Re: [Qemu-devel] [PATCH 28/35] kvm: x86: Introduce kvmclock device to save/restore its state

[Jan Kiszka]

On 2011-01-18 18:09, Anthony Liguori wrote:
> On 01/18/2011 10:56 AM, Jan Kiszka wrote:
>>
>>> The device model topology is 100% a hidden architectural detail.
>>>      
>> This is true for the sysbus, it is obviously not the case for PCI and
>> similarly discoverable buses. There we have a guest-explorable topology
>> that is currently equivalent to the the qdev layout.
>>    
> 
> But we also don't do PCI passthrough so we really haven't even explored 
> how that maps in qdev.  I don't know if qemu-kvm has attempted to 
> qdev-ify it.

It is. And even if it weren't or the current version in qemu-kvm was not
perfect, we need to consider those uses cases now as we are about to
define a generic model for kvm device integration. That's the point of
this discussion.

> 
>>>> Management and analysis tools must be able to traverse the system buses
>>>> and find guest devices this way.
>>>>        
>>> We need to provide a compatible interface to the guest.  If you agree
>>> with my above statements, then you'll also agree that we can do this
>>> without keeping the device model topology stable.
>>>
>>> But we also need to provide a compatible interface to management tools.
>>> Exposing the device model topology as a compatible interface
>>> artificially limits us.  It's far better to provide higher level
>>> supported interfaces to give us the flexibility to change the device
>>> model as we need to.
>>>      
>> How do you want to change qdev to keep the guest and management tool
>> view stable while branching off kvm sub-buses?
> 
> The qdev device model is not a stable interface.  I think that's been 
> clear from the very beginning.

Internals aren't stable, but they should only be changed for a good
reason, specifically when the change may impact the whole set of device
models.

> 
>>   Please propose such
>> extensions so that they can be discussed. IIUC, that would be second
>> relation between qdev and qbus instances besides the physical topology.
>> What further use cases (besides passing kvm_state around) do you have in
>> mind?
>>    
> 
> The -device interface is a stable interface.  Right now, you don't 
> specify any type of identifier of the pci bus when you create a PCI 
> device.  It's implied in the interface.

Which only works as along as we expose a single bus. You don't need to
be an oracle to predict that this is not a stable interface.

> 
>>    
>>>      
>>>>    If they create a device on bus X, it
>>>> must never end up on bus Y just because it happens to be KVM-assisted or
>>>> has some other property.
>>>>        
>>> Nope.  This is exactly what should happen.
>>>
>>> 90% of the devices in the device model are not created by management
>>> tools.  They're part of a chipset.  The chipset has well defined
>>> extension points and we provide management interfaces to create devices
>>> on those extension points.  That is, interfaces to create PCI devices.
>>>
>>>      
>> Creating kvm irqchips via the management tool would be one simple way
>> (not the only one, though) to enable/disable kvm assistance for those
>> devices.
>>    
> 
> It's automatically created as part of the CPUs or as part of the 
> chipset.  How to enable/disable kvm assistance is a property of the CPU 
> and/or chipset.

If we exclude creation via command line / config files, we could also
pass the kvm_state directly from the machine or chipset setup code and
save us at least the kvm system buses.

Jan

-- 
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux