possible deadlock in configfs_depend_item_unlocked

Status: upstream: reported on 2026/07/25 12:59
Reported-by: syzbot+35c6202df82fb8c32975@syzkaller.appspotmail.com
First crash: 1d18h, last: 23h37m
Similar bugs (2)
Kernel Title Rank 🛈 Repro Cause bisect Fix bisect Count Last Reported Patched Status
linux-6.1 possible deadlock in configfs_depend_item_unlocked 4 6 1d19h 1d23h 0/3 upstream: reported on 2026/07/25 08:24
linux-5.15 possible deadlock in configfs_depend_item_unlocked 4 7 1d12h 2d00h 0/3 upstream: reported on 2026/07/25 06:55

Sample crash report:
db_root: cannot be changed because it's in use
======================================================
WARNING: possible circular locking dependency detected
syzkaller #0 Not tainted
------------------------------------------------------
syz.8.1274/12323 is trying to acquire lock:
ffff8880200ef910 (&sb->s_type->i_mutex_key#32){++++}-{3:3}, at: inode_lock include/linux/fs.h:804 [inline]
ffff8880200ef910 (&sb->s_type->i_mutex_key#32){++++}-{3:3}, at: configfs_depend_item_unlocked+0x157/0x430 fs/configfs/dir.c:1255

but task is already holding lock:
ffff888026498920 (&opts->dep_lock){+.+.}-{3:3}, at: usbg_make_tpg+0x148/0x580 drivers/usb/gadget/function/f_tcm.c:1333

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 (&opts->dep_lock){+.+.}-{3:3}:
       __mutex_lock_common kernel/locking/mutex.c:603 [inline]
       __mutex_lock+0x135/0xc80 kernel/locking/mutex.c:747
       tcm_set_name+0x2a/0x90 drivers/usb/gadget/function/f_tcm.c:2188
       function_make+0x199/0x350 drivers/usb/gadget/configfs.c:635
       configfs_mkdir+0x515/0x9e0 fs/configfs/dir.c:1360
       vfs_mkdir+0x2a5/0x470 fs/namei.c:4125
       do_mkdirat+0x1dc/0x450 fs/namei.c:4148
       __do_sys_mkdir fs/namei.c:4168 [inline]
       __se_sys_mkdir fs/namei.c:4166 [inline]
       __x64_sys_mkdir+0x6e/0x80 fs/namei.c:4166
       do_syscall_x64 arch/x86/entry/common.c:46 [inline]
       do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:76
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #1 (gadget_subsys.su_mutex){+.+.}-{3:3}:
       __mutex_lock_common kernel/locking/mutex.c:603 [inline]
       __mutex_lock+0x135/0xc80 kernel/locking/mutex.c:747
       uvcg_streaming_header_allow_link+0x71/0x4a0 drivers/usb/gadget/function/uvc_configfs.c:1747
       configfs_symlink+0x5a5/0xfe0 fs/configfs/symlink.c:202
       vfs_symlink+0x138/0x2b0 fs/namei.c:4485
       do_symlinkat+0x1ba/0x400 fs/namei.c:4511
       __do_sys_symlink fs/namei.c:4532 [inline]
       __se_sys_symlink fs/namei.c:4530 [inline]
       __x64_sys_symlink+0x7e/0x90 fs/namei.c:4530
       do_syscall_x64 arch/x86/entry/common.c:46 [inline]
       do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:76
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #0 (&sb->s_type->i_mutex_key#32){++++}-{3:3}:
       check_prev_add kernel/locking/lockdep.c:3134 [inline]
       check_prevs_add kernel/locking/lockdep.c:3253 [inline]
       validate_chain kernel/locking/lockdep.c:3869 [inline]
       __lock_acquire+0x37ef/0x7d80 kernel/locking/lockdep.c:5137
       lock_acquire+0x19e/0x420 kernel/locking/lockdep.c:5754
       down_write+0x97/0x200 kernel/locking/rwsem.c:1573
       inode_lock include/linux/fs.h:804 [inline]
       configfs_depend_item_unlocked+0x157/0x430 fs/configfs/dir.c:1255
       usbg_make_tpg+0x1f7/0x580 drivers/usb/gadget/function/f_tcm.c:1341
       target_fabric_make_tpg+0xa4/0x6d0 drivers/target/target_core_fabric_configfs.c:939
       configfs_mkdir+0x515/0x9e0 fs/configfs/dir.c:1360
       vfs_mkdir+0x2a5/0x470 fs/namei.c:4125
       do_mkdirat+0x1dc/0x450 fs/namei.c:4148
       __do_sys_mkdir fs/namei.c:4168 [inline]
       __se_sys_mkdir fs/namei.c:4166 [inline]
       __x64_sys_mkdir+0x6e/0x80 fs/namei.c:4166
       do_syscall_x64 arch/x86/entry/common.c:46 [inline]
       do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:76
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

other info that might help us debug this:

Chain exists of:
  &sb->s_type->i_mutex_key#32 --> gadget_subsys.su_mutex --> &opts->dep_lock

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&opts->dep_lock);
                               lock(gadget_subsys.su_mutex);
                               lock(&opts->dep_lock);
  lock(&sb->s_type->i_mutex_key#32);

 *** DEADLOCK ***

5 locks held by syz.8.1274/12323:
 #0: ffff88801eae6418 (sb_writers#26){.+.+}-{0:0}, at: mnt_want_write+0x41/0x90 fs/namespace.c:412
 #1: ffff88805464aef0 (&sb->s_type->i_mutex_key#7/1){+.+.}-{3:3}, at: inode_lock_nested include/linux/fs.h:839 [inline]
 #1: ffff88805464aef0 (&sb->s_type->i_mutex_key#7/1){+.+.}-{3:3}, at: filename_create+0x20f/0x450 fs/namei.c:3894
 #2: ffffffff8db2b450 (&subsys->su_mutex){+.+.}-{3:3}, at: configfs_mkdir+0x4b9/0x9e0 fs/configfs/dir.c:1358
 #3: ffffffff8de58aa8 (tpg_instances_lock){+.+.}-{3:3}, at: usbg_make_tpg+0xf8/0x580 drivers/usb/gadget/function/f_tcm.c:1324
 #4: ffff888026498920 (&opts->dep_lock){+.+.}-{3:3}, at: usbg_make_tpg+0x148/0x580 drivers/usb/gadget/function/f_tcm.c:1333

stack backtrace:
CPU: 1 PID: 12323 Comm: syz.8.1274 Not tainted syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/16/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0x18c/0x250 lib/dump_stack.c:106
 check_noncircular+0x2fc/0x400 kernel/locking/lockdep.c:2187
 check_prev_add kernel/locking/lockdep.c:3134 [inline]
 check_prevs_add kernel/locking/lockdep.c:3253 [inline]
 validate_chain kernel/locking/lockdep.c:3869 [inline]
 __lock_acquire+0x37ef/0x7d80 kernel/locking/lockdep.c:5137
 lock_acquire+0x19e/0x420 kernel/locking/lockdep.c:5754
 down_write+0x97/0x200 kernel/locking/rwsem.c:1573
 inode_lock include/linux/fs.h:804 [inline]
 configfs_depend_item_unlocked+0x157/0x430 fs/configfs/dir.c:1255
 usbg_make_tpg+0x1f7/0x580 drivers/usb/gadget/function/f_tcm.c:1341
 target_fabric_make_tpg+0xa4/0x6d0 drivers/target/target_core_fabric_configfs.c:939
 configfs_mkdir+0x515/0x9e0 fs/configfs/dir.c:1360
 vfs_mkdir+0x2a5/0x470 fs/namei.c:4125
 do_mkdirat+0x1dc/0x450 fs/namei.c:4148
 __do_sys_mkdir fs/namei.c:4168 [inline]
 __se_sys_mkdir fs/namei.c:4166 [inline]
 __x64_sys_mkdir+0x6e/0x80 fs/namei.c:4166
 do_syscall_x64 arch/x86/entry/common.c:46 [inline]
 do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:76
 entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7fc86279de99
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 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 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fc86375c028 EFLAGS: 00000246 ORIG_RAX: 0000000000000053
RAX: ffffffffffffffda RBX: 00007fc862a25fa0 RCX: 00007fc86279de99
RDX: 0000000000000000 RSI: 00000000000001ff RDI: 0000200000000c00
RBP: 00007fc862833eaf R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007fc862a26038 R14: 00007fc862a25fa0 R15: 00007ffca3d8e358
 </TASK>
Rounding down aligned max_sectors from 4294967295 to 4294967288

Crashes (5):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2026/07/26 08:05 linux-6.6.y c5596480c50e 492bab15 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in configfs_depend_item_unlocked
2026/07/26 02:25 linux-6.6.y c5596480c50e 492bab15 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in configfs_depend_item_unlocked
2026/07/26 00:29 linux-6.6.y c5596480c50e 492bab15 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in configfs_depend_item_unlocked
2026/07/25 12:58 linux-6.6.y c5596480c50e 492bab15 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in configfs_depend_item_unlocked
2026/07/25 12:58 linux-6.6.y c5596480c50e 492bab15 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in configfs_depend_item_unlocked
* Struck through repros no longer work on HEAD.