[libvirt] [RFC 4/6] conf: Expose GIC capabilities

[Andrea Bolognani]

On Wed, 2016-03-30 at 16:10 -0400, John Ferlan wrote:
> > diff --git a/tests/domaincapsschemadata/domaincaps-qemu_1.6.50-1.xml b/tests/domaincapsschemadata/domaincaps-
> > qemu_1.6.50-1.xml
> > index 37d2102..990661b 100644
> > --- a/tests/domaincapsschemadata/domaincaps-qemu_1.6.50-1.xml
> > +++ b/tests/domaincapsschemadata/domaincaps-qemu_1.6.50-1.xml
> > @@ -56,4 +56,7 @@
> >        </enum>
> >      </hostdev>
> >    </devices>
> > +  <features>
> > +    <gic supported='no'/>
> > +  </features>
> >  </domainCapabilities>
> 
> I first wondered why the XML file was modified, but later understood -
> sort of...

We will need more tests for the new feature, of course.

I wanted to gather feedback on the interface with the RFC
before I spent too much time testing or documenting it.

> Still not fully clear on how what this schema output should
> look like... For prior to 2.6 should it even be printed?  For 2.6 and
> beyond what would it look like?

There might be a case for not printing it, but we don't
perform such optimization for the rest of the information
that's part of domain capabilities, so I'm not sure
whether we should bother in this case.

Versions prior to 2.6, and 2.6 itself on non-ARM
architectures would yield

  <features>
    <gic supported='no'/>
  </features>

whereas 2.6 onward on ARM would yield something like

  <features>
    <gic supported='yes'>
      <enum name='version'>
        <value>2</value>
      </enum>
    </gic>
  </features>

which mirrors exactly the type of <gic> element the
user can expect to be able to use inside the <feature>
element in the domain XML, eg.

  <features>
    <gic version='2'/>
  </features>

or none at all.

Cheers.

-- 
Andrea Bolognani
Software Engineer - Virtualization Team