Merge branch 'kvm-updates/2.6.30' of git://git.kernel.org/pub/scm/virt/kvm/kvm

* 'kvm-updates/2.6.30' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (113 commits)
  KVM: VMX: Don't allow uninhibited access to EFER on i386
  KVM: Correct deassign device ioctl to IOW
  KVM: ppc: e500: Fix the bug that KVM is unstable in SMP
  KVM: ppc: e500: Fix the bug that mas0 update to wrong value when read TLB entry
  KVM: Fix missing smp tlb flush in invlpg
  KVM: Get support IRQ routing entry counts
  KVM: fix sparse warnings: Should it be static?
  KVM: fix sparse warnings: context imbalance
  KVM: is_long_mode() should check for EFER.LMA
  KVM: VMX: Update necessary state when guest enters long mode
  KVM: ia64: Fix the build errors due to lack of macros related to MSI.
  ia64: Move the macro definitions related to MSI to one header file.
  KVM: fix kvm_vm_ioctl_deassign_device
  KVM: define KVM_CAP_DEVICE_DEASSIGNMENT
  KVM: ppc: Add emulation of E500 register mmucsr0
  KVM: Report IRQ injection status for MSI delivered interrupts
  KVM: MMU: Fix another largepage memory leak
  KVM: SVM: set accessed bit for VMCB segment selectors
  KVM: Report IRQ injection status to userspace.
  KVM: MMU: remove assertion in kvm_mmu_alloc_page
  ...

arch/ia64/include/asm/kvm.h

@@ -166,7 +166,40 @@ struct saved_vpd {
 	unsigned long  vcpuid[5];
 	unsigned long  vpsr;
 	unsigned long  vpr;
-	unsigned long  vcr[128];
+	union {
+		unsigned long  vcr[128];
+		struct {
+			unsigned long dcr;
+			unsigned long itm;
+			unsigned long iva;
+			unsigned long rsv1[5];
+			unsigned long pta;
+			unsigned long rsv2[7];
+			unsigned long ipsr;
+			unsigned long isr;
+			unsigned long rsv3;
+			unsigned long iip;
+			unsigned long ifa;
+			unsigned long itir;
+			unsigned long iipa;
+			unsigned long ifs;
+			unsigned long iim;
+			unsigned long iha;
+			unsigned long rsv4[38];
+			unsigned long lid;
+			unsigned long ivr;
+			unsigned long tpr;
+			unsigned long eoi;
+			unsigned long irr[4];
+			unsigned long itv;
+			unsigned long pmv;
+			unsigned long cmcv;
+			unsigned long rsv5[5];
+			unsigned long lrr0;
+			unsigned long lrr1;
+			unsigned long rsv6[46];
+		};
+	};
 };
 
 struct kvm_regs {
@@ -214,4 +247,18 @@ struct kvm_sregs {
 struct kvm_fpu {
 };
 
+#define KVM_IA64_VCPU_STACK_SHIFT	16
+#define KVM_IA64_VCPU_STACK_SIZE	(1UL << KVM_IA64_VCPU_STACK_SHIFT)
+
+struct kvm_ia64_vcpu_stack {
+	unsigned char stack[KVM_IA64_VCPU_STACK_SIZE];
+};
+
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif

arch/ia64/include/asm/kvm_host.h

@@ -112,7 +112,11 @@
 #define VCPU_STRUCT_SHIFT	16
 #define VCPU_STRUCT_SIZE	(__IA64_UL_CONST(1) << VCPU_STRUCT_SHIFT)
 
-#define KVM_STK_OFFSET		VCPU_STRUCT_SIZE
+/*
+ * This must match KVM_IA64_VCPU_STACK_{SHIFT,SIZE} arch/ia64/include/asm/kvm.h
+ */
+#define KVM_STK_SHIFT		16
+#define KVM_STK_OFFSET		(__IA64_UL_CONST(1)<< KVM_STK_SHIFT)
 
 #define KVM_VM_STRUCT_SHIFT	19
 #define KVM_VM_STRUCT_SIZE	(__IA64_UL_CONST(1) << KVM_VM_STRUCT_SHIFT)
@@ -153,10 +157,10 @@ struct kvm_vm_data {
 	struct kvm_vcpu_data vcpu_data[KVM_MAX_VCPUS];
 };
 
-#define VCPU_BASE(n)	KVM_VM_DATA_BASE + \
-				offsetof(struct kvm_vm_data, vcpu_data[n])
-#define VM_BASE		KVM_VM_DATA_BASE + \
-				offsetof(struct kvm_vm_data, kvm_vm_struct)
+#define VCPU_BASE(n)	(KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, vcpu_data[n]))
+#define KVM_VM_BASE	(KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, kvm_vm_struct))
 #define KVM_MEM_DIRTY_LOG_BASE	KVM_VM_DATA_BASE + \
 				offsetof(struct kvm_vm_data, kvm_mem_dirty_log)
 
@@ -235,8 +239,6 @@ struct kvm_vm_data {
 
 struct kvm;
 struct kvm_vcpu;
-struct kvm_guest_debug{
-};
 
 struct kvm_mmio_req {
 	uint64_t addr;          /*  physical address		*/
@@ -462,6 +464,8 @@ struct kvm_arch {
 	unsigned long	metaphysical_rr4;
 	unsigned long	vmm_init_rr;
 
+	int		online_vcpus;
+
 	struct kvm_ioapic *vioapic;
 	struct kvm_vm_stat stat;
 	struct kvm_sal_data rdv_sal_data;

arch/ia64/include/asm/msidef.h

@@ -0,0 +1,42 @@
+#ifndef _IA64_MSI_DEF_H
+#define _IA64_MSI_DEF_H
+
+/*
+ * Shifts for APIC-based data
+ */
+
+#define     MSI_DATA_VECTOR_SHIFT	0
+#define	    MSI_DATA_VECTOR(v)		(((u8)v) << MSI_DATA_VECTOR_SHIFT)
+#define     MSI_DATA_VECTOR_MASK	0xffffff00
+
+#define     MSI_DATA_DELIVERY_MODE_SHIFT	8
+#define     MSI_DATA_DELIVERY_FIXED	(0 << MSI_DATA_DELIVERY_MODE_SHIFT)
+#define     MSI_DATA_DELIVERY_LOWPRI	(1 << MSI_DATA_DELIVERY_MODE_SHIFT)
+
+#define     MSI_DATA_LEVEL_SHIFT	14
+#define     MSI_DATA_LEVEL_DEASSERT	(0 << MSI_DATA_LEVEL_SHIFT)
+#define     MSI_DATA_LEVEL_ASSERT	(1 << MSI_DATA_LEVEL_SHIFT)
+
+#define     MSI_DATA_TRIGGER_SHIFT	15
+#define     MSI_DATA_TRIGGER_EDGE	(0 << MSI_DATA_TRIGGER_SHIFT)
+#define     MSI_DATA_TRIGGER_LEVEL	(1 << MSI_DATA_TRIGGER_SHIFT)
+
+/*
+ * Shift/mask fields for APIC-based bus address
+ */
+
+#define     MSI_ADDR_DEST_ID_SHIFT	4
+#define     MSI_ADDR_HEADER		0xfee00000
+
+#define     MSI_ADDR_DEST_ID_MASK	0xfff0000f
+#define     MSI_ADDR_DEST_ID_CPU(cpu)	((cpu) << MSI_ADDR_DEST_ID_SHIFT)
+
+#define     MSI_ADDR_DEST_MODE_SHIFT	2
+#define     MSI_ADDR_DEST_MODE_PHYS	(0 << MSI_ADDR_DEST_MODE_SHIFT)
+#define	    MSI_ADDR_DEST_MODE_LOGIC	(1 << MSI_ADDR_DEST_MODE_SHIFT)
+
+#define     MSI_ADDR_REDIRECTION_SHIFT	3
+#define     MSI_ADDR_REDIRECTION_CPU	(0 << MSI_ADDR_REDIRECTION_SHIFT)
+#define     MSI_ADDR_REDIRECTION_LOWPRI	(1 << MSI_ADDR_REDIRECTION_SHIFT)
+
+#endif/* _IA64_MSI_DEF_H */

arch/ia64/kernel/msi_ia64.c

@@ -7,44 +7,7 @@
 #include <linux/msi.h>
 #include <linux/dmar.h>
 #include <asm/smp.h>
-
-/*
- * Shifts for APIC-based data
- */
-
-#define MSI_DATA_VECTOR_SHIFT		0
-#define	    MSI_DATA_VECTOR(v)		(((u8)v) << MSI_DATA_VECTOR_SHIFT)
-#define MSI_DATA_VECTOR_MASK		0xffffff00
-
-#define MSI_DATA_DELIVERY_SHIFT		8
-#define     MSI_DATA_DELIVERY_FIXED	(0 << MSI_DATA_DELIVERY_SHIFT)
-#define     MSI_DATA_DELIVERY_LOWPRI	(1 << MSI_DATA_DELIVERY_SHIFT)
-
-#define MSI_DATA_LEVEL_SHIFT		14
-#define     MSI_DATA_LEVEL_DEASSERT	(0 << MSI_DATA_LEVEL_SHIFT)
-#define     MSI_DATA_LEVEL_ASSERT	(1 << MSI_DATA_LEVEL_SHIFT)
-
-#define MSI_DATA_TRIGGER_SHIFT		15
-#define     MSI_DATA_TRIGGER_EDGE	(0 << MSI_DATA_TRIGGER_SHIFT)
-#define     MSI_DATA_TRIGGER_LEVEL	(1 << MSI_DATA_TRIGGER_SHIFT)
-
-/*
- * Shift/mask fields for APIC-based bus address
- */
-
-#define MSI_TARGET_CPU_SHIFT		4
-#define MSI_ADDR_HEADER			0xfee00000
-
-#define MSI_ADDR_DESTID_MASK		0xfff0000f
-#define     MSI_ADDR_DESTID_CPU(cpu)	((cpu) << MSI_TARGET_CPU_SHIFT)
-
-#define MSI_ADDR_DESTMODE_SHIFT		2
-#define     MSI_ADDR_DESTMODE_PHYS	(0 << MSI_ADDR_DESTMODE_SHIFT)
-#define	    MSI_ADDR_DESTMODE_LOGIC	(1 << MSI_ADDR_DESTMODE_SHIFT)
-
-#define MSI_ADDR_REDIRECTION_SHIFT	3
-#define     MSI_ADDR_REDIRECTION_CPU	(0 << MSI_ADDR_REDIRECTION_SHIFT)
-#define     MSI_ADDR_REDIRECTION_LOWPRI	(1 << MSI_ADDR_REDIRECTION_SHIFT)
+#include <asm/msidef.h>
 
 static struct irq_chip	ia64_msi_chip;
 
@@ -65,8 +28,8 @@ static void ia64_set_msi_irq_affinity(unsigned int irq,
 	read_msi_msg(irq, &msg);
 
 	addr = msg.address_lo;
-	addr &= MSI_ADDR_DESTID_MASK;
-	addr |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
+	addr &= MSI_ADDR_DEST_ID_MASK;
+	addr |= MSI_ADDR_DEST_ID_CPU(cpu_physical_id(cpu));
 	msg.address_lo = addr;
 
 	data = msg.data;
@@ -98,9 +61,9 @@ int ia64_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
 	msg.address_hi = 0;
 	msg.address_lo =
 		MSI_ADDR_HEADER |
-		MSI_ADDR_DESTMODE_PHYS |
+		MSI_ADDR_DEST_MODE_PHYS |
 		MSI_ADDR_REDIRECTION_CPU |
-		MSI_ADDR_DESTID_CPU(dest_phys_id);
+		MSI_ADDR_DEST_ID_CPU(dest_phys_id);
 
 	msg.data =
 		MSI_DATA_TRIGGER_EDGE |
@@ -183,8 +146,8 @@ static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
 	msg.data |= MSI_DATA_VECTOR(cfg->vector);
-	msg.address_lo &= ~MSI_ADDR_DESTID_MASK;
-	msg.address_lo |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
+	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
+	msg.address_lo |= MSI_ADDR_DEST_ID_CPU(cpu_physical_id(cpu));
 
 	dmar_msi_write(irq, &msg);
 	irq_desc[irq].affinity = *mask;
@@ -215,9 +178,9 @@ msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg)
 	msg->address_hi = 0;
 	msg->address_lo =
 		MSI_ADDR_HEADER |
-		MSI_ADDR_DESTMODE_PHYS |
+		MSI_ADDR_DEST_MODE_PHYS |
 		MSI_ADDR_REDIRECTION_CPU |
-		MSI_ADDR_DESTID_CPU(dest);
+		MSI_ADDR_DEST_ID_CPU(dest);
 
 	msg->data =
 		MSI_DATA_TRIGGER_EDGE |

arch/ia64/kvm/Kconfig

@@ -4,6 +4,10 @@
 config HAVE_KVM
 	bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+       default y
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	depends on HAVE_KVM || IA64

arch/ia64/kvm/irq.h

@@ -23,6 +23,8 @@
 #ifndef __IRQ_H
 #define __IRQ_H
 
+#include "lapic.h"
+
 static inline int irqchip_in_kernel(struct kvm *kvm)
 {
 	return 1;

arch/ia64/kvm/kvm-ia64.c

@@ -182,7 +182,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	switch (ext) {
 	case KVM_CAP_IRQCHIP:
 	case KVM_CAP_MP_STATE:
-
+	case KVM_CAP_IRQ_INJECT_STATUS:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -314,7 +314,7 @@ static struct kvm_vcpu *lid_to_vcpu(struct kvm *kvm, unsigned long id,
 	union ia64_lid lid;
 	int i;
 
-	for (i = 0; i < KVM_MAX_VCPUS; i++) {
+	for (i = 0; i < kvm->arch.online_vcpus; i++) {
 		if (kvm->vcpus[i]) {
 			lid.val = VCPU_LID(kvm->vcpus[i]);
 			if (lid.id == id && lid.eid == eid)
@@ -388,7 +388,7 @@ static int handle_global_purge(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 
 	call_data.ptc_g_data = p->u.ptc_g_data;
 
-	for (i = 0; i < KVM_MAX_VCPUS; i++) {
+	for (i = 0; i < kvm->arch.online_vcpus; i++) {
 		if (!kvm->vcpus[i] || kvm->vcpus[i]->arch.mp_state ==
 						KVM_MP_STATE_UNINITIALIZED ||
 					vcpu == kvm->vcpus[i])
@@ -788,6 +788,8 @@ struct  kvm *kvm_arch_create_vm(void)
 		return ERR_PTR(-ENOMEM);
 	kvm_init_vm(kvm);
 
+	kvm->arch.online_vcpus = 0;
+
 	return kvm;
 
 }
@@ -919,7 +921,13 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = kvm_ioapic_init(kvm);
 		if (r)
 			goto out;
+		r = kvm_setup_default_irq_routing(kvm);
+		if (r) {
+			kfree(kvm->arch.vioapic);
+			goto out;
+		}
 		break;
+	case KVM_IRQ_LINE_STATUS:
 	case KVM_IRQ_LINE: {
 		struct kvm_irq_level irq_event;
 
@@ -927,10 +935,17 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		if (copy_from_user(&irq_event, argp, sizeof irq_event))
 			goto out;
 		if (irqchip_in_kernel(kvm)) {
+			__s32 status;
 			mutex_lock(&kvm->lock);
-			kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
+			status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
 				    irq_event.irq, irq_event.level);
 			mutex_unlock(&kvm->lock);
+			if (ioctl == KVM_IRQ_LINE_STATUS) {
+				irq_event.status = status;
+				if (copy_to_user(argp, &irq_event,
+							sizeof irq_event))
+					goto out;
+			}
 			r = 0;
 		}
 		break;
@@ -1149,7 +1164,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 
 		/*Initialize itc offset for vcpus*/
 		itc_offset = 0UL - ia64_getreg(_IA64_REG_AR_ITC);
-		for (i = 0; i < KVM_MAX_VCPUS; i++) {
+		for (i = 0; i < kvm->arch.online_vcpus; i++) {
 			v = (struct kvm_vcpu *)((char *)vcpu +
 					sizeof(struct kvm_vcpu_data) * i);
 			v->arch.itc_offset = itc_offset;
@@ -1283,6 +1298,8 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 		goto fail;
 	}
 
+	kvm->arch.online_vcpus++;
+
 	return vcpu;
 fail:
 	return ERR_PTR(r);
@@ -1303,8 +1320,8 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 	return -EINVAL;
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-		struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	return -EINVAL;
 }
@@ -1421,6 +1438,23 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 	return 0;
 }
 
+int kvm_arch_vcpu_ioctl_get_stack(struct kvm_vcpu *vcpu,
+				  struct kvm_ia64_vcpu_stack *stack)
+{
+	memcpy(stack, vcpu, sizeof(struct kvm_ia64_vcpu_stack));
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_set_stack(struct kvm_vcpu *vcpu,
+				  struct kvm_ia64_vcpu_stack *stack)
+{
+	memcpy(vcpu + 1, &stack->stack[0] + sizeof(struct kvm_vcpu),
+	       sizeof(struct kvm_ia64_vcpu_stack) - sizeof(struct kvm_vcpu));
+
+	vcpu->arch.exit_data = ((struct kvm_vcpu *)stack)->arch.exit_data;
+	return 0;
+}
+
 void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 {
 
@@ -1430,9 +1464,78 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 
 
 long kvm_arch_vcpu_ioctl(struct file *filp,
-		unsigned int ioctl, unsigned long arg)
+			 unsigned int ioctl, unsigned long arg)
 {
-	return -EINVAL;
+	struct kvm_vcpu *vcpu = filp->private_data;
+	void __user *argp = (void __user *)arg;
+	struct kvm_ia64_vcpu_stack *stack = NULL;
+	long r;
+
+	switch (ioctl) {
+	case KVM_IA64_VCPU_GET_STACK: {
+		struct kvm_ia64_vcpu_stack __user *user_stack;
+	        void __user *first_p = argp;
+
+		r = -EFAULT;
+		if (copy_from_user(&user_stack, first_p, sizeof(void *)))
+			goto out;
+
+		if (!access_ok(VERIFY_WRITE, user_stack,
+			       sizeof(struct kvm_ia64_vcpu_stack))) {
+			printk(KERN_INFO "KVM_IA64_VCPU_GET_STACK: "
+			       "Illegal user destination address for stack\n");
+			goto out;
+		}
+		stack = kzalloc(sizeof(struct kvm_ia64_vcpu_stack), GFP_KERNEL);
+		if (!stack) {
+			r = -ENOMEM;
+			goto out;
+		}
+
+		r = kvm_arch_vcpu_ioctl_get_stack(vcpu, stack);
+		if (r)
+			goto out;
+
+		if (copy_to_user(user_stack, stack,
+				 sizeof(struct kvm_ia64_vcpu_stack)))
+			goto out;
+
+		break;
+	}
+	case KVM_IA64_VCPU_SET_STACK: {
+		struct kvm_ia64_vcpu_stack __user *user_stack;
+	        void __user *first_p = argp;
+
+		r = -EFAULT;
+		if (copy_from_user(&user_stack, first_p, sizeof(void *)))
+			goto out;
+
+		if (!access_ok(VERIFY_READ, user_stack,
+			    sizeof(struct kvm_ia64_vcpu_stack))) {
+			printk(KERN_INFO "KVM_IA64_VCPU_SET_STACK: "
+			       "Illegal user address for stack\n");
+			goto out;
+		}
+		stack = kmalloc(sizeof(struct kvm_ia64_vcpu_stack), GFP_KERNEL);
+		if (!stack) {
+			r = -ENOMEM;
+			goto out;
+		}
+		if (copy_from_user(stack, user_stack,
+				   sizeof(struct kvm_ia64_vcpu_stack)))
+			goto out;
+
+		r = kvm_arch_vcpu_ioctl_set_stack(vcpu, stack);
+		break;
+	}
+
+	default:
+		r = -EINVAL;
+	}
+
+out:
+	kfree(stack);
+	return r;
 }
 
 int kvm_arch_set_memory_region(struct kvm *kvm,
@@ -1472,7 +1575,7 @@ void kvm_arch_flush_shadow(struct kvm *kvm)
 }
 
 long kvm_arch_dev_ioctl(struct file *filp,
-		unsigned int ioctl, unsigned long arg)
+			unsigned int ioctl, unsigned long arg)
 {
 	return -EINVAL;
 }
@@ -1737,7 +1840,7 @@ struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
 	struct kvm_vcpu *lvcpu = kvm->vcpus[0];
 	int i;
 
-	for (i = 1; i < KVM_MAX_VCPUS; i++) {
+	for (i = 1; i < kvm->arch.online_vcpus; i++) {
 		if (!kvm->vcpus[i])
 			continue;
 		if (lvcpu->arch.xtp > kvm->vcpus[i]->arch.xtp)

arch/ia64/kvm/kvm_fw.c

@@ -227,6 +227,18 @@ static struct ia64_pal_retval pal_proc_get_features(struct kvm_vcpu *vcpu)
 	return result;
 }
 
+static struct ia64_pal_retval pal_register_info(struct kvm_vcpu *vcpu)
+{
+
+	struct ia64_pal_retval result = {0, 0, 0, 0};
+	long in0, in1, in2, in3;
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	result.status = ia64_pal_register_info(in1, &result.v1, &result.v2);
+
+	return result;
+}
+
 static struct ia64_pal_retval pal_cache_info(struct kvm_vcpu *vcpu)
 {
 
@@ -268,8 +280,12 @@ static struct ia64_pal_retval pal_vm_summary(struct kvm_vcpu *vcpu)
 static struct ia64_pal_retval pal_vm_info(struct kvm_vcpu *vcpu)
 {
 	struct ia64_pal_retval result;
+	unsigned long in0, in1, in2, in3;
 
-	INIT_PAL_STATUS_UNIMPLEMENTED(result);
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+
+	result.status = ia64_pal_vm_info(in1, in2,
+			(pal_tc_info_u_t *)&result.v1, &result.v2);
 
 	return result;
 }
@@ -292,6 +308,108 @@ static void prepare_for_halt(struct kvm_vcpu *vcpu)
 	vcpu->arch.timer_fired = 0;
 }
 
+static struct ia64_pal_retval pal_perf_mon_info(struct kvm_vcpu *vcpu)
+{
+	long status;
+	unsigned long in0, in1, in2, in3, r9;
+	unsigned long pm_buffer[16];
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	status = ia64_pal_perf_mon_info(pm_buffer,
+				(pal_perf_mon_info_u_t *) &r9);
+	if (status != 0) {
+		printk(KERN_DEBUG"PAL_PERF_MON_INFO fails ret=%ld\n", status);
+	} else {
+		if (in1)
+			memcpy((void *)in1, pm_buffer, sizeof(pm_buffer));
+		else {
+			status = PAL_STATUS_EINVAL;
+			printk(KERN_WARNING"Invalid parameters "
+						"for PAL call:0x%lx!\n", in0);
+		}
+	}
+	return (struct ia64_pal_retval){status, r9, 0, 0};
+}
+
+static struct ia64_pal_retval pal_halt_info(struct kvm_vcpu *vcpu)
+{
+	unsigned long in0, in1, in2, in3;
+	long status;
+	unsigned long res = 1000UL | (1000UL << 16) | (10UL << 32)
+					| (1UL << 61) | (1UL << 60);
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	if (in1) {
+		memcpy((void *)in1, &res, sizeof(res));
+		status = 0;
+	} else{
+		status = PAL_STATUS_EINVAL;
+		printk(KERN_WARNING"Invalid parameters "
+					"for PAL call:0x%lx!\n", in0);
+	}
+
+	return (struct ia64_pal_retval){status, 0, 0, 0};
+}
+
+static struct ia64_pal_retval pal_mem_attrib(struct kvm_vcpu *vcpu)
+{
+	unsigned long r9;
+	long status;
+
+	status = ia64_pal_mem_attrib(&r9);
+
+	return (struct ia64_pal_retval){status, r9, 0, 0};
+}
+
+static void remote_pal_prefetch_visibility(void *v)
+{
+	s64 trans_type = (s64)v;
+	ia64_pal_prefetch_visibility(trans_type);
+}
+
+static struct ia64_pal_retval pal_prefetch_visibility(struct kvm_vcpu *vcpu)
+{
+	struct ia64_pal_retval result = {0, 0, 0, 0};
+	unsigned long in0, in1, in2, in3;
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	result.status = ia64_pal_prefetch_visibility(in1);
+	if (result.status == 0) {
+		/* Must be performed on all remote processors
+		in the coherence domain. */
+		smp_call_function(remote_pal_prefetch_visibility,
+					(void *)in1, 1);
+		/* Unnecessary on remote processor for other vcpus!*/
+		result.status = 1;
+	}
+	return result;
+}
+
+static void remote_pal_mc_drain(void *v)
+{
+	ia64_pal_mc_drain();
+}
+
+static struct ia64_pal_retval pal_get_brand_info(struct kvm_vcpu *vcpu)
+{
+	struct ia64_pal_retval result = {0, 0, 0, 0};
+	unsigned long in0, in1, in2, in3;
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+
+	if (in1 == 0 && in2) {
+		char brand_info[128];
+		result.status = ia64_pal_get_brand_info(brand_info);
+		if (result.status == PAL_STATUS_SUCCESS)
+			memcpy((void *)in2, brand_info, 128);
+	} else {
+		result.status = PAL_STATUS_REQUIRES_MEMORY;
+		printk(KERN_WARNING"Invalid parameters for "
+					"PAL call:0x%lx!\n", in0);
+	}
+
+	return result;
+}
+
 int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
 
@@ -300,14 +418,22 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	int ret = 1;
 
 	gr28 = kvm_get_pal_call_index(vcpu);
-	/*printk("pal_call index:%lx\n",gr28);*/
 	switch (gr28) {
 	case PAL_CACHE_FLUSH:
 		result = pal_cache_flush(vcpu);
 		break;
+	case PAL_MEM_ATTRIB:
+		result = pal_mem_attrib(vcpu);
+		break;
 	case PAL_CACHE_SUMMARY:
 		result = pal_cache_summary(vcpu);
 		break;
+	case PAL_PERF_MON_INFO:
+		result = pal_perf_mon_info(vcpu);
+		break;
+	case PAL_HALT_INFO:
+		result = pal_halt_info(vcpu);
+		break;
 	case PAL_HALT_LIGHT:
 	{
 		INIT_PAL_STATUS_SUCCESS(result);
@@ -317,6 +443,16 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	}
 		break;
 
+	case PAL_PREFETCH_VISIBILITY:
+		result = pal_prefetch_visibility(vcpu);
+		break;
+	case PAL_MC_DRAIN:
+		result.status = ia64_pal_mc_drain();
+		/* FIXME: All vcpus likely call PAL_MC_DRAIN.
+		   That causes the congestion. */
+		smp_call_function(remote_pal_mc_drain, NULL, 1);
+		break;
+
 	case PAL_FREQ_RATIOS:
 		result = pal_freq_ratios(vcpu);
 		break;
@@ -346,6 +482,9 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		INIT_PAL_STATUS_SUCCESS(result);
 		result.v1 = (1L << 32) | 1L;
 		break;
+	case PAL_REGISTER_INFO:
+		result = pal_register_info(vcpu);
+		break;
 	case PAL_VM_PAGE_SIZE:
 		result.status = ia64_pal_vm_page_size(&result.v0,
 							&result.v1);
@@ -365,12 +504,18 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		result.status = ia64_pal_version(
 				(pal_version_u_t *)&result.v0,
 				(pal_version_u_t *)&result.v1);
-
 		break;
 	case PAL_FIXED_ADDR:
 		result.status = PAL_STATUS_SUCCESS;
 		result.v0 = vcpu->vcpu_id;
 		break;
+	case PAL_BRAND_INFO:
+		result = pal_get_brand_info(vcpu);
+		break;
+	case PAL_GET_PSTATE:
+	case PAL_CACHE_SHARED_INFO:
+		INIT_PAL_STATUS_UNIMPLEMENTED(result);
+		break;
 	default:
 		INIT_PAL_STATUS_UNIMPLEMENTED(result);
 		printk(KERN_WARNING"kvm: Unsupported pal call,"

arch/ia64/kvm/process.c

@@ -167,7 +167,6 @@ static u64 vcpu_get_itir_on_fault(struct kvm_vcpu *vcpu, u64 ifa)
 	return (rr1.val);
 }
 
-
 /*
  * Set vIFA & vITIR & vIHA, when vPSR.ic =1
  * Parameter:
@@ -222,8 +221,6 @@ void itlb_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	inject_guest_interruption(vcpu, IA64_INST_TLB_VECTOR);
 }
 
-
-
 /*
  * Data Nested TLB Fault
  *  @ Data Nested TLB Vector
@@ -245,7 +242,6 @@ void alt_dtlb(struct kvm_vcpu *vcpu, u64 vadr)
 	inject_guest_interruption(vcpu, IA64_ALT_DATA_TLB_VECTOR);
 }
 
-
 /*
  * Data TLB Fault
  *  @ Data TLB vector
@@ -265,8 +261,6 @@ static void _vhpt_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	/* If vPSR.ic, IFA, ITIR, IHA*/
 	set_ifa_itir_iha(vcpu, vadr, 1, 1, 1);
 	inject_guest_interruption(vcpu, IA64_VHPT_TRANS_VECTOR);
-
-
 }
 
 /*
@@ -279,7 +273,6 @@ void ivhpt_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	_vhpt_fault(vcpu, vadr);
 }
 
-
 /*
  * VHPT Data Fault
  *  @ VHPT Translation vector
@@ -290,8 +283,6 @@ void dvhpt_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	_vhpt_fault(vcpu, vadr);
 }
 
-
-
 /*
  * Deal with:
  *  General Exception vector
@@ -301,7 +292,6 @@ void _general_exception(struct kvm_vcpu *vcpu)
 	inject_guest_interruption(vcpu, IA64_GENEX_VECTOR);
 }
 
-
 /*
  * Illegal Operation Fault
  *  @ General Exception Vector
@@ -419,19 +409,16 @@ static void __page_not_present(struct kvm_vcpu *vcpu, u64 vadr)
 	inject_guest_interruption(vcpu, IA64_PAGE_NOT_PRESENT_VECTOR);
 }
 
-
 void data_page_not_present(struct kvm_vcpu *vcpu, u64 vadr)
 {
 	__page_not_present(vcpu, vadr);
 }
 
-
 void inst_page_not_present(struct kvm_vcpu *vcpu, u64 vadr)
 {
 	__page_not_present(vcpu, vadr);
 }
 
-
 /* Deal with
  *  Data access rights vector
  */
@@ -563,22 +550,64 @@ void reflect_interruption(u64 ifa, u64 isr, u64 iim,
 	inject_guest_interruption(vcpu, vector);
 }
 
+static unsigned long kvm_trans_pal_call_args(struct kvm_vcpu *vcpu,
+						unsigned long arg)
+{
+	struct thash_data *data;
+	unsigned long gpa, poff;
+
+	if (!is_physical_mode(vcpu)) {
+		/* Depends on caller to provide the DTR or DTC mapping.*/
+		data = vtlb_lookup(vcpu, arg, D_TLB);
+		if (data)
+			gpa = data->page_flags & _PAGE_PPN_MASK;
+		else {
+			data = vhpt_lookup(arg);
+			if (!data)
+				return 0;
+			gpa = data->gpaddr & _PAGE_PPN_MASK;
+		}
+
+		poff = arg & (PSIZE(data->ps) - 1);
+		arg = PAGEALIGN(gpa, data->ps) | poff;
+	}
+	arg = kvm_gpa_to_mpa(arg << 1 >> 1);
+
+	return (unsigned long)__va(arg);
+}
+
 static void set_pal_call_data(struct kvm_vcpu *vcpu)
 {
 	struct exit_ctl_data *p = &vcpu->arch.exit_data;
+	unsigned long gr28 = vcpu_get_gr(vcpu, 28);
+	unsigned long gr29 = vcpu_get_gr(vcpu, 29);
+	unsigned long gr30 = vcpu_get_gr(vcpu, 30);
 
 	/*FIXME:For static and stacked convention, firmware
 	 * has put the parameters in gr28-gr31 before
 	 * break to vmm  !!*/
 
-	p->u.pal_data.gr28 = vcpu_get_gr(vcpu, 28);
-	p->u.pal_data.gr29 = vcpu_get_gr(vcpu, 29);
-	p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
+	switch (gr28) {
+	case PAL_PERF_MON_INFO:
+	case PAL_HALT_INFO:
+		p->u.pal_data.gr29 =  kvm_trans_pal_call_args(vcpu, gr29);
+		p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
+		break;
+	case PAL_BRAND_INFO:
+		p->u.pal_data.gr29 = gr29;;
+		p->u.pal_data.gr30 = kvm_trans_pal_call_args(vcpu, gr30);
+		break;
+	default:
+		p->u.pal_data.gr29 = gr29;;
+		p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
+	}
+	p->u.pal_data.gr28 = gr28;
 	p->u.pal_data.gr31 = vcpu_get_gr(vcpu, 31);
+
 	p->exit_reason = EXIT_REASON_PAL_CALL;
 }
 
-static void set_pal_call_result(struct kvm_vcpu *vcpu)
+static void get_pal_call_result(struct kvm_vcpu *vcpu)
 {
 	struct exit_ctl_data *p = &vcpu->arch.exit_data;
 
@@ -606,7 +635,7 @@ static void set_sal_call_data(struct kvm_vcpu *vcpu)
 	p->exit_reason = EXIT_REASON_SAL_CALL;
 }
 
-static void set_sal_call_result(struct kvm_vcpu *vcpu)
+static void get_sal_call_result(struct kvm_vcpu *vcpu)
 {
 	struct exit_ctl_data *p = &vcpu->arch.exit_data;
 
@@ -629,13 +658,13 @@ void  kvm_ia64_handle_break(unsigned long ifa, struct kvm_pt_regs *regs,
 		if (iim == DOMN_PAL_REQUEST) {
 			set_pal_call_data(v);
 			vmm_transition(v);
-			set_pal_call_result(v);
+			get_pal_call_result(v);
 			vcpu_increment_iip(v);
 			return;
 		} else if (iim == DOMN_SAL_REQUEST) {
 			set_sal_call_data(v);
 			vmm_transition(v);
-			set_sal_call_result(v);
+			get_sal_call_result(v);
 			vcpu_increment_iip(v);
 			return;
 		}
@@ -703,7 +732,6 @@ void vhpi_detection(struct kvm_vcpu *vcpu)
 	}
 }
 
-
 void leave_hypervisor_tail(void)
 {
 	struct kvm_vcpu *v = current_vcpu;
@@ -737,7 +765,6 @@ void leave_hypervisor_tail(void)
 	}
 }
 
-
 static inline void handle_lds(struct kvm_pt_regs *regs)
 {
 	regs->cr_ipsr |= IA64_PSR_ED;

arch/ia64/kvm/vcpu.c

@@ -112,7 +112,6 @@ void switch_to_physical_rid(struct kvm_vcpu *vcpu)
 	return;
 }
 
-
 void switch_to_virtual_rid(struct kvm_vcpu *vcpu)
 {
 	unsigned long psr;
@@ -166,8 +165,6 @@ void switch_mm_mode(struct kvm_vcpu *vcpu, struct ia64_psr old_psr,
 	return;
 }
 
-
-
 /*
  * In physical mode, insert tc/tr for region 0 and 4 uses
  * RID[0] and RID[4] which is for physical mode emulation.
@@ -269,7 +266,6 @@ static inline unsigned long fph_index(struct kvm_pt_regs *regs,
 	return rotate_reg(96, rrb_fr, (regnum - IA64_FIRST_ROTATING_FR));
 }
 
-
 /*
  * The inverse of the above: given bspstore and the number of
  * registers, calculate ar.bsp.
@@ -811,12 +807,15 @@ static inline void vcpu_set_itm(struct kvm_vcpu *vcpu, u64 val);
 static void vcpu_set_itc(struct kvm_vcpu *vcpu, u64 val)
 {
 	struct kvm_vcpu *v;
+	struct kvm *kvm;
 	int i;
 	long itc_offset = val - ia64_getreg(_IA64_REG_AR_ITC);
 	unsigned long vitv = VCPU(vcpu, itv);
 
+	kvm = (struct kvm *)KVM_VM_BASE;
+
 	if (vcpu->vcpu_id == 0) {
-		for (i = 0; i < KVM_MAX_VCPUS; i++) {
+		for (i = 0; i < kvm->arch.online_vcpus; i++) {
 			v = (struct kvm_vcpu *)((char *)vcpu +
 					sizeof(struct kvm_vcpu_data) * i);
 			VMX(v, itc_offset) = itc_offset;
@@ -1039,8 +1038,6 @@ u64 vcpu_tak(struct kvm_vcpu *vcpu, u64 vadr)
 	return key;
 }
 
-
-
 void kvm_thash(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long thash, vadr;
@@ -1050,7 +1047,6 @@ void kvm_thash(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_gr(vcpu, inst.M46.r1, thash, 0);
 }
 
-
 void kvm_ttag(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long tag, vadr;
@@ -1131,7 +1127,6 @@ int vcpu_tpa(struct kvm_vcpu *vcpu, u64 vadr, u64 *padr)
 	return IA64_NO_FAULT;
 }
 
-
 int kvm_tpa(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r1, r3;
@@ -1154,7 +1149,6 @@ void kvm_tak(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_gr(vcpu, inst.M46.r1, r1, 0);
 }
 
-
 /************************************
  * Insert/Purge translation register/cache
  ************************************/
@@ -1385,7 +1379,6 @@ void kvm_mov_to_ar_reg(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_itc(vcpu, r2);
 }
 
-
 void kvm_mov_from_ar_reg(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r1;
@@ -1393,8 +1386,9 @@ void kvm_mov_from_ar_reg(struct kvm_vcpu *vcpu, INST64 inst)
 	r1 = vcpu_get_itc(vcpu);
 	vcpu_set_gr(vcpu, inst.M31.r1, r1, 0);
 }
+
 /**************************************************************************
-  struct kvm_vcpu*protection key register access routines
+  struct kvm_vcpu protection key register access routines
  **************************************************************************/
 
 unsigned long vcpu_get_pkr(struct kvm_vcpu *vcpu, unsigned long reg)
@@ -1407,20 +1401,6 @@ void vcpu_set_pkr(struct kvm_vcpu *vcpu, unsigned long reg, unsigned long val)
 	ia64_set_pkr(reg, val);
 }
 
-
-unsigned long vcpu_get_itir_on_fault(struct kvm_vcpu *vcpu, unsigned long ifa)
-{
-	union ia64_rr rr, rr1;
-
-	rr.val = vcpu_get_rr(vcpu, ifa);
-	rr1.val = 0;
-	rr1.ps = rr.ps;
-	rr1.rid = rr.rid;
-	return (rr1.val);
-}
-
-
-
 /********************************
  * Moves to privileged registers
  ********************************/
@@ -1464,8 +1444,6 @@ unsigned long vcpu_set_rr(struct kvm_vcpu *vcpu, unsigned long reg,
 	return (IA64_NO_FAULT);
 }
 
-
-
 void kvm_mov_to_rr(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r3, r2;
@@ -1510,8 +1488,6 @@ void kvm_mov_to_pkr(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_pkr(vcpu, r3, r2);
 }
 
-
-
 void kvm_mov_from_rr(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r3, r1;
@@ -1557,7 +1533,6 @@ void kvm_mov_from_pmc(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_gr(vcpu, inst.M43.r1, r1, 0);
 }
 
-
 unsigned long vcpu_get_cpuid(struct kvm_vcpu *vcpu, unsigned long reg)
 {
 	/* FIXME: This could get called as a result of a rsvd-reg fault */
@@ -1609,7 +1584,6 @@ unsigned long kvm_mov_to_cr(struct kvm_vcpu *vcpu, INST64 inst)
 	return 0;
 }
 
-
 unsigned long kvm_mov_from_cr(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long tgt = inst.M33.r1;
@@ -1633,8 +1607,6 @@ unsigned long kvm_mov_from_cr(struct kvm_vcpu *vcpu, INST64 inst)
 	return 0;
 }
 
-
-
 void vcpu_set_psr(struct kvm_vcpu *vcpu, unsigned long val)
 {
 
@@ -1776,9 +1748,6 @@ void vcpu_bsw1(struct kvm_vcpu *vcpu)
 	}
 }
 
-
-
-
 void vcpu_rfi(struct kvm_vcpu *vcpu)
 {
 	unsigned long ifs, psr;
@@ -1796,7 +1765,6 @@ void vcpu_rfi(struct kvm_vcpu *vcpu)
 	regs->cr_iip = VCPU(vcpu, iip);
 }
 
-
 /*
    VPSR can't keep track of below bits of guest PSR
    This function gets guest PSR

arch/ia64/kvm/vcpu.h

@@ -703,7 +703,7 @@ extern u64 guest_vhpt_lookup(u64 iha, u64 *pte);
 extern void thash_purge_entries(struct kvm_vcpu *v, u64 va, u64 ps);
 extern void thash_purge_entries_remote(struct kvm_vcpu *v, u64 va, u64 ps);
 extern u64 translate_phy_pte(u64 *pte, u64 itir, u64 va);
-extern int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte,
+extern void thash_purge_and_insert(struct kvm_vcpu *v, u64 pte,
 		u64 itir, u64 ifa, int type);
 extern void thash_purge_all(struct kvm_vcpu *v);
 extern struct thash_data *vtlb_lookup(struct kvm_vcpu *v,
@@ -738,7 +738,7 @@ void kvm_init_vhpt(struct kvm_vcpu *v);
 void thash_init(struct thash_cb *hcb, u64 sz);
 
 void panic_vm(struct kvm_vcpu *v, const char *fmt, ...);
-
+u64 kvm_gpa_to_mpa(u64 gpa);
 extern u64 ia64_call_vsa(u64 proc, u64 arg1, u64 arg2, u64 arg3,
 		u64 arg4, u64 arg5, u64 arg6, u64 arg7);
 

arch/ia64/kvm/vtlb.c

@@ -164,11 +164,11 @@ static void vhpt_insert(u64 pte, u64 itir, u64 ifa, u64 gpte)
 	unsigned long ps, gpaddr;
 
 	ps = itir_ps(itir);
+	rr.val = ia64_get_rr(ifa);
 
-	gpaddr = ((gpte & _PAGE_PPN_MASK) >> ps << ps) |
-		(ifa & ((1UL << ps) - 1));
+	 gpaddr = ((gpte & _PAGE_PPN_MASK) >> ps << ps) |
+					(ifa & ((1UL << ps) - 1));
 
-	rr.val = ia64_get_rr(ifa);
 	head = (struct thash_data *)ia64_thash(ifa);
 	head->etag = INVALID_TI_TAG;
 	ia64_mf();
@@ -412,16 +412,14 @@ u64 translate_phy_pte(u64 *pte, u64 itir, u64 va)
 
 /*
  * Purge overlap TCs and then insert the new entry to emulate itc ops.
- *    Notes: Only TC entry can purge and insert.
- *    1 indicates this is MMIO
+ * Notes: Only TC entry can purge and insert.
  */
-int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
+void  thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
 						u64 ifa, int type)
 {
 	u64 ps;
 	u64 phy_pte, io_mask, index;
 	union ia64_rr vrr, mrr;
-	int ret = 0;
 
 	ps = itir_ps(itir);
 	vrr.val = vcpu_get_rr(v, ifa);
@@ -441,25 +439,19 @@ int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
 		phy_pte &= ~_PAGE_MA_MASK;
 	}
 
-	if (pte & VTLB_PTE_IO)
-		ret = 1;
-
 	vtlb_purge(v, ifa, ps);
 	vhpt_purge(v, ifa, ps);
 
