possible deadlock in io_timeout_fn

Status: fixed on 2020/09/16 22:51
Subsystems: io-uring fs
[Documentation on labels]
Reported-by: syzbot+ef4b654b49ed7ff049bf@syzkaller.appspotmail.com
Fix commit: 51a4cc112c7a io_uring: defer file table grabbing request cleanup for locked requests
First crash: 2193d, last: 2189d
Cause bisection: introduced by (bisect log) :
commit e62753e4e2926f249d088cc0517be5ed4efec6d6
Author: Bijan Mottahedeh <bijan.mottahedeh@oracle.com>
Date: Sat May 23 04:31:18 2020 +0000

  io_uring: call statx directly

Crash: possible deadlock in io_timeout_fn (log)
Repro: C syz .config
  
▶ ▼ Discussions (1)
Title Replies (including bot) Last reply
possible deadlock in io_timeout_fn 1 (2) 2020/08/10 16:53
▶ ▼ Similar bugs (1)
Kernel Title Rank 🛈 Repro Cause bisect Fix bisect Count Last Reported Patched Status
upstream possible deadlock in io_timeout_fn (2) fs 4 4 2041d 2048d 19/29 fixed on 2021/03/10 01:48

Sample crash report:
========================================================
WARNING: possible irq lock inversion dependency detected
5.8.0-syzkaller #0 Not tainted
--------------------------------------------------------
syz-executor046/6843 just changed the state of lock:
ffff8880a6b534d8 (&ctx->completion_lock){-...}-{2:2}, at: io_timeout_fn+0x6c/0x3f0 fs/io_uring.c:4999
but this lock took another, HARDIRQ-unsafe lock in the past:
 (&fs->lock){+.+.}-{2:2}


and interrupts could create inverse lock ordering between them.


other info that might help us debug this:
 Possible interrupt unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&fs->lock);
                               local_irq_disable();
                               lock(&ctx->completion_lock);
                               lock(&fs->lock);
  <Interrupt>
    lock(&ctx->completion_lock);

 *** DEADLOCK ***

1 lock held by syz-executor046/6843:
 #0: ffff8880a6b53428 (&ctx->uring_lock){+.+.}-{3:3}, at: __do_sys_io_uring_enter+0xdb7/0x1ae0 fs/io_uring.c:8035

the shortest dependencies between 2nd lock and 1st lock:
 -> (&fs->lock){+.+.}-{2:2} {
    HARDIRQ-ON-W at:
                      lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
                      __raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline]
                      _raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:151
                      spin_lock include/linux/spinlock.h:354 [inline]
                      set_fs_pwd+0x85/0x290 fs/fs_struct.c:39
                      init_chdir+0x106/0x14e fs/init.c:54
                      devtmpfs_setup drivers/base/devtmpfs.c:415 [inline]
                      devtmpfsd+0x76/0x333 drivers/base/devtmpfs.c:430
                      kthread+0x3b5/0x4a0 kernel/kthread.c:292
                      ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
    SOFTIRQ-ON-W at:
                      lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
                      __raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline]
                      _raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:151
                      spin_lock include/linux/spinlock.h:354 [inline]
                      set_fs_pwd+0x85/0x290 fs/fs_struct.c:39
                      init_chdir+0x106/0x14e fs/init.c:54
                      devtmpfs_setup drivers/base/devtmpfs.c:415 [inline]
                      devtmpfsd+0x76/0x333 drivers/base/devtmpfs.c:430
                      kthread+0x3b5/0x4a0 kernel/kthread.c:292
                      ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
    INITIAL USE at:
                     lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
                     __raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline]
                     _raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:151
                     spin_lock include/linux/spinlock.h:354 [inline]
                     set_fs_pwd+0x85/0x290 fs/fs_struct.c:39
                     init_chdir+0x106/0x14e fs/init.c:54
                     devtmpfs_setup drivers/base/devtmpfs.c:415 [inline]
                     devtmpfsd+0x76/0x333 drivers/base/devtmpfs.c:430
                     kthread+0x3b5/0x4a0 kernel/kthread.c:292
                     ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
  }
  ... key      at: [<ffffffff8c98c220>] __key.1+0x0/0x40
  ... acquired at:
   __raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline]
   _raw_spin_lock+0x2a/0x40 kernel/locking/spinlock.c:151
   spin_lock include/linux/spinlock.h:354 [inline]
   io_req_clean_work fs/io_uring.c:1126 [inline]
   io_dismantle_req+0x3ec/0x9e0 fs/io_uring.c:1544
   __io_free_req+0x16/0x3c0 fs/io_uring.c:1562
   __io_double_put_req fs/io_uring.c:1909 [inline]
   __io_fail_links+0x433/0x5b0 fs/io_uring.c:1659
   io_fail_links fs/io_uring.c:1675 [inline]
   __io_req_find_next+0x368/0x460 fs/io_uring.c:1698
   io_req_find_next fs/io_uring.c:1706 [inline]
   io_steal_work fs/io_uring.c:1897 [inline]
   io_wq_submit_work+0x33c/0x3d0 fs/io_uring.c:5792
   io_worker_handle_work+0xa45/0x13f0 fs/io-wq.c:527
   io_wqe_worker+0xbf0/0x10e0 fs/io-wq.c:569
   kthread+0x3b5/0x4a0 kernel/kthread.c:292
   ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294

