virtio-devices: Handle virtio queues interrupts from transport layer

Instead of relying on the virtio-queue crate to store the information
about the MSI-X vectors for each queue, we handle this directly from the
PCI transport layer.

This is the first step in getting closer to the upstream version of
virtio-queue so that we can eventually move fully to the upstream
version.

Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>

virtio-queue/src/state.rs

@@ -17,8 +17,8 @@ use vm_memory::{Address, Bytes, GuestAddress, GuestMemory};
 use crate::defs::{
     DEFAULT_AVAIL_RING_ADDR, DEFAULT_DESC_TABLE_ADDR, DEFAULT_USED_RING_ADDR,
     VIRTQ_AVAIL_ELEMENT_SIZE, VIRTQ_AVAIL_RING_HEADER_SIZE, VIRTQ_AVAIL_RING_META_SIZE,
-    VIRTQ_MSI_NO_VECTOR, VIRTQ_USED_ELEMENT_SIZE, VIRTQ_USED_F_NO_NOTIFY,
-    VIRTQ_USED_RING_HEADER_SIZE, VIRTQ_USED_RING_META_SIZE,
+    VIRTQ_USED_ELEMENT_SIZE, VIRTQ_USED_F_NO_NOTIFY, VIRTQ_USED_RING_HEADER_SIZE,
+    VIRTQ_USED_RING_META_SIZE,
 };
 use crate::{
     error, AccessPlatform, AvailIter, Descriptor, Error, QueueStateGuard, QueueStateT,
@@ -58,9 +58,6 @@ pub struct QueueState {
     /// Guest physical address of the used ring.
     pub used_ring: GuestAddress,
 
-    /// Interrupt vector
-    pub vector: u16,
-
     /// Access platform handler
     pub access_platform: Option<Arc<dyn AccessPlatform>>,
 }
@@ -200,7 +197,6 @@ impl QueueStateT for QueueState {
             next_used: Wrapping(0),
             event_idx_enabled: false,
             signalled_used: None,
-            vector: VIRTQ_MSI_NO_VECTOR,
             access_platform: None,
         }
     }
@@ -266,7 +262,6 @@ impl QueueStateT for QueueState {
         self.next_used = Wrapping(0);
         self.signalled_used = None;
         self.event_idx_enabled = false;
-        self.vector = VIRTQ_MSI_NO_VECTOR;
     }
 
     fn lock(&mut self) -> <Self as QueueStateGuard>::G {