Re: [Qemu-devel] [PATCH 6/6] spec/vhost-user spec: Add IOMMU support

[Michael S. Tsirkin]

On Fri, May 19, 2017 at 03:46:36PM +0800, Jason Wang wrote:
> 
> 
> On 2017年05月18日 16:43, Maxime Coquelin wrote:
> > > > 
> > > > +When the VHOST_USER_PROTOCOL_F_SLAVE_REQ is supported by the
> > > > slave, and the
> > > > +master initiated the slave to master communication channel using the
> > > > +VHOST_USER_SET_SLAVE_REQ_FD request, the slave can send IOTLB
> > > > miss and access
> > > > +failure events by sending VHOST_USER_SLAVE_IOTLB_MSG requests
> > > > to the master
> > > > +with a struct vhost_iotlb_msg payload. For miss events, the
> > > > iotlb payload has
> > > > +to be filled with the miss message type (1), the I/O virtual
> > > > address and the
> > > > +permissions flags. For access failure event, the iotlb payload
> > > > has to be
> > > > +filled with the access failure message type (4), the I/O
> > > > virtual address and
> > > > +the permissions flags.
> > > 
> > > I don't think slave should cache invalid entries. If it does not,
> > > how can it detect access failure as opposed to a miss?
> > 
> > Of course, invalid cache entries should not be cached.
> > The VHOST_IOTLB_ACCESS_FAIL has been specified for the Kernel backend,
> > even if the latter does not implement it yet.
> 
> Yes, I leave this for future use e.g reporting copy_to_user() failure to
> userspace.
> 
> Thanks

Interesting. And it's not handled now.
So let's add a text "reserved for reporting internal access
errors in the future. Should not be used for now.".

-- 
MST