[Top] [All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] Doamin crash when trying to install disk encryption (PointSec) on Windows HVM

On 22/04/2009 15:40, "Tom Rotenberg" <tom.rotenberg@xxxxxxxxx> wrote:

> So, do u have any suggestion for me, on how to advance with this issue? should
> i debug the real mode emulation myself in order to track it down? will u be
> able top assist me in any way with this issue?

If specific questions arise then certainly yes.

> BTW - what is exactly the problem which causes this exception from the CPU?
> Tim mentioned something about segment state - can u please elaborate on this
> issue?

Bit 17 in EFLAGS is set, meaning the guest is in vm86 mode. In this mode, on
vmentry, the processor applies specific checks to guest segment register
state. These are listed in SDM Vol 3B, but they include checking that
segment selector equals segment base address shifted right four bit
positions, for example. Clearly the segment state in the dump you emailed is
not valid for vm86 mode. Quite possibly PointSec doesn't expect to be in
vm86 mode at all, and something may have gone wrong to set that bit in
EFLAGS. But it's hard to say at arm's length.

 -- Keir

> Tom
> 
> 2009/4/22 Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
>> Then it probably is a mis-emulation somewhere down the line. Unfortunately
>> that could be hard to track down, even if we had PointSec software to hand,
>> which we do not.
>> 
>>  -- Keir
>> 
>> On 22/04/2009 15:20, "Tom Rotenberg" <tom.rotenberg@xxxxxxxxx> wrote:
>> 
>>> Well, just tried this patch, and it doesn't seem to work.
>>> I get the following error:
>>> 
>>> (XEN) HVM1: Booting from 0000:7c00
>>> (XEN) Failed vm entry (exit reason 0x80000021) caused by invalid guest state
>>> (0).
>>> (XEN) ************* VMCS Area **************
>>> (XEN) *** Guest State ***
>>> (XEN) CR0: actual=0x0000000080010039, shadow=0x0000000080000019,
>>> gh_mask=ffffffffffffffff
>>> (XEN) CR4: actual=0x0000000000002060, shadow=0x0000000000000000,
>>> gh_mask=ffffffffffffffff
>>> (XEN) CR3: actual=0x000000000a22fa20, target_count=0
>>> (XEN)      target0=0000000000000000, target1=0000000000000000
>>> (XEN)      target2=0000000000000000, target3=0000000000000000
>>> (XEN) RSP = 0x0000000000000080 (0x0000000000000080)  RIP =
>>> 0x000000000000002a
>>> (0x000000000000002a)
>>> (XEN) RFLAGS=0x0000000000023202 (0x0000000000023202)  DR7 =
>>> 0x0000000000000400
>>> (XEN) Sysenter RSP=0000000000000000 CS:RIP=0000:0000000000000000
>>> (XEN) CS: sel=0x0060, attr=0x0c09b, limit=0xffffffff,
>>> base=0x0000000000200000
>>> (XEN) DS: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>> base=0x0000000000200000
>>> (XEN) SS: sel=0x0070, attr=0x0c093, limit=0xfc000fff,
>>> base=0x000000000020ba62
>>> (XEN) ES: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>> base=0x0000000000200000
>>> (XEN) FS: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>> base=0x0000000000200000
>>> (XEN) GS: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>> base=0x0000000000200000
>>> (XEN) GDTR:                           limit=0x00001dd8,
>>> base=0x0000000000200000
>>> (XEN) LDTR: sel=0x0000, attr=0x1c000, limit=0xffffffff,
>>> base=0x0000000000000000
>>> (XEN) IDTR:                           limit=0x00000188,
>>> base=0x0000000000201df0
>>> (XEN) TR: sel=0x0058, attr=0x0008b, limit=0x0000ffff,
>>> base=0x0000000000201ff2
>>> (XEN) Guest PAT = 0x0000000000000000
>>> (XEN) TSC Offset = ffffffe2242f086e
>>> (XEN) DebugCtl=0000000000000000 DebugExceptions=0000000000000000
>>> (XEN) Interruptibility=0001 ActivityState=0000
>>> (XEN) *** Host State ***
>>> (XEN) RSP = 0xffff83007e4f7fa0  RIP = 0xffff828c8019aa20
>>> (XEN) CS=e008 DS=0000 ES=0000 FS=0000 GS=0000 SS=0000 TR=e040
>>> (XEN) FSBase=0000000000000000 GSBase=0000000000000000
>>> TRBase=ffff828c802a8b00
>>> (XEN) GDTBase=ffff83007e9a3000 IDTBase=ffff83007e62e010
>>> (XEN) CR0=0000000080050033 CR3=000000007cfd8000 CR4=00000000000026f0
>>> (XEN) Sysenter RSP=ffff83007e4f7fd0 CS:RIP=e008:ffff828c801c7290
>>> (XEN) Host PAT = 0x0000000000000000
>>> (XEN) *** Control State ***
>>> (XEN) PinBased=0000003f CPUBased=b6a1e7fe SecondaryExec=00000041
>>> (XEN) EntryControls=000011ff ExitControls=0003efff
>>> (XEN) ExceptionBitmap=00044080
>>> (XEN) VMEntry: intr_info=80000b0b errcode=00001eac ilen=00000000
>>> (XEN) VMExit: intr_info=00000000 errcode=00008000 ilen=00000000
>>> (XEN)         reason=80000021 qualification=00000000
>>> (XEN) IDTVectoring: info=00000000 errcode=00000000
>>> (XEN) TPR Threshold = 0x00
>>> (XEN) EPT pointer = 0x0000000000000000
>>> (XEN) Virtual processor ID = 0x0000
>>> (XEN) **************************************
>>> (XEN) domain_crash called from vmx.c:2218
>>> (XEN) Domain 1 (vcpu#0) crashed on cpu#1:
>>> (XEN) ----[ Xen-3.4.0-rc3-pre  x86_64  debug=n  Not tainted ]----
>>> (XEN) CPU:    1
>>> (XEN) RIP:    0060:[<000000000000002a>]
>>> (XEN) RFLAGS: 0000000000023202   CONTEXT: hvm guest
>>> (XEN) rax: 0000000000000007   rbx: 0000000000001490   rcx: 0000000000000000
>>> (XEN) rdx: 0000000000001da8   rsi: 0000000000000000   rdi: 0000000000000000
>>> (XEN) rbp: 0000000000008ebf   rsp: 0000000000000080   r8:  0000000000000000
>>> (XEN) r9:  0000000000000000   r10: 0000000000000000   r11: 0000000000000000
>>> (XEN) r12: 0000000000000000   r13: 0000000000000000   r14: 0000000000000000
>>> (XEN) r15: 0000000000000000   cr0: 0000000080000019   cr4: 0000000000000000
>>> (XEN) cr3: 0000000001443000   cr2: 0000000000000000
>>> (XEN) ds: 0068   es: 0068   fs: 0068   gs: 0068   ss: 0070   cs: 0060
>>> 
>>> 
>>> Any ideas?
>>> 
>>> Tom
>>> 
>>> 
>>> 2009/4/22 Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
>>>> That should do it.
>>>> 
>>>>  K.
>>>> 
>>>> 
>>>> On 22/04/2009 15:04, "Tom Rotenberg" <tom.rotenberg@xxxxxxxxx> wrote:
>>>> 
>>>>> Keir,
>>>>> Just to make sure, i am using the following patch, in order to disable the
>>>>> vm86 acceleration:
>>>>> 
>>>>> diff -r cdc044f665dc xen/arch/x86/hvm/vmx/vmx.c
>>>>> --- a/xen/arch/x86/hvm/vmx/vmx.c    Wed Apr 22 11:26:37 2009 +0100
>>>>> +++ b/xen/arch/x86/hvm/vmx/vmx.c    Wed Apr 22 17:03:20 2009 +0300
>>>>> @@ -829,7 +829,7 @@
>>>>>         
>>>>>          if ( seg == x86_seg_tr )
>>>>>          {
>>>>> -            if ( v->domain->arch.hvm_domain.params[HVM_PARAM_VM86_TSS] )
>>>>> +            if (0)
>>>>>              {
>>>>>                  sel = 0;
>>>>>                  attr = vm86_tr_attr;
>>>>> 
>>>>> Is this OK?
>>>>> 
>>>>> Tom
>>>>> 
>>>>> 2009/4/22 Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
>>>>>> Yes, the safest way to be sure is probably to replace the if() statement
>>>>>> in
>>>>>> vmx_set_segment_register() that tests HVM_PARAM_VM86_TSS with if(0). That
>>>>>> is
>>>>>> the only place in Xen that checks HVM_PARAM_VM86_TSS. Then you can
>>>>>> re-build/install Xen and be sure that vm86 accel must be disabled.
>>>>>> 
>>>>>>  -- Keir
>>>>>> 
>>>>>> On 22/04/2009 14:52, "Tom Rotenberg" <tom.rotenberg@xxxxxxxxx> wrote:
>>>>>> 
>>>>>>> So, do u suggest, that i will set HVM_PARAM_VM86_TSS to 0, and re-check
>>>>>>> it?
>>>>>>> 
>>>>>>> 2009/4/22 Tim Deegan <Tim.Deegan@xxxxxxxxxx>
>>>>>>>> At 14:34 +0100 on 22 Apr (1240410866), Keir Fraser wrote:
>>>>>>>>> It could be an issue with the vm86 acceleration, possibly. I'm pretty
>>>>>>>>> sure
>>>>>>>>> the guest would have to IRET from protected mode to enter vm86 mode
>>>>>>>>> itself,
>>>>>>>>> and we don't emulate that. Tim: what would we need to do to disable
>>>>>>>>> the
>>>>>>>>> vm86
>>>>>>>>> acceleration for testing purposes? You suggested not setting VM86_TSS
>>>>>>>>> param
>>>>>>>>> from hvmloader, but I couldn't convince myself what effect that would
>>>>>>>>> actually have as the logic in Xen is non-trivial.
>>>>>>>> 
>>>>>>>> Yes; if HVM_PARAM_VM86_TSS is zero, vmx_set_segment_register() will
>>>>>>>> always set the tss bit in the bitmap of segments that aren't safe to
>>>>>>>> enter VM86 with.
>>>>>>>> 
>>>>>>>> Tim.
>>>>>>>> 
>>>>>>>>> 
>>>>>>>>>  -- Keir
>>>>>>>>> 
>>>>>>>>> On 22/04/2009 14:23, "Tom Rotenberg" <tom.rotenberg@xxxxxxxxx> wrote:
>>>>>>>>> 
>>>>>>>>>> Tim,
>>>>>>>>>> 
>>>>>>>>>> so what does it mean? could it be that we have a bug in the real mode
>>>>>>>>>> emulation, which causes the segment state to be invalid (maybe it's
>>>>>>>>>> because
>>>>>>>>>> of
>>>>>>>>>> a bug in the patch that Keir made for me, which emulated the LLDT,
>>>>>>>>>> and
>>>>>>>>>> the
>>>>>>>>>> LTR
>>>>>>>>>> instructions)?
>>>>>>>>>> 
>>>>>>>>>> Keir suggested to trace back where the problem (segment state)
>>>>>>>>>> occured,
>>>>>>>>>> and
>>>>>>>>>> from there to try and find the bug which caused it. Do u have any
>>>>>>>>>> better
>>>>>>>>>> suggestion for solving this?
>>>>>>>>>> 
>>>>>>>>>> Tom
>>>>>>>>>> 
>>>>>>>>>> 2009/4/22 Tim Deegan <Tim.Deegan@xxxxxxxxxx>
>>>>>>>>>>> At 13:39 +0100 on 22 Apr (1240407546), Tom Rotenberg wrote:
>>>>>>>>>>>> Keir,
>>>>>>>>>>>> 
>>>>>>>>>>>> I have tried your latest patch, and it looks like now it passes the
>>>>>>>>>>>> emulation problem. However,  now the domain crashes with the
>>>>>>>>>>>> following
>>>>>>>>>>>> error:
>>>>>>>>>>>> 
>>>>>>>>>>>> (XEN) HVM1: Booting from 0000:7c00
>>>>>>>>>>>> (XEN) Failed vm entry (exit reason 0x80000021) caused by invalid
>>>>>>>>>>>> guest
>>>>>>>>>>>> state
>>>>>>>>>>>> (0).
>>>>>>>>>>>> (XEN) ************* VMCS Area **************
>>>>>>>>>>>> (XEN) *** Guest State ***
>>>>>>>>>>>> (XEN) CR0: actual=0x0000000080010039, shadow=0x0000000080000019,
>>>>>>>>>>>> gh_mask=ffffffffffffffff
>>>>>>>>>>>> (XEN) CR4: actual=0x0000000000002060, shadow=0x0000000000000000,
>>>>>>>>>>>> gh_mask=ffffffffffffffff
>>>>>>>>>>>> (XEN) CR3: actual=0x000000000a213a20, target_count=0
>>>>>>>>>>>> (XEN)      target0=0000000000000000, target1=0000000000000000
>>>>>>>>>>>> (XEN)      target2=0000000000000000, target3=0000000000000000
>>>>>>>>>>>> (XEN) RSP = 0x0000000000000080 (0x0000000000000080)  RIP =
>>>>>>>>>>>> 0x000000000000002a (0x000000000000002a)
>>>>>>>>>>>> (XEN) RFLAGS=0x0000000000023202 (0x0000000000023202)  DR7 =
>>>>>>>>>>>> 0x0000000000000400
>>>>>>>>>>> 
>>>>>>>>>>> Looks like we're trying to VMENTER in virtual 8086 mode but without
>>>>>>>>>>> tidying up the segment state.  That could either be the guest
>>>>>>>>>>> entering
>>>>>>>>>>> virtual 8086 mode itself or Xen entering vitrual 8086 mode to
>>>>>>>>>>> emulate
>>>>>>>>>>> real mode, but Xen is always careful to make the segment state agree
>>>>>>>>>>> with Intel's rather strict requrements when it does that.
>>>>>>>>>>> 
>>>>>>>>>>> Tim.
>>>>>>>>>>> 
>>>>>>>>>>> 
>>>>>>>>>>>> (XEN) Sysenter RSP=0000000000000000 CS:RIP=0000:0000000000000000
>>>>>>>>>>>> (XEN) CS: sel=0x0060, attr=0x0c09b, limit=0xffffffff,
>>>>>>>>>>>> base=0x0000000000200000
>>>>>>>>>>>> (XEN) DS: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>>>>>>>>>>> base=0x0000000000200000
