Re: [syzbot] [ocfs2?] possible deadlock in shrinker_free

From: syzbot

Date: Sat Nov 21 2026 - 21:47:47 EST


syzbot has found a reproducer for the following issue on:

HEAD commit: f802b6558eb8 Merge tag 'xfs-fixes-7.4-rc4' of git://git.ke..
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
console output: https://syzkaller.appspot.com/x/log.txt?x=141d98e9580000
kernel config: https://syzkaller.appspot.com/x/.config?x=3cc6a7ee2d24d208
dashboard link: https://syzkaller.appspot.com/bug?extid=38fdbb551940270021f1
compiler: Debian clang version 22.1.8 (1~deb13u4), Debian LLD 22.1.8
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10918a03580000

The issue was bisected by AI to:

commit f277230d002d03ed98b814b735f9d3d01b42020a
Author: Cen Zhang <zzzccc427@xxxxxxxxx>
Date: Fri Oct 30 04:48:56 2026 +0000

ocfs2: serialize inode journal access with shutdown

AI reasoning: https://syzkaller.appspot.com/ai_job?id=0e68af1a-9a2b-4aad-b8ed-125ab8c4ed0a

AI-generated root cause summary:
ocfs2_journal_shutdown() holds down_write(&osb->journal_shutdown_sem) across
jbd2_journal_destroy(), which invokes shrinker_free() and acquires
shrinker_mutex, establishing &osb->journal_shutdown_sem -> shrinker_mutex.
Meanwhile, shrinker_mutex is held across GFP_KERNEL allocations (e.g.,
alloc_shrinker_info()), establishing shrinker_mutex -> fs_reclaim, and memory
reclaim can reach ocfs2_evict_inode() / ocfs2_clear_inode() via
super_cache_scan() and acquire down_read(&osb->journal_shutdown_sem),
establishing fs_reclaim -> &osb->journal_shutdown_sem. Lockdep reports this
3-lock cycle as a circular locking dependency because it does not track the
mutual exclusion provided by sb->s_umount via super_trylock_shared().

full analysis: https://syzkaller.appspot.com/x/ai_root_cause?x=38fdbb551940270021f1

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+38fdbb551940270021f1@xxxxxxxxxxxxxxxxxxxxxxxxx

The cause commit was identified by AI and may be wrong.
If you agree with it, please also add:
Fixes: f277230d002d ("ocfs2: serialize inode journal access with shutdown")

======================================================
WARNING: possible circular locking dependency detected
syzkaller #0 Not tainted
------------------------------------------------------
syz-executor875/5689 is trying to acquire lock:
ffffffff902617a0 (shrinker_mutex){+.+.}-{4:4}, at: shrinker_free+0xdb/0x3b0 mm/shrinker.c:809

but task is already holding lock:
ffff8880421282f8 (&osb->journal_shutdown_sem){++++}-{4:4}, at: ocfs2_journal_shutdown+0x4ec/0xb40 fs/ocfs2/journal.c:1122

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 (&osb->journal_shutdown_sem){++++}-{4:4}:
down_read+0x45/0x320 kernel/locking/rwsem.c:1574
ocfs2_clear_inode fs/ocfs2/inode.c:1326 [inline]
ocfs2_evict_inode+0x2dd8/0x43c0 fs/ocfs2/inode.c:1343
evict+0x6bf/0xc20 fs/inode.c:829
ocfs2_dentry_iput+0x24d/0x390 fs/ocfs2/dcache.c:407
dentry_kill+0x1b9/0x910 fs/dcache.c:857
shrink_kill fs/dcache.c:1317 [inline]
shrink_dentry_list+0x209/0x310 fs/dcache.c:1334
prune_dcache_sb+0x119/0x180 fs/dcache.c:1416
super_cache_scan+0x362/0x4a0 fs/super.c:225
do_shrink_slab+0x6d0/0x10d0 mm/shrinker.c:452
shrink_slab_memcg mm/shrinker.c:569 [inline]
shrink_slab+0x838/0x1210 mm/shrinker.c:647
shrink_one+0x396/0x10f0 mm/vmscan.c:5320
shrink_many mm/vmscan.c:5381 [inline]
lru_gen_shrink_node mm/vmscan.c:5459 [inline]
shrink_node+0x631e/0x6c50 mm/vmscan.c:6465
__node_reclaim+0x2f4/0xa10 mm/vmscan.c:8012
user_proactive_reclaim+0x699/0x8d0 mm/vmscan.c:8202
reclaim_store+0x69/0xb0 mm/vmscan.c:8268
kernfs_fop_write_iter+0x3a4/0x540 fs/kernfs/file.c:345
new_sync_write fs/read_write.c:575 [inline]
vfs_write+0x612/0xba0 fs/read_write.c:667
ksys_write+0x150/0x270 fs/read_write.c:719
do_syscall_x64 arch/x86/entry/syscall_64.c:61 [inline]
do_syscall_64+0x171/0x540 arch/x86/entry/syscall_64.c:89
entry_SYSCALL_64_after_hwframe+0x77/0x7f

