Re: possible deadlock in svm_vm_copy_asid_from

[Wanpeng Li]

On Tue, 4 Jan 2022 at 19:01, kvartet <xyru1999@xxxxxxxxx> wrote:
>
> Hello,
>
> When using Syzkaller to fuzz the latest Linux kernel, the following
> crash was triggered.
>
> HEAD commit: a7904a538933 Linux 5.16-rc6
> git tree: upstream
> console output: https://paste.ubuntu.com/p/GCRXrYQmMN/plain/
> kernel config: https://paste.ubuntu.com/p/FDDNHDxtwz/plain/
> C reproducer: https://paste.ubuntu.com/p/gD2D5wthDK/plain/
> Syzlang reproducer: https://paste.ubuntu.com/p/hTnbvmsW8r/plain/
>
> ============================================
> WARNING: possible recursive locking detected
> 5.16.0-rc6 #9 Not tainted
> --------------------------------------------
> syz-executor.6/4919 is trying to acquire lock:
> ffffc9000afbb250 (&kvm->lock){+.+.}-{3:3}, at: sev_lock_two_vms
> arch/x86/kvm/svm/sev.c:1568 [inline]
> ffffc9000afbb250 (&kvm->lock){+.+.}-{3:3}, at:
> svm_vm_copy_asid_from+0x1bd/0x380 arch/x86/kvm/svm/sev.c:1988
>
> but task is already holding lock:
> ffffc9000a703250 (&kvm->lock){+.+.}-{3:3}, at: sev_lock_two_vms
> arch/x86/kvm/svm/sev.c:1566 [inline]
> ffffc9000a703250 (&kvm->lock){+.+.}-{3:3}, at:
> svm_vm_copy_asid_from+0x188/0x380 arch/x86/kvm/svm/sev.c:1988
>
> other info that might help us debug this:
>  Possible unsafe locking scenario:
>
>        CPU0
>        ----
>   lock(&kvm->lock);
>   lock(&kvm->lock);
>
>  *** DEADLOCK ***
>
>  May be due to missing lock nesting notation
>
> 1 lock held by syz-executor.6/4919:
>  #0: ffffc9000a703250 (&kvm->lock){+.+.}-{3:3}, at: sev_lock_two_vms
> arch/x86/kvm/svm/sev.c:1566 [inline]
>  #0: ffffc9000a703250 (&kvm->lock){+.+.}-{3:3}, at:
> svm_vm_copy_asid_from+0x188/0x380 arch/x86/kvm/svm/sev.c:1988
>
> stack backtrace:
> CPU: 1 PID: 4919 Comm: syz-executor.6 Not tainted 5.16.0-rc6 #9
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> 1.13.0-1ubuntu1.1 04/01/2014
> Call Trace:
>  <TASK>
>  __dump_stack lib/dump_stack.c:88 [inline]
>  dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
>  print_deadlock_bug kernel/locking/lockdep.c:2956 [inline]
>  check_deadlock kernel/locking/lockdep.c:2999 [inline]
>  validate_chain kernel/locking/lockdep.c:3788 [inline]
>  __lock_acquire.cold+0x168/0x3c3 kernel/locking/lockdep.c:5027
>  lock_acquire kernel/locking/lockdep.c:5637 [inline]
>  lock_acquire+0x1ab/0x520 kernel/locking/lockdep.c:5602
>  __mutex_lock_common kernel/locking/mutex.c:607 [inline]
>  __mutex_lock+0x151/0x1610 kernel/locking/mutex.c:740
>  sev_lock_two_vms arch/x86/kvm/svm/sev.c:1568 [inline]
>  svm_vm_copy_asid_from+0x1bd/0x380 arch/x86/kvm/svm/sev.c:1988
>  kvm_vm_ioctl_enable_cap+0xf8/0xc40 arch/x86/kvm/x86.c:5829
>  kvm_vm_ioctl_enable_cap_generic
> arch/x86/kvm/../../../virt/kvm/kvm_main.c:4241 [inline]
>  kvm_vm_ioctl+0x3dd/0x23a0 arch/x86/kvm/../../../virt/kvm/kvm_main.c:4300
>  vfs_ioctl fs/ioctl.c:51 [inline]
>  __do_sys_ioctl fs/ioctl.c:874 [inline]
>  __se_sys_ioctl fs/ioctl.c:860 [inline]
>  __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:860
>  do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>  do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>  entry_SYSCALL_64_after_hwframe+0x44/0xae
> RIP: 0033:0x7fc4241dc89d
> Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 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 bc ff ff ff f7 d8 64 89 01 48
> RSP: 002b:00007fc422b4dc28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
> RAX: ffffffffffffffda RBX: 00007fc4242fbf60 RCX: 00007fc4241dc89d
> RDX: 0000000020000080 RSI: 000000004068aea3 RDI: 0000000000000004
> RBP: 00007fc42424900d R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
> R13: 00007ffc9816f67f R14: 00007fc4242fbf60 R15: 00007fc422b4ddc0
>  </TASK>

Maybe mutex_lock_killable_nested(), I will have a try.

    Wanpeng