1 files changed, 52 insertions, 19 deletions

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 1320e0f8e611..138ceffc6377 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -503,7 +503,7 @@ static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu)
                                 [number##_HIGH] = VMCS12_OFFSET(name)+4
 
 
-static const unsigned long shadow_read_only_fields[] = {
+static unsigned long shadow_read_only_fields[] = {
         /*
          * We do NOT shadow fields that are modified when L0
          * traps and emulates any vmx instruction (e.g. VMPTRLD,
@@ -526,10 +526,10 @@ static const unsigned long shadow_read_only_fields[] = {
         GUEST_LINEAR_ADDRESS,
         GUEST_PHYSICAL_ADDRESS
 };
-static const int max_shadow_read_only_fields =
+static int max_shadow_read_only_fields =
         ARRAY_SIZE(shadow_read_only_fields);
 
-static const unsigned long shadow_read_write_fields[] = {
+static unsigned long shadow_read_write_fields[] = {
         GUEST_RIP,
         GUEST_RSP,
         GUEST_CR0,
@@ -558,7 +558,7 @@ static const unsigned long shadow_read_write_fields[] = {
         HOST_FS_SELECTOR,
         HOST_GS_SELECTOR
 };
-static const int max_shadow_read_write_fields =
+static int max_shadow_read_write_fields =
         ARRAY_SIZE(shadow_read_write_fields);
 
 static const unsigned short vmcs_field_to_offset_table[] = {
@@ -3009,6 +3009,41 @@ static void free_kvm_area(void)
         }
 }
 
+static void init_vmcs_shadow_fields(void)
+{
+        int i, j;
+
+        /* No checks for read only fields yet */
+
+        for (i = j = 0; i < max_shadow_read_write_fields; i++) {
+                switch (shadow_read_write_fields[i]) {
+                case GUEST_BNDCFGS:
+                        if (!vmx_mpx_supported())
+                                continue;
+                        break;
+                default:
+                        break;
+                }
+
+                if (j < i)
+                        shadow_read_write_fields[j] =
+                                shadow_read_write_fields[i];
+                j++;
+        }
+        max_shadow_read_write_fields = j;
+
+        /* shadowed fields guest access without vmexit */
+        for (i = 0; i < max_shadow_read_write_fields; i++) {
+                clear_bit(shadow_read_write_fields[i],
+                          vmx_vmwrite_bitmap);
+                clear_bit(shadow_read_write_fields[i],
+                          vmx_vmread_bitmap);
+        }
+        for (i = 0; i < max_shadow_read_only_fields; i++)
+                clear_bit(shadow_read_only_fields[i],
+                          vmx_vmread_bitmap);
+}
+
 static __init int alloc_kvm_area(void)
 {
         int cpu;
@@ -3039,6 +3074,8 @@ static __init int hardware_setup(void)
                 enable_vpid = 0;
         if (!cpu_has_vmx_shadow_vmcs())
                 enable_shadow_vmcs = 0;
+        if (enable_shadow_vmcs)
+                init_vmcs_shadow_fields();
 
         if (!cpu_has_vmx_ept() ||
             !cpu_has_vmx_ept_4levels()) {
@@ -3484,13 +3521,14 @@ static int vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
                         hw_cr4 &= ~X86_CR4_PAE;
                         hw_cr4 |= X86_CR4_PSE;
                         /*
-                         * SMEP is disabled if CPU is in non-paging mode in
-                         * hardware. However KVM always uses paging mode to
+                         * SMEP/SMAP is disabled if CPU is in non-paging mode
+                         * in hardware. However KVM always uses paging mode to
                          * emulate guest non-paging mode with TDP.
-                         * To emulate this behavior, SMEP needs to be manually
-                         * disabled when guest switches to non-paging mode.
+                         * To emulate this behavior, SMEP/SMAP needs to be
+                         * manually disabled when guest switches to non-paging
+                         * mode.
                          */
-                        hw_cr4 &= ~X86_CR4_SMEP;
+                        hw_cr4 &= ~(X86_CR4_SMEP | X86_CR4_SMAP);
                 } else if (!(cr4 & X86_CR4_PAE)) {
                         hw_cr4 &= ~X86_CR4_PAE;
                 }
@@ -7740,7 +7778,8 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 
         exec_control = vmcs12->pin_based_vm_exec_control;
         exec_control |= vmcs_config.pin_based_exec_ctrl;
-        exec_control &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
+        exec_control &= ~(PIN_BASED_VMX_PREEMPTION_TIMER |
+                          PIN_BASED_POSTED_INTR);
         vmcs_write32(PIN_BASED_VM_EXEC_CONTROL, exec_control);
 
         vmx->nested.preemption_timer_expired = false;
@@ -7777,7 +7816,9 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
                 if (!vmx->rdtscp_enabled)
                         exec_control &= ~SECONDARY_EXEC_RDTSCP;
                 /* Take the following fields only from vmcs12 */
-                exec_control &= ~SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES;
+                exec_control &= ~(SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
+                                  SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY |
+                                  SECONDARY_EXEC_APIC_REGISTER_VIRT);
                 if (nested_cpu_has(vmcs12,
                                 CPU_BASED_ACTIVATE_SECONDARY_CONTROLS))
                         exec_control |= vmcs12->secondary_vm_exec_control;
@@ -8802,14 +8843,6 @@ static int __init vmx_init(void)
 
         memset(vmx_vmread_bitmap, 0xff, PAGE_SIZE);
         memset(vmx_vmwrite_bitmap, 0xff, PAGE_SIZE);
-        /* shadowed read/write fields */
-        for (i = 0; i < max_shadow_read_write_fields; i++) {
-                clear_bit(shadow_read_write_fields[i], vmx_vmwrite_bitmap);
-                clear_bit(shadow_read_write_fields[i], vmx_vmread_bitmap);
-        }
-        /* shadowed read only fields */
-        for (i = 0; i < max_shadow_read_only_fields; i++)
-                clear_bit(shadow_read_only_fields[i], vmx_vmread_bitmap);
 
         /*
          * Allow direct access to the PC debug port (it is often used for I/O