KVM: nVMX: update last_nonleaf_level when initializing nested EPT
authorLadi Prosek <lprosek@redhat.com>
Thu, 5 Oct 2017 09:10:22 +0000 (11:10 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 18 Oct 2017 07:35:42 +0000 (09:35 +0200)
commit fd19d3b45164466a4adce7cbff448ba9189e1427 upstream.

The function updates context->root_level but didn't call
update_last_nonleaf_level so the previous and potentially wrong value
was used for page walks.  For example, a zero value of last_nonleaf_level
would allow a potential out-of-bounds access in arch/x86/mmu/paging_tmpl.h's
walk_addr_generic function (CVE-2017-12188).

Fixes: 155a97a3d7c78b46cef6f1a973c831bc5a4f82bb
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/x86/kvm/mmu.c

index ea3f8f7..d29c745 100644 (file)
@@ -4169,6 +4169,7 @@ void kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, bool execonly)
 
        update_permission_bitmask(vcpu, context, true);
        update_pkru_bitmask(vcpu, context, true);
+       update_last_nonleaf_level(vcpu, context);
        reset_rsvds_bits_mask_ept(vcpu, context, execonly);
        reset_ept_shadow_zero_bits_mask(vcpu, context, execonly);
 }