vdpa: Fix memory listener deletions of iova tree

vhost_vdpa_listener_region_del is always deleting the first iova entry
of the tree, since it's using the needle iova instead of the result's
one.

This was detected using a vga virtual device in the VM using vdpa SVQ.
It makes some extra memory adding and deleting, so the wrong one was
mapped / unmapped. This was undetected before since all the memory was
mappend and unmapped totally without that device, but other conditions
could trigger it too:

* mem_region was with .iova = 0, .translated_addr = (correct GPA).
* iova_tree_find_iova returned right result, but does not update
  mem_region.
* iova_tree_remove always removed region with .iova = 0. Right iova were
  sent to the device.
* Next map will fill the first region with .iova = 0, causing a mapping
  with the same iova and device complains, if the next action is a map.
* Next unmap will cause to try to unmap again iova = 0, causing the
  device to complain that no region was mapped at iova = 0.

Fixes: 34e3c94 ("vdpa: Add custom IOTLB translations to SVQ")
Reported-by: Lei Yang <[email protected]>
Signed-off-by: Eugenio Pérez <[email protected]>
Signed-off-by: Jason Wang <[email protected]>

hw/virtio/vhost-vdpa.c

@@ -290,7 +290,7 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
 
         result = vhost_iova_tree_find_iova(v->iova_tree, &mem_region);
         iova = result->iova;
-        vhost_iova_tree_remove(v->iova_tree, &mem_region);
+        vhost_iova_tree_remove(v->iova_tree, result);
     }
     vhost_vdpa_iotlb_batch_begin_once(v);
     ret = vhost_vdpa_dma_unmap(v, iova, int128_get64(llsize));