KVM_REQ_GET_NESTED_STATE_PAGES is also used with VM entries that happen
on exit from SMM mode, and in this case PDPTRS must be always reloaded.
Thanks to Sean Christopherson for pointing this out.
Fixes: 0f85722341b0 ("KVM: nVMX: delay loading of PDPTRs to KVM_REQ_GET_NESTED_STATE_PAGES")
Signed-off-by: Maxim Levitsky <mlevitsk@xxxxxxxxxx>
---
arch/x86/kvm/vmx/vmx.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index fada1055f325..4194fbf5e5d6 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -7504,6 +7504,13 @@ static int vmx_leave_smm(struct kvm_vcpu *vcpu, const char *smstate)
}
if (vmx->nested.smm.guest_mode) {
+
+ /* Exit from the SMM to the non root mode also uses
+ * the KVM_REQ_GET_NESTED_STATE_PAGES request,
+ * but in this case the pdptrs must be always reloaded
+ */
+ vcpu->arch.pdptrs_from_userspace = false;
+
ret = nested_vmx_enter_non_root_mode(vcpu, false);
if (ret)
return ret;
--
2.26.3
