Commit b89f625e authored by Ming Lei's avatar Ming Lei Committed by Jens Axboe

block: don't release queue's sysfs lock during switching elevator

cecf5d87 ("block: split .sysfs_lock into two locks") starts to
release & acquire sysfs_lock before registering/un-registering elevator
queue during switching elevator for avoiding potential deadlock from
showing & storing 'queue/iosched' attributes and removing elevator's
kobject.

Turns out there isn't such deadlock because 'q->sysfs_lock' isn't
required in .show & .store of queue/iosched's attributes, and just
elevator's sysfs lock is acquired in elv_iosched_store() and
elv_iosched_show(). So it is safe to hold queue's sysfs lock when
registering/un-registering elevator queue.

The biggest issue is that commit cecf5d87 assumes that concurrent
write on 'queue/scheduler' can't happen. However, this assumption isn't
true, because kernfs_fop_write() only guarantees that concurrent write
aren't called on the same open file, but the write could be from
different open on the file. So we can't release & re-acquire queue's
sysfs lock during switching elevator, otherwise use-after-free on
elevator could be triggered.

Fixes the issue by not releasing queue's sysfs lock during switching
elevator.

Fixes: cecf5d87 ("block: split .sysfs_lock into two locks")
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Mike Snitzer <snitzer@redhat.com>
Reviewed-by: default avatarBart Van Assche <bvanassche@acm.org>
Signed-off-by: default avatarMing Lei <ming.lei@redhat.com>
Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent 284b94be
...@@ -989,13 +989,11 @@ int blk_register_queue(struct gendisk *disk) ...@@ -989,13 +989,11 @@ int blk_register_queue(struct gendisk *disk)
blk_mq_debugfs_register(q); blk_mq_debugfs_register(q);
} }
/* mutex_lock(&q->sysfs_lock);
* The flag of QUEUE_FLAG_REGISTERED isn't set yet, so elevator
* switch won't happen at all.
*/
if (q->elevator) { if (q->elevator) {
ret = elv_register_queue(q, false); ret = elv_register_queue(q, false);
if (ret) { if (ret) {
mutex_unlock(&q->sysfs_lock);
mutex_unlock(&q->sysfs_dir_lock); mutex_unlock(&q->sysfs_dir_lock);
kobject_del(&q->kobj); kobject_del(&q->kobj);
blk_trace_remove_sysfs(dev); blk_trace_remove_sysfs(dev);
...@@ -1005,7 +1003,6 @@ int blk_register_queue(struct gendisk *disk) ...@@ -1005,7 +1003,6 @@ int blk_register_queue(struct gendisk *disk)
has_elevator = true; has_elevator = true;
} }
mutex_lock(&q->sysfs_lock);
blk_queue_flag_set(QUEUE_FLAG_REGISTERED, q); blk_queue_flag_set(QUEUE_FLAG_REGISTERED, q);
wbt_enable_default(q); wbt_enable_default(q);
blk_throtl_register_queue(q); blk_throtl_register_queue(q);
...@@ -1062,12 +1059,10 @@ void blk_unregister_queue(struct gendisk *disk) ...@@ -1062,12 +1059,10 @@ void blk_unregister_queue(struct gendisk *disk)
kobject_del(&q->kobj); kobject_del(&q->kobj);
blk_trace_remove_sysfs(disk_to_dev(disk)); blk_trace_remove_sysfs(disk_to_dev(disk));
/* mutex_lock(&q->sysfs_lock);
* q->kobj has been removed, so it is safe to check if elevator
* exists without holding q->sysfs_lock.
*/
if (q->elevator) if (q->elevator)
elv_unregister_queue(q); elv_unregister_queue(q);
mutex_unlock(&q->sysfs_lock);
mutex_unlock(&q->sysfs_dir_lock); mutex_unlock(&q->sysfs_dir_lock);
kobject_put(&disk_to_dev(disk)->kobj); kobject_put(&disk_to_dev(disk)->kobj);
......
...@@ -503,9 +503,7 @@ int elv_register_queue(struct request_queue *q, bool uevent) ...@@ -503,9 +503,7 @@ int elv_register_queue(struct request_queue *q, bool uevent)
if (uevent) if (uevent)
kobject_uevent(&e->kobj, KOBJ_ADD); kobject_uevent(&e->kobj, KOBJ_ADD);
mutex_lock(&q->sysfs_lock);
e->registered = 1; e->registered = 1;
mutex_unlock(&q->sysfs_lock);
} }
return error; return error;
} }
...@@ -523,11 +521,9 @@ void elv_unregister_queue(struct request_queue *q) ...@@ -523,11 +521,9 @@ void elv_unregister_queue(struct request_queue *q)
kobject_uevent(&e->kobj, KOBJ_REMOVE); kobject_uevent(&e->kobj, KOBJ_REMOVE);
kobject_del(&e->kobj); kobject_del(&e->kobj);
mutex_lock(&q->sysfs_lock);
e->registered = 0; e->registered = 0;
/* Re-enable throttling in case elevator disabled it */ /* Re-enable throttling in case elevator disabled it */
wbt_enable_default(q); wbt_enable_default(q);
mutex_unlock(&q->sysfs_lock);
} }
} }
...@@ -590,32 +586,11 @@ int elevator_switch_mq(struct request_queue *q, ...@@ -590,32 +586,11 @@ int elevator_switch_mq(struct request_queue *q,
lockdep_assert_held(&q->sysfs_lock); lockdep_assert_held(&q->sysfs_lock);
if (q->elevator) { if (q->elevator) {
if (q->elevator->registered) { if (q->elevator->registered)
mutex_unlock(&q->sysfs_lock);
/*
* Concurrent elevator switch can't happen becasue
* sysfs write is always exclusively on same file.
*
* Also the elevator queue won't be freed after
* sysfs_lock is released becasue kobject_del() in
* blk_unregister_queue() waits for completion of
* .store & .show on its attributes.
*/
elv_unregister_queue(q); elv_unregister_queue(q);
mutex_lock(&q->sysfs_lock);
}
ioc_clear_queue(q); ioc_clear_queue(q);
elevator_exit(q, q->elevator); elevator_exit(q, q->elevator);
/*
* sysfs_lock may be dropped, so re-check if queue is
* unregistered. If yes, don't switch to new elevator
* any more
*/
if (!blk_queue_registered(q))
return 0;
} }
ret = blk_mq_init_sched(q, new_e); ret = blk_mq_init_sched(q, new_e);
...@@ -623,11 +598,7 @@ int elevator_switch_mq(struct request_queue *q, ...@@ -623,11 +598,7 @@ int elevator_switch_mq(struct request_queue *q,
goto out; goto out;
if (new_e) { if (new_e) {
mutex_unlock(&q->sysfs_lock);
ret = elv_register_queue(q, true); ret = elv_register_queue(q, true);
mutex_lock(&q->sysfs_lock);
if (ret) { if (ret) {
elevator_exit(q, q->elevator); elevator_exit(q, q->elevator);
goto out; goto out;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment