[syzbot] possible deadlock in rfcomm_dlc_exists

12 views

Skip to first unread message

syzbot

unread,

Jan 20, 2023, 11:28:51 AM1/20/23

to da...@davemloft.net, edum...@google.com, johan....@gmail.com, ku...@kernel.org, linux-b...@vger.kernel.org, linux-...@vger.kernel.org, luiz....@gmail.com, mar...@holtmann.org, net...@vger.kernel.org, pab...@redhat.com, syzkall...@googlegroups.com, yangyi...@huawei.com

Hello,

syzbot found the following issue on:

HEAD commit: c12e2e5b76b2 Add linux-next specific files for 20230116
git tree: linux-next
console output: https://syzkaller.appspot.com/x/log.txt?x=154e0046480000
kernel config: https://syzkaller.appspot.com/x/.config?x=ef6b6ac6c6c96c0e
dashboard link: https://syzkaller.appspot.com/bug?extid=b69a625d06e8ece26415
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2

Unfortunately, I don't have any reproducer for this issue yet.

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/4fb49204daa9/disk-c12e2e5b.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/7b02ca8eacc0/vmlinux-c12e2e5b.xz
kernel image: https://storage.googleapis.com/syzbot-assets/94539232cf54/bzImage-c12e2e5b.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+b69a62...@syzkaller.appspotmail.com

======================================================
WARNING: possible circular locking dependency detected
6.2.0-rc4-next-20230116-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor.2/27510 is trying to acquire lock:
ffffffff8e322188 (rfcomm_mutex){+.+.}-{3:3}, at: rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:542

but task is already holding lock:
ffffffff8e3270c8 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
ffffffff8e3270c8 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #3 (rfcomm_ioctl_mutex){+.+.}-{3:3}:
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880
sock_do_ioctl+0xcc/0x230 net/socket.c:1194
sock_ioctl+0x1f8/0x680 net/socket.c:1311
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #2 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}:
lock_sock_nested+0x3a/0xf0 net/core/sock.c:3470
lock_sock include/net/sock.h:1725 [inline]
rfcomm_sk_state_change+0x6d/0x3a0 net/bluetooth/rfcomm/sock.c:73
__rfcomm_dlc_close+0x1b1/0x890 net/bluetooth/rfcomm/core.c:489
rfcomm_dlc_close+0x1e9/0x240 net/bluetooth/rfcomm/core.c:520
__rfcomm_sock_close+0x17a/0x2f0 net/bluetooth/rfcomm/sock.c:220
rfcomm_sock_shutdown+0xd8/0x230 net/bluetooth/rfcomm/sock.c:912
rfcomm_sock_release+0x68/0x140 net/bluetooth/rfcomm/sock.c:933
__sock_release+0xcd/0x280 net/socket.c:651
sock_close+0x1c/0x20 net/socket.c:1390
__fput+0x27c/0xa90 fs/file_table.c:321
task_work_run+0x16f/0x270 kernel/task_work.c:179
get_signal+0x1c7/0x24f0 kernel/signal.c:2635
arch_do_signal_or_restart+0x79/0x5c0 arch/x86/kernel/signal.c:306
exit_to_user_mode_loop kernel/entry/common.c:168 [inline]
exit_to_user_mode_prepare+0x11f/0x240 kernel/entry/common.c:204
__syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline]
syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:297
do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #1 (&d->lock){+.+.}-{3:3}:
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
__rfcomm_dlc_close+0x15d/0x890 net/bluetooth/rfcomm/core.c:487
rfcomm_dlc_close+0x1e9/0x240 net/bluetooth/rfcomm/core.c:520
__rfcomm_sock_close+0x17a/0x2f0 net/bluetooth/rfcomm/sock.c:220
rfcomm_sock_shutdown+0xd8/0x230 net/bluetooth/rfcomm/sock.c:912
rfcomm_sock_release+0x68/0x140 net/bluetooth/rfcomm/sock.c:933
__sock_release+0xcd/0x280 net/socket.c:651
sock_close+0x1c/0x20 net/socket.c:1390
__fput+0x27c/0xa90 fs/file_table.c:321
task_work_run+0x16f/0x270 kernel/task_work.c:179
get_signal+0x1c7/0x24f0 kernel/signal.c:2635
arch_do_signal_or_restart+0x79/0x5c0 arch/x86/kernel/signal.c:306
exit_to_user_mode_loop kernel/entry/common.c:168 [inline]
exit_to_user_mode_prepare+0x11f/0x240 kernel/entry/common.c:204
__syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline]
syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:297
do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #0 (rfcomm_mutex){+.+.}-{3:3}:
check_prev_add kernel/locking/lockdep.c:3107 [inline]
check_prevs_add kernel/locking/lockdep.c:3226 [inline]
validate_chain kernel/locking/lockdep.c:3841 [inline]
__lock_acquire+0x2a9d/0x5780 kernel/locking/lockdep.c:5073
lock_acquire.part.0+0x11c/0x350 kernel/locking/lockdep.c:5690
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:542
__rfcomm_create_dev net/bluetooth/rfcomm/tty.c:414 [inline]
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:485 [inline]
rfcomm_dev_ioctl+0x966/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880
sock_do_ioctl+0xcc/0x230 net/socket.c:1194
sock_ioctl+0x1f8/0x680 net/socket.c:1311
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

other info that might help us debug this:

Chain exists of:
rfcomm_mutex --> sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM --> rfcomm_ioctl_mutex

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(rfcomm_ioctl_mutex);
lock(sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM);
lock(rfcomm_ioctl_mutex);
lock(rfcomm_mutex);

*** DEADLOCK ***

2 locks held by syz-executor.2/27510:
#0: ffff88804d0ad130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1725 [inline]
#0: ffff88804d0ad130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: rfcomm_sock_ioctl+0xaa/0xe0 net/bluetooth/rfcomm/sock.c:879
#1: ffffffff8e3270c8 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
#1: ffffffff8e3270c8 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587

stack backtrace:
CPU: 1 PID: 27510 Comm: syz-executor.2 Not tainted 6.2.0-rc4-next-20230116-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xd1/0x138 lib/dump_stack.c:106
check_noncircular+0x25f/0x2e0 kernel/locking/lockdep.c:2187
check_prev_add kernel/locking/lockdep.c:3107 [inline]
check_prevs_add kernel/locking/lockdep.c:3226 [inline]
validate_chain kernel/locking/lockdep.c:3841 [inline]
__lock_acquire+0x2a9d/0x5780 kernel/locking/lockdep.c:5073
lock_acquire.part.0+0x11c/0x350 kernel/locking/lockdep.c:5690
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:542
__rfcomm_create_dev net/bluetooth/rfcomm/tty.c:414 [inline]
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:485 [inline]
rfcomm_dev_ioctl+0x966/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880
sock_do_ioctl+0xcc/0x230 net/socket.c:1194
sock_ioctl+0x1f8/0x680 net/socket.c:1311
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7efe0208c0c9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 19 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007efe02d1a168 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007efe021ac120 RCX: 00007efe0208c0c9
RDX: 0000000020000100 RSI: 00000000400452c8 RDI: 0000000000000008
RBP: 00007efe020e7ae9 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007ffe7923c9af R14: 00007efe02d1a300 R15: 0000000000022000
</TASK>

---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

syzbot

unread,

Mar 30, 2023, 8:21:47 PM3/30/23

syzbot has found a reproducer for the following issue on:

HEAD commit: a6d9e3034536 Add linux-next specific files for 20230330
git tree: linux-next
console+strace: https://syzkaller.appspot.com/x/log.txt?x=152adc3ec80000
kernel config: https://syzkaller.appspot.com/x/.config?x=aceb117f7924508e

dashboard link: https://syzkaller.appspot.com/bug?extid=b69a625d06e8ece26415
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2

syz repro: https://syzkaller.appspot.com/x/repro.syz?x=153acb85c80000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=179d1ed1c80000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/ec1f900ea929/disk-a6d9e303.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/fabbf89c0d22/vmlinux-a6d9e303.xz
kernel image: https://storage.googleapis.com/syzbot-assets/1ed05d6192fa/bzImage-a6d9e303.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+b69a62...@syzkaller.appspotmail.com

======================================================
WARNING: possible circular locking dependency detected

