OSDN Git Service

KVM: nSVM: always intercept VMLOAD/VMSAVE when nested (CVE-2021-3656)
authorMaxim Levitsky <mlevitsk@redhat.com>
Mon, 19 Jul 2021 13:05:00 +0000 (16:05 +0300)
committerPaolo Bonzini <pbonzini@redhat.com>
Mon, 16 Aug 2021 13:48:37 +0000 (09:48 -0400)
commitc7dfa4009965a9b2d7b329ee970eb8da0d32f0bc
tree42c6b74cec101cdffce4c14e608b4c672adb4b61
parent0f923e07124df069ba68d8bb12324398f4b6b709
KVM: nSVM: always intercept VMLOAD/VMSAVE when nested (CVE-2021-3656)

If L1 disables VMLOAD/VMSAVE intercepts, and doesn't enable
Virtual VMLOAD/VMSAVE (currently not supported for the nested hypervisor),
then VMLOAD/VMSAVE must operate on the L1 physical memory, which is only
possible by making L0 intercept these instructions.

Failure to do so allowed the nested guest to run VMLOAD/VMSAVE unintercepted,
and thus read/write portions of the host physical memory.

Fixes: 89c8a4984fc9 ("KVM: SVM: Enable Virtual VMLOAD VMSAVE feature")

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/kvm/svm/nested.c