-> (&ctx->completion_lock){-...}-{2:2} {
   IN-HARDIRQ-W at:
                    lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
                    __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
                    _raw_spin_lock_irqsave+0x8c/0xc0 kernel/locking/spinlock.c:159
                    io_timeout_fn+0x6c/0x3f0 fs/io_uring.c:4999
                    __run_hrtimer kernel/time/hrtimer.c:1520 [inline]
                    __hrtimer_run_queues+0x6a9/0xfc0 kernel/time/hrtimer.c:1584
                    hrtimer_interrupt+0x32a/0x930 kernel/time/hrtimer.c:1646
                    local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1079 [inline]
                    __sysvec_apic_timer_interrupt+0x142/0x5e0 arch/x86/kernel/apic/apic.c:1096
                    asm_call_on_stack+0xf/0x20 arch/x86/entry/entry_64.S:706
                    __run_on_irqstack arch/x86/include/asm/irq_stack.h:22 [inline]
                    run_on_irqstack_cond arch/x86/include/asm/irq_stack.h:48 [inline]
                    sysvec_apic_timer_interrupt+0xb2/0xf0 arch/x86/kernel/apic/apic.c:1090
                    asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:581
                    arch_local_irq_enable arch/x86/include/asm/paravirt.h:780 [inline]
                    __raw_spin_unlock_irq include/linux/spinlock_api_smp.h:168 [inline]
                    _raw_spin_unlock_irq+0x4b/0x80 kernel/locking/spinlock.c:199
                    spin_unlock_irq include/linux/spinlock.h:404 [inline]
                    io_timeout fs/io_uring.c:5162 [inline]
                    io_issue_sqe+0x2de6/0x60d0 fs/io_uring.c:5594
                    __io_queue_sqe+0x284/0x1190 fs/io_uring.c:5981
                    io_queue_sqe+0x73e/0x1130 fs/io_uring.c:6060
                    io_submit_sqe fs/io_uring.c:6130 [inline]
                    io_submit_sqes+0x1794/0x2380 fs/io_uring.c:6327
                    __do_sys_io_uring_enter+0xdc7/0x1ae0 fs/io_uring.c:8036
                    do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
                    entry_SYSCALL_64_after_hwframe+0x44/0xa9
   INITIAL USE at:
                   lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
                   __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
                   _raw_spin_lock_irqsave+0x8c/0xc0 kernel/locking/spinlock.c:159
                   io_cqring_add_event fs/io_uring.c:1419 [inline]
                   __io_req_complete fs/io_uring.c:1458 [inline]
                   __io_req_complete fs/io_uring.c:1454 [inline]
                   io_req_complete fs/io_uring.c:1472 [inline]
                   io_madvise fs/io_uring.c:3737 [inline]
                   io_issue_sqe+0xcfd/0x60d0 fs/io_uring.c:5682
                   io_wq_submit_work+0x183/0x3d0 fs/io_uring.c:5775
                   io_worker_handle_work+0xa45/0x13f0 fs/io-wq.c:527
                   io_wqe_worker+0xbf0/0x10e0 fs/io-wq.c:569
                   kthread+0x3b5/0x4a0 kernel/kthread.c:292
                   ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
 }
 ... key      at: [<ffffffff8c98dc80>] __key.9+0x0/0x40
 ... acquired at:
   mark_lock_irq kernel/locking/lockdep.c:3568 [inline]
   mark_lock+0x54b/0x1710 kernel/locking/lockdep.c:4006
   mark_usage kernel/locking/lockdep.c:3902 [inline]
   __lock_acquire+0x13ad/0x5640 kernel/locking/lockdep.c:4380
   lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
   __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
   _raw_spin_lock_irqsave+0x8c/0xc0 kernel/locking/spinlock.c:159
   io_timeout_fn+0x6c/0x3f0 fs/io_uring.c:4999
   __run_hrtimer kernel/time/hrtimer.c:1520 [inline]
   __hrtimer_run_queues+0x6a9/0xfc0 kernel/time/hrtimer.c:1584
   hrtimer_interrupt+0x32a/0x930 kernel/time/hrtimer.c:1646
   local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1079 [inline]
   __sysvec_apic_timer_interrupt+0x142/0x5e0 arch/x86/kernel/apic/apic.c:1096
   asm_call_on_stack+0xf/0x20 arch/x86/entry/entry_64.S:706
   __run_on_irqstack arch/x86/include/asm/irq_stack.h:22 [inline]
   run_on_irqstack_cond arch/x86/include/asm/irq_stack.h:48 [inline]
   sysvec_apic_timer_interrupt+0xb2/0xf0 arch/x86/kernel/apic/apic.c:1090
   asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:581
   arch_local_irq_enable arch/x86/include/asm/paravirt.h:780 [inline]
   __raw_spin_unlock_irq include/linux/spinlock_api_smp.h:168 [inline]
   _raw_spin_unlock_irq+0x4b/0x80 kernel/locking/spinlock.c:199
   spin_unlock_irq include/linux/spinlock.h:404 [inline]
   io_timeout fs/io_uring.c:5162 [inline]
   io_issue_sqe+0x2de6/0x60d0 fs/io_uring.c:5594
   __io_queue_sqe+0x284/0x1190 fs/io_uring.c:5981
   io_queue_sqe+0x73e/0x1130 fs/io_uring.c:6060
   io_submit_sqe fs/io_uring.c:6130 [inline]
   io_submit_sqes+0x1794/0x2380 fs/io_uring.c:6327
   __do_sys_io_uring_enter+0xdc7/0x1ae0 fs/io_uring.c:8036
   do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
   entry_SYSCALL_64_after_hwframe+0x44/0xa9


