Merge remote-tracking branch 'kvm/next' into queue

Conflicts:
	arch/arm64/include/asm/kvm_host.h
	virt/kvm/arm/vgic.c

Documentation/virtual/kvm/api.txt

@@ -2565,6 +2565,120 @@ associated with the service will be forgotten, and subsequent RTAS
 calls by the guest for that service will be passed to userspace to be
 handled.
 
+4.87 KVM_SET_GUEST_DEBUG
+
+Capability: KVM_CAP_SET_GUEST_DEBUG
+Architectures: x86, s390, ppc
+Type: vcpu ioctl
+Parameters: struct kvm_guest_debug (in)
+Returns: 0 on success; -1 on error
+
+struct kvm_guest_debug {
+       __u32 control;
+       __u32 pad;
+       struct kvm_guest_debug_arch arch;
+};
+
+Set up the processor specific debug registers and configure vcpu for
+handling guest debug events. There are two parts to the structure, the
+first a control bitfield indicates the type of debug events to handle
+when running. Common control bits are:
+
+  - KVM_GUESTDBG_ENABLE:        guest debugging is enabled
+  - KVM_GUESTDBG_SINGLESTEP:    the next run should single-step
+
+The top 16 bits of the control field are architecture specific control
+flags which can include the following:
+
+  - KVM_GUESTDBG_USE_SW_BP:     using software breakpoints [x86]
+  - KVM_GUESTDBG_USE_HW_BP:     using hardware breakpoints [x86, s390]
+  - KVM_GUESTDBG_INJECT_DB:     inject DB type exception [x86]
+  - KVM_GUESTDBG_INJECT_BP:     inject BP type exception [x86]
+  - KVM_GUESTDBG_EXIT_PENDING:  trigger an immediate guest exit [s390]
+
+For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
+are enabled in memory so we need to ensure breakpoint exceptions are
+correctly trapped and the KVM run loop exits at the breakpoint and not
+running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
+we need to ensure the guest vCPUs architecture specific registers are
+updated to the correct (supplied) values.
+
+The second part of the structure is architecture specific and
+typically contains a set of debug registers.
+
+When debug events exit the main run loop with the reason
+KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
+structure containing architecture specific debug information.
+
+4.88 KVM_GET_EMULATED_CPUID
+
+Capability: KVM_CAP_EXT_EMUL_CPUID
+Architectures: x86
+Type: system ioctl
+Parameters: struct kvm_cpuid2 (in/out)
+Returns: 0 on success, -1 on error
+
+struct kvm_cpuid2 {
+	__u32 nent;
+	__u32 flags;
+	struct kvm_cpuid_entry2 entries[0];
+};
+
+The member 'flags' is used for passing flags from userspace.
+
+#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX		BIT(0)
+#define KVM_CPUID_FLAG_STATEFUL_FUNC		BIT(1)
+#define KVM_CPUID_FLAG_STATE_READ_NEXT		BIT(2)
+
+struct kvm_cpuid_entry2 {
+	__u32 function;
+	__u32 index;
+	__u32 flags;
+	__u32 eax;
+	__u32 ebx;
+	__u32 ecx;
+	__u32 edx;
+	__u32 padding[3];
+};
+
+This ioctl returns x86 cpuid features which are emulated by
+kvm.Userspace can use the information returned by this ioctl to query
+which features are emulated by kvm instead of being present natively.
+
+Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
+structure with the 'nent' field indicating the number of entries in
+the variable-size array 'entries'. If the number of entries is too low
+to describe the cpu capabilities, an error (E2BIG) is returned. If the
+number is too high, the 'nent' field is adjusted and an error (ENOMEM)
+is returned. If the number is just right, the 'nent' field is adjusted
+to the number of valid entries in the 'entries' array, which is then
+filled.
+
+The entries returned are the set CPUID bits of the respective features
+which kvm emulates, as returned by the CPUID instruction, with unknown
+or unsupported feature bits cleared.
+
+Features like x2apic, for example, may not be present in the host cpu
+but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
+emulated efficiently and thus not included here.
+
+The fields in each entry are defined as follows:
+
+  function: the eax value used to obtain the entry
+  index: the ecx value used to obtain the entry (for entries that are
+         affected by ecx)
+  flags: an OR of zero or more of the following:
+        KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
+           if the index field is valid
+        KVM_CPUID_FLAG_STATEFUL_FUNC:
+           if cpuid for this function returns different values for successive
+           invocations; there will be several entries with the same function,
+           all with this flag set
+        KVM_CPUID_FLAG_STATE_READ_NEXT:
+           for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
+           the first entry to be read by a cpu
+   eax, ebx, ecx, edx: the values returned by the cpuid instruction for
+         this function/index combination
 
 5. The kvm_run structure
 ------------------------
@@ -2861,78 +2975,12 @@ kvm_valid_regs for specific bits. These bits are architecture specific
 and usually define the validity of a groups of registers. (e.g. one bit
  for general purpose registers)
 
-};
-
+Please note that the kernel is allowed to use the kvm_run structure as the
+primary storage for certain register types. Therefore, the kernel may use the
+values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
 
-4.81 KVM_GET_EMULATED_CPUID
-
-Capability: KVM_CAP_EXT_EMUL_CPUID
-Architectures: x86
-Type: system ioctl
-Parameters: struct kvm_cpuid2 (in/out)
-Returns: 0 on success, -1 on error
-
-struct kvm_cpuid2 {
-	__u32 nent;
-	__u32 flags;
-	struct kvm_cpuid_entry2 entries[0];
 };
 
-The member 'flags' is used for passing flags from userspace.
-
-#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX		BIT(0)
-#define KVM_CPUID_FLAG_STATEFUL_FUNC		BIT(1)
-#define KVM_CPUID_FLAG_STATE_READ_NEXT		BIT(2)
-
-struct kvm_cpuid_entry2 {
-	__u32 function;
-	__u32 index;
-	__u32 flags;
-	__u32 eax;
-	__u32 ebx;
-	__u32 ecx;
-	__u32 edx;
-	__u32 padding[3];
-};
-
-This ioctl returns x86 cpuid features which are emulated by
-kvm.Userspace can use the information returned by this ioctl to query
-which features are emulated by kvm instead of being present natively.
-
-Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
-structure with the 'nent' field indicating the number of entries in
-the variable-size array 'entries'. If the number of entries is too low
-to describe the cpu capabilities, an error (E2BIG) is returned. If the
-number is too high, the 'nent' field is adjusted and an error (ENOMEM)
-is returned. If the number is just right, the 'nent' field is adjusted
-to the number of valid entries in the 'entries' array, which is then
-filled.
-
-The entries returned are the set CPUID bits of the respective features
-which kvm emulates, as returned by the CPUID instruction, with unknown
-or unsupported feature bits cleared.
-
-Features like x2apic, for example, may not be present in the host cpu
-but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
-emulated efficiently and thus not included here.
-
-The fields in each entry are defined as follows:
-
-  function: the eax value used to obtain the entry
-  index: the ecx value used to obtain the entry (for entries that are
-         affected by ecx)
-  flags: an OR of zero or more of the following:
-        KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
-           if the index field is valid
-        KVM_CPUID_FLAG_STATEFUL_FUNC:
-           if cpuid for this function returns different values for successive
-           invocations; there will be several entries with the same function,
-           all with this flag set
-        KVM_CPUID_FLAG_STATE_READ_NEXT:
-           for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
-           the first entry to be read by a cpu
-   eax, ebx, ecx, edx: the values returned by the cpuid instruction for
-         this function/index combination
 
 
 6. Capabilities that can be enabled on vCPUs

Documentation/virtual/kvm/mmu.txt

@@ -425,6 +425,20 @@ fault through the slow path.
 Since only 19 bits are used to store generation-number on mmio spte, all
 pages are zapped when there is an overflow.
 
+Unfortunately, a single memory access might access kvm_memslots(kvm) multiple
+times, the last one happening when the generation number is retrieved and
+stored into the MMIO spte.  Thus, the MMIO spte might be created based on
+out-of-date information, but with an up-to-date generation number.
+
+To avoid this, the generation number is incremented again after synchronize_srcu
+returns; thus, the low bit of kvm_memslots(kvm)->generation is only 1 during a
+memslot update, while some SRCU readers might be using the old copy.  We do not
+want to use an MMIO sptes created with an odd generation number, and we can do
+this without losing a bit in the MMIO spte.  The low bit of the generation
+is not stored in MMIO spte, and presumed zero when it is extracted out of the
+spte.  If KVM is unlucky and creates an MMIO spte while the low bit is 1,
+the next access to the spte will always be a cache miss.
+
 
 Further reading
 ===============

arch/arm/include/asm/kvm_host.h

@@ -19,6 +19,8 @@
 #ifndef __ARM_KVM_HOST_H__
 #define __ARM_KVM_HOST_H__
 
+#include <linux/types.h>
+#include <linux/kvm_types.h>
 #include <asm/kvm.h>
 #include <asm/kvm_asm.h>
 #include <asm/kvm_mmio.h>
@@ -40,7 +42,6 @@
 
 #include <kvm/arm_vgic.h>
 
-struct kvm_vcpu;
 u32 *kvm_vcpu_reg(struct kvm_vcpu *vcpu, u8 reg_num, u32 mode);
 int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
@@ -149,20 +150,17 @@ struct kvm_vcpu_stat {
 	u32 halt_wakeup;
 };
 
-struct kvm_vcpu_init;
 int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 			const struct kvm_vcpu_init *init);
 int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
 unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
 int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
-struct kvm_one_reg;
 int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
 int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
 u64 kvm_call_hyp(void *hypfn, ...);
 void force_vm_exit(const cpumask_t *mask);
 
 #define KVM_ARCH_WANT_MMU_NOTIFIER
-struct kvm;
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_unmap_hva_range(struct kvm *kvm,
 			unsigned long start, unsigned long end);
@@ -187,7 +185,6 @@ struct kvm_vcpu __percpu **kvm_get_running_vcpus(void);
 
 int kvm_arm_copy_coproc_indices(struct kvm_vcpu *vcpu, u64 __user *uindices);
 unsigned long kvm_arm_num_coproc_regs(struct kvm_vcpu *vcpu);
-struct kvm_one_reg;
 int kvm_arm_coproc_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *);
 int kvm_arm_coproc_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *);
 
@@ -233,4 +230,10 @@ static inline void vgic_arch_setup(const struct vgic_params *vgic)
 int kvm_perf_init(void);
 int kvm_perf_teardown(void);
 
+static inline void kvm_arch_hardware_disable(void) {}
+static inline void kvm_arch_hardware_unsetup(void) {}
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
+
 #endif /* __ARM_KVM_HOST_H__ */

arch/arm/kvm/arm.c

@@ -87,7 +87,7 @@ struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void)
 	return &kvm_arm_running_vcpu;
 }
 
-int kvm_arch_hardware_enable(void *garbage)
+int kvm_arch_hardware_enable(void)
 {
 	return 0;
 }
@@ -97,27 +97,16 @@ int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
 	return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE;
 }
 
-void kvm_arch_hardware_disable(void *garbage)
-{
-}
-
 int kvm_arch_hardware_setup(void)
 {
 	return 0;
 }
 
-void kvm_arch_hardware_unsetup(void)
-{
-}
-
 void kvm_arch_check_processor_compat(void *rtn)
 {
 	*(int *)rtn = 0;
 }
 
-void kvm_arch_sync_events(struct kvm *kvm)
-{
-}
 
 /**
  * kvm_arch_init_vm - initializes a VM data structure
@@ -285,14 +274,6 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
-{
-}
-
-void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu)
-{
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	vcpu->cpu = cpu;

arch/arm64/include/asm/kvm_host.h

@@ -22,6 +22,8 @@
 #ifndef __ARM64_KVM_HOST_H__
 #define __ARM64_KVM_HOST_H__
 
+#include <linux/types.h>
+#include <linux/kvm_types.h>
 #include <asm/kvm.h>
 #include <asm/kvm_asm.h>
 #include <asm/kvm_mmio.h>
@@ -41,7 +43,6 @@
 
 #define KVM_VCPU_MAX_FEATURES 3
 
-struct kvm_vcpu;
 int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
 int kvm_arch_dev_ioctl_check_extension(long ext);
@@ -164,18 +165,15 @@ struct kvm_vcpu_stat {
 	u32 halt_wakeup;
 };
 
-struct kvm_vcpu_init;
 int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 			const struct kvm_vcpu_init *init);
 int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
 unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
 int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
-struct kvm_one_reg;
 int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
 int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
 
 #define KVM_ARCH_WANT_MMU_NOTIFIER
-struct kvm;
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_unmap_hva_range(struct kvm *kvm,
 			unsigned long start, unsigned long end);
@@ -244,4 +242,10 @@ static inline void vgic_arch_setup(const struct vgic_params *vgic)
 	}
 }
 
+static inline void kvm_arch_hardware_disable(void) {}
+static inline void kvm_arch_hardware_unsetup(void) {}
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
+
 #endif /* __ARM64_KVM_HOST_H__ */

arch/ia64/include/asm/kvm_host.h

@@ -234,9 +234,6 @@ struct kvm_vm_data {
 #define KVM_REQ_PTC_G		32
 #define KVM_REQ_RESUME		33
 
