syzbot


possible deadlock in set_blocksize

Status: upstream: reported on 2025/10/30 02:22
Reported-by: syzbot+b871b6319e7734bba0f9@syzkaller.appspotmail.com
First crash: 135d, last: 81d

Sample crash report:
loop3: detected capacity change from 0 to 16
erofs: DAX enabled. Warning: EXPERIMENTAL, use at your own risk
======================================================
WARNING: possible circular locking dependency detected
syzkaller #0 Not tainted
------------------------------------------------------
syz.3.1195/10398 is trying to acquire lock:
ffff88801fc47ab0 (&sb->s_type->i_mutex_key#8){++++}-{3:3}, at: inode_lock include/linux/fs.h:804 [inline]
ffff88801fc47ab0 (&sb->s_type->i_mutex_key#8){++++}-{3:3}, at: set_blocksize+0x201/0x4b0 block/bdev.c:160

but task is already holding lock:
ffff88807b4380e0 (&type->s_umount_key#72){++++}-{3:3}, at: __super_lock fs/super.c:56 [inline]
ffff88807b4380e0 (&type->s_umount_key#72){++++}-{3:3}, at: __super_lock_excl fs/super.c:71 [inline]
ffff88807b4380e0 (&type->s_umount_key#72){++++}-{3:3}, at: get_tree_bdev+0x344/0x510 fs/super.c:1589

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #4 (&type->s_umount_key#72){++++}-{3:3}:
       down_read+0x46/0x2e0 kernel/locking/rwsem.c:1520
       __super_lock fs/super.c:58 [inline]
       super_lock+0x167/0x360 fs/super.c:117
       super_lock_shared fs/super.c:146 [inline]
       super_lock_shared_active fs/super.c:1442 [inline]
       fs_bdev_mark_dead+0xad/0x1f0 fs/super.c:1458
       bdev_mark_dead+0x83/0x1a0 block/bdev.c:1030
       disk_force_media_change+0x132/0x1b0 block/disk-events.c:302
       __loop_clr_fd+0x396/0x660 drivers/block/loop.c:1170
       loop_clr_fd drivers/block/loop.c:1253 [inline]
       lo_ioctl+0x11e8/0x19f0 drivers/block/loop.c:1578
       blkdev_ioctl+0x58b/0x6f0 block/ioctl.c:639
       vfs_ioctl fs/ioctl.c:51 [inline]
       __do_sys_ioctl fs/ioctl.c:871 [inline]
       __se_sys_ioctl+0xfd/0x170 fs/ioctl.c:857
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:81
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #3 (&bdev->bd_holder_lock){+.+.}-{3:3}:
       __mutex_lock_common kernel/locking/mutex.c:603 [inline]
       __mutex_lock+0x129/0xcc0 kernel/locking/mutex.c:747
       bd_finish_claiming+0x22f/0x3f0 block/bdev.c:585
       blkdev_get_by_dev+0x45c/0x600 block/bdev.c:818
       bdev_open_by_dev+0x77/0x100 block/bdev.c:859
       setup_bdev_super+0x59/0x660 fs/super.c:1496
       mount_bdev+0x1dd/0x2d0 fs/super.c:1640
       legacy_get_tree+0xea/0x180 fs/fs_context.c:662
       vfs_get_tree+0x8c/0x280 fs/super.c:1764
       do_new_mount+0x24b/0xa40 fs/namespace.c:3386
       init_mount+0xd2/0x120 fs/init.c:25
       do_mount_root+0x97/0x230 init/do_mounts.c:166
       mount_root_generic+0x195/0x3c0 init/do_mounts.c:205
       prepare_namespace+0xc2/0x100 init/do_mounts.c:489
       kernel_init_freeable+0x413/0x570 init/main.c:1578
       kernel_init+0x1d/0x1c0 init/main.c:1455
       ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:152
       ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:293

-> #2 (bdev_lock){+.+.}-{3:3}:
       __mutex_lock_common kernel/locking/mutex.c:603 [inline]
       __mutex_lock+0x129/0xcc0 kernel/locking/mutex.c:747
       bd_prepare_to_claim+0x1ba/0x480 block/bdev.c:527
       truncate_bdev_range+0x4e/0x260 block/bdev.c:105
       blkdev_fallocate+0x50d/0x6a0 block/fops.c:798
       vfs_fallocate+0x58e/0x700 fs/open.c:324
       ksys_fallocate fs/open.c:347 [inline]
       __do_sys_fallocate fs/open.c:355 [inline]
       __se_sys_fallocate fs/open.c:353 [inline]
       __x64_sys_fallocate+0xc1/0x110 fs/open.c:353
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:81
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #1 (mapping.invalidate_lock){++++}-{3:3}:
       down_write+0x97/0x1f0 kernel/locking/rwsem.c:1573
       filemap_invalidate_lock include/linux/fs.h:849 [inline]
       set_blocksize+0x249/0x4b0 block/bdev.c:161
       sb_set_blocksize block/bdev.c:178 [inline]
       sb_min_blocksize+0xbe/0x190 block/bdev.c:194
       ext4_load_super fs/ext4/super.c:5042 [inline]
       __ext4_fill_super fs/ext4/super.c:5251 [inline]
       ext4_fill_super+0x6df/0x66c0 fs/ext4/super.c:5724
       get_tree_bdev+0x3e4/0x510 fs/super.c:1591
       vfs_get_tree+0x8c/0x280 fs/super.c:1764
       do_new_mount+0x24b/0xa40 fs/namespace.c:3386
       init_mount+0xd2/0x120 fs/init.c:25
       do_mount_root+0x97/0x230 init/do_mounts.c:166
       mount_root_generic+0x195/0x3c0 init/do_mounts.c:205
       prepare_namespace+0xc2/0x100 init/do_mounts.c:489
       kernel_init_freeable+0x413/0x570 init/main.c:1578
       kernel_init+0x1d/0x1c0 init/main.c:1455
       ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:152
       ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:293

-> #0 (&sb->s_type->i_mutex_key#8){++++}-{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+0x2ddb/0x7c80 kernel/locking/lockdep.c:5137
       lock_acquire+0x197/0x410 kernel/locking/lockdep.c:5754
       down_write+0x97/0x1f0 kernel/locking/rwsem.c:1573
       inode_lock include/linux/fs.h:804 [inline]
       set_blocksize+0x201/0x4b0 block/bdev.c:160
       sb_set_blocksize+0x4a/0x100 block/bdev.c:178
       erofs_fc_fill_super+0x12a2/0x1f50 fs/erofs/super.c:621
       get_tree_bdev+0x3e4/0x510 fs/super.c:1591
       vfs_get_tree+0x8c/0x280 fs/super.c:1764
       do_new_mount+0x24b/0xa40 fs/namespace.c:3386
       do_mount fs/namespace.c:3726 [inline]
       __do_sys_mount fs/namespace.c:3935 [inline]
       __se_sys_mount+0x2da/0x3c0 fs/namespace.c:3912
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:81
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

other info that might help us debug this:

Chain exists of:
  &sb->s_type->i_mutex_key#8 --> &bdev->bd_holder_lock --> &type->s_umount_key#72

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&type->s_umount_key#72);
                               lock(&bdev->bd_holder_lock);
                               lock(&type->s_umount_key#72);
  lock(&sb->s_type->i_mutex_key#8);

 *** DEADLOCK ***

1 lock held by syz.3.1195/10398:
 #0: ffff88807b4380e0 (&type->s_umount_key#72){++++}-{3:3}, at: __super_lock fs/super.c:56 [inline]
 #0: ffff88807b4380e0 (&type->s_umount_key#72){++++}-{3:3}, at: __super_lock_excl fs/super.c:71 [inline]
 #0: ffff88807b4380e0 (&type->s_umount_key#72){++++}-{3:3}, at: get_tree_bdev+0x344/0x510 fs/super.c:1589

stack backtrace:
CPU: 0 PID: 10398 Comm: syz.3.1195 Not tainted syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/25/2025
Call Trace:
 <TASK>
 dump_stack_lvl+0x16c/0x230 lib/dump_stack.c:106
 check_noncircular+0x2bd/0x3c0 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+0x2ddb/0x7c80 kernel/locking/lockdep.c:5137
 lock_acquire+0x197/0x410 kernel/locking/lockdep.c:5754
 down_write+0x97/0x1f0 kernel/locking/rwsem.c:1573
 inode_lock include/linux/fs.h:804 [inline]
 set_blocksize+0x201/0x4b0 block/bdev.c:160
 sb_set_blocksize+0x4a/0x100 block/bdev.c:178
 erofs_fc_fill_super+0x12a2/0x1f50 fs/erofs/super.c:621
 get_tree_bdev+0x3e4/0x510 fs/super.c:1591
 vfs_get_tree+0x8c/0x280 fs/super.c:1764
 do_new_mount+0x24b/0xa40 fs/namespace.c:3386
 do_mount fs/namespace.c:3726 [inline]
 __do_sys_mount fs/namespace.c:3935 [inline]
 __se_sys_mount+0x2da/0x3c0 fs/namespace.c:3912
 do_syscall_x64 arch/x86/entry/common.c:51 [inline]
 do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:81
 entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7fa193f90eea
Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb a6 e8 de 1a 00 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fa194e45e68 EFLAGS: 00000246 ORIG_RAX: 00000000000000a5
RAX: ffffffffffffffda RBX: 00007fa194e45ef0 RCX: 00007fa193f90eea
RDX: 0000200000000000 RSI: 0000200000000080 RDI: 00007fa194e45eb0
RBP: 0000200000000000 R08: 00007fa194e45ef0 R09: 000000000000000c
R10: 000000000000000c R11: 0000000000000246 R12: 0000200000000080
R13: 00007fa194e45eb0 R14: 00000000000001b1 R15: 0000200000000340
 </TASK>
erofs: DAX unsupported by block device. Turning off DAX.

Crashes (3):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2025/12/22 14:58 linux-6.6.y 5fa4793a2d2d d6526ea3 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in set_blocksize
2025/10/30 02:23 linux-6.6.y e5bbb12db2c7 fd2207e7 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in set_blocksize
2025/10/30 02:22 linux-6.6.y e5bbb12db2c7 fd2207e7 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan possible deadlock in set_blocksize
* Struck through repros no longer work on HEAD.