Re: qemu-kvm-1.1.0 crashing with kernel 3.5.0-rc6

[Chris Clayton <chris2553@googlemail.com>]

On 07/11/12 08:22, Gleb Natapov wrote:
> On Wed, Jul 11, 2012 at 08:18:17AM +0100, Chris Clayton wrote:
>> On 07/11/12 08:12, Gleb Natapov wrote:
>>> On Wed, Jul 11, 2012 at 08:09:42AM +0100, Chris Clayton wrote:
>>>> Ping.
>>>>
>>>> Have I committed a bug-reporting sin in the mail below or is
>>>> everyone simply too busy to look at this kvm-related crash?
>>>>
>>> Since you have good and bad points can you bisect the problem?
>>>
>>
>> Yes, I can bisect, but since the crash occurs with only only one
>> combination of qemu-kvm (1.1.0) and kernel (3.5.0-rc6), I'm not sure
>> which of those I should bisect. Any ideas on how I could narrow that
>> down.
>>
> Bisect qemu between qemu-kvm-1.0.1 & qemu-kvm-1.1.0.
>

Change of diagnostics, unfortunately. qemu-kvm-1.0.1 can, in fact, crash 
on 3.5.0-rc6 (and rc7). I didn't get it earlier because it takes many 
times more invocations before the crash occurs with 1.0.1 and I haven't 
used qemu-kvm much in the past few weeks.

I'm now checking whether I can get crashes (with 1.0.1 and/or 1.1.0) on 
linux-3.4.4. I'll report back in a day or two.

>> Thanks.
>>>> On 07/09/12 11:57, Chris Clayton wrote:
>>>>> Hi,
>>>>>
>>>>> When I run WinXP SP3 through qemu-kvm-1.1.0 on linux kernel 3.5.0-rc6, I
>>>>> get a segmentation fault within 3 or 4 minutes maximum. In dmesg I see:
>>>>>
>>>>> qemu-kvm: sending ioctl 5326 to a partition!
>>>>> qemu-kvm: sending ioctl 801c0204 to a partition!
>>>>> qemu-kvm: sending ioctl 5326 to a partition!
>>>>> qemu-kvm: sending ioctl 801c0204 to a partition!
>>>>> qemu-kvm: sending ioctl 5326 to a partition!
>>>>> qemu-kvm: sending ioctl 801c0204 to a partition!
>>>>> qemu-kvm: sending ioctl 5326 to a partition!
>>>>> qemu-kvm: sending ioctl 801c0204 to a partition!
>>>>> qemu-kvm[860] general protection ip:b6abad77 sp:b52ff09c error:0 in
>>>>> libc-2.16.so[b697d000+1b4000]
>>>>>
>>>>> The crash does not occur with qemu-kvm-1.0.1 on rc6. Nor does it occur
>>>>> qemu-kvm-1.0.1 or qemu-kvm-1.1.0 on kernel 3.4.4. All three combinations
>>>>> survive for 15 minutes or more
>>>>>
>>>>> When I try to get a backtrace with gdb, the screen on which konsole and
>>>>> qemu are running locks up until I kill qemu in another console.
>>>>> Consequently I can't get a full BT, but, although probably not very
>>>>> helpful, what I did get is:
>>>>>
>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>> [Switching to Thread 0xb6946b40 (LWP 506)]
>>>>> 0xb7705d77 in __strcmp_sse4_2 () from /lib/libc.so.6
>>>>> (gdb) bt
>>>>> #0  0xb7705d77 in __strcmp_sse4_2 () from /lib/libc.so.6
>>>>> #1  0xb7e8d6e3 in g_str_equal () from /usr/lib/libglib-2.0.so.0
>>>>> #2  0xb7e8c94c in g_hash_table_lookup () from /usr/lib/libglib-2.0.so.0
>>>>> Cannot access memory at address 0xb694610c
>>>>> (gdb)
>>>>>
>>>>> Note that the gdb and dmesg outputs above are not from the same crash
>>>>> instance.
>>>>>
>>>>> I'm not subscribed,so please cc me on any reply.
>>>>>
>>>>> Happy to provide any additional diagnostics (but may need help on how to
>>>>> get them) or test patches, etc
>>>>>
>>>>> Thanks
>>>>>
>>>>> Chris Clayton
>>>>
>>>>
>>>> --
>>>> To unsubscribe from this list: send the line "unsubscribe kvm" in
>>>> the body of a message to majordomo@vger.kernel.org
>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>
>>> --
>>> 			Gleb.
>>>
>>
>
> --
> 			Gleb.
>