-struct kvm;
-struct kvm_vcpu;
-
 struct kvm_mmio_req {
 	uint64_t addr;          /*  physical address		*/
 	uint64_t size;          /*  size in bytes		*/
@@ -595,6 +592,18 @@ void kvm_sal_emul(struct kvm_vcpu *vcpu);
 struct kvm *kvm_arch_alloc_vm(void);
 void kvm_arch_free_vm(struct kvm *kvm);
 
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu) {}
+static inline void kvm_arch_free_memslot(struct kvm *kvm,
+		struct kvm_memory_slot *free, struct kvm_memory_slot *dont) {}
+static inline void kvm_arch_memslots_updated(struct kvm *kvm) {}
+static inline void kvm_arch_commit_memory_region(struct kvm *kvm,
+		struct kvm_userspace_memory_region *mem,
+		const struct kvm_memory_slot *old,
+		enum kvm_mr_change change) {}
+static inline void kvm_arch_hardware_unsetup(void) {}
+
 #endif /* __ASSEMBLY__*/
 
 #endif

arch/ia64/kvm/kvm-ia64.c

@@ -125,7 +125,7 @@ long ia64_pal_vp_create(u64 *vpd, u64 *host_iva, u64 *opt_handler)
 
 static  DEFINE_SPINLOCK(vp_lock);
 
-int kvm_arch_hardware_enable(void *garbage)
+int kvm_arch_hardware_enable(void)
 {
 	long  status;
 	long  tmp_base;
@@ -160,7 +160,7 @@ int kvm_arch_hardware_enable(void *garbage)
 	return 0;
 }
 
-void kvm_arch_hardware_disable(void *garbage)
+void kvm_arch_hardware_disable(void)
 {
 
 	long status;
@@ -1364,10 +1364,6 @@ static void kvm_release_vm_pages(struct kvm *kvm)
 	}
 }
 
-void kvm_arch_sync_events(struct kvm *kvm)
-{
-}
-
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
 	kvm_iommu_unmap_guest(kvm);
@@ -1376,10 +1372,6 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kvm_release_vm_pages(kvm);
 }
 
-void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
-{
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	if (cpu != vcpu->cpu) {
@@ -1468,7 +1460,6 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 	kfree(vcpu->arch.apic);
 }
 
-
 long kvm_arch_vcpu_ioctl(struct file *filp,
 			 unsigned int ioctl, unsigned long arg)
 {
@@ -1551,21 +1542,12 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
-			   struct kvm_memory_slot *dont)
-{
-}
-
 int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 			    unsigned long npages)
 {
 	return 0;
 }
 
-void kvm_arch_memslots_updated(struct kvm *kvm)
-{
-}
-
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 		struct kvm_memory_slot *memslot,
 		struct kvm_userspace_memory_region *mem,
@@ -1597,14 +1579,6 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 	return 0;
 }
 
-void kvm_arch_commit_memory_region(struct kvm *kvm,
-		struct kvm_userspace_memory_region *mem,
-		const struct kvm_memory_slot *old,
-		enum kvm_mr_change change)
-{
-	return;
-}
-
 void kvm_arch_flush_shadow_all(struct kvm *kvm)
 {
 	kvm_flush_remote_tlbs(kvm);
@@ -1853,10 +1827,6 @@ int kvm_arch_hardware_setup(void)
 	return 0;
 }
 
-void kvm_arch_hardware_unsetup(void)
-{
-}
-
 int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq)
 {
 	return __apic_accept_irq(vcpu, irq->vector);

arch/mips/include/asm/kvm_host.h

@@ -96,11 +96,6 @@
 #define CAUSEB_DC			27
 #define CAUSEF_DC			(_ULCAST_(1) << 27)
 
-struct kvm;
-struct kvm_run;
-struct kvm_vcpu;
-struct kvm_interrupt;
-
 extern atomic_t kvm_mips_instance;
 extern pfn_t(*kvm_mips_gfn_to_pfn) (struct kvm *kvm, gfn_t gfn);
 extern void (*kvm_mips_release_pfn_clean) (pfn_t pfn);
@@ -767,5 +762,16 @@ extern int kvm_mips_trans_mtc0(uint32_t inst, uint32_t *opc,
 extern void kvm_mips_dump_stats(struct kvm_vcpu *vcpu);
 extern unsigned long kvm_mips_get_ramsize(struct kvm *kvm);
 
+static inline void kvm_arch_hardware_disable(void) {}
+static inline void kvm_arch_hardware_unsetup(void) {}
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_free_memslot(struct kvm *kvm,
+		struct kvm_memory_slot *free, struct kvm_memory_slot *dont) {}
+static inline void kvm_arch_memslots_updated(struct kvm *kvm) {}
+static inline void kvm_arch_flush_shadow_all(struct kvm *kvm) {}
+static inline void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
+		struct kvm_memory_slot *slot) {}
+static inline void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
 
 #endif /* __MIPS_KVM_HOST_H__ */

arch/mips/kvm/mips.c

@@ -77,24 +77,16 @@ int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
 	return 1;
 }
 
-int kvm_arch_hardware_enable(void *garbage)
+int kvm_arch_hardware_enable(void)
 {
 	return 0;
 }
 
-void kvm_arch_hardware_disable(void *garbage)
-{
-}
-
 int kvm_arch_hardware_setup(void)
 {
 	return 0;
 }
 
-void kvm_arch_hardware_unsetup(void)
-{
-}
-
 void kvm_arch_check_processor_compat(void *rtn)
 {
 	*(int *)rtn = 0;
@@ -163,10 +155,6 @@ void kvm_mips_free_vcpus(struct kvm *kvm)
 	mutex_unlock(&kvm->lock);
 }
 
-void kvm_arch_sync_events(struct kvm *kvm)
-{
-}
-
 static void kvm_mips_uninit_tlbs(void *arg)
 {
 	/* Restore wired count */
@@ -194,21 +182,12 @@ long kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl,
 	return -ENOIOCTLCMD;
 }
 
-void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
-			   struct kvm_memory_slot *dont)
-{
-}
-
 int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 			    unsigned long npages)
 {
 	return 0;
 }
 
-void kvm_arch_memslots_updated(struct kvm *kvm)
-{
-}
-
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				   struct kvm_memory_slot *memslot,
 				   struct kvm_userspace_memory_region *mem,
@@ -254,19 +233,6 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 	}
 }
 
-void kvm_arch_flush_shadow_all(struct kvm *kvm)
-{
-}
-
-void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
-				   struct kvm_memory_slot *slot)
-{
-}
-
-void kvm_arch_flush_shadow(struct kvm *kvm)
-{
-}
-
 struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
 {
 	int err, size, offset;
@@ -998,14 +964,6 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
-{
-}
-
-void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu)
-{
-}
-
 int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
 				  struct kvm_translation *tr)
 {

arch/powerpc/include/asm/kvm_host.h

@@ -53,7 +53,6 @@
 
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 
-struct kvm;
 extern int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 extern int kvm_unmap_hva_range(struct kvm *kvm,
 			       unsigned long start, unsigned long end);
@@ -76,10 +75,6 @@ extern void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 /* Physical Address Mask - allowed range of real mode RAM access */
 #define KVM_PAM			0x0fffffffffffffffULL
 
-struct kvm;
-struct kvm_run;
-struct kvm_vcpu;
-
 struct lppaca;
 struct slb_shadow;
 struct dtl_entry;
@@ -687,4 +682,12 @@ struct kvm_vcpu_arch {
 #define __KVM_HAVE_ARCH_WQP
 #define __KVM_HAVE_CREATE_DEVICE
 
+static inline void kvm_arch_hardware_disable(void) {}
+static inline void kvm_arch_hardware_unsetup(void) {}
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_memslots_updated(struct kvm *kvm) {}
+static inline void kvm_arch_flush_shadow_all(struct kvm *kvm) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
+static inline void kvm_arch_exit(void) {}
+
 #endif /* __POWERPC_KVM_HOST_H__ */

arch/powerpc/kvm/powerpc.c

@@ -384,24 +384,16 @@ int kvmppc_ld(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
 }
 EXPORT_SYMBOL_GPL(kvmppc_ld);
 
-int kvm_arch_hardware_enable(void *garbage)
+int kvm_arch_hardware_enable(void)
 {
 	return 0;
 }
 
-void kvm_arch_hardware_disable(void *garbage)
-{
-}
-
 int kvm_arch_hardware_setup(void)
 {
 	return 0;
 }
 
-void kvm_arch_hardware_unsetup(void)
-{
-}
-
 void kvm_arch_check_processor_compat(void *rtn)
 {
 	*(int *)rtn = kvmppc_core_check_processor_compat();
@@ -462,10 +454,6 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	module_put(kvm->arch.kvm_ops->owner);
 }
 
-void kvm_arch_sync_events(struct kvm *kvm)
-{
-}
-
 int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 {
 	int r;
@@ -608,10 +596,6 @@ int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 	return kvmppc_core_create_memslot(kvm, slot, npages);
 }
 
-void kvm_arch_memslots_updated(struct kvm *kvm)
-{
-}
-
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				   struct kvm_memory_slot *memslot,
 				   struct kvm_userspace_memory_region *mem,
@@ -628,10 +612,6 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 	kvmppc_core_commit_memory_region(kvm, mem, old);
 }
 
-void kvm_arch_flush_shadow_all(struct kvm *kvm)
-{
-}
-
 void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
 				   struct kvm_memory_slot *slot)
 {
@@ -720,10 +700,6 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 	kvmppc_subarch_vcpu_uninit(vcpu);
 }
 
-void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu)
-{
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 #ifdef CONFIG_BOOKE
@@ -1347,9 +1323,4 @@ int kvm_arch_init(void *opaque)
 	return 0;
 }
 
-void kvm_arch_exit(void)
-{
-
-}
-
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ppc_instr);

arch/s390/include/asm/kvm_host.h

@@ -13,8 +13,11 @@
 
 #ifndef ASM_KVM_HOST_H
 #define ASM_KVM_HOST_H
+
+#include <linux/types.h>
 #include <linux/hrtimer.h>
 #include <linux/interrupt.h>
+#include <linux/kvm_types.h>
 #include <linux/kvm_host.h>
 #include <linux/kvm.h>
 #include <asm/debug.h>
@@ -154,7 +157,9 @@ struct kvm_s390_sie_block {
 	__u8	armid;			/* 0x00e3 */
 	__u8	reservede4[4];		/* 0x00e4 */
 	__u64	tecmc;			/* 0x00e8 */
-	__u8	reservedf0[16];		/* 0x00f0 */
+	__u8	reservedf0[12];		/* 0x00f0 */
+#define CRYCB_FORMAT1 0x00000001
+	__u32	crycbd;			/* 0x00fc */
 	__u64	gcr[16];		/* 0x0100 */
 	__u64	gbea;			/* 0x0180 */
 	__u8	reserved188[24];	/* 0x0188 */
@@ -407,6 +412,15 @@ struct s390_io_adapter {
 #define MAX_S390_IO_ADAPTERS ((MAX_ISC + 1) * 8)
 #define MAX_S390_ADAPTER_MAPS 256
 
+struct kvm_s390_crypto {
+	struct kvm_s390_crypto_cb *crycb;
+	__u32 crycbd;
+};
+
+struct kvm_s390_crypto_cb {
+	__u8    reserved00[128];                /* 0x0000 */
+};
+
 struct kvm_arch{
 	struct sca_block *sca;
 	debug_info_t *dbf;
@@ -420,6 +434,7 @@ struct kvm_arch{
 	struct s390_io_adapter *adapters[MAX_S390_IO_ADAPTERS];
 	wait_queue_head_t ipte_wq;
 	spinlock_t start_stop_lock;
+	struct kvm_s390_crypto crypto;
 };
 
 #define KVM_HVA_ERR_BAD		(-1UL)
@@ -431,8 +446,6 @@ static inline bool kvm_is_error_hva(unsigned long addr)
 }
 
 #define ASYNC_PF_PER_VCPU	64
-struct kvm_vcpu;
-struct kvm_async_pf;
 struct kvm_arch_async_pf {
 	unsigned long pfault_token;
 };
@@ -450,4 +463,18 @@ void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
 
 extern int sie64a(struct kvm_s390_sie_block *, u64 *);
 extern char sie_exit;
+
+static inline void kvm_arch_hardware_disable(void) {}
+static inline void kvm_arch_check_processor_compat(void *rtn) {}
+static inline void kvm_arch_exit(void) {}
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
+static inline void kvm_arch_free_memslot(struct kvm *kvm,
+		struct kvm_memory_slot *free, struct kvm_memory_slot *dont) {}
+static inline void kvm_arch_memslots_updated(struct kvm *kvm) {}
+static inline void kvm_arch_flush_shadow_all(struct kvm *kvm) {}
+static inline void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
+		struct kvm_memory_slot *slot) {}
+
 #endif

arch/s390/include/asm/pgalloc.h

@@ -18,9 +18,9 @@
 unsigned long *crst_table_alloc(struct mm_struct *);
 void crst_table_free(struct mm_struct *, unsigned long *);
 
