Avi Kivity wrote:
> Jan Kiszka wrote:
>>>> Note that both SVM and VTX are supported, but only the latter was
>>>> tested
>>>> yet. Based on the experience with all those VTX corner case, I would be
>>>> fairly surprised if SVM will work out of the box.
>>>>
>>>>         
>>> I'd like svm to work before applying.
>>>     
>>
>> To validate the design?
>>
>>   
> 
> To make sure it doesn't oops, and to make sure it works.
> 
>> I will see if I can organize an SVM box, but I can't promise when I'll
>> be able to do the testing. Anyone willing to contribute time on this
>> would be warmly welcome!
>>
>>   
> 
> I happen to have an svm box.  How do you test this? play with gdb inside
> and outside the guest?

For inside the box, I also use the kgdb testsuite (kgbts).

For guest debugging check, some manual interaction is required: Set a
breakpoint on a prominent symbol (my favorite is sys_execve), do some
single stepping, check that smp works (breakpoint hits on all vcpus),
add some watchpoint and check if it triggers at reasonable spots (I
recently did a short demo with first trapping hrtimer_start, grabbing
&timer->state and then set a watchpoint on that address to catch changes).

> 
> You could also try qemu's svm emulation, but it may be a little slow,
> and probably hasn't been well tested in these areas.
> 

Yeah, could be a supplemental test.

Jan

-- 
Siemens AG, Corporate Technology, CT SE 26
Corporate Competence Center Embedded Linux
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to