syzbot


possible deadlock in vfs_rename (2)

Status: upstream: reported on 2025/08/05 14:59
Reported-by: syzbot+3b3acb90faf05334e0b6@syzkaller.appspotmail.com
First crash: 20d, last: 20d
Similar bugs (5)
Kernel Title Rank 🛈 Repro Cause bisect Fix bisect Count Last Reported Patched Status
linux-5.15 possible deadlock in vfs_rename 4 13 161d 474d 0/3 auto-obsoleted due to no activity on 2025/06/25 12:29
upstream possible deadlock in vfs_rename kernel 4 1 929d 929d 0/29 closed as invalid on 2023/02/08 16:28
linux-6.1 possible deadlock in vfs_rename (2) origin:lts-only 4 C inconclusive 6 72d 237d 0/3 upstream: reported C repro on 2024/12/31 00:31
linux-6.1 possible deadlock in vfs_rename 4 8 343d 378d 0/3 auto-obsoleted due to no activity on 2024/12/25 19:29
upstream possible deadlock in vfs_rename (2) ntfs3 4 3 220d 268d 0/29 auto-obsoleted due to no activity on 2025/04/27 15:12

Sample crash report:
======================================================
WARNING: possible circular locking dependency detected
5.15.189-syzkaller #0 Not tainted
------------------------------------------------------
syz.1.92/4543 is trying to acquire lock:
ffff0000ea71c520 (&type->i_mutex_dir_key#9){++++}-{3:3}, at: inode_lock include/linux/fs.h:787 [inline]
ffff0000ea71c520 (&type->i_mutex_dir_key#9){++++}-{3:3}, at: vfs_rename+0x670/0xdcc fs/namei.c:4797

but task is already holding lock:
ffff0000ea71d260 (&type->i_mutex_dir_key#9/2){+.+.}-{3:3}, at: inode_lock_nested include/linux/fs.h:822 [inline]
ffff0000ea71d260 (&type->i_mutex_dir_key#9/2){+.+.}-{3:3}, at: vfs_rename+0x60c/0xdcc fs/namei.c:4795

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 (&type->i_mutex_dir_key#9/2){+.+.}-{3:3}:
       down_write_nested+0xc0/0x130 kernel/locking/rwsem.c:1667
       inode_lock_nested include/linux/fs.h:822 [inline]
       xattr_rmdir fs/reiserfs/xattr.c:106 [inline]
       delete_one_xattr+0xe8/0x2c8 fs/reiserfs/xattr.c:338
       reiserfs_for_each_xattr+0x568/0x68c fs/reiserfs/xattr.c:311
       reiserfs_delete_xattrs+0x2c/0xa4 fs/reiserfs/xattr.c:364
       reiserfs_evict_inode+0x1d8/0x3e8 fs/reiserfs/inode.c:53
       evict+0x3c8/0x810 fs/inode.c:647
       iput_final fs/inode.c:1769 [inline]
       iput+0x6c4/0x77c fs/inode.c:1795
       d_delete_notify include/linux/fsnotify.h:267 [inline]
       vfs_rmdir+0x2f4/0x3ec fs/namei.c:4162
       do_rmdir+0x240/0x634 fs/namei.c:4210
       __do_sys_unlinkat fs/namei.c:4390 [inline]
       __se_sys_unlinkat fs/namei.c:4384 [inline]
       __arm64_sys_unlinkat+0xcc/0xfc fs/namei.c:4384
       __invoke_syscall arch/arm64/kernel/syscall.c:38 [inline]
       invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:52
       el0_svc_common+0x138/0x258 arch/arm64/kernel/syscall.c:142
       do_el0_svc+0x58/0x14c arch/arm64/kernel/syscall.c:181
       el0_svc+0x78/0x1e0 arch/arm64/kernel/entry-common.c:608
       el0t_64_sync_handler+0xcc/0xe4 arch/arm64/kernel/entry-common.c:626
       el0t_64_sync+0x1a0/0x1a4 arch/arm64/kernel/entry.S:584