-unsigned long *page_table_alloc(struct mm_struct *, unsigned long);
+unsigned long *page_table_alloc(struct mm_struct *);
 void page_table_free(struct mm_struct *, unsigned long *);
-void page_table_free_rcu(struct mmu_gather *, unsigned long *);
+void page_table_free_rcu(struct mmu_gather *, unsigned long *, unsigned long);
 
 void page_table_reset_pgste(struct mm_struct *, unsigned long, unsigned long,
 			    bool init_skey);
@@ -145,8 +145,8 @@ static inline void pmd_populate(struct mm_struct *mm,
 /*
  * page table entry allocation/free routines.
  */
-#define pte_alloc_one_kernel(mm, vmaddr) ((pte_t *) page_table_alloc(mm, vmaddr))
-#define pte_alloc_one(mm, vmaddr) ((pte_t *) page_table_alloc(mm, vmaddr))
+#define pte_alloc_one_kernel(mm, vmaddr) ((pte_t *) page_table_alloc(mm))
+#define pte_alloc_one(mm, vmaddr) ((pte_t *) page_table_alloc(mm))
 
 #define pte_free_kernel(mm, pte) page_table_free(mm, (unsigned long *) pte)
 #define pte_free(mm, pte) page_table_free(mm, (unsigned long *) pte)

arch/s390/include/asm/pgtable.h

@@ -30,6 +30,7 @@
 #include <linux/sched.h>
 #include <linux/mm_types.h>
 #include <linux/page-flags.h>
+#include <linux/radix-tree.h>
 #include <asm/bug.h>
 #include <asm/page.h>
 
@@ -789,82 +790,67 @@ static inline pgste_t pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry)
 
 /**
  * struct gmap_struct - guest address space
+ * @crst_list: list of all crst tables used in the guest address space
  * @mm: pointer to the parent mm_struct
+ * @guest_to_host: radix tree with guest to host address translation
+ * @host_to_guest: radix tree with pointer to segment table entries
+ * @guest_table_lock: spinlock to protect all entries in the guest page table
  * @table: pointer to the page directory
  * @asce: address space control element for gmap page table
- * @crst_list: list of all crst tables used in the guest address space
  * @pfault_enabled: defines if pfaults are applicable for the guest
  */
 struct gmap {
 	struct list_head list;
+	struct list_head crst_list;
 	struct mm_struct *mm;
+	struct radix_tree_root guest_to_host;
+	struct radix_tree_root host_to_guest;
+	spinlock_t guest_table_lock;
 	unsigned long *table;
 	unsigned long asce;
+	unsigned long asce_end;
 	void *private;
-	struct list_head crst_list;
 	bool pfault_enabled;
 };
 
-/**
- * struct gmap_rmap - reverse mapping for segment table entries
- * @gmap: pointer to the gmap_struct
- * @entry: pointer to a segment table entry
- * @vmaddr: virtual address in the guest address space
- */
-struct gmap_rmap {
-	struct list_head list;
-	struct gmap *gmap;
-	unsigned long *entry;
-	unsigned long vmaddr;
-};
-
-/**
- * struct gmap_pgtable - gmap information attached to a page table
- * @vmaddr: address of the 1MB segment in the process virtual memory
- * @mapper: list of segment table entries mapping a page table
- */
-struct gmap_pgtable {
-	unsigned long vmaddr;
-	struct list_head mapper;
-};
-
 /**
  * struct gmap_notifier - notify function block for page invalidation
  * @notifier_call: address of callback function
  */
 struct gmap_notifier {
 	struct list_head list;
-	void (*notifier_call)(struct gmap *gmap, unsigned long address);
+	void (*notifier_call)(struct gmap *gmap, unsigned long gaddr);
 };
 
-struct gmap *gmap_alloc(struct mm_struct *mm);
+struct gmap *gmap_alloc(struct mm_struct *mm, unsigned long limit);
 void gmap_free(struct gmap *gmap);
 void gmap_enable(struct gmap *gmap);
 void gmap_disable(struct gmap *gmap);
 int gmap_map_segment(struct gmap *gmap, unsigned long from,
 		     unsigned long to, unsigned long len);
 int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len);
-unsigned long __gmap_translate(unsigned long address, struct gmap *);
-unsigned long gmap_translate(unsigned long address, struct gmap *);
-unsigned long __gmap_fault(unsigned long address, struct gmap *);
-unsigned long gmap_fault(unsigned long address, struct gmap *);
-void gmap_discard(unsigned long from, unsigned long to, struct gmap *);
-void __gmap_zap(unsigned long address, struct gmap *);
+unsigned long __gmap_translate(struct gmap *, unsigned long gaddr);
+unsigned long gmap_translate(struct gmap *, unsigned long gaddr);
+int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr);
+int gmap_fault(struct gmap *, unsigned long gaddr, unsigned int fault_flags);
+void gmap_discard(struct gmap *, unsigned long from, unsigned long to);
+void __gmap_zap(struct gmap *, unsigned long gaddr);
 bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *);
 
 
 void gmap_register_ipte_notifier(struct gmap_notifier *);
 void gmap_unregister_ipte_notifier(struct gmap_notifier *);
 int gmap_ipte_notify(struct gmap *, unsigned long start, unsigned long len);
-void gmap_do_ipte_notify(struct mm_struct *, pte_t *);
+void gmap_do_ipte_notify(struct mm_struct *, unsigned long addr, pte_t *);
 
 static inline pgste_t pgste_ipte_notify(struct mm_struct *mm,
+					unsigned long addr,
 					pte_t *ptep, pgste_t pgste)
 {
 #ifdef CONFIG_PGSTE
 	if (pgste_val(pgste) & PGSTE_IN_BIT) {
 		pgste_val(pgste) &= ~PGSTE_IN_BIT;
-		gmap_do_ipte_notify(mm, ptep);
+		gmap_do_ipte_notify(mm, addr, ptep);
 	}
 #endif
 	return pgste;
@@ -1110,7 +1096,7 @@ static inline int ptep_test_and_clear_user_dirty(struct mm_struct *mm,
 	pgste_val(pgste) &= ~PGSTE_UC_BIT;
 	pte = *ptep;
 	if (dirty && (pte_val(pte) & _PAGE_PRESENT)) {
-		pgste = pgste_ipte_notify(mm, ptep, pgste);
+		pgste = pgste_ipte_notify(mm, addr, ptep, pgste);
 		__ptep_ipte(addr, ptep);
 		if (MACHINE_HAS_ESOP || !(pte_val(pte) & _PAGE_WRITE))
 			pte_val(pte) |= _PAGE_PROTECT;
@@ -1132,7 +1118,7 @@ static inline int ptep_test_and_clear_young(struct vm_area_struct *vma,
 
 	if (mm_has_pgste(vma->vm_mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(vma->vm_mm, ptep, pgste);
+		pgste = pgste_ipte_notify(vma->vm_mm, addr, ptep, pgste);
 	}
 
 	pte = *ptep;
@@ -1178,7 +1164,7 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
 
 	if (mm_has_pgste(mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(mm, ptep, pgste);
+		pgste = pgste_ipte_notify(mm, address, ptep, pgste);
 	}
 
 	pte = *ptep;
@@ -1202,7 +1188,7 @@ static inline pte_t ptep_modify_prot_start(struct mm_struct *mm,
 
 	if (mm_has_pgste(mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste_ipte_notify(mm, ptep, pgste);
+		pgste_ipte_notify(mm, address, ptep, pgste);
 	}
 
 	pte = *ptep;
@@ -1239,7 +1225,7 @@ static inline pte_t ptep_clear_flush(struct vm_area_struct *vma,
 
 	if (mm_has_pgste(vma->vm_mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(vma->vm_mm, ptep, pgste);
+		pgste = pgste_ipte_notify(vma->vm_mm, address, ptep, pgste);
 	}
 
 	pte = *ptep;
@@ -1273,7 +1259,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
 
 	if (!full && mm_has_pgste(mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(mm, ptep, pgste);
+		pgste = pgste_ipte_notify(mm, address, ptep, pgste);
 	}
 
 	pte = *ptep;
@@ -1298,7 +1284,7 @@ static inline pte_t ptep_set_wrprotect(struct mm_struct *mm,
 	if (pte_write(pte)) {
 		if (mm_has_pgste(mm)) {
 			pgste = pgste_get_lock(ptep);
-			pgste = pgste_ipte_notify(mm, ptep, pgste);
+			pgste = pgste_ipte_notify(mm, address, ptep, pgste);
 		}
 
 		ptep_flush_lazy(mm, address, ptep);
@@ -1324,7 +1310,7 @@ static inline int ptep_set_access_flags(struct vm_area_struct *vma,
 		return 0;
 	if (mm_has_pgste(vma->vm_mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste = pgste_ipte_notify(vma->vm_mm, ptep, pgste);
+		pgste = pgste_ipte_notify(vma->vm_mm, address, ptep, pgste);
 	}
 
 	ptep_flush_direct(vma->vm_mm, address, ptep);

arch/s390/include/asm/tlb.h

@@ -105,7 +105,7 @@ static inline void tlb_remove_page(struct mmu_gather *tlb, struct page *page)
 static inline void pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
 				unsigned long address)
 {
-	page_table_free_rcu(tlb, (unsigned long *) pte);
+	page_table_free_rcu(tlb, (unsigned long *) pte, address);
 }
 
 /*

arch/s390/include/uapi/asm/kvm.h

@@ -111,12 +111,22 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GPRS   (1UL << 1)
 #define KVM_SYNC_ACRS   (1UL << 2)
 #define KVM_SYNC_CRS    (1UL << 3)
+#define KVM_SYNC_ARCH0  (1UL << 4)
+#define KVM_SYNC_PFAULT (1UL << 5)
 /* definition of registers in kvm_run */
 struct kvm_sync_regs {
 	__u64 prefix;	/* prefix register */
 	__u64 gprs[16];	/* general purpose registers */
 	__u32 acrs[16];	/* access registers */
 	__u64 crs[16];	/* control registers */
+	__u64 todpr;	/* tod programmable register [ARCH0] */
+	__u64 cputm;	/* cpu timer [ARCH0] */
+	__u64 ckc;	/* clock comparator [ARCH0] */
+	__u64 pp;	/* program parameter [ARCH0] */
+	__u64 gbea;	/* guest breaking-event address [ARCH0] */
+	__u64 pft;	/* pfault token [PFAULT] */
+	__u64 pfs;	/* pfault select [PFAULT] */
+	__u64 pfc;	/* pfault compare [PFAULT] */
 };
 
 #define KVM_REG_S390_TODPR	(KVM_REG_S390 | KVM_REG_SIZE_U32 | 0x1)

arch/s390/kvm/diag.c

@@ -28,22 +28,32 @@ static int diag_release_pages(struct kvm_vcpu *vcpu)
 	start = vcpu->run->s.regs.gprs[(vcpu->arch.sie_block->ipa & 0xf0) >> 4];
 	end = vcpu->run->s.regs.gprs[vcpu->arch.sie_block->ipa & 0xf] + 4096;
 
-	if (start & ~PAGE_MASK || end & ~PAGE_MASK || start > end
+	if (start & ~PAGE_MASK || end & ~PAGE_MASK || start >= end
 	    || start < 2 * PAGE_SIZE)
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
 
 	VCPU_EVENT(vcpu, 5, "diag release pages %lX %lX", start, end);
 	vcpu->stat.diagnose_10++;
 
-	/* we checked for start > end above */
-	if (end < prefix || start >= prefix + 2 * PAGE_SIZE) {
-		gmap_discard(start, end, vcpu->arch.gmap);
+	/*
+	 * We checked for start >= end above, so lets check for the
+	 * fast path (no prefix swap page involved)
+	 */
+	if (end <= prefix || start >= prefix + 2 * PAGE_SIZE) {
+		gmap_discard(vcpu->arch.gmap, start, end);
 	} else {
-		if (start < prefix)
-			gmap_discard(start, prefix, vcpu->arch.gmap);
-		if (end >= prefix)
-			gmap_discard(prefix + 2 * PAGE_SIZE,
-				     end, vcpu->arch.gmap);
+		/*
+		 * This is slow path.  gmap_discard will check for start
+		 * so lets split this into before prefix, prefix, after
+		 * prefix and let gmap_discard make some of these calls
+		 * NOPs.
+		 */
+		gmap_discard(vcpu->arch.gmap, start, prefix);
+		if (start <= prefix)
+			gmap_discard(vcpu->arch.gmap, 0, 4096);
+		if (end > prefix + 4096)
+			gmap_discard(vcpu->arch.gmap, 4096, 8192);
+		gmap_discard(vcpu->arch.gmap, prefix + 2 * PAGE_SIZE, end);
 	}
 	return 0;
 }

arch/s390/kvm/gaccess.c

@@ -254,8 +254,7 @@ static void ipte_unlock_simple(struct kvm_vcpu *vcpu)
 		new = old = ACCESS_ONCE(*ic);
 		new.k = 0;
 	} while (cmpxchg(&ic->val, old.val, new.val) != old.val);
-	if (!ipte_lock_count)
-		wake_up(&vcpu->kvm->arch.ipte_wq);
+	wake_up(&vcpu->kvm->arch.ipte_wq);
 out:
 	mutex_unlock(&ipte_mutex);
 }

arch/s390/kvm/interrupt.c

@@ -26,8 +26,9 @@
 #define IOINT_SSID_MASK 0x00030000
 #define IOINT_CSSID_MASK 0x03fc0000
 #define IOINT_AI_MASK 0x04000000
+#define PFAULT_INIT 0x0600
 
-static void deliver_ckc_interrupt(struct kvm_vcpu *vcpu);
+static int __must_check deliver_ckc_interrupt(struct kvm_vcpu *vcpu);
 
 static int is_ioint(u64 type)
 {
@@ -76,7 +77,7 @@ static u64 int_word_to_isc_bits(u32 int_word)
 	return (0x80 >> isc) << 24;
 }
 
-static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu,
+static int __must_check __interrupt_is_deliverable(struct kvm_vcpu *vcpu,
 				      struct kvm_s390_interrupt_info *inti)
 {
 	switch (inti->type) {
@@ -85,6 +86,7 @@ static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu,
 			return 0;
 		if (vcpu->arch.sie_block->gcr[0] & 0x2000ul)
 			return 1;
+		return 0;
 	case KVM_S390_INT_EMERGENCY:
 		if (psw_extint_disabled(vcpu))
 			return 0;
@@ -205,11 +207,30 @@ static void __set_intercept_indicator(struct kvm_vcpu *vcpu,
 	}
 }
 
-static int __deliver_prog_irq(struct kvm_vcpu *vcpu,
-			      struct kvm_s390_pgm_info *pgm_info)
+static u16 get_ilc(struct kvm_vcpu *vcpu)
 {
 	const unsigned short table[] = { 2, 4, 4, 6 };
+
+	switch (vcpu->arch.sie_block->icptcode) {
+	case ICPT_INST:
+	case ICPT_INSTPROGI:
+	case ICPT_OPEREXC:
+	case ICPT_PARTEXEC:
+	case ICPT_IOINST:
+		/* last instruction only stored for these icptcodes */
+		return table[vcpu->arch.sie_block->ipa >> 14];
+	case ICPT_PROGI:
+		return vcpu->arch.sie_block->pgmilc;
+	default:
+		return 0;
+	}
+}
+
+static int __must_check __deliver_prog_irq(struct kvm_vcpu *vcpu,
+			      struct kvm_s390_pgm_info *pgm_info)
+{
 	int rc = 0;
+	u16 ilc = get_ilc(vcpu);
 
 	switch (pgm_info->code & ~PGM_PER) {
 	case PGM_AFX_TRANSLATION:
@@ -276,25 +297,7 @@ static int __deliver_prog_irq(struct kvm_vcpu *vcpu,
 				   (u8 *) __LC_PER_ACCESS_ID);
 	}
 
-	switch (vcpu->arch.sie_block->icptcode) {
-	case ICPT_INST:
-	case ICPT_INSTPROGI:
-	case ICPT_OPEREXC:
-	case ICPT_PARTEXEC:
-	case ICPT_IOINST:
-		/* last instruction only stored for these icptcodes */
-		rc |= put_guest_lc(vcpu, table[vcpu->arch.sie_block->ipa >> 14],
-				   (u16 *) __LC_PGM_ILC);
-		break;
-	case ICPT_PROGI:
-		rc |= put_guest_lc(vcpu, vcpu->arch.sie_block->pgmilc,
-				   (u16 *) __LC_PGM_ILC);
-		break;
-	default:
-		rc |= put_guest_lc(vcpu, 0,
-				   (u16 *) __LC_PGM_ILC);
-	}
-
+	rc |= put_guest_lc(vcpu, ilc, (u16 *) __LC_PGM_ILC);
 	rc |= put_guest_lc(vcpu, pgm_info->code,
 			   (u16 *)__LC_PGM_INT_CODE);
 	rc |= write_guest_lc(vcpu, __LC_PGM_OLD_PSW,
@@ -305,7 +308,7 @@ static int __deliver_prog_irq(struct kvm_vcpu *vcpu,
 	return rc;
 }
 
-static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
+static int __must_check __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 				   struct kvm_s390_interrupt_info *inti)
 {
 	const unsigned short table[] = { 2, 4, 4, 6 };
@@ -343,7 +346,7 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 	case KVM_S390_INT_CLOCK_COMP:
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
 						 inti->ext.ext_params, 0);
-		deliver_ckc_interrupt(vcpu);
+		rc = deliver_ckc_interrupt(vcpu);
 		break;
 	case KVM_S390_INT_CPU_TIMER:
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
@@ -376,8 +379,9 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 	case KVM_S390_INT_PFAULT_INIT:
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, 0,
 						 inti->ext.ext_params2);
-		rc  = put_guest_lc(vcpu, 0x2603, (u16 *) __LC_EXT_INT_CODE);
-		rc |= put_guest_lc(vcpu, 0x0600, (u16 *) __LC_EXT_CPU_ADDR);
+		rc  = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE,
+				   (u16 *) __LC_EXT_INT_CODE);
+		rc |= put_guest_lc(vcpu, PFAULT_INIT, (u16 *) __LC_EXT_CPU_ADDR);
 		rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
 				     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
 		rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
@@ -501,14 +505,11 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 	default:
 		BUG();
 	}
-	if (rc) {
-		printk("kvm: The guest lowcore is not mapped during interrupt "
-		       "delivery, killing userspace\n");
-		do_exit(SIGKILL);
-	}
+
+	return rc;
 }
 
-static void deliver_ckc_interrupt(struct kvm_vcpu *vcpu)
+static int __must_check deliver_ckc_interrupt(struct kvm_vcpu *vcpu)
 {
 	int rc;
 
@@ -518,11 +519,7 @@ static void deliver_ckc_interrupt(struct kvm_vcpu *vcpu)
 	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
 			    &vcpu->arch.sie_block->gpsw,
 			    sizeof(psw_t));
-	if (rc) {
-		printk("kvm: The guest lowcore is not mapped during interrupt "
-			"delivery, killing userspace\n");
-		do_exit(SIGKILL);
-	}
+	return rc;
 }
 
 /* Check whether SIGP interpretation facility has an external call pending */
@@ -661,12 +658,13 @@ void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu)
 			  &vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].ctrl);
 }
 
-void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
+int __must_check kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
 {
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
 	struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int;
 	struct kvm_s390_interrupt_info  *n, *inti = NULL;
 	int deliver;
+	int rc = 0;
 
 	__reset_intercept_indicators(vcpu);
 	if (atomic_read(&li->active)) {
@@ -685,16 +683,16 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
 				atomic_set(&li->active, 0);
 			spin_unlock(&li->lock);
 			if (deliver) {
-				__do_deliver_interrupt(vcpu, inti);
+				rc = __do_deliver_interrupt(vcpu, inti);
 				kfree(inti);
 			}
-		} while (deliver);
+		} while (!rc && deliver);
 	}
 
-	if (kvm_cpu_has_pending_timer(vcpu))
-		deliver_ckc_interrupt(vcpu);
+	if (!rc && kvm_cpu_has_pending_timer(vcpu))
+		rc = deliver_ckc_interrupt(vcpu);
 
-	if (atomic_read(&fi->active)) {
+	if (!rc && atomic_read(&fi->active)) {
 		do {
 			deliver = 0;
 			spin_lock(&fi->lock);
@@ -711,67 +709,13 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
 				atomic_set(&fi->active, 0);
 			spin_unlock(&fi->lock);
 			if (deliver) {
-				__do_deliver_interrupt(vcpu, inti);
-				kfree(inti);
-			}
-		} while (deliver);
-	}
-}
-
-void kvm_s390_deliver_pending_machine_checks(struct kvm_vcpu *vcpu)
-{
-	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
-	struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int;
-	struct kvm_s390_interrupt_info  *n, *inti = NULL;
-	int deliver;
-
-	__reset_intercept_indicators(vcpu);
-	if (atomic_read(&li->active)) {
-		do {
-			deliver = 0;
-			spin_lock(&li->lock);
-			list_for_each_entry_safe(inti, n, &li->list, list) {
-				if ((inti->type == KVM_S390_MCHK) &&
-				    __interrupt_is_deliverable(vcpu, inti)) {
-					list_del(&inti->list);
-					deliver = 1;
-					break;
-				}
-				__set_intercept_indicator(vcpu, inti);
-			}
-			if (list_empty(&li->list))
-				atomic_set(&li->active, 0);
-			spin_unlock(&li->lock);
-			if (deliver) {
-				__do_deliver_interrupt(vcpu, inti);
+				rc = __do_deliver_interrupt(vcpu, inti);
 				kfree(inti);
 			}
-		} while (deliver);
+		} while (!rc && deliver);
 	}
 
-	if (atomic_read(&fi->active)) {
-		do {
-			deliver = 0;
-			spin_lock(&fi->lock);
-			list_for_each_entry_safe(inti, n, &fi->list, list) {
-				if ((inti->type == KVM_S390_MCHK) &&
-				    __interrupt_is_deliverable(vcpu, inti)) {
-					list_del(&inti->list);
-					fi->irq_count--;
-					deliver = 1;
-					break;
-				}
-				__set_intercept_indicator(vcpu, inti);
-			}
-			if (list_empty(&fi->list))
-				atomic_set(&fi->active, 0);
-			spin_unlock(&fi->lock);
-			if (deliver) {
-				__do_deliver_interrupt(vcpu, inti);
-				kfree(inti);
-			}
-		} while (deliver);
-	}
+	return rc;
 }
 
 int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code)
@@ -1048,7 +992,6 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, s390int->type, s390int->parm,
 				   s390int->parm64, 2);
 
-	mutex_lock(&vcpu->kvm->lock);
 	li = &vcpu->arch.local_int;
 	spin_lock(&li->lock);
 	if (inti->type == KVM_S390_PROGRAM_INT)
@@ -1060,7 +1003,6 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 		li->action_bits |= ACTION_STOP_ON_STOP;
 	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
 	spin_unlock(&li->lock);
-	mutex_unlock(&vcpu->kvm->lock);
 	kvm_s390_vcpu_wakeup(vcpu);
 	return 0;
 }
@@ -1300,7 +1242,7 @@ static int kvm_s390_adapter_map(struct kvm *kvm, unsigned int id, __u64 addr)
 	}
 	INIT_LIST_HEAD(&map->list);
 	map->guest_addr = addr;
-	map->addr = gmap_translate(addr, kvm->arch.gmap);
+	map->addr = gmap_translate(kvm->arch.gmap, addr);
 	if (map->addr == -EFAULT) {
 		ret = -EFAULT;
 		goto out;
@@ -1410,7 +1352,6 @@ static int flic_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 		r = enqueue_floating_irq(dev, attr);
 		break;
 	case KVM_DEV_FLIC_CLEAR_IRQS:
-		r = 0;
 		kvm_s390_clear_float_irqs(dev->kvm);
 		break;
 	case KVM_DEV_FLIC_APF_ENABLE:

arch/s390/kvm/kvm-s390.c

@@ -100,16 +100,12 @@ int test_vfacility(unsigned long nr)
 }
 
 /* Section: not file related */
-int kvm_arch_hardware_enable(void *garbage)
+int kvm_arch_hardware_enable(void)
 {
 	/* every s390 is virtualization enabled ;-) */
 	return 0;
 }
 
-void kvm_arch_hardware_disable(void *garbage)
-{
-}
-
 static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address);
 
 int kvm_arch_hardware_setup(void)
@@ -124,17 +120,10 @@ void kvm_arch_hardware_unsetup(void)
 	gmap_unregister_ipte_notifier(&gmap_notifier);
 }
 
-void kvm_arch_check_processor_compat(void *rtn)
-{
-}
-
 int kvm_arch_init(void *opaque)
 {
-	return 0;
-}
-
-void kvm_arch_exit(void)
-{
+	/* Register floating interrupt controller interface. */
+	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
 }
 
 /* Section: device related */
@@ -404,6 +393,22 @@ long kvm_arch_vm_ioctl(struct file *filp,
 	return r;
 }
 
+static int kvm_s390_crypto_init(struct kvm *kvm)
+{
+	if (!test_vfacility(76))
+		return 0;
+
+	kvm->arch.crypto.crycb = kzalloc(sizeof(*kvm->arch.crypto.crycb),
+					 GFP_KERNEL | GFP_DMA);
+	if (!kvm->arch.crypto.crycb)
+		return -ENOMEM;
+
+	kvm->arch.crypto.crycbd = (__u32) (unsigned long) kvm->arch.crypto.crycb |
+				  CRYCB_FORMAT1;
+
+	return 0;
+}
+
 int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
 	int rc;
@@ -441,6 +446,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	if (!kvm->arch.dbf)
 		goto out_nodbf;
 
+	if (kvm_s390_crypto_init(kvm) < 0)
+		goto out_crypto;
+
 	spin_lock_init(&kvm->arch.float_int.lock);
 	INIT_LIST_HEAD(&kvm->arch.float_int.list);
 	init_waitqueue_head(&kvm->arch.ipte_wq);
@@ -451,7 +459,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	if (type & KVM_VM_S390_UCONTROL) {
 		kvm->arch.gmap = NULL;
 	} else {
-		kvm->arch.gmap = gmap_alloc(current->mm);
+		kvm->arch.gmap = gmap_alloc(current->mm, (1UL << 44) - 1);
 		if (!kvm->arch.gmap)
 			goto out_nogmap;
 		kvm->arch.gmap->private = kvm;
@@ -465,6 +473,8 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	return 0;
 out_nogmap:
+	kfree(kvm->arch.crypto.crycb);
+out_crypto:
 	debug_unregister(kvm->arch.dbf);
 out_nodbf:
 	free_page((unsigned long)(kvm->arch.sca));
@@ -514,15 +524,12 @@ static void kvm_free_vcpus(struct kvm *kvm)
 	mutex_unlock(&kvm->lock);
 }
 
