syzbot


possible deadlock in ext4_move_extents (3)

Status: upstream: reported on 2024/10/24 22:08
Reported-by: syzbot+4f2662aee9760cf70151@syzkaller.appspotmail.com
First crash: 27d, last: 27d
Similar bugs (6)
Kernel Title Repro Cause bisect Fix bisect Count Last Reported Patched Status
linux-6.1 possible deadlock in ext4_move_extents 2 476d 565d 0/3 auto-obsoleted due to no activity on 2023/11/11 09:48
upstream possible deadlock in ext4_move_extents ext4 133 576d 747d 0/28 auto-obsoleted due to no activity on 2023/08/22 15:17
linux-6.1 possible deadlock in ext4_move_extents (2) 6 164d 343d 0/3 auto-obsoleted due to no activity on 2024/09/18 11:30
linux-5.15 possible deadlock in ext4_move_extents 3 501d 577d 0/3 auto-obsoleted due to no activity on 2023/10/16 22:21
linux-5.15 possible deadlock in ext4_move_extents (2) 2 304d 373d 0/3 auto-obsoleted due to no activity on 2024/05/01 17:40
linux-5.15 possible deadlock in ext4_move_extents (3) 5 48d 176d 0/3 upstream: reported on 2024/05/28 19:59

Sample crash report:
ext4 filesystem being mounted at /77/file0aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa supports timestamps until 2038-01-19 (0x7fffffff)
======================================================
WARNING: possible circular locking dependency detected
6.1.114-syzkaller #0 Not tainted
------------------------------------------------------
syz.0.4614/18279 is trying to acquire lock:
ffff888071f65cd0 (&ei->i_data_sem/1){+.+.}-{3:3}, at: ext4_move_extents+0x3a0/0xe10 fs/ext4/move_extent.c:615

but task is already holding lock:
ffff888071f62aa8 (&ei->i_data_sem/2){++++}-{3:3}, at: ext4_double_down_write_data_sem+0x28/0x40

