vhost: force vhost off for non-MSI guests
When MSI is off, each interrupt needs to be bounced through the io
thread when it's set/cleared, so vhost-net causes more context switches and
higher CPU utilization than userspace virtio which handles networking in
the same thread.
We'll need to fix this by adding level irq support in kvm irqfd,
for now disable vhost-net in these configurations.
Added a vhostforce flag to force vhost-net back on.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
diff --git a/hw/virtio-pci.c b/hw/virtio-pci.c
index d07ff97..3911b09 100644
--- a/hw/virtio-pci.c
+++ b/hw/virtio-pci.c
@@ -595,6 +595,12 @@
return 0;
}
+static bool virtio_pci_query_guest_notifiers(void *opaque)
+{
+ VirtIOPCIProxy *proxy = opaque;
+ return msix_enabled(&proxy->pci_dev);
+}
+
static int virtio_pci_set_guest_notifiers(void *opaque, bool assign)
{
VirtIOPCIProxy *proxy = opaque;
@@ -658,6 +664,7 @@
.save_queue = virtio_pci_save_queue,
.load_queue = virtio_pci_load_queue,
.get_features = virtio_pci_get_features,
+ .query_guest_notifiers = virtio_pci_query_guest_notifiers,
.set_host_notifier = virtio_pci_set_host_notifier,
.set_guest_notifiers = virtio_pci_set_guest_notifiers,
.vmstate_change = virtio_pci_vmstate_change,