-	if (ps == mrr.ps) {
-		if (!(pte&VTLB_PTE_IO)) {
-			vhpt_insert(phy_pte, itir, ifa, pte);
-		} else {
-			vtlb_insert(v, pte, itir, ifa);
-			vcpu_quick_region_set(VMX(v, tc_regions), ifa);
-		}
-	} else if (ps > mrr.ps) {
+	if ((ps != mrr.ps) || (pte & VTLB_PTE_IO)) {
 		vtlb_insert(v, pte, itir, ifa);
 		vcpu_quick_region_set(VMX(v, tc_regions), ifa);
-		if (!(pte&VTLB_PTE_IO))
-			vhpt_insert(phy_pte, itir, ifa, pte);
-	} else {
+	}
+	if (pte & VTLB_PTE_IO)
+		return;
+
+	if (ps >= mrr.ps)
+		vhpt_insert(phy_pte, itir, ifa, pte);
+	else {
 		u64 psr;
 		phy_pte  &= ~PAGE_FLAGS_RV_MASK;
 		psr = ia64_clear_ic();
@@ -469,7 +461,6 @@ int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
 	if (!(pte&VTLB_PTE_IO))
 		mark_pages_dirty(v, pte, ps);
 
-	return ret;
 }
 
 /*
@@ -509,7 +500,6 @@ void thash_purge_all(struct kvm_vcpu *v)
 	local_flush_tlb_all();
 }
 
-
 /*
  * Lookup the hash table and its collision chain to find an entry
  * covering this address rid:va or the entry.
@@ -517,7 +507,6 @@ void thash_purge_all(struct kvm_vcpu *v)
  * INPUT:
  *  in: TLB format for both VHPT & TLB.
  */
-
 struct thash_data *vtlb_lookup(struct kvm_vcpu *v, u64 va, int is_data)
 {
 	struct thash_data  *cch;
@@ -547,7 +536,6 @@ struct thash_data *vtlb_lookup(struct kvm_vcpu *v, u64 va, int is_data)
 	return NULL;
 }
 
-
 /*
  * Initialize internal control data before service.
  */
@@ -573,6 +561,10 @@ void thash_init(struct thash_cb *hcb, u64 sz)
 u64 kvm_get_mpt_entry(u64 gpfn)
 {
 	u64 *base = (u64 *) KVM_P2M_BASE;
+
+	if (gpfn >= (KVM_P2M_SIZE >> 3))
+		panic_vm(current_vcpu, "Invalid gpfn =%lx\n", gpfn);
+
 	return *(base + gpfn);
 }
 
@@ -589,7 +581,6 @@ u64 kvm_gpa_to_mpa(u64 gpa)
 	return (pte >> PAGE_SHIFT << PAGE_SHIFT) | (gpa & ~PAGE_MASK);
 }
 
-
 /*
  * Fetch guest bundle code.
  * INPUT:
@@ -631,7 +622,6 @@ int fetch_code(struct kvm_vcpu *vcpu, u64 gip, IA64_BUNDLE *pbundle)
 	return IA64_NO_FAULT;
 }
 
-
 void kvm_init_vhpt(struct kvm_vcpu *v)
 {
 	v->arch.vhpt.num = VHPT_NUM_ENTRIES;

arch/powerpc/include/asm/kvm.h

@@ -52,4 +52,11 @@ struct kvm_fpu {
 	__u64 fpr[32];
 };
 
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif /* __LINUX_KVM_POWERPC_H */

arch/powerpc/include/asm/kvm_44x.h

@@ -28,6 +28,13 @@
  * need to find some way of advertising it. */
 #define KVM44x_GUEST_TLB_SIZE 64
 
+struct kvmppc_44x_tlbe {
+	u32 tid; /* Only the low 8 bits are used. */
+	u32 word0;
+	u32 word1;
+	u32 word2;
+};
+
 struct kvmppc_44x_shadow_ref {
 	struct page *page;
 	u16 gtlb_index;

arch/powerpc/include/asm/kvm_asm.h

@@ -42,7 +42,12 @@
 #define BOOKE_INTERRUPT_DTLB_MISS 13
 #define BOOKE_INTERRUPT_ITLB_MISS 14
 #define BOOKE_INTERRUPT_DEBUG 15
-#define BOOKE_MAX_INTERRUPT 15
+
+/* E500 */
+#define BOOKE_INTERRUPT_SPE_UNAVAIL 32
+#define BOOKE_INTERRUPT_SPE_FP_DATA 33
+#define BOOKE_INTERRUPT_SPE_FP_ROUND 34
+#define BOOKE_INTERRUPT_PERFORMANCE_MONITOR 35
 
 #define RESUME_FLAG_NV          (1<<0)  /* Reload guest nonvolatile state? */
 #define RESUME_FLAG_HOST        (1<<1)  /* Resume host? */

arch/powerpc/include/asm/kvm_e500.h

@@ -0,0 +1,67 @@
+/*
+ * Copyright (C) 2008 Freescale Semiconductor, Inc. All rights reserved.
+ *
+ * Author: Yu Liu, <yu.liu@freescale.com>
+ *
+ * Description:
+ * This file is derived from arch/powerpc/include/asm/kvm_44x.h,
+ * by Hollis Blanchard <hollisb@us.ibm.com>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_KVM_E500_H__
+#define __ASM_KVM_E500_H__
+
+#include <linux/kvm_host.h>
+
+#define BOOKE_INTERRUPT_SIZE 36
+
+#define E500_PID_NUM   3
+#define E500_TLB_NUM   2
+
+struct tlbe{
+	u32 mas1;
+	u32 mas2;
+	u32 mas3;
+	u32 mas7;
+};
+
+struct kvmppc_vcpu_e500 {
+	/* Unmodified copy of the guest's TLB. */
+	struct tlbe *guest_tlb[E500_TLB_NUM];
+	/* TLB that's actually used when the guest is running. */
+	struct tlbe *shadow_tlb[E500_TLB_NUM];
+	/* Pages which are referenced in the shadow TLB. */
+	struct page **shadow_pages[E500_TLB_NUM];
+
+	unsigned int guest_tlb_size[E500_TLB_NUM];
+	unsigned int shadow_tlb_size[E500_TLB_NUM];
+	unsigned int guest_tlb_nv[E500_TLB_NUM];
+
+	u32 host_pid[E500_PID_NUM];
+	u32 pid[E500_PID_NUM];
+
+	u32 mas0;
+	u32 mas1;
+	u32 mas2;
+	u32 mas3;
+	u32 mas4;
+	u32 mas5;
+	u32 mas6;
+	u32 mas7;
+	u32 l1csr1;
+	u32 hid0;
+	u32 hid1;
+
+	struct kvm_vcpu vcpu;
+};
+
+static inline struct kvmppc_vcpu_e500 *to_e500(struct kvm_vcpu *vcpu)
+{
+	return container_of(vcpu, struct kvmppc_vcpu_e500, vcpu);
+}
+
+#endif /* __ASM_KVM_E500_H__ */

arch/powerpc/include/asm/kvm_host.h

@@ -64,13 +64,6 @@ struct kvm_vcpu_stat {
 	u32 halt_wakeup;
 };
 
-struct kvmppc_44x_tlbe {
-	u32 tid; /* Only the low 8 bits are used. */
-	u32 word0;
-	u32 word1;
-	u32 word2;
-};
-
 enum kvm_exit_types {
 	MMIO_EXITS,
 	DCR_EXITS,
@@ -118,11 +111,6 @@ struct kvm_arch {
 struct kvm_vcpu_arch {
 	u32 host_stack;
 	u32 host_pid;
-	u32 host_dbcr0;
-	u32 host_dbcr1;
-	u32 host_dbcr2;
-	u32 host_iac[4];
-	u32 host_msr;
 
 	u64 fpr[32];
 	ulong gpr[32];
@@ -157,7 +145,7 @@ struct kvm_vcpu_arch {
 	u32 tbu;
 	u32 tcr;
 	u32 tsr;
-	u32 ivor[16];
+	u32 ivor[64];
 	ulong ivpr;
 	u32 pir;
 
@@ -170,6 +158,7 @@ struct kvm_vcpu_arch {
 	u32 ccr1;
 	u32 dbcr0;
 	u32 dbcr1;
+	u32 dbsr;
 
 #ifdef CONFIG_KVM_EXIT_TIMING
 	struct kvmppc_exit_timing timing_exit;
@@ -200,10 +189,4 @@ struct kvm_vcpu_arch {
 	unsigned long pending_exceptions;
 };
 
-struct kvm_guest_debug {
-	int enabled;
-	unsigned long bp[4];
-	int singlestep;
-};
-
 #endif /* __POWERPC_KVM_HOST_H__ */

arch/powerpc/include/asm/kvm_ppc.h

@@ -52,13 +52,19 @@ extern int kvmppc_emulate_instruction(struct kvm_run *run,
 extern int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu);
 extern void kvmppc_emulate_dec(struct kvm_vcpu *vcpu);
 
+/* Core-specific hooks */
+
 extern void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gpa_t gpaddr,
-                           u64 asid, u32 flags, u32 max_bytes,
                            unsigned int gtlb_idx);
 extern void kvmppc_mmu_priv_switch(struct kvm_vcpu *vcpu, int usermode);
 extern void kvmppc_mmu_switch_pid(struct kvm_vcpu *vcpu, u32 pid);
-
-/* Core-specific hooks */
+extern void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu);
+extern int kvmppc_mmu_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr);
+extern int kvmppc_mmu_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr);
+extern gpa_t kvmppc_mmu_xlate(struct kvm_vcpu *vcpu, unsigned int gtlb_index,
+                              gva_t eaddr);
+extern void kvmppc_mmu_dtlb_miss(struct kvm_vcpu *vcpu);
+extern void kvmppc_mmu_itlb_miss(struct kvm_vcpu *vcpu);
 
 extern struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm,
                                                 unsigned int id);
@@ -71,9 +77,6 @@ extern int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
 extern void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
 extern void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu);
 
-extern void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu);
-extern void kvmppc_core_load_host_debugstate(struct kvm_vcpu *vcpu);
-
 extern void kvmppc_core_deliver_interrupts(struct kvm_vcpu *vcpu);
 extern int kvmppc_core_pending_dec(struct kvm_vcpu *vcpu);
 extern void kvmppc_core_queue_program(struct kvm_vcpu *vcpu);

arch/powerpc/include/asm/mmu-fsl-booke.h

@@ -75,6 +75,8 @@
 
 #ifndef __ASSEMBLY__
 
+extern unsigned int tlbcam_index;
+
 typedef struct {
 	unsigned int	id;
 	unsigned int	active;

arch/powerpc/kernel/asm-offsets.c

@@ -49,7 +49,7 @@
 #include <asm/iseries/alpaca.h>
 #endif
 #ifdef CONFIG_KVM
-#include <asm/kvm_44x.h>
+#include <linux/kvm_host.h>
 #endif
 
 #if defined(CONFIG_BOOKE) || defined(CONFIG_40x)
@@ -361,8 +361,6 @@ int main(void)
 	DEFINE(PTE_SIZE, sizeof(pte_t));
 
 #ifdef CONFIG_KVM
-	DEFINE(TLBE_BYTES, sizeof(struct kvmppc_44x_tlbe));
-
 	DEFINE(VCPU_HOST_STACK, offsetof(struct kvm_vcpu, arch.host_stack));
 	DEFINE(VCPU_HOST_PID, offsetof(struct kvm_vcpu, arch.host_pid));
 	DEFINE(VCPU_GPRS, offsetof(struct kvm_vcpu, arch.gpr));

arch/powerpc/kvm/44x.c

@@ -28,72 +28,6 @@
 
 #include "44x_tlb.h"
 
-/* Note: clearing MSR[DE] just means that the debug interrupt will not be
- * delivered *immediately*. Instead, it simply sets the appropriate DBSR bits.
- * If those DBSR bits are still set when MSR[DE] is re-enabled, the interrupt
- * will be delivered as an "imprecise debug event" (which is indicated by
- * DBSR[IDE].
- */
-static void kvm44x_disable_debug_interrupts(void)
-{
-	mtmsr(mfmsr() & ~MSR_DE);
-}
-
-void kvmppc_core_load_host_debugstate(struct kvm_vcpu *vcpu)
-{
-	kvm44x_disable_debug_interrupts();
-
-	mtspr(SPRN_IAC1, vcpu->arch.host_iac[0]);
-	mtspr(SPRN_IAC2, vcpu->arch.host_iac[1]);
-	mtspr(SPRN_IAC3, vcpu->arch.host_iac[2]);
-	mtspr(SPRN_IAC4, vcpu->arch.host_iac[3]);
-	mtspr(SPRN_DBCR1, vcpu->arch.host_dbcr1);
-	mtspr(SPRN_DBCR2, vcpu->arch.host_dbcr2);
-	mtspr(SPRN_DBCR0, vcpu->arch.host_dbcr0);
-	mtmsr(vcpu->arch.host_msr);
-}
-
-void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu)
-{
-	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
-	u32 dbcr0 = 0;
-
-	vcpu->arch.host_msr = mfmsr();
-	kvm44x_disable_debug_interrupts();
-
-	/* Save host debug register state. */
-	vcpu->arch.host_iac[0] = mfspr(SPRN_IAC1);
-	vcpu->arch.host_iac[1] = mfspr(SPRN_IAC2);
-	vcpu->arch.host_iac[2] = mfspr(SPRN_IAC3);
-	vcpu->arch.host_iac[3] = mfspr(SPRN_IAC4);
-	vcpu->arch.host_dbcr0 = mfspr(SPRN_DBCR0);
-	vcpu->arch.host_dbcr1 = mfspr(SPRN_DBCR1);
-	vcpu->arch.host_dbcr2 = mfspr(SPRN_DBCR2);
-
-	/* set registers up for guest */
-
-	if (dbg->bp[0]) {
-		mtspr(SPRN_IAC1, dbg->bp[0]);
-		dbcr0 |= DBCR0_IAC1 | DBCR0_IDM;
-	}
-	if (dbg->bp[1]) {
-		mtspr(SPRN_IAC2, dbg->bp[1]);
-		dbcr0 |= DBCR0_IAC2 | DBCR0_IDM;
-	}
-	if (dbg->bp[2]) {
-		mtspr(SPRN_IAC3, dbg->bp[2]);
-		dbcr0 |= DBCR0_IAC3 | DBCR0_IDM;
-	}
-	if (dbg->bp[3]) {
-		mtspr(SPRN_IAC4, dbg->bp[3]);
-		dbcr0 |= DBCR0_IAC4 | DBCR0_IDM;
-	}
-
-	mtspr(SPRN_DBCR0, dbcr0);
-	mtspr(SPRN_DBCR1, 0);
-	mtspr(SPRN_DBCR2, 0);
-}
-
 void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	kvmppc_44x_tlb_load(vcpu);
@@ -149,8 +83,6 @@ int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu)
 int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
                                struct kvm_translation *tr)
 {
-	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
-	struct kvmppc_44x_tlbe *gtlbe;
 	int index;
 	gva_t eaddr;
 	u8 pid;
@@ -166,9 +98,7 @@ int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
 		return 0;
 	}
 
-	gtlbe = &vcpu_44x->guest_tlb[index];
-
-	tr->physical_address = tlb_xlate(gtlbe, eaddr);
+	tr->physical_address = kvmppc_mmu_xlate(vcpu, index, eaddr);
 	/* XXX what does "writeable" and "usermode" even mean? */
 	tr->valid = 1;
 

arch/powerpc/kvm/44x_emulate.c

@@ -27,25 +27,12 @@
 #include "booke.h"
 #include "44x_tlb.h"
 
-#define OP_RFI      19
-
-#define XOP_RFI     50
-#define XOP_MFMSR   83
-#define XOP_WRTEE   131
-#define XOP_MTMSR   146
-#define XOP_WRTEEI  163
 #define XOP_MFDCR   323
 #define XOP_MTDCR   451
 #define XOP_TLBSX   914
 #define XOP_ICCCI   966
 #define XOP_TLBWE   978
 
-static void kvmppc_emul_rfi(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.pc = vcpu->arch.srr0;
-	kvmppc_set_msr(vcpu, vcpu->arch.srr1);
-}
-
 int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
                            unsigned int inst, int *advance)
 {
@@ -59,48 +46,9 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	int ws;
 
 	switch (get_op(inst)) {
-	case OP_RFI:
-		switch (get_xop(inst)) {
-		case XOP_RFI:
-			kvmppc_emul_rfi(vcpu);
-			kvmppc_set_exit_type(vcpu, EMULATED_RFI_EXITS);
-			*advance = 0;
-			break;
-
-		default:
-			emulated = EMULATE_FAIL;
-			break;
-		}
-		break;
-
 	case 31:
 		switch (get_xop(inst)) {
 
-		case XOP_MFMSR:
-			rt = get_rt(inst);
-			vcpu->arch.gpr[rt] = vcpu->arch.msr;
-			kvmppc_set_exit_type(vcpu, EMULATED_MFMSR_EXITS);
-			break;
-
-		case XOP_MTMSR:
-			rs = get_rs(inst);
-			kvmppc_set_exit_type(vcpu, EMULATED_MTMSR_EXITS);
-			kvmppc_set_msr(vcpu, vcpu->arch.gpr[rs]);
-			break;
-
-		case XOP_WRTEE:
-			rs = get_rs(inst);
-			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
-							 | (vcpu->arch.gpr[rs] & MSR_EE);
-			kvmppc_set_exit_type(vcpu, EMULATED_WRTEE_EXITS);
-			break;
-
-		case XOP_WRTEEI:
-			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
-							 | (inst & MSR_EE);
-			kvmppc_set_exit_type(vcpu, EMULATED_WRTEE_EXITS);
-			break;
-
 		case XOP_MFDCR:
 			dcrn = get_dcrn(inst);
 			rt = get_rt(inst);
@@ -186,186 +134,51 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		emulated = EMULATE_FAIL;
 	}
 
+	if (emulated == EMULATE_FAIL)
+		emulated = kvmppc_booke_emulate_op(run, vcpu, inst, advance);
+
 	return emulated;
 }
 
 int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs)
 {
+	int emulated = EMULATE_DONE;
+
 	switch (sprn) {
-	case SPRN_MMUCR:
-		vcpu->arch.mmucr = vcpu->arch.gpr[rs]; break;
 	case SPRN_PID:
 		kvmppc_set_pid(vcpu, vcpu->arch.gpr[rs]); break;
+	case SPRN_MMUCR:
+		vcpu->arch.mmucr = vcpu->arch.gpr[rs]; break;
 	case SPRN_CCR0:
 		vcpu->arch.ccr0 = vcpu->arch.gpr[rs]; break;
 	case SPRN_CCR1:
 		vcpu->arch.ccr1 = vcpu->arch.gpr[rs]; break;
-	case SPRN_DEAR:
-		vcpu->arch.dear = vcpu->arch.gpr[rs]; break;
-	case SPRN_ESR:
-		vcpu->arch.esr = vcpu->arch.gpr[rs]; break;
-	case SPRN_DBCR0:
-		vcpu->arch.dbcr0 = vcpu->arch.gpr[rs]; break;
-	case SPRN_DBCR1:
-		vcpu->arch.dbcr1 = vcpu->arch.gpr[rs]; break;
-	case SPRN_TSR:
-		vcpu->arch.tsr &= ~vcpu->arch.gpr[rs]; break;
-	case SPRN_TCR:
-		vcpu->arch.tcr = vcpu->arch.gpr[rs];
-		kvmppc_emulate_dec(vcpu);
-		break;
-
-	/* Note: SPRG4-7 are user-readable. These values are
-	 * loaded into the real SPRGs when resuming the
-	 * guest. */
-	case SPRN_SPRG4:
-		vcpu->arch.sprg4 = vcpu->arch.gpr[rs]; break;
-	case SPRN_SPRG5:
-		vcpu->arch.sprg5 = vcpu->arch.gpr[rs]; break;
-	case SPRN_SPRG6:
-		vcpu->arch.sprg6 = vcpu->arch.gpr[rs]; break;
-	case SPRN_SPRG7:
-		vcpu->arch.sprg7 = vcpu->arch.gpr[rs]; break;
-
-	case SPRN_IVPR:
-		vcpu->arch.ivpr = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR0:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_CRITICAL] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR1:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_MACHINE_CHECK] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR2:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_DATA_STORAGE] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR3:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_INST_STORAGE] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR4:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_EXTERNAL] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR5:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_ALIGNMENT] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR6:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_PROGRAM] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR7:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_FP_UNAVAIL] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR8:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_SYSCALL] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR9:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_AP_UNAVAIL] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR10:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_DECREMENTER] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR11:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_FIT] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR12:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_WATCHDOG] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR13:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR14:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS] = vcpu->arch.gpr[rs];
-		break;
-	case SPRN_IVOR15:
-		vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG] = vcpu->arch.gpr[rs];
-		break;
-
 	default:
-		return EMULATE_FAIL;
+		emulated = kvmppc_booke_emulate_mtspr(vcpu, sprn, rs);
 	}
 
 	kvmppc_set_exit_type(vcpu, EMULATED_MTSPR_EXITS);
-	return EMULATE_DONE;
+	return emulated;
 }
 
 int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt)
 {
+	int emulated = EMULATE_DONE;
+
 	switch (sprn) {
-	/* 440 */
+	case SPRN_PID:
+		vcpu->arch.gpr[rt] = vcpu->arch.pid; break;
 	case SPRN_MMUCR:
 		vcpu->arch.gpr[rt] = vcpu->arch.mmucr; break;
 	case SPRN_CCR0:
 		vcpu->arch.gpr[rt] = vcpu->arch.ccr0; break;
 	case SPRN_CCR1:
 		vcpu->arch.gpr[rt] = vcpu->arch.ccr1; break;
-
-	/* Book E */
-	case SPRN_PID:
-		vcpu->arch.gpr[rt] = vcpu->arch.pid; break;
-	case SPRN_IVPR:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivpr; break;
-	case SPRN_DEAR:
-		vcpu->arch.gpr[rt] = vcpu->arch.dear; break;
-	case SPRN_ESR:
-		vcpu->arch.gpr[rt] = vcpu->arch.esr; break;
-	case SPRN_DBCR0:
-		vcpu->arch.gpr[rt] = vcpu->arch.dbcr0; break;
-	case SPRN_DBCR1:
-		vcpu->arch.gpr[rt] = vcpu->arch.dbcr1; break;
-
-	case SPRN_IVOR0:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_CRITICAL];
-		break;
-	case SPRN_IVOR1:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_MACHINE_CHECK];
-		break;
-	case SPRN_IVOR2:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DATA_STORAGE];
-		break;
-	case SPRN_IVOR3:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_INST_STORAGE];
-		break;
-	case SPRN_IVOR4:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_EXTERNAL];
-		break;
-	case SPRN_IVOR5:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_ALIGNMENT];
-		break;
-	case SPRN_IVOR6:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_PROGRAM];
-		break;
-	case SPRN_IVOR7:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_FP_UNAVAIL];
-		break;
-	case SPRN_IVOR8:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_SYSCALL];
-		break;
-	case SPRN_IVOR9:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_AP_UNAVAIL];
-		break;
-	case SPRN_IVOR10:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DECREMENTER];
-		break;
-	case SPRN_IVOR11:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_FIT];
-		break;
-	case SPRN_IVOR12:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_WATCHDOG];
-		break;
-	case SPRN_IVOR13:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS];
-		break;
-	case SPRN_IVOR14:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS];
-		break;
-	case SPRN_IVOR15:
-		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG];
-		break;
-
 	default:
-		return EMULATE_FAIL;
+		emulated = kvmppc_booke_emulate_mfspr(vcpu, sprn, rt);
 	}
 
 	kvmppc_set_exit_type(vcpu, EMULATED_MFSPR_EXITS);
-	return EMULATE_DONE;
+	return emulated;
 }
 

arch/powerpc/kvm/44x_tlb.c

@@ -208,20 +208,38 @@ int kvmppc_44x_tlb_index(struct kvm_vcpu *vcpu, gva_t eaddr, unsigned int pid,
 	return -1;
 }
 
-int kvmppc_44x_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
+gpa_t kvmppc_mmu_xlate(struct kvm_vcpu *vcpu, unsigned int gtlb_index,
+                       gva_t eaddr)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	struct kvmppc_44x_tlbe *gtlbe = &vcpu_44x->guest_tlb[gtlb_index];
+	unsigned int pgmask = get_tlb_bytes(gtlbe) - 1;
+
+	return get_tlb_raddr(gtlbe) | (eaddr & pgmask);
+}
+
+int kvmppc_mmu_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
 {
 	unsigned int as = !!(vcpu->arch.msr & MSR_IS);
 
 	return kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as);
 }
 
-int kvmppc_44x_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
+int kvmppc_mmu_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
 {
 	unsigned int as = !!(vcpu->arch.msr & MSR_DS);
 
 	return kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as);
 }
 
+void kvmppc_mmu_itlb_miss(struct kvm_vcpu *vcpu)
+{
+}
+
+void kvmppc_mmu_dtlb_miss(struct kvm_vcpu *vcpu)
+{
+}
+
 static void kvmppc_44x_shadow_release(struct kvmppc_vcpu_44x *vcpu_44x,
                                       unsigned int stlb_index)
 {
@@ -248,7 +266,7 @@ static void kvmppc_44x_shadow_release(struct kvmppc_vcpu_44x *vcpu_44x,
 	KVMTRACE_1D(STLB_INVAL, &vcpu_44x->vcpu, stlb_index, handler);
 }
 
-void kvmppc_core_destroy_mmu(struct kvm_vcpu *vcpu)
+void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 	int i;
@@ -269,15 +287,19 @@ void kvmppc_core_destroy_mmu(struct kvm_vcpu *vcpu)
  * Caller must ensure that the specified guest TLB entry is safe to insert into
  * the shadow TLB.
  */
-void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gpa_t gpaddr, u64 asid,
-                    u32 flags, u32 max_bytes, unsigned int gtlb_index)
+void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gpa_t gpaddr,
+                    unsigned int gtlb_index)
 {
 	struct kvmppc_44x_tlbe stlbe;
 	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	struct kvmppc_44x_tlbe *gtlbe = &vcpu_44x->guest_tlb[gtlb_index];
 	struct kvmppc_44x_shadow_ref *ref;
 	struct page *new_page;
 	hpa_t hpaddr;
 	gfn_t gfn;
+	u32 asid = gtlbe->tid;
+	u32 flags = gtlbe->word2;
+	u32 max_bytes = get_tlb_bytes(gtlbe);
 	unsigned int victim;
 
 	/* Select TLB entry to clobber. Indirectly guard against races with the TLB
@@ -448,10 +470,8 @@ int kvmppc_44x_emul_tlbwe(struct kvm_vcpu *vcpu, u8 ra, u8 rs, u8 ws)
 	}
 
 	if (tlbe_is_host_safe(vcpu, tlbe)) {
-		u64 asid;
 		gva_t eaddr;
 		gpa_t gpaddr;
-		u32 flags;
 		u32 bytes;
 
 		eaddr = get_tlb_eaddr(tlbe);
@@ -462,10 +482,7 @@ int kvmppc_44x_emul_tlbwe(struct kvm_vcpu *vcpu, u8 ra, u8 rs, u8 ws)
 		eaddr &= ~(bytes - 1);
 		gpaddr &= ~(bytes - 1);
 
-		asid = (tlbe->word0 & PPC44x_TLB_TS) | tlbe->tid;
-		flags = tlbe->word2 & 0xffff;
-
-		kvmppc_mmu_map(vcpu, eaddr, gpaddr, asid, flags, bytes, gtlb_index);
+		kvmppc_mmu_map(vcpu, eaddr, gpaddr, gtlb_index);
 	}
 
 	KVMTRACE_5D(GTLB_WRITE, vcpu, gtlb_index, tlbe->tid, tlbe->word0,

arch/powerpc/kvm/44x_tlb.h

@@ -25,8 +25,6 @@
 
 extern int kvmppc_44x_tlb_index(struct kvm_vcpu *vcpu, gva_t eaddr,
                                 unsigned int pid, unsigned int as);
-extern int kvmppc_44x_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr);
-extern int kvmppc_44x_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr);
 
 extern int kvmppc_44x_emul_tlbsx(struct kvm_vcpu *vcpu, u8 rt, u8 ra, u8 rb,
                                  u8 rc);
@@ -85,11 +83,4 @@ static inline unsigned int get_mmucr_sts(const struct kvm_vcpu *vcpu)
 	return (vcpu->arch.mmucr >> 16) & 0x1;
 }
 
-static inline gpa_t tlb_xlate(struct kvmppc_44x_tlbe *tlbe, gva_t eaddr)
-{
-	unsigned int pgmask = get_tlb_bytes(tlbe) - 1;
-
-	return get_tlb_raddr(tlbe) | (eaddr & pgmask);
-}
-
 #endif /* __KVM_POWERPC_TLB_H__ */

arch/powerpc/kvm/Kconfig

@@ -2,6 +2,9 @@
 # KVM configuration
 #
 
+config HAVE_KVM_IRQCHIP
+       bool
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	---help---
@@ -43,6 +46,19 @@ config KVM_EXIT_TIMING
 
 	  If unsure, say N.
 
+config KVM_E500
+	bool "KVM support for PowerPC E500 processors"
+	depends on EXPERIMENTAL && E500
+	select KVM
+	---help---
+	  Support running unmodified E500 guest kernels in virtual machines on
+	  E500 host processors.
+
+	  This module provides access to the hardware capabilities through
+	  a character device node named /dev/kvm.
+
+	  If unsure, say N.
+
 config KVM_TRACE
 	bool "KVM trace support"
 	depends on KVM && MARKERS && SYSFS

arch/powerpc/kvm/Makefile

@@ -16,8 +16,18 @@ AFLAGS_booke_interrupts.o := -I$(obj)
 
 kvm-440-objs := \
 	booke.o \
+	booke_emulate.o \
 	booke_interrupts.o \
 	44x.o \
 	44x_tlb.o \
 	44x_emulate.o
 obj-$(CONFIG_KVM_440) += kvm-440.o
+
+kvm-e500-objs := \
+	booke.o \
+	booke_emulate.o \
+	booke_interrupts.o \
+	e500.o \
+	e500_tlb.o \
+	e500_emulate.o
+obj-$(CONFIG_KVM_E500) += kvm-e500.o

arch/powerpc/kvm/booke.c

@@ -30,10 +30,8 @@
 #include <asm/kvm_ppc.h>
 #include "timing.h"
 #include <asm/cacheflush.h>
-#include <asm/kvm_44x.h>
 
 #include "booke.h"
-#include "44x_tlb.h"
 
 unsigned long kvmppc_booke_handlers;
 
@@ -120,6 +118,9 @@ static int kvmppc_booke_irqprio_deliver(struct kvm_vcpu *vcpu,
 	case BOOKE_IRQPRIO_DATA_STORAGE:
 	case BOOKE_IRQPRIO_INST_STORAGE:
 	case BOOKE_IRQPRIO_FP_UNAVAIL:
+	case BOOKE_IRQPRIO_SPE_UNAVAIL:
+	case BOOKE_IRQPRIO_SPE_FP_DATA:
+	case BOOKE_IRQPRIO_SPE_FP_ROUND:
 	case BOOKE_IRQPRIO_AP_UNAVAIL:
 	case BOOKE_IRQPRIO_ALIGNMENT:
 		allowed = 1;
@@ -165,7 +166,7 @@ void kvmppc_core_deliver_interrupts(struct kvm_vcpu *vcpu)
 	unsigned int priority;
 
 	priority = __ffs(*pending);
-	while (priority <= BOOKE_MAX_INTERRUPT) {
+	while (priority <= BOOKE_IRQPRIO_MAX) {
 		if (kvmppc_booke_irqprio_deliver(vcpu, priority))
 			break;
 
@@ -263,6 +264,21 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		r = RESUME_GUEST;
 		break;
 
+	case BOOKE_INTERRUPT_SPE_UNAVAIL:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_SPE_UNAVAIL);
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_SPE_FP_DATA:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_SPE_FP_DATA);
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_SPE_FP_ROUND:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_SPE_FP_ROUND);
+		r = RESUME_GUEST;
+		break;
+
 	case BOOKE_INTERRUPT_DATA_STORAGE:
 		vcpu->arch.dear = vcpu->arch.fault_dear;
 		vcpu->arch.esr = vcpu->arch.fault_esr;
@@ -284,29 +300,27 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		r = RESUME_GUEST;
 		break;
 
-	/* XXX move to a 440-specific file. */
 	case BOOKE_INTERRUPT_DTLB_MISS: {
-		struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
-		struct kvmppc_44x_tlbe *gtlbe;
 		unsigned long eaddr = vcpu->arch.fault_dear;
 		int gtlb_index;
+		gpa_t gpaddr;
 		gfn_t gfn;
 
 		/* Check the guest TLB. */
-		gtlb_index = kvmppc_44x_dtlb_index(vcpu, eaddr);
+		gtlb_index = kvmppc_mmu_dtlb_index(vcpu, eaddr);
 		if (gtlb_index < 0) {
 			/* The guest didn't have a mapping for it. */
 			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_DTLB_MISS);
 			vcpu->arch.dear = vcpu->arch.fault_dear;
 			vcpu->arch.esr = vcpu->arch.fault_esr;
+			kvmppc_mmu_dtlb_miss(vcpu);
 			kvmppc_account_exit(vcpu, DTLB_REAL_MISS_EXITS);
 			r = RESUME_GUEST;
 			break;
 		}
 
-		gtlbe = &vcpu_44x->guest_tlb[gtlb_index];
-		vcpu->arch.paddr_accessed = tlb_xlate(gtlbe, eaddr);
-		gfn = vcpu->arch.paddr_accessed >> PAGE_SHIFT;
+		gpaddr = kvmppc_mmu_xlate(vcpu, gtlb_index, eaddr);
+		gfn = gpaddr >> PAGE_SHIFT;
 
 		if (kvm_is_visible_gfn(vcpu->kvm, gfn)) {
 			/* The guest TLB had a mapping, but the shadow TLB
@@ -315,13 +329,13 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			 * b) the guest used a large mapping which we're faking
 			 * Either way, we need to satisfy the fault without
 			 * invoking the guest. */
-			kvmppc_mmu_map(vcpu, eaddr, vcpu->arch.paddr_accessed, gtlbe->tid,
-			               gtlbe->word2, get_tlb_bytes(gtlbe), gtlb_index);
+			kvmppc_mmu_map(vcpu, eaddr, gpaddr, gtlb_index);
 			kvmppc_account_exit(vcpu, DTLB_VIRT_MISS_EXITS);
 			r = RESUME_GUEST;
 		} else {
 			/* Guest has mapped and accessed a page which is not
 			 * actually RAM. */
+			vcpu->arch.paddr_accessed = gpaddr;
 			r = kvmppc_emulate_mmio(run, vcpu);
 			kvmppc_account_exit(vcpu, MMIO_EXITS);
 		}
@@ -329,10 +343,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		break;
 	}
 
-	/* XXX move to a 440-specific file. */
 	case BOOKE_INTERRUPT_ITLB_MISS: {
-		struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
-		struct kvmppc_44x_tlbe *gtlbe;
 		unsigned long eaddr = vcpu->arch.pc;
 		gpa_t gpaddr;
 		gfn_t gfn;
@@ -341,18 +352,18 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		r = RESUME_GUEST;
 
 		/* Check the guest TLB. */
-		gtlb_index = kvmppc_44x_itlb_index(vcpu, eaddr);
+		gtlb_index = kvmppc_mmu_itlb_index(vcpu, eaddr);
 		if (gtlb_index < 0) {
 			/* The guest didn't have a mapping for it. */
 			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_ITLB_MISS);
+			kvmppc_mmu_itlb_miss(vcpu);
 			kvmppc_account_exit(vcpu, ITLB_REAL_MISS_EXITS);
 			break;
 		}
 
 		kvmppc_account_exit(vcpu, ITLB_VIRT_MISS_EXITS);
 
-		gtlbe = &vcpu_44x->guest_tlb[gtlb_index];
-		gpaddr = tlb_xlate(gtlbe, eaddr);
+		gpaddr = kvmppc_mmu_xlate(vcpu, gtlb_index, eaddr);
 		gfn = gpaddr >> PAGE_SHIFT;
 
 		if (kvm_is_visible_gfn(vcpu->kvm, gfn)) {
@@ -362,8 +373,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			 * b) the guest used a large mapping which we're faking
 			 * Either way, we need to satisfy the fault without
 			 * invoking the guest. */
-			kvmppc_mmu_map(vcpu, eaddr, gpaddr, gtlbe->tid,
-			               gtlbe->word2, get_tlb_bytes(gtlbe), gtlb_index);
+			kvmppc_mmu_map(vcpu, eaddr, gpaddr, gtlb_index);
 		} else {
 			/* Guest mapped and leaped at non-RAM! */
 			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_MACHINE_CHECK);

arch/powerpc/kvm/booke.h

@@ -22,6 +22,7 @@
 
 #include <linux/types.h>
 #include <linux/kvm_host.h>
+#include <asm/kvm_ppc.h>
 #include "timing.h"
 
 /* interrupt priortity ordering */
@@ -30,17 +31,24 @@
 #define BOOKE_IRQPRIO_ALIGNMENT 2
 #define BOOKE_IRQPRIO_PROGRAM 3
 #define BOOKE_IRQPRIO_FP_UNAVAIL 4
-#define BOOKE_IRQPRIO_SYSCALL 5
-#define BOOKE_IRQPRIO_AP_UNAVAIL 6
-#define BOOKE_IRQPRIO_DTLB_MISS 7
-#define BOOKE_IRQPRIO_ITLB_MISS 8
-#define BOOKE_IRQPRIO_MACHINE_CHECK 9
-#define BOOKE_IRQPRIO_DEBUG 10
-#define BOOKE_IRQPRIO_CRITICAL 11
-#define BOOKE_IRQPRIO_WATCHDOG 12
-#define BOOKE_IRQPRIO_EXTERNAL 13
-#define BOOKE_IRQPRIO_FIT 14
-#define BOOKE_IRQPRIO_DECREMENTER 15
+#define BOOKE_IRQPRIO_SPE_UNAVAIL 5
+#define BOOKE_IRQPRIO_SPE_FP_DATA 6
+#define BOOKE_IRQPRIO_SPE_FP_ROUND 7
+#define BOOKE_IRQPRIO_SYSCALL 8
+#define BOOKE_IRQPRIO_AP_UNAVAIL 9
+#define BOOKE_IRQPRIO_DTLB_MISS 10
+#define BOOKE_IRQPRIO_ITLB_MISS 11
+#define BOOKE_IRQPRIO_MACHINE_CHECK 12
+#define BOOKE_IRQPRIO_DEBUG 13
+#define BOOKE_IRQPRIO_CRITICAL 14
+#define BOOKE_IRQPRIO_WATCHDOG 15
+#define BOOKE_IRQPRIO_EXTERNAL 16
+#define BOOKE_IRQPRIO_FIT 17
+#define BOOKE_IRQPRIO_DECREMENTER 18
+#define BOOKE_IRQPRIO_PERFORMANCE_MONITOR 19
+#define BOOKE_IRQPRIO_MAX 19
+
+extern unsigned long kvmppc_booke_handlers;
 
 /* Helper function for "full" MSR writes. No need to call this if only EE is
  * changing. */
@@ -57,4 +65,9 @@ static inline void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr)
 	};
 }
 
+int kvmppc_booke_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
+                            unsigned int inst, int *advance);
+int kvmppc_booke_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt);
+int kvmppc_booke_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs);
+
 #endif /* __KVM_BOOKE_H__ */

arch/powerpc/kvm/booke_emulate.c

@@ -0,0 +1,266 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#include <linux/kvm_host.h>
+#include <asm/disassemble.h>
+
+#include "booke.h"
+
+#define OP_19_XOP_RFI     50
+
+#define OP_31_XOP_MFMSR   83
+#define OP_31_XOP_WRTEE   131
+#define OP_31_XOP_MTMSR   146
+#define OP_31_XOP_WRTEEI  163
+
+static void kvmppc_emul_rfi(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.pc = vcpu->arch.srr0;
+	kvmppc_set_msr(vcpu, vcpu->arch.srr1);
+}
+
+int kvmppc_booke_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
+                            unsigned int inst, int *advance)
+{
+	int emulated = EMULATE_DONE;
+	int rs;
+	int rt;
+
+	switch (get_op(inst)) {
+	case 19:
+		switch (get_xop(inst)) {
+		case OP_19_XOP_RFI:
+			kvmppc_emul_rfi(vcpu);
+			kvmppc_set_exit_type(vcpu, EMULATED_RFI_EXITS);
+			*advance = 0;
+			break;
+
+		default:
+			emulated = EMULATE_FAIL;
+			break;
+		}
+		break;
+
+	case 31:
+		switch (get_xop(inst)) {
+
+		case OP_31_XOP_MFMSR:
+			rt = get_rt(inst);
+			vcpu->arch.gpr[rt] = vcpu->arch.msr;
+			kvmppc_set_exit_type(vcpu, EMULATED_MFMSR_EXITS);
+			break;
+
+		case OP_31_XOP_MTMSR:
+			rs = get_rs(inst);
+			kvmppc_set_exit_type(vcpu, EMULATED_MTMSR_EXITS);
+			kvmppc_set_msr(vcpu, vcpu->arch.gpr[rs]);
+			break;
+
+		case OP_31_XOP_WRTEE:
+			rs = get_rs(inst);
+			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
+							 | (vcpu->arch.gpr[rs] & MSR_EE);
+			kvmppc_set_exit_type(vcpu, EMULATED_WRTEE_EXITS);
+			break;
+
+		case OP_31_XOP_WRTEEI:
+			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
+							 | (inst & MSR_EE);
+			kvmppc_set_exit_type(vcpu, EMULATED_WRTEE_EXITS);
+			break;
+
+		default:
+			emulated = EMULATE_FAIL;
+		}
+
+		break;
+
+	default:
+		emulated = EMULATE_FAIL;
+	}
+
+	return emulated;
+}
+
+int kvmppc_booke_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs)
+{
+	int emulated = EMULATE_DONE;
+
+	switch (sprn) {
+	case SPRN_DEAR:
+		vcpu->arch.dear = vcpu->arch.gpr[rs]; break;
+	case SPRN_ESR:
+		vcpu->arch.esr = vcpu->arch.gpr[rs]; break;
+	case SPRN_DBCR0:
+		vcpu->arch.dbcr0 = vcpu->arch.gpr[rs]; break;
+	case SPRN_DBCR1:
+		vcpu->arch.dbcr1 = vcpu->arch.gpr[rs]; break;
+	case SPRN_DBSR:
+		vcpu->arch.dbsr &= ~vcpu->arch.gpr[rs]; break;
+	case SPRN_TSR:
+		vcpu->arch.tsr &= ~vcpu->arch.gpr[rs]; break;
+	case SPRN_TCR:
+		vcpu->arch.tcr = vcpu->arch.gpr[rs];
+		kvmppc_emulate_dec(vcpu);
+		break;
+
+	/* Note: SPRG4-7 are user-readable. These values are
+	 * loaded into the real SPRGs when resuming the
+	 * guest. */
+	case SPRN_SPRG4:
+		vcpu->arch.sprg4 = vcpu->arch.gpr[rs]; break;
+	case SPRN_SPRG5:
+		vcpu->arch.sprg5 = vcpu->arch.gpr[rs]; break;
+	case SPRN_SPRG6:
+		vcpu->arch.sprg6 = vcpu->arch.gpr[rs]; break;
+	case SPRN_SPRG7:
+		vcpu->arch.sprg7 = vcpu->arch.gpr[rs]; break;
+
+	case SPRN_IVPR:
+		vcpu->arch.ivpr = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR0:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_CRITICAL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR1:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_MACHINE_CHECK] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR2:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DATA_STORAGE] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR3:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_INST_STORAGE] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR4:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_EXTERNAL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR5:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_ALIGNMENT] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR6:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_PROGRAM] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR7:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_FP_UNAVAIL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR8:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_SYSCALL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR9:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_AP_UNAVAIL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR10:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DECREMENTER] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR11:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_FIT] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR12:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_WATCHDOG] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR13:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR14:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR15:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG] = vcpu->arch.gpr[rs];
+		break;
+
+	default:
+		emulated = EMULATE_FAIL;
+	}
+
+	return emulated;
+}
+
+int kvmppc_booke_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt)
+{
+	int emulated = EMULATE_DONE;
+
+	switch (sprn) {
+	case SPRN_IVPR:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivpr; break;
+	case SPRN_DEAR:
+		vcpu->arch.gpr[rt] = vcpu->arch.dear; break;
+	case SPRN_ESR:
+		vcpu->arch.gpr[rt] = vcpu->arch.esr; break;
+	case SPRN_DBCR0:
+		vcpu->arch.gpr[rt] = vcpu->arch.dbcr0; break;
+	case SPRN_DBCR1:
+		vcpu->arch.gpr[rt] = vcpu->arch.dbcr1; break;
+	case SPRN_DBSR:
+		vcpu->arch.gpr[rt] = vcpu->arch.dbsr; break;
+
+	case SPRN_IVOR0:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_CRITICAL];
+		break;
+	case SPRN_IVOR1:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_MACHINE_CHECK];
+		break;
+	case SPRN_IVOR2:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DATA_STORAGE];
+		break;
+	case SPRN_IVOR3:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_INST_STORAGE];
+		break;
+	case SPRN_IVOR4:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_EXTERNAL];
+		break;
+	case SPRN_IVOR5:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_ALIGNMENT];
+		break;
+	case SPRN_IVOR6:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_PROGRAM];
+		break;
+	case SPRN_IVOR7:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_FP_UNAVAIL];
+		break;
+	case SPRN_IVOR8:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_SYSCALL];
+		break;
+	case SPRN_IVOR9:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_AP_UNAVAIL];
+		break;
+	case SPRN_IVOR10:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DECREMENTER];
+		break;
+	case SPRN_IVOR11:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_FIT];
+		break;
+	case SPRN_IVOR12:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_WATCHDOG];
+		break;
+	case SPRN_IVOR13:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS];
+		break;
+	case SPRN_IVOR14:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS];
+		break;
+	case SPRN_IVOR15:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG];
+		break;
+
+	default:
+		emulated = EMULATE_FAIL;
+	}
+
+	return emulated;
+}

arch/powerpc/kvm/booke_interrupts.S

@@ -86,6 +86,9 @@ KVM_HANDLER BOOKE_INTERRUPT_WATCHDOG
 KVM_HANDLER BOOKE_INTERRUPT_DTLB_MISS
 KVM_HANDLER BOOKE_INTERRUPT_ITLB_MISS
 KVM_HANDLER BOOKE_INTERRUPT_DEBUG
+KVM_HANDLER BOOKE_INTERRUPT_SPE_UNAVAIL
+KVM_HANDLER BOOKE_INTERRUPT_SPE_FP_DATA
+KVM_HANDLER BOOKE_INTERRUPT_SPE_FP_ROUND
 
