[PATCH] hw/virtio/vhost: check nvqs at dev_start

[Albert Esteve]

While this is not expected to happen, it could still
be that a vhost_dev did not set its nvqs member.

Since `vhost_dev_start` access the device's vqs array
later without checking its size, it would cause a
Segmentation fault when nvqs is 0.

To avoid this `rare` case and made the code safer,
add a clause that ensures nvqs has been set, and
warn the user if it has not.

Signed-off-by: Albert Esteve <aesteve@redhat.com>
---
 hw/virtio/vhost.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index e2f6ffb446..78805fe5b7 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -1935,6 +1935,11 @@ int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice 
*vdev, bool vrings)
     hdev->started = true;
     hdev->vdev = vdev;
 
+    if (!hdev->nvqs) {
+        error_report("device nvqs not set");
+        goto fail_nvqs;
+    }
+
     r = vhost_dev_set_features(hdev, hdev->log_enabled);
     if (r < 0) {
         goto fail_features;
@@ -2028,6 +2033,7 @@ fail_mem:
     if (vhost_dev_has_iommu(hdev)) {
         memory_listener_unregister(&hdev->iommu_listener);
     }
+fail_nvqs:
 fail_features:
     vdev->vhost_started = false;
     hdev->started = false;
-- 
2.41.0