EXT4-fs (loop6): mounting ext2 file system using the ext4 subsystem
EXT4-fs (loop3): get root inode failed
EXT4-fs (loop3): mount failed
======================================================
WARNING: possible circular locking dependency detected
5.18.0-rc2-next-20220414 #1 Not tainted
------------------------------------------------------
kworker/0:0/15458 is trying to acquire lock:
ffff8880104fc0e0 (&type->s_umount_key#42){++++}-{3:3}, at: deactivate_super+0xa5/0xd0

but task is already holding lock:
ffff88804787fdb0 ((delayed_mntput_work).work){+.+.}-{0:0}, at: process_one_work+0x949/0x16a0

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 ((delayed_mntput_work).work){+.+.}-{0:0}:
       process_one_work+0x9a0/0x16a0
       worker_thread+0x637/0x1250
       kthread+0x2f2/0x3b0
       ret_from_fork+0x22/0x30

-> #1 ((wq_completion)events){+.+.}-{0:0}:
       flush_workqueue+0x144/0x11a0
       ext4_put_super+0x9d/0x1050
       generic_shutdown_super+0x14f/0x410
EXT4-fs (loop6): mounted filesystem without journal. Quota mode: none.
       kill_block_super+0x9d/0xf0
       deactivate_locked_super+0x99/0x160
       deactivate_super+0xad/0xd0
       cleanup_mnt+0x39a/0x510
       task_work_run+0xe2/0x1a0
       exit_to_user_mode_prepare+0x199/0x1a0
       syscall_exit_to_user_mode+0x19/0x40
       do_syscall_64+0x48/0x90
       entry_SYSCALL_64_after_hwframe+0x44/0xae

-> #0 (&type->s_umount_key#42){++++}-{3:3}:
       __lock_acquire+0x2c2f/0x6120
       lock_acquire+0x1a2/0x530
       down_write+0x90/0x150
       deactivate_super+0xa5/0xd0
       cleanup_mnt+0x39a/0x510
       delayed_mntput+0x62/0x80
       process_one_work+0xa1c/0x16a0
       worker_thread+0x637/0x1250
       kthread+0x2f2/0x3b0
       ret_from_fork+0x22/0x30

other info that might help us debug this:

Chain exists of:
  &type->s_umount_key#42 --> (wq_completion)events --> (delayed_mntput_work).work

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock((delayed_mntput_work).work);
                               lock((wq_completion)events);
                               lock((delayed_mntput_work).work);
  lock(&type->s_umount_key#42);

 *** DEADLOCK ***

2 locks held by kworker/0:0/15458:
 #0: ffff888007858d38 ((wq_completion)events){+.+.}-{0:0}, at: process_one_work+0x915/0x16a0
 #1: ffff88804787fdb0 ((delayed_mntput_work).work){+.+.}-{0:0}, at: process_one_work+0x949/0x16a0

stack backtrace:
CPU: 0 PID: 15458 Comm: kworker/0:0 Not tainted 5.18.0-rc2-next-20220414 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
Workqueue: events delayed_mntput
Call Trace:
 <TASK>
 dump_stack_lvl+0x8b/0xb3
 check_noncircular+0x25f/0x2e0
 __lock_acquire+0x2c2f/0x6120
 lock_acquire+0x1a2/0x530
 down_write+0x90/0x150
 deactivate_super+0xa5/0xd0
 cleanup_mnt+0x39a/0x510
 delayed_mntput+0x62/0x80
 process_one_work+0xa1c/0x16a0
 worker_thread+0x637/0x1250
 kthread+0x2f2/0x3b0
 ret_from_fork+0x22/0x30
 </TASK>
loop3: detected capacity change from 0 to 512
EXT4-fs error (device loop3): __ext4_fill_super:5311: inode #2: comm syz-executor.3: iget: root inode unallocated
EXT4-fs (loop3): get root inode failed
EXT4-fs (loop3): mount failed
loop3: detected capacity change from 0 to 512
EXT4-fs error (device loop3): __ext4_fill_super:5311: inode #2: comm syz-executor.3: iget: root inode unallocated
EXT4-fs (loop3): get root inode failed
EXT4-fs (loop3): mount failed