6.3.0-rc4-next-20230330-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor363/5115 is trying to acquire lock:
ffffffff8e357cc8 (rfcomm_mutex){+.+.}-{3:3}, at: rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:546

but task is already holding lock:

ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #3 (rfcomm_ioctl_mutex){+.+.}-{3:3}:
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880

sock_do_ioctl+0xcc/0x230 net/socket.c:1199
sock_ioctl+0x1f8/0x680 net/socket.c:1316

lock_sock_nested+0x3a/0xf0 net/core/sock.c:3474
lock_sock include/net/sock.h:1697 [inline]
rfcomm_sk_state_change+0x6d/0x3a0 net/bluetooth/rfcomm/sock.c:73
__rfcomm_dlc_close+0x1b1/0x890 net/bluetooth/rfcomm/core.c:493
rfcomm_dlc_close+0x1e9/0x240 net/bluetooth/rfcomm/core.c:524

__rfcomm_sock_close+0x17a/0x2f0 net/bluetooth/rfcomm/sock.c:220
rfcomm_sock_shutdown+0xd8/0x230 net/bluetooth/rfcomm/sock.c:912
rfcomm_sock_release+0x68/0x140 net/bluetooth/rfcomm/sock.c:933

__sock_release+0xcd/0x290 net/socket.c:653
sock_close+0x1c/0x20 net/socket.c:1395

__fput+0x27c/0xa90 fs/file_table.c:321
task_work_run+0x16f/0x270 kernel/task_work.c:179

exit_task_work include/linux/task_work.h:38 [inline]
do_exit+0xb0d/0x29f0 kernel/exit.c:869
do_group_exit+0xd4/0x2a0 kernel/exit.c:1019
get_signal+0x2315/0x25b0 kernel/signal.c:2859
arch_do_signal_or_restart+0x79/0x5c0 arch/x86/kernel/signal.c:307

exit_to_user_mode_loop kernel/entry/common.c:168 [inline]
exit_to_user_mode_prepare+0x11f/0x240 kernel/entry/common.c:204
__syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline]
syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:297
do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #1 (&d->lock){+.+.}-{3:3}:
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747

__rfcomm_dlc_close+0x15d/0x890 net/bluetooth/rfcomm/core.c:491
rfcomm_dlc_close+0x1e9/0x240 net/bluetooth/rfcomm/core.c:524

__rfcomm_sock_close+0x17a/0x2f0 net/bluetooth/rfcomm/sock.c:220
rfcomm_sock_shutdown+0xd8/0x230 net/bluetooth/rfcomm/sock.c:912
rfcomm_sock_release+0x68/0x140 net/bluetooth/rfcomm/sock.c:933

__sock_release+0xcd/0x290 net/socket.c:653
sock_close+0x1c/0x20 net/socket.c:1395

__fput+0x27c/0xa90 fs/file_table.c:321
task_work_run+0x16f/0x270 kernel/task_work.c:179

check_prev_add kernel/locking/lockdep.c:3108 [inline]
check_prevs_add kernel/locking/lockdep.c:3227 [inline]
validate_chain kernel/locking/lockdep.c:3842 [inline]
__lock_acquire+0x2f21/0x5df0 kernel/locking/lockdep.c:5074
lock_acquire.part.0+0x11c/0x370 kernel/locking/lockdep.c:5691

__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747

rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:546

__rfcomm_create_dev net/bluetooth/rfcomm/tty.c:414 [inline]
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:485 [inline]
rfcomm_dev_ioctl+0x966/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880

sock_do_ioctl+0xcc/0x230 net/socket.c:1199
sock_ioctl+0x1f8/0x680 net/socket.c:1316

vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

other info that might help us debug this:

Chain exists of:
rfcomm_mutex --> sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM --> rfcomm_ioctl_mutex

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(rfcomm_ioctl_mutex);
lock(sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM);
lock(rfcomm_ioctl_mutex);
lock(rfcomm_mutex);

*** DEADLOCK ***

2 locks held by syz-executor363/5115:
#0: ffff888146eb7130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1697 [inline]
#0: ffff888146eb7130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: rfcomm_sock_ioctl+0xaa/0xe0 net/bluetooth/rfcomm/sock.c:879
#1: ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
#1: ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587

stack backtrace:
CPU: 0 PID: 5115 Comm: syz-executor363 Not tainted 6.3.0-rc4-next-20230330-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/02/2023

Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]

dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
check_noncircular+0x25f/0x2e0 kernel/locking/lockdep.c:2188
check_prev_add kernel/locking/lockdep.c:3108 [inline]
check_prevs_add kernel/locking/lockdep.c:3227 [inline]
validate_chain kernel/locking/lockdep.c:3842 [inline]
__lock_acquire+0x2f21/0x5df0 kernel/locking/lockdep.c:5074
lock_acquire.part.0+0x11c/0x370 kernel/locking/lockdep.c:5691

__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747

rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:546

sock_do_ioctl+0xcc/0x230 net/socket.c:1199
sock_ioctl+0x1f8/0x680 net/socket.c:1316

RIP: 0033:0x7fdd90dc3379
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 d1 15 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffc287c58b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007fdd90dc3379
RDX: 0000000020000100 RSI: 00000000400452c8 RDI: 0000000000000006
RBP: 0000000000000000 R08: 0000000000000000 R09: 00007ffc287c58e8
R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffc287c5900
R13: 00007ffc287c5910 R14: 000000000001c039 R15: 00007ffc287c58d0
</TASK>

Hillf Danton

unread,

Mar 30, 2023, 11:20:21 PM3/30/23

to syzbot, linux-...@vger.kernel.org, syzkall...@googlegroups.com

On 30 Mar 2023 17:21:46 -0700

> HEAD commit: a6d9e3034536 Add linux-next specific files for 20230330
> git tree: linux-next

> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=179d1ed1c80000

Lock sock without dlc locked.

#syz test https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git a6d9e3034536

--- x/net/bluetooth/rfcomm/core.c
+++ y/net/bluetooth/rfcomm/core.c
@@ -490,8 +490,8 @@ static int __rfcomm_dlc_close(struct rfc

rfcomm_dlc_lock(d);
d->state = BT_CLOSED;
- d->state_change(d, err);
rfcomm_dlc_unlock(d);
+ d->state_change(d, err);

skb_queue_purge(&d->tx_queue);
rfcomm_dlc_unlink(d);
--

syzbot

unread,

Mar 30, 2023, 11:37:22 PM3/30/23

to hda...@sina.com, linux-...@vger.kernel.org, syzkall...@googlegroups.com

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
possible deadlock in rfcomm_sk_state_change

======================================================
WARNING: possible circular locking dependency detected

6.3.0-rc4-next-20230330-syzkaller-dirty #0 Not tainted
------------------------------------------------------
syz-executor.0/5614 is trying to acquire lock:
ffff888028fb5130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1697 [inline]
ffff888028fb5130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: rfcomm_sk_state_change+0x6d/0x3a0 net/bluetooth/rfcomm/sock.c:73

but task is already holding lock:

ffffffff8e357cc8 (rfcomm_mutex){+.+.}-{3:3}, at: rfcomm_dlc_close+0x33/0x240 net/bluetooth/rfcomm/core.c:511

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #2 (rfcomm_mutex){+.+.}-{3:3}:

__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
rfcomm_dlc_exists+0x58/0x190 net/bluetooth/rfcomm/core.c:546
__rfcomm_create_dev net/bluetooth/rfcomm/tty.c:414 [inline]
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:485 [inline]
rfcomm_dev_ioctl+0x966/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880
sock_do_ioctl+0xcc/0x230 net/socket.c:1199
sock_ioctl+0x1f8/0x680 net/socket.c:1316
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #1 (rfcomm_ioctl_mutex){+.+.}-{3:3}:

__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:484 [inline]
rfcomm_dev_ioctl+0x8a2/0x1c00 net/bluetooth/rfcomm/tty.c:587
rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880
sock_do_ioctl+0xcc/0x230 net/socket.c:1199
sock_ioctl+0x1f8/0x680 net/socket.c:1316
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #0 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}:

lock_sock_nested+0x3a/0xf0 net/core/sock.c:3474
lock_sock include/net/sock.h:1697 [inline]
rfcomm_sk_state_change+0x6d/0x3a0 net/bluetooth/rfcomm/sock.c:73