-void kvm_arch_sync_events(struct kvm *kvm)
-{
-}
-
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
 	kvm_free_vcpus(kvm);
 	free_page((unsigned long)(kvm->arch.sca));
 	debug_unregister(kvm->arch.dbf);
+	kfree(kvm->arch.crypto.crycb);
 	if (!kvm_is_ucontrol(kvm))
 		gmap_free(kvm->arch.gmap);
 	kvm_s390_destroy_adapters(kvm);
@@ -535,7 +542,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
 	kvm_clear_async_pf_completion_queue(vcpu);
 	if (kvm_is_ucontrol(vcpu->kvm)) {
-		vcpu->arch.gmap = gmap_alloc(current->mm);
+		vcpu->arch.gmap = gmap_alloc(current->mm, -1UL);
 		if (!vcpu->arch.gmap)
 			return -ENOMEM;
 		vcpu->arch.gmap->private = vcpu->kvm;
@@ -546,19 +553,12 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
 				    KVM_SYNC_GPRS |
 				    KVM_SYNC_ACRS |
-				    KVM_SYNC_CRS;
+				    KVM_SYNC_CRS |
+				    KVM_SYNC_ARCH0 |
+				    KVM_SYNC_PFAULT;
 	return 0;
 }
 
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
-{
-	/* Nothing todo */
-}
-
-void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu)
-{
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	save_fp_ctl(&vcpu->arch.host_fpregs.fpc);
@@ -611,6 +611,14 @@ int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
+{
+	if (!test_vfacility(76))
+		return;
+
+	vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
+}
+
 void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu)
 {
 	free_page(vcpu->arch.sie_block->cbrlo);
@@ -657,6 +665,9 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
 	get_cpu_id(&vcpu->arch.cpu_id);
 	vcpu->arch.cpu_id.version = 0xff;
+
+	kvm_s390_vcpu_crypto_setup(vcpu);
+
 	return rc;
 }
 
@@ -1053,6 +1064,11 @@ retry:
 		goto retry;
 	}
 
+	if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
+		vcpu->arch.sie_block->ihcpu = 0xffff;
+		goto retry;
+	}
+
 	if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
 		if (!ibs_enabled(vcpu)) {
 			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
@@ -1089,18 +1105,8 @@ retry:
  */
 long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable)
 {
-	struct mm_struct *mm = current->mm;
-	hva_t hva;
-	long rc;
-
-	hva = gmap_fault(gpa, vcpu->arch.gmap);
-	if (IS_ERR_VALUE(hva))
-		return (long)hva;
-	down_read(&mm->mmap_sem);
-	rc = get_user_pages(current, mm, hva, 1, writable, 0, NULL, NULL);
-	up_read(&mm->mmap_sem);
-
-	return rc < 0 ? rc : 0;
+	return gmap_fault(vcpu->arch.gmap, gpa,
+			  writable ? FAULT_FLAG_WRITE : 0);
 }
 
 static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
@@ -1195,8 +1201,11 @@ static int vcpu_pre_run(struct kvm_vcpu *vcpu)
 	if (test_cpu_flag(CIF_MCCK_PENDING))
 		s390_handle_mcck();
 
-	if (!kvm_is_ucontrol(vcpu->kvm))
-		kvm_s390_deliver_pending_interrupts(vcpu);
+	if (!kvm_is_ucontrol(vcpu->kvm)) {
+		rc = kvm_s390_deliver_pending_interrupts(vcpu);
+		if (rc)
+			return rc;
+	}
 
 	rc = kvm_s390_handle_requests(vcpu);
 	if (rc)
@@ -1300,6 +1309,48 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
 	return rc;
 }
 
+static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	vcpu->arch.sie_block->gpsw.mask = kvm_run->psw_mask;
+	vcpu->arch.sie_block->gpsw.addr = kvm_run->psw_addr;
+	if (kvm_run->kvm_dirty_regs & KVM_SYNC_PREFIX)
+		kvm_s390_set_prefix(vcpu, kvm_run->s.regs.prefix);
+	if (kvm_run->kvm_dirty_regs & KVM_SYNC_CRS) {
+		memcpy(&vcpu->arch.sie_block->gcr, &kvm_run->s.regs.crs, 128);
+		/* some control register changes require a tlb flush */
+		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
+	}
+	if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
+		vcpu->arch.sie_block->cputm = kvm_run->s.regs.cputm;
+		vcpu->arch.sie_block->ckc = kvm_run->s.regs.ckc;
+		vcpu->arch.sie_block->todpr = kvm_run->s.regs.todpr;
+		vcpu->arch.sie_block->pp = kvm_run->s.regs.pp;
+		vcpu->arch.sie_block->gbea = kvm_run->s.regs.gbea;
+	}
+	if (kvm_run->kvm_dirty_regs & KVM_SYNC_PFAULT) {
+		vcpu->arch.pfault_token = kvm_run->s.regs.pft;
+		vcpu->arch.pfault_select = kvm_run->s.regs.pfs;
+		vcpu->arch.pfault_compare = kvm_run->s.regs.pfc;
+	}
+	kvm_run->kvm_dirty_regs = 0;
+}
+
+static void store_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	kvm_run->psw_mask = vcpu->arch.sie_block->gpsw.mask;
+	kvm_run->psw_addr = vcpu->arch.sie_block->gpsw.addr;
+	kvm_run->s.regs.prefix = kvm_s390_get_prefix(vcpu);
+	memcpy(&kvm_run->s.regs.crs, &vcpu->arch.sie_block->gcr, 128);
+	kvm_run->s.regs.cputm = vcpu->arch.sie_block->cputm;
+	kvm_run->s.regs.ckc = vcpu->arch.sie_block->ckc;
+	kvm_run->s.regs.todpr = vcpu->arch.sie_block->todpr;
+	kvm_run->s.regs.pp = vcpu->arch.sie_block->pp;
+	kvm_run->s.regs.gbea = vcpu->arch.sie_block->gbea;
+	kvm_run->s.regs.pft = vcpu->arch.pfault_token;
+	kvm_run->s.regs.pfs = vcpu->arch.pfault_select;
+	kvm_run->s.regs.pfc = vcpu->arch.pfault_compare;
+}
+
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	int rc;
@@ -1321,30 +1372,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		return -EINVAL;
 	}
 
-	switch (kvm_run->exit_reason) {
-	case KVM_EXIT_S390_SIEIC:
-	case KVM_EXIT_UNKNOWN:
-	case KVM_EXIT_INTR:
-	case KVM_EXIT_S390_RESET:
-	case KVM_EXIT_S390_UCONTROL:
-	case KVM_EXIT_S390_TSCH:
-	case KVM_EXIT_DEBUG:
-		break;
-	default:
-		BUG();
-	}
-
-	vcpu->arch.sie_block->gpsw.mask = kvm_run->psw_mask;
-	vcpu->arch.sie_block->gpsw.addr = kvm_run->psw_addr;
-	if (kvm_run->kvm_dirty_regs & KVM_SYNC_PREFIX) {
-		kvm_run->kvm_dirty_regs &= ~KVM_SYNC_PREFIX;
-		kvm_s390_set_prefix(vcpu, kvm_run->s.regs.prefix);
-	}
-	if (kvm_run->kvm_dirty_regs & KVM_SYNC_CRS) {
-		kvm_run->kvm_dirty_regs &= ~KVM_SYNC_CRS;
-		memcpy(&vcpu->arch.sie_block->gcr, &kvm_run->s.regs.crs, 128);
-		kvm_s390_set_prefix(vcpu, kvm_run->s.regs.prefix);
-	}
+	sync_regs(vcpu, kvm_run);
 
 	might_fault();
 	rc = __vcpu_run(vcpu);
@@ -1374,10 +1402,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		rc = 0;
 	}
 
-	kvm_run->psw_mask     = vcpu->arch.sie_block->gpsw.mask;
-	kvm_run->psw_addr     = vcpu->arch.sie_block->gpsw.addr;
-	kvm_run->s.regs.prefix = kvm_s390_get_prefix(vcpu);
-	memcpy(&kvm_run->s.regs.crs, &vcpu->arch.sie_block->gcr, 128);
+	store_regs(vcpu, kvm_run);
 
 	if (vcpu->sigset_active)
 		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
@@ -1506,7 +1531,7 @@ void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
 	 * Another VCPU might have used IBS while we were offline.
 	 * Let's play safe and flush the VCPU at startup.
 	 */
-	vcpu->arch.sie_block->ihcpu  = 0xffff;
+	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
 	return;
 }
@@ -1661,9 +1686,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 	}
 #endif
 	case KVM_S390_VCPU_FAULT: {
-		r = gmap_fault(arg, vcpu->arch.gmap);
-		if (!IS_ERR_VALUE(r))
-			r = 0;
+		r = gmap_fault(vcpu->arch.gmap, arg, 0);
 		break;
 	}
 	case KVM_ENABLE_CAP:
@@ -1694,21 +1717,12 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
-			   struct kvm_memory_slot *dont)
-{
-}
-
 int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 			    unsigned long npages)
 {
 	return 0;
 }
 
-void kvm_arch_memslots_updated(struct kvm *kvm)
-{
-}
-
 /* Section: memory related */
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				   struct kvm_memory_slot *memslot,
@@ -1754,15 +1768,6 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 	return;
 }
 
-void kvm_arch_flush_shadow_all(struct kvm *kvm)
-{
-}
-
-void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
-				   struct kvm_memory_slot *slot)
-{
-}
-
 static int __init kvm_s390_init(void)
 {
 	int ret;

arch/s390/kvm/kvm-s390.h

@@ -70,7 +70,7 @@ static inline u32 kvm_s390_get_prefix(struct kvm_vcpu *vcpu)
 static inline void kvm_s390_set_prefix(struct kvm_vcpu *vcpu, u32 prefix)
 {
 	vcpu->arch.sie_block->prefix = prefix >> GUEST_PREFIX_SHIFT;
-	vcpu->arch.sie_block->ihcpu  = 0xffff;
+	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 	kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu);
 }
 
@@ -138,8 +138,7 @@ static inline int kvm_s390_user_cpu_state_ctrl(struct kvm *kvm)
 int kvm_s390_handle_wait(struct kvm_vcpu *vcpu);
 void kvm_s390_vcpu_wakeup(struct kvm_vcpu *vcpu);
 enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer);
-void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu);
-void kvm_s390_deliver_pending_machine_checks(struct kvm_vcpu *vcpu);
+int __must_check kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu);
 void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu);
 void kvm_s390_clear_float_irqs(struct kvm *kvm);
 int __must_check kvm_s390_inject_vm(struct kvm *kvm,
@@ -228,6 +227,7 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int psw_extint_disabled(struct kvm_vcpu *vcpu);
 void kvm_s390_destroy_adapters(struct kvm *kvm);
 int kvm_s390_si_ext_call_pending(struct kvm_vcpu *vcpu);
+extern struct kvm_device_ops kvm_flic_ops;
 
 /* implemented in guestdbg.c */
 void kvm_s390_backup_guest_per_regs(struct kvm_vcpu *vcpu);

arch/s390/kvm/priv.c

@@ -352,13 +352,6 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-static void handle_new_psw(struct kvm_vcpu *vcpu)
-{
-	/* Check whether the new psw is enabled for machine checks. */
-	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_MCHECK)
-		kvm_s390_deliver_pending_machine_checks(vcpu);
-}
-
 #define PSW_MASK_ADDR_MODE (PSW_MASK_EA | PSW_MASK_BA)
 #define PSW_MASK_UNASSIGNED 0xb80800fe7fffffffUL
 #define PSW_ADDR_24 0x0000000000ffffffUL
@@ -405,7 +398,6 @@ int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu)
 	gpsw->addr = new_psw.addr & ~PSW32_ADDR_AMODE;
 	if (!is_valid_psw(gpsw))
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
-	handle_new_psw(vcpu);
 	return 0;
 }
 
@@ -427,7 +419,6 @@ static int handle_lpswe(struct kvm_vcpu *vcpu)
 	vcpu->arch.sie_block->gpsw = new_psw;
 	if (!is_valid_psw(&vcpu->arch.sie_block->gpsw))
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
-	handle_new_psw(vcpu);
 	return 0;
 }
 
@@ -738,7 +729,7 @@ static int handle_essa(struct kvm_vcpu *vcpu)
 			/* invalid entry */
 			break;
 		/* try to free backing */
-		__gmap_zap(cbrle, gmap);
+		__gmap_zap(gmap, cbrle);
 	}
 	up_read(&gmap->mm->mmap_sem);
 	if (i < entries)

arch/s390/mm/fault.c

@@ -442,18 +442,15 @@ static inline int do_exception(struct pt_regs *regs, int access)
 	down_read(&mm->mmap_sem);
 
