KVM: nVMX: Fully emulate preemption timer

We cannot rely on the hardware-provided preemption timer support because
we are holding L2 in HLT outside non-root mode. Furthermore, emulating
the preemption will resolve tick rate errata on older Intel CPUs.

The emulation is based on hrtimer which is started on L2 entry, stopped
on L2 exit and evaluated via the new check_nested_events hook. As we no
longer rely on hardware features, we can enable both the preemption
timer support and value saving unconditionally.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

arch/x86/kvm/vmx.c

@@ -31,6 +31,7 @@
 #include <linux/ftrace_event.h>
 #include <linux/slab.h>
 #include <linux/tboot.h>
+#include <linux/hrtimer.h>
 #include "kvm_cache_regs.h"
 #include "x86.h"
 
@@ -110,6 +111,8 @@ module_param(nested, bool, S_IRUGO);
 
 #define RMODE_GUEST_OWNED_EFLAGS_BITS (~(X86_EFLAGS_IOPL | X86_EFLAGS_VM))
 
+#define VMX_MISC_EMULATED_PREEMPTION_TIMER_RATE 5
+
 /*
  * These 2 parameters are used to config the controls for Pause-Loop Exiting:
  * ple_gap:    upper bound on the amount of time between two successive
@@ -374,6 +377,9 @@ struct nested_vmx {
 	 */
 	struct page *apic_access_page;
 	u64 msr_ia32_feature_control;
+
+	struct hrtimer preemption_timer;
+	bool preemption_timer_expired;
 };
 
 #define POSTED_INTR_ON  0
@@ -1048,6 +1054,12 @@ static inline bool nested_cpu_has_virtual_nmis(struct vmcs12 *vmcs12)
 	return vmcs12->pin_based_vm_exec_control & PIN_BASED_VIRTUAL_NMIS;
 }
 
+static inline bool nested_cpu_has_preemption_timer(struct vmcs12 *vmcs12)
+{
+	return vmcs12->pin_based_vm_exec_control &
+		PIN_BASED_VMX_PREEMPTION_TIMER;
+}
+
 static inline int nested_cpu_has_ept(struct vmcs12 *vmcs12)
 {
 	return nested_cpu_has2(vmcs12, SECONDARY_EXEC_ENABLE_EPT);
@@ -2253,9 +2265,9 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 	 */
 	nested_vmx_pinbased_ctls_low |= PIN_BASED_ALWAYSON_WITHOUT_TRUE_MSR;
 	nested_vmx_pinbased_ctls_high &= PIN_BASED_EXT_INTR_MASK |
-		PIN_BASED_NMI_EXITING | PIN_BASED_VIRTUAL_NMIS |
+		PIN_BASED_NMI_EXITING | PIN_BASED_VIRTUAL_NMIS;
+	nested_vmx_pinbased_ctls_high |= PIN_BASED_ALWAYSON_WITHOUT_TRUE_MSR |
 		PIN_BASED_VMX_PREEMPTION_TIMER;
-	nested_vmx_pinbased_ctls_high |= PIN_BASED_ALWAYSON_WITHOUT_TRUE_MSR;
 
 	/*
 	 * Exit controls
@@ -2270,15 +2282,10 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 #ifdef CONFIG_X86_64
 		VM_EXIT_HOST_ADDR_SPACE_SIZE |
 #endif
-		VM_EXIT_LOAD_IA32_PAT | VM_EXIT_SAVE_IA32_PAT |
+		VM_EXIT_LOAD_IA32_PAT | VM_EXIT_SAVE_IA32_PAT;
+	nested_vmx_exit_ctls_high |= VM_EXIT_ALWAYSON_WITHOUT_TRUE_MSR |
+		VM_EXIT_LOAD_IA32_EFER | VM_EXIT_SAVE_IA32_EFER |
 		VM_EXIT_SAVE_VMX_PREEMPTION_TIMER;
-	if (!(nested_vmx_pinbased_ctls_high & PIN_BASED_VMX_PREEMPTION_TIMER) ||
-	    !(nested_vmx_exit_ctls_high & VM_EXIT_SAVE_VMX_PREEMPTION_TIMER)) {
-		nested_vmx_exit_ctls_high &= ~VM_EXIT_SAVE_VMX_PREEMPTION_TIMER;
-		nested_vmx_pinbased_ctls_high &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
-	}
-	nested_vmx_exit_ctls_high |= (VM_EXIT_ALWAYSON_WITHOUT_TRUE_MSR |
-		VM_EXIT_LOAD_IA32_EFER | VM_EXIT_SAVE_IA32_EFER);
 
 	/* entry controls */
 	rdmsr(MSR_IA32_VMX_ENTRY_CTLS,
@@ -2347,9 +2354,9 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 
 	/* miscellaneous data */
 	rdmsr(MSR_IA32_VMX_MISC, nested_vmx_misc_low, nested_vmx_misc_high);
-	nested_vmx_misc_low &= VMX_MISC_PREEMPTION_TIMER_RATE_MASK |
-		VMX_MISC_SAVE_EFER_LMA;
-	nested_vmx_misc_low |= VMX_MISC_ACTIVITY_HLT;
+	nested_vmx_misc_low &= VMX_MISC_SAVE_EFER_LMA;
+	nested_vmx_misc_low |= VMX_MISC_EMULATED_PREEMPTION_TIMER_RATE |
+		VMX_MISC_ACTIVITY_HLT;
 	nested_vmx_misc_high = 0;
 }
 
@@ -5713,6 +5720,18 @@ static void nested_vmx_failValid(struct kvm_vcpu *vcpu,
 	 */
 }
 
+static enum hrtimer_restart vmx_preemption_timer_fn(struct hrtimer *timer)
+{
+	struct vcpu_vmx *vmx =
+		container_of(timer, struct vcpu_vmx, nested.preemption_timer);
+
+	vmx->nested.preemption_timer_expired = true;
+	kvm_make_request(KVM_REQ_EVENT, &vmx->vcpu);
+	kvm_vcpu_kick(&vmx->vcpu);
+
+	return HRTIMER_NORESTART;
+}
+
 /*
  * Emulate the VMXON instruction.
  * Currently, we just remember that VMX is active, and do not save or even
@@ -5777,6 +5796,10 @@ static int handle_vmon(struct kvm_vcpu *vcpu)
 	INIT_LIST_HEAD(&(vmx->nested.vmcs02_pool));
 	vmx->nested.vmcs02_num = 0;
 
+	hrtimer_init(&vmx->nested.preemption_timer, CLOCK_MONOTONIC,
+		     HRTIMER_MODE_REL);
+	vmx->nested.preemption_timer.function = vmx_preemption_timer_fn;
+
 	vmx->nested.vmxon = true;
 
 	skip_emulated_instruction(vcpu);
@@ -6753,9 +6776,6 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu)
 		 * table is L0's fault.
 		 */
 		return 0;
-	case EXIT_REASON_PREEMPTION_TIMER:
-		return vmcs12->pin_based_vm_exec_control &
-			PIN_BASED_VMX_PREEMPTION_TIMER;
 	case EXIT_REASON_WBINVD:
 		return nested_cpu_has2(vmcs12, SECONDARY_EXEC_WBINVD_EXITING);
 	case EXIT_REASON_XSETBV:
@@ -6771,27 +6791,6 @@ static void vmx_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2)
 	*info2 = vmcs_read32(VM_EXIT_INTR_INFO);
 }
 
-static void nested_adjust_preemption_timer(struct kvm_vcpu *vcpu)
-{
-	u64 delta_tsc_l1;
-	u32 preempt_val_l1, preempt_val_l2, preempt_scale;
-
-	if (!(get_vmcs12(vcpu)->pin_based_vm_exec_control &
-			PIN_BASED_VMX_PREEMPTION_TIMER))
-		return;
-	preempt_scale = native_read_msr(MSR_IA32_VMX_MISC) &
-			MSR_IA32_VMX_MISC_PREEMPTION_TIMER_SCALE;
-	preempt_val_l2 = vmcs_read32(VMX_PREEMPTION_TIMER_VALUE);
-	delta_tsc_l1 = vmx_read_l1_tsc(vcpu, native_read_tsc())
-		- vcpu->arch.last_guest_tsc;
-	preempt_val_l1 = delta_tsc_l1 >> preempt_scale;
-	if (preempt_val_l2 <= preempt_val_l1)
-		preempt_val_l2 = 0;
-	else
-		preempt_val_l2 -= preempt_val_l1;
-	vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, preempt_val_l2);
-}
-
 /*
  * The guest has exited.  See if we can fix it or if we need userspace
  * assistance.
@@ -7210,8 +7209,6 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	atomic_switch_perf_msrs(vmx);
 	debugctlmsr = get_debugctlmsr();
 
-	if (is_guest_mode(vcpu) && !vmx->nested.nested_run_pending)
-		nested_adjust_preemption_timer(vcpu);
 	vmx->__launched = vmx->loaded_vmcs->launched;
 	asm(
 		/* Store host registers */
@@ -7608,6 +7605,28 @@ static void vmx_inject_page_fault_nested(struct kvm_vcpu *vcpu,
 		kvm_inject_page_fault(vcpu, fault);
 }
 
+static void vmx_start_preemption_timer(struct kvm_vcpu *vcpu)
+{
+	u64 preemption_timeout = get_vmcs12(vcpu)->vmx_preemption_timer_value;
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (vcpu->arch.virtual_tsc_khz == 0)
+		return;
+
+	/* Make sure short timeouts reliably trigger an immediate vmexit.
+	 * hrtimer_start does not guarantee this. */
+	if (preemption_timeout <= 1) {
+		vmx_preemption_timer_fn(&vmx->nested.preemption_timer);
+		return;
+	}
+
+	preemption_timeout <<= VMX_MISC_EMULATED_PREEMPTION_TIMER_RATE;
+	preemption_timeout *= 1000000;
+	do_div(preemption_timeout, vcpu->arch.virtual_tsc_khz);
+	hrtimer_start(&vmx->nested.preemption_timer,
+		      ns_to_ktime(preemption_timeout), HRTIMER_MODE_REL);
+}
+
 /*
  * prepare_vmcs02 is called when the L1 guest hypervisor runs its nested
  * L2 guest. L1 has a vmcs for L2 (vmcs12), and this function "merges" it
@@ -7621,7 +7640,6 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u32 exec_control;
-	u32 exit_control;
 
 	vmcs_write16(GUEST_ES_SELECTOR, vmcs12->guest_es_selector);
 	vmcs_write16(GUEST_CS_SELECTOR, vmcs12->guest_cs_selector);
@@ -7679,13 +7697,14 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 
 	vmcs_write64(VMCS_LINK_POINTER, -1ull);
 
-	vmcs_write32(PIN_BASED_VM_EXEC_CONTROL,
-		(vmcs_config.pin_based_exec_ctrl |
-		 vmcs12->pin_based_vm_exec_control));
+	exec_control = vmcs12->pin_based_vm_exec_control;
+	exec_control |= vmcs_config.pin_based_exec_ctrl;
+	exec_control &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
+	vmcs_write32(PIN_BASED_VM_EXEC_CONTROL, exec_control);
 
-	if (vmcs12->pin_based_vm_exec_control & PIN_BASED_VMX_PREEMPTION_TIMER)
-		vmcs_write32(VMX_PREEMPTION_TIMER_VALUE,
-			     vmcs12->vmx_preemption_timer_value);
+	vmx->nested.preemption_timer_expired = false;
+	if (nested_cpu_has_preemption_timer(vmcs12))
+		vmx_start_preemption_timer(vcpu);
 