>>>>>>>>>>>> (XEN) SS: sel=0x0070, attr=0x0c093, limit=0xfc000fff,
>>>>>>>>>>>> base=0x000000000020ba62
>>>>>>>>>>>> (XEN) ES: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>>>>>>>>>>> base=0x0000000000200000
>>>>>>>>>>>> (XEN) FS: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>>>>>>>>>>> base=0x0000000000200000
>>>>>>>>>>>> (XEN) GS: sel=0x0068, attr=0x0c093, limit=0xffffffff,
>>>>>>>>>>>> base=0x0000000000200000
>>>>>>>>>>>> (XEN) GDTR:                           limit=0x00001dd8,
>>>>>>>>>>>> base=0x0000000000200000
>>>>>>>>>>>> (XEN) LDTR: sel=0x0000, attr=0x1c000, limit=0xffffffff,
>>>>>>>>>>>> base=0x0000000000000000
>>>>>>>>>>>> (XEN) IDTR:                           limit=0x00000188,
>>>>>>>>>>>> base=0x0000000000201df0
>>>>>>>>>>>> (XEN) TR: sel=0x0058, attr=0x0008b, limit=0x0000ffff,
>>>>>>>>>>>> base=0x0000000000201ff2
>>>>>>>>>>>> (XEN) Guest PAT = 0x0000000000000000
>>>>>>>>>>>> (XEN) TSC Offset = ffffffe4920110b7
>>>>>>>>>>>> (XEN) DebugCtl=0000000000000000 DebugExceptions=0000000000000000
>>>>>>>>>>>> (XEN) Interruptibility=0001 ActivityState=0000
>>>>>>>>>>>> (XEN) *** Host State ***
>>>>>>>>>>>> (XEN) RSP = 0xffff83007e4f7fa0  RIP = 0xffff828c8019aa20
>>>>>>>>>>>> (XEN) CS=e008 DS=0000 ES=0000 FS=0000 GS=0000 SS=0000 TR=e040
>>>>>>>>>>>> (XEN) FSBase=0000000000000000 GSBase=0000000000000000
>>>>>>>>>>>> TRBase=ffff828c802a8b00
>>>>>>>>>>>> (XEN) GDTBase=ffff83007e9a3000 IDTBase=ffff83007e62e010
>>>>>>>>>>>> (XEN) CR0=0000000080050033 CR3=000000007cfdc000
>>>>>>>>>>>> CR4=00000000000026f0
>>>>>>>>>>>> (XEN) Sysenter RSP=ffff83007e4f7fd0 CS:RIP=e008:ffff828c801c7290
>>>>>>>>>>>> (XEN) Host PAT = 0x0000000000000000
>>>>>>>>>>>> (XEN) *** Control State ***
>>>>>>>>>>>> (XEN) PinBased=0000003f CPUBased=b6a1e7fe SecondaryExec=00000041
>>>>>>>>>>>> (XEN) EntryControls=000011ff ExitControls=0003efff
>>>>>>>>>>>> (XEN) ExceptionBitmap=00044080
>>>>>>>>>>>> (XEN) VMEntry: intr_info=80000b0b errcode=00001eac ilen=00000000
>>>>>>>>>>>> (XEN) VMExit: intr_info=00000000 errcode=00008000 ilen=00000000
>>>>>>>>>>>> (XEN)         reason=80000021 qualification=00000000
>>>>>>>>>>>> (XEN) IDTVectoring: info=00000000 errcode=00000000
>>>>>>>>>>>> (XEN) TPR Threshold = 0x00
>>>>>>>>>>>> (XEN) EPT pointer = 0x0000000000000000
>>>>>>>>>>>> (XEN) Virtual processor ID = 0x0000
>>>>>>>>>>>> (XEN) **************************************
>>>>>>>>>>>> (XEN) domain_crash called from vmx.c:2218
>>>>>>>>>>>> (XEN) Domain 1 (vcpu#0) crashed on cpu#1:
>>>>>>>>>>>> (XEN) ----[ Xen-3.4.0-rc3-pre  x86_64  debug=n  Not tainted ]----
>>>>>>>>>>>> (XEN) CPU:    1
>>>>>>>>>>>> (XEN) RIP:    0060:[<000000000000002a>]
>>>>>>>>>>>> (XEN) RFLAGS: 0000000000023202   CONTEXT: hvm guest
