syzbot


possible deadlock in proc_tid_io_accounting

Status: auto-closed as invalid on 2021/07/31 05:58
Reported-by: syzbot+95cfd7e72b9dbdaa4bee@syzkaller.appspotmail.com
First crash: 541d, last: 461d
similar bugs (3):
Kernel Title Repro Cause bisect Fix bisect Count Last Reported Patched Status
linux-4.19 possible deadlock in proc_tid_io_accounting 36 5d09h 654d 0/1 upstream: reported on 2020/09/21 07:35
linux-4.14 possible deadlock in proc_tid_io_accounting (2) 1 40d 40d 0/1 upstream: reported on 2022/05/27 23:12
upstream possible deadlock in proc_tid_io_accounting 1 987d 983d 0/22 auto-closed as invalid on 2020/02/21 09:02

Sample crash report:
Quota error (device loop2): v2_read_file_info: Number of blocks too big for quota file size (52959232 > 6144).
EXT4-fs warning (device loop2): ext4_enable_quotas:5758: Failed to enable quota tracking (type=-1, err=-117). Please run e2fsck to fix.
======================================================
WARNING: possible circular locking dependency detected
4.14.228-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor.0/18454 is trying to acquire lock:
 (&sig->cred_guard_mutex){+.+.}, at: [<ffffffff81a06bc4>] do_io_accounting fs/proc/base.c:2725 [inline]
 (&sig->cred_guard_mutex){+.+.}, at: [<ffffffff81a06bc4>] proc_tid_io_accounting+0x184/0x290 fs/proc/base.c:2768
EXT4-fs (loop2): mount failed

but task is already holding lock:
 (&p->lock){+.+.}, at: [<ffffffff818e708a>] seq_read+0xba/0x1120 fs/seq_file.c:165

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #3 (&p->lock){+.+.}:
       __mutex_lock_common kernel/locking/mutex.c:756 [inline]
       __mutex_lock+0xc4/0x1310 kernel/locking/mutex.c:893
       seq_read+0xba/0x1120 fs/seq_file.c:165
       proc_reg_read+0xee/0x1a0 fs/proc/inode.c:217
       do_loop_readv_writev fs/read_write.c:695 [inline]
       do_loop_readv_writev fs/read_write.c:682 [inline]
       do_iter_read+0x3eb/0x5b0 fs/read_write.c:919
       vfs_readv+0xc8/0x120 fs/read_write.c:981
       kernel_readv fs/splice.c:361 [inline]
       default_file_splice_read+0x418/0x910 fs/splice.c:416
       do_splice_to+0xfb/0x140 fs/splice.c:880
       splice_direct_to_actor+0x207/0x730 fs/splice.c:952
       do_splice_direct+0x164/0x210 fs/splice.c:1061
       do_sendfile+0x47f/0xb30 fs/read_write.c:1441
       SYSC_sendfile64 fs/read_write.c:1502 [inline]
       SyS_sendfile64+0xff/0x110 fs/read_write.c:1488
       do_syscall_64+0x1d5/0x640 arch/x86/entry/common.c:292
       entry_SYSCALL_64_after_hwframe+0x46/0xbb

-> #2 (sb_writers#3){.+.+}:
       percpu_down_read_preempt_disable include/linux/percpu-rwsem.h:36 [inline]
       percpu_down_read include/linux/percpu-rwsem.h:59 [inline]
       __sb_start_write+0x64/0x260 fs/super.c:1342
       sb_start_write include/linux/fs.h:1549 [inline]
       mnt_want_write+0x3a/0xb0 fs/namespace.c:386
       ovl_setattr+0x52/0x500 fs/overlayfs/inode.c:38
       notify_change+0x56b/0xd10 fs/attr.c:313
       chown_common+0x40b/0x4b0 fs/open.c:631
       SYSC_fchownat fs/open.c:661 [inline]
       SyS_fchownat+0xf6/0x190 fs/open.c:641
       do_syscall_64+0x1d5/0x640 arch/x86/entry/common.c:292
       entry_SYSCALL_64_after_hwframe+0x46/0xbb

-> #1 (&ovl_i_mutex_dir_key[depth]#2){++++}:
       down_read+0x36/0x80 kernel/locking/rwsem.c:24
       inode_lock_shared include/linux/fs.h:729 [inline]
       do_last fs/namei.c:3333 [inline]
       path_openat+0x149b/0x2970 fs/namei.c:3569
       do_filp_open+0x179/0x3c0 fs/namei.c:3603
       do_open_execat+0xd3/0x450 fs/exec.c:849
       do_execveat_common+0x711/0x1f30 fs/exec.c:1755
       do_execve fs/exec.c:1860 [inline]
       SYSC_execve fs/exec.c:1941 [inline]
       SyS_execve+0x3b/0x50 fs/exec.c:1936
       do_syscall_64+0x1d5/0x640 arch/x86/entry/common.c:292
       entry_SYSCALL_64_after_hwframe+0x46/0xbb

-> #0 (&sig->cred_guard_mutex){+.+.}:
       lock_acquire+0x170/0x3f0 kernel/locking/lockdep.c:3998
       __mutex_lock_common kernel/locking/mutex.c:756 [inline]
       __mutex_lock+0xc4/0x1310 kernel/locking/mutex.c:893
       do_io_accounting fs/proc/base.c:2725 [inline]
       proc_tid_io_accounting+0x184/0x290 fs/proc/base.c:2768
       proc_single_show+0xe7/0x150 fs/proc/base.c:761
       traverse+0x306/0x850 fs/seq_file.c:111
       seq_read+0x94b/0x1120 fs/seq_file.c:192
       do_loop_readv_writev fs/read_write.c:695 [inline]
       do_loop_readv_writev fs/read_write.c:682 [inline]
       do_iter_read+0x3eb/0x5b0 fs/read_write.c:919
       vfs_readv+0xc8/0x120 fs/read_write.c:981
       do_preadv fs/read_write.c:1065 [inline]
       SYSC_preadv fs/read_write.c:1115 [inline]
       SyS_preadv+0x15a/0x200 fs/read_write.c:1110
       do_syscall_64+0x1d5/0x640 arch/x86/entry/common.c:292
       entry_SYSCALL_64_after_hwframe+0x46/0xbb

other info that might help us debug this:

Chain exists of:
  &sig->cred_guard_mutex --> sb_writers#3 --> &p->lock

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&p->lock);
                               lock(sb_writers#3);
                               lock(&p->lock);
  lock(&sig->cred_guard_mutex);

 *** DEADLOCK ***

1 lock held by syz-executor.0/18454:
 #0:  (&p->lock){+.+.}, at: [<ffffffff818e708a>] seq_read+0xba/0x1120 fs/seq_file.c:165

stack backtrace:
CPU: 1 PID: 18454 Comm: syz-executor.0 Not tainted 4.14.228-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:17 [inline]
 dump_stack+0x1b2/0x281 lib/dump_stack.c:58
 print_circular_bug.constprop.0.cold+0x2d7/0x41e kernel/locking/lockdep.c:1258
 check_prev_add kernel/locking/lockdep.c:1905 [inline]
 check_prevs_add kernel/locking/lockdep.c:2022 [inline]
 validate_chain kernel/locking/lockdep.c:2464 [inline]
 __lock_acquire+0x2e0e/0x3f20 kernel/locking/lockdep.c:3491
 lock_acquire+0x170/0x3f0 kernel/locking/lockdep.c:3998
 __mutex_lock_common kernel/locking/mutex.c:756 [inline]
 __mutex_lock+0xc4/0x1310 kernel/locking/mutex.c:893
 do_io_accounting fs/proc/base.c:2725 [inline]
 proc_tid_io_accounting+0x184/0x290 fs/proc/base.c:2768
 proc_single_show+0xe7/0x150 fs/proc/base.c:761
 traverse+0x306/0x850 fs/seq_file.c:111
 seq_read+0x94b/0x1120 fs/seq_file.c:192
 do_loop_readv_writev fs/read_write.c:695 [inline]
 do_loop_readv_writev fs/read_write.c:682 [inline]
 do_iter_read+0x3eb/0x5b0 fs/read_write.c:919
 vfs_readv+0xc8/0x120 fs/read_write.c:981
 do_preadv fs/read_write.c:1065 [inline]
 SYSC_preadv fs/read_write.c:1115 [inline]
 SyS_preadv+0x15a/0x200 fs/read_write.c:1110
 do_syscall_64+0x1d5/0x640 arch/x86/entry/common.c:292
 entry_SYSCALL_64_after_hwframe+0x46/0xbb
RIP: 0033:0x466459
RSP: 002b:00007fccdf9a9188 EFLAGS: 00000246 ORIG_RAX: 0000000000000127
RAX: ffffffffffffffda RBX: 000000000056bf60 RCX: 0000000000466459
RDX: 0000000000000009 RSI: 0000000020004980 RDI: 0000000000000004
RBP: 00000000004bf9fb R08: 000000000000007f R09: 0000000000000000
R10: 0000000000000006 R11: 0000000000000246 R12: 000000000056bf60
R13: 00007ffc760c835f R14: 00007fccdf9a9300 R15: 0000000000022000
Quota error (device loop2): v2_read_file_info: Number of blocks too big for quota file size (52959232 > 6144).
EXT4-fs warning (device loop2): ext4_enable_quotas:5758: Failed to enable quota tracking (type=-1, err=-117). Please run e2fsck to fix.
EXT4-fs (loop2): mount failed
FAT-fs (loop2): Unrecognized mount option ">bi$dK?:w׈I|m!	.IO<}IES(Jh<L?\ㆡ2=ÝT655^&{=b-%K$" or missing value
caif:caif_disconnect_client(): nothing to disconnect
caif:caif_disconnect_client(): nothing to disconnect
IPVS: ftp: loaded support on port[0] = 21
FAT-fs (loop0): IO charset cp932H$ not found
FAT-fs (loop0): IO charset cp932H$ not found
IPVS: ftp: loaded support on port[0] = 21
xt_helper: cannot load conntrack support for proto=10
xt_helper: cannot load conntrack support for proto=10
overlayfs: upperdir is in-use by another mount, mount with '-o index=off' to override exclusive upperdir protection.
overlayfs: filesystem on './bus' not supported as upperdir
ufs: ufs was compiled with read-only support, can't be mounted as read-write
overlayfs: filesystem on './bus' not supported as upperdir
 loop2: [POWERTEC] p1 p2
loop2: p1 start 134152192 is beyond EOD, truncated
loop2: p2 size 2005401600 extends beyond EOD, truncated
NILFS (loop0): unable to read secondary superblock (blocksize = 1024)
NILFS (loop0): broken superblock, retrying with spare superblock (blocksize = 1024)
NILFS (loop0): unrecognized mount option "y6dx9+<9[F'wХu)CVFi<_X]ehsdY/CRO7@%9 D<-r3LBAU3{'aްTۧ-Sfe4"
NILFS (loop0): unable to read secondary superblock (blocksize = 1024)
NILFS (loop0): broken superblock, retrying with spare superblock (blocksize = 1024)
NILFS (loop0): unrecognized mount option "y6dx9+<9[F'wХu)CVFi<_X]ehsdY/CRO7@%9 D<-r3LBAU3{'aްTۧ-Sfe4"
libceph: resolve 'd ' (ret=-3): failed
libceph: parse_ips bad ip '[d :7]'
FAT-fs (loop2): IO charset cp932H$ not found
FAT-fs (loop2): IO charset cp932 not found

Crashes (3):
Manager Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Title
ci2-linux-4-14 2021/04/02 05:57 linux-4.14.y bd634aa64163 6a81331a .config log report info possible deadlock in proc_tid_io_accounting
ci2-linux-4-14 2021/04/01 07:42 linux-4.14.y bd634aa64163 6a81331a .config log report info possible deadlock in proc_tid_io_accounting
ci2-linux-4-14 2021/01/12 05:52 linux-4.14.y ec822b3e8bf4 2c1f2513 .config log report info