-> #1 (&type->i_mutex_dir_key#9/3){+.+.}-{3:3}:
       down_write_nested+0xc0/0x130 kernel/locking/rwsem.c:1667
       inode_lock_nested include/linux/fs.h:822 [inline]
       open_xa_root fs/reiserfs/xattr.c:127 [inline]
       open_xa_dir+0x118/0x57c fs/reiserfs/xattr.c:152
       lookup_and_delete_xattr fs/reiserfs/xattr.c:489 [inline]
       reiserfs_xattr_set_handle+0xf8/0xb34 fs/reiserfs/xattr.c:536
       __reiserfs_set_acl+0x428/0x5b0 fs/reiserfs/xattr_acl.c:276
       reiserfs_set_acl+0x3ac/0x540 fs/reiserfs/xattr_acl.c:50
       set_posix_acl fs/posix_acl.c:947 [inline]
       posix_acl_xattr_set+0x2cc/0x378 fs/posix_acl.c:966
       __vfs_removexattr+0x370/0x38c fs/xattr.c:470
       __vfs_removexattr_locked+0x1d0/0x208 fs/xattr.c:505
       vfs_removexattr+0x80/0x190 fs/xattr.c:527
       ovl_do_removexattr fs/overlayfs/overlayfs.h:231 [inline]
       ovl_workdir_create+0x2d8/0x578 fs/overlayfs/super.c:818
       ovl_make_workdir fs/overlayfs/super.c:1376 [inline]
       ovl_get_workdir+0x2fc/0xe8c fs/overlayfs/super.c:1526
       ovl_fill_super+0x139c/0x2294 fs/overlayfs/super.c:2082
       mount_nodev+0x68/0x104 fs/super.c:1446
       ovl_mount+0x3c/0x50 fs/overlayfs/super.c:2173
       legacy_get_tree+0xd4/0x16c fs/fs_context.c:611
       vfs_get_tree+0x90/0x274 fs/super.c:1530
       do_new_mount+0x228/0x810 fs/namespace.c:3014
       path_mount+0x5b4/0x1000 fs/namespace.c:3344
       do_mount fs/namespace.c:3357 [inline]
       __do_sys_mount fs/namespace.c:3565 [inline]
       __se_sys_mount fs/namespace.c:3542 [inline]
       __arm64_sys_mount+0x514/0x5e4 fs/namespace.c:3542
       __invoke_syscall arch/arm64/kernel/syscall.c:38 [inline]
       invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:52
       el0_svc_common+0x138/0x258 arch/arm64/kernel/syscall.c:142
       do_el0_svc+0x58/0x14c arch/arm64/kernel/syscall.c:181
       el0_svc+0x78/0x1e0 arch/arm64/kernel/entry-common.c:608
       el0t_64_sync_handler+0xcc/0xe4 arch/arm64/kernel/entry-common.c:626
       el0t_64_sync+0x1a0/0x1a4 arch/arm64/kernel/entry.S:584

-> #0 (&type->i_mutex_dir_key#9){++++}-{3:3}:
       check_prev_add kernel/locking/lockdep.c:3053 [inline]
       check_prevs_add kernel/locking/lockdep.c:3172 [inline]
       validate_chain kernel/locking/lockdep.c:3788 [inline]
       __lock_acquire+0x2928/0x651c kernel/locking/lockdep.c:5012
       lock_acquire+0x1f4/0x620 kernel/locking/lockdep.c:5623
       down_write+0xbc/0x12c kernel/locking/rwsem.c:1551
       inode_lock include/linux/fs.h:787 [inline]
       vfs_rename+0x670/0xdcc fs/namei.c:4797
       do_renameat2+0x74c/0xcdc fs/namei.c:4985
       __do_sys_renameat2 fs/namei.c:5018 [inline]
       __se_sys_renameat2 fs/namei.c:5015 [inline]
       __arm64_sys_renameat2+0xe0/0xfc fs/namei.c:5015
       __invoke_syscall arch/arm64/kernel/syscall.c:38 [inline]
       invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:52
       el0_svc_common+0x138/0x258 arch/arm64/kernel/syscall.c:142
       do_el0_svc+0x58/0x14c arch/arm64/kernel/syscall.c:181
       el0_svc+0x78/0x1e0 arch/arm64/kernel/entry-common.c:608
       el0t_64_sync_handler+0xcc/0xe4 arch/arm64/kernel/entry-common.c:626
       el0t_64_sync+0x1a0/0x1a4 arch/arm64/kernel/entry.S:584

other info that might help us debug this:

Chain exists of:
  &type->i_mutex_dir_key#9 --> &type->i_mutex_dir_key#9/3 --> &type->i_mutex_dir_key#9/2

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&type->i_mutex_dir_key#9/2);
                               lock(&type->i_mutex_dir_key#9/3);
                               lock(&type->i_mutex_dir_key#9/2);
  lock(&type->i_mutex_dir_key#9);

