[v6.1] possible deadlock in unix_gc

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: dcbc050cb0d3 Linux 6.1.89
git tree: linux-6.1.y
console output: https://syzkaller.appspot.com/x/log.txt?x=1771bcf8980000
kernel config: https://syzkaller.appspot.com/x/.config?x=22f9bec33cc10172
dashboard link: https://syzkaller.appspot.com/bug?extid=b08b278d99eaf14f330d
compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40

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

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/62590cf0da60/disk-dcbc050c.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/f48fe2388a70/vmlinux-dcbc050c.xz
kernel image: https://storage.googleapis.com/syzbot-assets/f4eb42b21e90/bzImage-dcbc050c.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+b08b27...@syzkaller.appspotmail.com

======================================================
WARNING: possible circular locking dependency detected
6.1.89-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor.4/10657 is trying to acquire lock:
ffff88807bd31e80 (&u->lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline]
ffff88807bd31e80 (&u->lock){+.+.}-{2:2}, at: unix_gc+0x441/0x13d0 net/unix/garbage.c:263

but task is already holding lock:
ffffffff8e397ab8 (unix_gc_lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline]
ffffffff8e397ab8 (unix_gc_lock){+.+.}-{2:2}, at: unix_gc+0x115/0x13d0 net/unix/garbage.c:215

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #1 (unix_gc_lock){+.+.}-{2:2}:
lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
__raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
_raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:154
spin_lock include/linux/spinlock.h:351 [inline]
unix_notinflight+0x20a/0x380 net/unix/scm.c:73
unix_detach_fds net/unix/scm.c:136 [inline]
unix_destruct_scm+0x21d/0x350 net/unix/scm.c:147
skb_release_head_state+0xf9/0x230 net/core/skbuff.c:843
skb_release_all net/core/skbuff.c:854 [inline]
__kfree_skb net/core/skbuff.c:870 [inline]
kfree_skb_reason+0x13d/0x390 net/core/skbuff.c:893
kfree_skb include/linux/skbuff.h:1220 [inline]
manage_oob net/unix/af_unix.c:2684 [inline]
unix_stream_read_generic+0xf91/0x2250 net/unix/af_unix.c:2761
unix_stream_recvmsg+0x222/0x2b0 net/unix/af_unix.c:2956
____sys_recvmsg+0x285/0x530
___sys_recvmsg net/socket.c:2775 [inline]
__sys_recvmsg+0x2e9/0x3d0 net/socket.c:2805
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x3b/0xb0 arch/x86/entry/common.c:81
entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #0 (&u->lock){+.+.}-{2:2}:
check_prev_add kernel/locking/lockdep.c:3090 [inline]
check_prevs_add kernel/locking/lockdep.c:3209 [inline]
validate_chain+0x1661/0x5950 kernel/locking/lockdep.c:3825
__lock_acquire+0x125b/0x1f80 kernel/locking/lockdep.c:5049
lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
__raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
_raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:154
spin_lock include/linux/spinlock.h:351 [inline]
unix_gc+0x441/0x13d0 net/unix/garbage.c:263
unix_release_sock+0xab8/0xcf0 net/unix/af_unix.c:669
unix_release+0x88/0xc0 net/unix/af_unix.c:1039
__sock_release net/socket.c:654 [inline]
sock_close+0xcd/0x230 net/socket.c:1400
__fput+0x3b7/0x890 fs/file_table.c:320
task_work_run+0x246/0x300 kernel/task_work.c:179
resume_user_mode_work include/linux/resume_user_mode.h:49 [inline]
exit_to_user_mode_loop+0xde/0x100 kernel/entry/common.c:177
exit_to_user_mode_prepare+0xb1/0x140 kernel/entry/common.c:210
__syscall_exit_to_user_mode_work kernel/entry/common.c:292 [inline]
syscall_exit_to_user_mode+0x60/0x270 kernel/entry/common.c:303
do_syscall_64+0x47/0xb0 arch/x86/entry/common.c:87
entry_SYSCALL_64_after_hwframe+0x68/0xd2

other info that might help us debug this:

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(unix_gc_lock);
lock(&u->lock);
lock(unix_gc_lock);
lock(&u->lock);

*** DEADLOCK ***

2 locks held by syz-executor.4/10657:
#0: ffff888058bc6210 (&sb->s_type->i_mutex_key#10){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:758 [inline]
#0: ffff888058bc6210 (&sb->s_type->i_mutex_key#10){+.+.}-{3:3}, at: __sock_release net/socket.c:653 [inline]
#0: ffff888058bc6210 (&sb->s_type->i_mutex_key#10){+.+.}-{3:3}, at: sock_close+0x98/0x230 net/socket.c:1400
#1: ffffffff8e397ab8 (unix_gc_lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline]
#1: ffffffff8e397ab8 (unix_gc_lock){+.+.}-{2:2}, at: unix_gc+0x115/0x13d0 net/unix/garbage.c:215

stack backtrace:
CPU: 0 PID: 10657 Comm: syz-executor.4 Not tainted 6.1.89-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1e3/0x2cb lib/dump_stack.c:106
check_noncircular+0x2fa/0x3b0 kernel/locking/lockdep.c:2170
check_prev_add kernel/locking/lockdep.c:3090 [inline]
check_prevs_add kernel/locking/lockdep.c:3209 [inline]
validate_chain+0x1661/0x5950 kernel/locking/lockdep.c:3825
__lock_acquire+0x125b/0x1f80 kernel/locking/lockdep.c:5049
lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
__raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
_raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:154
spin_lock include/linux/spinlock.h:351 [inline]
unix_gc+0x441/0x13d0 net/unix/garbage.c:263
unix_release_sock+0xab8/0xcf0 net/unix/af_unix.c:669
unix_release+0x88/0xc0 net/unix/af_unix.c:1039
__sock_release net/socket.c:654 [inline]
sock_close+0xcd/0x230 net/socket.c:1400
__fput+0x3b7/0x890 fs/file_table.c:320
task_work_run+0x246/0x300 kernel/task_work.c:179
resume_user_mode_work include/linux/resume_user_mode.h:49 [inline]
exit_to_user_mode_loop+0xde/0x100 kernel/entry/common.c:177
exit_to_user_mode_prepare+0xb1/0x140 kernel/entry/common.c:210
__syscall_exit_to_user_mode_work kernel/entry/common.c:292 [inline]
syscall_exit_to_user_mode+0x60/0x270 kernel/entry/common.c:303
do_syscall_64+0x47/0xb0 arch/x86/entry/common.c:87
entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7f8c2967dea9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 e1 20 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 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f8c2a47c0c8 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4
RAX: 0000000000000000 RBX: 00007f8c297abf80 RCX: 00007f8c2967dea9
RDX: 0000000000000000 RSI: ffffffffffffffff RDI: 000000000000000a
RBP: 00007f8c296ca4a4 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 000000000000000b R14: 00007f8c297abf80 R15: 00007ffe0ac61468
</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.

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup