KVM: arm64: Clear __hyp_running_vcpu when flushing the pKVM hyp vCPU
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
KVM: arm64: Clear __hyp_running_vcpu when flushing the pKVM hyp vCPU
flush_hyp_vcpu() copies the host vCPU context into the hyp's private vCPU on every run. ctxt_to_vcpu() expects a guest context to have a NULL __hyp_running_vcpu, which is only ever set on the host context, so that it resolves the vCPU via container_of(). While this is generally the case, flush_hyp_vcpu() copies the context verbatim and does not enforce this, so a value provided by the host is dereferenced at EL2 (host -> EL2).
Fix by clearing __hyp_running_vcpu after the copy.
From the CVE record · Linux
The KVM arm64 pKVM code copies a host vCPU context into a private hypervisor vCPU without clearing __hyp_running_vcpu. On a vCPU run, a host-provided value in that field can be dereferenced at EL2. The input does not specify attacker access requirements or user interaction.
Engineers running Linux kernels with KVM arm64 pKVM should check their versions; affected versions are not specified.