[RFC/RFT PATCH 0/3] arm64: KVM: work around incoherency with uncached guest mappings

[Catalin Marinas]

On Wed, Mar 04, 2015 at 06:03:11PM +0100, Paolo Bonzini wrote:
> 
> 
> On 04/03/2015 15:29, Catalin Marinas wrote:
> > I disagree it is 100% a host-side issue. It is a host-side issue _if_
> > the host tells the guest that the (virtual) device is non-coherent (or,
> > more precisely, it does not explicitly tell the guest that the device is
> > coherent). If the guest thinks the (virtual) device is non-coherent
> > because of information passed by the host, I fully agree that the host
> > needs to manage the cache coherency.
> > 
> > However, the host could also pass a "dma-coherent" property in the DT
> > given to the guest and avoid any form of cache maintenance. If the guest
> > does not honour such coherency property, it's a guest problem and it
> > needs fixing in the guest. This isn't any different from a real physical
> > device behaviour.
> 
> Can you add that property to the device tree for PCI devices too?

Yes but not with mainline yet:

http://thread.gmane.org/gmane.linux.kernel.iommu/8935

We can add the property at the PCI host bridge level (with the drawback
that it covers all the PCI devices), like here:

Documentation/devicetree/bindings/pci/xgene-pci.txt

-- 
Catalin