__rfcomm_dlc_close+0x1b9/0x890 net/bluetooth/rfcomm/core.c:494

rfcomm_dlc_close+0x1e9/0x240 net/bluetooth/rfcomm/core.c:524
__rfcomm_sock_close+0x17a/0x2f0 net/bluetooth/rfcomm/sock.c:220
rfcomm_sock_shutdown+0xd8/0x230 net/bluetooth/rfcomm/sock.c:912
rfcomm_sock_release+0x68/0x140 net/bluetooth/rfcomm/sock.c:933
__sock_release+0xcd/0x290 net/socket.c:653
sock_close+0x1c/0x20 net/socket.c:1395
__fput+0x27c/0xa90 fs/file_table.c:321
task_work_run+0x16f/0x270 kernel/task_work.c:179

get_signal+0x1c7/0x25b0 kernel/signal.c:2635

arch_do_signal_or_restart+0x79/0x5c0 arch/x86/kernel/signal.c:307
exit_to_user_mode_loop kernel/entry/common.c:168 [inline]
exit_to_user_mode_prepare+0x11f/0x240 kernel/entry/common.c:204
__syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline]
syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:297
do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x63/0xcd

other info that might help us debug this:

Chain exists of:

sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM --> rfcomm_ioctl_mutex --> rfcomm_mutex

Possible unsafe locking scenario:

CPU0 CPU1
---- ----

lock(rfcomm_mutex);
lock(rfcomm_ioctl_mutex);
lock(rfcomm_mutex);
lock(sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM);

*** DEADLOCK ***

2 locks held by syz-executor.0/5614:
#0: ffff888074585c10 (&sb->s_type->i_mutex_key#10){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:761 [inline]
#0: ffff888074585c10 (&sb->s_type->i_mutex_key#10){+.+.}-{3:3}, at: __sock_release+0x86/0x290 net/socket.c:652
#1: ffffffff8e357cc8 (rfcomm_mutex){+.+.}-{3:3}, at: rfcomm_dlc_close+0x33/0x240 net/bluetooth/rfcomm/core.c:511

stack backtrace:
CPU: 1 PID: 5614 Comm: syz-executor.0 Not tainted 6.3.0-rc4-next-20230330-syzkaller-dirty #0

Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/02/2023
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
check_noncircular+0x25f/0x2e0 kernel/locking/lockdep.c:2188
check_prev_add kernel/locking/lockdep.c:3108 [inline]
check_prevs_add kernel/locking/lockdep.c:3227 [inline]
validate_chain kernel/locking/lockdep.c:3842 [inline]
__lock_acquire+0x2f21/0x5df0 kernel/locking/lockdep.c:5074
lock_acquire.part.0+0x11c/0x370 kernel/locking/lockdep.c:5691

lock_sock_nested+0x3a/0xf0 net/core/sock.c:3474
lock_sock include/net/sock.h:1697 [inline]
rfcomm_sk_state_change+0x6d/0x3a0 net/bluetooth/rfcomm/sock.c:73

__rfcomm_dlc_close+0x1b9/0x890 net/bluetooth/rfcomm/core.c:494

get_signal+0x1c7/0x25b0 kernel/signal.c:2635

RIP: 0033:0x7fe4d528c0f9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 19 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fe4d602c168 EFLAGS: 00000246 ORIG_RAX: 000000000000002a
RAX: fffffffffffffffc RBX: 00007fe4d53abf80 RCX: 00007fe4d528c0f9
RDX: 000000000000000a RSI: 0000000020000040 RDI: 0000000000000003
RBP: 00007fe4d52e7b39 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007ffe4d1bd46f R14: 00007fe4d602c300 R15: 0000000000022000
</TASK>

Tested on:

commit: a6d9e303 Add linux-next specific files for 20230330
git tree: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
console output: https://syzkaller.appspot.com/x/log.txt?x=15f008c9c80000

kernel config: https://syzkaller.appspot.com/x/.config?x=aceb117f7924508e
dashboard link: https://syzkaller.appspot.com/bug?extid=b69a625d06e8ece26415
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2

patch: https://syzkaller.appspot.com/x/patch.diff?x=1032965dc80000

Hillf Danton

unread,

Mar 31, 2023, 4:01:29 AM3/31/23

to syzbot, linux-...@vger.kernel.org, syzkall...@googlegroups.com

On 30 Mar 2023 17:21:46 -0700

> HEAD commit: a6d9e3034536 Add linux-next specific files for 20230330
> git tree: linux-next

--- x/net/bluetooth/rfcomm/tty.c
+++ y/net/bluetooth/rfcomm/tty.c
@@ -394,6 +394,7 @@ static int __rfcomm_create_dev(struct so
struct rfcomm_dlc *dlc;
int id;

+ lockdep_assert_held(&sk->sk_lock);
if (copy_from_user(&req, arg, sizeof(req)))
return -EFAULT;

@@ -410,8 +411,10 @@ static int __rfcomm_create_dev(struct so
dlc = rfcomm_pi(sk)->dlc;
rfcomm_dlc_hold(dlc);
} else {
+ release_sock(sk);
/* Validate the channel is unused */
dlc = rfcomm_dlc_exists(&req.src, &req.dst, req.channel);
+ lock_sock(sk);
if (IS_ERR(dlc))
return PTR_ERR(dlc);
if (dlc)
--

syzbot

unread,

Mar 31, 2023, 4:16:26 AM3/31/23

to hda...@sina.com, linux-...@vger.kernel.org, syzkall...@googlegroups.com

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:

possible deadlock in rfcomm_dev_ioctl

======================================================
WARNING: possible circular locking dependency detected
6.3.0-rc4-next-20230330-syzkaller-dirty #0 Not tainted
------------------------------------------------------

syz-executor.0/5603 is trying to acquire lock:
ffff88801e179130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1697 [inline]
ffff88801e179130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: __rfcomm_create_dev net/bluetooth/rfcomm/tty.c:417 [inline]
ffff88801e179130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:488 [inline]
ffff88801e179130 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}, at: rfcomm_dev_ioctl+0x9ca/0x1cb0 net/bluetooth/rfcomm/tty.c:590

but task is already holding lock:

ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:487 [inline]
ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_dev_ioctl+0x8a2/0x1cb0 net/bluetooth/rfcomm/tty.c:590

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (rfcomm_ioctl_mutex){+.+.}-{3:3}:
__mutex_lock_common kernel/locking/mutex.c:603 [inline]
__mutex_lock+0x12f/0x1350 kernel/locking/mutex.c:747

rfcomm_create_dev net/bluetooth/rfcomm/tty.c:487 [inline]
rfcomm_dev_ioctl+0x8a2/0x1cb0 net/bluetooth/rfcomm/tty.c:590

rfcomm_sock_ioctl+0xb7/0xe0 net/bluetooth/rfcomm/sock.c:880
sock_do_ioctl+0xcc/0x230 net/socket.c:1199
sock_ioctl+0x1f8/0x680 net/socket.c:1316
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x197/0x210 fs/ioctl.c:856
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #0 (sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM){+.+.}-{0:0}:
check_prev_add kernel/locking/lockdep.c:3108 [inline]
check_prevs_add kernel/locking/lockdep.c:3227 [inline]
validate_chain kernel/locking/lockdep.c:3842 [inline]
__lock_acquire+0x2f21/0x5df0 kernel/locking/lockdep.c:5074
lock_acquire.part.0+0x11c/0x370 kernel/locking/lockdep.c:5691
lock_sock_nested+0x3a/0xf0 net/core/sock.c:3474
lock_sock include/net/sock.h:1697 [inline]

__rfcomm_create_dev net/bluetooth/rfcomm/tty.c:417 [inline]
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:488 [inline]
rfcomm_dev_ioctl+0x9ca/0x1cb0 net/bluetooth/rfcomm/tty.c:590

other info that might help us debug this:

Possible unsafe locking scenario:

CPU0 CPU1
---- ----

lock(rfcomm_ioctl_mutex);
lock(sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM);
lock(rfcomm_ioctl_mutex);
lock(sk_lock-AF_BLUETOOTH-BTPROTO_RFCOMM);

*** DEADLOCK ***

1 lock held by syz-executor.0/5603:
#0: ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_create_dev net/bluetooth/rfcomm/tty.c:487 [inline]
#0: ffffffff8e35cc88 (rfcomm_ioctl_mutex){+.+.}-{3:3}, at: rfcomm_dev_ioctl+0x8a2/0x1cb0 net/bluetooth/rfcomm/tty.c:590

stack backtrace:
CPU: 0 PID: 5603 Comm: syz-executor.0 Not tainted 6.3.0-rc4-next-20230330-syzkaller-dirty #0

__rfcomm_create_dev net/bluetooth/rfcomm/tty.c:417 [inline]
rfcomm_create_dev net/bluetooth/rfcomm/tty.c:488 [inline]
rfcomm_dev_ioctl+0x9ca/0x1cb0 net/bluetooth/rfcomm/tty.c:590

RIP: 0033:0x7f94d848c0f9

Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 19 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48

RSP: 002b:00007f94d911c168 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007f94d85ac050 RCX: 00007f94d848c0f9

RDX: 0000000020000100 RSI: 00000000400452c8 RDI: 0000000000000006

RBP: 00007f94d84e7b39 R08: 0000000000000000 R09: 0000000000000000

R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000

R13: 00007ffef88f28bf R14: 00007f94d911c300 R15: 0000000000022000

</TASK>

Tested on:

commit: a6d9e303 Add linux-next specific files for 20230330
git tree: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git

console output: https://syzkaller.appspot.com/x/log.txt?x=14b657d5c80000

patch: https://syzkaller.appspot.com/x/patch.diff?x=1741d1cdc80000

Hillf Danton

unread,

Mar 31, 2023, 4:49:19 AM3/31/23

to syzbot, linux-...@vger.kernel.org, syzkall...@googlegroups.com

On 30 Mar 2023 17:21:46 -0700

> HEAD commit: a6d9e3034536 Add linux-next specific files for 20230330
> git tree: linux-next

--- x/net/bluetooth/rfcomm/sock.c
+++ y/net/bluetooth/rfcomm/sock.c
@@ -876,9 +876,7 @@ static int rfcomm_sock_ioctl(struct sock

if (err == -ENOIOCTLCMD) {
#ifdef CONFIG_BT_RFCOMM_TTY
- lock_sock(sk);
err = rfcomm_dev_ioctl(sk, cmd, (void __user *) arg);
- release_sock(sk);
#else
err = -EOPNOTSUPP;
#endif
--- x/net/bluetooth/rfcomm/tty.c
+++ y/net/bluetooth/rfcomm/tty.c
@@ -403,9 +403,12 @@ static int __rfcomm_create_dev(struct so
return -EPERM;

if (req.flags & (1 << RFCOMM_REUSE_DLC)) {
+ lock_sock(sk);
/* Socket must be connected */
- if (sk->sk_state != BT_CONNECTED)
+ if (sk->sk_state != BT_CONNECTED) {
+ release_sock(sk);
return -EBADFD;
+ }

dlc = rfcomm_pi(sk)->dlc;
rfcomm_dlc_hold(dlc);

@@ -422,13 +425,13 @@ static int __rfcomm_create_dev(struct so
}

id = rfcomm_dev_add(&req, dlc);
- if (id < 0)
- return id;

if (req.flags & (1 << RFCOMM_REUSE_DLC)) {
/* DLC is now used by device.
* Socket must be disconnected */
- sk->sk_state = BT_CLOSED;
+ if (!(id < 0))
+ sk->sk_state = BT_CLOSED;
+ release_sock(sk);
}

return id;
--

syzbot

unread,

Mar 31, 2023, 5:18:27 AM3/31/23

to hda...@sina.com, linux-...@vger.kernel.org, syzkall...@googlegroups.com

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-and-tested-by: syzbot+b69a62...@syzkaller.appspotmail.com

Tested on:

commit: a6d9e303 Add linux-next specific files for 20230330
git tree: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git

console output: https://syzkaller.appspot.com/x/log.txt?x=134609cdc80000

patch: https://syzkaller.appspot.com/x/patch.diff?x=133a6835c80000

Note: testing is done by a robot and is best-effort only.

Reply all

Reply to author

Forward

0 new messages