Re: [syzbot] WARNING in io_wq_put

[Jens Axboe]

On 3/25/21 11:58 AM, syzbot wrote:
> Hello,
> 
> syzbot has tested the proposed patch but the reproducer is still triggering an issue:
> WARNING in kvm_wait
> 
> ------------[ cut here ]------------
> raw_local_irq_restore() called with IRQs enabled
> WARNING: CPU: 1 PID: 191 at kernel/locking/irqflag-debug.c:10 warn_bogus_irq_restore+0x1f/0x30 kernel/locking/irqflag-debug.c:10
> Modules linked in:
> CPU: 1 PID: 191 Comm: kworker/u4:4 Not tainted 5.12.0-rc2-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> Workqueue:  0x0 (events_unbound)
> RIP: 0010:warn_bogus_irq_restore+0x1f/0x30 kernel/locking/irqflag-debug.c:10
> Code: cc cc cc cc cc cc cc cc cc cc cc 80 3d c7 af b1 03 00 74 01 c3 c6 05 bd af b1 03 01 48 c7 c7 c0 5f ae 89 31 c0 e8 d1 dd f6 f7 <0f> 0b c3 cc cc cc cc cc cc cc cc cc cc cc cc cc cc 41 56 53 48 83
> RSP: 0018:ffffc90000dc0a08 EFLAGS: 00010246
> RAX: 6a712abdc5855100 RBX: ffffffff8f982d60 RCX: ffff8880118bb880
> RDX: 0000000000000103 RSI: 0000000000000103 RDI: 0000000000000000
> RBP: 1ffff920001b8142 R08: ffffffff81609502 R09: ffffed10173e5fe8
> R10: ffffed10173e5fe8 R11: 0000000000000000 R12: 0000000000000003
> R13: ffff88823ffe6880 R14: 0000000000000246 R15: dffffc0000000000
> FS:  0000000000000000(0000) GS:ffff8880b9f00000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f44d117d718 CR3: 000000001340a000 CR4: 00000000001506e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
>  <IRQ>
>  kvm_wait+0x10e/0x160 arch/x86/kernel/kvm.c:860
>  pv_wait arch/x86/include/asm/paravirt.h:564 [inline]
>  pv_wait_head_or_lock kernel/locking/qspinlock_paravirt.h:470 [inline]
>  __pv_queued_spin_lock_slowpath+0x6b5/0xb90 kernel/locking/qspinlock.c:508
>  pv_queued_spin_lock_slowpath arch/x86/include/asm/paravirt.h:554 [inline]
>  queued_spin_lock_slowpath arch/x86/include/asm/qspinlock.h:51 [inline]
>  queued_spin_lock include/asm-generic/qspinlock.h:85 [inline]
>  do_raw_spin_lock+0x430/0x810 kernel/locking/spinlock_debug.c:113
>  spin_lock include/linux/spinlock.h:354 [inline]
>  mac80211_hwsim_tx_frame_no_nl+0x60e/0x1860 drivers/net/wireless/mac80211_hwsim.c:1514
>  mac80211_hwsim_tx_frame+0x143/0x180 drivers/net/wireless/mac80211_hwsim.c:1775
>  mac80211_hwsim_beacon_tx+0x4b9/0x870 drivers/net/wireless/mac80211_hwsim.c:1829
>  __iterate_interfaces+0x23e/0x4b0 net/mac80211/util.c:793
>  ieee80211_iterate_active_interfaces_atomic+0x9b/0x120 net/mac80211/util.c:829
>  mac80211_hwsim_beacon+0xa4/0x180 drivers/net/wireless/mac80211_hwsim.c:1852
>  __run_hrtimer kernel/time/hrtimer.c:1519 [inline]
>  __hrtimer_run_queues+0x4c9/0xa00 kernel/time/hrtimer.c:1583
>  hrtimer_run_softirq+0x176/0x1e0 kernel/time/hrtimer.c:1600
>  __do_softirq+0x318/0x714 kernel/softirq.c:345
>  invoke_softirq kernel/softirq.c:221 [inline]
>  __irq_exit_rcu+0x1d8/0x200 kernel/softirq.c:422
>  irq_exit_rcu+0x5/0x20 kernel/softirq.c:434
>  sysvec_apic_timer_interrupt+0x91/0xb0 arch/x86/kernel/apic/apic.c:1100
>  </IRQ>
>  asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:632
> RIP: 0010:__raw_spin_unlock_irq include/linux/spinlock_api_smp.h:169 [inline]
> RIP: 0010:_raw_spin_unlock_irq+0x25/0x40 kernel/locking/spinlock.c:199
> Code: b4 fd ff 66 90 53 48 89 fb 48 83 c7 18 48 8b 74 24 08 e8 0e 56 09 f8 48 89 df e8 56 2b 0b f8 e8 41 9f 2b f8 fb bf 01 00 00 00 <e8> c6 3b ff f7 65 8b 05 c7 9f ae 76 85 c0 74 02 5b c3 e8 7b fb ac
> RSP: 0018:ffffc9000143fca0 EFLAGS: 00000286
> RAX: 6a712abdc5855100 RBX: ffff8880b9f34c40 RCX: ffffffff8f59cb03
> RDX: 0000000040000000 RSI: 0000000000000002 RDI: 0000000000000001
> RBP: ffffc9000143fd00 R08: ffffffff817eef20 R09: ffffed10173e6989
> R10: ffffed10173e6989 R11: 0000000000000000 R12: ffff8880b9f34c40
> R13: ffff8880118bb880 R14: dffffc0000000000 R15: 0000000000000000
>  finish_task_switch+0x145/0x620 kernel/sched/core.c:4193
>  context_switch kernel/sched/core.c:4327 [inline]
>  __schedule+0x9a1/0xe70 kernel/sched/core.c:5075
>  schedule+0x14b/0x200 kernel/sched/core.c:5154
>  worker_thread+0xfe6/0x1300 kernel/workqueue.c:2442
>  kthread+0x39a/0x3c0 kernel/kthread.c:292
>  ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294

This is a different issue, not related to this code at all.

-- 
Jens Axboe