Re: kvm.git next: KVM internal error. Suberror: 1

[Jan Kiszka <jan.kiszka@siemens.com>]

Gleb Natapov wrote:
> On Mon, Feb 08, 2010 at 01:45:01PM +0100, Jan Kiszka wrote:
>> Avi Kivity wrote:
>>> On 02/08/2010 02:04 PM, Jan Kiszka wrote:
>>>> Avi,
>>>>
>>>> with 2c8232f over kvm-kmod and "qemu-system-x86_64 -m 256 vm-image.qcow2 -snapshot -serial stdio -s -smp 2" I just got this:
>>>>
>>>>    
>>> What is vm-image.qcow2?
>>>
>>>> KVM internal error. Suberror: 1
>>>> rax 0000000000000000 rbx 0000000000006f08 rcx 0000000000000000 rdx 0000000000000052
>>>> rsi 0000000000000000 rdi 00000000000f4fd4 rsp 0000000000006ed8 rbp 00000000000f7280
>>>> r8  0000000000000000 r9  0000000000000000 r10 0000000000000000 r11 0000000000000000
>>>> r12 0000000000000000 r13 0000000000000000 r14 0000000000000000 r15 0000000000000000
>>>> rip 00000000f000ff53 rflags 00010016
>>>>    
>>> ffffff53 is an 'iret'.  But f000ff53 doesn't make sense.
>>>
>>>> cs 0008 (00000000/ffffffff p 1 dpl 0 db 1 s 1 type b l 0 g 1 avl 0)
>>>> ds 0010 (00000000/ffffffff p 1 dpl 0 db 1 s 1 type 3 l 0 g 1 avl 0)
>>>> es 0010 (00000000/ffffffff p 1 dpl 0 db 1 s 1 type 3 l 0 g 1 avl 0)
>>>> ss 0010 (00000000/ffffffff p 1 dpl 0 db 1 s 1 type 3 l 0 g 1 avl 0)
>>>> fs 0010 (00000000/ffffffff p 1 dpl 0 db 1 s 1 type 3 l 0 g 1 avl 0)
>>>> gs 0010 (00000000/ffffffff p 1 dpl 0 db 1 s 1 type 3 l 0 g 1 avl 0)
>>>> tr 0000 (00000000/0000ffff p 1 dpl 0 db 0 s 0 type b l 0 g 0 avl 0)
>>>> ldt 0000 (00000000/0000ffff p 1 dpl 0 db 0 s 0 type 2 l 0 g 0 avl 0)
>>>> gdt f7a20/37
>>>> idt f8aa0/0
>>>> cr0 11 cr2 0 cr3 0 cr4 0 cr8 0 efer 0
>>>> emulation failure
>>>>
>>>> Before that run, I started the very same VM and shut it down via
>>>> system_powerdown. This is reproducible!
>>>>    
>>> Not sure I understand.  This is with -snapshot, so how can a previous 
>>> run have any effect?
>>>
>>>> Maybe it's the same issue that causes the #UD regression with
>>>> -no-kvm-irqchip.
>>> I wasn't able to reproduce.
>>>
>> Looks like tried to outsource my own bugs: I was on queues/vcpu-state,
>> ie. my state writeback rework, and I'm unable to reproduce over qemu-kvm
>> master. Will do my homework.
>>
> Saw this behaviour in my testing. Sometimes during startup I get similar
> errors. Wasn't sure that this is not my fault, so didn't report.

So far it looks like the final patch is yet too early, something of the
initialization is still fragile.

Jan

-- 
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux