[PATCH] elevator: Fix a race in elevator switching and md device initialization

[Tomoki Sekiyama <tomoki.sekiyama@hds.com>]

The soft lockup below happes at the boot time of the system using dm
multipath and automated elevator switching udev rules.

[  356.127001] BUG: soft lockup - CPU#3 stuck for 22s! [sh:483]
[  356.127001] RIP: 0010:[<ffffffff81072a7d>]  [<ffffffff81072a7d>] lock_timer_base.isra.35+0x1d/0x50
...
[  356.127001] Call Trace:
[  356.127001]  [<ffffffff81073810>] try_to_del_timer_sync+0x20/0x70
[  356.127001]  [<ffffffff8118b08a>] ? kmem_cache_alloc_node_trace+0x20a/0x230
[  356.127001]  [<ffffffff810738b2>] del_timer_sync+0x52/0x60
[  356.127001]  [<ffffffff812ece22>] cfq_exit_queue+0x32/0xf0
[  356.127001]  [<ffffffff812c98df>] elevator_exit+0x2f/0x50
[  356.127001]  [<ffffffff812c9f21>] elevator_change+0xf1/0x1c0
[  356.127001]  [<ffffffff812caa50>] elv_iosched_store+0x20/0x50
[  356.127001]  [<ffffffff812d1d09>] queue_attr_store+0x59/0xb0
[  356.127001]  [<ffffffff812143f6>] sysfs_write_file+0xc6/0x140
[  356.127001]  [<ffffffff811a326d>] vfs_write+0xbd/0x1e0
[  356.127001]  [<ffffffff811a3ca9>] SyS_write+0x49/0xa0
[  356.127001]  [<ffffffff8164e899>] system_call_fastpath+0x16/0x1b

This is caused by a race between md device initialization and sysfs knob
to switch the scheduler.

* multipathd:
 SyS_ioctl -> do_vfs_ioctl -> dm_ctl_ioctl -> ctl_ioctl ->  table_load
  -> dm_setup_md_queue -> blk_init_allocated_queue -> elevator_init:

    q->elevator = elevator_alloc(q, e); // not yet initialized

* sh -c 'echo deadline > /sys/$DEVPATH/queue/scheduler'
 SyS_write -> vfs_write -> sysfs_write_file -> queue_attr_store
     ( mutex_lock(&q->sysfs_lock) here. )
  -> elv_iosched_store -> elevator_change:

  elevator_exit(old); // try to de-init uninitialized elevator and hang up

This patch adds acquisition of q->sysfs_lock in blk_init_allocated_queue().
This also adds the lock into elevator_change() to ensure locking from the
other path, as it is exposed function (and queue_attr_store will uses
__elevator_change() now, the non-locking version of elevator_change()).

Signed-off-by: Tomoki Sekiyama <tomoki.sekiyama@hds.com>
---
 block/blk-core.c |    6 +++++-
 block/elevator.c |   16 ++++++++++++++--
 2 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 93a18d1..2323ec3 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -739,9 +739,13 @@ blk_init_allocated_queue(struct request_queue *q, request_fn_proc *rfn,
 
 	q->sg_reserved_size = INT_MAX;
 
+	/* Protect q->elevator from elevator_change */
+	mutex_lock(&q->sysfs_lock);
 	/* init elevator */
 	if (elevator_init(q, NULL))
-		return NULL;
+		q = NULL;
+	mutex_unlock(&q->sysfs_lock);
+
 	return q;
 }
 EXPORT_SYMBOL(blk_init_allocated_queue);
diff --git a/block/elevator.c b/block/elevator.c
index 668394d..5232565 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -959,7 +959,7 @@ fail_init:
 /*
  * Switch this queue to the given IO scheduler.
  */
-int elevator_change(struct request_queue *q, const char *name)
+static int __elevator_change(struct request_queue *q, const char *name)
 {
 	char elevator_name[ELV_NAME_MAX];
 	struct elevator_type *e;
@@ -981,6 +981,18 @@ int elevator_change(struct request_queue *q, const char *name)
 
 	return elevator_switch(q, e);
 }
+
+int elevator_change(struct request_queue *q, const char *name)
+{
+	int ret;
+
+	/* Protect q->elevator from blk_init_allocated_queue() */
+	mutex_lock(&q->sysfs_lock);
+	ret = __elevator_change(q, name);
+	mutex_unlock(&q->sysfs_lock);
+
+	return ret;
+}
 EXPORT_SYMBOL(elevator_change);
 
 ssize_t elv_iosched_store(struct request_queue *q, const char *name,
@@ -991,7 +1003,7 @@ ssize_t elv_iosched_store(struct request_queue *q, const char *name,
 	if (!q->elevator)
 		return count;
 
-	ret = elevator_change(q, name);
+	ret = __elevator_change(q, name);
 	if (!ret)
 		return count;