[v5.15] WARNING: held lock freed in mptcp_worker

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: c79648372d02 Linux 5.15.189
git tree: linux-5.15.y
console output: https://syzkaller.appspot.com/x/log.txt?x=12fb85a2580000
kernel config: https://syzkaller.appspot.com/x/.config?x=f3b1215b3cf4119f
dashboard link: https://syzkaller.appspot.com/bug?extid=b4b7d8a00eb1891989c5
compiler: Debian clang version 20.1.7 (++20250616065708+6146a88f6049-1~exp1~20250616065826.132), Debian LLD 20.1.7

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

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/fa511d2a8893/disk-c7964837.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/f14b6c77bbb9/vmlinux-c7964837.xz
kernel image: https://storage.googleapis.com/syzbot-assets/388987ba15c6/bzImage-c7964837.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+b4b7d8...@syzkaller.appspotmail.com

=========================
WARNING: held lock freed!
5.15.189-syzkaller #0 Not tainted
-------------------------
kworker/1:4/4226 is freeing memory ffff88805a731800-ffff88805a73237f, with a lock still held there!
ffff88805a731920 (sk_lock-AF_INET6){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1694 [inline]
ffff88805a731920 (sk_lock-AF_INET6){+.+.}-{0:0}, at: mptcp_worker+0x31/0xc60 net/mptcp/protocol.c:2507
3 locks held by kworker/1:4/4226:
#0: ffff888016870938 ((wq_completion)events){+.+.}-{0:0}, at: process_one_work+0x760/0x1000 kernel/workqueue.c:-1
#1: ffffc9000350fd00 ((work_completion)(&msk->work)){+.+.}-{0:0}, at: process_one_work+0x7a3/0x1000 kernel/workqueue.c:2285
#2: ffff88805a731920 (sk_lock-AF_INET6){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1694 [inline]
#2: ffff88805a731920 (sk_lock-AF_INET6){+.+.}-{0:0}, at: mptcp_worker+0x31/0xc60 net/mptcp/protocol.c:2507

stack backtrace:
CPU: 1 PID: 4226 Comm: kworker/1:4 Not tainted 5.15.189-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/12/2025
Workqueue: events mptcp_worker
Call Trace:
<TASK>
dump_stack_lvl+0x168/0x230 lib/dump_stack.c:106
print_freed_lock_bug kernel/locking/lockdep.c:6401 [inline]
debug_check_no_locks_freed+0x2f2/0x3c0 kernel/locking/lockdep.c:6434
slab_free_hook mm/slub.c:1677 [inline]
slab_free_freelist_hook+0x76/0x170 mm/slub.c:1731
slab_free mm/slub.c:3499 [inline]
kmem_cache_free+0x8f/0x210 mm/slub.c:3515
sk_prot_free net/core/sock.c:1898 [inline]
__sk_destruct+0x54b/0x820 net/core/sock.c:1986
sock_put include/net/sock.h:1939 [inline]
__mptcp_destroy_sock+0x5c1/0x720 net/mptcp/protocol.c:2775
mptcp_worker+0x870/0xc60 net/mptcp/protocol.c:2533
process_one_work+0x863/0x1000 kernel/workqueue.c:2310
worker_thread+0xaa8/0x12a0 kernel/workqueue.c:2457
kthread+0x436/0x520 kernel/kthread.c:334
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:287
</TASK>
------------[ cut here ]------------
refcount_t: underflow; use-after-free.
WARNING: CPU: 1 PID: 4226 at lib/refcount.c:28 refcount_warn_saturate+0x11b/0x1a0 lib/refcount.c:28
Modules linked in:
CPU: 1 PID: 4226 Comm: kworker/1:4 Not tainted 5.15.189-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/12/2025
Workqueue: events mptcp_worker
RIP: 0010:refcount_warn_saturate+0x11b/0x1a0 lib/refcount.c:28
Code: 09 01 48 c7 c7 60 56 59 8a e8 c1 f1 c2 05 0f 0b eb c4 e8 e8 4d 9d fd c6 05 47 54 7b 09 01 48 c7 c7 c0 56 59 8a e8 a5 f1 c2 05 <0f> 0b eb a8 e8 cc 4d 9d fd c6 05 28 54 7b 09 01 48 c7 c7 00 56 59
RSP: 0018:ffffc9000350fc28 EFLAGS: 00010246

RAX: 424bfd1515c27000 RBX: 0000000000000003 RCX: ffff8880619c9dc0
RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000
RBP: ffffc9000350fdb8 R08: dffffc0000000000 R09: ffffed10172267a8
R10: ffffed10172267a8 R11: 1ffff110172267a7 R12: dffffc0000000000
R13: ffff88802333db18 R14: 0000000000000003 R15: ffff8880b913f008
FS: 0000000000000000(0000) GS:ffff8880b9100000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f5e3cabe2f8 CR3: 00000000630dd000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
process_one_work+0x863/0x1000 kernel/workqueue.c:2310
worker_thread+0xaa8/0x12a0 kernel/workqueue.c:2457
kthread+0x436/0x520 kernel/kthread.c:334
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:287
</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

[syzbot]

Auto-closing this bug as obsolete.
Crashes did not happen for a while, no reproducer and no activity.