 #ifdef CONFIG_PGSTE
-	gmap = (struct gmap *)
-		((current->flags & PF_VCPU) ? S390_lowcore.gmap : 0);
+	gmap = (current->flags & PF_VCPU) ?
+		(struct gmap *) S390_lowcore.gmap : NULL;
 	if (gmap) {
-		address = __gmap_fault(address, gmap);
+		current->thread.gmap_addr = address;
+		address = __gmap_translate(gmap, address);
 		if (address == -EFAULT) {
 			fault = VM_FAULT_BADMAP;
 			goto out_up;
 		}
-		if (address == -ENOMEM) {
-			fault = VM_FAULT_OOM;
-			goto out_up;
-		}
 		if (gmap->pfault_enabled)
 			flags |= FAULT_FLAG_RETRY_NOWAIT;
 	}
@@ -530,6 +527,20 @@ retry:
 			goto retry;
 		}
 	}
+#ifdef CONFIG_PGSTE
+	if (gmap) {
+		address =  __gmap_link(gmap, current->thread.gmap_addr,
+				       address);
+		if (address == -EFAULT) {
+			fault = VM_FAULT_BADMAP;
+			goto out_up;
+		}
+		if (address == -ENOMEM) {
+			fault = VM_FAULT_OOM;
+			goto out_up;
+		}
+	}
+#endif
 	fault = 0;
 out_up:
 	up_read(&mm->mmap_sem);

arch/s390/mm/pgtable.c

@@ -145,30 +145,56 @@ void crst_table_downgrade(struct mm_struct *mm, unsigned long limit)
 /**
  * gmap_alloc - allocate a guest address space
  * @mm: pointer to the parent mm_struct
+ * @limit: maximum size of the gmap address space
  *
  * Returns a guest address space structure.
  */
-struct gmap *gmap_alloc(struct mm_struct *mm)
+struct gmap *gmap_alloc(struct mm_struct *mm, unsigned long limit)
 {
 	struct gmap *gmap;
 	struct page *page;
 	unsigned long *table;
-
+	unsigned long etype, atype;
+
+	if (limit < (1UL << 31)) {
+		limit = (1UL << 31) - 1;
+		atype = _ASCE_TYPE_SEGMENT;
+		etype = _SEGMENT_ENTRY_EMPTY;
+	} else if (limit < (1UL << 42)) {
+		limit = (1UL << 42) - 1;
+		atype = _ASCE_TYPE_REGION3;
+		etype = _REGION3_ENTRY_EMPTY;
+	} else if (limit < (1UL << 53)) {
+		limit = (1UL << 53) - 1;
+		atype = _ASCE_TYPE_REGION2;
+		etype = _REGION2_ENTRY_EMPTY;
+	} else {
+		limit = -1UL;
+		atype = _ASCE_TYPE_REGION1;
+		etype = _REGION1_ENTRY_EMPTY;
+	}
 	gmap = kzalloc(sizeof(struct gmap), GFP_KERNEL);
 	if (!gmap)
 		goto out;
 	INIT_LIST_HEAD(&gmap->crst_list);
+	INIT_RADIX_TREE(&gmap->guest_to_host, GFP_KERNEL);
+	INIT_RADIX_TREE(&gmap->host_to_guest, GFP_ATOMIC);
+	spin_lock_init(&gmap->guest_table_lock);
 	gmap->mm = mm;
 	page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
 	if (!page)
 		goto out_free;
+	page->index = 0;
 	list_add(&page->lru, &gmap->crst_list);
 	table = (unsigned long *) page_to_phys(page);
-	crst_table_init(table, _REGION1_ENTRY_EMPTY);
+	crst_table_init(table, etype);
 	gmap->table = table;
-	gmap->asce = _ASCE_TYPE_REGION1 | _ASCE_TABLE_LENGTH |
-		     _ASCE_USER_BITS | __pa(table);
+	gmap->asce = atype | _ASCE_TABLE_LENGTH |
+		_ASCE_USER_BITS | __pa(table);
+	gmap->asce_end = limit;
+	down_write(&mm->mmap_sem);
 	list_add(&gmap->list, &mm->context.gmap_list);
+	up_write(&mm->mmap_sem);
 	return gmap;
 
 out_free:
@@ -178,36 +204,38 @@ out:
 }
 EXPORT_SYMBOL_GPL(gmap_alloc);
 
-static int gmap_unlink_segment(struct gmap *gmap, unsigned long *table)
-{
-	struct gmap_pgtable *mp;
-	struct gmap_rmap *rmap;
-	struct page *page;
-
-	if (*table & _SEGMENT_ENTRY_INVALID)
-		return 0;
-	page = pfn_to_page(*table >> PAGE_SHIFT);
-	mp = (struct gmap_pgtable *) page->index;
-	list_for_each_entry(rmap, &mp->mapper, list) {
-		if (rmap->entry != table)
-			continue;
-		list_del(&rmap->list);
-		kfree(rmap);
-		break;
-	}
-	*table = mp->vmaddr | _SEGMENT_ENTRY_INVALID | _SEGMENT_ENTRY_PROTECT;
-	return 1;
-}
-
 static void gmap_flush_tlb(struct gmap *gmap)
 {
 	if (MACHINE_HAS_IDTE)
-		__tlb_flush_asce(gmap->mm, (unsigned long) gmap->table |
-				 _ASCE_TYPE_REGION1);
+		__tlb_flush_asce(gmap->mm, gmap->asce);
 	else
 		__tlb_flush_global();
 }
 
+static void gmap_radix_tree_free(struct radix_tree_root *root)
+{
+	struct radix_tree_iter iter;
+	unsigned long indices[16];
+	unsigned long index;
+	void **slot;
+	int i, nr;
+
+	/* A radix tree is freed by deleting all of its entries */
+	index = 0;
+	do {
+		nr = 0;
+		radix_tree_for_each_slot(slot, root, &iter, index) {
+			indices[nr] = iter.index;
+			if (++nr == 16)
+				break;
+		}
+		for (i = 0; i < nr; i++) {
+			index = indices[i];
+			radix_tree_delete(root, index);
+		}
+	} while (nr > 0);
+}
+
 /**
  * gmap_free - free a guest address space
  * @gmap: pointer to the guest address space structure
@@ -215,31 +243,21 @@ static void gmap_flush_tlb(struct gmap *gmap)
 void gmap_free(struct gmap *gmap)
 {
 	struct page *page, *next;
-	unsigned long *table;
-	int i;
-
 
 	/* Flush tlb. */
 	if (MACHINE_HAS_IDTE)
-		__tlb_flush_asce(gmap->mm, (unsigned long) gmap->table |
-				 _ASCE_TYPE_REGION1);
+		__tlb_flush_asce(gmap->mm, gmap->asce);
 	else
 		__tlb_flush_global();
 
 	/* Free all segment & region tables. */
-	down_read(&gmap->mm->mmap_sem);
-	spin_lock(&gmap->mm->page_table_lock);
-	list_for_each_entry_safe(page, next, &gmap->crst_list, lru) {
-		table = (unsigned long *) page_to_phys(page);
-		if ((*table & _REGION_ENTRY_TYPE_MASK) == 0)
-			/* Remove gmap rmap structures for segment table. */
-			for (i = 0; i < PTRS_PER_PMD; i++, table++)
-				gmap_unlink_segment(gmap, table);
+	list_for_each_entry_safe(page, next, &gmap->crst_list, lru)
 		__free_pages(page, ALLOC_ORDER);
-	}
-	spin_unlock(&gmap->mm->page_table_lock);
-	up_read(&gmap->mm->mmap_sem);
+	gmap_radix_tree_free(&gmap->guest_to_host);
+	gmap_radix_tree_free(&gmap->host_to_guest);
+	down_write(&gmap->mm->mmap_sem);
 	list_del(&gmap->list);
+	up_write(&gmap->mm->mmap_sem);
 	kfree(gmap);
 }
 EXPORT_SYMBOL_GPL(gmap_free);
@@ -267,42 +285,97 @@ EXPORT_SYMBOL_GPL(gmap_disable);
 /*
  * gmap_alloc_table is assumed to be called with mmap_sem held
  */
-static int gmap_alloc_table(struct gmap *gmap,
-			    unsigned long *table, unsigned long init)
-	__releases(&gmap->mm->page_table_lock)
-	__acquires(&gmap->mm->page_table_lock)
+static int gmap_alloc_table(struct gmap *gmap, unsigned long *table,
+			    unsigned long init, unsigned long gaddr)
 {
 	struct page *page;
 	unsigned long *new;
 
 	/* since we dont free the gmap table until gmap_free we can unlock */
-	spin_unlock(&gmap->mm->page_table_lock);
 	page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
-	spin_lock(&gmap->mm->page_table_lock);
 	if (!page)
 		return -ENOMEM;
 	new = (unsigned long *) page_to_phys(page);
 	crst_table_init(new, init);
+	spin_lock(&gmap->mm->page_table_lock);
 	if (*table & _REGION_ENTRY_INVALID) {
 		list_add(&page->lru, &gmap->crst_list);
 		*table = (unsigned long) new | _REGION_ENTRY_LENGTH |
 			(*table & _REGION_ENTRY_TYPE_MASK);
-	} else
+		page->index = gaddr;
+		page = NULL;
+	}
+	spin_unlock(&gmap->mm->page_table_lock);
+	if (page)
 		__free_pages(page, ALLOC_ORDER);
 	return 0;
 }
 
+/**
+ * __gmap_segment_gaddr - find virtual address from segment pointer
+ * @entry: pointer to a segment table entry in the guest address space
+ *
+ * Returns the virtual address in the guest address space for the segment
+ */
+static unsigned long __gmap_segment_gaddr(unsigned long *entry)
+{
+	struct page *page;
+	unsigned long offset;
+
+	offset = (unsigned long) entry / sizeof(unsigned long);
+	offset = (offset & (PTRS_PER_PMD - 1)) * PMD_SIZE;
+	page = pmd_to_page((pmd_t *) entry);
+	return page->index + offset;
+}
+
+/**
+ * __gmap_unlink_by_vmaddr - unlink a single segment via a host address
+ * @gmap: pointer to the guest address space structure
+ * @vmaddr: address in the host process address space
+ *
+ * Returns 1 if a TLB flush is required
+ */
+static int __gmap_unlink_by_vmaddr(struct gmap *gmap, unsigned long vmaddr)
+{
+	unsigned long *entry;
+	int flush = 0;
+
+	spin_lock(&gmap->guest_table_lock);
+	entry = radix_tree_delete(&gmap->host_to_guest, vmaddr >> PMD_SHIFT);
+	if (entry) {
+		flush = (*entry != _SEGMENT_ENTRY_INVALID);
+		*entry = _SEGMENT_ENTRY_INVALID;
+	}
+	spin_unlock(&gmap->guest_table_lock);
+	return flush;
+}
+
+/**
+ * __gmap_unmap_by_gaddr - unmap a single segment via a guest address
+ * @gmap: pointer to the guest address space structure
+ * @gaddr: address in the guest address space
+ *
+ * Returns 1 if a TLB flush is required
+ */
+static int __gmap_unmap_by_gaddr(struct gmap *gmap, unsigned long gaddr)
+{
+	unsigned long vmaddr;
+
+	vmaddr = (unsigned long) radix_tree_delete(&gmap->guest_to_host,
+						   gaddr >> PMD_SHIFT);
+	return vmaddr ? __gmap_unlink_by_vmaddr(gmap, vmaddr) : 0;
+}
+
 /**
  * gmap_unmap_segment - unmap segment from the guest address space
  * @gmap: pointer to the guest address space structure
- * @addr: address in the guest address space
+ * @to: address in the guest address space
  * @len: length of the memory area to unmap
  *
  * Returns 0 if the unmap succeeded, -EINVAL if not.
  */
 int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len)
 {
-	unsigned long *table;
 	unsigned long off;
 	int flush;
 
@@ -312,31 +385,10 @@ int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len)
 		return -EINVAL;
 
 	flush = 0;
-	down_read(&gmap->mm->mmap_sem);
-	spin_lock(&gmap->mm->page_table_lock);
-	for (off = 0; off < len; off += PMD_SIZE) {
-		/* Walk the guest addr space page table */
-		table = gmap->table + (((to + off) >> 53) & 0x7ff);
-		if (*table & _REGION_ENTRY_INVALID)
-			goto out;
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + (((to + off) >> 42) & 0x7ff);
-		if (*table & _REGION_ENTRY_INVALID)
-			goto out;
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + (((to + off) >> 31) & 0x7ff);
-		if (*table & _REGION_ENTRY_INVALID)
-			goto out;
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + (((to + off) >> 20) & 0x7ff);
-
-		/* Clear segment table entry in guest address space. */
-		flush |= gmap_unlink_segment(gmap, table);
-		*table = _SEGMENT_ENTRY_INVALID;
-	}
-out:
-	spin_unlock(&gmap->mm->page_table_lock);
-	up_read(&gmap->mm->mmap_sem);
+	down_write(&gmap->mm->mmap_sem);
+	for (off = 0; off < len; off += PMD_SIZE)
+		flush |= __gmap_unmap_by_gaddr(gmap, to + off);
+	up_write(&gmap->mm->mmap_sem);
 	if (flush)
 		gmap_flush_tlb(gmap);
 	return 0;
@@ -348,87 +400,47 @@ EXPORT_SYMBOL_GPL(gmap_unmap_segment);
  * @gmap: pointer to the guest address space structure
  * @from: source address in the parent address space
  * @to: target address in the guest address space
+ * @len: length of the memory area to map
  *
  * Returns 0 if the mmap succeeded, -EINVAL or -ENOMEM if not.
  */
 int gmap_map_segment(struct gmap *gmap, unsigned long from,
 		     unsigned long to, unsigned long len)
 {
-	unsigned long *table;
 	unsigned long off;
 	int flush;
 
 	if ((from | to | len) & (PMD_SIZE - 1))
 		return -EINVAL;
-	if (len == 0 || from + len > TASK_MAX_SIZE ||
-	    from + len < from || to + len < to)
+	if (len == 0 || from + len < from || to + len < to ||
+	    from + len > TASK_MAX_SIZE || to + len > gmap->asce_end)
 		return -EINVAL;
 
 	flush = 0;
-	down_read(&gmap->mm->mmap_sem);
-	spin_lock(&gmap->mm->page_table_lock);
+	down_write(&gmap->mm->mmap_sem);
 	for (off = 0; off < len; off += PMD_SIZE) {
-		/* Walk the gmap address space page table */
-		table = gmap->table + (((to + off) >> 53) & 0x7ff);
-		if ((*table & _REGION_ENTRY_INVALID) &&
-		    gmap_alloc_table(gmap, table, _REGION2_ENTRY_EMPTY))
-			goto out_unmap;
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + (((to + off) >> 42) & 0x7ff);
-		if ((*table & _REGION_ENTRY_INVALID) &&
-		    gmap_alloc_table(gmap, table, _REGION3_ENTRY_EMPTY))
-			goto out_unmap;
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + (((to + off) >> 31) & 0x7ff);
-		if ((*table & _REGION_ENTRY_INVALID) &&
-		    gmap_alloc_table(gmap, table, _SEGMENT_ENTRY_EMPTY))
-			goto out_unmap;
-		table = (unsigned long *) (*table & _REGION_ENTRY_ORIGIN);
-		table = table + (((to + off) >> 20) & 0x7ff);
-
-		/* Store 'from' address in an invalid segment table entry. */
-		flush |= gmap_unlink_segment(gmap, table);
-		*table =  (from + off) | (_SEGMENT_ENTRY_INVALID |
-					  _SEGMENT_ENTRY_PROTECT);
+		/* Remove old translation */
+		flush |= __gmap_unmap_by_gaddr(gmap, to + off);
+		/* Store new translation */
+		if (radix_tree_insert(&gmap->guest_to_host,
+				      (to + off) >> PMD_SHIFT,
+				      (void *) from + off))
+			break;
 	}
-	spin_unlock(&gmap->mm->page_table_lock);
-	up_read(&gmap->mm->mmap_sem);
+	up_write(&gmap->mm->mmap_sem);
 	if (flush)
 		gmap_flush_tlb(gmap);
-	return 0;
-
-out_unmap:
-	spin_unlock(&gmap->mm->page_table_lock);
-	up_read(&gmap->mm->mmap_sem);
+	if (off >= len)
+		return 0;
 	gmap_unmap_segment(gmap, to, len);
 	return -ENOMEM;
 }
 EXPORT_SYMBOL_GPL(gmap_map_segment);
 
-static unsigned long *gmap_table_walk(unsigned long address, struct gmap *gmap)
-{
-	unsigned long *table;
-
-	table = gmap->table + ((address >> 53) & 0x7ff);
-	if (unlikely(*table & _REGION_ENTRY_INVALID))
-		return ERR_PTR(-EFAULT);
-	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-	table = table + ((address >> 42) & 0x7ff);
-	if (unlikely(*table & _REGION_ENTRY_INVALID))
-		return ERR_PTR(-EFAULT);
-	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-	table = table + ((address >> 31) & 0x7ff);
-	if (unlikely(*table & _REGION_ENTRY_INVALID))
-		return ERR_PTR(-EFAULT);
-	table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-	table = table + ((address >> 20) & 0x7ff);
-	return table;
-}
-
 /**
  * __gmap_translate - translate a guest address to a user space address
- * @address: guest address
  * @gmap: pointer to guest mapping meta data structure
+ * @gaddr: guest address
  *
  * Returns user space address which corresponds to the guest address or
  * -EFAULT if no such mapping exists.
@@ -436,168 +448,161 @@ static unsigned long *gmap_table_walk(unsigned long address, struct gmap *gmap)
  * The mmap_sem of the mm that belongs to the address space must be held
  * when this function gets called.
  */
-unsigned long __gmap_translate(unsigned long address, struct gmap *gmap)
+unsigned long __gmap_translate(struct gmap *gmap, unsigned long gaddr)
 {
-	unsigned long *segment_ptr, vmaddr, segment;
-	struct gmap_pgtable *mp;
-	struct page *page;
+	unsigned long vmaddr;
 
-	current->thread.gmap_addr = address;
-	segment_ptr = gmap_table_walk(address, gmap);
-	if (IS_ERR(segment_ptr))
-		return PTR_ERR(segment_ptr);
-	/* Convert the gmap address to an mm address. */
-	segment = *segment_ptr;
-	if (!(segment & _SEGMENT_ENTRY_INVALID)) {
-		page = pfn_to_page(segment >> PAGE_SHIFT);
-		mp = (struct gmap_pgtable *) page->index;
-		return mp->vmaddr | (address & ~PMD_MASK);
-	} else if (segment & _SEGMENT_ENTRY_PROTECT) {
-		vmaddr = segment & _SEGMENT_ENTRY_ORIGIN;
-		return vmaddr | (address & ~PMD_MASK);
-	}
-	return -EFAULT;
+	vmaddr = (unsigned long)
+		radix_tree_lookup(&gmap->guest_to_host, gaddr >> PMD_SHIFT);
+	return vmaddr ? (vmaddr | (gaddr & ~PMD_MASK)) : -EFAULT;
 }
 EXPORT_SYMBOL_GPL(__gmap_translate);
 
 /**
  * gmap_translate - translate a guest address to a user space address
- * @address: guest address
  * @gmap: pointer to guest mapping meta data structure
+ * @gaddr: guest address
  *
  * Returns user space address which corresponds to the guest address or
  * -EFAULT if no such mapping exists.
  * This function does not establish potentially missing page table entries.
  */
-unsigned long gmap_translate(unsigned long address, struct gmap *gmap)
+unsigned long gmap_translate(struct gmap *gmap, unsigned long gaddr)
 {
 	unsigned long rc;
 
 	down_read(&gmap->mm->mmap_sem);
-	rc = __gmap_translate(address, gmap);
+	rc = __gmap_translate(gmap, gaddr);
 	up_read(&gmap->mm->mmap_sem);
 	return rc;
 }
 EXPORT_SYMBOL_GPL(gmap_translate);
 
-static int gmap_connect_pgtable(unsigned long address, unsigned long segment,
-				unsigned long *segment_ptr, struct gmap *gmap)
+/**
+ * gmap_unlink - disconnect a page table from the gmap shadow tables
+ * @gmap: pointer to guest mapping meta data structure
+ * @table: pointer to the host page table
+ * @vmaddr: vm address associated with the host page table
+ */
+static void gmap_unlink(struct mm_struct *mm, unsigned long *table,
+			unsigned long vmaddr)
+{
+	struct gmap *gmap;
+	int flush;
+
+	list_for_each_entry(gmap, &mm->context.gmap_list, list) {
+		flush = __gmap_unlink_by_vmaddr(gmap, vmaddr);
+		if (flush)
+			gmap_flush_tlb(gmap);
+	}
+}
+
+/**
+ * gmap_link - set up shadow page tables to connect a host to a guest address
+ * @gmap: pointer to guest mapping meta data structure
+ * @gaddr: guest address
+ * @vmaddr: vm address
+ *
+ * Returns 0 on success, -ENOMEM for out of memory conditions, and -EFAULT
+ * if the vm address is already mapped to a different guest segment.
+ * The mmap_sem of the mm that belongs to the address space must be held
+ * when this function gets called.
+ */
+int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr)
 {
-	unsigned long vmaddr;
-	struct vm_area_struct *vma;
-	struct gmap_pgtable *mp;
-	struct gmap_rmap *rmap;
 	struct mm_struct *mm;
-	struct page *page;
+	unsigned long *table;
+	spinlock_t *ptl;
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
+	int rc;
 
-	mm = gmap->mm;
-	vmaddr = segment & _SEGMENT_ENTRY_ORIGIN;
-	vma = find_vma(mm, vmaddr);
-	if (!vma || vma->vm_start > vmaddr)
-		return -EFAULT;
+	/* Create higher level tables in the gmap page table */
+	table = gmap->table;
+	if ((gmap->asce & _ASCE_TYPE_MASK) >= _ASCE_TYPE_REGION1) {
+		table += (gaddr >> 53) & 0x7ff;
+		if ((*table & _REGION_ENTRY_INVALID) &&
+		    gmap_alloc_table(gmap, table, _REGION2_ENTRY_EMPTY,
+				     gaddr & 0xffe0000000000000))
+			return -ENOMEM;
+		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
+	}
+	if ((gmap->asce & _ASCE_TYPE_MASK) >= _ASCE_TYPE_REGION2) {
+		table += (gaddr >> 42) & 0x7ff;
+		if ((*table & _REGION_ENTRY_INVALID) &&
+		    gmap_alloc_table(gmap, table, _REGION3_ENTRY_EMPTY,
+				     gaddr & 0xfffffc0000000000))
+			return -ENOMEM;
+		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
+	}
+	if ((gmap->asce & _ASCE_TYPE_MASK) >= _ASCE_TYPE_REGION3) {
+		table += (gaddr >> 31) & 0x7ff;
+		if ((*table & _REGION_ENTRY_INVALID) &&
+		    gmap_alloc_table(gmap, table, _SEGMENT_ENTRY_EMPTY,
+				     gaddr & 0xffffffff80000000))
+			return -ENOMEM;
+		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
+	}
+	table += (gaddr >> 20) & 0x7ff;
 	/* Walk the parent mm page table */
+	mm = gmap->mm;
 	pgd = pgd_offset(mm, vmaddr);
-	pud = pud_alloc(mm, pgd, vmaddr);
-	if (!pud)
-		return -ENOMEM;
-	pmd = pmd_alloc(mm, pud, vmaddr);
-	if (!pmd)
-		return -ENOMEM;
-	if (!pmd_present(*pmd) &&
-	    __pte_alloc(mm, vma, pmd, vmaddr))
-		return -ENOMEM;
+	VM_BUG_ON(pgd_none(*pgd));
+	pud = pud_offset(pgd, vmaddr);
+	VM_BUG_ON(pud_none(*pud));
+	pmd = pmd_offset(pud, vmaddr);
+	VM_BUG_ON(pmd_none(*pmd));
 	/* large pmds cannot yet be handled */
 	if (pmd_large(*pmd))
 		return -EFAULT;
-	/* pmd now points to a valid segment table entry. */
-	rmap = kmalloc(sizeof(*rmap), GFP_KERNEL|__GFP_REPEAT);
-	if (!rmap)
-		return -ENOMEM;
 	/* Link gmap segment table entry location to page table. */
-	page = pmd_page(*pmd);
-	mp = (struct gmap_pgtable *) page->index;
-	rmap->gmap = gmap;
-	rmap->entry = segment_ptr;
-	rmap->vmaddr = address & PMD_MASK;
-	spin_lock(&mm->page_table_lock);
-	if (*segment_ptr == segment) {
-		list_add(&rmap->list, &mp->mapper);
-		/* Set gmap segment table entry to page table. */
-		*segment_ptr = pmd_val(*pmd) & PAGE_MASK;
-		rmap = NULL;
-	}
-	spin_unlock(&mm->page_table_lock);
-	kfree(rmap);
-	return 0;
-}
-
-static void gmap_disconnect_pgtable(struct mm_struct *mm, unsigned long *table)
-{
-	struct gmap_rmap *rmap, *next;
-	struct gmap_pgtable *mp;
-	struct page *page;
-	int flush;
-
-	flush = 0;
-	spin_lock(&mm->page_table_lock);
-	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
-	mp = (struct gmap_pgtable *) page->index;
-	list_for_each_entry_safe(rmap, next, &mp->mapper, list) {
-		*rmap->entry = mp->vmaddr | (_SEGMENT_ENTRY_INVALID |
-					     _SEGMENT_ENTRY_PROTECT);
-		list_del(&rmap->list);
-		kfree(rmap);
-		flush = 1;
-	}
-	spin_unlock(&mm->page_table_lock);
-	if (flush)
-		__tlb_flush_global();
+	rc = radix_tree_preload(GFP_KERNEL);
+	if (rc)
+		return rc;
+	ptl = pmd_lock(mm, pmd);
+	spin_lock(&gmap->guest_table_lock);
+	if (*table == _SEGMENT_ENTRY_INVALID) {
+		rc = radix_tree_insert(&gmap->host_to_guest,
+				       vmaddr >> PMD_SHIFT, table);
+		if (!rc)
+			*table = pmd_val(*pmd);
+	} else
+		rc = 0;
+	spin_unlock(&gmap->guest_table_lock);
+	spin_unlock(ptl);
+	radix_tree_preload_end();
+	return rc;
 }
 
