Re: [syzbot] Monthly kvm report (Jul 2023)

[Sean Christopherson <seanjc@google.com>]

On Mon, Jul 10, 2023, syzbot wrote:
> Hello kvm maintainers/developers,
> 
> This is a 31-day syzbot report for the kvm subsystem.
> All related reports/information can be found at:
> https://syzkaller.appspot.com/upstream/s/kvm
> 
> During the period, 2 new issues were detected and 0 were fixed.
> In total, 5 issues are still open and 111 have been fixed so far.
> 
> Some of the still happening issues:
> 
> Ref Crashes Repro Title
> <1> 138     Yes   WARNING in kvm_arch_vcpu_ioctl_run (5)
>                   https://syzkaller.appspot.com/bug?extid=5feef0b9ee9c8e9e5689

This will *hopefully* be resolved when the unrestricted guest patches[*] are
applied.  This WARN is a bit of a catch-all.  The upside is that it detects a lot
of bugs, the downside is completely unrelated bugs can have the same signature,
i.e. there may be more bugs lurking.

[*] https://lore.kernel.org/all/20230613203037.1968489-1-seanjc@google.com

> <2> 89      Yes   WARNING in handle_exception_nmi (2)
>                   https://syzkaller.appspot.com/bug?extid=4688c50a9c8e68e7aaa1

This is likely a bug in the underlying GCE kernel, i.e. in L0.  Not sure how to
bin this one.

> <3> 58      No    WARNING in kthread_bind_mask
>                   https://syzkaller.appspot.com/bug?extid=087b7effddeec0697c66

This is more than likely a bug in either the workqueues or in sched, e.g. btrfs hit
the same WARN a while back with a similarly innocuous alloc_workqueue() call.
I'll (try to) re-label to "kernel".