>>>>>>>>>>>> (XEN) rax: 0000000000000007   rbx: 0000000000001490   rcx:
>>>>>>>>>>>> 0000000000000000
>>>>>>>>>>>> (XEN) rdx: 0000000000001da8   rsi: 0000000000000000   rdi:
>>>>>>>>>>>> 0000000000000000
>>>>>>>>>>>> (XEN) rbp: 0000000000008ebf   rsp: 0000000000000080   r8:
>>>>>>>>>>>>  0000000000000000
>>>>>>>>>>>> (XEN) r9:  0000000000000000   r10: 0000000000000000   r11:
>>>>>>>>>>>> 0000000000000000
>>>>>>>>>>>> (XEN) r12: 0000000000000000   r13: 0000000000000000   r14:
>>>>>>>>>>>> 0000000000000000
>>>>>>>>>>>> (XEN) r15: 0000000000000000   cr0: 0000000080000019   cr4:
>>>>>>>>>>>> 0000000000000000
>>>>>>>>>>>> (XEN) cr3: 0000000001443000   cr2: 0000000000000000
>>>>>>>>>>>> (XEN) ds: 0068   es: 0068   fs: 0068   gs: 0068   ss: 0070   cs:
>>>>>>>>>>>> 0060
>>>>>>>>>>>> 
>>>>>>>>>>>> Could it be, that the real mode emulation code has a bug? What does
>>>>>>>>>>>> this
>>>>>>>>>>>> error means?
>>>>>>>>>>>> 
>>>>>>>>>>>> Tom
>>>>>>>>>>>> 
>>>>>>>>>>>> 2009/4/22 Keir Fraser
>>>>>>>>>>>> <keir.fraser@xxxxxxxxxxxxx<mailto:keir.fraser@xxxxxxxxxxxxx>>
>>>>>>>>>>>> On 22/04/2009 12:18, "Tom Rotenberg"
>>>>>>>>>>>> <tom.rotenberg@xxxxxxxxx<mailto:tom.rotenberg@xxxxxxxxx>> wrote:
>>>>>>>>>>>> 
>>>>>>>>>>>>> Keir,
>>>>>>>>>>>>> 
>>>>>>>>>>>>> I have applied your patch, and it seemed to work. However, the
>>>>>>>>>>>>> domain
>>>>>>>>>>>>> still
>>>>>>>>>>>>> crashes, and now it looks like it's because of the 'LTR'
>>>>>>>>>>>>> instruction.
>>>>>>>>>>>> 
>>>>>>>>>>>> Try the attached patch. It replaces the one I sent last time, and
>>>>>>>>>>>> emulates
>>>>>>>>>>>> both LLDT and LTR.
>>>>>>>>>>>> 
>>>>>>>>>>>>  -- Keir
>>>>>>>>>>>> 
>>>>>>>>>>> 
>>>>>>>>>>> Content-Description: ATT00001.txt
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> Xen-devel mailing list
>>>>>>>>>>>> Xen-devel@xxxxxxxxxxxxxxxxxxx
>>>>>>>>>>>> http://lists.xensource.com/xen-devel
>>>>>>>>>>> 
>>>>>>>>>>> 
>>>>>>>>>>> --
>>>>>>>>>>> Tim Deegan <Tim.Deegan@xxxxxxxxxx>
>>>>>>>>>>> Principal Software Engineer, Citrix Systems (R&D) Ltd.
>>>>>>>>>>> [Company #02300071, SL9 0DZ, UK.]
>>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>> 
>>>>>>>> --
>>>>>>>> Tim Deegan <Tim.Deegan@xxxxxxxxxx>
>>>>>>>> Principal Software Engineer, Citrix Systems (R&D) Ltd.
>>>>>>>> [Company #02300071, SL9 0DZ, UK.]
>>>>>>> 
>>>>>> 
>>>>>> 
>>>>> 
>>>> 
>>>> 
>>> 
>> 
>> 
> 



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel

 

©2013 Xen Project, A Linux Foundation Collaborative Project. All Rights Reserved.
Linux Foundation is a registered trademark of The Linux Foundation.
Xen Project is a trademark of The Linux Foundation.

Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.