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #3 (&ei->i_data_sem/2){++++}-{3:3}:
       lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
       down_read+0xad/0xa30 kernel/locking/rwsem.c:1520
       ext4_map_blocks+0x379/0x1cb0 fs/ext4/inode.c:603
       ext4_getblk+0x1eb/0x7c0 fs/ext4/inode.c:889
       ext4_bread+0x2a/0x170 fs/ext4/inode.c:945
       ext4_quota_write+0x21e/0x570 fs/ext4/super.c:7171
       write_blk fs/quota/quota_tree.c:70 [inline]
       get_free_dqblk+0x340/0x6c0 fs/quota/quota_tree.c:136
       do_insert_tree+0x298/0x1440 fs/quota/quota_tree.c:347
       do_insert_tree+0xb94/0x1440 fs/quota/quota_tree.c:402
       do_insert_tree+0xb94/0x1440 fs/quota/quota_tree.c:402
       do_insert_tree+0xb94/0x1440 fs/quota/quota_tree.c:402
       dq_insert_tree fs/quota/quota_tree.c:432 [inline]
       qtree_write_dquot+0x483/0x5a0 fs/quota/quota_tree.c:451
       v2_write_dquot+0x11c/0x190 fs/quota/quota_v2.c:361
       dquot_acquire+0x34d/0x680 fs/quota/dquot.c:472
       ext4_acquire_dquot+0x2eb/0x4a0 fs/ext4/super.c:6797
       dqget+0x762/0xe90 fs/quota/dquot.c:986
       __dquot_initialize+0x45e/0xea0 fs/quota/dquot.c:1515
       ext4_rename fs/ext4/namei.c:3881 [inline]
       ext4_rename2+0xcb4/0x4420 fs/ext4/namei.c:4258
       vfs_rename+0xd32/0x10f0 fs/namei.c:4874
       do_renameat2+0xde0/0x1440 fs/namei.c:5027
       __do_sys_rename fs/namei.c:5073 [inline]
       __se_sys_rename fs/namei.c:5071 [inline]
       __x64_sys_rename+0x82/0x90 fs/namei.c:5071
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x3b/0xb0 arch/x86/entry/common.c:81
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #2 (&s->s_dquot.dqio_sem){++++}-{3:3}:
       lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
       down_read+0xad/0xa30 kernel/locking/rwsem.c:1520
       v2_read_dquot+0x4a/0x100 fs/quota/quota_v2.c:335
       dquot_acquire+0x188/0x680 fs/quota/dquot.c:463
       ext4_acquire_dquot+0x2eb/0x4a0 fs/ext4/super.c:6797
       dqget+0x762/0xe90 fs/quota/dquot.c:986
       __dquot_initialize+0x2d9/0xea0 fs/quota/dquot.c:1515
       ext4_rename fs/ext4/namei.c:3868 [inline]
       ext4_rename2+0x957/0x4420 fs/ext4/namei.c:4258
       vfs_rename+0xd32/0x10f0 fs/namei.c:4874
       do_renameat2+0xde0/0x1440 fs/namei.c:5027
       __do_sys_rename fs/namei.c:5073 [inline]
       __se_sys_rename fs/namei.c:5071 [inline]
       __x64_sys_rename+0x82/0x90 fs/namei.c:5071
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x3b/0xb0 arch/x86/entry/common.c:81
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #1 (&dquot->dq_lock){+.+.}-{3:3}:
       lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
       __mutex_lock_common kernel/locking/mutex.c:603 [inline]
       __mutex_lock+0x132/0xd80 kernel/locking/mutex.c:747
       dquot_commit+0x57/0x510 fs/quota/dquot.c:507
       ext4_write_dquot+0x1fd/0x360 fs/ext4/super.c:6777
       mark_dquot_dirty fs/quota/dquot.c:372 [inline]
       mark_all_dquot_dirty+0xf7/0x400 fs/quota/dquot.c:412
       __dquot_alloc_space+0xb68/0x10e0 fs/quota/dquot.c:1750
       dquot_alloc_space_nodirty include/linux/quotaops.h:300 [inline]
       dquot_alloc_space include/linux/quotaops.h:313 [inline]
       dquot_alloc_block include/linux/quotaops.h:337 [inline]
       ext4_mb_new_blocks+0x1302/0x4b80 fs/ext4/mballoc.c:5727
       ext4_new_meta_blocks+0x206/0x4f0 fs/ext4/balloc.c:739
       ext4_ext_grow_indepth fs/ext4/extents.c:1331 [inline]
       ext4_ext_create_new_leaf fs/ext4/extents.c:1437 [inline]
       ext4_ext_insert_extent+0xf4e/0x4e40 fs/ext4/extents.c:2105
       ext4_split_extent_at+0x825/0x11b0 fs/ext4/extents.c:3235
       ext4_force_split_extent_at fs/ext4/extents.c:338 [inline]
       ext4_swap_extents+0x12fb/0x2280 fs/ext4/extents.c:5770
       move_extent_per_page+0x24cb/0x4c30 fs/ext4/move_extent.c:367
       ext4_move_extents+0x9ab/0xe10 fs/ext4/move_extent.c:678
       __ext4_ioctl fs/ext4/ioctl.c:1358 [inline]
       ext4_ioctl+0x3a8d/0x5f60 fs/ext4/ioctl.c:1614
       vfs_ioctl fs/ioctl.c:51 [inline]
       __do_sys_ioctl fs/ioctl.c:870 [inline]
       __se_sys_ioctl+0xf1/0x160 fs/ioctl.c:856
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x3b/0xb0 arch/x86/entry/common.c:81
       entry_SYSCALL_64_after_hwframe+0x68/0xd2

