KVM: nSVM: Check the value written to MSR_VM_HSAVE_PA

APM states that #GP is raised upon write to MSR_VM_HSAVE_PA when
the supplied address is not page-aligned or is outside of "maximum
supported physical address for this implementation".
page_address_valid() check seems suitable. Also, forcefully page-align
the address when it's written from VMM.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210628104425.391276-2-vkuznets@redhat.com>
Cc: stable@vger.kernel.org
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
[Add comment about behavior for host-provided values. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index 63488d3..b6e91de 100644 (file)
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -2954,7 +2954,16 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)
                        svm_disable_lbrv(vcpu);
                break;
        case MSR_VM_HSAVE_PA:
-               svm->nested.hsave_msr = data;
+               /*
+                * Old kernels did not validate the value written to
+                * MSR_VM_HSAVE_PA.  Allow KVM_SET_MSR to set an invalid
+                * value to allow live migrating buggy or malicious guests
+                * originating from those kernels.
+                */
+               if (!msr->host_initiated && !page_address_valid(vcpu, data))
+                       return 1;
+
+               svm->nested.hsave_msr = data & PAGE_MASK;
                break;
        case MSR_VM_CR:
                return svm_set_vm_cr(vcpu, data);