stack backtrace:
CPU: 0 PID: 6843 Comm: syz-executor046 Not tainted 5.8.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 <IRQ>
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x18f/0x20d lib/dump_stack.c:118
 print_irq_inversion_bug kernel/locking/lockdep.c:3429 [inline]
 check_usage_forwards.cold+0x1e/0x27 kernel/locking/lockdep.c:3453
 mark_lock_irq kernel/locking/lockdep.c:3568 [inline]
 mark_lock+0x54b/0x1710 kernel/locking/lockdep.c:4006
 mark_usage kernel/locking/lockdep.c:3902 [inline]
 __lock_acquire+0x13ad/0x5640 kernel/locking/lockdep.c:4380
 lock_acquire+0x1f1/0xad0 kernel/locking/lockdep.c:5005
 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
 _raw_spin_lock_irqsave+0x8c/0xc0 kernel/locking/spinlock.c:159
 io_timeout_fn+0x6c/0x3f0 fs/io_uring.c:4999
 __run_hrtimer kernel/time/hrtimer.c:1520 [inline]
 __hrtimer_run_queues+0x6a9/0xfc0 kernel/time/hrtimer.c:1584
 hrtimer_interrupt+0x32a/0x930 kernel/time/hrtimer.c:1646
 local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1079 [inline]
 __sysvec_apic_timer_interrupt+0x142/0x5e0 arch/x86/kernel/apic/apic.c:1096
 asm_call_on_stack+0xf/0x20 arch/x86/entry/entry_64.S:706
 </IRQ>
 __run_on_irqstack arch/x86/include/asm/irq_stack.h:22 [inline]
 run_on_irqstack_cond arch/x86/include/asm/irq_stack.h:48 [inline]
 sysvec_apic_timer_interrupt+0xb2/0xf0 arch/x86/kernel/apic/apic.c:1090
 asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:581
RIP: 0010:__raw_spin_unlock_irq include/linux/spinlock_api_smp.h:169 [inline]
RIP: 0010:_raw_spin_unlock_irq+0x4b/0x80 kernel/locking/spinlock.c:199
Code: c0 58 34 b6 89 48 ba 00 00 00 00 00 fc ff df 48 c1 e8 03 80 3c 10 00 75 31 48 83 3d 06 e0 c2 01 00 74 25 fb 66 0f 1f 44 00 00 <bf> 01 00 00 00 e8 bb 6c 5c f9 65 8b 05 54 aa 0e 78 85 c0 74 02 5d
RSP: 0018:ffffc900054af8b0 EFLAGS: 00000286
RAX: 1ffffffff136c68b RBX: 0000000000000000 RCX: 0000000000000006
RDX: dffffc0000000000 RSI: 0000000000000000 RDI: ffffffff87f3542f
RBP: ffff8880a6b534c0 R08: 0000000000000001 R09: ffffffff8c5eaadf
R10: fffffbfff18bd55b R11: 000000000001f8e0 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: ffff8880a61c4d00
 spin_unlock_irq include/linux/spinlock.h:404 [inline]
 io_timeout fs/io_uring.c:5162 [inline]
 io_issue_sqe+0x2de6/0x60d0 fs/io_uring.c:5594
 __io_queue_sqe+0x284/0x1190 fs/io_uring.c:5981
 io_queue_sqe+0x73e/0x1130 fs/io_uring.c:6060
 io_submit_sqe fs/io_uring.c:6130 [inline]
 io_submit_sqes+0x1794/0x2380 fs/io_uring.c:6327
 __do_sys_io_uring_enter+0xdc7/0x1ae0 fs/io_uring.c:8036
 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x440b99
Code: 18 89 d0 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 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 0f 83 7b 13 fc ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007ffc41ef4908 EFLAGS: 00000246 ORIG_RAX: 00000000000001aa
RAX: ffffffffffffff

Crashes (4):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2020/08/11 05:31 upstream fc80c51fd4b2 7adc7b65 .config console log report syz C ci-upstream-kasan-gce-root
2020/08/09 19:50 upstream 06a81c1c7db9 70301872 .config console log report syz C ci-upstream-kasan-gce-selinux-root
2020/08/09 03:50 upstream 449dc8c97089 f721e4a0 .config console log report syz C ci-upstream-kasan-gce-smack-root
2020/08/13 03:45 linux-next bc09acc9f224 bc15f7db .config console log report syz C ci-upstream-linux-next-kasan-gce-root
* Struck through repros no longer work on HEAD.