-> #1 (fs_reclaim){+.+.}-{0:0}:
__fs_reclaim_acquire mm/page_alloc.c:4388 [inline]
fs_reclaim_acquire+0x71/0x100 mm/page_alloc.c:4402
might_alloc include/linux/sched/mm.h:316 [inline]
slab_pre_alloc_hook mm/slub.c:4673 [inline]
slab_alloc_node mm/slub.c:5011 [inline]
__do_kmalloc_node mm/slub.c:5530 [inline]
__kvmalloc_node_noprof+0x134/0x820 mm/slub.c:7143
alloc_shrinker_info+0xe9/0x400 mm/shrinker.c:101
mem_cgroup_css_online+0x84/0x7b0 mm/memcontrol.c:4396
online_css+0xbf/0x310 kernel/cgroup/cgroup.c:5903
css_create kernel/cgroup/cgroup.c:5983 [inline]
cgroup_apply_control_enable+0x9ed/0xda0 kernel/cgroup/cgroup.c:3474
cgroup_mkdir+0x1220/0x14b0 kernel/cgroup/cgroup.c:6188
kernfs_iop_mkdir+0x213/0x360 fs/kernfs/dir.c:1304
vfs_mkdir+0x411/0x660 fs/namei.c:5428
filename_mkdirat+0x284/0x550 fs/namei.c:5461
__do_sys_mkdirat fs/namei.c:5482 [inline]
__se_sys_mkdirat+0x35/0x150 fs/namei.c:5479
do_syscall_x64 arch/x86/entry/syscall_64.c:61 [inline]
do_syscall_64+0x171/0x540 arch/x86/entry/syscall_64.c:89
entry_SYSCALL_64_after_hwframe+0x77/0x7f

-> #0 (shrinker_mutex){+.+.}-{4:4}:
check_prev_add kernel/locking/lockdep.c:3209 [inline]
check_prevs_add kernel/locking/lockdep.c:3328 [inline]
validate_chain kernel/locking/lockdep.c:3952 [inline]
__lock_acquire+0x164c/0x2de0 kernel/locking/lockdep.c:5288
lock_acquire+0x115/0x350 kernel/locking/lockdep.c:5942
__mutex_lock_common kernel/locking/mutex.c:646 [inline]
__mutex_lock+0x197/0x15a0 kernel/locking/mutex.c:821
shrinker_free+0xdb/0x3b0 mm/shrinker.c:809
jbd2_journal_destroy+0x606/0x7e0 fs/jbd2/journal.c:2178
ocfs2_journal_shutdown+0x517/0xb40 fs/ocfs2/journal.c:1125
ocfs2_dismount_volume+0x3c9/0x860 fs/ocfs2/super.c:1874
generic_shutdown_super+0x140/0x2d0 fs/super.c:723
kill_block_super+0x44/0xa0 fs/super.c:1852
deactivate_locked_super+0xbe/0x110 fs/super.c:558
cleanup_mnt+0x204/0x2a0 fs/namespace.c:1396
task_work_run+0x1d9/0x260 kernel/task_work.c:233
resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
__exit_to_user_mode_loop kernel/entry/common.c:70 [inline]
exit_to_user_mode_loop+0x204/0x770 kernel/entry/common.c:101
__exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline]
syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:230 [inline]
syscall_exit_to_user_mode include/linux/entry-common.h:343 [inline]
do_syscall_64+0x336/0x540 arch/x86/entry/syscall_64.c:107
entry_SYSCALL_64_after_hwframe+0x77/0x7f

other info that might help us debug this:

Chain exists of:
shrinker_mutex --> fs_reclaim --> &osb->journal_shutdown_sem

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(&osb->journal_shutdown_sem);
lock(fs_reclaim);
lock(&osb->journal_shutdown_sem);
lock(shrinker_mutex);

*** DEADLOCK ***

locks held by syz-executor875/5689: 2, on CPU#0:
#0: ffff88804f39a100 (&type->s_umount_key#56){++++}-{4:4}, at: __super_lock fs/super.c:60 [inline]
#0: ffff88804f39a100 (&type->s_umount_key#56){++++}-{4:4}, at: __super_lock_excl fs/super.c:75 [inline]
#0: ffff88804f39a100 (&type->s_umount_key#56){++++}-{4:4}, at: deactivate_super+0xac/0xe0 fs/super.c:585
#1: ffff8880421282f8 (&osb->journal_shutdown_sem){++++}-{4:4}, at: ocfs2_journal_shutdown+0x4ec/0xb40 fs/ocfs2/journal.c:1122

stack backtrace:
CPU: 0 UID: 0 PID: 5689 Comm: syz-executor875 Not tainted syzkaller #0 PREEMPT(full)
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
print_circular_bug+0x2e2/0x300 kernel/locking/lockdep.c:2087
check_noncircular+0x12f/0x150 kernel/locking/lockdep.c:2219
check_prev_add kernel/locking/lockdep.c:3209 [inline]
check_prevs_add kernel/locking/lockdep.c:3328 [inline]
validate_chain kernel/locking/lockdep.c:3952 [inline]
__lock_acquire+0x164c/0x2de0 kernel/locking/lockdep.c:5288
lock_acquire+0x115/0x350 kernel/locking/lockdep.c:5942
__mutex_lock_common kernel/locking/mutex.c:646 [inline]
__mutex_lock+0x197/0x15a0 kernel/locking/mutex.c:821
shrinker_free+0xdb/0x3b0 mm/shrinker.c:809
jbd2_journal_destroy+0x606/0x7e0 fs/jbd2/journal.c:2178
ocfs2_journal_shutdown+0x517/0xb40 fs/ocfs2/journal.c:1125
ocfs2_dismount_volume+0x3c9/0x860 fs/ocfs2/super.c:1874
generic_shutdown_super+0x140/0x2d0 fs/super.c:723
kill_block_super+0x44/0xa0 fs/super.c:1852
deactivate_locked_super+0xbe/0x110 fs/super.c:558
cleanup_mnt+0x204/0x2a0 fs/namespace.c:1396
task_work_run+0x1d9/0x260 kernel/task_work.c:233
resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
__exit_to_user_mode_loop kernel/entry/common.c:70 [inline]
exit_to_user_mode_loop+0x204/0x770 kernel/entry/common.c:101
__exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline]
syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:230 [inline]
syscall_exit_to_user_mode include/linux/entry-common.h:343 [inline]
do_syscall_64+0x336/0x540 arch/x86/entry/syscall_64.c:107
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7ff9dea5b657
Code: a2 c7 05 24 da 09 00 00 00 00 00 eb 96 e8 51 09 00 00 90 31 f6 e9 09 00 00 00 66 0f 1f 84 00 00 00 00 00 b8 a6 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 01 c3 48 c7 c2 d8 ff ff ff f7 d8 64 89 02 b8
RSP: 002b:00007fff91194d68 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6
RAX: 0000000000000000 RBX: 00007fff91195610 RCX: 00007ff9dea5b657
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 00007ff9deac527a
RBP: 00007fff91195210 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 000000000f1aea66
R13: 00007ff9deac527a R14: 00007ff9deac52fe R15: 00005555828d9760
</TASK>
ocfs2: Unmounting device (7,0) on (node local)