 _GLOBAL(kvmppc_handler_len)
 	.long kvmppc_handler_1 - kvmppc_handler_0
@@ -347,7 +350,9 @@ lightweight_exit:
 	lwz	r3, VCPU_SHADOW_PID(r4)
 	mtspr	SPRN_PID, r3
 
+#ifdef CONFIG_44x
 	iccci	0, 0 /* XXX hack */
+#endif
 
 	/* Load some guest volatiles. */
 	lwz	r0, VCPU_GPR(r0)(r4)

arch/powerpc/kvm/e500.c

@@ -0,0 +1,169 @@
+/*
+ * Copyright (C) 2008 Freescale Semiconductor, Inc. All rights reserved.
+ *
+ * Author: Yu Liu, <yu.liu@freescale.com>
+ *
+ * Description:
+ * This file is derived from arch/powerpc/kvm/44x.c,
+ * by Hollis Blanchard <hollisb@us.ibm.com>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kvm_host.h>
+#include <linux/err.h>
+
+#include <asm/reg.h>
+#include <asm/cputable.h>
+#include <asm/tlbflush.h>
+#include <asm/kvm_e500.h>
+#include <asm/kvm_ppc.h>
+
+#include "booke.h"
+#include "e500_tlb.h"
+
+void kvmppc_core_load_host_debugstate(struct kvm_vcpu *vcpu)
+{
+}
+
+void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu)
+{
+}
+
+void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	kvmppc_e500_tlb_load(vcpu, cpu);
+}
+
+void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	kvmppc_e500_tlb_put(vcpu);
+}
+
+int kvmppc_core_check_processor_compat(void)
+{
+	int r;
+
+	if (strcmp(cur_cpu_spec->cpu_name, "e500v2") == 0)
+		r = 0;
+	else
+		r = -ENOTSUPP;
+
+	return r;
+}
+
+int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+
+	kvmppc_e500_tlb_setup(vcpu_e500);
+
+	/* Use the same core vertion as host's */
+	vcpu->arch.pvr = mfspr(SPRN_PVR);
+
+	return 0;
+}
+
+/* 'linear_address' is actually an encoding of AS|PID|EADDR . */
+int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
+                               struct kvm_translation *tr)
+{
+	int index;
+	gva_t eaddr;
+	u8 pid;
+	u8 as;
+
+	eaddr = tr->linear_address;
+	pid = (tr->linear_address >> 32) & 0xff;
+	as = (tr->linear_address >> 40) & 0x1;
+
+	index = kvmppc_e500_tlb_search(vcpu, eaddr, pid, as);
+	if (index < 0) {
+		tr->valid = 0;
+		return 0;
+	}
+
+	tr->physical_address = kvmppc_mmu_xlate(vcpu, index, eaddr);
+	/* XXX what does "writeable" and "usermode" even mean? */
+	tr->valid = 1;
+
+	return 0;
+}
+
+struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500;
+	struct kvm_vcpu *vcpu;
+	int err;
+
+	vcpu_e500 = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
+	if (!vcpu_e500) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	vcpu = &vcpu_e500->vcpu;
+	err = kvm_vcpu_init(vcpu, kvm, id);
+	if (err)
+		goto free_vcpu;
+
+	err = kvmppc_e500_tlb_init(vcpu_e500);
+	if (err)
+		goto uninit_vcpu;
+
+	return vcpu;
+
+uninit_vcpu:
+	kvm_vcpu_uninit(vcpu);
+free_vcpu:
+	kmem_cache_free(kvm_vcpu_cache, vcpu_e500);
+out:
+	return ERR_PTR(err);
+}
+
+void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+
+	kvmppc_e500_tlb_uninit(vcpu_e500);
+	kvm_vcpu_uninit(vcpu);
+	kmem_cache_free(kvm_vcpu_cache, vcpu_e500);
+}
+
+static int kvmppc_e500_init(void)
+{
+	int r, i;
+	unsigned long ivor[3];
+	unsigned long max_ivor = 0;
+
+	r = kvmppc_booke_init();
+	if (r)
+		return r;
+
+	/* copy extra E500 exception handlers */
+	ivor[0] = mfspr(SPRN_IVOR32);
+	ivor[1] = mfspr(SPRN_IVOR33);
+	ivor[2] = mfspr(SPRN_IVOR34);
+	for (i = 0; i < 3; i++) {
+		if (ivor[i] > max_ivor)
+			max_ivor = ivor[i];
+
+		memcpy((void *)kvmppc_booke_handlers + ivor[i],
+		       kvmppc_handlers_start + (i + 16) * kvmppc_handler_len,
+		       kvmppc_handler_len);
+	}
+	flush_icache_range(kvmppc_booke_handlers,
+			kvmppc_booke_handlers + max_ivor + kvmppc_handler_len);
+
+	return kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), THIS_MODULE);
+}
+
+static void kvmppc_e500_exit(void)
+{
+	kvmppc_booke_exit();
+}
+
+module_init(kvmppc_e500_init);
+module_exit(kvmppc_e500_exit);

arch/powerpc/kvm/e500_emulate.c

@@ -0,0 +1,202 @@
+/*
+ * Copyright (C) 2008 Freescale Semiconductor, Inc. All rights reserved.
+ *
+ * Author: Yu Liu, <yu.liu@freescale.com>
+ *
+ * Description:
+ * This file is derived from arch/powerpc/kvm/44x_emulate.c,
+ * by Hollis Blanchard <hollisb@us.ibm.com>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/kvm_ppc.h>
+#include <asm/disassemble.h>
+#include <asm/kvm_e500.h>
+
+#include "booke.h"
+#include "e500_tlb.h"
+
+#define XOP_TLBIVAX 786
+#define XOP_TLBSX   914
+#define XOP_TLBRE   946
+#define XOP_TLBWE   978
+
+int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
+                           unsigned int inst, int *advance)
+{
+	int emulated = EMULATE_DONE;
+	int ra;
+	int rb;
+
+	switch (get_op(inst)) {
+	case 31:
+		switch (get_xop(inst)) {
+
+		case XOP_TLBRE:
+			emulated = kvmppc_e500_emul_tlbre(vcpu);
+			break;
+
+		case XOP_TLBWE:
+			emulated = kvmppc_e500_emul_tlbwe(vcpu);
+			break;
+
+		case XOP_TLBSX:
+			rb = get_rb(inst);
+			emulated = kvmppc_e500_emul_tlbsx(vcpu,rb);
+			break;
+
+		case XOP_TLBIVAX:
+			ra = get_ra(inst);
+			rb = get_rb(inst);
+			emulated = kvmppc_e500_emul_tlbivax(vcpu, ra, rb);
+			break;
+
+		default:
+			emulated = EMULATE_FAIL;
+		}
+
+		break;
+
+	default:
+		emulated = EMULATE_FAIL;
+	}
+
+	if (emulated == EMULATE_FAIL)
+		emulated = kvmppc_booke_emulate_op(run, vcpu, inst, advance);
+
+	return emulated;
+}
+
+int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int emulated = EMULATE_DONE;
+
+	switch (sprn) {
+	case SPRN_PID:
+		vcpu_e500->pid[0] = vcpu->arch.shadow_pid =
+			vcpu->arch.pid = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_PID1:
+		vcpu_e500->pid[1] = vcpu->arch.gpr[rs]; break;
+	case SPRN_PID2:
+		vcpu_e500->pid[2] = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS0:
+		vcpu_e500->mas0 = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS1:
+		vcpu_e500->mas1 = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS2:
+		vcpu_e500->mas2 = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS3:
+		vcpu_e500->mas3 = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS4:
+		vcpu_e500->mas4 = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS6:
+		vcpu_e500->mas6 = vcpu->arch.gpr[rs]; break;
+	case SPRN_MAS7:
+		vcpu_e500->mas7 = vcpu->arch.gpr[rs]; break;
+	case SPRN_L1CSR1:
+		vcpu_e500->l1csr1 = vcpu->arch.gpr[rs]; break;
+	case SPRN_HID0:
+		vcpu_e500->hid0 = vcpu->arch.gpr[rs]; break;
+	case SPRN_HID1:
+		vcpu_e500->hid1 = vcpu->arch.gpr[rs]; break;
+
+	case SPRN_MMUCSR0:
+		emulated = kvmppc_e500_emul_mt_mmucsr0(vcpu_e500,
+				vcpu->arch.gpr[rs]);
+		break;
+
+	/* extra exceptions */
+	case SPRN_IVOR32:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_UNAVAIL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR33:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_FP_DATA] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR34:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_FP_ROUND] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR35:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_PERFORMANCE_MONITOR] = vcpu->arch.gpr[rs];
+		break;
+
+	default:
+		emulated = kvmppc_booke_emulate_mtspr(vcpu, sprn, rs);
+	}
+
+	return emulated;
+}
+
+int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int emulated = EMULATE_DONE;
+
+	switch (sprn) {
+	case SPRN_PID:
+		vcpu->arch.gpr[rt] = vcpu_e500->pid[0]; break;
+	case SPRN_PID1:
+		vcpu->arch.gpr[rt] = vcpu_e500->pid[1]; break;
+	case SPRN_PID2:
+		vcpu->arch.gpr[rt] = vcpu_e500->pid[2]; break;
+	case SPRN_MAS0:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas0; break;
+	case SPRN_MAS1:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas1; break;
+	case SPRN_MAS2:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas2; break;
+	case SPRN_MAS3:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas3; break;
+	case SPRN_MAS4:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas4; break;
+	case SPRN_MAS6:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas6; break;
+	case SPRN_MAS7:
+		vcpu->arch.gpr[rt] = vcpu_e500->mas7; break;
+
+	case SPRN_TLB0CFG:
+		vcpu->arch.gpr[rt] = mfspr(SPRN_TLB0CFG);
+		vcpu->arch.gpr[rt] &= ~0xfffUL;
+		vcpu->arch.gpr[rt] |= vcpu_e500->guest_tlb_size[0];
+		break;
+
+	case SPRN_TLB1CFG:
+		vcpu->arch.gpr[rt] = mfspr(SPRN_TLB1CFG);
+		vcpu->arch.gpr[rt] &= ~0xfffUL;
+		vcpu->arch.gpr[rt] |= vcpu_e500->guest_tlb_size[1];
+		break;
+
+	case SPRN_L1CSR1:
+		vcpu->arch.gpr[rt] = vcpu_e500->l1csr1; break;
+	case SPRN_HID0:
+		vcpu->arch.gpr[rt] = vcpu_e500->hid0; break;
+	case SPRN_HID1:
+		vcpu->arch.gpr[rt] = vcpu_e500->hid1; break;
+
+	case SPRN_MMUCSR0:
+		vcpu->arch.gpr[rt] = 0; break;
+
+	/* extra exceptions */
+	case SPRN_IVOR32:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_UNAVAIL];
+		break;
+	case SPRN_IVOR33:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_FP_DATA];
+		break;
+	case SPRN_IVOR34:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_FP_ROUND];
+		break;
+	case SPRN_IVOR35:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_PERFORMANCE_MONITOR];
+		break;
+	default:
+		emulated = kvmppc_booke_emulate_mfspr(vcpu, sprn, rt);
+	}
+
+	return emulated;
+}
+

arch/powerpc/kvm/e500_tlb.c

@@ -0,0 +1,757 @@
+/*
+ * Copyright (C) 2008 Freescale Semiconductor, Inc. All rights reserved.
+ *
+ * Author: Yu Liu, yu.liu@freescale.com
+ *
+ * Description:
+ * This file is based on arch/powerpc/kvm/44x_tlb.c,
+ * by Hollis Blanchard <hollisb@us.ibm.com>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <linux/highmem.h>
+#include <asm/kvm_ppc.h>
+#include <asm/kvm_e500.h>
+
+#include "../mm/mmu_decl.h"
+#include "e500_tlb.h"
+
+#define to_htlb1_esel(esel) (tlb1_entry_num - (esel) - 1)
+
+static unsigned int tlb1_entry_num;
+
+void kvmppc_dump_tlbs(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	struct tlbe *tlbe;
+	int i, tlbsel;
+
+	printk("| %8s | %8s | %8s | %8s | %8s |\n",
+			"nr", "mas1", "mas2", "mas3", "mas7");
+
+	for (tlbsel = 0; tlbsel < 2; tlbsel++) {
+		printk("Guest TLB%d:\n", tlbsel);
+		for (i = 0; i < vcpu_e500->guest_tlb_size[tlbsel]; i++) {
+			tlbe = &vcpu_e500->guest_tlb[tlbsel][i];
+			if (tlbe->mas1 & MAS1_VALID)
+				printk(" G[%d][%3d] |  %08X | %08X | %08X | %08X |\n",
+					tlbsel, i, tlbe->mas1, tlbe->mas2,
+					tlbe->mas3, tlbe->mas7);
+		}
+	}
+
+	for (tlbsel = 0; tlbsel < 2; tlbsel++) {
+		printk("Shadow TLB%d:\n", tlbsel);
+		for (i = 0; i < vcpu_e500->shadow_tlb_size[tlbsel]; i++) {
+			tlbe = &vcpu_e500->shadow_tlb[tlbsel][i];
+			if (tlbe->mas1 & MAS1_VALID)
+				printk(" S[%d][%3d] |  %08X | %08X | %08X | %08X |\n",
+					tlbsel, i, tlbe->mas1, tlbe->mas2,
+					tlbe->mas3, tlbe->mas7);
+		}
+	}
+}
+
+static inline unsigned int tlb0_get_next_victim(
+		struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	unsigned int victim;
+
+	victim = vcpu_e500->guest_tlb_nv[0]++;
+	if (unlikely(vcpu_e500->guest_tlb_nv[0] >= KVM_E500_TLB0_WAY_NUM))
+		vcpu_e500->guest_tlb_nv[0] = 0;
+
+	return victim;
+}
+
+static inline unsigned int tlb1_max_shadow_size(void)
+{
+	return tlb1_entry_num - tlbcam_index;
+}
+
+static inline int tlbe_is_writable(struct tlbe *tlbe)
+{
+	return tlbe->mas3 & (MAS3_SW|MAS3_UW);
+}
+
+static inline u32 e500_shadow_mas3_attrib(u32 mas3, int usermode)
+{
+	/* Mask off reserved bits. */
+	mas3 &= MAS3_ATTRIB_MASK;
+
+	if (!usermode) {
+		/* Guest is in supervisor mode,
+		 * so we need to translate guest
+		 * supervisor permissions into user permissions. */
+		mas3 &= ~E500_TLB_USER_PERM_MASK;
+		mas3 |= (mas3 & E500_TLB_SUPER_PERM_MASK) << 1;
+	}
+
+	return mas3 | E500_TLB_SUPER_PERM_MASK;
+}
+
+static inline u32 e500_shadow_mas2_attrib(u32 mas2, int usermode)
+{
+#ifdef CONFIG_SMP
+	return (mas2 & MAS2_ATTRIB_MASK) | MAS2_M;
+#else
+	return mas2 & MAS2_ATTRIB_MASK;
+#endif
+}
+
+/*
+ * writing shadow tlb entry to host TLB
+ */
+static inline void __write_host_tlbe(struct tlbe *stlbe)
+{
+	mtspr(SPRN_MAS1, stlbe->mas1);
+	mtspr(SPRN_MAS2, stlbe->mas2);
+	mtspr(SPRN_MAS3, stlbe->mas3);
+	mtspr(SPRN_MAS7, stlbe->mas7);
+	__asm__ __volatile__ ("tlbwe\n" : : );
+}
+
+static inline void write_host_tlbe(struct kvmppc_vcpu_e500 *vcpu_e500,
+		int tlbsel, int esel)
+{
+	struct tlbe *stlbe = &vcpu_e500->shadow_tlb[tlbsel][esel];
+
+	local_irq_disable();
+	if (tlbsel == 0) {
+		__write_host_tlbe(stlbe);
+	} else {
+		unsigned register mas0;
+
+		mas0 = mfspr(SPRN_MAS0);
+
+		mtspr(SPRN_MAS0, MAS0_TLBSEL(1) | MAS0_ESEL(to_htlb1_esel(esel)));
+		__write_host_tlbe(stlbe);
+
+		mtspr(SPRN_MAS0, mas0);
+	}
+	local_irq_enable();
+}
+
+void kvmppc_e500_tlb_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int i;
+	unsigned register mas0;
+
+	/* Load all valid TLB1 entries to reduce guest tlb miss fault */
+	local_irq_disable();
+	mas0 = mfspr(SPRN_MAS0);
+	for (i = 0; i < tlb1_max_shadow_size(); i++) {
+		struct tlbe *stlbe = &vcpu_e500->shadow_tlb[1][i];
+
+		if (get_tlb_v(stlbe)) {
+			mtspr(SPRN_MAS0, MAS0_TLBSEL(1)
+					| MAS0_ESEL(to_htlb1_esel(i)));
+			__write_host_tlbe(stlbe);
+		}
+	}
+	mtspr(SPRN_MAS0, mas0);
+	local_irq_enable();
+}
+
+void kvmppc_e500_tlb_put(struct kvm_vcpu *vcpu)
+{
+	_tlbil_all();
+}
+
+/* Search the guest TLB for a matching entry. */
+static int kvmppc_e500_tlb_index(struct kvmppc_vcpu_e500 *vcpu_e500,
+		gva_t eaddr, int tlbsel, unsigned int pid, int as)
+{
+	int i;
+
+	/* XXX Replace loop with fancy data structures. */
+	for (i = 0; i < vcpu_e500->guest_tlb_size[tlbsel]; i++) {
+		struct tlbe *tlbe = &vcpu_e500->guest_tlb[tlbsel][i];
+		unsigned int tid;
+
+		if (eaddr < get_tlb_eaddr(tlbe))
+			continue;
+
+		if (eaddr > get_tlb_end(tlbe))
+			continue;
+
+		tid = get_tlb_tid(tlbe);
+		if (tid && (tid != pid))
+			continue;
+
+		if (!get_tlb_v(tlbe))
+			continue;
+
+		if (get_tlb_ts(tlbe) != as && as != -1)
+			continue;
+
+		return i;
+	}
+
+	return -1;
+}
+
+static void kvmppc_e500_shadow_release(struct kvmppc_vcpu_e500 *vcpu_e500,
+		int tlbsel, int esel)
+{
+	struct tlbe *stlbe = &vcpu_e500->shadow_tlb[tlbsel][esel];
+	struct page *page = vcpu_e500->shadow_pages[tlbsel][esel];
+
+	if (page) {
+		vcpu_e500->shadow_pages[tlbsel][esel] = NULL;
+
+		if (get_tlb_v(stlbe)) {
+			if (tlbe_is_writable(stlbe))
+				kvm_release_page_dirty(page);
+			else
+				kvm_release_page_clean(page);
+		}
+	}
+}
+
+static void kvmppc_e500_stlbe_invalidate(struct kvmppc_vcpu_e500 *vcpu_e500,
+		int tlbsel, int esel)
+{
+	struct tlbe *stlbe = &vcpu_e500->shadow_tlb[tlbsel][esel];
+
+	kvmppc_e500_shadow_release(vcpu_e500, tlbsel, esel);
+	stlbe->mas1 = 0;
+	KVMTRACE_5D(STLB_INVAL, &vcpu_e500->vcpu, index_of(tlbsel, esel),
+			stlbe->mas1, stlbe->mas2, stlbe->mas3, stlbe->mas7,
+			handler);
+}
+
+static void kvmppc_e500_tlb1_invalidate(struct kvmppc_vcpu_e500 *vcpu_e500,
+		gva_t eaddr, gva_t eend, u32 tid)
+{
+	unsigned int pid = tid & 0xff;
+	unsigned int i;
+
+	/* XXX Replace loop with fancy data structures. */
+	for (i = 0; i < vcpu_e500->guest_tlb_size[1]; i++) {
+		struct tlbe *stlbe = &vcpu_e500->shadow_tlb[1][i];
+		unsigned int tid;
+
+		if (!get_tlb_v(stlbe))
+			continue;
+
+		if (eend < get_tlb_eaddr(stlbe))
+			continue;
+
+		if (eaddr > get_tlb_end(stlbe))
+			continue;
+
+		tid = get_tlb_tid(stlbe);
+		if (tid && (tid != pid))
+			continue;
+
+		kvmppc_e500_stlbe_invalidate(vcpu_e500, 1, i);
+		write_host_tlbe(vcpu_e500, 1, i);
+	}
+}
+
+static inline void kvmppc_e500_deliver_tlb_miss(struct kvm_vcpu *vcpu,
+		unsigned int eaddr, int as)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	unsigned int victim, pidsel, tsized;
+	int tlbsel;
+
+	/* since we only have two TLBs, only lower bit is used. */
+	tlbsel = (vcpu_e500->mas4 >> 28) & 0x1;
+	victim = (tlbsel == 0) ? tlb0_get_next_victim(vcpu_e500) : 0;
+	pidsel = (vcpu_e500->mas4 >> 16) & 0xf;
+	tsized = (vcpu_e500->mas4 >> 8) & 0xf;
+
+	vcpu_e500->mas0 = MAS0_TLBSEL(tlbsel) | MAS0_ESEL(victim)
+		| MAS0_NV(vcpu_e500->guest_tlb_nv[tlbsel]);
+	vcpu_e500->mas1 = MAS1_VALID | (as ? MAS1_TS : 0)
+		| MAS1_TID(vcpu_e500->pid[pidsel])
+		| MAS1_TSIZE(tsized);
+	vcpu_e500->mas2 = (eaddr & MAS2_EPN)
+		| (vcpu_e500->mas4 & MAS2_ATTRIB_MASK);
+	vcpu_e500->mas3 &= MAS3_U0 | MAS3_U1 | MAS3_U2 | MAS3_U3;
+	vcpu_e500->mas6 = (vcpu_e500->mas6 & MAS6_SPID1)
+		| (get_cur_pid(vcpu) << 16)
+		| (as ? MAS6_SAS : 0);
+	vcpu_e500->mas7 = 0;
+}
+
+static inline void kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500,
+	u64 gvaddr, gfn_t gfn, struct tlbe *gtlbe, int tlbsel, int esel)
+{
+	struct page *new_page;
+	struct tlbe *stlbe;
+	hpa_t hpaddr;
+
+	stlbe = &vcpu_e500->shadow_tlb[tlbsel][esel];
+
+	/* Get reference to new page. */
+	new_page = gfn_to_page(vcpu_e500->vcpu.kvm, gfn);
+	if (is_error_page(new_page)) {
+		printk(KERN_ERR "Couldn't get guest page for gfn %lx!\n", gfn);
+		kvm_release_page_clean(new_page);
+		return;
+	}
+	hpaddr = page_to_phys(new_page);
+
+	/* Drop reference to old page. */
+	kvmppc_e500_shadow_release(vcpu_e500, tlbsel, esel);
+
+	vcpu_e500->shadow_pages[tlbsel][esel] = new_page;
+
+	/* Force TS=1 IPROT=0 TSIZE=4KB for all guest mappings. */
+	stlbe->mas1 = MAS1_TSIZE(BOOKE_PAGESZ_4K)
+		| MAS1_TID(get_tlb_tid(gtlbe)) | MAS1_TS | MAS1_VALID;
+	stlbe->mas2 = (gvaddr & MAS2_EPN)
+		| e500_shadow_mas2_attrib(gtlbe->mas2,
+				vcpu_e500->vcpu.arch.msr & MSR_PR);
+	stlbe->mas3 = (hpaddr & MAS3_RPN)
+		| e500_shadow_mas3_attrib(gtlbe->mas3,
+				vcpu_e500->vcpu.arch.msr & MSR_PR);
+	stlbe->mas7 = (hpaddr >> 32) & MAS7_RPN;
+
+	KVMTRACE_5D(STLB_WRITE, &vcpu_e500->vcpu, index_of(tlbsel, esel),
+			stlbe->mas1, stlbe->mas2, stlbe->mas3, stlbe->mas7,
+			handler);
+}
+
+/* XXX only map the one-one case, for now use TLB0 */
+static int kvmppc_e500_stlbe_map(struct kvmppc_vcpu_e500 *vcpu_e500,
+		int tlbsel, int esel)
+{
+	struct tlbe *gtlbe;
+
+	gtlbe = &vcpu_e500->guest_tlb[tlbsel][esel];
+
+	kvmppc_e500_shadow_map(vcpu_e500, get_tlb_eaddr(gtlbe),
+			get_tlb_raddr(gtlbe) >> PAGE_SHIFT,
+			gtlbe, tlbsel, esel);
+
+	return esel;
+}
+
+/* Caller must ensure that the specified guest TLB entry is safe to insert into
+ * the shadow TLB. */
+/* XXX for both one-one and one-to-many , for now use TLB1 */
+static int kvmppc_e500_tlb1_map(struct kvmppc_vcpu_e500 *vcpu_e500,
+		u64 gvaddr, gfn_t gfn, struct tlbe *gtlbe)
+{
+	unsigned int victim;
+
+	victim = vcpu_e500->guest_tlb_nv[1]++;
+
+	if (unlikely(vcpu_e500->guest_tlb_nv[1] >= tlb1_max_shadow_size()))
+		vcpu_e500->guest_tlb_nv[1] = 0;
+
+	kvmppc_e500_shadow_map(vcpu_e500, gvaddr, gfn, gtlbe, 1, victim);
+
+	return victim;
+}
+
+/* Invalidate all guest kernel mappings when enter usermode,
+ * so that when they fault back in they will get the
+ * proper permission bits. */
+void kvmppc_mmu_priv_switch(struct kvm_vcpu *vcpu, int usermode)
+{
+	if (usermode) {
+		struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+		int i;
+
+		/* XXX Replace loop with fancy data structures. */
+		for (i = 0; i < tlb1_max_shadow_size(); i++)
+			kvmppc_e500_stlbe_invalidate(vcpu_e500, 1, i);
+
+		_tlbil_all();
+	}
+}
+
+static int kvmppc_e500_gtlbe_invalidate(struct kvmppc_vcpu_e500 *vcpu_e500,
+		int tlbsel, int esel)
+{
+	struct tlbe *gtlbe = &vcpu_e500->guest_tlb[tlbsel][esel];
+
+	if (unlikely(get_tlb_iprot(gtlbe)))
+		return -1;
+
+	if (tlbsel == 1) {
+		kvmppc_e500_tlb1_invalidate(vcpu_e500, get_tlb_eaddr(gtlbe),
+				get_tlb_end(gtlbe),
+				get_tlb_tid(gtlbe));
+	} else {
+		kvmppc_e500_stlbe_invalidate(vcpu_e500, tlbsel, esel);
+	}
+
+	gtlbe->mas1 = 0;
+
+	return 0;
+}
+
+int kvmppc_e500_emul_mt_mmucsr0(struct kvmppc_vcpu_e500 *vcpu_e500, ulong value)
+{
+	int esel;
+
+	if (value & MMUCSR0_TLB0FI)
+		for (esel = 0; esel < vcpu_e500->guest_tlb_size[0]; esel++)
+			kvmppc_e500_gtlbe_invalidate(vcpu_e500, 0, esel);
+	if (value & MMUCSR0_TLB1FI)
+		for (esel = 0; esel < vcpu_e500->guest_tlb_size[1]; esel++)
+			kvmppc_e500_gtlbe_invalidate(vcpu_e500, 1, esel);
+
+	_tlbil_all();
+
+	return EMULATE_DONE;
+}
+
+int kvmppc_e500_emul_tlbivax(struct kvm_vcpu *vcpu, int ra, int rb)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	unsigned int ia;
+	int esel, tlbsel;
+	gva_t ea;
+
+	ea = ((ra) ? vcpu->arch.gpr[ra] : 0) + vcpu->arch.gpr[rb];
+
+	ia = (ea >> 2) & 0x1;
+
+	/* since we only have two TLBs, only lower bit is used. */
+	tlbsel = (ea >> 3) & 0x1;
+
+	if (ia) {
+		/* invalidate all entries */
+		for (esel = 0; esel < vcpu_e500->guest_tlb_size[tlbsel]; esel++)
+			kvmppc_e500_gtlbe_invalidate(vcpu_e500, tlbsel, esel);
+	} else {
+		ea &= 0xfffff000;
+		esel = kvmppc_e500_tlb_index(vcpu_e500, ea, tlbsel,
+				get_cur_pid(vcpu), -1);
+		if (esel >= 0)
+			kvmppc_e500_gtlbe_invalidate(vcpu_e500, tlbsel, esel);
+	}
+
+	_tlbil_all();
+
+	return EMULATE_DONE;
+}
+
+int kvmppc_e500_emul_tlbre(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int tlbsel, esel;
+	struct tlbe *gtlbe;
+
+	tlbsel = get_tlb_tlbsel(vcpu_e500);
+	esel = get_tlb_esel(vcpu_e500, tlbsel);
+
+	gtlbe = &vcpu_e500->guest_tlb[tlbsel][esel];
+	vcpu_e500->mas0 &= ~MAS0_NV(~0);
+	vcpu_e500->mas0 |= MAS0_NV(vcpu_e500->guest_tlb_nv[tlbsel]);
+	vcpu_e500->mas1 = gtlbe->mas1;
+	vcpu_e500->mas2 = gtlbe->mas2;
+	vcpu_e500->mas3 = gtlbe->mas3;
+	vcpu_e500->mas7 = gtlbe->mas7;
+
+	return EMULATE_DONE;
+}
+
+int kvmppc_e500_emul_tlbsx(struct kvm_vcpu *vcpu, int rb)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int as = !!get_cur_sas(vcpu_e500);
+	unsigned int pid = get_cur_spid(vcpu_e500);
+	int esel, tlbsel;
+	struct tlbe *gtlbe = NULL;
+	gva_t ea;
+
+	ea = vcpu->arch.gpr[rb];
+
+	for (tlbsel = 0; tlbsel < 2; tlbsel++) {
+		esel = kvmppc_e500_tlb_index(vcpu_e500, ea, tlbsel, pid, as);
+		if (esel >= 0) {
+			gtlbe = &vcpu_e500->guest_tlb[tlbsel][esel];
+			break;
+		}
+	}
+
+	if (gtlbe) {
+		vcpu_e500->mas0 = MAS0_TLBSEL(tlbsel) | MAS0_ESEL(esel)
+			| MAS0_NV(vcpu_e500->guest_tlb_nv[tlbsel]);
+		vcpu_e500->mas1 = gtlbe->mas1;
+		vcpu_e500->mas2 = gtlbe->mas2;
+		vcpu_e500->mas3 = gtlbe->mas3;
+		vcpu_e500->mas7 = gtlbe->mas7;
+	} else {
+		int victim;
+
+		/* since we only have two TLBs, only lower bit is used. */
+		tlbsel = vcpu_e500->mas4 >> 28 & 0x1;
+		victim = (tlbsel == 0) ? tlb0_get_next_victim(vcpu_e500) : 0;
+
+		vcpu_e500->mas0 = MAS0_TLBSEL(tlbsel) | MAS0_ESEL(victim)
+			| MAS0_NV(vcpu_e500->guest_tlb_nv[tlbsel]);
+		vcpu_e500->mas1 = (vcpu_e500->mas6 & MAS6_SPID0)
+			| (vcpu_e500->mas6 & (MAS6_SAS ? MAS1_TS : 0))
+			| (vcpu_e500->mas4 & MAS4_TSIZED(~0));
+		vcpu_e500->mas2 &= MAS2_EPN;
+		vcpu_e500->mas2 |= vcpu_e500->mas4 & MAS2_ATTRIB_MASK;
+		vcpu_e500->mas3 &= MAS3_U0 | MAS3_U1 | MAS3_U2 | MAS3_U3;
+		vcpu_e500->mas7 = 0;
+	}
+
+	return EMULATE_DONE;
+}
+
+int kvmppc_e500_emul_tlbwe(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	u64 eaddr;
+	u64 raddr;
+	u32 tid;
+	struct tlbe *gtlbe;
+	int tlbsel, esel, stlbsel, sesel;
+
+	tlbsel = get_tlb_tlbsel(vcpu_e500);
+	esel = get_tlb_esel(vcpu_e500, tlbsel);
+
+	gtlbe = &vcpu_e500->guest_tlb[tlbsel][esel];
+
+	if (get_tlb_v(gtlbe) && tlbsel == 1) {
+		eaddr = get_tlb_eaddr(gtlbe);
+		tid = get_tlb_tid(gtlbe);
+		kvmppc_e500_tlb1_invalidate(vcpu_e500, eaddr,
+				get_tlb_end(gtlbe), tid);
+	}
+
+	gtlbe->mas1 = vcpu_e500->mas1;
+	gtlbe->mas2 = vcpu_e500->mas2;
+	gtlbe->mas3 = vcpu_e500->mas3;
+	gtlbe->mas7 = vcpu_e500->mas7;
+
+	KVMTRACE_5D(GTLB_WRITE, vcpu, vcpu_e500->mas0,
+			gtlbe->mas1, gtlbe->mas2, gtlbe->mas3, gtlbe->mas7,
+			handler);
+
+	/* Invalidate shadow mappings for the about-to-be-clobbered TLBE. */
+	if (tlbe_is_host_safe(vcpu, gtlbe)) {
+		switch (tlbsel) {
+		case 0:
+			/* TLB0 */
+			gtlbe->mas1 &= ~MAS1_TSIZE(~0);
+			gtlbe->mas1 |= MAS1_TSIZE(BOOKE_PAGESZ_4K);
+
+			stlbsel = 0;
+			sesel = kvmppc_e500_stlbe_map(vcpu_e500, 0, esel);
+
+			break;
+
+		case 1:
+			/* TLB1 */
+			eaddr = get_tlb_eaddr(gtlbe);
+			raddr = get_tlb_raddr(gtlbe);
+
+			/* Create a 4KB mapping on the host.
+			 * If the guest wanted a large page,
+			 * only the first 4KB is mapped here and the rest
+			 * are mapped on the fly. */
+			stlbsel = 1;
+			sesel = kvmppc_e500_tlb1_map(vcpu_e500, eaddr,
+					raddr >> PAGE_SHIFT, gtlbe);
+			break;
+
+		default:
+			BUG();
+		}
+		write_host_tlbe(vcpu_e500, stlbsel, sesel);
+	}
+
+	return EMULATE_DONE;
+}
+
+int kvmppc_mmu_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
+{
+	unsigned int as = !!(vcpu->arch.msr & MSR_IS);
+
+	return kvmppc_e500_tlb_search(vcpu, eaddr, get_cur_pid(vcpu), as);
+}
+
+int kvmppc_mmu_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
+{
+	unsigned int as = !!(vcpu->arch.msr & MSR_DS);
+
+	return kvmppc_e500_tlb_search(vcpu, eaddr, get_cur_pid(vcpu), as);
+}
+
+void kvmppc_mmu_itlb_miss(struct kvm_vcpu *vcpu)
+{
+	unsigned int as = !!(vcpu->arch.msr & MSR_IS);
+
+	kvmppc_e500_deliver_tlb_miss(vcpu, vcpu->arch.pc, as);
+}
+
+void kvmppc_mmu_dtlb_miss(struct kvm_vcpu *vcpu)
+{
+	unsigned int as = !!(vcpu->arch.msr & MSR_DS);
+
+	kvmppc_e500_deliver_tlb_miss(vcpu, vcpu->arch.fault_dear, as);
+}
+
+gpa_t kvmppc_mmu_xlate(struct kvm_vcpu *vcpu, unsigned int index,
+			gva_t eaddr)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	struct tlbe *gtlbe =
+		&vcpu_e500->guest_tlb[tlbsel_of(index)][esel_of(index)];
+	u64 pgmask = get_tlb_bytes(gtlbe) - 1;
+
+	return get_tlb_raddr(gtlbe) | (eaddr & pgmask);
+}
+
+void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int tlbsel, i;
+
+	for (tlbsel = 0; tlbsel < 2; tlbsel++)
+		for (i = 0; i < vcpu_e500->guest_tlb_size[tlbsel]; i++)
+			kvmppc_e500_shadow_release(vcpu_e500, tlbsel, i);
+
+	/* discard all guest mapping */
+	_tlbil_all();
+}
+
+void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 eaddr, gpa_t gpaddr,
+			unsigned int index)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int tlbsel = tlbsel_of(index);
+	int esel = esel_of(index);
+	int stlbsel, sesel;
+
+	switch (tlbsel) {
+	case 0:
+		stlbsel = 0;
+		sesel = esel;
+		break;
+
+	case 1: {
+		gfn_t gfn = gpaddr >> PAGE_SHIFT;
+		struct tlbe *gtlbe
+			= &vcpu_e500->guest_tlb[tlbsel][esel];
+
+		stlbsel = 1;
+		sesel = kvmppc_e500_tlb1_map(vcpu_e500, eaddr, gfn, gtlbe);
+		break;
+	}
+
+	default:
+		BUG();
+		break;
+	}
+	write_host_tlbe(vcpu_e500, stlbsel, sesel);
+}
+
+int kvmppc_e500_tlb_search(struct kvm_vcpu *vcpu,
+				gva_t eaddr, unsigned int pid, int as)
+{
+	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
+	int esel, tlbsel;
+
+	for (tlbsel = 0; tlbsel < 2; tlbsel++) {
+		esel = kvmppc_e500_tlb_index(vcpu_e500, eaddr, tlbsel, pid, as);
+		if (esel >= 0)
+			return index_of(tlbsel, esel);
+	}
+
+	return -1;
+}
+
+void kvmppc_e500_tlb_setup(struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	struct tlbe *tlbe;
+
+	/* Insert large initial mapping for guest. */
+	tlbe = &vcpu_e500->guest_tlb[1][0];
+	tlbe->mas1 = MAS1_VALID | MAS1_TSIZE(BOOKE_PAGESZ_256M);
+	tlbe->mas2 = 0;
+	tlbe->mas3 = E500_TLB_SUPER_PERM_MASK;
+	tlbe->mas7 = 0;
+
+	/* 4K map for serial output. Used by kernel wrapper. */
+	tlbe = &vcpu_e500->guest_tlb[1][1];
+	tlbe->mas1 = MAS1_VALID | MAS1_TSIZE(BOOKE_PAGESZ_4K);
+	tlbe->mas2 = (0xe0004500 & 0xFFFFF000) | MAS2_I | MAS2_G;
+	tlbe->mas3 = (0xe0004500 & 0xFFFFF000) | E500_TLB_SUPER_PERM_MASK;
+	tlbe->mas7 = 0;
+}
+
+int kvmppc_e500_tlb_init(struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	tlb1_entry_num = mfspr(SPRN_TLB1CFG) & 0xFFF;
+
+	vcpu_e500->guest_tlb_size[0] = KVM_E500_TLB0_SIZE;
+	vcpu_e500->guest_tlb[0] =
+		kzalloc(sizeof(struct tlbe) * KVM_E500_TLB0_SIZE, GFP_KERNEL);
+	if (vcpu_e500->guest_tlb[0] == NULL)
+		goto err_out;
+
+	vcpu_e500->shadow_tlb_size[0] = KVM_E500_TLB0_SIZE;
+	vcpu_e500->shadow_tlb[0] =
+		kzalloc(sizeof(struct tlbe) * KVM_E500_TLB0_SIZE, GFP_KERNEL);
+	if (vcpu_e500->shadow_tlb[0] == NULL)
+		goto err_out_guest0;
+
+	vcpu_e500->guest_tlb_size[1] = KVM_E500_TLB1_SIZE;
+	vcpu_e500->guest_tlb[1] =
+		kzalloc(sizeof(struct tlbe) * KVM_E500_TLB1_SIZE, GFP_KERNEL);
+	if (vcpu_e500->guest_tlb[1] == NULL)
+		goto err_out_shadow0;
+
+	vcpu_e500->shadow_tlb_size[1] = tlb1_entry_num;
+	vcpu_e500->shadow_tlb[1] =
+		kzalloc(sizeof(struct tlbe) * tlb1_entry_num, GFP_KERNEL);
+	if (vcpu_e500->shadow_tlb[1] == NULL)
+		goto err_out_guest1;
+
+	vcpu_e500->shadow_pages[0] = (struct page **)
+		kzalloc(sizeof(struct page *) * KVM_E500_TLB0_SIZE, GFP_KERNEL);
+	if (vcpu_e500->shadow_pages[0] == NULL)
+		goto err_out_shadow1;
+
+	vcpu_e500->shadow_pages[1] = (struct page **)
+		kzalloc(sizeof(struct page *) * tlb1_entry_num, GFP_KERNEL);
+	if (vcpu_e500->shadow_pages[1] == NULL)
+		goto err_out_page0;
+
+	return 0;
+
+err_out_page0:
+	kfree(vcpu_e500->shadow_pages[0]);
+err_out_shadow1:
+	kfree(vcpu_e500->shadow_tlb[1]);
+err_out_guest1:
+	kfree(vcpu_e500->guest_tlb[1]);
+err_out_shadow0:
+	kfree(vcpu_e500->shadow_tlb[0]);
+err_out_guest0:
+	kfree(vcpu_e500->guest_tlb[0]);
+err_out:
+	return -1;
+}
+
+void kvmppc_e500_tlb_uninit(struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	kfree(vcpu_e500->shadow_pages[1]);
+	kfree(vcpu_e500->shadow_pages[0]);
+	kfree(vcpu_e500->shadow_tlb[1]);
+	kfree(vcpu_e500->guest_tlb[1]);
+	kfree(vcpu_e500->shadow_tlb[0]);
+	kfree(vcpu_e500->guest_tlb[0]);
+}

arch/powerpc/kvm/e500_tlb.h

@@ -0,0 +1,185 @@
+/*
+ * Copyright (C) 2008 Freescale Semiconductor, Inc. All rights reserved.
+ *
+ * Author: Yu Liu, yu.liu@freescale.com
+ *
+ * Description:
+ * This file is based on arch/powerpc/kvm/44x_tlb.h,
+ * by Hollis Blanchard <hollisb@us.ibm.com>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __KVM_E500_TLB_H__
+#define __KVM_E500_TLB_H__
+
+#include <linux/kvm_host.h>
+#include <asm/mmu-fsl-booke.h>
+#include <asm/tlb.h>
+#include <asm/kvm_e500.h>
+
+#define KVM_E500_TLB0_WAY_SIZE_BIT	7	/* Fixed */
+#define KVM_E500_TLB0_WAY_SIZE		(1UL << KVM_E500_TLB0_WAY_SIZE_BIT)
+#define KVM_E500_TLB0_WAY_SIZE_MASK	(KVM_E500_TLB0_WAY_SIZE - 1)
+
+#define KVM_E500_TLB0_WAY_NUM_BIT	1	/* No greater than 7 */
+#define KVM_E500_TLB0_WAY_NUM		(1UL << KVM_E500_TLB0_WAY_NUM_BIT)
+#define KVM_E500_TLB0_WAY_NUM_MASK	(KVM_E500_TLB0_WAY_NUM - 1)
+
+#define KVM_E500_TLB0_SIZE  (KVM_E500_TLB0_WAY_SIZE * KVM_E500_TLB0_WAY_NUM)
+#define KVM_E500_TLB1_SIZE  16
+
+#define index_of(tlbsel, esel)	(((tlbsel) << 16) | ((esel) & 0xFFFF))
+#define tlbsel_of(index)	((index) >> 16)
+#define esel_of(index)		((index) & 0xFFFF)
+
+#define E500_TLB_USER_PERM_MASK (MAS3_UX|MAS3_UR|MAS3_UW)
+#define E500_TLB_SUPER_PERM_MASK (MAS3_SX|MAS3_SR|MAS3_SW)
+#define MAS2_ATTRIB_MASK \
+	  (MAS2_X0 | MAS2_X1)
+#define MAS3_ATTRIB_MASK \
+	  (MAS3_U0 | MAS3_U1 | MAS3_U2 | MAS3_U3 \
+	   | E500_TLB_USER_PERM_MASK | E500_TLB_SUPER_PERM_MASK)
+
+extern void kvmppc_dump_tlbs(struct kvm_vcpu *);
+extern int kvmppc_e500_emul_mt_mmucsr0(struct kvmppc_vcpu_e500 *, ulong);
+extern int kvmppc_e500_emul_tlbwe(struct kvm_vcpu *);
+extern int kvmppc_e500_emul_tlbre(struct kvm_vcpu *);
+extern int kvmppc_e500_emul_tlbivax(struct kvm_vcpu *, int, int);
+extern int kvmppc_e500_emul_tlbsx(struct kvm_vcpu *, int);
+extern int kvmppc_e500_tlb_search(struct kvm_vcpu *, gva_t, unsigned int, int);
+extern void kvmppc_e500_tlb_put(struct kvm_vcpu *);
+extern void kvmppc_e500_tlb_load(struct kvm_vcpu *, int);
+extern int kvmppc_e500_tlb_init(struct kvmppc_vcpu_e500 *);
+extern void kvmppc_e500_tlb_uninit(struct kvmppc_vcpu_e500 *);
+extern void kvmppc_e500_tlb_setup(struct kvmppc_vcpu_e500 *);
+
+/* TLB helper functions */
+static inline unsigned int get_tlb_size(const struct tlbe *tlbe)
+{
+	return (tlbe->mas1 >> 8) & 0xf;
+}
+
+static inline gva_t get_tlb_eaddr(const struct tlbe *tlbe)
+{
+	return tlbe->mas2 & 0xfffff000;
+}
+
+static inline u64 get_tlb_bytes(const struct tlbe *tlbe)
+{
+	unsigned int pgsize = get_tlb_size(tlbe);
+	return 1ULL << 10 << (pgsize << 1);
+}
+
+static inline gva_t get_tlb_end(const struct tlbe *tlbe)
+{
+	u64 bytes = get_tlb_bytes(tlbe);
+	return get_tlb_eaddr(tlbe) + bytes - 1;
+}
+
+static inline u64 get_tlb_raddr(const struct tlbe *tlbe)
+{
+	u64 rpn = tlbe->mas7;
+	return (rpn << 32) | (tlbe->mas3 & 0xfffff000);
+}
+
+static inline unsigned int get_tlb_tid(const struct tlbe *tlbe)
+{
+	return (tlbe->mas1 >> 16) & 0xff;
+}
+
+static inline unsigned int get_tlb_ts(const struct tlbe *tlbe)
+{
+	return (tlbe->mas1 >> 12) & 0x1;
+}
+
+static inline unsigned int get_tlb_v(const struct tlbe *tlbe)
+{
+	return (tlbe->mas1 >> 31) & 0x1;
+}
+
+static inline unsigned int get_tlb_iprot(const struct tlbe *tlbe)
+{
+	return (tlbe->mas1 >> 30) & 0x1;
+}
+
+static inline unsigned int get_cur_pid(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.pid & 0xff;
+}
+
+static inline unsigned int get_cur_spid(
+		const struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	return (vcpu_e500->mas6 >> 16) & 0xff;
+}
+
+static inline unsigned int get_cur_sas(
+		const struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	return vcpu_e500->mas6 & 0x1;
+}
+
+static inline unsigned int get_tlb_tlbsel(
+		const struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	/*
+	 * Manual says that tlbsel has 2 bits wide.
+	 * Since we only have two TLBs, only lower bit is used.
+	 */
+	return (vcpu_e500->mas0 >> 28) & 0x1;
+}
+
+static inline unsigned int get_tlb_nv_bit(
+		const struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	return vcpu_e500->mas0 & 0xfff;
+}
+
+static inline unsigned int get_tlb_esel_bit(
+		const struct kvmppc_vcpu_e500 *vcpu_e500)
+{
+	return (vcpu_e500->mas0 >> 16) & 0xfff;
+}
+
+static inline unsigned int get_tlb_esel(
+		const struct kvmppc_vcpu_e500 *vcpu_e500,
+		int tlbsel)
+{
+	unsigned int esel = get_tlb_esel_bit(vcpu_e500);
+
+	if (tlbsel == 0) {
+		esel &= KVM_E500_TLB0_WAY_NUM_MASK;
+		esel |= ((vcpu_e500->mas2 >> 12) & KVM_E500_TLB0_WAY_SIZE_MASK)
+				<< KVM_E500_TLB0_WAY_NUM_BIT;
+	} else {
+		esel &= KVM_E500_TLB1_SIZE - 1;
+	}
+
+	return esel;
+}
+
+static inline int tlbe_is_host_safe(const struct kvm_vcpu *vcpu,
+			const struct tlbe *tlbe)
+{
+	gpa_t gpa;
+
+	if (!get_tlb_v(tlbe))
+		return 0;
+
+	/* Does it match current guest AS? */
+	/* XXX what about IS != DS? */
+	if (get_tlb_ts(tlbe) != !!(vcpu->arch.msr & MSR_IS))
+		return 0;
+
+	gpa = get_tlb_raddr(tlbe);
+	if (!gfn_to_memslot(vcpu->kvm, gpa >> PAGE_SHIFT))
+		/* Mapping is not for RAM. */
+		return 0;
+
+	return 1;
+}
+
+#endif /* __KVM_E500_TLB_H__ */

arch/powerpc/kvm/emulate.c

@@ -30,6 +30,39 @@
 #include <asm/disassemble.h>
 #include "timing.h"
 
+#define OP_TRAP 3
+
+#define OP_31_XOP_LWZX      23
+#define OP_31_XOP_LBZX      87
+#define OP_31_XOP_STWX      151
+#define OP_31_XOP_STBX      215
+#define OP_31_XOP_STBUX     247
+#define OP_31_XOP_LHZX      279
+#define OP_31_XOP_LHZUX     311
+#define OP_31_XOP_MFSPR     339
+#define OP_31_XOP_STHX      407
+#define OP_31_XOP_STHUX     439
+#define OP_31_XOP_MTSPR     467
+#define OP_31_XOP_DCBI      470
+#define OP_31_XOP_LWBRX     534
+#define OP_31_XOP_TLBSYNC   566
+#define OP_31_XOP_STWBRX    662
+#define OP_31_XOP_LHBRX     790
+#define OP_31_XOP_STHBRX    918
+
+#define OP_LWZ  32
+#define OP_LWZU 33
+#define OP_LBZ  34
+#define OP_LBZU 35
+#define OP_STW  36
+#define OP_STWU 37
+#define OP_STB  38
+#define OP_STBU 39
+#define OP_LHZ  40
+#define OP_LHZU 41
+#define OP_STH  44
+#define OP_STHU 45
+
 void kvmppc_emulate_dec(struct kvm_vcpu *vcpu)
 {
 	if (vcpu->arch.tcr & TCR_DIE) {
@@ -78,7 +111,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	kvmppc_set_exit_type(vcpu, EMULATED_INST_EXITS);
 
 	switch (get_op(inst)) {
-	case 3:                                             /* trap */
+	case OP_TRAP:
 		vcpu->arch.esr |= ESR_PTR;
 		kvmppc_core_queue_program(vcpu);
 		advance = 0;
@@ -87,31 +120,31 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	case 31:
 		switch (get_xop(inst)) {
 
-		case 23:                                        /* lwzx */
+		case OP_31_XOP_LWZX:
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 4, 1);
 			break;
 
-		case 87:                                        /* lbzx */
+		case OP_31_XOP_LBZX:
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1);
 			break;
 
-		case 151:                                       /* stwx */
+		case OP_31_XOP_STWX:
 			rs = get_rs(inst);
 			emulated = kvmppc_handle_store(run, vcpu,
 			                               vcpu->arch.gpr[rs],
 			                               4, 1);
 			break;
 
-		case 215:                                       /* stbx */
+		case OP_31_XOP_STBX:
 			rs = get_rs(inst);
 			emulated = kvmppc_handle_store(run, vcpu,
 			                               vcpu->arch.gpr[rs],
 			                               1, 1);
 			break;
 
-		case 247:                                       /* stbux */
+		case OP_31_XOP_STBUX:
 			rs = get_rs(inst);
 			ra = get_ra(inst);
 			rb = get_rb(inst);
@@ -126,12 +159,12 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			vcpu->arch.gpr[rs] = ea;
 			break;
 
-		case 279:                                       /* lhzx */
+		case OP_31_XOP_LHZX:
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1);
 			break;
 
-		case 311:                                       /* lhzux */
+		case OP_31_XOP_LHZUX:
 			rt = get_rt(inst);
 			ra = get_ra(inst);
 			rb = get_rb(inst);
@@ -144,7 +177,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			vcpu->arch.gpr[ra] = ea;
 			break;
 
-		case 339:                                       /* mfspr */
+		case OP_31_XOP_MFSPR:
 			sprn = get_sprn(inst);
 			rt = get_rt(inst);
 
@@ -185,7 +218,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			}
 			break;
 
-		case 407:                                       /* sthx */
+		case OP_31_XOP_STHX:
 			rs = get_rs(inst);
 			ra = get_ra(inst);
 			rb = get_rb(inst);
@@ -195,7 +228,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			                               2, 1);
 			break;
 
-		case 439:                                       /* sthux */
+		case OP_31_XOP_STHUX:
 			rs = get_rs(inst);
 			ra = get_ra(inst);
 			rb = get_rb(inst);
@@ -210,7 +243,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			vcpu->arch.gpr[ra] = ea;
 			break;
 
-		case 467:                                       /* mtspr */
+		case OP_31_XOP_MTSPR:
 			sprn = get_sprn(inst);
 			rs = get_rs(inst);
 			switch (sprn) {
@@ -246,7 +279,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			}
 			break;
 
-		case 470:                                       /* dcbi */
+		case OP_31_XOP_DCBI:
 			/* Do nothing. The guest is performing dcbi because
 			 * hardware DMA is not snooped by the dcache, but
 			 * emulated DMA either goes through the dcache as
@@ -254,15 +287,15 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			 * coherence. */
 			break;
 
-		case 534:                                       /* lwbrx */
+		case OP_31_XOP_LWBRX:
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 4, 0);
 			break;
 
-		case 566:                                       /* tlbsync */
+		case OP_31_XOP_TLBSYNC:
 			break;
 
-		case 662:                                       /* stwbrx */
+		case OP_31_XOP_STWBRX:
 			rs = get_rs(inst);
 			ra = get_ra(inst);
 			rb = get_rb(inst);
@@ -272,12 +305,12 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 			                               4, 0);
 			break;
 
-		case 790:                                       /* lhbrx */
+		case OP_31_XOP_LHBRX:
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 2, 0);
 			break;
 
-		case 918:                                       /* sthbrx */
+		case OP_31_XOP_STHBRX:
 			rs = get_rs(inst);
 			ra = get_ra(inst);
 			rb = get_rb(inst);
@@ -293,37 +326,37 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		}
 		break;
 
-	case 32:                                                /* lwz */
+	case OP_LWZ:
 		rt = get_rt(inst);
 		emulated = kvmppc_handle_load(run, vcpu, rt, 4, 1);
 		break;
 
-	case 33:                                                /* lwzu */
+	case OP_LWZU:
 		ra = get_ra(inst);
 		rt = get_rt(inst);
 		emulated = kvmppc_handle_load(run, vcpu, rt, 4, 1);
 		vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed;
 		break;
 
-	case 34:                                                /* lbz */
+	case OP_LBZ:
 		rt = get_rt(inst);
 		emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1);
 		break;
 
-	case 35:                                                /* lbzu */
+	case OP_LBZU:
 		ra = get_ra(inst);
 		rt = get_rt(inst);
 		emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1);
 		vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed;
 		break;
 
-	case 36:                                                /* stw */
+	case OP_STW:
 		rs = get_rs(inst);
 		emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs],
 		                               4, 1);
 		break;
 
-	case 37:                                                /* stwu */
+	case OP_STWU:
 		ra = get_ra(inst);
 		rs = get_rs(inst);
 		emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs],
@@ -331,13 +364,13 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed;
 		break;
 
-	case 38:                                                /* stb */
+	case OP_STB:
 		rs = get_rs(inst);
 		emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs],
 		                               1, 1);
 		break;
 
-	case 39:                                                /* stbu */
+	case OP_STBU:
 		ra = get_ra(inst);
 		rs = get_rs(inst);
 		emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs],
@@ -345,25 +378,25 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed;
 		break;
 
-	case 40:                                                /* lhz */
+	case OP_LHZ:
 		rt = get_rt(inst);
 		emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1);
 		break;
 
-	case 41:                                                /* lhzu */
+	case OP_LHZU:
 		ra = get_ra(inst);
 		rt = get_rt(inst);
 		emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1);
 		vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed;
 		break;
 
-	case 44:                                                /* sth */
+	case OP_STH:
 		rs = get_rs(inst);
 		emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs],
 		                               2, 1);
 		break;
 
-	case 45:                                                /* sthu */
+	case OP_STHU:
 		ra = get_ra(inst);
 		rs = get_rs(inst);
 		emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs],

arch/powerpc/kvm/powerpc.c

@@ -216,46 +216,23 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 
 void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 {
-	kvmppc_core_destroy_mmu(vcpu);
+	kvmppc_mmu_destroy(vcpu);
 }
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
-	if (vcpu->guest_debug.enabled)
-		kvmppc_core_load_guest_debugstate(vcpu);
-
 	kvmppc_core_vcpu_load(vcpu, cpu);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
-	if (vcpu->guest_debug.enabled)
-		kvmppc_core_load_host_debugstate(vcpu);
-
-	/* Don't leave guest TLB entries resident when being de-scheduled. */
-	/* XXX It would be nice to differentiate between heavyweight exit and
-	 * sched_out here, since we could avoid the TLB flush for heavyweight
-	 * exits. */
-	_tlbil_all();
 	kvmppc_core_vcpu_put(vcpu);
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-                                    struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+                                        struct kvm_guest_debug *dbg)
 {
-	int i;
-
-	vcpu->guest_debug.enabled = dbg->enabled;
-	if (vcpu->guest_debug.enabled) {
-		for (i=0; i < ARRAY_SIZE(vcpu->guest_debug.bp); i++) {
-			if (dbg->breakpoints[i].enabled)
-				vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address;
-			else
-				vcpu->guest_debug.bp[i] = 0;
-		}
-	}
-
-	return 0;
+	return -EINVAL;
 }
 
 static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,

arch/s390/include/asm/kvm.h

@@ -42,4 +42,11 @@ struct kvm_fpu {
 	__u64 fprs[16];
 };
 
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif

arch/s390/include/asm/kvm_host.h

@@ -21,9 +21,6 @@
 /* memory slots that does not exposed to userspace */
 #define KVM_PRIVATE_MEM_SLOTS 4
 
-struct kvm_guest_debug {
-};
-
 struct sca_entry {
 	atomic_t scn;
 	__u64	reserved;

arch/s390/kvm/Kconfig

@@ -4,6 +4,9 @@
 config HAVE_KVM
        bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	default y

arch/s390/kvm/intercept.c

@@ -103,7 +103,7 @@ static int handle_lctl(struct kvm_vcpu *vcpu)
 static intercept_handler_t instruction_handlers[256] = {
 	[0x83] = kvm_s390_handle_diag,
 	[0xae] = kvm_s390_handle_sigp,
-	[0xb2] = kvm_s390_handle_priv,
+	[0xb2] = kvm_s390_handle_b2,
 	[0xb7] = handle_lctl,
 	[0xeb] = handle_lctlg,
 };

arch/s390/kvm/interrupt.c

@@ -555,9 +555,14 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 		VCPU_EVENT(vcpu, 3, "inject: program check %d (from user)",
 			   s390int->parm);
 		break;
+	case KVM_S390_SIGP_SET_PREFIX:
+		inti->prefix.address = s390int->parm;
+		inti->type = s390int->type;
+		VCPU_EVENT(vcpu, 3, "inject: set prefix to %x (from user)",
+			   s390int->parm);
+		break;
 	case KVM_S390_SIGP_STOP:
 	case KVM_S390_RESTART:
-	case KVM_S390_SIGP_SET_PREFIX:
 	case KVM_S390_INT_EMERGENCY:
 		VCPU_EVENT(vcpu, 3, "inject: type %x", s390int->type);
 		inti->type = s390int->type;

arch/s390/kvm/kvm-s390.c

@@ -422,8 +422,8 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
 	return -EINVAL; /* not implemented yet */
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	return -EINVAL; /* not implemented yet */
 }

arch/s390/kvm/kvm-s390.h

@@ -50,7 +50,7 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code);
 
 /* implemented in priv.c */
-int kvm_s390_handle_priv(struct kvm_vcpu *vcpu);
+int kvm_s390_handle_b2(struct kvm_vcpu *vcpu);
 
 /* implemented in sigp.c */
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu);

arch/s390/kvm/priv.c

@@ -304,12 +304,24 @@ static intercept_handler_t priv_handlers[256] = {
 	[0xb1] = handle_stfl,
 };
 
-int kvm_s390_handle_priv(struct kvm_vcpu *vcpu)
+int kvm_s390_handle_b2(struct kvm_vcpu *vcpu)
 {
 	intercept_handler_t handler;
 
+	/*
+	 * a lot of B2 instructions are priviledged. We first check for
+	 * the priviledges ones, that we can handle in the kernel. If the
+	 * kernel can handle this instruction, we check for the problem
+	 * state bit and (a) handle the instruction or (b) send a code 2
+	 * program check.
+	 * Anything else goes to userspace.*/
 	handler = priv_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
-	if (handler)
-		return handler(vcpu);
+	if (handler) {
+		if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+			return kvm_s390_inject_program_int(vcpu,
+						   PGM_PRIVILEGED_OPERATION);
+		else
+			return handler(vcpu);
+	}
 	return -ENOTSUPP;
 }

arch/s390/kvm/sigp.c

@@ -153,8 +153,6 @@ static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter)
 
 	switch (parameter & 0xff) {
 	case 0:
-		printk(KERN_WARNING "kvm: request to switch to ESA/390 mode"
-							" not supported");
 		rc = 3; /* not operational */
 		break;
 	case 1:

arch/x86/include/asm/kvm.h

@@ -15,6 +15,7 @@
 #define __KVM_HAVE_DEVICE_ASSIGNMENT
 #define __KVM_HAVE_MSI
 #define __KVM_HAVE_USER_NMI
+#define __KVM_HAVE_GUEST_DEBUG
 
 /* Architectural interrupt line count. */
 #define KVM_NR_INTERRUPTS 256
@@ -212,7 +213,30 @@ struct kvm_pit_channel_state {
 	__s64 count_load_time;
 };
 
+struct kvm_debug_exit_arch {
+	__u32 exception;
+	__u32 pad;
+	__u64 pc;
+	__u64 dr6;
+	__u64 dr7;
+};
+
+#define KVM_GUESTDBG_USE_SW_BP		0x00010000
+#define KVM_GUESTDBG_USE_HW_BP		0x00020000
+#define KVM_GUESTDBG_INJECT_DB		0x00040000
+#define KVM_GUESTDBG_INJECT_BP		0x00080000
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+	__u64 debugreg[8];
+};
+
 struct kvm_pit_state {
 	struct kvm_pit_channel_state channels[3];
 };
+
+struct kvm_reinject_control {
+	__u8 pit_reinject;
+	__u8 reserved[31];
+};
 #endif /* _ASM_X86_KVM_H */

arch/x86/include/asm/kvm_host.h

@@ -22,6 +22,7 @@
 #include <asm/pvclock-abi.h>
 #include <asm/desc.h>
 #include <asm/mtrr.h>
+#include <asm/msr-index.h>
 
 #define KVM_MAX_VCPUS 16
 #define KVM_MEMORY_SLOTS 32
@@ -134,11 +135,18 @@ enum {
 
 #define KVM_NR_MEM_OBJS 40
 
-struct kvm_guest_debug {
-	int enabled;
-	unsigned long bp[4];
-	int singlestep;
-};
+#define KVM_NR_DB_REGS	4
+
+#define DR6_BD		(1 << 13)
+#define DR6_BS		(1 << 14)
+#define DR6_FIXED_1	0xffff0ff0
+#define DR6_VOLATILE	0x0000e00f
+
+#define DR7_BP_EN_MASK	0x000000ff
+#define DR7_GE		(1 << 9)
+#define DR7_GD		(1 << 13)
+#define DR7_FIXED_1	0x00000400
+#define DR7_VOLATILE	0xffff23ff
 
 /*
  * We don't want allocation failures within the mmu code, so we preallocate
@@ -162,7 +170,8 @@ struct kvm_pte_chain {
  *   bits 0:3 - total guest paging levels (2-4, or zero for real mode)
  *   bits 4:7 - page table level for this shadow (1-4)
  *   bits 8:9 - page table quadrant for 2-level guests
- *   bit   16 - "metaphysical" - gfn is not a real page (huge page/real mode)
+ *   bit   16 - direct mapping of virtual to physical mapping at gfn
+ *              used for real mode and two-dimensional paging
  *   bits 17:19 - common access permissions for all ptes in this shadow page
  */
 union kvm_mmu_page_role {
@@ -172,9 +181,10 @@ union kvm_mmu_page_role {
 		unsigned level:4;
 		unsigned quadrant:2;
 		unsigned pad_for_nice_hex_output:6;
-		unsigned metaphysical:1;
+		unsigned direct:1;
 		unsigned access:3;
 		unsigned invalid:1;
+		unsigned cr4_pge:1;
 	};
 };
 
@@ -218,6 +228,18 @@ struct kvm_pv_mmu_op_buffer {
 	char buf[512] __aligned(sizeof(long));
 };
 
+struct kvm_pio_request {
+	unsigned long count;
+	int cur_count;
+	gva_t guest_gva;
+	int in;
+	int port;
+	int size;
+	int string;
+	int down;
+	int rep;
+};
+
 /*
  * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
  * 32-bit).  The kvm_mmu structure abstracts the details of the current mmu
@@ -236,6 +258,7 @@ struct kvm_mmu {
 	hpa_t root_hpa;
 	int root_level;
 	int shadow_root_level;
+	union kvm_mmu_page_role base_role;
 
 	u64 *pae_root;
 };
@@ -258,6 +281,7 @@ struct kvm_vcpu_arch {
 	unsigned long cr3;
 	unsigned long cr4;
 	unsigned long cr8;
+	u32 hflags;
 	u64 pdptrs[4]; /* pae */
 	u64 shadow_efer;
 	u64 apic_base;
@@ -338,6 +362,15 @@ struct kvm_vcpu_arch {
 
 	struct mtrr_state_type mtrr_state;
 	u32 pat;
+
+	int switch_db_regs;
+	unsigned long host_db[KVM_NR_DB_REGS];
+	unsigned long host_dr6;
+	unsigned long host_dr7;
+	unsigned long db[KVM_NR_DB_REGS];
+	unsigned long dr6;
+	unsigned long dr7;
+	unsigned long eff_db[KVM_NR_DB_REGS];
 };
 
 struct kvm_mem_alias {
@@ -378,6 +411,7 @@ struct kvm_arch{
 
 	unsigned long irq_sources_bitmap;
 	unsigned long irq_states[KVM_IOAPIC_NUM_PINS];
+	u64 vm_init_tsc;
 };
 
 struct kvm_vm_stat {
@@ -446,8 +480,7 @@ struct kvm_x86_ops {
 	void (*vcpu_put)(struct kvm_vcpu *vcpu);
 
 	int (*set_guest_debug)(struct kvm_vcpu *vcpu,
-			       struct kvm_debug_guest *dbg);
-	void (*guest_debug_pre)(struct kvm_vcpu *vcpu);
+			       struct kvm_guest_debug *dbg);
 	int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
 	int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
 	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
@@ -583,16 +616,12 @@ void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
 void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long cr2,
 			   u32 error_code);
 
-void kvm_pic_set_irq(void *opaque, int irq, int level);
+int kvm_pic_set_irq(void *opaque, int irq, int level);
 
 void kvm_inject_nmi(struct kvm_vcpu *vcpu);
 
 void fx_init(struct kvm_vcpu *vcpu);
 
-int emulator_read_std(unsigned long addr,
-		      void *val,
-		      unsigned int bytes,
-		      struct kvm_vcpu *vcpu);
 int emulator_write_emulated(unsigned long addr,
 			    const void *val,
 			    unsigned int bytes,
@@ -737,6 +766,10 @@ enum {
 	TASK_SWITCH_GATE = 3,
 };
 
+#define HF_GIF_MASK		(1 << 0)
+#define HF_HIF_MASK		(1 << 1)
+#define HF_VINTR_MASK		(1 << 2)
+
 /*
  * Hardware virtualization extension instructions may fault if a
  * reboot turns off virtualization while processes are running.

arch/x86/include/asm/msr-index.h

@@ -18,11 +18,15 @@
 #define _EFER_LME		8  /* Long mode enable */
 #define _EFER_LMA		10 /* Long mode active (read-only) */
 #define _EFER_NX		11 /* No execute enable */
+#define _EFER_SVME		12 /* Enable virtualization */
+#define _EFER_FFXSR		14 /* Enable Fast FXSAVE/FXRSTOR */
 
 #define EFER_SCE		(1<<_EFER_SCE)
 #define EFER_LME		(1<<_EFER_LME)
 #define EFER_LMA		(1<<_EFER_LMA)
 #define EFER_NX			(1<<_EFER_NX)
+#define EFER_SVME		(1<<_EFER_SVME)
+#define EFER_FFXSR		(1<<_EFER_FFXSR)
 
 /* Intel MSRs. Some also available on other CPUs */
 #define MSR_IA32_PERFCTR0		0x000000c1
@@ -360,4 +364,9 @@
 #define MSR_IA32_VMX_PROCBASED_CTLS2    0x0000048b
 #define MSR_IA32_VMX_EPT_VPID_CAP       0x0000048c
 
+/* AMD-V MSRs */
+
+#define MSR_VM_CR                       0xc0010114
+#define MSR_VM_HSAVE_PA                 0xc0010117
+
 #endif /* _ASM_X86_MSR_INDEX_H */

arch/x86/include/asm/svm.h

@@ -174,10 +174,6 @@ struct __attribute__ ((__packed__)) vmcb {
 #define SVM_CPUID_FEATURE_SHIFT 2
 #define SVM_CPUID_FUNC 0x8000000a
 
-#define MSR_EFER_SVME_MASK (1ULL << 12)
-#define MSR_VM_CR       0xc0010114
-#define MSR_VM_HSAVE_PA 0xc0010117ULL
-
 #define SVM_VM_CR_SVM_DISABLE 4
 
 #define SVM_SELECTOR_S_SHIFT 4

arch/x86/include/asm/virtext.h

@@ -118,7 +118,7 @@ static inline void cpu_svm_disable(void)
 
 	wrmsrl(MSR_VM_HSAVE_PA, 0);
 	rdmsrl(MSR_EFER, efer);
-	wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK);
+	wrmsrl(MSR_EFER, efer & ~EFER_SVME);
 }
 
 /** Makes sure SVM is disabled, if it is supported on the CPU

arch/x86/include/asm/vmx.h

@@ -270,8 +270,9 @@ enum vmcs_field {
 
 #define INTR_TYPE_EXT_INTR              (0 << 8) /* external interrupt */
 #define INTR_TYPE_NMI_INTR		(2 << 8) /* NMI */
-#define INTR_TYPE_EXCEPTION             (3 << 8) /* processor exception */
+#define INTR_TYPE_HARD_EXCEPTION	(3 << 8) /* processor exception */
 #define INTR_TYPE_SOFT_INTR             (4 << 8) /* software interrupt */
+#define INTR_TYPE_SOFT_EXCEPTION	(6 << 8) /* software exception */
 
 /* GUEST_INTERRUPTIBILITY_INFO flags. */
 #define GUEST_INTR_STATE_STI		0x00000001
@@ -311,7 +312,7 @@ enum vmcs_field {
 #define DEBUG_REG_ACCESS_TYPE           0x10    /* 4, direction of access */
 #define TYPE_MOV_TO_DR                  (0 << 4)
 #define TYPE_MOV_FROM_DR                (1 << 4)
-#define DEBUG_REG_ACCESS_REG            0xf00   /* 11:8, general purpose reg. */
+#define DEBUG_REG_ACCESS_REG(eq)        (((eq) >> 8) & 0xf) /* 11:8, general purpose reg. */
 
 
 /* segment AR */

arch/x86/kvm/Kconfig

@@ -4,6 +4,10 @@
 config HAVE_KVM
        bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+       default y
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	depends on HAVE_KVM || X86

arch/x86/kvm/i8254.c

@@ -201,6 +201,9 @@ static int __pit_timer_fn(struct kvm_kpit_state *ps)
 	if (!atomic_inc_and_test(&pt->pending))
 		set_bit(KVM_REQ_PENDING_TIMER, &vcpu0->requests);
 
+	if (!pt->reinject)
+		atomic_set(&pt->pending, 1);
+
 	if (vcpu0 && waitqueue_active(&vcpu0->wq))
 		wake_up_interruptible(&vcpu0->wq);
 
@@ -536,6 +539,16 @@ void kvm_pit_reset(struct kvm_pit *pit)
 	pit->pit_state.irq_ack = 1;
 }
 
+static void pit_mask_notifer(struct kvm_irq_mask_notifier *kimn, bool mask)
+{
+	struct kvm_pit *pit = container_of(kimn, struct kvm_pit, mask_notifier);
+
+	if (!mask) {
+		atomic_set(&pit->pit_state.pit_timer.pending, 0);
+		pit->pit_state.irq_ack = 1;
+	}
+}
+
 struct kvm_pit *kvm_create_pit(struct kvm *kvm)
 {
 	struct kvm_pit *pit;
@@ -545,9 +558,7 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm)
 	if (!pit)
 		return NULL;
 
-	mutex_lock(&kvm->lock);
 	pit->irq_source_id = kvm_request_irq_source_id(kvm);
-	mutex_unlock(&kvm->lock);
 	if (pit->irq_source_id < 0) {
 		kfree(pit);
 		return NULL;
@@ -580,10 +591,14 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm)
 	pit_state->irq_ack_notifier.gsi = 0;
 	pit_state->irq_ack_notifier.irq_acked = kvm_pit_ack_irq;
 	kvm_register_irq_ack_notifier(kvm, &pit_state->irq_ack_notifier);
+	pit_state->pit_timer.reinject = true;
 	mutex_unlock(&pit->pit_state.lock);
 
 	kvm_pit_reset(pit);
 
+	pit->mask_notifier.func = pit_mask_notifer;
+	kvm_register_irq_mask_notifier(kvm, 0, &pit->mask_notifier);
+
 	return pit;
 }
 
@@ -592,6 +607,8 @@ void kvm_free_pit(struct kvm *kvm)
 	struct hrtimer *timer;
 
 	if (kvm->arch.vpit) {
+		kvm_unregister_irq_mask_notifier(kvm, 0,
+					       &kvm->arch.vpit->mask_notifier);
 		mutex_lock(&kvm->arch.vpit->pit_state.lock);
 		timer = &kvm->arch.vpit->pit_state.pit_timer.timer;
 		hrtimer_cancel(timer);

arch/x86/kvm/i8254.h

@@ -9,6 +9,7 @@ struct kvm_kpit_timer {
 	s64 period; /* unit: ns */
 	s64 scheduled;
 	atomic_t pending;
+	bool reinject;
 };
 
 struct kvm_kpit_channel_state {
@@ -45,6 +46,7 @@ struct kvm_pit {
 	struct kvm *kvm;
 	struct kvm_kpit_state pit_state;
 	int irq_source_id;
+	struct kvm_irq_mask_notifier mask_notifier;
 };
 
 #define KVM_PIT_BASE_ADDRESS	    0x40

arch/x86/kvm/i8259.c

@@ -32,11 +32,13 @@
 #include <linux/kvm_host.h>
 
 static void pic_lock(struct kvm_pic *s)
+	__acquires(&s->lock)
 {
 	spin_lock(&s->lock);
 }
 
 static void pic_unlock(struct kvm_pic *s)
+	__releases(&s->lock)
 {
 	struct kvm *kvm = s->kvm;
 	unsigned acks = s->pending_acks;
@@ -49,7 +51,8 @@ static void pic_unlock(struct kvm_pic *s)
 	spin_unlock(&s->lock);
 
 	while (acks) {
-		kvm_notify_acked_irq(kvm, __ffs(acks));
+		kvm_notify_acked_irq(kvm, SELECT_PIC(__ffs(acks)),
+				     __ffs(acks));
 		acks &= acks - 1;
 	}
 
@@ -76,12 +79,13 @@ void kvm_pic_clear_isr_ack(struct kvm *kvm)
 /*
  * set irq level. If an edge is detected, then the IRR is set to 1
  */
-static inline void pic_set_irq1(struct kvm_kpic_state *s, int irq, int level)
+static inline int pic_set_irq1(struct kvm_kpic_state *s, int irq, int level)
 {
-	int mask;
+	int mask, ret = 1;
 	mask = 1 << irq;
 	if (s->elcr & mask)	/* level triggered */
 		if (level) {
+			ret = !(s->irr & mask);
 			s->irr |= mask;
 			s->last_irr |= mask;
 		} else {
@@ -90,11 +94,15 @@ static inline void pic_set_irq1(struct kvm_kpic_state *s, int irq, int level)
 		}
 	else	/* edge triggered */
 		if (level) {
-			if ((s->last_irr & mask) == 0)
+			if ((s->last_irr & mask) == 0) {
+				ret = !(s->irr & mask);
 				s->irr |= mask;
+			}
 			s->last_irr |= mask;
 		} else
 			s->last_irr &= ~mask;
+
+	return (s->imr & mask) ? -1 : ret;
 }
 
 /*
@@ -171,16 +179,19 @@ void kvm_pic_update_irq(struct kvm_pic *s)
 	pic_unlock(s);
 }
 
-void kvm_pic_set_irq(void *opaque, int irq, int level)
+int kvm_pic_set_irq(void *opaque, int irq, int level)
 {
 	struct kvm_pic *s = opaque;
+	int ret = -1;
 
 	pic_lock(s);
 	if (irq >= 0 && irq < PIC_NUM_PINS) {
-		pic_set_irq1(&s->pics[irq >> 3], irq & 7, level);
+		ret = pic_set_irq1(&s->pics[irq >> 3], irq & 7, level);
 		pic_update_irq(s);
 	}
 	pic_unlock(s);
+
+	return ret;
 }
 
 /*
@@ -232,7 +243,7 @@ int kvm_pic_read_irq(struct kvm *kvm)
 	}
 	pic_update_irq(s);
 	pic_unlock(s);
-	kvm_notify_acked_irq(kvm, irq);
+	kvm_notify_acked_irq(kvm, SELECT_PIC(irq), irq);
 
 	return intno;
 }

arch/x86/kvm/irq.h

@@ -32,6 +32,8 @@
 #include "lapic.h"
 
 #define PIC_NUM_PINS 16
+#define SELECT_PIC(irq) \
+	((irq) < 8 ? KVM_IRQCHIP_PIC_MASTER : KVM_IRQCHIP_PIC_SLAVE)
 
 struct kvm;
 struct kvm_vcpu;

arch/x86/kvm/kvm_svm.h

@@ -18,7 +18,6 @@ static const u32 host_save_user_msrs[] = {
 };
 
 #define NR_HOST_SAVE_USER_MSRS ARRAY_SIZE(host_save_user_msrs)
-#define NUM_DB_REGS 4
 
 struct kvm_vcpu;
 
@@ -29,18 +28,23 @@ struct vcpu_svm {
 	struct svm_cpu_data *svm_data;
 	uint64_t asid_generation;
 
-	unsigned long db_regs[NUM_DB_REGS];
-
 	u64 next_rip;
 
 	u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS];
 	u64 host_gs_base;
 	unsigned long host_cr2;
-	unsigned long host_db_regs[NUM_DB_REGS];
-	unsigned long host_dr6;
-	unsigned long host_dr7;
 
 	u32 *msrpm;
+	struct vmcb *hsave;
+	u64 hsave_msr;
+
+	u64 nested_vmcb;
+
+	/* These are the merged vectors */
+	u32 *nested_msrpm;
+
+	/* gpa pointers to the real vectors */
+	u64 nested_vmcb_msrpm;
 };
 
 #endif

arch/x86/kvm/mmu.c

@@ -145,11 +145,20 @@ struct kvm_rmap_desc {
 	struct kvm_rmap_desc *more;
 };
 
-struct kvm_shadow_walk {
-	int (*entry)(struct kvm_shadow_walk *walk, struct kvm_vcpu *vcpu,
-		     u64 addr, u64 *spte, int level);
+struct kvm_shadow_walk_iterator {
+	u64 addr;
+	hpa_t shadow_addr;
+	int level;
+	u64 *sptep;
+	unsigned index;
 };
 
+#define for_each_shadow_entry(_vcpu, _addr, _walker)    \
+	for (shadow_walk_init(&(_walker), _vcpu, _addr);	\
+	     shadow_walk_okay(&(_walker));			\
+	     shadow_walk_next(&(_walker)))
+
+
 struct kvm_unsync_walk {
 	int (*entry) (struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk);
 };
@@ -343,7 +352,6 @@ static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc,
 
 	BUG_ON(!mc->nobjs);
 	p = mc->objects[--mc->nobjs];
-	memset(p, 0, size);
 	return p;
 }
 
@@ -794,10 +802,8 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
 	set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
 	list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
 	INIT_LIST_HEAD(&sp->oos_link);
-	ASSERT(is_empty_shadow_page(sp->spt));
 	bitmap_zero(sp->slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS);
 	sp->multimapped = 0;
-	sp->global = 1;
 	sp->parent_pte = parent_pte;
 	--vcpu->kvm->arch.n_free_mmu_pages;
 	return sp;
@@ -983,8 +989,8 @@ struct kvm_mmu_pages {
 	     idx < 512;					\
 	     idx = find_next_bit(bitmap, 512, idx+1))
 
-int mmu_pages_add(struct kvm_mmu_pages *pvec, struct kvm_mmu_page *sp,
-		   int idx)
+static int mmu_pages_add(struct kvm_mmu_pages *pvec, struct kvm_mmu_page *sp,
+			 int idx)
 {
 	int i;
 
@@ -1059,7 +1065,7 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn)
 	index = kvm_page_table_hashfn(gfn);
 	bucket = &kvm->arch.mmu_page_hash[index];
 	hlist_for_each_entry(sp, node, bucket, hash_link)
-		if (sp->gfn == gfn && !sp->role.metaphysical
+		if (sp->gfn == gfn && !sp->role.direct
 		    && !sp->role.invalid) {
 			pgprintk("%s: found role %x\n",
 				 __func__, sp->role.word);
@@ -1115,8 +1121,9 @@ struct mmu_page_path {
 			i < pvec.nr && ({ sp = pvec.page[i].sp; 1;});	\
 			i = mmu_pages_next(&pvec, &parents, i))
 
-int mmu_pages_next(struct kvm_mmu_pages *pvec, struct mmu_page_path *parents,
-		   int i)
+static int mmu_pages_next(struct kvm_mmu_pages *pvec,
+			  struct mmu_page_path *parents,
+			  int i)
 {
 	int n;
 
@@ -1135,7 +1142,7 @@ int mmu_pages_next(struct kvm_mmu_pages *pvec, struct mmu_page_path *parents,
 	return n;
 }
 
-void mmu_pages_clear_parents(struct mmu_page_path *parents)
+static void mmu_pages_clear_parents(struct mmu_page_path *parents)
 {
 	struct kvm_mmu_page *sp;
 	unsigned int level = 0;
@@ -1193,7 +1200,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
 					     gfn_t gfn,
 					     gva_t gaddr,
 					     unsigned level,
-					     int metaphysical,
+					     int direct,
 					     unsigned access,
 					     u64 *parent_pte)
 {
@@ -1204,10 +1211,9 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
 	struct kvm_mmu_page *sp;
 	struct hlist_node *node, *tmp;
 
-	role.word = 0;
-	role.glevels = vcpu->arch.mmu.root_level;
+	role = vcpu->arch.mmu.base_role;
 	role.level = level;
-	role.metaphysical = metaphysical;
+	role.direct = direct;
 	role.access = access;
 	if (vcpu->arch.mmu.root_level <= PT32_ROOT_LEVEL) {
 		quadrant = gaddr >> (PAGE_SHIFT + (PT64_PT_BITS * level));
@@ -1242,8 +1248,9 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
 	pgprintk("%s: adding gfn %lx role %x\n", __func__, gfn, role.word);
 	sp->gfn = gfn;
 	sp->role = role;
+	sp->global = role.cr4_pge;
 	hlist_add_head(&sp->hash_link, bucket);
-	if (!metaphysical) {
+	if (!direct) {
 		if (rmap_write_protect(vcpu->kvm, gfn))
 			kvm_flush_remote_tlbs(vcpu->kvm);
 		account_shadowed(vcpu->kvm, gfn);
@@ -1255,35 +1262,35 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
 	return sp;
 }
 
-static int walk_shadow(struct kvm_shadow_walk *walker,
-		       struct kvm_vcpu *vcpu, u64 addr)
+static void shadow_walk_init(struct kvm_shadow_walk_iterator *iterator,
+			     struct kvm_vcpu *vcpu, u64 addr)
 {
-	hpa_t shadow_addr;
-	int level;
-	int r;
-	u64 *sptep;
-	unsigned index;
-
-	shadow_addr = vcpu->arch.mmu.root_hpa;
-	level = vcpu->arch.mmu.shadow_root_level;
-	if (level == PT32E_ROOT_LEVEL) {
-		shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3];
-		shadow_addr &= PT64_BASE_ADDR_MASK;
-		if (!shadow_addr)
-			return 1;
-		--level;
+	iterator->addr = addr;
+	iterator->shadow_addr = vcpu->arch.mmu.root_hpa;
+	iterator->level = vcpu->arch.mmu.shadow_root_level;
+	if (iterator->level == PT32E_ROOT_LEVEL) {
+		iterator->shadow_addr
+			= vcpu->arch.mmu.pae_root[(addr >> 30) & 3];
+		iterator->shadow_addr &= PT64_BASE_ADDR_MASK;
+		--iterator->level;
+		if (!iterator->shadow_addr)
+			iterator->level = 0;
 	}
+}
 
