On Wed, 11 Mar 2015 10:49:41 -0300
Eduardo Habkost <ehabk...@redhat.com> wrote:
> On Wed, Mar 11, 2015 at 02:36:08PM +0100, Igor Mammedov wrote:
> [...]
> > > - cpu = cpu_x86_create(cpu_model, icc_bridge, &local_err);
> > > + if (icc_bridge == NULL) {
> > turn it into assert(icc_bridge != NULL)
> > that should never be NULL or we have a codding error somewhere in code.
>
> See changelog and the v1 discussion:
Ok, I see
Since it's just code movement it's fine as well.
We could do assert and object_unref changes on top
so it would be easier to review.
> > > Changes v1 -> v2:
> > > * Keep existing check for NULL icc_bridge and error reporting, instead
> > > of assing assert(icc_bridge)
>
> (I just noticed the typo, and it was _not_ intentional :)
>
