[PATCH v2] bluetooth: hci: Fix UAF from MGMT_OP_REMOVE_ADV_MONITOR during closure

[Mazin Al Haddad <mazin@getstate.dev>]

When hci_dev is closing down, mgmt_power_off will free parts of the device.
The freed memory can then be accessed when processing pending
MGMT_OP_REMOVE_ADV_MONITOR cmds. Since submitting the command is allowed
when it is powered off (as in previous discussions linked below), fix
this by returning MGMT_STATUS_BUSY to pending MGMT_OP_REMOVE_ADV_MONITOR
operations submitted as hci_dev_close_sync is running. Avoid processing
pending cmds since doing so will lead to reacquiring the same lock. Add
a sanity check within mgmt_remove_adv_monitor to ensure the cmd is still
valid and exit early if not.

BUG: KASAN: slab-use-after-free in mgmt_remove_adv_monitor_sync+0x3a/0xd0
net/bluetooth/mgmt.c:5543
Read of size 8 at addr ffff88814128f898 by task kworker/u9:4/5961

 <TASK>
 __dump_stack lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0x169/0x550 mm/kasan/report.c:489
 kasan_report+0x143/0x180 mm/kasan/report.c:602
 mgmt_remove_adv_monitor_sync+0x3a/0xd0 net/bluetooth/mgmt.c:5543
 hci_cmd_sync_work+0x22b/0x400 net/bluetooth/hci_sync.c:332
 process_one_work kernel/workqueue.c:3229 [inline]
 process_scheduled_works+0xa63/0x1850 kernel/workqueue.c:3310
 worker_thread+0x870/0xd30 kernel/workqueue.c:3391
 kthread+0x2f0/0x390 kernel/kthread.c:389
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
 </TASK>

Freed by task 16022:
 kasan_save_stack mm/kasan/common.c:47 [inline]
 kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
 kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:582
 poison_slab_object mm/kasan/common.c:247 [inline]
 __kasan_slab_free+0x59/0x70 mm/kasan/common.c:264
 kasan_slab_free include/linux/kasan.h:233 [inline]
 slab_free_hook mm/slub.c:2338 [inline]
 slab_free mm/slub.c:4598 [inline]
 kfree+0x196/0x420 mm/slub.c:4746
 mgmt_pending_foreach+0xd1/0x130 net/bluetooth/mgmt_util.c:259
 __mgmt_power_off+0x183/0x430 net/bluetooth/mgmt.c:9550
 hci_dev_close_sync+0x6c4/0x11c0 net/bluetooth/hci_sync.c:5208
 hci_dev_do_close net/bluetooth/hci_core.c:483 [inline]
 hci_dev_close+0x112/0x210 net/bluetooth/hci_core.c:508
 sock_do_ioctl+0x158/0x460 net/socket.c:1209
 sock_ioctl+0x626/0x8e0 net/socket.c:1328
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:906 [inline]
 __se_sys_ioctl+0xf5/0x170 fs/ioctl.c:892
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Link: https://lore.kernel.org/lkml/20240424135903.24169-1-jlee@suse.com/
Reported-by: syzbot+479aff51bb361ef5aa18@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=479aff51bb361ef5aa18
Signed-off-by: Mazin Al Haddad <mazin@getstate.dev>
---
Changes since v1:
 * Change return code to ECANCELED
 * Send out MGMT_STATUS_CANCELLED instead of MGMT_STATUS_BUSY
 * Style fixes 

 net/bluetooth/hci_sync.c |  5 +++--
 net/bluetooth/mgmt.c     | 20 ++++++++++++++++++--
 2 files changed, 21 insertions(+), 4 deletions(-)

diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index c86f4e42e69c..aa5aa3fed32d 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -5197,6 +5197,9 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 	 */
 	drain_workqueue(hdev->workqueue);
 
+	/* flush cmd  work */
+	flush_work(&hdev->cmd_work);
+
 	hci_dev_lock(hdev);
 
 	hci_discovery_set_state(hdev, DISCOVERY_STOPPED);
@@ -5234,8 +5237,6 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 		clear_bit(HCI_INIT, &hdev->flags);
 	}
 
-	/* flush cmd  work */
-	flush_work(&hdev->cmd_work);
 
 	/* Drop queues */
 	skb_queue_purge(&hdev->rx_q);
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index b31192d473d0..ec86ae851e56 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -5519,9 +5519,17 @@ static void mgmt_remove_adv_monitor_complete(struct hci_dev *hdev,
 {
 	struct mgmt_rp_remove_adv_monitor rp;
 	struct mgmt_pending_cmd *cmd = data;
-	struct mgmt_cp_remove_adv_monitor *cp = cmd->param;
+	struct mgmt_cp_remove_adv_monitor *cp;
+
+	// if executing while device is closing down, status could
+	// be invalid as pending cmd could be removed by __mgmt_power_off
+	// so exit early if the device was busy.
+	if (status == -ECANCELED ||
+	    cmd != pending_find(MGMT_OP_REMOVE_ADV_MONITOR, hdev))
+		return;
 
 	hci_dev_lock(hdev);
+	cp = cmd->param;
 
 	rp.monitor_handle = cp->monitor_handle;
 
@@ -5540,6 +5548,10 @@ static void mgmt_remove_adv_monitor_complete(struct hci_dev *hdev,
 static int mgmt_remove_adv_monitor_sync(struct hci_dev *hdev, void *data)
 {
 	struct mgmt_pending_cmd *cmd = data;
+
+	if (cmd != pending_find(MGMT_OP_REMOVE_ADV_MONITOR, hdev))
+		return -ECANCELED;
+
 	struct mgmt_cp_remove_adv_monitor *cp = cmd->param;
 	u16 handle = __le16_to_cpu(cp->monitor_handle);
 
@@ -9544,8 +9556,12 @@ void __mgmt_power_off(struct hci_dev *hdev)
 	 */
 	if (hci_dev_test_flag(hdev, HCI_UNREGISTER))
 		match.mgmt_status = MGMT_STATUS_INVALID_INDEX;
-	else
+	else {
+		match.mgmt_status = MGMT_STATUS_CANCELLED;
+		mgmt_pending_foreach(MGMT_OP_REMOVE_ADV_MONITOR, hdev,
+				     cmd_status_rsp, &match);
 		match.mgmt_status = MGMT_STATUS_NOT_POWERED;
+	}
 
 	mgmt_pending_foreach(0, hdev, cmd_complete_rsp, &match);
 

base-commit: 499551201b5f4fd3c0618a3e95e3d0d15ea18f31
-- 
2.46.0