-> #0 (&ei->i_data_sem/1){+.+.}-{3:3}:
       check_prev_add kernel/locking/lockdep.c:3090 [inline]
       check_prevs_add kernel/locking/lockdep.c:3209 [inline]
       validate_chain+0x1661/0x5950 kernel/locking/lockdep.c:3825
       __lock_acquire+0x125b/0x1f80 kernel/locking/lockdep.c:5049
       lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
       down_write_nested+0x39/0x60 kernel/locking/rwsem.c:1689
       ext4_move_extents+0x3a0/0xe10 fs/ext4/move_extent.c:615
       __ext4_ioctl fs/ext4/ioctl.c:1358 [inline]
       ext4_ioctl+0x3a8d/0x5f60 fs/ext4/ioctl.c:1614
       vfs_ioctl fs/ioctl.c:51 [inline]
       __do_sys_ioctl fs/ioctl.c:870 [inline]
       __se_sys_ioctl+0xf1/0x160 fs/ioctl.c:856
       do_syscall_x64 arch/x86/entry/common.c:51 [inline]
       do_syscall_64+0x3b/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:
  &ei->i_data_sem/1 --> &s->s_dquot.dqio_sem --> &ei->i_data_sem/2

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&ei->i_data_sem/2);
                               lock(&s->s_dquot.dqio_sem);
                               lock(&ei->i_data_sem/2);
  lock(&ei->i_data_sem/1);

 *** DEADLOCK ***

4 locks held by syz.0.4614/18279:
 #0: ffff88807c010460 (sb_writers#4){++++}-{0:0}, at: mnt_want_write_file+0x5a/0x1f0 fs/namespace.c:437
 #1: ffff888071f62c20 (&sb->s_type->i_mutex_key#8){++++}-{3:3}, at: inode_lock include/linux/fs.h:758 [inline]
 #1: ffff888071f62c20 (&sb->s_type->i_mutex_key#8){++++}-{3:3}, at: lock_two_nondirectories+0xde/0x130 fs/inode.c:1204
 #2: ffff888071f65e48 (&sb->s_type->i_mutex_key#8/4){+.+.}-{3:3}, at: ext4_move_extents+0x385/0xe10 fs/ext4/move_extent.c:608
 #3: ffff888071f62aa8 (&ei->i_data_sem/2){++++}-{3:3}, at: ext4_double_down_write_data_sem+0x28/0x40

stack backtrace:
CPU: 0 PID: 18279 Comm: syz.0.4614 Not tainted 6.1.114-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x1e3/0x2cb lib/dump_stack.c:106
 check_noncircular+0x2fa/0x3b0 kernel/locking/lockdep.c:2170
 check_prev_add kernel/locking/lockdep.c:3090 [inline]
 check_prevs_add kernel/locking/lockdep.c:3209 [inline]
 validate_chain+0x1661/0x5950 kernel/locking/lockdep.c:3825
 __lock_acquire+0x125b/0x1f80 kernel/locking/lockdep.c:5049
 lock_acquire+0x1f8/0x5a0 kernel/locking/lockdep.c:5662
 down_write_nested+0x39/0x60 kernel/locking/rwsem.c:1689
 ext4_move_extents+0x3a0/0xe10 fs/ext4/move_extent.c:615
 __ext4_ioctl fs/ext4/ioctl.c:1358 [inline]
 ext4_ioctl+0x3a8d/0x5f60 fs/ext4/ioctl.c:1614
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:870 [inline]
 __se_sys_ioctl+0xf1/0x160 fs/ioctl.c:856
 do_syscall_x64 arch/x86/entry/common.c:51 [inline]
 do_syscall_64+0x3b/0xb0 arch/x86/entry/common.c:81
 entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7f493cd7dff9
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 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 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f493db7b038 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007f493cf35f80 RCX: 00007f493cd7dff9
RDX: 0000000020000040 RSI: 00000000c028660f RDI: 0000000000000007
RBP: 00007f493cdf0296 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 00007f493cf35f80 R15: 00007ffc299f70a8
 </TASK>

Crashes (1):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2024/10/24 22:07 linux-6.1.y 7ec6f9fa3d97 0d144d1a .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-1-kasan possible deadlock in ext4_move_extents
* Struck through repros no longer work on HEAD.