A flaw was found in the KVM's AMD code for supporting SVM nested
virtualization. The flaw occurs when processing the VMCB (virtual machine
control block) provided by the L1 guest to spawn/handle a nested guest
(L2). Due to improper validation of the "int_ctl" field, this issue could
allow a malicious L1 to enable AVIC support (Advanced Virtual Interrupt
Controller) for the L2 guest. As a result, the L2 guest would be allowed to
read/write physical pages of the host, resulting in a crash of the entire
system, leak of sensitive data or potential guest-to-host escape. This flaw
affects Linux kernel versions prior to 5.14-rc7.

Maxim Levitsky discovered that the KVM hypervisor implementation for AMD
processors in the Linux kernel did not properly prevent a guest VM from
enabling AVIC in nested guest VMs. An attacker in a guest VM could use this
to write to portions of the host's physical memory.

Disable nested virtualisation when loading the KVM AMD module:

 modprobe kvm_amd nested=0

cascardo

cascardo

trusty libvirt/qemu does not create nested capable VMs by default