Re: [Xen-devel] Xenalyze questions

[George Dunlap <George.Dunlap@xxxxxxxxxxxxx>]

On Fri, Jun 7, 2013 at 10:14 AM, Simon Graham <simon.graham@xxxxxxxxxx> wrote:
>> > init_pcpus: through first trace write, done for now.
>> > hvm_generic_postprocess_init: Strange, h->postprocess set!
>>
>> Hmm, it looks like this is the problem here.
>>
>> Can you try adding this patch, and then running the following?
>>
>> xenalyze -a -s --tolerance=1 [filename] > [filename.dump]?
>>
>> The resulting file may be pretty big; I'll just need the last 50 or so lines.
>>
>
> Actually it was pretty small - must be the 1st record that is bad!
>
> Tolerating errors at or below 1
> scan_for_new_pcpu: Activating pcpu 0 at offset 0
> Creating vcpu 0 for dom 32768
> scan_for_new_pcpu: Activating pcpu 1 at offset 1560
> Creating vcpu 1 for dom 32768
> scan_for_new_pcpu: Activating pcpu 2 at offset 1772
> Creating vcpu 2 for dom 32768
> scan_for_new_pcpu: Activating pcpu 3 at offset 68428
> Creating vcpu 3 for dom 32768
> scan_for_new_pcpu: Activating pcpu 5 at offset 68520
> Creating vcpu 5 for dom 32768
> scan_for_new_pcpu: Activating pcpu 10 at offset 73632
> Creating vcpu 10 for dom 32768
> scan_for_new_pcpu: Activating pcpu 11 at offset 90588
> Creating vcpu 11 for dom 32768
> init_pcpus: through first trace write, done for now.
> ]               .... .    x. d32768v10 vmexit exit_reason EXCEPTION_NMI eip 
> fffff88005a25ad0
> ]               .... .    x. d32768v10 fast mmio va fffff88008808df0
> ]               .... .    x. d32768v10 cpuid [ 1 206c2 400800 83b82203 
> 178bfbff ]
> hvm_generic_postprocess_init: Strange, h->postprocess set!

...I'm trying to figure out how on earth you got a trace like that.
It looks like:
1. The VM took a page fault (EXCEPTION_NMI)
2. Xen determined that it was an MMIO, which would mean that the guest
PT was valid, but pointed to PFN space that Xen didn't recognize
3. It somehow emulated a CPUID instruction???

CPUID doesn't do any memory accesses, so it shouldn't be able to
trigger an MMIO fault like this.

Do you have any way of telling what instruction was at the address at
the EIP (fffff88005a25ad0)?

Also, could you run xenalyze as follows, and attach the first couple
hundred lines:

xenalyze -a [filename] > [filename.dump]

(Not having the -s will remove the path that is crashing, to allow us
to see what happened after this trace.)

 -George

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel