======================================================
WARNING: possible circular locking dependency detected
5.19.0-rc5-next-20220707 #1 Not tainted
------------------------------------------------------
syz-executor.3/4472 is trying to acquire lock:
ffff8880451e0130 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}, at: sco_conn_del+0x11c/0x2c0

but task is already holding lock:
ffffffff857d26e8 (hci_cb_list_lock){+.+.}-{3:3}, at: hci_conn_hash_flush+0xc7/0x230

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 (hci_cb_list_lock){+.+.}-{3:3}:
       __mutex_lock+0x136/0x14d0
       hci_connect_cfm+0x26/0x140
       hci_remote_features_evt+0x479/0x870
       hci_event_packet+0x779/0xee0
       hci_rx_work+0x499/0xba0
       process_one_work+0xa0f/0x1690
       worker_thread+0x637/0x1250
       kthread+0x2ed/0x3a0
       ret_from_fork+0x22/0x30

-> #1 (&hdev->lock){+.+.}-{3:3}:
       __mutex_lock+0x136/0x14d0
       sco_sock_connect+0x1e0/0xa60
       __sys_connect_file+0x151/0x190
       __sys_connect+0x161/0x190
       __x64_sys_connect+0x6f/0xb0
       do_syscall_64+0x3b/0x90
       entry_SYSCALL_64_after_hwframe+0x46/0xb0

-> #0 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}:
       __lock_acquire+0x29fe/0x5e70
       lock_acquire+0x1a2/0x530
       lock_sock_nested+0x3d/0xf0
       sco_conn_del+0x11c/0x2c0
       sco_disconn_cfm+0x62/0x80
       hci_conn_hash_flush+0x119/0x230
       hci_dev_close_sync+0x4dc/0xf20
       hci_unregister_dev+0x10f/0x380
       vhci_release+0x7c/0xf0
       __fput+0x272/0x9d0
       task_work_run+0xe2/0x1a0
       do_exit+0xaf4/0x27d0
       do_group_exit+0xd2/0x2f0
       get_signal+0x2201/0x24b0
       arch_do_signal_or_restart+0x89/0x1be0
       exit_to_user_mode_prepare+0x131/0x1a0
       syscall_exit_to_user_mode+0x19/0x40
       do_syscall_64+0x48/0x90
       entry_SYSCALL_64_after_hwframe+0x46/0xb0

other info that might help us debug this:

Chain exists of:
  sk_lock-AF_BLUETOOTH-BTPROTO_SCO --> &hdev->lock --> hci_cb_list_lock

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(hci_cb_list_lock);
                               lock(&hdev->lock);
                               lock(hci_cb_list_lock);
  lock(sk_lock-AF_BLUETOOTH-BTPROTO_SCO);

 *** DEADLOCK ***

3 locks held by syz-executor.3/4472:
 #0: ffff888016e2d048 (&hdev->req_lock){+.+.}-{3:3}, at: hci_unregister_dev+0x107/0x380
 #1: ffff888016e2c078 (&hdev->lock){+.+.}-{3:3}, at: hci_dev_close_sync+0x241/0xf20
 #2: ffffffff857d26e8 (hci_cb_list_lock){+.+.}-{3:3}, at: hci_conn_hash_flush+0xc7/0x230

stack backtrace:
CPU: 1 PID: 4472 Comm: syz-executor.3 Not tainted 5.19.0-rc5-next-20220707 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
Call Trace:
 <TASK>
 dump_stack_lvl+0x8b/0xb3
 check_noncircular+0x25f/0x2e0
 __lock_acquire+0x29fe/0x5e70
 lock_acquire+0x1a2/0x530
 lock_sock_nested+0x3d/0xf0
 sco_conn_del+0x11c/0x2c0
 sco_disconn_cfm+0x62/0x80
 hci_conn_hash_flush+0x119/0x230
 hci_dev_close_sync+0x4dc/0xf20
 hci_unregister_dev+0x10f/0x380
 vhci_release+0x7c/0xf0
 __fput+0x272/0x9d0
 task_work_run+0xe2/0x1a0
 do_exit+0xaf4/0x27d0
 do_group_exit+0xd2/0x2f0
 get_signal+0x2201/0x24b0
 arch_do_signal_or_restart+0x89/0x1be0
 exit_to_user_mode_prepare+0x131/0x1a0
 syscall_exit_to_user_mode+0x19/0x40
 do_syscall_64+0x48/0x90
 entry_SYSCALL_64_after_hwframe+0x46/0xb0
RIP: 0033:0x7f7e70e18b19
Code: Unable to access opcode bytes at RIP 0x7f7e70e18aef.
RSP: 002b:00007f7e6e36d218 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca
RAX: fffffffffffffe00 RBX: 00007f7e70f2c028 RCX: 00007f7e70e18b19
RDX: 0000000000000000 RSI: 0000000000000080 RDI: 00007f7e70f2c028
RBP: 00007f7e70f2c020 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f7e70f2c02c
R13: 00007ffe7d2add5f R14: 00007f7e6e36d300 R15: 0000000000022000
 </TASK>
Bluetooth: hci4: Opcode 0x c03 failed: -110
Bluetooth: hci4: Opcode 0x c03 failed: -110