queue/KVM-nVMX-initialize-PML-fields-in-vmcs02.patch

From 1fb883bb827ee8efc1cc9ea0154f953f8a219d38 Mon Sep 17 00:00:00 2001
From: Ladi Prosek <lprosek@redhat.com>
Date: Tue, 4 Apr 2017 14:18:53 +0200
Subject: [PATCH] KVM: nVMX: initialize PML fields in vmcs02
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

commit 1fb883bb827ee8efc1cc9ea0154f953f8a219d38 upstream.

L2 was running with uninitialized PML fields which led to incomplete
dirty bitmap logging. This manifested as all kinds of subtle erratic
behavior of the nested guest.

Fixes: 843e4330573c ("KVM: VMX: Add PML support in VMX")
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 605183291069..259e9b28ccf8 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -10270,6 +10270,18 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12,
 
 	}
 
+	if (enable_pml) {
+		/*
+		 * Conceptually we want to copy the PML address and index from
+		 * vmcs01 here, and then back to vmcs01 on nested vmexit. But,
+		 * since we always flush the log on each vmexit, this happens
+		 * to be equivalent to simply resetting the fields in vmcs02.
+		 */
+		ASSERT(vmx->pml_pg);
+		vmcs_write64(PML_ADDRESS, page_to_phys(vmx->pml_pg));
+		vmcs_write16(GUEST_PML_INDEX, PML_ENTITY_NUM - 1);
+	}
+
 	if (nested_cpu_has_ept(vmcs12)) {
 		kvm_mmu_unload(vcpu);
 		nested_ept_init_mmu_context(vcpu);
-- 
2.12.0