Please modify vcpu model in guest xml to this: <cpu mode='custom' match='exact'> <model fallback='allow'>kvm64</model> </cpu> Then observe what happens. If the issue is reproduced, then please modify guest xml to change vcpus topology to this: <cpu mode='host-model'> <model fallback='allow'/> <topology sockets='1' cores='2' threads='1'/> </cpu> observe what happens.