On Tue, Jul 11, 2023 at 10:32:31AM +0200, David Hildenbrand wrote:
> On 10.07.23 23:40, Michael S. Tsirkin wrote:
> > > @@ -2855,12 +2796,11 @@ static void
> > > virt_machine_device_plug_cb(HotplugHandler *hotplug_dev,
> > > SYS_BUS_DEVICE(dev));
> > > }
> > > }
> > > +
> > > if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
> > > virt_memory_plug(hotplug_dev, dev, errp);
> > > - }
> > > -
> > > - if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {
> > > - virt_virtio_md_pci_plug(hotplug_dev, dev, errp);
> > > + } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_PCI)) {
> > > + virtio_md_pci_plug(VIRTIO_MD_PCI(dev), MACHINE(hotplug_dev),
> > > errp);
> > > }
> > > if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {
> >
> >
> > How is this supposed to link if virtio-md is disabled at compile time?
> >
>
> Good point.
>
> The old code unconditionally enabled MEM_DEVICE, so we never required subs
> for that.
>
> We either need stubs or have to wrap this in #ifdef.
>
> Stubs sound cleaner.
That is what we usually do, yes.
--
MST
