Bug 1029732

My understanding is that it *can't* be changed live (i.e. without hot-unplugging the device and plugging in a new device). The plumbing for a macvtap interface is more or less set in stone at the time the device is created.

For this reason, I think the only reasonable thing to do here is to log an error if someone tries to make a live change to the number of queues.

Fix pushed upstream:

commit 5e12641e0ea49fa1f0b4338f999ec63f41efec4c
Author: Laine Stump <laine>
Date:   Tue Dec 3 12:40:38 2013 +0200

    qemu: report error on attempt to live change virtio-net queues

Verify it as follows. The result is expected. 


Version:
libvirt-1.2.8-6.el7.x86_64
qemu-kvm-rhev-2.1.2-8.el7.x86_64
qemu-kvm-1.5.3-79.el7.x86_64


Steps:
# virsh start rhel7.0
Domain rhel7.0 started

# virsh dumpxml rhel7.0 |grep -aA8 interface
    <interface type='network'>
      <mac address='52:54:00:db:0c:00'/>
      <source network='default'/>
      <target dev='vnet0'/>
      <model type='virtio'/>
      <driver name='vhost' queues='2'/>
      <alias name='net0'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
    </interface>
...

# cat nic.xml 
    <interface type='network'>
      <mac address='52:54:00:db:0c:00'/>
      <source network='default'/>
      <target dev='vnet0'/>
      <model type='virtio'/>
      <driver name='vhost' queues='4'/>
      <alias name='net0'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
    </interface>

# virsh update-device rhel7.0 nic.xml --current
error: Failed to update device from nic.xml
error: Operation not supported: cannot modify virtio network device driver attributes

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHSA-2015-0323.html

The domain is at running, can not update virtio-net queues living. The libvirt code as following:

  if (virDomainNetIsVirtioModel(olddev) &&
        (olddev->driver.virtio.name != newdev->driver.virtio.name ||
         olddev->driver.virtio.txmode != newdev->driver.virtio.txmode ||
         olddev->driver.virtio.ioeventfd != newdev->driver.virtio.ioeventfd ||
         olddev->driver.virtio.event_idx != newdev->driver.virtio.event_idx ||

         olddev->driver.virtio.queues != newdev->driver.virtio.queues ||

         olddev->driver.virtio.rx_queue_size != newdev->driver.virtio.rx_queue_size ||
         olddev->driver.virtio.tx_queue_size != newdev->driver.virtio.tx_queue_size ||
         olddev->driver.virtio.host.csum != newdev->driver.virtio.host.csum ||
         olddev->driver.virtio.host.gso != newdev->driver.virtio.host.gso ||
         olddev->driver.virtio.host.tso4 != newdev->driver.virtio.host.tso4 ||
         olddev->driver.virtio.host.tso6 != newdev->driver.virtio.host.tso6 ||
         olddev->driver.virtio.host.ecn != newdev->driver.virtio.host.ecn ||
         olddev->driver.virtio.host.ufo != newdev->driver.virtio.host.ufo ||
         olddev->driver.virtio.host.mrg_rxbuf != newdev->driver.virtio.host.mrg_rxbuf ||
         olddev->driver.virtio.guest.csum != newdev->driver.virtio.guest.csum ||
         olddev->driver.virtio.guest.tso4 != newdev->driver.virtio.guest.tso4 ||
         olddev->driver.virtio.guest.tso6 != newdev->driver.virtio.guest.tso6 ||
         olddev->driver.virtio.guest.ecn != newdev->driver.virtio.guest.ecn ||
         olddev->driver.virtio.guest.ufo != newdev->driver.virtio.guest.ufo)) {
        virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s",
                       _("cannot modify virtio network device driver attributes"));
        goto cleanup;