-	while (level >= PT_PAGE_TABLE_LEVEL) {
-		index = SHADOW_PT_INDEX(addr, level);
-		sptep = ((u64 *)__va(shadow_addr)) + index;
-		r = walker->entry(walker, vcpu, addr, sptep, level);
-		if (r)
-			return r;
-		shadow_addr = *sptep & PT64_BASE_ADDR_MASK;
-		--level;
-	}
-	return 0;
+static bool shadow_walk_okay(struct kvm_shadow_walk_iterator *iterator)
+{
+	if (iterator->level < PT_PAGE_TABLE_LEVEL)
+		return false;
+	iterator->index = SHADOW_PT_INDEX(iterator->addr, iterator->level);
+	iterator->sptep	= ((u64 *)__va(iterator->shadow_addr)) + iterator->index;
+	return true;
+}
+
+static void shadow_walk_next(struct kvm_shadow_walk_iterator *iterator)
+{
+	iterator->shadow_addr = *iterator->sptep & PT64_BASE_ADDR_MASK;
+	--iterator->level;
 }
 
 static void kvm_mmu_page_unlink_children(struct kvm *kvm,
@@ -1388,7 +1395,7 @@ static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
 	kvm_mmu_page_unlink_children(kvm, sp);
 	kvm_mmu_unlink_parents(kvm, sp);
 	kvm_flush_remote_tlbs(kvm);
-	if (!sp->role.invalid && !sp->role.metaphysical)
+	if (!sp->role.invalid && !sp->role.direct)
 		unaccount_shadowed(kvm, sp->gfn);
 	if (sp->unsync)
 		kvm_unlink_unsync_page(kvm, sp);
@@ -1451,7 +1458,7 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
 	index = kvm_page_table_hashfn(gfn);
 	bucket = &kvm->arch.mmu_page_hash[index];
 	hlist_for_each_entry_safe(sp, node, n, bucket, hash_link)
-		if (sp->gfn == gfn && !sp->role.metaphysical) {
+		if (sp->gfn == gfn && !sp->role.direct) {
 			pgprintk("%s: gfn %lx role %x\n", __func__, gfn,
 				 sp->role.word);
 			r = 1;
@@ -1463,11 +1470,20 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
 
 static void mmu_unshadow(struct kvm *kvm, gfn_t gfn)
 {
+	unsigned index;
+	struct hlist_head *bucket;
 	struct kvm_mmu_page *sp;
+	struct hlist_node *node, *nn;
 
-	while ((sp = kvm_mmu_lookup_page(kvm, gfn)) != NULL) {
-		pgprintk("%s: zap %lx %x\n", __func__, gfn, sp->role.word);
-		kvm_mmu_zap_page(kvm, sp);
+	index = kvm_page_table_hashfn(gfn);
+	bucket = &kvm->arch.mmu_page_hash[index];
+	hlist_for_each_entry_safe(sp, node, nn, bucket, hash_link) {
+		if (sp->gfn == gfn && !sp->role.direct
+		    && !sp->role.invalid) {
+			pgprintk("%s: zap %lx %x\n",
+				 __func__, gfn, sp->role.word);
+			kvm_mmu_zap_page(kvm, sp);
+		}
 	}
 }
 
@@ -1622,7 +1638,7 @@ static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
 	/* don't unsync if pagetable is shadowed with multiple roles */
 	hlist_for_each_entry_safe(s, node, n, bucket, hash_link) {
-		if (s->gfn != sp->gfn || s->role.metaphysical)
+		if (s->gfn != sp->gfn || s->role.direct)
 			continue;
 		if (s->role.word != sp->role.word)
 			return 1;
@@ -1669,8 +1685,6 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
 	u64 mt_mask = shadow_mt_mask;
 	struct kvm_mmu_page *sp = page_header(__pa(shadow_pte));
 
-	if (!(vcpu->arch.cr4 & X86_CR4_PGE))
-		global = 0;
 	if (!global && sp->global) {
 		sp->global = 0;
 		if (sp->unsync) {
@@ -1777,12 +1791,8 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
 			pgprintk("hfn old %lx new %lx\n",
 				 spte_to_pfn(*shadow_pte), pfn);
 			rmap_remove(vcpu->kvm, shadow_pte);
-		} else {
-			if (largepage)
-				was_rmapped = is_large_pte(*shadow_pte);
-			else
-				was_rmapped = 1;
-		}
+		} else
+			was_rmapped = 1;
 	}
 	if (set_spte(vcpu, shadow_pte, pte_access, user_fault, write_fault,
 		      dirty, largepage, global, gfn, pfn, speculative, true)) {
@@ -1820,67 +1830,42 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
 {
 }
 
-struct direct_shadow_walk {
-	struct kvm_shadow_walk walker;
-	pfn_t pfn;
-	int write;
-	int largepage;
-	int pt_write;
-};
-
-static int direct_map_entry(struct kvm_shadow_walk *_walk,
-			    struct kvm_vcpu *vcpu,
-			    u64 addr, u64 *sptep, int level)
+static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
+			int largepage, gfn_t gfn, pfn_t pfn)
 {
-	struct direct_shadow_walk *walk =
-		container_of(_walk, struct direct_shadow_walk, walker);
+	struct kvm_shadow_walk_iterator iterator;
 	struct kvm_mmu_page *sp;
+	int pt_write = 0;
 	gfn_t pseudo_gfn;
-	gfn_t gfn = addr >> PAGE_SHIFT;
-
-	if (level == PT_PAGE_TABLE_LEVEL
-	    || (walk->largepage && level == PT_DIRECTORY_LEVEL)) {
-		mmu_set_spte(vcpu, sptep, ACC_ALL, ACC_ALL,
-			     0, walk->write, 1, &walk->pt_write,
-			     walk->largepage, 0, gfn, walk->pfn, false);
-		++vcpu->stat.pf_fixed;
-		return 1;
-	}
 
-	if (*sptep == shadow_trap_nonpresent_pte) {
-		pseudo_gfn = (addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT;
-		sp = kvm_mmu_get_page(vcpu, pseudo_gfn, (gva_t)addr, level - 1,
-				      1, ACC_ALL, sptep);
-		if (!sp) {
-			pgprintk("nonpaging_map: ENOMEM\n");
-			kvm_release_pfn_clean(walk->pfn);
-			return -ENOMEM;
+	for_each_shadow_entry(vcpu, (u64)gfn << PAGE_SHIFT, iterator) {
+		if (iterator.level == PT_PAGE_TABLE_LEVEL
+		    || (largepage && iterator.level == PT_DIRECTORY_LEVEL)) {
+			mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL,
+				     0, write, 1, &pt_write,
+				     largepage, 0, gfn, pfn, false);
+			++vcpu->stat.pf_fixed;
+			break;
 		}
 
-		set_shadow_pte(sptep,
-			       __pa(sp->spt)
-			       | PT_PRESENT_MASK | PT_WRITABLE_MASK
-			       | shadow_user_mask | shadow_x_mask);
-	}
-	return 0;
-}
+		if (*iterator.sptep == shadow_trap_nonpresent_pte) {
+			pseudo_gfn = (iterator.addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT;
+			sp = kvm_mmu_get_page(vcpu, pseudo_gfn, iterator.addr,
+					      iterator.level - 1,
+					      1, ACC_ALL, iterator.sptep);
+			if (!sp) {
+				pgprintk("nonpaging_map: ENOMEM\n");
+				kvm_release_pfn_clean(pfn);
+				return -ENOMEM;
+			}
 
-static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
-			int largepage, gfn_t gfn, pfn_t pfn)
-{
-	int r;
-	struct direct_shadow_walk walker = {
-		.walker = { .entry = direct_map_entry, },
-		.pfn = pfn,
-		.largepage = largepage,
-		.write = write,
-		.pt_write = 0,
-	};
-
-	r = walk_shadow(&walker.walker, vcpu, gfn << PAGE_SHIFT);
-	if (r < 0)
-		return r;
-	return walker.pt_write;
+			set_shadow_pte(iterator.sptep,
+				       __pa(sp->spt)
+				       | PT_PRESENT_MASK | PT_WRITABLE_MASK
+				       | shadow_user_mask | shadow_x_mask);
+		}
+	}
+	return pt_write;
 }
 
 static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
@@ -1962,7 +1947,7 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
 	int i;
 	gfn_t root_gfn;
 	struct kvm_mmu_page *sp;
-	int metaphysical = 0;
+	int direct = 0;
 
 	root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT;
 
@@ -1971,18 +1956,18 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
 
 		ASSERT(!VALID_PAGE(root));
 		if (tdp_enabled)
-			metaphysical = 1;
+			direct = 1;
 		sp = kvm_mmu_get_page(vcpu, root_gfn, 0,
-				      PT64_ROOT_LEVEL, metaphysical,
+				      PT64_ROOT_LEVEL, direct,
 				      ACC_ALL, NULL);
 		root = __pa(sp->spt);
 		++sp->root_count;
 		vcpu->arch.mmu.root_hpa = root;
 		return;
 	}
-	metaphysical = !is_paging(vcpu);
+	direct = !is_paging(vcpu);
 	if (tdp_enabled)
-		metaphysical = 1;
+		direct = 1;
 	for (i = 0; i < 4; ++i) {
 		hpa_t root = vcpu->arch.mmu.pae_root[i];
 
@@ -1996,7 +1981,7 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
 		} else if (vcpu->arch.mmu.root_level == 0)
 			root_gfn = 0;
 		sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
-				      PT32_ROOT_LEVEL, metaphysical,
+				      PT32_ROOT_LEVEL, direct,
 				      ACC_ALL, NULL);
 		root = __pa(sp->spt);
 		++sp->root_count;
@@ -2251,17 +2236,23 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu)
 
 static int init_kvm_softmmu(struct kvm_vcpu *vcpu)
 {
+	int r;
+
 	ASSERT(vcpu);
 	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
 
 	if (!is_paging(vcpu))
-		return nonpaging_init_context(vcpu);
+		r = nonpaging_init_context(vcpu);
 	else if (is_long_mode(vcpu))
-		return paging64_init_context(vcpu);
+		r = paging64_init_context(vcpu);
 	else if (is_pae(vcpu))
-		return paging32E_init_context(vcpu);
+		r = paging32E_init_context(vcpu);
 	else
-		return paging32_init_context(vcpu);
+		r = paging32_init_context(vcpu);
+
+	vcpu->arch.mmu.base_role.glevels = vcpu->arch.mmu.root_level;
+
+	return r;
 }
 
 static int init_kvm_mmu(struct kvm_vcpu *vcpu)
@@ -2492,7 +2483,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 	index = kvm_page_table_hashfn(gfn);
 	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
 	hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) {
-		if (sp->gfn != gfn || sp->role.metaphysical || sp->role.invalid)
+		if (sp->gfn != gfn || sp->role.direct || sp->role.invalid)
 			continue;
 		pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8;
 		misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1);
@@ -3130,7 +3121,7 @@ static void audit_write_protection(struct kvm_vcpu *vcpu)
 	gfn_t gfn;
 
 	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
-		if (sp->role.metaphysical)
+		if (sp->role.direct)
 			continue;
 
 		gfn = unalias_gfn(vcpu->kvm, sp->gfn);

arch/x86/kvm/mmu.h

@@ -54,7 +54,7 @@ static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
 static inline int is_long_mode(struct kvm_vcpu *vcpu)
 {
 #ifdef CONFIG_X86_64
-	return vcpu->arch.shadow_efer & EFER_LME;
+	return vcpu->arch.shadow_efer & EFER_LMA;
 #else
 	return 0;
 #endif

arch/x86/kvm/paging_tmpl.h

@@ -25,7 +25,6 @@
 #if PTTYPE == 64
 	#define pt_element_t u64
 	#define guest_walker guest_walker64
-	#define shadow_walker shadow_walker64
 	#define FNAME(name) paging##64_##name
 	#define PT_BASE_ADDR_MASK PT64_BASE_ADDR_MASK
 	#define PT_DIR_BASE_ADDR_MASK PT64_DIR_BASE_ADDR_MASK
@@ -42,7 +41,6 @@
 #elif PTTYPE == 32
 	#define pt_element_t u32
 	#define guest_walker guest_walker32
-	#define shadow_walker shadow_walker32
 	#define FNAME(name) paging##32_##name
 	#define PT_BASE_ADDR_MASK PT32_BASE_ADDR_MASK
 	#define PT_DIR_BASE_ADDR_MASK PT32_DIR_BASE_ADDR_MASK
@@ -73,18 +71,6 @@ struct guest_walker {
 	u32 error_code;
 };
 
-struct shadow_walker {
-	struct kvm_shadow_walk walker;
-	struct guest_walker *guest_walker;
-	int user_fault;
-	int write_fault;
-	int largepage;
-	int *ptwrite;
-	pfn_t pfn;
-	u64 *sptep;
-	gpa_t pte_gpa;
-};
-
 static gfn_t gpte_to_gfn(pt_element_t gpte)
 {
 	return (gpte & PT_BASE_ADDR_MASK) >> PAGE_SHIFT;
@@ -283,91 +269,79 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
 /*
  * Fetch a shadow pte for a specific level in the paging hierarchy.
  */
-static int FNAME(shadow_walk_entry)(struct kvm_shadow_walk *_sw,
-				    struct kvm_vcpu *vcpu, u64 addr,
-				    u64 *sptep, int level)
+static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
+			 struct guest_walker *gw,
+			 int user_fault, int write_fault, int largepage,
+			 int *ptwrite, pfn_t pfn)
 {
-	struct shadow_walker *sw =
-		container_of(_sw, struct shadow_walker, walker);
-	struct guest_walker *gw = sw->guest_walker;
 	unsigned access = gw->pt_access;
 	struct kvm_mmu_page *shadow_page;
-	u64 spte;
-	int metaphysical;
+	u64 spte, *sptep;
+	int direct;
 	gfn_t table_gfn;
 	int r;
+	int level;
 	pt_element_t curr_pte;
+	struct kvm_shadow_walk_iterator iterator;
 
-	if (level == PT_PAGE_TABLE_LEVEL
-	    || (sw->largepage && level == PT_DIRECTORY_LEVEL)) {
-		mmu_set_spte(vcpu, sptep, access, gw->pte_access & access,
-			     sw->user_fault, sw->write_fault,
-			     gw->ptes[gw->level-1] & PT_DIRTY_MASK,
-			     sw->ptwrite, sw->largepage,
-			     gw->ptes[gw->level-1] & PT_GLOBAL_MASK,
-			     gw->gfn, sw->pfn, false);
-		sw->sptep = sptep;
-		return 1;
-	}
+	if (!is_present_pte(gw->ptes[gw->level - 1]))
+		return NULL;
 
-	if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep))
-		return 0;
+	for_each_shadow_entry(vcpu, addr, iterator) {
+		level = iterator.level;
+		sptep = iterator.sptep;
+		if (level == PT_PAGE_TABLE_LEVEL
+		    || (largepage && level == PT_DIRECTORY_LEVEL)) {
+			mmu_set_spte(vcpu, sptep, access,
+				     gw->pte_access & access,
+				     user_fault, write_fault,
+				     gw->ptes[gw->level-1] & PT_DIRTY_MASK,
+				     ptwrite, largepage,
+				     gw->ptes[gw->level-1] & PT_GLOBAL_MASK,
+				     gw->gfn, pfn, false);
+			break;
+		}
 
-	if (is_large_pte(*sptep)) {
-		set_shadow_pte(sptep, shadow_trap_nonpresent_pte);
-		kvm_flush_remote_tlbs(vcpu->kvm);
-		rmap_remove(vcpu->kvm, sptep);
-	}
+		if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep))
+			continue;
 
-	if (level == PT_DIRECTORY_LEVEL && gw->level == PT_DIRECTORY_LEVEL) {
-		metaphysical = 1;
-		if (!is_dirty_pte(gw->ptes[level - 1]))
-			access &= ~ACC_WRITE_MASK;
-		table_gfn = gpte_to_gfn(gw->ptes[level - 1]);
-	} else {
-		metaphysical = 0;
-		table_gfn = gw->table_gfn[level - 2];
-	}
-	shadow_page = kvm_mmu_get_page(vcpu, table_gfn, (gva_t)addr, level-1,
-				       metaphysical, access, sptep);
-	if (!metaphysical) {
-		r = kvm_read_guest_atomic(vcpu->kvm, gw->pte_gpa[level - 2],
-					  &curr_pte, sizeof(curr_pte));
-		if (r || curr_pte != gw->ptes[level - 2]) {
-			kvm_mmu_put_page(shadow_page, sptep);
-			kvm_release_pfn_clean(sw->pfn);
-			sw->sptep = NULL;
-			return 1;
+		if (is_large_pte(*sptep)) {
+			rmap_remove(vcpu->kvm, sptep);
+			set_shadow_pte(sptep, shadow_trap_nonpresent_pte);
+			kvm_flush_remote_tlbs(vcpu->kvm);
 		}
-	}
 
-	spte = __pa(shadow_page->spt) | PT_PRESENT_MASK | PT_ACCESSED_MASK
-		| PT_WRITABLE_MASK | PT_USER_MASK;
-	*sptep = spte;
-	return 0;
-}
-
-static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
-			 struct guest_walker *guest_walker,
-			 int user_fault, int write_fault, int largepage,
-			 int *ptwrite, pfn_t pfn)
-{
-	struct shadow_walker walker = {
-		.walker = { .entry = FNAME(shadow_walk_entry), },
-		.guest_walker = guest_walker,
-		.user_fault = user_fault,
-		.write_fault = write_fault,
-		.largepage = largepage,
-		.ptwrite = ptwrite,
-		.pfn = pfn,
-	};
-
-	if (!is_present_pte(guest_walker->ptes[guest_walker->level - 1]))
-		return NULL;
+		if (level == PT_DIRECTORY_LEVEL
+		    && gw->level == PT_DIRECTORY_LEVEL) {
+			direct = 1;
+			if (!is_dirty_pte(gw->ptes[level - 1]))
+				access &= ~ACC_WRITE_MASK;
+			table_gfn = gpte_to_gfn(gw->ptes[level - 1]);
+		} else {
+			direct = 0;
+			table_gfn = gw->table_gfn[level - 2];
+		}
+		shadow_page = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1,
+					       direct, access, sptep);
+		if (!direct) {
+			r = kvm_read_guest_atomic(vcpu->kvm,
+						  gw->pte_gpa[level - 2],
+						  &curr_pte, sizeof(curr_pte));
+			if (r || curr_pte != gw->ptes[level - 2]) {
+				kvm_mmu_put_page(shadow_page, sptep);
+				kvm_release_pfn_clean(pfn);
+				sptep = NULL;
+				break;
+			}
+		}
 
-	walk_shadow(&walker.walker, vcpu, addr);
+		spte = __pa(shadow_page->spt)
+			| PT_PRESENT_MASK | PT_ACCESSED_MASK
+			| PT_WRITABLE_MASK | PT_USER_MASK;
+		*sptep = spte;
+	}
 
-	return walker.sptep;
+	return sptep;
 }
 
 /*
@@ -465,54 +439,56 @@ out_unlock:
 	return 0;
 }
 
-static int FNAME(shadow_invlpg_entry)(struct kvm_shadow_walk *_sw,
-				      struct kvm_vcpu *vcpu, u64 addr,
-				      u64 *sptep, int level)
+static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva)
 {
-	struct shadow_walker *sw =
-		container_of(_sw, struct shadow_walker, walker);
-
-	/* FIXME: properly handle invlpg on large guest pages */
-	if (level == PT_PAGE_TABLE_LEVEL ||
-	    ((level == PT_DIRECTORY_LEVEL) && is_large_pte(*sptep))) {
-		struct kvm_mmu_page *sp = page_header(__pa(sptep));
+	struct kvm_shadow_walk_iterator iterator;
+	pt_element_t gpte;
+	gpa_t pte_gpa = -1;
+	int level;
+	u64 *sptep;
+	int need_flush = 0;
 
-		sw->pte_gpa = (sp->gfn << PAGE_SHIFT);
-		sw->pte_gpa += (sptep - sp->spt) * sizeof(pt_element_t);
+	spin_lock(&vcpu->kvm->mmu_lock);
 
-		if (is_shadow_present_pte(*sptep)) {
-			rmap_remove(vcpu->kvm, sptep);
-			if (is_large_pte(*sptep))
-				--vcpu->kvm->stat.lpages;
+	for_each_shadow_entry(vcpu, gva, iterator) {
+		level = iterator.level;
+		sptep = iterator.sptep;
+
+		/* FIXME: properly handle invlpg on large guest pages */
+		if (level == PT_PAGE_TABLE_LEVEL ||
+		    ((level == PT_DIRECTORY_LEVEL) && is_large_pte(*sptep))) {
+			struct kvm_mmu_page *sp = page_header(__pa(sptep));
+
+			pte_gpa = (sp->gfn << PAGE_SHIFT);
+			pte_gpa += (sptep - sp->spt) * sizeof(pt_element_t);
+
+			if (is_shadow_present_pte(*sptep)) {
+				rmap_remove(vcpu->kvm, sptep);
+				if (is_large_pte(*sptep))
+					--vcpu->kvm->stat.lpages;
+				need_flush = 1;
+			}
+			set_shadow_pte(sptep, shadow_trap_nonpresent_pte);
+			break;
 		}
-		set_shadow_pte(sptep, shadow_trap_nonpresent_pte);
-		return 1;
-	}
-	if (!is_shadow_present_pte(*sptep))
-		return 1;
-	return 0;
-}
 
-static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva)
-{
-	pt_element_t gpte;
-	struct shadow_walker walker = {
-		.walker = { .entry = FNAME(shadow_invlpg_entry), },
-		.pte_gpa = -1,
-	};
+		if (!is_shadow_present_pte(*sptep))
+			break;
+	}
 
-	spin_lock(&vcpu->kvm->mmu_lock);
-	walk_shadow(&walker.walker, vcpu, gva);
+	if (need_flush)
+		kvm_flush_remote_tlbs(vcpu->kvm);
 	spin_unlock(&vcpu->kvm->mmu_lock);
-	if (walker.pte_gpa == -1)
+
+	if (pte_gpa == -1)
 		return;
-	if (kvm_read_guest_atomic(vcpu->kvm, walker.pte_gpa, &gpte,
+	if (kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &gpte,
 				  sizeof(pt_element_t)))
 		return;
 	if (is_present_pte(gpte) && (gpte & PT_ACCESSED_MASK)) {
 		if (mmu_topup_memory_caches(vcpu))
 			return;
-		kvm_mmu_pte_write(vcpu, walker.pte_gpa, (const u8 *)&gpte,
+		kvm_mmu_pte_write(vcpu, pte_gpa, (const u8 *)&gpte,
 				  sizeof(pt_element_t), 0);
 	}
 }
@@ -540,7 +516,7 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
 	pt_element_t pt[256 / sizeof(pt_element_t)];
 	gpa_t pte_gpa;
 
-	if (sp->role.metaphysical
+	if (sp->role.direct
 	    || (PTTYPE == 32 && sp->role.level > PT_PAGE_TABLE_LEVEL)) {
 		nonpaging_prefetch_page(vcpu, sp);
 		return;
@@ -619,7 +595,6 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 
 #undef pt_element_t
 #undef guest_walker
-#undef shadow_walker
 #undef FNAME
 #undef PT_BASE_ADDR_MASK
 #undef PT_INDEX

arch/x86/kvm/svm.c

@@ -38,9 +38,6 @@ MODULE_LICENSE("GPL");
 #define IOPM_ALLOC_ORDER 2
 #define MSRPM_ALLOC_ORDER 1
 
-#define DR7_GD_MASK (1 << 13)
-#define DR6_BD_MASK (1 << 13)
-
 #define SEG_TYPE_LDT 2
 #define SEG_TYPE_BUSY_TSS16 3
 
@@ -50,6 +47,15 @@ MODULE_LICENSE("GPL");
 
 #define DEBUGCTL_RESERVED_BITS (~(0x3fULL))
 
+/* Turn on to get debugging output*/
+/* #define NESTED_DEBUG */
+
+#ifdef NESTED_DEBUG
+#define nsvm_printk(fmt, args...) printk(KERN_INFO fmt, ## args)
+#else
+#define nsvm_printk(fmt, args...) do {} while(0)
+#endif
+
 /* enable NPT for AMD64 and X86 with PAE */
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
 static bool npt_enabled = true;
@@ -60,14 +66,29 @@ static int npt = 1;
 
 module_param(npt, int, S_IRUGO);
 
+static int nested = 0;
+module_param(nested, int, S_IRUGO);
+
 static void kvm_reput_irq(struct vcpu_svm *svm);
 static void svm_flush_tlb(struct kvm_vcpu *vcpu);
 
+static int nested_svm_exit_handled(struct vcpu_svm *svm, bool kvm_override);
+static int nested_svm_vmexit(struct vcpu_svm *svm);
+static int nested_svm_vmsave(struct vcpu_svm *svm, void *nested_vmcb,
+			     void *arg2, void *opaque);
+static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr,
+				      bool has_error_code, u32 error_code);
+
 static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu)
 {
 	return container_of(vcpu, struct vcpu_svm, vcpu);
 }
 
+static inline bool is_nested(struct vcpu_svm *svm)
+{
+	return svm->nested_vmcb;
+}
+
 static unsigned long iopm_base;
 
 struct kvm_ldttss_desc {
@@ -157,32 +178,6 @@ static inline void kvm_write_cr2(unsigned long val)
 	asm volatile ("mov %0, %%cr2" :: "r" (val));
 }
 
-static inline unsigned long read_dr6(void)
-{
-	unsigned long dr6;
-
-	asm volatile ("mov %%dr6, %0" : "=r" (dr6));
-	return dr6;
-}
-
-static inline void write_dr6(unsigned long val)
-{
-	asm volatile ("mov %0, %%dr6" :: "r" (val));
-}
-
-static inline unsigned long read_dr7(void)
-{
-	unsigned long dr7;
-
-	asm volatile ("mov %%dr7, %0" : "=r" (dr7));
-	return dr7;
-}
-
-static inline void write_dr7(unsigned long val)
-{
-	asm volatile ("mov %0, %%dr7" :: "r" (val));
-}
-
 static inline void force_new_asid(struct kvm_vcpu *vcpu)
 {
 	to_svm(vcpu)->asid_generation--;
@@ -198,7 +193,7 @@ static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer)
 	if (!npt_enabled && !(efer & EFER_LMA))
 		efer &= ~EFER_LME;
 
-	to_svm(vcpu)->vmcb->save.efer = efer | MSR_EFER_SVME_MASK;
+	to_svm(vcpu)->vmcb->save.efer = efer | EFER_SVME;
 	vcpu->arch.shadow_efer = efer;
 }
 
@@ -207,6 +202,11 @@ static void svm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
 
+	/* If we are within a nested VM we'd better #VMEXIT and let the
+	   guest handle the exception */
+	if (nested_svm_check_exception(svm, nr, has_error_code, error_code))
+		return;
+
 	svm->vmcb->control.event_inj = nr
 		| SVM_EVTINJ_VALID
 		| (has_error_code ? SVM_EVTINJ_VALID_ERR : 0)
@@ -242,7 +242,7 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
 	kvm_rip_write(vcpu, svm->next_rip);
 	svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
 
-	vcpu->arch.interrupt_window_open = 1;
+	vcpu->arch.interrupt_window_open = (svm->vcpu.arch.hflags & HF_GIF_MASK);
 }
 
 static int has_svm(void)
@@ -250,7 +250,7 @@ static int has_svm(void)
 	const char *msg;
 
 	if (!cpu_has_svm(&msg)) {
-		printk(KERN_INFO "has_svn: %s\n", msg);
+		printk(KERN_INFO "has_svm: %s\n", msg);
 		return 0;
 	}
 
@@ -292,7 +292,7 @@ static void svm_hardware_enable(void *garbage)
 	svm_data->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS);
 
 	rdmsrl(MSR_EFER, efer);
-	wrmsrl(MSR_EFER, efer | MSR_EFER_SVME_MASK);
+	wrmsrl(MSR_EFER, efer | EFER_SVME);
 
 	wrmsrl(MSR_VM_HSAVE_PA,
 	       page_to_pfn(svm_data->save_area) << PAGE_SHIFT);
@@ -417,6 +417,14 @@ static __init int svm_hardware_setup(void)
 	if (boot_cpu_has(X86_FEATURE_NX))
 		kvm_enable_efer_bits(EFER_NX);
 
+	if (boot_cpu_has(X86_FEATURE_FXSR_OPT))
+		kvm_enable_efer_bits(EFER_FFXSR);
+
+	if (nested) {
+		printk(KERN_INFO "kvm: Nested Virtualization enabled\n");
+		kvm_enable_efer_bits(EFER_SVME);
+	}
+
 	for_each_online_cpu(cpu) {
 		r = svm_cpu_init(cpu);
 		if (r)
@@ -559,7 +567,7 @@ static void init_vmcb(struct vcpu_svm *svm)
 	init_sys_seg(&save->ldtr, SEG_TYPE_LDT);
 	init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16);
 
-	save->efer = MSR_EFER_SVME_MASK;
+	save->efer = EFER_SVME;
 	save->dr6 = 0xffff0ff0;
 	save->dr7 = 0x400;
 	save->rflags = 2;
@@ -591,6 +599,9 @@ static void init_vmcb(struct vcpu_svm *svm)
 		save->cr4 = 0;
 	}
 	force_new_asid(&svm->vcpu);
+
+	svm->nested_vmcb = 0;
+	svm->vcpu.arch.hflags = HF_GIF_MASK;
 }
 
 static int svm_vcpu_reset(struct kvm_vcpu *vcpu)
@@ -615,6 +626,8 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
 	struct vcpu_svm *svm;
 	struct page *page;
 	struct page *msrpm_pages;
+	struct page *hsave_page;
+	struct page *nested_msrpm_pages;
 	int err;
 
 	svm = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
@@ -637,14 +650,25 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
 	msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER);
 	if (!msrpm_pages)
 		goto uninit;
+
+	nested_msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER);
+	if (!nested_msrpm_pages)
+		goto uninit;
+
 	svm->msrpm = page_address(msrpm_pages);
 	svm_vcpu_init_msrpm(svm->msrpm);
 
+	hsave_page = alloc_page(GFP_KERNEL);
+	if (!hsave_page)
+		goto uninit;
+	svm->hsave = page_address(hsave_page);
+
+	svm->nested_msrpm = page_address(nested_msrpm_pages);
+
 	svm->vmcb = page_address(page);
 	clear_page(svm->vmcb);
 	svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT;
 	svm->asid_generation = 0;
-	memset(svm->db_regs, 0, sizeof(svm->db_regs));
 	init_vmcb(svm);
 
 	fx_init(&svm->vcpu);
@@ -669,6 +693,8 @@ static void svm_free_vcpu(struct kvm_vcpu *vcpu)
 
 	__free_page(pfn_to_page(svm->vmcb_pa >> PAGE_SHIFT));
 	__free_pages(virt_to_page(svm->msrpm), MSRPM_ALLOC_ORDER);
+	__free_page(virt_to_page(svm->hsave));
+	__free_pages(virt_to_page(svm->nested_msrpm), MSRPM_ALLOC_ORDER);
 	kvm_vcpu_uninit(vcpu);
 	kmem_cache_free(kvm_vcpu_cache, svm);
 }
@@ -718,6 +744,16 @@ static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
 	to_svm(vcpu)->vmcb->save.rflags = rflags;
 }
 
+static void svm_set_vintr(struct vcpu_svm *svm)
+{
+	svm->vmcb->control.intercept |= 1ULL << INTERCEPT_VINTR;
+}
+
+static void svm_clear_vintr(struct vcpu_svm *svm)
+{
+	svm->vmcb->control.intercept &= ~(1ULL << INTERCEPT_VINTR);
+}
+
 static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg)
 {
 	struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save;
@@ -760,20 +796,37 @@ static void svm_get_segment(struct kvm_vcpu *vcpu,
 	var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1;
 	var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1;
 
-	/*
-	 * SVM always stores 0 for the 'G' bit in the CS selector in
-	 * the VMCB on a VMEXIT. This hurts cross-vendor migration:
-	 * Intel's VMENTRY has a check on the 'G' bit.
-	 */
-	if (seg == VCPU_SREG_CS)
+	switch (seg) {
+	case VCPU_SREG_CS:
+		/*
+		 * SVM always stores 0 for the 'G' bit in the CS selector in
+		 * the VMCB on a VMEXIT. This hurts cross-vendor migration:
+		 * Intel's VMENTRY has a check on the 'G' bit.
+		 */
 		var->g = s->limit > 0xfffff;
-
-	/*
-	 * Work around a bug where the busy flag in the tr selector
-	 * isn't exposed
-	 */
-	if (seg == VCPU_SREG_TR)
+		break;
+	case VCPU_SREG_TR:
+		/*
+		 * Work around a bug where the busy flag in the tr selector
+		 * isn't exposed
+		 */
 		var->type |= 0x2;
+		break;
+	case VCPU_SREG_DS:
+	case VCPU_SREG_ES:
+	case VCPU_SREG_FS:
+	case VCPU_SREG_GS:
+		/*
+		 * The accessed bit must always be set in the segment
+		 * descriptor cache, although it can be cleared in the
+		 * descriptor, the cached bit always remains at 1. Since
+		 * Intel has a check on this, set it here to support
+		 * cross-vendor migration.
+		 */
+		if (!var->unusable)
+			var->type |= 0x1;
+		break;
+	}
 
 	var->unusable = !var->present;
 }
@@ -905,9 +958,37 @@ static void svm_set_segment(struct kvm_vcpu *vcpu,
 
 }
 
-static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
+static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg)
 {
-	return -EOPNOTSUPP;
+	int old_debug = vcpu->guest_debug;
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	vcpu->guest_debug = dbg->control;
+
+	svm->vmcb->control.intercept_exceptions &=
+		~((1 << DB_VECTOR) | (1 << BP_VECTOR));
+	if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) {
+		if (vcpu->guest_debug &
+		    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+			svm->vmcb->control.intercept_exceptions |=
+				1 << DB_VECTOR;
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			svm->vmcb->control.intercept_exceptions |=
+				1 << BP_VECTOR;
+	} else
+		vcpu->guest_debug = 0;
+
+	if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)
+		svm->vmcb->save.dr7 = dbg->arch.debugreg[7];
+	else
+		svm->vmcb->save.dr7 = vcpu->arch.dr7;
+
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		svm->vmcb->save.rflags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
+	else if (old_debug & KVM_GUESTDBG_SINGLESTEP)
+		svm->vmcb->save.rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
+
+	return 0;
 }
 
 static int svm_get_irq(struct kvm_vcpu *vcpu)
@@ -949,7 +1030,29 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *svm_data)
 
 static unsigned long svm_get_dr(struct kvm_vcpu *vcpu, int dr)
 {
-	unsigned long val = to_svm(vcpu)->db_regs[dr];
+	struct vcpu_svm *svm = to_svm(vcpu);
+	unsigned long val;
+
+	switch (dr) {
+	case 0 ... 3:
+		val = vcpu->arch.db[dr];
+		break;
+	case 6:
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)
+			val = vcpu->arch.dr6;
+		else
+			val = svm->vmcb->save.dr6;
+		break;
+	case 7:
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)
+			val = vcpu->arch.dr7;
+		else
+			val = svm->vmcb->save.dr7;
+		break;
+	default:
+		val = 0;
+	}
+
 	KVMTRACE_2D(DR_READ, vcpu, (u32)dr, (u32)val, handler);
 	return val;
 }
@@ -959,33 +1062,40 @@ static void svm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long value,
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
 
-	*exception = 0;
+	KVMTRACE_2D(DR_WRITE, vcpu, (u32)dr, (u32)value, handler);
 
-	if (svm->vmcb->save.dr7 & DR7_GD_MASK) {
-		svm->vmcb->save.dr7 &= ~DR7_GD_MASK;
-		svm->vmcb->save.dr6 |= DR6_BD_MASK;
-		*exception = DB_VECTOR;
-		return;
-	}
+	*exception = 0;
 
 	switch (dr) {
 	case 0 ... 3:
-		svm->db_regs[dr] = value;
+		vcpu->arch.db[dr] = value;
+		if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP))
+			vcpu->arch.eff_db[dr] = value;
 		return;
 	case 4 ... 5:
-		if (vcpu->arch.cr4 & X86_CR4_DE) {
+		if (vcpu->arch.cr4 & X86_CR4_DE)
 			*exception = UD_VECTOR;
+		return;
+	case 6:
+		if (value & 0xffffffff00000000ULL) {
+			*exception = GP_VECTOR;
 			return;
 		}
-	case 7: {
-		if (value & ~((1ULL << 32) - 1)) {
+		vcpu->arch.dr6 = (value & DR6_VOLATILE) | DR6_FIXED_1;
+		return;
+	case 7:
+		if (value & 0xffffffff00000000ULL) {
 			*exception = GP_VECTOR;
 			return;
 		}
-		svm->vmcb->save.dr7 = value;
+		vcpu->arch.dr7 = (value & DR7_VOLATILE) | DR7_FIXED_1;
+		if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) {
+			svm->vmcb->save.dr7 = vcpu->arch.dr7;
+			vcpu->arch.switch_db_regs = (value & DR7_BP_EN_MASK);
+		}
 		return;
-	}
 	default:
+		/* FIXME: Possible case? */
 		printk(KERN_DEBUG "%s: unexpected dr %u\n",
 		       __func__, dr);
 		*exception = UD_VECTOR;
@@ -1031,6 +1141,27 @@ static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 	return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code);
 }
 
+static int db_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (!(svm->vcpu.guest_debug &
+	      (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))) {
+		kvm_queue_exception(&svm->vcpu, DB_VECTOR);
+		return 1;
+	}
+	kvm_run->exit_reason = KVM_EXIT_DEBUG;
+	kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip;
+	kvm_run->debug.arch.exception = DB_VECTOR;
+	return 0;
+}
+
+static int bp_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	kvm_run->exit_reason = KVM_EXIT_DEBUG;
+	kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip;
+	kvm_run->debug.arch.exception = BP_VECTOR;
+	return 0;
+}
+
 static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 {
 	int er;
@@ -1080,7 +1211,7 @@ static int shutdown_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 static int io_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 {
 	u32 io_info = svm->vmcb->control.exit_info_1; /* address size bug? */
-	int size, down, in, string, rep;
+	int size, in, string;
 	unsigned port;
 
 	++svm->vcpu.stat.io_exits;
@@ -1099,8 +1230,6 @@ static int io_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 	in = (io_info & SVM_IOIO_TYPE_MASK) != 0;
 	port = io_info >> 16;
 	size = (io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT;
-	rep = (io_info & SVM_IOIO_REP_MASK) != 0;
-	down = (svm->vmcb->save.rflags & X86_EFLAGS_DF) != 0;
 
 	skip_emulated_instruction(&svm->vcpu);
 	return kvm_emulate_pio(&svm->vcpu, kvm_run, in, size, port);
@@ -1139,6 +1268,567 @@ static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 	return 1;
 }
 
+static int nested_svm_check_permissions(struct vcpu_svm *svm)
+{
+	if (!(svm->vcpu.arch.shadow_efer & EFER_SVME)
+	    || !is_paging(&svm->vcpu)) {
+		kvm_queue_exception(&svm->vcpu, UD_VECTOR);
+		return 1;
+	}
+
+	if (svm->vmcb->save.cpl) {
+		kvm_inject_gp(&svm->vcpu, 0);
+		return 1;
+	}
+
+       return 0;
+}
+
+static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr,
+				      bool has_error_code, u32 error_code)
+{
+	if (is_nested(svm)) {
+		svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + nr;
+		svm->vmcb->control.exit_code_hi = 0;
+		svm->vmcb->control.exit_info_1 = error_code;
+		svm->vmcb->control.exit_info_2 = svm->vcpu.arch.cr2;
+		if (nested_svm_exit_handled(svm, false)) {
+			nsvm_printk("VMexit -> EXCP 0x%x\n", nr);
+
+			nested_svm_vmexit(svm);
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
+static inline int nested_svm_intr(struct vcpu_svm *svm)
+{
+	if (is_nested(svm)) {
+		if (!(svm->vcpu.arch.hflags & HF_VINTR_MASK))
+			return 0;
+
+		if (!(svm->vcpu.arch.hflags & HF_HIF_MASK))
+			return 0;
+
+		svm->vmcb->control.exit_code = SVM_EXIT_INTR;
+
+		if (nested_svm_exit_handled(svm, false)) {
+			nsvm_printk("VMexit -> INTR\n");
+			nested_svm_vmexit(svm);
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
+static struct page *nested_svm_get_page(struct vcpu_svm *svm, u64 gpa)
+{
+	struct page *page;
+
+	down_read(&current->mm->mmap_sem);
+	page = gfn_to_page(svm->vcpu.kvm, gpa >> PAGE_SHIFT);
+	up_read(&current->mm->mmap_sem);
+
+	if (is_error_page(page)) {
+		printk(KERN_INFO "%s: could not find page at 0x%llx\n",
+		       __func__, gpa);
+		kvm_release_page_clean(page);
+		kvm_inject_gp(&svm->vcpu, 0);
+		return NULL;
+	}
+	return page;
+}
+
+static int nested_svm_do(struct vcpu_svm *svm,
+			 u64 arg1_gpa, u64 arg2_gpa, void *opaque,
+			 int (*handler)(struct vcpu_svm *svm,
+					void *arg1,
+					void *arg2,
+					void *opaque))
+{
+	struct page *arg1_page;
+	struct page *arg2_page = NULL;
+	void *arg1;
+	void *arg2 = NULL;
+	int retval;
+
+	arg1_page = nested_svm_get_page(svm, arg1_gpa);
+	if(arg1_page == NULL)
+		return 1;
+
+	if (arg2_gpa) {
+		arg2_page = nested_svm_get_page(svm, arg2_gpa);
+		if(arg2_page == NULL) {
+			kvm_release_page_clean(arg1_page);
+			return 1;
+		}
+	}
+
+	arg1 = kmap_atomic(arg1_page, KM_USER0);
+	if (arg2_gpa)
+		arg2 = kmap_atomic(arg2_page, KM_USER1);
+
+	retval = handler(svm, arg1, arg2, opaque);
+
+	kunmap_atomic(arg1, KM_USER0);
+	if (arg2_gpa)
+		kunmap_atomic(arg2, KM_USER1);
+
+	kvm_release_page_dirty(arg1_page);
+	if (arg2_gpa)
+		kvm_release_page_dirty(arg2_page);
+
+	return retval;
+}
+
+static int nested_svm_exit_handled_real(struct vcpu_svm *svm,
+					void *arg1,
+					void *arg2,
+					void *opaque)
+{
+	struct vmcb *nested_vmcb = (struct vmcb *)arg1;
+	bool kvm_overrides = *(bool *)opaque;
+	u32 exit_code = svm->vmcb->control.exit_code;
+
+	if (kvm_overrides) {
+		switch (exit_code) {
+		case SVM_EXIT_INTR:
+		case SVM_EXIT_NMI:
+			return 0;
+		/* For now we are always handling NPFs when using them */
+		case SVM_EXIT_NPF:
+			if (npt_enabled)
+				return 0;
+			break;
+		/* When we're shadowing, trap PFs */
+		case SVM_EXIT_EXCP_BASE + PF_VECTOR:
+			if (!npt_enabled)
+				return 0;
+			break;
+		default:
+			break;
+		}
+	}
+
+	switch (exit_code) {
+	case SVM_EXIT_READ_CR0 ... SVM_EXIT_READ_CR8: {
+		u32 cr_bits = 1 << (exit_code - SVM_EXIT_READ_CR0);
+		if (nested_vmcb->control.intercept_cr_read & cr_bits)
+			return 1;
+		break;
+	}
+	case SVM_EXIT_WRITE_CR0 ... SVM_EXIT_WRITE_CR8: {
+		u32 cr_bits = 1 << (exit_code - SVM_EXIT_WRITE_CR0);
+		if (nested_vmcb->control.intercept_cr_write & cr_bits)
+			return 1;
+		break;
+	}
+	case SVM_EXIT_READ_DR0 ... SVM_EXIT_READ_DR7: {
+		u32 dr_bits = 1 << (exit_code - SVM_EXIT_READ_DR0);
+		if (nested_vmcb->control.intercept_dr_read & dr_bits)
+			return 1;
+		break;
+	}
+	case SVM_EXIT_WRITE_DR0 ... SVM_EXIT_WRITE_DR7: {
+		u32 dr_bits = 1 << (exit_code - SVM_EXIT_WRITE_DR0);
+		if (nested_vmcb->control.intercept_dr_write & dr_bits)
+			return 1;
+		break;
+	}
+	case SVM_EXIT_EXCP_BASE ... SVM_EXIT_EXCP_BASE + 0x1f: {
+		u32 excp_bits = 1 << (exit_code - SVM_EXIT_EXCP_BASE);
+		if (nested_vmcb->control.intercept_exceptions & excp_bits)
+			return 1;
+		break;
+	}
+	default: {
+		u64 exit_bits = 1ULL << (exit_code - SVM_EXIT_INTR);
+		nsvm_printk("exit code: 0x%x\n", exit_code);
+		if (nested_vmcb->control.intercept & exit_bits)
+			return 1;
+	}
+	}
+
+	return 0;
+}
+
+static int nested_svm_exit_handled_msr(struct vcpu_svm *svm,
+				       void *arg1, void *arg2,
+				       void *opaque)
+{
+	struct vmcb *nested_vmcb = (struct vmcb *)arg1;
+	u8 *msrpm = (u8 *)arg2;
+        u32 t0, t1;
+	u32 msr = svm->vcpu.arch.regs[VCPU_REGS_RCX];
+	u32 param = svm->vmcb->control.exit_info_1 & 1;
+
+	if (!(nested_vmcb->control.intercept & (1ULL << INTERCEPT_MSR_PROT)))
+		return 0;
+
+	switch(msr) {
+	case 0 ... 0x1fff:
+		t0 = (msr * 2) % 8;
+		t1 = msr / 8;
+		break;
+	case 0xc0000000 ... 0xc0001fff:
+		t0 = (8192 + msr - 0xc0000000) * 2;
+		t1 = (t0 / 8);
+		t0 %= 8;
+		break;
+	case 0xc0010000 ... 0xc0011fff:
+		t0 = (16384 + msr - 0xc0010000) * 2;
+		t1 = (t0 / 8);
+		t0 %= 8;
+		break;
+	default:
+		return 1;
+		break;
+	}
+	if (msrpm[t1] & ((1 << param) << t0))
+		return 1;
+
+	return 0;
+}
+
+static int nested_svm_exit_handled(struct vcpu_svm *svm, bool kvm_override)
+{
+	bool k = kvm_override;
+
+	switch (svm->vmcb->control.exit_code) {
+	case SVM_EXIT_MSR:
+		return nested_svm_do(svm, svm->nested_vmcb,
+				     svm->nested_vmcb_msrpm, NULL,
+				     nested_svm_exit_handled_msr);
+	default: break;
+	}
+
+	return nested_svm_do(svm, svm->nested_vmcb, 0, &k,
+			     nested_svm_exit_handled_real);
+}
+
+static int nested_svm_vmexit_real(struct vcpu_svm *svm, void *arg1,
+				  void *arg2, void *opaque)
+{
+	struct vmcb *nested_vmcb = (struct vmcb *)arg1;
+	struct vmcb *hsave = svm->hsave;
+	u64 nested_save[] = { nested_vmcb->save.cr0,
+			      nested_vmcb->save.cr3,
+			      nested_vmcb->save.cr4,
+			      nested_vmcb->save.efer,
+			      nested_vmcb->control.intercept_cr_read,
+			      nested_vmcb->control.intercept_cr_write,
+			      nested_vmcb->control.intercept_dr_read,
+			      nested_vmcb->control.intercept_dr_write,
+			      nested_vmcb->control.intercept_exceptions,
+			      nested_vmcb->control.intercept,
+			      nested_vmcb->control.msrpm_base_pa,
+			      nested_vmcb->control.iopm_base_pa,
+			      nested_vmcb->control.tsc_offset };
+
+	/* Give the current vmcb to the guest */
+	memcpy(nested_vmcb, svm->vmcb, sizeof(struct vmcb));
+	nested_vmcb->save.cr0 = nested_save[0];
+	if (!npt_enabled)
+		nested_vmcb->save.cr3 = nested_save[1];
+	nested_vmcb->save.cr4 = nested_save[2];
+	nested_vmcb->save.efer = nested_save[3];
+	nested_vmcb->control.intercept_cr_read = nested_save[4];
+	nested_vmcb->control.intercept_cr_write = nested_save[5];
+	nested_vmcb->control.intercept_dr_read = nested_save[6];
+	nested_vmcb->control.intercept_dr_write = nested_save[7];
+	nested_vmcb->control.intercept_exceptions = nested_save[8];
+	nested_vmcb->control.intercept = nested_save[9];
+	nested_vmcb->control.msrpm_base_pa = nested_save[10];
+	nested_vmcb->control.iopm_base_pa = nested_save[11];
+	nested_vmcb->control.tsc_offset = nested_save[12];
+
+	/* We always set V_INTR_MASKING and remember the old value in hflags */
+	if (!(svm->vcpu.arch.hflags & HF_VINTR_MASK))
+		nested_vmcb->control.int_ctl &= ~V_INTR_MASKING_MASK;
+
+	if ((nested_vmcb->control.int_ctl & V_IRQ_MASK) &&
+	    (nested_vmcb->control.int_vector)) {
+		nsvm_printk("WARNING: IRQ 0x%x still enabled on #VMEXIT\n",
+				nested_vmcb->control.int_vector);
+	}
+
+	/* Restore the original control entries */
+	svm->vmcb->control = hsave->control;
+
+	/* Kill any pending exceptions */
+	if (svm->vcpu.arch.exception.pending == true)
+		nsvm_printk("WARNING: Pending Exception\n");
+	svm->vcpu.arch.exception.pending = false;
+
+	/* Restore selected save entries */
+	svm->vmcb->save.es = hsave->save.es;
+	svm->vmcb->save.cs = hsave->save.cs;
+	svm->vmcb->save.ss = hsave->save.ss;
+	svm->vmcb->save.ds = hsave->save.ds;
+	svm->vmcb->save.gdtr = hsave->save.gdtr;
+	svm->vmcb->save.idtr = hsave->save.idtr;
+	svm->vmcb->save.rflags = hsave->save.rflags;
+	svm_set_efer(&svm->vcpu, hsave->save.efer);
+	svm_set_cr0(&svm->vcpu, hsave->save.cr0 | X86_CR0_PE);
+	svm_set_cr4(&svm->vcpu, hsave->save.cr4);
+	if (npt_enabled) {
+		svm->vmcb->save.cr3 = hsave->save.cr3;
+		svm->vcpu.arch.cr3 = hsave->save.cr3;
+	} else {
+		kvm_set_cr3(&svm->vcpu, hsave->save.cr3);
+	}
+	kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, hsave->save.rax);
+	kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, hsave->save.rsp);
+	kvm_register_write(&svm->vcpu, VCPU_REGS_RIP, hsave->save.rip);
+	svm->vmcb->save.dr7 = 0;
+	svm->vmcb->save.cpl = 0;
+	svm->vmcb->control.exit_int_info = 0;
+
+	svm->vcpu.arch.hflags &= ~HF_GIF_MASK;
+	/* Exit nested SVM mode */
+	svm->nested_vmcb = 0;
+
+	return 0;
+}
+
+static int nested_svm_vmexit(struct vcpu_svm *svm)
+{
+	nsvm_printk("VMexit\n");
+	if (nested_svm_do(svm, svm->nested_vmcb, 0,
+			  NULL, nested_svm_vmexit_real))
+		return 1;
+
+	kvm_mmu_reset_context(&svm->vcpu);
+	kvm_mmu_load(&svm->vcpu);
+
+	return 0;
+}
+
+static int nested_svm_vmrun_msrpm(struct vcpu_svm *svm, void *arg1,
+				  void *arg2, void *opaque)
+{
+	int i;
+	u32 *nested_msrpm = (u32*)arg1;
+	for (i=0; i< PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER) / 4; i++)
+		svm->nested_msrpm[i] = svm->msrpm[i] | nested_msrpm[i];
+	svm->vmcb->control.msrpm_base_pa = __pa(svm->nested_msrpm);
+
+	return 0;
+}
+
+static int nested_svm_vmrun(struct vcpu_svm *svm, void *arg1,
+			    void *arg2, void *opaque)
+{
+	struct vmcb *nested_vmcb = (struct vmcb *)arg1;
+	struct vmcb *hsave = svm->hsave;
+
+	/* nested_vmcb is our indicator if nested SVM is activated */
+	svm->nested_vmcb = svm->vmcb->save.rax;
+
+	/* Clear internal status */
+	svm->vcpu.arch.exception.pending = false;
+
+	/* Save the old vmcb, so we don't need to pick what we save, but
+	   can restore everything when a VMEXIT occurs */
+	memcpy(hsave, svm->vmcb, sizeof(struct vmcb));
+	/* We need to remember the original CR3 in the SPT case */
+	if (!npt_enabled)
+		hsave->save.cr3 = svm->vcpu.arch.cr3;
+	hsave->save.cr4 = svm->vcpu.arch.cr4;
+	hsave->save.rip = svm->next_rip;
+
+	if (svm->vmcb->save.rflags & X86_EFLAGS_IF)
+		svm->vcpu.arch.hflags |= HF_HIF_MASK;
+	else
+		svm->vcpu.arch.hflags &= ~HF_HIF_MASK;
+
+	/* Load the nested guest state */
+	svm->vmcb->save.es = nested_vmcb->save.es;
+	svm->vmcb->save.cs = nested_vmcb->save.cs;
+	svm->vmcb->save.ss = nested_vmcb->save.ss;
+	svm->vmcb->save.ds = nested_vmcb->save.ds;
+	svm->vmcb->save.gdtr = nested_vmcb->save.gdtr;
+	svm->vmcb->save.idtr = nested_vmcb->save.idtr;
+	svm->vmcb->save.rflags = nested_vmcb->save.rflags;
+	svm_set_efer(&svm->vcpu, nested_vmcb->save.efer);
+	svm_set_cr0(&svm->vcpu, nested_vmcb->save.cr0);
+	svm_set_cr4(&svm->vcpu, nested_vmcb->save.cr4);
+	if (npt_enabled) {
+		svm->vmcb->save.cr3 = nested_vmcb->save.cr3;
+		svm->vcpu.arch.cr3 = nested_vmcb->save.cr3;
+	} else {
+		kvm_set_cr3(&svm->vcpu, nested_vmcb->save.cr3);
+		kvm_mmu_reset_context(&svm->vcpu);
+	}
+	svm->vmcb->save.cr2 = nested_vmcb->save.cr2;
+	kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, nested_vmcb->save.rax);
+	kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, nested_vmcb->save.rsp);
+	kvm_register_write(&svm->vcpu, VCPU_REGS_RIP, nested_vmcb->save.rip);
+	/* In case we don't even reach vcpu_run, the fields are not updated */
+	svm->vmcb->save.rax = nested_vmcb->save.rax;
+	svm->vmcb->save.rsp = nested_vmcb->save.rsp;
+	svm->vmcb->save.rip = nested_vmcb->save.rip;
+	svm->vmcb->save.dr7 = nested_vmcb->save.dr7;
+	svm->vmcb->save.dr6 = nested_vmcb->save.dr6;
+	svm->vmcb->save.cpl = nested_vmcb->save.cpl;
+
+	/* We don't want a nested guest to be more powerful than the guest,
+	   so all intercepts are ORed */
+	svm->vmcb->control.intercept_cr_read |=
+		nested_vmcb->control.intercept_cr_read;
+	svm->vmcb->control.intercept_cr_write |=
+		nested_vmcb->control.intercept_cr_write;
+	svm->vmcb->control.intercept_dr_read |=
+		nested_vmcb->control.intercept_dr_read;
+	svm->vmcb->control.intercept_dr_write |=
+		nested_vmcb->control.intercept_dr_write;
+	svm->vmcb->control.intercept_exceptions |=
+		nested_vmcb->control.intercept_exceptions;
+
+	svm->vmcb->control.intercept |= nested_vmcb->control.intercept;
+
+	svm->nested_vmcb_msrpm = nested_vmcb->control.msrpm_base_pa;
+
+	force_new_asid(&svm->vcpu);
+	svm->vmcb->control.exit_int_info = nested_vmcb->control.exit_int_info;
+	svm->vmcb->control.exit_int_info_err = nested_vmcb->control.exit_int_info_err;
+	svm->vmcb->control.int_ctl = nested_vmcb->control.int_ctl | V_INTR_MASKING_MASK;
+	if (nested_vmcb->control.int_ctl & V_IRQ_MASK) {
+		nsvm_printk("nSVM Injecting Interrupt: 0x%x\n",
+				nested_vmcb->control.int_ctl);
+	}
+	if (nested_vmcb->control.int_ctl & V_INTR_MASKING_MASK)
+		svm->vcpu.arch.hflags |= HF_VINTR_MASK;
+	else
+		svm->vcpu.arch.hflags &= ~HF_VINTR_MASK;
+
+	nsvm_printk("nSVM exit_int_info: 0x%x | int_state: 0x%x\n",
+			nested_vmcb->control.exit_int_info,
+			nested_vmcb->control.int_state);
+
+	svm->vmcb->control.int_vector = nested_vmcb->control.int_vector;
+	svm->vmcb->control.int_state = nested_vmcb->control.int_state;
+	svm->vmcb->control.tsc_offset += nested_vmcb->control.tsc_offset;
+	if (nested_vmcb->control.event_inj & SVM_EVTINJ_VALID)
+		nsvm_printk("Injecting Event: 0x%x\n",
+				nested_vmcb->control.event_inj);
+	svm->vmcb->control.event_inj = nested_vmcb->control.event_inj;
+	svm->vmcb->control.event_inj_err = nested_vmcb->control.event_inj_err;
+
+	svm->vcpu.arch.hflags |= HF_GIF_MASK;
+
+	return 0;
+}
+
+static int nested_svm_vmloadsave(struct vmcb *from_vmcb, struct vmcb *to_vmcb)
+{
+	to_vmcb->save.fs = from_vmcb->save.fs;
+	to_vmcb->save.gs = from_vmcb->save.gs;
+	to_vmcb->save.tr = from_vmcb->save.tr;
+	to_vmcb->save.ldtr = from_vmcb->save.ldtr;
+	to_vmcb->save.kernel_gs_base = from_vmcb->save.kernel_gs_base;
+	to_vmcb->save.star = from_vmcb->save.star;
+	to_vmcb->save.lstar = from_vmcb->save.lstar;
+	to_vmcb->save.cstar = from_vmcb->save.cstar;
+	to_vmcb->save.sfmask = from_vmcb->save.sfmask;
+	to_vmcb->save.sysenter_cs = from_vmcb->save.sysenter_cs;
+	to_vmcb->save.sysenter_esp = from_vmcb->save.sysenter_esp;
+	to_vmcb->save.sysenter_eip = from_vmcb->save.sysenter_eip;
+
+	return 1;
+}
+
+static int nested_svm_vmload(struct vcpu_svm *svm, void *nested_vmcb,
+			     void *arg2, void *opaque)
+{
+	return nested_svm_vmloadsave((struct vmcb *)nested_vmcb, svm->vmcb);
+}
+
+static int nested_svm_vmsave(struct vcpu_svm *svm, void *nested_vmcb,
+			     void *arg2, void *opaque)
+{
+	return nested_svm_vmloadsave(svm->vmcb, (struct vmcb *)nested_vmcb);
+}
+
+static int vmload_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (nested_svm_check_permissions(svm))
+		return 1;
+
+	svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
+	skip_emulated_instruction(&svm->vcpu);
+
+	nested_svm_do(svm, svm->vmcb->save.rax, 0, NULL, nested_svm_vmload);
+
+	return 1;
+}
+
+static int vmsave_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (nested_svm_check_permissions(svm))
+		return 1;
+
+	svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
+	skip_emulated_instruction(&svm->vcpu);
+
+	nested_svm_do(svm, svm->vmcb->save.rax, 0, NULL, nested_svm_vmsave);
+
+	return 1;
+}
+
+static int vmrun_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	nsvm_printk("VMrun\n");
+	if (nested_svm_check_permissions(svm))
+		return 1;
+
+	svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
+	skip_emulated_instruction(&svm->vcpu);
+
+	if (nested_svm_do(svm, svm->vmcb->save.rax, 0,
+			  NULL, nested_svm_vmrun))
+		return 1;
+
+	if (nested_svm_do(svm, svm->nested_vmcb_msrpm, 0,
+		      NULL, nested_svm_vmrun_msrpm))
+		return 1;
+
+	return 1;
+}
+
+static int stgi_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (nested_svm_check_permissions(svm))
+		return 1;
+
+	svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
+	skip_emulated_instruction(&svm->vcpu);
+
+	svm->vcpu.arch.hflags |= HF_GIF_MASK;
+
+	return 1;
+}
+
+static int clgi_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (nested_svm_check_permissions(svm))
+		return 1;
+
+	svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
+	skip_emulated_instruction(&svm->vcpu);
+
+	svm->vcpu.arch.hflags &= ~HF_GIF_MASK;
+
+	/* After a CLGI no interrupts should come */
+	svm_clear_vintr(svm);
+	svm->vmcb->control.int_ctl &= ~V_IRQ_MASK;
+
+	return 1;
+}
+
 static int invalid_op_interception(struct vcpu_svm *svm,
 				   struct kvm_run *kvm_run)
 {
@@ -1250,6 +1940,15 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data)
 	case MSR_IA32_LASTINTTOIP:
 		*data = svm->vmcb->save.last_excp_to;
 		break;
+	case MSR_VM_HSAVE_PA:
+		*data = svm->hsave_msr;
+		break;
+	case MSR_VM_CR:
+		*data = 0;
+		break;
+	case MSR_IA32_UCODE_REV:
+		*data = 0x01000065;
+		break;
 	default:
 		return kvm_get_msr_common(vcpu, ecx, data);
 	}
@@ -1343,6 +2042,9 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data)
 		 */
 		pr_unimpl(vcpu, "unimplemented perfctr wrmsr: 0x%x data 0x%llx\n", ecx, data);
 
+		break;
+	case MSR_VM_HSAVE_PA:
+		svm->hsave_msr = data;
 		break;
 	default:
 		return kvm_set_msr_common(vcpu, ecx, data);
@@ -1380,7 +2082,7 @@ static int interrupt_window_interception(struct vcpu_svm *svm,
 {
 	KVMTRACE_0D(PEND_INTR, &svm->vcpu, handler);
 
-	svm->vmcb->control.intercept &= ~(1ULL << INTERCEPT_VINTR);
+	svm_clear_vintr(svm);
 	svm->vmcb->control.int_ctl &= ~V_IRQ_MASK;
 	/*
 	 * If the user space waits to inject interrupts, exit as soon as
@@ -1417,6 +2119,8 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm,
 	[SVM_EXIT_WRITE_DR3]			= emulate_on_interception,
 	[SVM_EXIT_WRITE_DR5]			= emulate_on_interception,
 	[SVM_EXIT_WRITE_DR7]			= emulate_on_interception,
+	[SVM_EXIT_EXCP_BASE + DB_VECTOR]	= db_interception,
+	[SVM_EXIT_EXCP_BASE + BP_VECTOR]	= bp_interception,
 	[SVM_EXIT_EXCP_BASE + UD_VECTOR]	= ud_interception,
 	[SVM_EXIT_EXCP_BASE + PF_VECTOR] 	= pf_interception,
 	[SVM_EXIT_EXCP_BASE + NM_VECTOR] 	= nm_interception,
@@ -1436,12 +2140,12 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm,
 	[SVM_EXIT_MSR]				= msr_interception,
 	[SVM_EXIT_TASK_SWITCH]			= task_switch_interception,
 	[SVM_EXIT_SHUTDOWN]			= shutdown_interception,
-	[SVM_EXIT_VMRUN]			= invalid_op_interception,
+	[SVM_EXIT_VMRUN]			= vmrun_interception,
 	[SVM_EXIT_VMMCALL]			= vmmcall_interception,
-	[SVM_EXIT_VMLOAD]			= invalid_op_interception,
-	[SVM_EXIT_VMSAVE]			= invalid_op_interception,
-	[SVM_EXIT_STGI]				= invalid_op_interception,
-	[SVM_EXIT_CLGI]				= invalid_op_interception,
+	[SVM_EXIT_VMLOAD]			= vmload_interception,
+	[SVM_EXIT_VMSAVE]			= vmsave_interception,
+	[SVM_EXIT_STGI]				= stgi_interception,
+	[SVM_EXIT_CLGI]				= clgi_interception,
 	[SVM_EXIT_SKINIT]			= invalid_op_interception,
 	[SVM_EXIT_WBINVD]                       = emulate_on_interception,
 	[SVM_EXIT_MONITOR]			= invalid_op_interception,
@@ -1457,6 +2161,17 @@ static int handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 	KVMTRACE_3D(VMEXIT, vcpu, exit_code, (u32)svm->vmcb->save.rip,
 		    (u32)((u64)svm->vmcb->save.rip >> 32), entryexit);
 
+	if (is_nested(svm)) {
+		nsvm_printk("nested handle_exit: 0x%x | 0x%lx | 0x%lx | 0x%lx\n",
+			    exit_code, svm->vmcb->control.exit_info_1,
+			    svm->vmcb->control.exit_info_2, svm->vmcb->save.rip);
+		if (nested_svm_exit_handled(svm, true)) {
+			nested_svm_vmexit(svm);
+			nsvm_printk("-> #VMEXIT\n");
+			return 1;
+		}
+	}
+
 	if (npt_enabled) {
 		int mmu_reload = 0;
 		if ((vcpu->arch.cr0 ^ svm->vmcb->save.cr0) & X86_CR0_PG) {
@@ -1544,6 +2259,8 @@ static void svm_set_irq(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
 
+	nested_svm_intr(svm);
+
 	svm_inject_irq(svm, irq);
 }
 
@@ -1589,11 +2306,17 @@ static void svm_intr_assist(struct kvm_vcpu *vcpu)
 	if (!kvm_cpu_has_interrupt(vcpu))
 		goto out;
 
+	if (nested_svm_intr(svm))
+		goto out;
+
+	if (!(svm->vcpu.arch.hflags & HF_GIF_MASK))
+		goto out;
+
 	if (!(vmcb->save.rflags & X86_EFLAGS_IF) ||
 	    (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) ||
 	    (vmcb->control.event_inj & SVM_EVTINJ_VALID)) {
 		/* unable to deliver irq, set pending irq */
-		vmcb->control.intercept |= (1ULL << INTERCEPT_VINTR);
+		svm_set_vintr(svm);
 		svm_inject_irq(svm, 0x0);
 		goto out;
 	}
@@ -1615,7 +2338,8 @@ static void kvm_reput_irq(struct vcpu_svm *svm)
 	}
 
 	svm->vcpu.arch.interrupt_window_open =
-		!(control->int_state & SVM_INTERRUPT_SHADOW_MASK);
+		!(control->int_state & SVM_INTERRUPT_SHADOW_MASK) &&
+		 (svm->vcpu.arch.hflags & HF_GIF_MASK);
 }
 
 static void svm_do_inject_vector(struct vcpu_svm *svm)
@@ -1637,9 +2361,13 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 	struct vcpu_svm *svm = to_svm(vcpu);
 	struct vmcb_control_area *control = &svm->vmcb->control;
 
+	if (nested_svm_intr(svm))
+		return;
+
 	svm->vcpu.arch.interrupt_window_open =
 		(!(control->int_state & SVM_INTERRUPT_SHADOW_MASK) &&
-		 (svm->vmcb->save.rflags & X86_EFLAGS_IF));
+		 (svm->vmcb->save.rflags & X86_EFLAGS_IF) &&
+		 (svm->vcpu.arch.hflags & HF_GIF_MASK));
 
 	if (svm->vcpu.arch.interrupt_window_open && svm->vcpu.arch.irq_summary)
 		/*
@@ -1652,9 +2380,9 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 	 */
 	if (!svm->vcpu.arch.interrupt_window_open &&
 	    (svm->vcpu.arch.irq_summary || kvm_run->request_interrupt_window))
-		control->intercept |= 1ULL << INTERCEPT_VINTR;
-	 else
-		control->intercept &= ~(1ULL << INTERCEPT_VINTR);
+		svm_set_vintr(svm);
+	else
+		svm_clear_vintr(svm);
 }
 
 static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr)
@@ -1662,22 +2390,6 @@ static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr)
 	return 0;
 }
 
-static void save_db_regs(unsigned long *db_regs)
-{
-	asm volatile ("mov %%dr0, %0" : "=r"(db_regs[0]));
-	asm volatile ("mov %%dr1, %0" : "=r"(db_regs[1]));
-	asm volatile ("mov %%dr2, %0" : "=r"(db_regs[2]));
-	asm volatile ("mov %%dr3, %0" : "=r"(db_regs[3]));
-}
-
-static void load_db_regs(unsigned long *db_regs)
-{
-	asm volatile ("mov %0, %%dr0" : : "r"(db_regs[0]));
-	asm volatile ("mov %0, %%dr1" : : "r"(db_regs[1]));
-	asm volatile ("mov %0, %%dr2" : : "r"(db_regs[2]));
-	asm volatile ("mov %0, %%dr3" : : "r"(db_regs[3]));
-}
-
 static void svm_flush_tlb(struct kvm_vcpu *vcpu)
 {
 	force_new_asid(vcpu);
@@ -1736,19 +2448,12 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	gs_selector = kvm_read_gs();
 	ldt_selector = kvm_read_ldt();
 	svm->host_cr2 = kvm_read_cr2();
-	svm->host_dr6 = read_dr6();
-	svm->host_dr7 = read_dr7();
-	svm->vmcb->save.cr2 = vcpu->arch.cr2;
+	if (!is_nested(svm))
+		svm->vmcb->save.cr2 = vcpu->arch.cr2;
 	/* required for live migration with NPT */
 	if (npt_enabled)
 		svm->vmcb->save.cr3 = vcpu->arch.cr3;
 
-	if (svm->vmcb->save.dr7 & 0xff) {
-		write_dr7(0);
-		save_db_regs(svm->host_db_regs);
-		load_db_regs(svm->db_regs);
-	}
-
 	clgi();
 
 	local_irq_enable();
@@ -1824,16 +2529,11 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 #endif
 		);
 
-	if ((svm->vmcb->save.dr7 & 0xff))
-		load_db_regs(svm->host_db_regs);
-
 	vcpu->arch.cr2 = svm->vmcb->save.cr2;
 	vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
 	vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
 	vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip;
 
-	write_dr6(svm->host_dr6);
-	write_dr7(svm->host_dr7);
 	kvm_write_cr2(svm->host_cr2);
 
 	kvm_load_fs(fs_selector);

arch/x86/kvm/vmx.c

@@ -91,6 +91,7 @@ struct vcpu_vmx {
 	} rmode;
 	int vpid;
 	bool emulation_required;
+	enum emulation_result invalid_state_emulation_result;
 
 	/* Support for vnmi-less CPUs */
 	int soft_vnmi_blocked;
@@ -189,21 +190,21 @@ static inline int is_page_fault(u32 intr_info)
 {
 	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
 			     INTR_INFO_VALID_MASK)) ==
-		(INTR_TYPE_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK);
+		(INTR_TYPE_HARD_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK);
 }
 
 static inline int is_no_device(u32 intr_info)
 {
 	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
 			     INTR_INFO_VALID_MASK)) ==
-		(INTR_TYPE_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK);
+		(INTR_TYPE_HARD_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK);
 }
 
 static inline int is_invalid_opcode(u32 intr_info)
 {
 	return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK |
 			     INTR_INFO_VALID_MASK)) ==
-		(INTR_TYPE_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK);
+		(INTR_TYPE_HARD_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK);
 }
 
 static inline int is_external_interrupt(u32 intr_info)
@@ -480,8 +481,13 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu)
 	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR);
 	if (!vcpu->fpu_active)
 		eb |= 1u << NM_VECTOR;
-	if (vcpu->guest_debug.enabled)
-		eb |= 1u << DB_VECTOR;
+	if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) {
+		if (vcpu->guest_debug &
+		    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+			eb |= 1u << DB_VECTOR;
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			eb |= 1u << BP_VECTOR;
+	}
 	if (vcpu->arch.rmode.active)
 		eb = ~0;
 	if (vm_need_ept())
@@ -747,29 +753,33 @@ static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
 				bool has_error_code, u32 error_code)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	u32 intr_info = nr | INTR_INFO_VALID_MASK;
 
-	if (has_error_code)
+	if (has_error_code) {
 		vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code);
+		intr_info |= INTR_INFO_DELIVER_CODE_MASK;
+	}
 
 	if (vcpu->arch.rmode.active) {
 		vmx->rmode.irq.pending = true;
 		vmx->rmode.irq.vector = nr;
 		vmx->rmode.irq.rip = kvm_rip_read(vcpu);
-		if (nr == BP_VECTOR)
+		if (nr == BP_VECTOR || nr == OF_VECTOR)
 			vmx->rmode.irq.rip++;
-		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
-			     nr | INTR_TYPE_SOFT_INTR
-			     | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0)
-			     | INTR_INFO_VALID_MASK);
+		intr_info |= INTR_TYPE_SOFT_INTR;
+		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, intr_info);
 		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1);
 		kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1);
 		return;
 	}
 
-	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
-		     nr | INTR_TYPE_EXCEPTION
-		     | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0)
-		     | INTR_INFO_VALID_MASK);
+	if (nr == BP_VECTOR || nr == OF_VECTOR) {
+		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1);
+		intr_info |= INTR_TYPE_SOFT_EXCEPTION;
+	} else
+		intr_info |= INTR_TYPE_HARD_EXCEPTION;
+
+	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, intr_info);
 }
 
 static bool vmx_exception_injected(struct kvm_vcpu *vcpu)
@@ -856,11 +866,8 @@ static u64 guest_read_tsc(void)
  * writes 'guest_tsc' into guest's timestamp counter "register"
  * guest_tsc = host_tsc + tsc_offset ==> tsc_offset = guest_tsc - host_tsc
  */
-static void guest_write_tsc(u64 guest_tsc)
+static void guest_write_tsc(u64 guest_tsc, u64 host_tsc)
 {
-	u64 host_tsc;
-
-	rdtscll(host_tsc);
 	vmcs_write64(TSC_OFFSET, guest_tsc - host_tsc);
 }
 
@@ -925,14 +932,15 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	struct kvm_msr_entry *msr;
+	u64 host_tsc;
 	int ret = 0;
 
 	switch (msr_index) {
-#ifdef CONFIG_X86_64
 	case MSR_EFER:
 		vmx_load_host_state(vmx);
 		ret = kvm_set_msr_common(vcpu, msr_index, data);
 		break;
+#ifdef CONFIG_X86_64
 	case MSR_FS_BASE:
 		vmcs_writel(GUEST_FS_BASE, data);
 		break;
@@ -950,7 +958,8 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
 		vmcs_writel(GUEST_SYSENTER_ESP, data);
 		break;
 	case MSR_IA32_TIME_STAMP_COUNTER:
-		guest_write_tsc(data);
+		rdtscll(host_tsc);
+		guest_write_tsc(data, host_tsc);
 		break;
 	case MSR_P6_PERFCTR0:
 	case MSR_P6_PERFCTR1:
@@ -999,40 +1008,28 @@ static void vmx_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg)
 	}
 }
 
-static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
+static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg)
 {
-	unsigned long dr7 = 0x400;
-	int old_singlestep;
-
-	old_singlestep = vcpu->guest_debug.singlestep;
-
-	vcpu->guest_debug.enabled = dbg->enabled;
-	if (vcpu->guest_debug.enabled) {
-		int i;
+	int old_debug = vcpu->guest_debug;
+	unsigned long flags;
 
-		dr7 |= 0x200;  /* exact */
-		for (i = 0; i < 4; ++i) {
-			if (!dbg->breakpoints[i].enabled)
-				continue;
-			vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address;
-			dr7 |= 2 << (i*2);    /* global enable */
-			dr7 |= 0 << (i*4+16); /* execution breakpoint */
-		}
+	vcpu->guest_debug = dbg->control;
+	if (!(vcpu->guest_debug & KVM_GUESTDBG_ENABLE))
+		vcpu->guest_debug = 0;
 
-		vcpu->guest_debug.singlestep = dbg->singlestep;
-	} else
-		vcpu->guest_debug.singlestep = 0;
-
-	if (old_singlestep && !vcpu->guest_debug.singlestep) {
-		unsigned long flags;
+	if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)
+		vmcs_writel(GUEST_DR7, dbg->arch.debugreg[7]);
+	else
+		vmcs_writel(GUEST_DR7, vcpu->arch.dr7);
 
-		flags = vmcs_readl(GUEST_RFLAGS);
+	flags = vmcs_readl(GUEST_RFLAGS);
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		flags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
+	else if (old_debug & KVM_GUESTDBG_SINGLESTEP)
 		flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
-		vmcs_writel(GUEST_RFLAGS, flags);
-	}
+	vmcs_writel(GUEST_RFLAGS, flags);
 
 	update_exception_bitmap(vcpu);
-	vmcs_writel(GUEST_DR7, dr7);
 
 	return 0;
 }
@@ -1433,6 +1430,29 @@ continue_rmode:
 	init_rmode(vcpu->kvm);
 }
 
+static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER);
+
+	vcpu->arch.shadow_efer = efer;
+	if (!msr)
+		return;
+	if (efer & EFER_LMA) {
+		vmcs_write32(VM_ENTRY_CONTROLS,
+			     vmcs_read32(VM_ENTRY_CONTROLS) |
+			     VM_ENTRY_IA32E_MODE);
+		msr->data = efer;
+	} else {
+		vmcs_write32(VM_ENTRY_CONTROLS,
+			     vmcs_read32(VM_ENTRY_CONTROLS) &
+			     ~VM_ENTRY_IA32E_MODE);
+
+		msr->data = efer & ~EFER_LME;
+	}
+	setup_msrs(vmx);
+}
+
 #ifdef CONFIG_X86_64
 
 static void enter_lmode(struct kvm_vcpu *vcpu)
@@ -1447,13 +1467,8 @@ static void enter_lmode(struct kvm_vcpu *vcpu)
 			     (guest_tr_ar & ~AR_TYPE_MASK)
 			     | AR_TYPE_BUSY_64_TSS);
 	}
-
 	vcpu->arch.shadow_efer |= EFER_LMA;
-
-	find_msr_entry(to_vmx(vcpu), MSR_EFER)->data |= EFER_LMA | EFER_LME;
-	vmcs_write32(VM_ENTRY_CONTROLS,
-		     vmcs_read32(VM_ENTRY_CONTROLS)
-		     | VM_ENTRY_IA32E_MODE);
+	vmx_set_efer(vcpu, vcpu->arch.shadow_efer);
 }
 
 static void exit_lmode(struct kvm_vcpu *vcpu)
@@ -1612,30 +1627,6 @@ static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
 	vmcs_writel(GUEST_CR4, hw_cr4);
 }
 
-static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer)
-{
-	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER);
-
-	vcpu->arch.shadow_efer = efer;
-	if (!msr)
-		return;
-	if (efer & EFER_LMA) {
-		vmcs_write32(VM_ENTRY_CONTROLS,
-				     vmcs_read32(VM_ENTRY_CONTROLS) |
-				     VM_ENTRY_IA32E_MODE);
-		msr->data = efer;
-
-	} else {
-		vmcs_write32(VM_ENTRY_CONTROLS,
-				     vmcs_read32(VM_ENTRY_CONTROLS) &
-				     ~VM_ENTRY_IA32E_MODE);
-
-		msr->data = efer & ~EFER_LME;
-	}
-	setup_msrs(vmx);
-}
-
 static u64 vmx_get_segment_base(struct kvm_vcpu *vcpu, int seg)
 {
 	struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
@@ -1653,7 +1644,7 @@ static void vmx_get_segment(struct kvm_vcpu *vcpu,
 	var->limit = vmcs_read32(sf->limit);
 	var->selector = vmcs_read16(sf->selector);
 	ar = vmcs_read32(sf->ar_bytes);
-	if (ar & AR_UNUSABLE_MASK)
+	if ((ar & AR_UNUSABLE_MASK) && !emulate_invalid_guest_state)
 		ar = 0;
 	var->type = ar & 15;
 	var->s = (ar >> 4) & 1;
@@ -1788,14 +1779,16 @@ static bool code_segment_valid(struct kvm_vcpu *vcpu)
 	vmx_get_segment(vcpu, &cs, VCPU_SREG_CS);
 	cs_rpl = cs.selector & SELECTOR_RPL_MASK;
 
+	if (cs.unusable)
+		return false;
 	if (~cs.type & (AR_TYPE_CODE_MASK|AR_TYPE_ACCESSES_MASK))
 		return false;
 	if (!cs.s)
 		return false;
-	if (!(~cs.type & (AR_TYPE_CODE_MASK|AR_TYPE_WRITEABLE_MASK))) {
+	if (cs.type & AR_TYPE_WRITEABLE_MASK) {
 		if (cs.dpl > cs_rpl)
 			return false;
-	} else if (cs.type & AR_TYPE_CODE_MASK) {
+	} else {
 		if (cs.dpl != cs_rpl)
 			return false;
 	}
@@ -1814,7 +1807,9 @@ static bool stack_segment_valid(struct kvm_vcpu *vcpu)
 	vmx_get_segment(vcpu, &ss, VCPU_SREG_SS);
 	ss_rpl = ss.selector & SELECTOR_RPL_MASK;
 
-	if ((ss.type != 3) || (ss.type != 7))
+	if (ss.unusable)
+		return true;
+	if (ss.type != 3 && ss.type != 7)
 		return false;
 	if (!ss.s)
 		return false;
@@ -1834,6 +1829,8 @@ static bool data_segment_valid(struct kvm_vcpu *vcpu, int seg)
 	vmx_get_segment(vcpu, &var, seg);
 	rpl = var.selector & SELECTOR_RPL_MASK;
 
+	if (var.unusable)
+		return true;
 	if (!var.s)
 		return false;
 	if (!var.present)
@@ -1855,9 +1852,11 @@ static bool tr_valid(struct kvm_vcpu *vcpu)
 
 	vmx_get_segment(vcpu, &tr, VCPU_SREG_TR);
 
+	if (tr.unusable)
+		return false;
 	if (tr.selector & SELECTOR_TI_MASK)	/* TI = 1 */
 		return false;
-	if ((tr.type != 3) || (tr.type != 11)) /* TODO: Check if guest is in IA32e mode */
+	if (tr.type != 3 && tr.type != 11) /* TODO: Check if guest is in IA32e mode */
 		return false;
 	if (!tr.present)
 		return false;
@@ -1871,6 +1870,8 @@ static bool ldtr_valid(struct kvm_vcpu *vcpu)
 
 	vmx_get_segment(vcpu, &ldtr, VCPU_SREG_LDTR);
 
+	if (ldtr.unusable)
+		return true;
 	if (ldtr.selector & SELECTOR_TI_MASK)	/* TI = 1 */
 		return false;
 	if (ldtr.type != 2)
@@ -2112,7 +2113,7 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
 {
 	u32 host_sysenter_cs, msr_low, msr_high;
 	u32 junk;
-	u64 host_pat;
+	u64 host_pat, tsc_this, tsc_base;
 	unsigned long a;
 	struct descriptor_table dt;
 	int i;
@@ -2240,6 +2241,12 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
 	vmcs_writel(CR0_GUEST_HOST_MASK, ~0UL);
 	vmcs_writel(CR4_GUEST_HOST_MASK, KVM_GUEST_CR4_MASK);
 
+	tsc_base = vmx->vcpu.kvm->arch.vm_init_tsc;
+	rdtscll(tsc_this);
+	if (tsc_this < vmx->vcpu.kvm->arch.vm_init_tsc)
+		tsc_base = tsc_this;
+
+	guest_write_tsc(0, tsc_base);
 
 	return 0;
 }
@@ -2319,7 +2326,6 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
 		kvm_rip_write(vcpu, 0);
 	kvm_register_write(vcpu, VCPU_REGS_RSP, 0);
 
-	/* todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 */
 	vmcs_writel(GUEST_DR7, 0x400);
 
 	vmcs_writel(GUEST_GDTR_BASE, 0);
@@ -2332,8 +2338,6 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
 	vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0);
 	vmcs_write32(GUEST_PENDING_DBG_EXCEPTIONS, 0);
 
-	guest_write_tsc(0);
-
 	/* Special registers */
 	vmcs_write64(GUEST_IA32_DEBUGCTL, 0);
 
@@ -2486,6 +2490,11 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 {
 	vmx_update_window_states(vcpu);
 
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO,
+				GUEST_INTR_STATE_STI |
+				GUEST_INTR_STATE_MOV_SS);
+
 	if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) {
 		if (vcpu->arch.interrupt.pending) {
 			enable_nmi_window(vcpu);
@@ -2536,24 +2545,6 @@ static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
 	return 0;
 }
 
-static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu)
-{
-	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
-
-	set_debugreg(dbg->bp[0], 0);
-	set_debugreg(dbg->bp[1], 1);
-	set_debugreg(dbg->bp[2], 2);
-	set_debugreg(dbg->bp[3], 3);
-
-	if (dbg->singlestep) {
-		unsigned long flags;
-
-		flags = vmcs_readl(GUEST_RFLAGS);
-		flags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
-		vmcs_writel(GUEST_RFLAGS, flags);
-	}
-}
-
 static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 				  int vec, u32 err_code)
 {
@@ -2570,9 +2561,17 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 	 *        the required debugging infrastructure rework.
 	 */
 	switch (vec) {
-	case DE_VECTOR:
 	case DB_VECTOR:
+		if (vcpu->guest_debug &
+		    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+			return 0;
+		kvm_queue_exception(vcpu, vec);
+		return 1;
 	case BP_VECTOR:
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			return 0;
+		/* fall through */
+	case DE_VECTOR:
 	case OF_VECTOR:
 	case BR_VECTOR:
 	case UD_VECTOR:
@@ -2589,8 +2588,8 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 intr_info, error_code;
-	unsigned long cr2, rip;
+	u32 intr_info, ex_no, error_code;
+	unsigned long cr2, rip, dr6;
 	u32 vect_info;
 	enum emulation_result er;
 
@@ -2649,14 +2648,30 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		return 1;
 	}
 
-	if ((intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK)) ==
-	    (INTR_TYPE_EXCEPTION | 1)) {
+	ex_no = intr_info & INTR_INFO_VECTOR_MASK;
+	switch (ex_no) {
+	case DB_VECTOR:
+		dr6 = vmcs_readl(EXIT_QUALIFICATION);
+		if (!(vcpu->guest_debug &
+		      (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))) {
+			vcpu->arch.dr6 = dr6 | DR6_FIXED_1;
+			kvm_queue_exception(vcpu, DB_VECTOR);
+			return 1;
+		}
+		kvm_run->debug.arch.dr6 = dr6 | DR6_FIXED_1;
+		kvm_run->debug.arch.dr7 = vmcs_readl(GUEST_DR7);
+		/* fall through */
+	case BP_VECTOR:
 		kvm_run->exit_reason = KVM_EXIT_DEBUG;
-		return 0;
+		kvm_run->debug.arch.pc = vmcs_readl(GUEST_CS_BASE) + rip;
+		kvm_run->debug.arch.exception = ex_no;
+		break;
+	default:
+		kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
+		kvm_run->ex.exception = ex_no;
+		kvm_run->ex.error_code = error_code;
+		break;
 	}
-	kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
-	kvm_run->ex.exception = intr_info & INTR_INFO_VECTOR_MASK;
-	kvm_run->ex.error_code = error_code;
 	return 0;
 }
 
@@ -2677,7 +2692,7 @@ static int handle_triple_fault(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	unsigned long exit_qualification;
-	int size, down, in, string, rep;
+	int size, in, string;
 	unsigned port;
 
 	++vcpu->stat.io_exits;
@@ -2693,8 +2708,6 @@ static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 
 	size = (exit_qualification & 7) + 1;
 	in = (exit_qualification & 8) != 0;
-	down = (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_DF) != 0;
-	rep = (exit_qualification & 32) != 0;
 	port = exit_qualification >> 16;
 
 	skip_emulated_instruction(vcpu);
@@ -2795,21 +2808,44 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	unsigned long val;
 	int dr, reg;
 
-	/*
-	 * FIXME: this code assumes the host is debugging the guest.
-	 *        need to deal with guest debugging itself too.
-	 */
+	dr = vmcs_readl(GUEST_DR7);
+	if (dr & DR7_GD) {
+		/*
+		 * As the vm-exit takes precedence over the debug trap, we
+		 * need to emulate the latter, either for the host or the
+		 * guest debugging itself.
+		 */
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) {
+			kvm_run->debug.arch.dr6 = vcpu->arch.dr6;
+			kvm_run->debug.arch.dr7 = dr;
+			kvm_run->debug.arch.pc =
+				vmcs_readl(GUEST_CS_BASE) +
+				vmcs_readl(GUEST_RIP);
+			kvm_run->debug.arch.exception = DB_VECTOR;
+			kvm_run->exit_reason = KVM_EXIT_DEBUG;
+			return 0;
+		} else {
+			vcpu->arch.dr7 &= ~DR7_GD;
+			vcpu->arch.dr6 |= DR6_BD;
+			vmcs_writel(GUEST_DR7, vcpu->arch.dr7);
+			kvm_queue_exception(vcpu, DB_VECTOR);
+			return 1;
+		}
+	}
+
 	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
-	dr = exit_qualification & 7;
-	reg = (exit_qualification >> 8) & 15;
-	if (exit_qualification & 16) {
-		/* mov from dr */
+	dr = exit_qualification & DEBUG_REG_ACCESS_NUM;
+	reg = DEBUG_REG_ACCESS_REG(exit_qualification);
+	if (exit_qualification & TYPE_MOV_FROM_DR) {
 		switch (dr) {
+		case 0 ... 3:
+			val = vcpu->arch.db[dr];
+			break;
 		case 6:
-			val = 0xffff0ff0;
+			val = vcpu->arch.dr6;
 			break;
 		case 7:
-			val = 0x400;
+			val = vcpu->arch.dr7;
 			break;
 		default:
 			val = 0;
@@ -2817,7 +2853,38 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		kvm_register_write(vcpu, reg, val);
 		KVMTRACE_2D(DR_READ, vcpu, (u32)dr, (u32)val, handler);
 	} else {
-		/* mov to dr */
+		val = vcpu->arch.regs[reg];
+		switch (dr) {
+		case 0 ... 3:
+			vcpu->arch.db[dr] = val;
+			if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP))
+				vcpu->arch.eff_db[dr] = val;
+			break;
+		case 4 ... 5:
+			if (vcpu->arch.cr4 & X86_CR4_DE)
+				kvm_queue_exception(vcpu, UD_VECTOR);
+			break;
+		case 6:
+			if (val & 0xffffffff00000000ULL) {
+				kvm_queue_exception(vcpu, GP_VECTOR);
+				break;
+			}
+			vcpu->arch.dr6 = (val & DR6_VOLATILE) | DR6_FIXED_1;
+			break;
+		case 7:
+			if (val & 0xffffffff00000000ULL) {
+				kvm_queue_exception(vcpu, GP_VECTOR);
+				break;
+			}
+			vcpu->arch.dr7 = (val & DR7_VOLATILE) | DR7_FIXED_1;
+			if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) {
+				vmcs_writel(GUEST_DR7, vcpu->arch.dr7);
+				vcpu->arch.switch_db_regs =
+					(val & DR7_BP_EN_MASK);
+			}
+			break;
+		}
+		KVMTRACE_2D(DR_WRITE, vcpu, (u32)dr, (u32)val, handler);
 	}
 	skip_emulated_instruction(vcpu);
 	return 1;
@@ -2968,17 +3035,25 @@ static int handle_task_switch(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	}
 	tss_selector = exit_qualification;
 
-	return kvm_task_switch(vcpu, tss_selector, reason);
+	if (!kvm_task_switch(vcpu, tss_selector, reason))
+		return 0;
+
+	/* clear all local breakpoint enable flags */
+	vmcs_writel(GUEST_DR7, vmcs_readl(GUEST_DR7) & ~55);
+
+	/*
+	 * TODO: What about debug traps on tss switch?
+	 *       Are we supposed to inject them and update dr6?
+	 */
+
+	return 1;
 }
 
 static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	u64 exit_qualification;
-	enum emulation_result er;
 	gpa_t gpa;
-	unsigned long hva;
 	int gla_validity;
-	int r;
 
 	exit_qualification = vmcs_read64(EXIT_QUALIFICATION);
 
@@ -3001,32 +3076,7 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	}
 
 	gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS);
-	hva = gfn_to_hva(vcpu->kvm, gpa >> PAGE_SHIFT);
-	if (!kvm_is_error_hva(hva)) {
-		r = kvm_mmu_page_fault(vcpu, gpa & PAGE_MASK, 0);
-		if (r < 0) {
-			printk(KERN_ERR "EPT: Not enough memory!\n");
-			return -ENOMEM;
-		}
-		return 1;
-	} else {
-		/* must be MMIO */
-		er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
-
-		if (er == EMULATE_FAIL) {
-			printk(KERN_ERR
-			 "EPT: Fail to handle EPT violation vmexit!er is %d\n",
-			 er);
-			printk(KERN_ERR "EPT: GPA: 0x%lx, GVA: 0x%lx\n",
-			 (long unsigned int)vmcs_read64(GUEST_PHYSICAL_ADDRESS),
-			 (long unsigned int)vmcs_read64(GUEST_LINEAR_ADDRESS));
-			printk(KERN_ERR "EPT: Exit qualification is 0x%lx\n",
-				(long unsigned int)exit_qualification);
-			return -ENOTSUPP;
-		} else if (er == EMULATE_DO_MMIO)
-			return 0;
-	}
-	return 1;
+	return kvm_mmu_page_fault(vcpu, gpa & PAGE_MASK, 0);
 }
 
 static int handle_nmi_window(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
@@ -3046,7 +3096,7 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu,
 				struct kvm_run *kvm_run)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	int err;
+	enum emulation_result err = EMULATE_DONE;
 
 	preempt_enable();
 	local_irq_enable();
@@ -3071,10 +3121,7 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu,
 	local_irq_disable();
 	preempt_disable();
 
-	/* Guest state should be valid now except if we need to
-	 * emulate an MMIO */
-	if (guest_state_valid(vcpu))
-		vmx->emulation_required = 0;
+	vmx->invalid_state_emulation_result = err;
 }
 
 /*
@@ -3123,8 +3170,11 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 
 	/* If we need to emulate an MMIO from handle_invalid_guest_state
 	 * we just return 0 */
-	if (vmx->emulation_required && emulate_invalid_guest_state)
-		return 0;
+	if (vmx->emulation_required && emulate_invalid_guest_state) {
+		if (guest_state_valid(vcpu))
+			vmx->emulation_required = 0;
+		return vmx->invalid_state_emulation_result != EMULATE_DO_MMIO;
+	}
 
 	/* Access CR3 don't cause VMExit in paging mode, so we need
 	 * to sync with guest real CR3. */
@@ -3238,7 +3288,8 @@ static void vmx_complete_interrupts(struct vcpu_vmx *vmx)
 			vmx->vcpu.arch.nmi_injected = false;
 	}
 	kvm_clear_exception_queue(&vmx->vcpu);
-	if (idtv_info_valid && type == INTR_TYPE_EXCEPTION) {
+	if (idtv_info_valid && (type == INTR_TYPE_HARD_EXCEPTION ||
+				type == INTR_TYPE_SOFT_EXCEPTION)) {
 		if (idt_vectoring_info & VECTORING_INFO_DELIVER_CODE_MASK) {
 			error = vmcs_read32(IDT_VECTORING_ERROR_CODE);
 			kvm_queue_exception_e(&vmx->vcpu, vector, error);
@@ -3259,6 +3310,11 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu)
 
 	vmx_update_window_states(vcpu);
 
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO,
+				GUEST_INTR_STATE_STI |
+				GUEST_INTR_STATE_MOV_SS);
+
 	if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) {
 		if (vcpu->arch.interrupt.pending) {
 			enable_nmi_window(vcpu);
@@ -3347,6 +3403,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	 */
 	vmcs_writel(HOST_CR0, read_cr0());
 
+	set_debugreg(vcpu->arch.dr6, 6);
+
 	asm(
 		/* Store host registers */
 		"push %%"R"dx; push %%"R"bp;"
@@ -3441,6 +3499,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP));
 	vcpu->arch.regs_dirty = 0;
 
+	get_debugreg(vcpu->arch.dr6, 6);
+
 	vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD);
 	if (vmx->rmode.irq.pending)
 		fixup_rmode_irq(vmx);
@@ -3595,7 +3655,6 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.vcpu_put = vmx_vcpu_put,
 
 	.set_guest_debug = set_guest_debug,
-	.guest_debug_pre = kvm_guest_debug_pre,
 	.get_msr = vmx_get_msr,
 	.set_msr = vmx_set_msr,
 	.get_segment_base = vmx_get_segment_base,

arch/x86/kvm/x86.c

@@ -36,6 +36,7 @@
 #include <linux/highmem.h>
 #include <linux/iommu.h>
 #include <linux/intel-iommu.h>
+#include <linux/cpufreq.h>
 
 #include <asm/uaccess.h>
 #include <asm/msr.h>
@@ -69,6 +70,8 @@ static u64 __read_mostly efer_reserved_bits = 0xfffffffffffffffeULL;
 
 static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
 				    struct kvm_cpuid_entry2 __user *entries);
+struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu,
+					      u32 function, u32 index);
 
 struct kvm_x86_ops *kvm_x86_ops;
 EXPORT_SYMBOL_GPL(kvm_x86_ops);
@@ -173,6 +176,7 @@ void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long addr,
 			   u32 error_code)
 {
 	++vcpu->stat.pf_guest;
+
 	if (vcpu->arch.exception.pending) {
 		if (vcpu->arch.exception.nr == PF_VECTOR) {
 			printk(KERN_DEBUG "kvm: inject_page_fault:"
@@ -361,6 +365,7 @@ void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
 	}
 	kvm_x86_ops->set_cr4(vcpu, cr4);
 	vcpu->arch.cr4 = cr4;
+	vcpu->arch.mmu.base_role.cr4_pge = (cr4 & X86_CR4_PGE) && !tdp_enabled;
 	kvm_mmu_sync_global(vcpu);
 	kvm_mmu_reset_context(vcpu);
 }
@@ -442,6 +447,11 @@ unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu)
 }
 EXPORT_SYMBOL_GPL(kvm_get_cr8);
 
+static inline u32 bit(int bitno)
+{
+	return 1 << (bitno & 31);
+}
+
 /*
  * List of msr numbers which we expose to userspace through KVM_GET_MSRS
  * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST.
@@ -456,7 +466,7 @@ static u32 msrs_to_save[] = {
 	MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR,
 #endif
 	MSR_IA32_TIME_STAMP_COUNTER, MSR_KVM_SYSTEM_TIME, MSR_KVM_WALL_CLOCK,
-	MSR_IA32_PERF_STATUS, MSR_IA32_CR_PAT
+	MSR_IA32_PERF_STATUS, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA
 };
 
 static unsigned num_msrs_to_save;
@@ -481,6 +491,28 @@ static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
 		return;
 	}
 
+	if (efer & EFER_FFXSR) {
+		struct kvm_cpuid_entry2 *feat;
+
+		feat = kvm_find_cpuid_entry(vcpu, 0x80000001, 0);
+		if (!feat || !(feat->edx & bit(X86_FEATURE_FXSR_OPT))) {
+			printk(KERN_DEBUG "set_efer: #GP, enable FFXSR w/o CPUID capability\n");
+			kvm_inject_gp(vcpu, 0);
+			return;
+		}
+	}
+
+	if (efer & EFER_SVME) {
+		struct kvm_cpuid_entry2 *feat;
+
+		feat = kvm_find_cpuid_entry(vcpu, 0x80000001, 0);
+		if (!feat || !(feat->ecx & bit(X86_FEATURE_SVM))) {
+			printk(KERN_DEBUG "set_efer: #GP, enable SVM w/o SVM\n");
+			kvm_inject_gp(vcpu, 0);
+			return;
+		}
+	}
+
 	kvm_x86_ops->set_efer(vcpu, efer);
 
 	efer &= ~EFER_LMA;
@@ -586,6 +618,8 @@ static void kvm_set_time_scale(uint32_t tsc_khz, struct pvclock_vcpu_time_info *
 		 hv_clock->tsc_to_system_mul);
 }
 
+static DEFINE_PER_CPU(unsigned long, cpu_tsc_khz);
+
 static void kvm_write_guest_time(struct kvm_vcpu *v)
 {
 	struct timespec ts;
@@ -596,9 +630,9 @@ static void kvm_write_guest_time(struct kvm_vcpu *v)
 	if ((!vcpu->time_page))
 		return;
 
-	if (unlikely(vcpu->hv_clock_tsc_khz != tsc_khz)) {
-		kvm_set_time_scale(tsc_khz, &vcpu->hv_clock);
-		vcpu->hv_clock_tsc_khz = tsc_khz;
+	if (unlikely(vcpu->hv_clock_tsc_khz != __get_cpu_var(cpu_tsc_khz))) {
+		kvm_set_time_scale(__get_cpu_var(cpu_tsc_khz), &vcpu->hv_clock);
+		vcpu->hv_clock_tsc_khz = __get_cpu_var(cpu_tsc_khz);
 	}
 
 	/* Keep irq disabled to prevent changes to the clock */
@@ -629,6 +663,16 @@ static void kvm_write_guest_time(struct kvm_vcpu *v)
 	mark_page_dirty(v->kvm, vcpu->time >> PAGE_SHIFT);
 }
 
+static int kvm_request_guest_time_update(struct kvm_vcpu *v)
+{
+	struct kvm_vcpu_arch *vcpu = &v->arch;
+
+	if (!vcpu->time_page)
+		return 0;
+	set_bit(KVM_REQ_KVMCLOCK_UPDATE, &v->requests);
+	return 1;
+}
+
 static bool msr_mtrr_valid(unsigned msr)
 {
 	switch (msr) {
@@ -722,6 +766,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 		break;
 	case MSR_IA32_UCODE_REV:
 	case MSR_IA32_UCODE_WRITE:
+	case MSR_VM_HSAVE_PA:
 		break;
 	case 0x200 ... 0x2ff:
 		return set_msr_mtrr(vcpu, msr, data);
@@ -758,7 +803,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 			vcpu->arch.time_page = NULL;
 		}
 
-		kvm_write_guest_time(vcpu);
+		kvm_request_guest_time_update(vcpu);
 		break;
 	}
 	default:
@@ -843,6 +888,7 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 	case MSR_IA32_LASTBRANCHTOIP:
 	case MSR_IA32_LASTINTFROMIP:
 	case MSR_IA32_LASTINTTOIP:
+	case MSR_VM_HSAVE_PA:
 		data = 0;
 		break;
 	case MSR_MTRRcap:
@@ -967,10 +1013,13 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
 	case KVM_CAP_SET_TSS_ADDR:
 	case KVM_CAP_EXT_CPUID:
+	case KVM_CAP_CLOCKSOURCE:
 	case KVM_CAP_PIT:
 	case KVM_CAP_NOP_IO_DELAY:
 	case KVM_CAP_MP_STATE:
 	case KVM_CAP_SYNC_MMU:
+	case KVM_CAP_REINJECT_CONTROL:
+	case KVM_CAP_IRQ_INJECT_STATUS:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -991,9 +1040,6 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_IOMMU:
 		r = iommu_found();
 		break;
-	case KVM_CAP_CLOCKSOURCE:
-		r = boot_cpu_has(X86_FEATURE_CONSTANT_TSC);
-		break;
 	default:
 		r = 0;
 		break;
@@ -1044,7 +1090,7 @@ long kvm_arch_dev_ioctl(struct file *filp,
 		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
 			goto out;
 		r = kvm_dev_ioctl_get_supported_cpuid(&cpuid,
-			cpuid_arg->entries);
+						      cpuid_arg->entries);
 		if (r)
 			goto out;
 
@@ -1064,7 +1110,7 @@ out:
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	kvm_x86_ops->vcpu_load(vcpu, cpu);
-	kvm_write_guest_time(vcpu);
+	kvm_request_guest_time_update(vcpu);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
@@ -1142,8 +1188,8 @@ out:
 }
 
 static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
-				    struct kvm_cpuid2 *cpuid,
-				    struct kvm_cpuid_entry2 __user *entries)
+				     struct kvm_cpuid2 *cpuid,
+				     struct kvm_cpuid_entry2 __user *entries)
 {
 	int r;
 
@@ -1162,8 +1208,8 @@ out:
 }
 
 static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
-				    struct kvm_cpuid2 *cpuid,
-				    struct kvm_cpuid_entry2 __user *entries)
+				     struct kvm_cpuid2 *cpuid,
+				     struct kvm_cpuid_entry2 __user *entries)
 {
 	int r;
 
@@ -1172,7 +1218,7 @@ static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
 		goto out;
 	r = -EFAULT;
 	if (copy_to_user(entries, &vcpu->arch.cpuid_entries,
-			   vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
+			 vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
 		goto out;
 	return 0;
 
@@ -1181,18 +1227,13 @@ out:
 	return r;
 }
 
-static inline u32 bit(int bitno)
-{
-	return 1 << (bitno & 31);
-}
-
 static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function,
-			  u32 index)
+			   u32 index)
 {
 	entry->function = function;
 	entry->index = index;
 	cpuid_count(entry->function, entry->index,
-		&entry->eax, &entry->ebx, &entry->ecx, &entry->edx);
+		    &entry->eax, &entry->ebx, &entry->ecx, &entry->edx);
 	entry->flags = 0;
 }
 
@@ -1222,15 +1263,17 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 #ifdef CONFIG_X86_64
 		bit(X86_FEATURE_LM) |
 #endif
+		bit(X86_FEATURE_FXSR_OPT) |
 		bit(X86_FEATURE_MMXEXT) |
 		bit(X86_FEATURE_3DNOWEXT) |
 		bit(X86_FEATURE_3DNOW);
 	const u32 kvm_supported_word3_x86_features =
 		bit(X86_FEATURE_XMM3) | bit(X86_FEATURE_CX16);
 	const u32 kvm_supported_word6_x86_features =
-		bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY);
+		bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY) |
+		bit(X86_FEATURE_SVM);
 
-	/* all func 2 cpuid_count() should be called on the same cpu */
+	/* all calls to cpuid_count() should be made on the same cpu */
 	get_cpu();
 	do_cpuid_1_ent(entry, function, index);
 	++*nent;
@@ -1304,7 +1347,7 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 }
 
 static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
-				    struct kvm_cpuid_entry2 __user *entries)
+				     struct kvm_cpuid_entry2 __user *entries)
 {
 	struct kvm_cpuid_entry2 *cpuid_entries;
 	int limit, nent = 0, r = -E2BIG;
@@ -1321,7 +1364,7 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
 	limit = cpuid_entries[0].eax;
 	for (func = 1; func <= limit && nent < cpuid->nent; ++func)
 		do_cpuid_ent(&cpuid_entries[nent], func, 0,
-				&nent, cpuid->nent);
+			     &nent, cpuid->nent);
 	r = -E2BIG;
 	if (nent >= cpuid->nent)
 		goto out_free;
@@ -1330,10 +1373,10 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
 	limit = cpuid_entries[nent - 1].eax;
 	for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func)
 		do_cpuid_ent(&cpuid_entries[nent], func, 0,
-			       &nent, cpuid->nent);
+			     &nent, cpuid->nent);
 	r = -EFAULT;
 	if (copy_to_user(entries, cpuid_entries,
-			nent * sizeof(struct kvm_cpuid_entry2)))
+			 nent * sizeof(struct kvm_cpuid_entry2)))
 		goto out_free;
 	cpuid->nent = nent;
 	r = 0;
@@ -1477,7 +1520,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
 			goto out;
 		r = kvm_vcpu_ioctl_set_cpuid2(vcpu, &cpuid,
-				cpuid_arg->entries);
+					      cpuid_arg->entries);
 		if (r)
 			goto out;
 		break;
@@ -1490,7 +1533,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
 			goto out;
 		r = kvm_vcpu_ioctl_get_cpuid2(vcpu, &cpuid,
-				cpuid_arg->entries);
+					      cpuid_arg->entries);
 		if (r)
 			goto out;
 		r = -EFAULT;
@@ -1710,6 +1753,15 @@ static int kvm_vm_ioctl_set_pit(struct kvm *kvm, struct kvm_pit_state *ps)
 	return r;
 }
 
+static int kvm_vm_ioctl_reinject(struct kvm *kvm,
+				 struct kvm_reinject_control *control)
+{
+	if (!kvm->arch.vpit)
+		return -ENXIO;
+	kvm->arch.vpit->pit_state.pit_timer.reinject = control->pit_reinject;
+	return 0;
+}
+
 /*
  * Get (and clear) the dirty memory log for a memory slot.
  */
@@ -1807,13 +1859,26 @@ long kvm_arch_vm_ioctl(struct file *filp,
 			}
 		} else
 			goto out;
+		r = kvm_setup_default_irq_routing(kvm);
+		if (r) {
+			kfree(kvm->arch.vpic);
+			kfree(kvm->arch.vioapic);
+			goto out;
+		}
 		break;
 	case KVM_CREATE_PIT:
+		mutex_lock(&kvm->lock);
+		r = -EEXIST;
+		if (kvm->arch.vpit)
+			goto create_pit_unlock;
 		r = -ENOMEM;
 		kvm->arch.vpit = kvm_create_pit(kvm);
 		if (kvm->arch.vpit)
 			r = 0;
+	create_pit_unlock:
+		mutex_unlock(&kvm->lock);
 		break;
+	case KVM_IRQ_LINE_STATUS:
 	case KVM_IRQ_LINE: {
 		struct kvm_irq_level irq_event;
 
@@ -1821,10 +1886,17 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		if (copy_from_user(&irq_event, argp, sizeof irq_event))
 			goto out;
 		if (irqchip_in_kernel(kvm)) {
+			__s32 status;
 			mutex_lock(&kvm->lock);
-			kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
-				    irq_event.irq, irq_event.level);
+			status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
+					irq_event.irq, irq_event.level);
 			mutex_unlock(&kvm->lock);
+			if (ioctl == KVM_IRQ_LINE_STATUS) {
+				irq_event.status = status;
+				if (copy_to_user(argp, &irq_event,
+							sizeof irq_event))
+					goto out;
+			}
 			r = 0;
 		}
 		break;
@@ -1907,6 +1979,17 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = 0;
 		break;
 	}
+	case KVM_REINJECT_CONTROL: {
+		struct kvm_reinject_control control;
+		r =  -EFAULT;
+		if (copy_from_user(&control, argp, sizeof(control)))
+			goto out;
+		r = kvm_vm_ioctl_reinject(kvm, &control);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
 	default:
 		;
 	}
@@ -1960,10 +2043,38 @@ static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu,
 	return dev;
 }
 
-int emulator_read_std(unsigned long addr,
-			     void *val,
-			     unsigned int bytes,
-			     struct kvm_vcpu *vcpu)
+static int kvm_read_guest_virt(gva_t addr, void *val, unsigned int bytes,
+			       struct kvm_vcpu *vcpu)
+{
+	void *data = val;
+	int r = X86EMUL_CONTINUE;
+
+	while (bytes) {
+		gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+		unsigned offset = addr & (PAGE_SIZE-1);
+		unsigned toread = min(bytes, (unsigned)PAGE_SIZE - offset);
+		int ret;
+
+		if (gpa == UNMAPPED_GVA) {
+			r = X86EMUL_PROPAGATE_FAULT;
+			goto out;
+		}
+		ret = kvm_read_guest(vcpu->kvm, gpa, data, toread);
+		if (ret < 0) {
+			r = X86EMUL_UNHANDLEABLE;
+			goto out;
+		}
+
+		bytes -= toread;
+		data += toread;
+		addr += toread;
+	}
+out:
+	return r;
+}
+
+static int kvm_write_guest_virt(gva_t addr, void *val, unsigned int bytes,
+				struct kvm_vcpu *vcpu)
 {
 	void *data = val;
 	int r = X86EMUL_CONTINUE;
@@ -1971,27 +2082,27 @@ int emulator_read_std(unsigned long addr,
 	while (bytes) {
 		gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
 		unsigned offset = addr & (PAGE_SIZE-1);
-		unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
+		unsigned towrite = min(bytes, (unsigned)PAGE_SIZE - offset);
 		int ret;
 
 		if (gpa == UNMAPPED_GVA) {
 			r = X86EMUL_PROPAGATE_FAULT;
 			goto out;
 		}
-		ret = kvm_read_guest(vcpu->kvm, gpa, data, tocopy);
+		ret = kvm_write_guest(vcpu->kvm, gpa, data, towrite);
 		if (ret < 0) {
 			r = X86EMUL_UNHANDLEABLE;
 			goto out;
 		}
 
-		bytes -= tocopy;
-		data += tocopy;
-		addr += tocopy;
+		bytes -= towrite;
+		data += towrite;
+		addr += towrite;
 	}
 out:
 	return r;
 }
-EXPORT_SYMBOL_GPL(emulator_read_std);
+
 
 static int emulator_read_emulated(unsigned long addr,
 				  void *val,
@@ -2013,8 +2124,8 @@ static int emulator_read_emulated(unsigned long addr,
 	if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
 		goto mmio;
 
-	if (emulator_read_std(addr, val, bytes, vcpu)
-			== X86EMUL_CONTINUE)
+	if (kvm_read_guest_virt(addr, val, bytes, vcpu)
+				== X86EMUL_CONTINUE)
 		return X86EMUL_CONTINUE;
 	if (gpa == UNMAPPED_GVA)
 		return X86EMUL_PROPAGATE_FAULT;
@@ -2217,7 +2328,7 @@ void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context)
 
 	rip_linear = rip + get_segment_base(vcpu, VCPU_SREG_CS);
 
-	emulator_read_std(rip_linear, (void *)opcodes, 4, vcpu);
+	kvm_read_guest_virt(rip_linear, (void *)opcodes, 4, vcpu);
 
 	printk(KERN_ERR "emulation failed (%s) rip %lx %02x %02x %02x %02x\n",
 	       context, rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]);
@@ -2225,7 +2336,7 @@ void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context)
 EXPORT_SYMBOL_GPL(kvm_report_emulation_failure);
 
 static struct x86_emulate_ops emulate_ops = {
-	.read_std            = emulator_read_std,
+	.read_std            = kvm_read_guest_virt,
 	.read_emulated       = emulator_read_emulated,
 	.write_emulated      = emulator_write_emulated,
 	.cmpxchg_emulated    = emulator_cmpxchg_emulated,
@@ -2327,40 +2438,19 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
 }
 EXPORT_SYMBOL_GPL(emulate_instruction);
 
-static void free_pio_guest_pages(struct kvm_vcpu *vcpu)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(vcpu->arch.pio.guest_pages); ++i)
-		if (vcpu->arch.pio.guest_pages[i]) {
-			kvm_release_page_dirty(vcpu->arch.pio.guest_pages[i]);
-			vcpu->arch.pio.guest_pages[i] = NULL;
-		}
-}
-
 static int pio_copy_data(struct kvm_vcpu *vcpu)
 {
 	void *p = vcpu->arch.pio_data;
-	void *q;
+	gva_t q = vcpu->arch.pio.guest_gva;
 	unsigned bytes;
-	int nr_pages = vcpu->arch.pio.guest_pages[1] ? 2 : 1;
+	int ret;
 
-	q = vmap(vcpu->arch.pio.guest_pages, nr_pages, VM_READ|VM_WRITE,
-		 PAGE_KERNEL);
-	if (!q) {
-		free_pio_guest_pages(vcpu);
-		return -ENOMEM;
-	}
-	q += vcpu->arch.pio.guest_page_offset;
 	bytes = vcpu->arch.pio.size * vcpu->arch.pio.cur_count;
 	if (vcpu->arch.pio.in)
-		memcpy(q, p, bytes);
+		ret = kvm_write_guest_virt(q, p, bytes, vcpu);
 	else
-		memcpy(p, q, bytes);
-	q -= vcpu->arch.pio.guest_page_offset;
-	vunmap(q);
-	free_pio_guest_pages(vcpu);
-	return 0;
+		ret = kvm_read_guest_virt(q, p, bytes, vcpu);
+	return ret;
 }
 
 int complete_pio(struct kvm_vcpu *vcpu)
@@ -2471,7 +2561,6 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 	vcpu->arch.pio.in = in;
 	vcpu->arch.pio.string = 0;
 	vcpu->arch.pio.down = 0;
-	vcpu->arch.pio.guest_page_offset = 0;
 	vcpu->arch.pio.rep = 0;
 
 	if (vcpu->run->io.direction == KVM_EXIT_IO_IN)
@@ -2499,9 +2588,7 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 		  gva_t address, int rep, unsigned port)
 {
 	unsigned now, in_page;
-	int i, ret = 0;
-	int nr_pages = 1;
-	struct page *page;
+	int ret = 0;
 	struct kvm_io_device *pio_dev;
 
 	vcpu->run->exit_reason = KVM_EXIT_IO;
@@ -2513,7 +2600,6 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 	vcpu->arch.pio.in = in;
 	vcpu->arch.pio.string = 1;
 	vcpu->arch.pio.down = down;
-	vcpu->arch.pio.guest_page_offset = offset_in_page(address);
 	vcpu->arch.pio.rep = rep;
 
 	if (vcpu->run->io.direction == KVM_EXIT_IO_IN)
@@ -2533,15 +2619,8 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 	else
 		in_page = offset_in_page(address) + size;
 	now = min(count, (unsigned long)in_page / size);
-	if (!now) {
-		/*
-		 * String I/O straddles page boundary.  Pin two guest pages
-		 * so that we satisfy atomicity constraints.  Do just one
-		 * transaction to avoid complexity.
-		 */
-		nr_pages = 2;
+	if (!now)
 		now = 1;
-	}
 	if (down) {
 		/*
 		 * String I/O in reverse.  Yuck.  Kill the guest, fix later.
@@ -2556,15 +2635,7 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 	if (vcpu->arch.pio.cur_count == vcpu->arch.pio.count)
 		kvm_x86_ops->skip_emulated_instruction(vcpu);
 
-	for (i = 0; i < nr_pages; ++i) {
-		page = gva_to_page(vcpu, address + i * PAGE_SIZE);
-		vcpu->arch.pio.guest_pages[i] = page;
-		if (!page) {
-			kvm_inject_gp(vcpu, 0);
-			free_pio_guest_pages(vcpu);
-			return 1;
-		}
-	}
+	vcpu->arch.pio.guest_gva = address;
 
 	pio_dev = vcpu_find_pio_dev(vcpu, port,
 				    vcpu->arch.pio.cur_count,
@@ -2572,7 +2643,11 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 	if (!vcpu->arch.pio.in) {
 		/* string PIO write */
 		ret = pio_copy_data(vcpu);
-		if (ret >= 0 && pio_dev) {
+		if (ret == X86EMUL_PROPAGATE_FAULT) {
+			kvm_inject_gp(vcpu, 0);
+			return 1;
+		}
+		if (ret == 0 && pio_dev) {
 			pio_string_write(pio_dev, vcpu);
 			complete_pio(vcpu);
 			if (vcpu->arch.pio.count == 0)
@@ -2587,9 +2662,72 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 }
 EXPORT_SYMBOL_GPL(kvm_emulate_pio_string);
 
+static void bounce_off(void *info)
+{
+	/* nothing */
+}
+
+static unsigned int  ref_freq;
+static unsigned long tsc_khz_ref;
+
+static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
+				     void *data)
+{
+	struct cpufreq_freqs *freq = data;
+	struct kvm *kvm;
+	struct kvm_vcpu *vcpu;
+	int i, send_ipi = 0;
+
+	if (!ref_freq)
+		ref_freq = freq->old;
+
+	if (val == CPUFREQ_PRECHANGE && freq->old > freq->new)
+		return 0;
+	if (val == CPUFREQ_POSTCHANGE && freq->old < freq->new)
+		return 0;
+	per_cpu(cpu_tsc_khz, freq->cpu) = cpufreq_scale(tsc_khz_ref, ref_freq, freq->new);
+
+	spin_lock(&kvm_lock);
+	list_for_each_entry(kvm, &vm_list, vm_list) {
+		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+			vcpu = kvm->vcpus[i];
+			if (!vcpu)
+				continue;
+			if (vcpu->cpu != freq->cpu)
+				continue;
+			if (!kvm_request_guest_time_update(vcpu))
+				continue;
+			if (vcpu->cpu != smp_processor_id())
+				send_ipi++;
+		}
+	}
+	spin_unlock(&kvm_lock);
+
+	if (freq->old < freq->new && send_ipi) {
+		/*
+		 * We upscale the frequency.  Must make the guest
+		 * doesn't see old kvmclock values while running with
+		 * the new frequency, otherwise we risk the guest sees
+		 * time go backwards.
+		 *
+		 * In case we update the frequency for another cpu
+		 * (which might be in guest context) send an interrupt
+		 * to kick the cpu out of guest context.  Next time
+		 * guest context is entered kvmclock will be updated,
+		 * so the guest will not see stale values.
+		 */
+		smp_call_function_single(freq->cpu, bounce_off, NULL, 1);
+	}
+	return 0;
+}
+
+static struct notifier_block kvmclock_cpufreq_notifier_block = {
+        .notifier_call  = kvmclock_cpufreq_notifier
+};
+
 int kvm_arch_init(void *opaque)
 {
-	int r;
+	int r, cpu;
 	struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque;
 
 	if (kvm_x86_ops) {
@@ -2620,6 +2758,15 @@ int kvm_arch_init(void *opaque)
 	kvm_mmu_set_base_ptes(PT_PRESENT_MASK);
 	kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK,
 			PT_DIRTY_MASK, PT64_NX_MASK, 0, 0);
+
+	for_each_possible_cpu(cpu)
+		per_cpu(cpu_tsc_khz, cpu) = tsc_khz;
+	if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) {
+		tsc_khz_ref = tsc_khz;
+		cpufreq_register_notifier(&kvmclock_cpufreq_notifier_block,
+					  CPUFREQ_TRANSITION_NOTIFIER);
+	}
+
 	return 0;
 
 out:
@@ -2827,25 +2974,20 @@ static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e,
 	if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index)
 		return 0;
 	if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) &&
-		!(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT))
+	    !(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT))
 		return 0;
 	return 1;
 }
 
-void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
+struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu,
+					      u32 function, u32 index)
 {
 	int i;
-	u32 function, index;
-	struct kvm_cpuid_entry2 *e, *best;
+	struct kvm_cpuid_entry2 *best = NULL;
 
-	function = kvm_register_read(vcpu, VCPU_REGS_RAX);
-	index = kvm_register_read(vcpu, VCPU_REGS_RCX);
-	kvm_register_write(vcpu, VCPU_REGS_RAX, 0);
-	kvm_register_write(vcpu, VCPU_REGS_RBX, 0);
-	kvm_register_write(vcpu, VCPU_REGS_RCX, 0);
-	kvm_register_write(vcpu, VCPU_REGS_RDX, 0);
-	best = NULL;
 	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
+		struct kvm_cpuid_entry2 *e;
+
 		e = &vcpu->arch.cpuid_entries[i];
 		if (is_matching_cpuid_entry(e, function, index)) {
 			if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC)
@@ -2860,6 +3002,21 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
 			if (!best || e->function > best->function)
 				best = e;
 	}
+	return best;
+}
+
+void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
+{
+	u32 function, index;
+	struct kvm_cpuid_entry2 *best;
+
+	function = kvm_register_read(vcpu, VCPU_REGS_RAX);
+	index = kvm_register_read(vcpu, VCPU_REGS_RCX);
+	kvm_register_write(vcpu, VCPU_REGS_RAX, 0);
+	kvm_register_write(vcpu, VCPU_REGS_RBX, 0);
+	kvm_register_write(vcpu, VCPU_REGS_RCX, 0);
+	kvm_register_write(vcpu, VCPU_REGS_RDX, 0);
+	best = kvm_find_cpuid_entry(vcpu, function, index);
 	if (best) {
 		kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax);
 		kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx);
@@ -2945,6 +3102,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	if (vcpu->requests) {
 		if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests))
 			__kvm_migrate_timers(vcpu);
+		if (test_and_clear_bit(KVM_REQ_KVMCLOCK_UPDATE, &vcpu->requests))
+			kvm_write_guest_time(vcpu);
 		if (test_and_clear_bit(KVM_REQ_MMU_SYNC, &vcpu->requests))
 			kvm_mmu_sync_roots(vcpu);
 		if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
@@ -2979,9 +3138,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		goto out;
 	}
 
-	if (vcpu->guest_debug.enabled)
-		kvm_x86_ops->guest_debug_pre(vcpu);
-
 	vcpu->guest_mode = 1;
 	/*
 	 * Make sure that guest_mode assignment won't happen after
@@ -3002,10 +3158,34 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 
 	kvm_guest_enter();
 
+	get_debugreg(vcpu->arch.host_dr6, 6);
+	get_debugreg(vcpu->arch.host_dr7, 7);
+	if (unlikely(vcpu->arch.switch_db_regs)) {
+		get_debugreg(vcpu->arch.host_db[0], 0);
+		get_debugreg(vcpu->arch.host_db[1], 1);
+		get_debugreg(vcpu->arch.host_db[2], 2);
+		get_debugreg(vcpu->arch.host_db[3], 3);
+
+		set_debugreg(0, 7);
+		set_debugreg(vcpu->arch.eff_db[0], 0);
+		set_debugreg(vcpu->arch.eff_db[1], 1);
+		set_debugreg(vcpu->arch.eff_db[2], 2);
+		set_debugreg(vcpu->arch.eff_db[3], 3);
+	}
 
 	KVMTRACE_0D(VMENTRY, vcpu, entryexit);
 	kvm_x86_ops->run(vcpu, kvm_run);
 
+	if (unlikely(vcpu->arch.switch_db_regs)) {
+		set_debugreg(0, 7);
+		set_debugreg(vcpu->arch.host_db[0], 0);
+		set_debugreg(vcpu->arch.host_db[1], 1);
+		set_debugreg(vcpu->arch.host_db[2], 2);
+		set_debugreg(vcpu->arch.host_db[3], 3);
+	}
+	set_debugreg(vcpu->arch.host_dr6, 6);
+	set_debugreg(vcpu->arch.host_dr7, 7);
+
 	vcpu->guest_mode = 0;
 	local_irq_enable();
 
@@ -3192,7 +3372,7 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 	/*
 	 * Don't leak debug flags in case they were set for guest debugging
 	 */
-	if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep)
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
 		regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
 
 	vcpu_put(vcpu);
@@ -3811,15 +3991,32 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
-	int r;
+	int i, r;
 
 	vcpu_load(vcpu);
 
+	if ((dbg->control & (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP)) ==
+	    (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP)) {
+		for (i = 0; i < KVM_NR_DB_REGS; ++i)
+			vcpu->arch.eff_db[i] = dbg->arch.debugreg[i];
+		vcpu->arch.switch_db_regs =
+			(dbg->arch.debugreg[7] & DR7_BP_EN_MASK);
+	} else {
+		for (i = 0; i < KVM_NR_DB_REGS; i++)
+			vcpu->arch.eff_db[i] = vcpu->arch.db[i];
+		vcpu->arch.switch_db_regs = (vcpu->arch.dr7 & DR7_BP_EN_MASK);
+	}
+
 	r = kvm_x86_ops->set_guest_debug(vcpu, dbg);
 
+	if (dbg->control & KVM_GUESTDBG_INJECT_DB)
+		kvm_queue_exception(vcpu, DB_VECTOR);
+	else if (dbg->control & KVM_GUESTDBG_INJECT_BP)
+		kvm_queue_exception(vcpu, BP_VECTOR);
+
 	vcpu_put(vcpu);
 
 	return r;
@@ -4007,6 +4204,11 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu)
 	vcpu->arch.nmi_pending = false;
 	vcpu->arch.nmi_injected = false;
 
+	vcpu->arch.switch_db_regs = 0;
+	memset(vcpu->arch.db, 0, sizeof(vcpu->arch.db));
+	vcpu->arch.dr6 = DR6_FIXED_1;
+	vcpu->arch.dr7 = DR7_FIXED_1;
+
 	return kvm_x86_ops->vcpu_reset(vcpu);
 }
 
@@ -4100,6 +4302,8 @@ struct  kvm *kvm_arch_create_vm(void)
 	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
 	set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap);
 
+	rdtscll(kvm->arch.vm_init_tsc);
+
 	return kvm;
 }
 

arch/x86/kvm/x86_emulate.c

@@ -178,7 +178,7 @@ static u32 opcode_table[256] = {
 	0, ImplicitOps | Stack, 0, 0,
 	ByteOp | DstMem | SrcImm | ModRM | Mov, DstMem | SrcImm | ModRM | Mov,
 	/* 0xC8 - 0xCF */
-	0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, ImplicitOps | Stack, 0, 0, 0, 0,
 	/* 0xD0 - 0xD7 */
 	ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM,
 	ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM,
@@ -1136,18 +1136,19 @@ static inline void emulate_push(struct x86_emulate_ctxt *ctxt)
 }
 
 static int emulate_pop(struct x86_emulate_ctxt *ctxt,
-		       struct x86_emulate_ops *ops)
+		       struct x86_emulate_ops *ops,
+		       void *dest, int len)
 {
 	struct decode_cache *c = &ctxt->decode;
 	int rc;
 
 	rc = ops->read_emulated(register_address(c, ss_base(ctxt),
 						 c->regs[VCPU_REGS_RSP]),
-				&c->src.val, c->src.bytes, ctxt->vcpu);
+				dest, len, ctxt->vcpu);
 	if (rc != 0)
 		return rc;
 
-	register_address_increment(c, &c->regs[VCPU_REGS_RSP], c->src.bytes);
+	register_address_increment(c, &c->regs[VCPU_REGS_RSP], len);
 	return rc;
 }
 
@@ -1157,11 +1158,9 @@ static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt,
 	struct decode_cache *c = &ctxt->decode;
 	int rc;
 
-	c->src.bytes = c->dst.bytes;
-	rc = emulate_pop(ctxt, ops);
+	rc = emulate_pop(ctxt, ops, &c->dst.val, c->dst.bytes);
 	if (rc != 0)
 		return rc;
-	c->dst.val = c->src.val;
 	return 0;
 }
 
@@ -1279,6 +1278,25 @@ static inline int emulate_grp9(struct x86_emulate_ctxt *ctxt,
 	return 0;
 }
 
+static int emulate_ret_far(struct x86_emulate_ctxt *ctxt,
+			   struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc;
+	unsigned long cs;
+
+	rc = emulate_pop(ctxt, ops, &c->eip, c->op_bytes);
+	if (rc)
+		return rc;
+	if (c->op_bytes == 4)
+		c->eip = (u32)c->eip;
+	rc = emulate_pop(ctxt, ops, &cs, c->op_bytes);
+	if (rc)
+		return rc;
+	rc = kvm_load_segment_descriptor(ctxt->vcpu, (u16)cs, 1, VCPU_SREG_CS);
+	return rc;
+}
+
 static inline int writeback(struct x86_emulate_ctxt *ctxt,
 			    struct x86_emulate_ops *ops)
 {
@@ -1467,11 +1485,9 @@ special_insn:
 		break;
 	case 0x58 ... 0x5f: /* pop reg */
 	pop_instruction:
-		c->src.bytes = c->op_bytes;
-		rc = emulate_pop(ctxt, ops);
+		rc = emulate_pop(ctxt, ops, &c->dst.val, c->op_bytes);
 		if (rc != 0)
 			goto done;
-		c->dst.val = c->src.val;
 		break;
 	case 0x63:		/* movsxd */
 		if (ctxt->mode != X86EMUL_MODE_PROT64)
@@ -1738,6 +1754,11 @@ special_insn:
 	mov:
 		c->dst.val = c->src.val;
 		break;
+	case 0xcb:		/* ret far */
+		rc = emulate_ret_far(ctxt, ops);
+		if (rc)
+			goto done;
+		break;
 	case 0xd0 ... 0xd1:	/* Grp2 */
 		c->src.val = 1;
 		emulate_grp2(ctxt);
@@ -1908,11 +1929,16 @@ twobyte_insn:
 			c->dst.type = OP_NONE;
 			break;
 		case 3: /* lidt/vmmcall */
-			if (c->modrm_mod == 3 && c->modrm_rm == 1) {
-				rc = kvm_fix_hypercall(ctxt->vcpu);
-				if (rc)
-					goto done;
-				kvm_emulate_hypercall(ctxt->vcpu);
+			if (c->modrm_mod == 3) {
+				switch (c->modrm_rm) {
+				case 1:
+					rc = kvm_fix_hypercall(ctxt->vcpu);
+					if (rc)
+						goto done;
+					break;
+				default:
+					goto cannot_emulate;
+				}
 			} else {
 				rc = read_descriptor(ctxt, ops, c->src.ptr,
 						     &size, &address,

include/linux/kvm.h

@@ -48,7 +48,10 @@ struct kvm_irq_level {
 	 * For IA-64 (APIC model) IOAPIC0: irq 0-23; IOAPIC1: irq 24-47..
 	 * For X86 (standard AT mode) PIC0/1: irq 0-15. IOAPIC0: 0-23..
 	 */
-	__u32 irq;
+	union {
+		__u32 irq;
+		__s32 status;
+	};
 	__u32 level;
 };
 
@@ -126,6 +129,7 @@ struct kvm_run {
 			__u64 data_offset; /* relative to kvm_run start */
 		} io;
 		struct {
+			struct kvm_debug_exit_arch arch;
 		} debug;
 		/* KVM_EXIT_MMIO */
 		struct {
@@ -217,21 +221,6 @@ struct kvm_interrupt {
 	__u32 irq;
 };
 
-struct kvm_breakpoint {
-	__u32 enabled;
-	__u32 padding;
-	__u64 address;
-};
-
-/* for KVM_DEBUG_GUEST */
-struct kvm_debug_guest {
-	/* int */
-	__u32 enabled;
-	__u32 pad;
-	struct kvm_breakpoint breakpoints[4];
-	__u32 singlestep;
-};
-
 /* for KVM_GET_DIRTY_LOG */
 struct kvm_dirty_log {
 	__u32 slot;
@@ -292,6 +281,17 @@ struct kvm_s390_interrupt {
 	__u64 parm64;
 };
 
+/* for KVM_SET_GUEST_DEBUG */
+
+#define KVM_GUESTDBG_ENABLE		0x00000001
+#define KVM_GUESTDBG_SINGLESTEP		0x00000002
+
+struct kvm_guest_debug {
+	__u32 control;
+	__u32 pad;
+	struct kvm_guest_debug_arch arch;
+};
+
 #define KVM_TRC_SHIFT           16
 /*
  * kvm trace categories
@@ -395,6 +395,57 @@ struct kvm_trace_rec {
 #define KVM_CAP_DESTROY_MEMORY_REGION_WORKS 21
 #ifdef __KVM_HAVE_USER_NMI
 #define KVM_CAP_USER_NMI 22
+#endif
+#ifdef __KVM_HAVE_GUEST_DEBUG
+#define KVM_CAP_SET_GUEST_DEBUG 23
+#endif
+#ifdef __KVM_HAVE_PIT
+#define KVM_CAP_REINJECT_CONTROL 24
+#endif
+#ifdef __KVM_HAVE_IOAPIC
+#define KVM_CAP_IRQ_ROUTING 25
+#endif
+#define KVM_CAP_IRQ_INJECT_STATUS 26
+#ifdef __KVM_HAVE_DEVICE_ASSIGNMENT
+#define KVM_CAP_DEVICE_DEASSIGNMENT 27
+#endif
+
+#ifdef KVM_CAP_IRQ_ROUTING
+
+struct kvm_irq_routing_irqchip {
+	__u32 irqchip;
+	__u32 pin;
+};
+
+struct kvm_irq_routing_msi {
+	__u32 address_lo;
+	__u32 address_hi;
+	__u32 data;
+	__u32 pad;
+};
+
+/* gsi routing entry types */
+#define KVM_IRQ_ROUTING_IRQCHIP 1
+#define KVM_IRQ_ROUTING_MSI 2
+
+struct kvm_irq_routing_entry {
+	__u32 gsi;
+	__u32 type;
+	__u32 flags;
+	__u32 pad;
+	union {
+		struct kvm_irq_routing_irqchip irqchip;
+		struct kvm_irq_routing_msi msi;
+		__u32 pad[8];
+	} u;
+};
+
+struct kvm_irq_routing {
+	__u32 nr;
+	__u32 flags;
+	struct kvm_irq_routing_entry entries[0];
+};
+
 #endif
 
 /*
@@ -421,14 +472,19 @@ struct kvm_trace_rec {
 #define KVM_CREATE_PIT		  _IO(KVMIO,  0x64)
 #define KVM_GET_PIT		  _IOWR(KVMIO, 0x65, struct kvm_pit_state)
 #define KVM_SET_PIT		  _IOR(KVMIO,  0x66, struct kvm_pit_state)
+#define KVM_IRQ_LINE_STATUS	  _IOWR(KVMIO, 0x67, struct kvm_irq_level)
 #define KVM_REGISTER_COALESCED_MMIO \
 			_IOW(KVMIO,  0x67, struct kvm_coalesced_mmio_zone)
 #define KVM_UNREGISTER_COALESCED_MMIO \
 			_IOW(KVMIO,  0x68, struct kvm_coalesced_mmio_zone)
 #define KVM_ASSIGN_PCI_DEVICE _IOR(KVMIO, 0x69, \
 				   struct kvm_assigned_pci_dev)
+#define KVM_SET_GSI_ROUTING       _IOW(KVMIO, 0x6a, struct kvm_irq_routing)
 #define KVM_ASSIGN_IRQ _IOR(KVMIO, 0x70, \
 			    struct kvm_assigned_irq)
+#define KVM_REINJECT_CONTROL      _IO(KVMIO, 0x71)
+#define KVM_DEASSIGN_PCI_DEVICE _IOW(KVMIO, 0x72, \
+				     struct kvm_assigned_pci_dev)
 
 /*
  * ioctls for vcpu fds
@@ -440,7 +496,8 @@ struct kvm_trace_rec {
 #define KVM_SET_SREGS             _IOW(KVMIO,  0x84, struct kvm_sregs)
 #define KVM_TRANSLATE             _IOWR(KVMIO, 0x85, struct kvm_translation)
 #define KVM_INTERRUPT             _IOW(KVMIO,  0x86, struct kvm_interrupt)
-#define KVM_DEBUG_GUEST           _IOW(KVMIO,  0x87, struct kvm_debug_guest)
+/* KVM_DEBUG_GUEST is no longer supported, use KVM_SET_GUEST_DEBUG instead */
+#define KVM_DEBUG_GUEST           __KVM_DEPRECATED_DEBUG_GUEST
 #define KVM_GET_MSRS              _IOWR(KVMIO, 0x88, struct kvm_msrs)
 #define KVM_SET_MSRS              _IOW(KVMIO,  0x89, struct kvm_msrs)
 #define KVM_SET_CPUID             _IOW(KVMIO,  0x8a, struct kvm_cpuid)
@@ -469,6 +526,29 @@ struct kvm_trace_rec {
 #define KVM_SET_MP_STATE          _IOW(KVMIO,  0x99, struct kvm_mp_state)
 /* Available with KVM_CAP_NMI */
 #define KVM_NMI                   _IO(KVMIO,  0x9a)
+/* Available with KVM_CAP_SET_GUEST_DEBUG */
+#define KVM_SET_GUEST_DEBUG       _IOW(KVMIO,  0x9b, struct kvm_guest_debug)
+
+/*
+ * Deprecated interfaces
+ */
+struct kvm_breakpoint {
+	__u32 enabled;
+	__u32 padding;
+	__u64 address;
+};
+
+struct kvm_debug_guest {
+	__u32 enabled;
+	__u32 pad;
+	struct kvm_breakpoint breakpoints[4];
+	__u32 singlestep;
+};
+
+#define __KVM_DEPRECATED_DEBUG_GUEST _IOW(KVMIO,  0x87, struct kvm_debug_guest)
+
+#define KVM_IA64_VCPU_GET_STACK   _IOR(KVMIO,  0x9a, void *)
+#define KVM_IA64_VCPU_SET_STACK   _IOW(KVMIO,  0x9b, void *)
 
 #define KVM_TRC_INJ_VIRQ         (KVM_TRC_HANDLER + 0x02)
 #define KVM_TRC_REDELIVER_EVT    (KVM_TRC_HANDLER + 0x03)
@@ -522,6 +602,7 @@ struct kvm_assigned_irq {
 
 #define KVM_DEV_ASSIGN_ENABLE_IOMMU	(1 << 0)
 
+#define KVM_DEV_IRQ_ASSIGN_MSI_ACTION	KVM_DEV_IRQ_ASSIGN_ENABLE_MSI
 #define KVM_DEV_IRQ_ASSIGN_ENABLE_MSI	(1 << 0)
 
 #endif

include/linux/kvm_host.h

@@ -37,6 +37,7 @@
 #define KVM_REQ_PENDING_TIMER      5
 #define KVM_REQ_UNHALT             6
 #define KVM_REQ_MMU_SYNC           7
+#define KVM_REQ_KVMCLOCK_UPDATE    8
 
 #define KVM_USERSPACE_IRQ_SOURCE_ID	0
 
@@ -73,7 +74,7 @@ struct kvm_vcpu {
 	struct kvm_run *run;
 	int guest_mode;
 	unsigned long requests;
-	struct kvm_guest_debug guest_debug;
+	unsigned long guest_debug;
 	int fpu_active;
 	int guest_fpu_loaded;
 	wait_queue_head_t wq;
@@ -107,6 +108,20 @@ struct kvm_memory_slot {
 	int user_alloc;
 };
 
+struct kvm_kernel_irq_routing_entry {
+	u32 gsi;
+	int (*set)(struct kvm_kernel_irq_routing_entry *e,
+		    struct kvm *kvm, int level);
+	union {
+		struct {
+			unsigned irqchip;
+			unsigned pin;
+		} irqchip;
+		struct msi_msg msi;
+	};
+	struct list_head link;
+};
+
 struct kvm {
 	struct mutex lock; /* protects the vcpus array and APIC accesses */
 	spinlock_t mmu_lock;
@@ -127,6 +142,11 @@ struct kvm {
 	struct kvm_coalesced_mmio_ring *coalesced_mmio_ring;
 #endif
 
+#ifdef CONFIG_HAVE_KVM_IRQCHIP
+	struct list_head irq_routing; /* of kvm_kernel_irq_routing_entry */
+	struct hlist_head mask_notifier_list;
+#endif
+
 #ifdef KVM_ARCH_WANT_MMU_NOTIFIER
 	struct mmu_notifier mmu_notifier;
 	unsigned long mmu_notifier_seq;
@@ -237,7 +257,6 @@ int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
 				   int user_alloc);
 long kvm_arch_vm_ioctl(struct file *filp,
 		       unsigned int ioctl, unsigned long arg);
-void kvm_arch_destroy_vm(struct kvm *kvm);
 
 int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
 int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
@@ -255,8 +274,8 @@ int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state);
 int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state);
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg);
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg);
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);
 
 int kvm_arch_init(void *opaque);
@@ -310,7 +329,6 @@ struct kvm_assigned_dev_kernel {
 	int host_irq;
 	bool host_irq_disabled;
 	int guest_irq;
-	struct msi_msg guest_msi;
 #define KVM_ASSIGNED_DEV_GUEST_INTX	(1 << 0)
 #define KVM_ASSIGNED_DEV_GUEST_MSI	(1 << 1)
 #define KVM_ASSIGNED_DEV_HOST_INTX	(1 << 8)
@@ -321,8 +339,21 @@ struct kvm_assigned_dev_kernel {
 	struct pci_dev *dev;
 	struct kvm *kvm;
 };
-void kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level);
-void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi);
+
+struct kvm_irq_mask_notifier {
+	void (*func)(struct kvm_irq_mask_notifier *kimn, bool masked);
+	int irq;
+	struct hlist_node link;
+};
+
+void kvm_register_irq_mask_notifier(struct kvm *kvm, int irq,
+				    struct kvm_irq_mask_notifier *kimn);
+void kvm_unregister_irq_mask_notifier(struct kvm *kvm, int irq,
+				      struct kvm_irq_mask_notifier *kimn);
+void kvm_fire_mask_notifiers(struct kvm *kvm, int irq, bool mask);
+
+int kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level);
+void kvm_notify_acked_irq(struct kvm *kvm, unsigned irqchip, unsigned pin);
 void kvm_register_irq_ack_notifier(struct kvm *kvm,
 				   struct kvm_irq_ack_notifier *kian);
 void kvm_unregister_irq_ack_notifier(struct kvm_irq_ack_notifier *kian);
@@ -464,4 +495,21 @@ static inline int mmu_notifier_retry(struct kvm_vcpu *vcpu, unsigned long mmu_se
 }
 #endif
 
+#ifdef CONFIG_HAVE_KVM_IRQCHIP
+
+#define KVM_MAX_IRQ_ROUTES 1024
+
+int kvm_setup_default_irq_routing(struct kvm *kvm);
+int kvm_set_irq_routing(struct kvm *kvm,
+			const struct kvm_irq_routing_entry *entries,
+			unsigned nr,
+			unsigned flags);
+void kvm_free_irq_routing(struct kvm *kvm);
+
+#else
+
+static inline void kvm_free_irq_routing(struct kvm *kvm) {}
+
+#endif
+
 #endif

include/linux/kvm_types.h

@@ -40,17 +40,4 @@ typedef unsigned long  hfn_t;
 
 typedef hfn_t pfn_t;
 
-struct kvm_pio_request {
-	unsigned long count;
-	int cur_count;
-	struct page *guest_pages[2];
-	unsigned guest_page_offset;
-	int in;
-	int port;
-	int size;
-	int string;
-	int down;
-	int rep;
-};
-
 #endif /* __KVM_TYPES_H__ */

virt/kvm/ioapic.c

@@ -83,24 +83,28 @@ static unsigned long ioapic_read_indirect(struct kvm_ioapic *ioapic,
 	return result;
 }
 
-static void ioapic_service(struct kvm_ioapic *ioapic, unsigned int idx)
+static int ioapic_service(struct kvm_ioapic *ioapic, unsigned int idx)
 {
 	union ioapic_redir_entry *pent;
+	int injected = -1;
 
 	pent = &ioapic->redirtbl[idx];
 
 	if (!pent->fields.mask) {
-		int injected = ioapic_deliver(ioapic, idx);
+		injected = ioapic_deliver(ioapic, idx);
 		if (injected && pent->fields.trig_mode == IOAPIC_LEVEL_TRIG)
 			pent->fields.remote_irr = 1;
 	}
 	if (!pent->fields.trig_mode)
 		ioapic->irr &= ~(1 << idx);
+
+	return injected;
 }
 
 static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val)
 {
 	unsigned index;
+	bool mask_before, mask_after;
 
 	switch (ioapic->ioregsel) {
 	case IOAPIC_REG_VERSION:
@@ -120,6 +124,7 @@ static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val)
 		ioapic_debug("change redir index %x val %x\n", index, val);
 		if (index >= IOAPIC_NUM_PINS)
 			return;
+		mask_before = ioapic->redirtbl[index].fields.mask;
 		if (ioapic->ioregsel & 1) {
 			ioapic->redirtbl[index].bits &= 0xffffffff;
 			ioapic->redirtbl[index].bits |= (u64) val << 32;
@@ -128,6 +133,9 @@ static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val)
 			ioapic->redirtbl[index].bits |= (u32) val;
 			ioapic->redirtbl[index].fields.remote_irr = 0;
 		}
+		mask_after = ioapic->redirtbl[index].fields.mask;
+		if (mask_before != mask_after)
+			kvm_fire_mask_notifiers(ioapic->kvm, index, mask_after);
 		if (ioapic->irr & (1 << index))
 			ioapic_service(ioapic, index);
 		break;
@@ -202,7 +210,7 @@ static int ioapic_deliver(struct kvm_ioapic *ioapic, int irq)
 	u8 trig_mode = ioapic->redirtbl[irq].fields.trig_mode;
 	u32 deliver_bitmask;
 	struct kvm_vcpu *vcpu;
-	int vcpu_id, r = 0;
+	int vcpu_id, r = -1;
 
 	ioapic_debug("dest=%x dest_mode=%x delivery_mode=%x "
 		     "vector=%x trig_mode=%x\n",
@@ -242,7 +250,9 @@ static int ioapic_deliver(struct kvm_ioapic *ioapic, int irq)
 			deliver_bitmask &= ~(1 << vcpu_id);
 			vcpu = ioapic->kvm->vcpus[vcpu_id];
 			if (vcpu) {
-				r = ioapic_inj_irq(ioapic, vcpu, vector,
+				if (r < 0)
+					r = 0;
+				r += ioapic_inj_irq(ioapic, vcpu, vector,
 					       trig_mode, delivery_mode);
 			}
 		}
@@ -253,8 +263,10 @@ static int ioapic_deliver(struct kvm_ioapic *ioapic, int irq)
 				continue;
 			deliver_bitmask &= ~(1 << vcpu_id);
 			vcpu = ioapic->kvm->vcpus[vcpu_id];
-			if (vcpu)
+			if (vcpu) {
 				ioapic_inj_nmi(vcpu);
+				r = 1;
+			}
 			else
 				ioapic_debug("NMI to vcpu %d failed\n",
 						vcpu->vcpu_id);
@@ -268,11 +280,12 @@ static int ioapic_deliver(struct kvm_ioapic *ioapic, int irq)
 	return r;
 }
 
-void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
+int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
 {
 	u32 old_irr = ioapic->irr;
 	u32 mask = 1 << irq;
 	union ioapic_redir_entry entry;
+	int ret = 1;
 
 	if (irq >= 0 && irq < IOAPIC_NUM_PINS) {
 		entry = ioapic->redirtbl[irq];
@@ -283,25 +296,26 @@ void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
 			ioapic->irr |= mask;
 			if ((!entry.fields.trig_mode && old_irr != ioapic->irr)
 			    || !entry.fields.remote_irr)
-				ioapic_service(ioapic, irq);
+				ret = ioapic_service(ioapic, irq);
 		}
 	}
+	return ret;
 }
 
-static void __kvm_ioapic_update_eoi(struct kvm_ioapic *ioapic, int gsi,
+static void __kvm_ioapic_update_eoi(struct kvm_ioapic *ioapic, int pin,
 				    int trigger_mode)
 {
 	union ioapic_redir_entry *ent;
 
-	ent = &ioapic->redirtbl[gsi];
+	ent = &ioapic->redirtbl[pin];
 
-	kvm_notify_acked_irq(ioapic->kvm, gsi);
+	kvm_notify_acked_irq(ioapic->kvm, KVM_IRQCHIP_IOAPIC, pin);
 
 	if (trigger_mode == IOAPIC_LEVEL_TRIG) {
 		ASSERT(ent->fields.trig_mode == IOAPIC_LEVEL_TRIG);
 		ent->fields.remote_irr = 0;
-		if (!ent->fields.mask && (ioapic->irr & (1 << gsi)))
-			ioapic_service(ioapic, gsi);
+		if (!ent->fields.mask && (ioapic->irr & (1 << pin)))
+			ioapic_service(ioapic, pin);
 	}
 }
 
@@ -426,3 +440,4 @@ int kvm_ioapic_init(struct kvm *kvm)
 	kvm_io_bus_register_dev(&kvm->mmio_bus, &ioapic->dev);
 	return 0;
 }
+

virt/kvm/ioapic.h

@@ -83,7 +83,7 @@ struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
 				       unsigned long bitmap);
 void kvm_ioapic_update_eoi(struct kvm *kvm, int vector, int trigger_mode);
 int kvm_ioapic_init(struct kvm *kvm);
-void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
+int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
 void kvm_ioapic_reset(struct kvm_ioapic *ioapic);
 u32 kvm_ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
 				u8 dest_mode);

virt/kvm/irq_comm.c

@@ -20,35 +20,132 @@
  */
 
 #include <linux/kvm_host.h>
+
+#include <asm/msidef.h>
+
 #include "irq.h"
 
 #include "ioapic.h"
 
-/* This should be called with the kvm->lock mutex held */
-void kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level)
+static int kvm_set_pic_irq(struct kvm_kernel_irq_routing_entry *e,
+			   struct kvm *kvm, int level)
+{
+#ifdef CONFIG_X86
+	return kvm_pic_set_irq(pic_irqchip(kvm), e->irqchip.pin, level);
+#else
+	return -1;
+#endif
+}
+
+static int kvm_set_ioapic_irq(struct kvm_kernel_irq_routing_entry *e,
+			      struct kvm *kvm, int level)
+{
+	return kvm_ioapic_set_irq(kvm->arch.vioapic, e->irqchip.pin, level);
+}
+
+static int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e,
+		       struct kvm *kvm, int level)
+{
+	int vcpu_id, r = -1;
+	struct kvm_vcpu *vcpu;
+	struct kvm_ioapic *ioapic = ioapic_irqchip(kvm);
+	int dest_id = (e->msi.address_lo & MSI_ADDR_DEST_ID_MASK)
+			>> MSI_ADDR_DEST_ID_SHIFT;
+	int vector = (e->msi.data & MSI_DATA_VECTOR_MASK)
+			>> MSI_DATA_VECTOR_SHIFT;
+	int dest_mode = test_bit(MSI_ADDR_DEST_MODE_SHIFT,
+				(unsigned long *)&e->msi.address_lo);
+	int trig_mode = test_bit(MSI_DATA_TRIGGER_SHIFT,
+				(unsigned long *)&e->msi.data);
+	int delivery_mode = test_bit(MSI_DATA_DELIVERY_MODE_SHIFT,
+				(unsigned long *)&e->msi.data);
+	u32 deliver_bitmask;
+
+	BUG_ON(!ioapic);
+
+	deliver_bitmask = kvm_ioapic_get_delivery_bitmask(ioapic,
+				dest_id, dest_mode);
+	/* IOAPIC delivery mode value is the same as MSI here */
+	switch (delivery_mode) {
+	case IOAPIC_LOWEST_PRIORITY:
+		vcpu = kvm_get_lowest_prio_vcpu(ioapic->kvm, vector,
+				deliver_bitmask);
+		if (vcpu != NULL)
+			r = kvm_apic_set_irq(vcpu, vector, trig_mode);
+		else
+			printk(KERN_INFO "kvm: null lowest priority vcpu!\n");
+		break;
+	case IOAPIC_FIXED:
+		for (vcpu_id = 0; deliver_bitmask != 0; vcpu_id++) {
+			if (!(deliver_bitmask & (1 << vcpu_id)))
+				continue;
+			deliver_bitmask &= ~(1 << vcpu_id);
+			vcpu = ioapic->kvm->vcpus[vcpu_id];
+			if (vcpu) {
+				if (r < 0)
+					r = 0;
+				r += kvm_apic_set_irq(vcpu, vector, trig_mode);
+			}
+		}
+		break;
+	default:
+		break;
+	}
+	return r;
+}
+
+/* This should be called with the kvm->lock mutex held
+ * Return value:
+ *  < 0   Interrupt was ignored (masked or not delivered for other reasons)
+ *  = 0   Interrupt was coalesced (previous irq is still pending)
+ *  > 0   Number of CPUs interrupt was delivered to
+ */
+int kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level)
 {
-	unsigned long *irq_state = (unsigned long *)&kvm->arch.irq_states[irq];
+	struct kvm_kernel_irq_routing_entry *e;
+	unsigned long *irq_state, sig_level;
+	int ret = -1;
+
+	if (irq < KVM_IOAPIC_NUM_PINS) {
+		irq_state = (unsigned long *)&kvm->arch.irq_states[irq];
 
-	/* Logical OR for level trig interrupt */
-	if (level)
-		set_bit(irq_source_id, irq_state);
-	else
-		clear_bit(irq_source_id, irq_state);
+		/* Logical OR for level trig interrupt */
+		if (level)
+			set_bit(irq_source_id, irq_state);
+		else
+			clear_bit(irq_source_id, irq_state);
+		sig_level = !!(*irq_state);
+	} else /* Deal with MSI/MSI-X */
+		sig_level = 1;
 
 	/* Not possible to detect if the guest uses the PIC or the
 	 * IOAPIC.  So set the bit in both. The guest will ignore
 	 * writes to the unused one.
 	 */
-	kvm_ioapic_set_irq(kvm->arch.vioapic, irq, !!(*irq_state));
-#ifdef CONFIG_X86
-	kvm_pic_set_irq(pic_irqchip(kvm), irq, !!(*irq_state));
-#endif
+	list_for_each_entry(e, &kvm->irq_routing, link)
+		if (e->gsi == irq) {
+			int r = e->set(e, kvm, sig_level);
+			if (r < 0)
+				continue;
+
+			ret = r + ((ret < 0) ? 0 : ret);
+		}
+	return ret;
 }
 
-void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi)
+void kvm_notify_acked_irq(struct kvm *kvm, unsigned irqchip, unsigned pin)
 {
+	struct kvm_kernel_irq_routing_entry *e;
 	struct kvm_irq_ack_notifier *kian;
 	struct hlist_node *n;
+	unsigned gsi = pin;
+
+	list_for_each_entry(e, &kvm->irq_routing, link)
+		if (e->irqchip.irqchip == irqchip &&
+		    e->irqchip.pin == pin) {
+			gsi = e->gsi;
+			break;
+		}
 
 	hlist_for_each_entry(kian, n, &kvm->arch.irq_ack_notifier_list, link)
 		if (kian->gsi == gsi)
@@ -99,3 +196,177 @@ void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id)
 		clear_bit(irq_source_id, &kvm->arch.irq_states[i]);
 	clear_bit(irq_source_id, &kvm->arch.irq_sources_bitmap);
 }
+
+void kvm_register_irq_mask_notifier(struct kvm *kvm, int irq,
+				    struct kvm_irq_mask_notifier *kimn)
+{
+	kimn->irq = irq;
+	hlist_add_head(&kimn->link, &kvm->mask_notifier_list);
+}
+
+void kvm_unregister_irq_mask_notifier(struct kvm *kvm, int irq,
+				      struct kvm_irq_mask_notifier *kimn)
+{
+	hlist_del(&kimn->link);
+}
+
+void kvm_fire_mask_notifiers(struct kvm *kvm, int irq, bool mask)
+{
+	struct kvm_irq_mask_notifier *kimn;
+	struct hlist_node *n;
+
+	hlist_for_each_entry(kimn, n, &kvm->mask_notifier_list, link)
+		if (kimn->irq == irq)
+			kimn->func(kimn, mask);
+}
+
+static void __kvm_free_irq_routing(struct list_head *irq_routing)
+{
+	struct kvm_kernel_irq_routing_entry *e, *n;
+
+	list_for_each_entry_safe(e, n, irq_routing, link)
+		kfree(e);
+}
+
+void kvm_free_irq_routing(struct kvm *kvm)
+{
+	__kvm_free_irq_routing(&kvm->irq_routing);
+}
+
+static int setup_routing_entry(struct kvm_kernel_irq_routing_entry *e,
+			       const struct kvm_irq_routing_entry *ue)
+{
+	int r = -EINVAL;
+	int delta;
+
+	e->gsi = ue->gsi;
+	switch (ue->type) {
+	case KVM_IRQ_ROUTING_IRQCHIP:
+		delta = 0;
+		switch (ue->u.irqchip.irqchip) {
+		case KVM_IRQCHIP_PIC_MASTER:
+			e->set = kvm_set_pic_irq;
+			break;
+		case KVM_IRQCHIP_PIC_SLAVE:
+			e->set = kvm_set_pic_irq;
+			delta = 8;
+			break;
+		case KVM_IRQCHIP_IOAPIC:
+				e->set = kvm_set_ioapic_irq;
+			break;
+		default:
+			goto out;
+		}
+		e->irqchip.irqchip = ue->u.irqchip.irqchip;
+		e->irqchip.pin = ue->u.irqchip.pin + delta;
+		break;
+	case KVM_IRQ_ROUTING_MSI:
+		e->set = kvm_set_msi;
+		e->msi.address_lo = ue->u.msi.address_lo;
+		e->msi.address_hi = ue->u.msi.address_hi;
+		e->msi.data = ue->u.msi.data;
+		break;
+	default:
+		goto out;
+	}
+	r = 0;
+out:
+	return r;
+}
+
+
+int kvm_set_irq_routing(struct kvm *kvm,
+			const struct kvm_irq_routing_entry *ue,
+			unsigned nr,
+			unsigned flags)
+{
+	struct list_head irq_list = LIST_HEAD_INIT(irq_list);
+	struct list_head tmp = LIST_HEAD_INIT(tmp);
+	struct kvm_kernel_irq_routing_entry *e = NULL;
+	unsigned i;
+	int r;
+
+	for (i = 0; i < nr; ++i) {
+		r = -EINVAL;
+		if (ue->gsi >= KVM_MAX_IRQ_ROUTES)
+			goto out;
+		if (ue->flags)
+			goto out;
+		r = -ENOMEM;
+		e = kzalloc(sizeof(*e), GFP_KERNEL);
+		if (!e)
+			goto out;
+		r = setup_routing_entry(e, ue);
+		if (r)
+			goto out;
+		++ue;
+		list_add(&e->link, &irq_list);
+		e = NULL;
+	}
+
+	mutex_lock(&kvm->lock);
+	list_splice(&kvm->irq_routing, &tmp);
+	INIT_LIST_HEAD(&kvm->irq_routing);
+	list_splice(&irq_list, &kvm->irq_routing);
+	INIT_LIST_HEAD(&irq_list);
+	list_splice(&tmp, &irq_list);
+	mutex_unlock(&kvm->lock);
+
+	r = 0;
+
+out:
+	kfree(e);
+	__kvm_free_irq_routing(&irq_list);
+	return r;
+}
+
+#define IOAPIC_ROUTING_ENTRY(irq) \
+	{ .gsi = irq, .type = KVM_IRQ_ROUTING_IRQCHIP,	\
+	  .u.irqchip.irqchip = KVM_IRQCHIP_IOAPIC, .u.irqchip.pin = (irq) }
+#define ROUTING_ENTRY1(irq) IOAPIC_ROUTING_ENTRY(irq)
+
+#ifdef CONFIG_X86
+#  define PIC_ROUTING_ENTRY(irq) \
+	{ .gsi = irq, .type = KVM_IRQ_ROUTING_IRQCHIP,	\
+	  .u.irqchip.irqchip = SELECT_PIC(irq), .u.irqchip.pin = (irq) % 8 }
+#  define ROUTING_ENTRY2(irq) \
+	IOAPIC_ROUTING_ENTRY(irq), PIC_ROUTING_ENTRY(irq)
+#else
+#  define ROUTING_ENTRY2(irq) \
+	IOAPIC_ROUTING_ENTRY(irq)
+#endif
+
+static const struct kvm_irq_routing_entry default_routing[] = {
+	ROUTING_ENTRY2(0), ROUTING_ENTRY2(1),
+	ROUTING_ENTRY2(2), ROUTING_ENTRY2(3),
+	ROUTING_ENTRY2(4), ROUTING_ENTRY2(5),
+	ROUTING_ENTRY2(6), ROUTING_ENTRY2(7),
+	ROUTING_ENTRY2(8), ROUTING_ENTRY2(9),
+	ROUTING_ENTRY2(10), ROUTING_ENTRY2(11),
+	ROUTING_ENTRY2(12), ROUTING_ENTRY2(13),
+	ROUTING_ENTRY2(14), ROUTING_ENTRY2(15),
+	ROUTING_ENTRY1(16), ROUTING_ENTRY1(17),
+	ROUTING_ENTRY1(18), ROUTING_ENTRY1(19),
+	ROUTING_ENTRY1(20), ROUTING_ENTRY1(21),
+	ROUTING_ENTRY1(22), ROUTING_ENTRY1(23),
+#ifdef CONFIG_IA64
+	ROUTING_ENTRY1(24), ROUTING_ENTRY1(25),
+	ROUTING_ENTRY1(26), ROUTING_ENTRY1(27),
+	ROUTING_ENTRY1(28), ROUTING_ENTRY1(29),
+	ROUTING_ENTRY1(30), ROUTING_ENTRY1(31),
+	ROUTING_ENTRY1(32), ROUTING_ENTRY1(33),
+	ROUTING_ENTRY1(34), ROUTING_ENTRY1(35),
+	ROUTING_ENTRY1(36), ROUTING_ENTRY1(37),
+	ROUTING_ENTRY1(38), ROUTING_ENTRY1(39),
+	ROUTING_ENTRY1(40), ROUTING_ENTRY1(41),
+	ROUTING_ENTRY1(42), ROUTING_ENTRY1(43),
+	ROUTING_ENTRY1(44), ROUTING_ENTRY1(45),
+	ROUTING_ENTRY1(46), ROUTING_ENTRY1(47),
+#endif
+};
+
+int kvm_setup_default_irq_routing(struct kvm *kvm)
+{
+	return kvm_set_irq_routing(kvm, default_routing,
+				   ARRAY_SIZE(default_routing), 0);
+}

virt/kvm/kvm_main.c

@@ -47,10 +47,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-#ifdef CONFIG_X86
-#include <asm/msidef.h>
-#endif
-
 #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
 #include "coalesced_mmio.h"
 #endif
@@ -85,57 +81,6 @@ static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl,
 static bool kvm_rebooting;
 
 #ifdef KVM_CAP_DEVICE_ASSIGNMENT
-
-#ifdef CONFIG_X86
-static void assigned_device_msi_dispatch(struct kvm_assigned_dev_kernel *dev)
-{
-	int vcpu_id;
-	struct kvm_vcpu *vcpu;
-	struct kvm_ioapic *ioapic = ioapic_irqchip(dev->kvm);
-	int dest_id = (dev->guest_msi.address_lo & MSI_ADDR_DEST_ID_MASK)
-			>> MSI_ADDR_DEST_ID_SHIFT;
-	int vector = (dev->guest_msi.data & MSI_DATA_VECTOR_MASK)
-			>> MSI_DATA_VECTOR_SHIFT;
-	int dest_mode = test_bit(MSI_ADDR_DEST_MODE_SHIFT,
-				(unsigned long *)&dev->guest_msi.address_lo);
-	int trig_mode = test_bit(MSI_DATA_TRIGGER_SHIFT,
-				(unsigned long *)&dev->guest_msi.data);
-	int delivery_mode = test_bit(MSI_DATA_DELIVERY_MODE_SHIFT,
-				(unsigned long *)&dev->guest_msi.data);
-	u32 deliver_bitmask;
-
-	BUG_ON(!ioapic);
-
-	deliver_bitmask = kvm_ioapic_get_delivery_bitmask(ioapic,
-				dest_id, dest_mode);
-	/* IOAPIC delivery mode value is the same as MSI here */
-	switch (delivery_mode) {
-	case IOAPIC_LOWEST_PRIORITY:
-		vcpu = kvm_get_lowest_prio_vcpu(ioapic->kvm, vector,
-				deliver_bitmask);
-		if (vcpu != NULL)
-			kvm_apic_set_irq(vcpu, vector, trig_mode);
-		else
-			printk(KERN_INFO "kvm: null lowest priority vcpu!\n");
-		break;
-	case IOAPIC_FIXED:
-		for (vcpu_id = 0; deliver_bitmask != 0; vcpu_id++) {
-			if (!(deliver_bitmask & (1 << vcpu_id)))
-				continue;
-			deliver_bitmask &= ~(1 << vcpu_id);
-			vcpu = ioapic->kvm->vcpus[vcpu_id];
-			if (vcpu)
-				kvm_apic_set_irq(vcpu, vector, trig_mode);
-		}
-		break;
-	default:
-		printk(KERN_INFO "kvm: unsupported MSI delivery mode\n");
-	}
-}
-#else
-static void assigned_device_msi_dispatch(struct kvm_assigned_dev_kernel *dev) {}
-#endif
-
 static struct kvm_assigned_dev_kernel *kvm_find_assigned_dev(struct list_head *head,
 						      int assigned_dev_id)
 {
@@ -162,13 +107,10 @@ static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
 	 * finer-grained lock, update this
 	 */
 	mutex_lock(&assigned_dev->kvm->lock);
-	if (assigned_dev->irq_requested_type & KVM_ASSIGNED_DEV_GUEST_INTX)
-		kvm_set_irq(assigned_dev->kvm,
-			    assigned_dev->irq_source_id,
-			    assigned_dev->guest_irq, 1);
-	else if (assigned_dev->irq_requested_type &
-				KVM_ASSIGNED_DEV_GUEST_MSI) {
-		assigned_device_msi_dispatch(assigned_dev);
+	kvm_set_irq(assigned_dev->kvm, assigned_dev->irq_source_id,
+		    assigned_dev->guest_irq, 1);
+
+	if (assigned_dev->irq_requested_type & KVM_ASSIGNED_DEV_GUEST_MSI) {
 		enable_irq(assigned_dev->host_irq);
 		assigned_dev->host_irq_disabled = false;
 	}
@@ -331,18 +273,24 @@ static int assigned_device_update_msi(struct kvm *kvm,
 {
 	int r;
 
+	adev->guest_irq = airq->guest_irq;
 	if (airq->flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSI) {
 		/* x86 don't care upper address of guest msi message addr */
 		adev->irq_requested_type |= KVM_ASSIGNED_DEV_GUEST_MSI;
 		adev->irq_requested_type &= ~KVM_ASSIGNED_DEV_GUEST_INTX;
-		adev->guest_msi.address_lo = airq->guest_msi.addr_lo;
-		adev->guest_msi.data = airq->guest_msi.data;
 		adev->ack_notifier.gsi = -1;
 	} else if (msi2intx) {
 		adev->irq_requested_type |= KVM_ASSIGNED_DEV_GUEST_INTX;
 		adev->irq_requested_type &= ~KVM_ASSIGNED_DEV_GUEST_MSI;
-		adev->guest_irq = airq->guest_irq;
 		adev->ack_notifier.gsi = airq->guest_irq;
+	} else {
+		/*
+		 * Guest require to disable device MSI, we disable MSI and
+		 * re-enable INTx by default again. Notice it's only for
+		 * non-msi2intx.
+		 */
+		assigned_device_update_intx(kvm, adev, airq);
+		return 0;
 	}
 
 	if (adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI)
@@ -379,6 +327,7 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 {
 	int r = 0;
 	struct kvm_assigned_dev_kernel *match;
+	u32 current_flags = 0, changed_flags;
 
 	mutex_lock(&kvm->lock);
 
@@ -416,8 +365,13 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 		}
 	}
 
-	if ((!msi2intx &&
-	     (assigned_irq->flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSI)) ||
+	if ((match->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI) &&
+		 (match->irq_requested_type & KVM_ASSIGNED_DEV_GUEST_MSI))
+		current_flags |= KVM_DEV_IRQ_ASSIGN_ENABLE_MSI;
+
+	changed_flags = assigned_irq->flags ^ current_flags;
+
+	if ((changed_flags & KVM_DEV_IRQ_ASSIGN_MSI_ACTION) ||
 	    (msi2intx && match->dev->msi_enabled)) {
 #ifdef CONFIG_X86
 		r = assigned_device_update_msi(kvm, match, assigned_irq);
@@ -563,7 +517,7 @@ static int kvm_vm_ioctl_deassign_device(struct kvm *kvm,
 		goto out;
 	}
 
-	if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU)
+	if (match->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU)
 		kvm_deassign_device(kvm, match);
 
 	kvm_free_assigned_device(kvm, match);
@@ -581,8 +535,10 @@ static inline int valid_vcpu(int n)
 
 inline int kvm_is_mmio_pfn(pfn_t pfn)
 {
-	if (pfn_valid(pfn))
-		return PageReserved(pfn_to_page(pfn));
+	if (pfn_valid(pfn)) {
+		struct page *page = compound_head(pfn_to_page(pfn));
+		return PageReserved(page);
+	}
 
 	return true;
 }
@@ -828,6 +784,10 @@ static struct kvm *kvm_create_vm(void)
 
 	if (IS_ERR(kvm))
 		goto out;
+#ifdef CONFIG_HAVE_KVM_IRQCHIP
+	INIT_LIST_HEAD(&kvm->irq_routing);
+	INIT_HLIST_HEAD(&kvm->mask_notifier_list);
+#endif
 
 #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
 	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
@@ -909,6 +869,7 @@ static void kvm_destroy_vm(struct kvm *kvm)
 	spin_lock(&kvm_lock);
 	list_del(&kvm->vm_list);
 	spin_unlock(&kvm_lock);
+	kvm_free_irq_routing(kvm);
 	kvm_io_bus_destroy(&kvm->pio_bus);
 	kvm_io_bus_destroy(&kvm->mmio_bus);
 #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
@@ -1755,13 +1716,13 @@ out_free2:
 		r = 0;
 		break;
 	}
-	case KVM_DEBUG_GUEST: {
-		struct kvm_debug_guest dbg;
+	case KVM_SET_GUEST_DEBUG: {
+		struct kvm_guest_debug dbg;
 
 		r = -EFAULT;
 		if (copy_from_user(&dbg, argp, sizeof dbg))
 			goto out;
-		r = kvm_arch_vcpu_ioctl_debug_guest(vcpu, &dbg);
+		r = kvm_arch_vcpu_ioctl_set_guest_debug(vcpu, &dbg);
 		if (r)
 			goto out;
 		r = 0;
@@ -1928,6 +1889,36 @@ static long kvm_vm_ioctl(struct file *filp,
 			goto out;
 		break;
 	}
+#endif
+#ifdef KVM_CAP_IRQ_ROUTING
+	case KVM_SET_GSI_ROUTING: {
+		struct kvm_irq_routing routing;
+		struct kvm_irq_routing __user *urouting;
+		struct kvm_irq_routing_entry *entries;
+
+		r = -EFAULT;
+		if (copy_from_user(&routing, argp, sizeof(routing)))
+			goto out;
+		r = -EINVAL;
+		if (routing.nr >= KVM_MAX_IRQ_ROUTES)
+			goto out;
+		if (routing.flags)
+			goto out;
+		r = -ENOMEM;
+		entries = vmalloc(routing.nr * sizeof(*entries));
+		if (!entries)
+			goto out;
+		r = -EFAULT;
+		urouting = argp;
+		if (copy_from_user(entries, urouting->entries,
+				   routing.nr * sizeof(*entries)))
+			goto out_free_irq_routing;
+		r = kvm_set_irq_routing(kvm, entries, routing.nr,
+					routing.flags);
+	out_free_irq_routing:
+		vfree(entries);
+		break;
+	}
 #endif
 	default:
 		r = kvm_arch_vm_ioctl(filp, ioctl, arg);
@@ -1995,6 +1986,10 @@ static long kvm_dev_ioctl_check_extension_generic(long arg)
 	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_DESTROY_MEMORY_REGION_WORKS:
 		return 1;
+#ifdef CONFIG_HAVE_KVM_IRQCHIP
+	case KVM_CAP_IRQ_ROUTING:
+		return KVM_MAX_IRQ_ROUTES;
+#endif
 	default:
 		break;
 	}