 *** DEADLOCK ***

5 locks held by syz.1.92/4543:
 #0: ffff0000c9b9c460 (sb_writers#13){.+.+}-{0:0}, at: mnt_want_write+0x44/0x9c fs/namespace.c:377
 #1: ffff0000c9b9c748 (&type->s_vfs_rename_key){+.+.}-{3:3}, at: lock_rename fs/namei.c:3016 [inline]
 #1: ffff0000c9b9c748 (&type->s_vfs_rename_key){+.+.}-{3:3}, at: do_renameat2+0x30c/0xcdc fs/namei.c:4924
 #2: ffff0000ea7196c0 (&type->i_mutex_dir_key#9/1){+.+.}-{3:3}, at: inode_lock_nested include/linux/fs.h:822 [inline]
 #2: ffff0000ea7196c0 (&type->i_mutex_dir_key#9/1){+.+.}-{3:3}, at: lock_two_directories fs/namei.c:-1 [inline]
 #2: ffff0000ea7196c0 (&type->i_mutex_dir_key#9/1){+.+.}-{3:3}, at: lock_rename fs/namei.c:3017 [inline]
 #2: ffff0000ea7196c0 (&type->i_mutex_dir_key#9/1){+.+.}-{3:3}, at: do_renameat2+0x388/0xcdc fs/namei.c:4924
 #3: ffff0000ea71cbc0 (&type->i_mutex_dir_key#9/5){+.+.}-{3:3}, at: lock_rename include/linux/fs.h:-1 [inline]
 #3: ffff0000ea71cbc0 (&type->i_mutex_dir_key#9/5){+.+.}-{3:3}, at: do_renameat2+0x3bc/0xcdc fs/namei.c:4924
 #4: ffff0000ea71d260 (&type->i_mutex_dir_key#9/2){+.+.}-{3:3}, at: inode_lock_nested include/linux/fs.h:822 [inline]
 #4: ffff0000ea71d260 (&type->i_mutex_dir_key#9/2){+.+.}-{3:3}, at: vfs_rename+0x60c/0xdcc fs/namei.c:4795

stack backtrace:
CPU: 0 PID: 4543 Comm: syz.1.92 Not tainted 5.15.189-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/12/2025
Call trace:
 dump_backtrace+0x0/0x43c arch/arm64/kernel/stacktrace.c:152
 show_stack+0x2c/0x3c arch/arm64/kernel/stacktrace.c:216
 __dump_stack+0x30/0x40 lib/dump_stack.c:88
 dump_stack_lvl+0xf8/0x160 lib/dump_stack.c:106
 dump_stack+0x1c/0x5c lib/dump_stack.c:113
 print_circular_bug+0x148/0x1b0 kernel/locking/lockdep.c:2011
 check_noncircular+0x240/0x2d4 kernel/locking/lockdep.c:2133
 check_prev_add kernel/locking/lockdep.c:3053 [inline]
 check_prevs_add kernel/locking/lockdep.c:3172 [inline]
 validate_chain kernel/locking/lockdep.c:3788 [inline]
 __lock_acquire+0x2928/0x651c kernel/locking/lockdep.c:5012
 lock_acquire+0x1f4/0x620 kernel/locking/lockdep.c:5623
 down_write+0xbc/0x12c kernel/locking/rwsem.c:1551
 inode_lock include/linux/fs.h:787 [inline]
 vfs_rename+0x670/0xdcc fs/namei.c:4797
 do_renameat2+0x74c/0xcdc fs/namei.c:4985
 __do_sys_renameat2 fs/namei.c:5018 [inline]
 __se_sys_renameat2 fs/namei.c:5015 [inline]
 __arm64_sys_renameat2+0xe0/0xfc fs/namei.c:5015
 __invoke_syscall arch/arm64/kernel/syscall.c:38 [inline]
 invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:52
 el0_svc_common+0x138/0x258 arch/arm64/kernel/syscall.c:142
 do_el0_svc+0x58/0x14c arch/arm64/kernel/syscall.c:181
 el0_svc+0x78/0x1e0 arch/arm64/kernel/entry-common.c:608
 el0t_64_sync_handler+0xcc/0xe4 arch/arm64/kernel/entry-common.c:626
 el0t_64_sync+0x1a0/0x1a4 arch/arm64/kernel/entry.S:584

Crashes (1):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2025/08/05 14:59 linux-5.15.y c79648372d02 37880f40 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-5-15-kasan-arm64 possible deadlock in vfs_rename
* Struck through repros no longer work on HEAD.