Re: [Xen-users] Domu windows 2012 crash.

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 19/03/2020 12:34, Johnny Ström wrote:
> On 3/18/20 5:36 PM, Andrew Cooper wrote:
>> On 18/03/2020 15:32, Johnny Ström wrote:
>>> On 3/14/20 2:54 PM, Hans van Kranenburg wrote:
>>>> Hi Johnny,
>>>>
>>>> On 2/20/20 10:10 AM, johnny Strom wrote:
>>>>> After the latest updates/security updates for debian 9 and debian 10,
>>>>> so will windows 2012 crash if more than 4096 memory is available to
>>>>> domu.
>>>> I am able to reproduce this with a Windows server essentials 2019 HVM
>>>> domU.
>>>>
>>>> An example from xl dmesg:
>>>>
>>>> (XEN) Failed to shatter gfn 105245: -12
>>>> (XEN) d75v1 EPT violation 0x19c (--x/rw-) gpa 0x00000105245760 mfn
>>>> 0x285245 type 0
>>>> (XEN) d75v1 Walking EPT tables for GFN 105245:
>>>> (XEN) d75v1  epte 9c000004105f9007
>>>> (XEN) d75v1  epte 9c000002800000f3
>>>> (XEN) d75v1  --- GLA 0x7ff98b40d760
>>>> (XEN) domain_crash called from vmx.c:3497
>>>> (XEN) Domain 75 (vcpu#1) crashed on cpu#4:
>>>> (XEN) ----[ Xen-4.11.4-pre  x86_64  debug=n   Not tainted ]----
>>>> (XEN) CPU:    4
>>>> (XEN) RIP:    0033:[<00007ff98b40d760>]
>>>> (XEN) RFLAGS: 0000000000010216   CONTEXT: hvm guest (d75v1)
>>>> (XEN) rax: 0000000000001212   rbx: 000000c714d9da58   rcx:
>>>> 0000023500001590
>>>> (XEN) rdx: 000000c700000001   rsi: 000000c714d9da18   rdi:
>>>> 000000c714d9db20
>>>> (XEN) rbp: 000000c714d9d950   rsp: 000000c714d9d918   r8:
>>>> 0000023500001470
>>>> (XEN) r9:  00000235000014f0   r10: 00007ff99c5c0923   r11:
>>>> 000000c714d9d970
>>>> (XEN) r12: 0000000000000000   r13: 000000c714d9d9d0   r14:
>>>> 000000c714d9da58
>>>> (XEN) r15: 0000000000000006   cr0: 0000000080050031   cr4:
>>>> 0000000000060678
>>>> (XEN) cr3: 00000001d9458002   cr2: 00007ff98b6fa048
>>>> (XEN) fsb: 0000000000000000   gsb: 000000c714e9e000   gss:
>>>> ffffcd015dc40000
>>>> (XEN) ds: 002b   es: 002b   fs: 0053   gs: 002b   ss: 002b   cs: 0033
>>>>
>>>>> So it's in debian 9 update: xen (4.8.5.final+shim4.10.4-1+deb9u12)
>>>>> and in debian 10 update:  xen (4.11.3+24-g14b62ab3e5-1) that is
>>>>> causing
>>>>> the crashes.
>>>> For 4.11, can you please try the workaround to enable "executable EPT
>>>> superpages" again?
>>>>
>>>> It's the ept=exec-sp hypervisor setting, and can also be set without
>>>> rebooting using:
>>>>
>>>>      xl set-parameters ept=exec-sp
>>>
>>> Hello
>>>
>>>
>>> I have been running test  with "xl set-parameters ept=exec-sp" and
>>> windows 2012 R2.
>>>
>>> And it dose not crash anymore.
>>>
>>> I used memory = 5120 that did make it crash within 2 minutes before.
>>>
>>> So xl set-parameters ept=exec-sp seems to be an workaround for now.
>> Right, but what you're doing is turning off the security fix for
>> XSA-304.
>>
>> Do you have any shadow_memory= option set in your VM's config file?  If
>> so, please remove it.
>
> Hi
>
> We had: shadow_memory = 8
>
> Rebooted the hypervisor and commented: out shadow_memory
>
> And no crash after some hours of running. So it seems to work.

Please can someone post this as a proper bug report to xen-devel@

We've clearly got a bug somewhere in the logic for determining and
enforcing the minimum safe shadow memory for a domain to use.

~Andrew

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-users