 	/*
 	 * Whether page-faults are trapped is determined by a combination of
@@ -7713,7 +7732,7 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 		enable_ept ? vmcs12->page_fault_error_code_match : 0);
 
 	if (cpu_has_secondary_exec_ctrls()) {
-		u32 exec_control = vmx_secondary_exec_control(vmx);
+		exec_control = vmx_secondary_exec_control(vmx);
 		if (!vmx->rdtscp_enabled)
 			exec_control &= ~SECONDARY_EXEC_RDTSCP;
 		/* Take the following fields only from vmcs12 */
@@ -7800,10 +7819,7 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	 * we should use its exit controls. Note that VM_EXIT_LOAD_IA32_EFER
 	 * bits are further modified by vmx_set_efer() below.
 	 */
-	exit_control = vmcs_config.vmexit_ctrl;
-	if (vmcs12->pin_based_vm_exec_control & PIN_BASED_VMX_PREEMPTION_TIMER)
-		exit_control |= VM_EXIT_SAVE_VMX_PREEMPTION_TIMER;
-	vm_exit_controls_init(vmx, exit_control);
+	vmcs_write32(VM_EXIT_CONTROLS, vmcs_config.vmexit_ctrl);
 
 	/* vmcs12's VM_ENTRY_LOAD_IA32_EFER and VM_ENTRY_IA32E_MODE are
 	 * emulated by vmx_set_efer(), below.
@@ -8151,6 +8167,14 @@ static int vmx_check_nested_events(struct kvm_vcpu *vcpu, bool external_intr)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
+	if (nested_cpu_has_preemption_timer(get_vmcs12(vcpu)) &&
+	    vmx->nested.preemption_timer_expired) {
+		if (vmx->nested.nested_run_pending)
+			return -EBUSY;
+		nested_vmx_vmexit(vcpu, EXIT_REASON_PREEMPTION_TIMER, 0, 0);
+		return 0;
+	}
+
 	if (vcpu->arch.nmi_pending && nested_exit_on_nmi(vcpu)) {
 		if (vmx->nested.nested_run_pending)
 			return -EBUSY;
@@ -8176,6 +8200,20 @@ static int vmx_check_nested_events(struct kvm_vcpu *vcpu, bool external_intr)
 	return 0;
 }
 
+static u32 vmx_get_preemption_timer_value(struct kvm_vcpu *vcpu)
+{
+	ktime_t remaining =
+		hrtimer_get_remaining(&to_vmx(vcpu)->nested.preemption_timer);
+	u64 value;
+
+	if (ktime_to_ns(remaining) <= 0)
+		return 0;
+
+	value = ktime_to_ns(remaining) * vcpu->arch.virtual_tsc_khz;
+	do_div(value, 1000000);
+	return value >> VMX_MISC_EMULATED_PREEMPTION_TIMER_RATE;
+}
+
 /*
  * prepare_vmcs12 is part of what we need to do when the nested L2 guest exits
  * and we want to prepare to run its L1 parent. L1 keeps a vmcs for L2 (vmcs12),
@@ -8246,10 +8284,13 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12,
 	else
 		vmcs12->guest_activity_state = GUEST_ACTIVITY_ACTIVE;
 
-	if ((vmcs12->pin_based_vm_exec_control & PIN_BASED_VMX_PREEMPTION_TIMER) &&
-	    (vmcs12->vm_exit_controls & VM_EXIT_SAVE_VMX_PREEMPTION_TIMER))
-		vmcs12->vmx_preemption_timer_value =
-			vmcs_read32(VMX_PREEMPTION_TIMER_VALUE);
+	if (nested_cpu_has_preemption_timer(vmcs12)) {
+		if (vmcs12->vm_exit_controls &
+		    VM_EXIT_SAVE_VMX_PREEMPTION_TIMER)
+			vmcs12->vmx_preemption_timer_value =
+				vmx_get_preemption_timer_value(vcpu);
+		hrtimer_cancel(&to_vmx(vcpu)->nested.preemption_timer);
+	}
 
 	/*
 	 * In some cases (usually, nested EPT), L2 is allowed to change its