-/*
- * this function is assumed to be called with mmap_sem held
+/**
+ * gmap_fault - resolve a fault on a guest address
+ * @gmap: pointer to guest mapping meta data structure
+ * @gaddr: guest address
+ * @fault_flags: flags to pass down to handle_mm_fault()
+ *
+ * Returns 0 on success, -ENOMEM for out of memory conditions, and -EFAULT
+ * if the vm address is already mapped to a different guest segment.
  */
-unsigned long __gmap_fault(unsigned long address, struct gmap *gmap)
+int gmap_fault(struct gmap *gmap, unsigned long gaddr,
+	       unsigned int fault_flags)
 {
-	unsigned long *segment_ptr, segment;
-	struct gmap_pgtable *mp;
-	struct page *page;
+	unsigned long vmaddr;
 	int rc;
 
-	current->thread.gmap_addr = address;
-	segment_ptr = gmap_table_walk(address, gmap);
-	if (IS_ERR(segment_ptr))
-		return -EFAULT;
-	/* Convert the gmap address to an mm address. */
-	while (1) {
-		segment = *segment_ptr;
-		if (!(segment & _SEGMENT_ENTRY_INVALID)) {
-			/* Page table is present */
-			page = pfn_to_page(segment >> PAGE_SHIFT);
-			mp = (struct gmap_pgtable *) page->index;
-			return mp->vmaddr | (address & ~PMD_MASK);
-		}
-		if (!(segment & _SEGMENT_ENTRY_PROTECT))
-			/* Nothing mapped in the gmap address space. */
-			break;
-		rc = gmap_connect_pgtable(address, segment, segment_ptr, gmap);
-		if (rc)
-			return rc;
-	}
-	return -EFAULT;
-}
-
-unsigned long gmap_fault(unsigned long address, struct gmap *gmap)
-{
-	unsigned long rc;
-
 	down_read(&gmap->mm->mmap_sem);
-	rc = __gmap_fault(address, gmap);
+	vmaddr = __gmap_translate(gmap, gaddr);
+	if (IS_ERR_VALUE(vmaddr)) {
+		rc = vmaddr;
+		goto out_up;
+	}
+	if (fixup_user_fault(current, gmap->mm, vmaddr, fault_flags)) {
+		rc = -EFAULT;
+		goto out_up;
+	}
+	rc = __gmap_link(gmap, gaddr, vmaddr);
+out_up:
 	up_read(&gmap->mm->mmap_sem);
-
 	return rc;
 }
 EXPORT_SYMBOL_GPL(gmap_fault);
@@ -617,17 +622,24 @@ static void gmap_zap_swap_entry(swp_entry_t entry, struct mm_struct *mm)
 	free_swap_and_cache(entry);
 }
 
-/**
- * The mm->mmap_sem lock must be held
+/*
+ * this function is assumed to be called with mmap_sem held
  */
-static void gmap_zap_unused(struct mm_struct *mm, unsigned long address)
+void __gmap_zap(struct gmap *gmap, unsigned long gaddr)
 {
-	unsigned long ptev, pgstev;
+	unsigned long vmaddr, ptev, pgstev;
+	pte_t *ptep, pte;
 	spinlock_t *ptl;
 	pgste_t pgste;
-	pte_t *ptep, pte;
 
-	ptep = get_locked_pte(mm, address, &ptl);
+	/* Find the vm address for the guest address */
+	vmaddr = (unsigned long) radix_tree_lookup(&gmap->guest_to_host,
+						   gaddr >> PMD_SHIFT);
+	if (!vmaddr)
+		return;
+	vmaddr |= gaddr & ~PMD_MASK;
+	/* Get pointer to the page table entry */
+	ptep = get_locked_pte(gmap->mm, vmaddr, &ptl);
 	if (unlikely(!ptep))
 		return;
 	pte = *ptep;
@@ -639,87 +651,34 @@ static void gmap_zap_unused(struct mm_struct *mm, unsigned long address)
 	ptev = pte_val(pte);
 	if (((pgstev & _PGSTE_GPS_USAGE_MASK) == _PGSTE_GPS_USAGE_UNUSED) ||
 	    ((pgstev & _PGSTE_GPS_ZERO) && (ptev & _PAGE_INVALID))) {
-		gmap_zap_swap_entry(pte_to_swp_entry(pte), mm);
-		pte_clear(mm, address, ptep);
+		gmap_zap_swap_entry(pte_to_swp_entry(pte), gmap->mm);
+		pte_clear(gmap->mm, vmaddr, ptep);
 	}
 	pgste_set_unlock(ptep, pgste);
 out_pte:
 	pte_unmap_unlock(*ptep, ptl);
 }
-
-/*
- * this function is assumed to be called with mmap_sem held
- */
-void __gmap_zap(unsigned long address, struct gmap *gmap)
-{
-	unsigned long *table, *segment_ptr;
-	unsigned long segment, pgstev, ptev;
-	struct gmap_pgtable *mp;
-	struct page *page;
-
-	segment_ptr = gmap_table_walk(address, gmap);
-	if (IS_ERR(segment_ptr))
-		return;
-	segment = *segment_ptr;
-	if (segment & _SEGMENT_ENTRY_INVALID)
-		return;
-	page = pfn_to_page(segment >> PAGE_SHIFT);
-	mp = (struct gmap_pgtable *) page->index;
-	address = mp->vmaddr | (address & ~PMD_MASK);
-	/* Page table is present */
-	table = (unsigned long *)(segment & _SEGMENT_ENTRY_ORIGIN);
-	table = table + ((address >> 12) & 0xff);
-	pgstev = table[PTRS_PER_PTE];
-	ptev = table[0];
-	/* quick check, checked again with locks held */
-	if (((pgstev & _PGSTE_GPS_USAGE_MASK) == _PGSTE_GPS_USAGE_UNUSED) ||
-	    ((pgstev & _PGSTE_GPS_ZERO) && (ptev & _PAGE_INVALID)))
-		gmap_zap_unused(gmap->mm, address);
-}
 EXPORT_SYMBOL_GPL(__gmap_zap);
 
-void gmap_discard(unsigned long from, unsigned long to, struct gmap *gmap)
+void gmap_discard(struct gmap *gmap, unsigned long from, unsigned long to)
 {
-
-	unsigned long *table, address, size;
+	unsigned long gaddr, vmaddr, size;
 	struct vm_area_struct *vma;
-	struct gmap_pgtable *mp;
-	struct page *page;
 
 	down_read(&gmap->mm->mmap_sem);
-	address = from;
-	while (address < to) {
-		/* Walk the gmap address space page table */
-		table = gmap->table + ((address >> 53) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INVALID)) {
-			address = (address + PMD_SIZE) & PMD_MASK;
-			continue;
-		}
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + ((address >> 42) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INVALID)) {
-			address = (address + PMD_SIZE) & PMD_MASK;
+	for (gaddr = from; gaddr < to;
+	     gaddr = (gaddr + PMD_SIZE) & PMD_MASK) {
+		/* Find the vm address for the guest address */
+		vmaddr = (unsigned long)
+			radix_tree_lookup(&gmap->guest_to_host,
+					  gaddr >> PMD_SHIFT);
+		if (!vmaddr)
 			continue;
-		}
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + ((address >> 31) & 0x7ff);
-		if (unlikely(*table & _REGION_ENTRY_INVALID)) {
-			address = (address + PMD_SIZE) & PMD_MASK;
-			continue;
-		}
-		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
-		table = table + ((address >> 20) & 0x7ff);
-		if (unlikely(*table & _SEGMENT_ENTRY_INVALID)) {
-			address = (address + PMD_SIZE) & PMD_MASK;
-			continue;
-		}
-		page = pfn_to_page(*table >> PAGE_SHIFT);
-		mp = (struct gmap_pgtable *) page->index;
-		vma = find_vma(gmap->mm, mp->vmaddr);
-		size = min(to - address, PMD_SIZE - (address & ~PMD_MASK));
-		zap_page_range(vma, mp->vmaddr | (address & ~PMD_MASK),
-			       size, NULL);
-		address = (address + PMD_SIZE) & PMD_MASK;
+		vmaddr |= gaddr & ~PMD_MASK;
+		/* Find vma in the parent mm */
+		vma = find_vma(gmap->mm, vmaddr);
+		size = min(to - gaddr, PMD_SIZE - (gaddr & ~PMD_MASK));
+		zap_page_range(vma, vmaddr, size, NULL);
 	}
 	up_read(&gmap->mm->mmap_sem);
 }
@@ -755,7 +714,7 @@ EXPORT_SYMBOL_GPL(gmap_unregister_ipte_notifier);
 /**
  * gmap_ipte_notify - mark a range of ptes for invalidation notification
  * @gmap: pointer to guest mapping meta data structure
- * @start: virtual address in the guest address space
+ * @gaddr: virtual address in the guest address space
  * @len: size of area
  *
  * Returns 0 if for each page in the given range a gmap mapping exists and
@@ -763,7 +722,7 @@ EXPORT_SYMBOL_GPL(gmap_unregister_ipte_notifier);
  * for one or more pages -EFAULT is returned. If no memory could be allocated
  * -ENOMEM is returned. This function establishes missing page table entries.
  */
-int gmap_ipte_notify(struct gmap *gmap, unsigned long start, unsigned long len)
+int gmap_ipte_notify(struct gmap *gmap, unsigned long gaddr, unsigned long len)
 {
 	unsigned long addr;
 	spinlock_t *ptl;
@@ -771,12 +730,12 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long start, unsigned long len)
 	pgste_t pgste;
 	int rc = 0;
 
-	if ((start & ~PAGE_MASK) || (len & ~PAGE_MASK))
+	if ((gaddr & ~PAGE_MASK) || (len & ~PAGE_MASK))
 		return -EINVAL;
 	down_read(&gmap->mm->mmap_sem);
 	while (len) {
 		/* Convert gmap address and connect the page tables */
-		addr = __gmap_fault(start, gmap);
+		addr = __gmap_translate(gmap, gaddr);
 		if (IS_ERR_VALUE(addr)) {
 			rc = addr;
 			break;
@@ -786,6 +745,9 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long start, unsigned long len)
 			rc = -EFAULT;
 			break;
 		}
+		rc = __gmap_link(gmap, gaddr, addr);
+		if (rc)
+			break;
 		/* Walk the process page table, lock and get pte pointer */
 		ptep = get_locked_pte(gmap->mm, addr, &ptl);
 		if (unlikely(!ptep))
@@ -796,7 +758,7 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long start, unsigned long len)
 			pgste = pgste_get_lock(ptep);
 			pgste_val(pgste) |= PGSTE_IN_BIT;
 			pgste_set_unlock(ptep, pgste);
-			start += PAGE_SIZE;
+			gaddr += PAGE_SIZE;
 			len -= PAGE_SIZE;
 		}
 		spin_unlock(ptl);
@@ -809,28 +771,30 @@ EXPORT_SYMBOL_GPL(gmap_ipte_notify);
 /**
  * gmap_do_ipte_notify - call all invalidation callbacks for a specific pte.
  * @mm: pointer to the process mm_struct
+ * @addr: virtual address in the process address space
  * @pte: pointer to the page table entry
  *
  * This function is assumed to be called with the page table lock held
  * for the pte to notify.
  */
-void gmap_do_ipte_notify(struct mm_struct *mm, pte_t *pte)
+void gmap_do_ipte_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte)
 {
-	unsigned long segment_offset;
+	unsigned long offset, gaddr;
+	unsigned long *table;
 	struct gmap_notifier *nb;
-	struct gmap_pgtable *mp;
-	struct gmap_rmap *rmap;
-	struct page *page;
+	struct gmap *gmap;
 
-	segment_offset = ((unsigned long) pte) & (255 * sizeof(pte_t));
-	segment_offset = segment_offset * (4096 / sizeof(pte_t));
-	page = pfn_to_page(__pa(pte) >> PAGE_SHIFT);
-	mp = (struct gmap_pgtable *) page->index;
+	offset = ((unsigned long) pte) & (255 * sizeof(pte_t));
+	offset = offset * (4096 / sizeof(pte_t));
 	spin_lock(&gmap_notifier_lock);
-	list_for_each_entry(rmap, &mp->mapper, list) {
+	list_for_each_entry(gmap, &mm->context.gmap_list, list) {
+		table = radix_tree_lookup(&gmap->host_to_guest,
+					  vmaddr >> PMD_SHIFT);
+		if (!table)
+			continue;
+		gaddr = __gmap_segment_gaddr(table) + offset;
 		list_for_each_entry(nb, &gmap_notifier_list, list)
-			nb->notifier_call(rmap->gmap,
-					  rmap->vmaddr + segment_offset);
+			nb->notifier_call(gmap, gaddr);
 	}
 	spin_unlock(&gmap_notifier_lock);
 }
@@ -841,29 +805,18 @@ static inline int page_table_with_pgste(struct page *page)
 	return atomic_read(&page->_mapcount) == 0;
 }
 
-static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
-						    unsigned long vmaddr)
+static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm)
 {
 	struct page *page;
 	unsigned long *table;
-	struct gmap_pgtable *mp;
 
 	page = alloc_page(GFP_KERNEL|__GFP_REPEAT);
 	if (!page)
 		return NULL;
-	mp = kmalloc(sizeof(*mp), GFP_KERNEL|__GFP_REPEAT);
-	if (!mp) {
-		__free_page(page);
-		return NULL;
-	}
 	if (!pgtable_page_ctor(page)) {
-		kfree(mp);
 		__free_page(page);
 		return NULL;
 	}
-	mp->vmaddr = vmaddr & PMD_MASK;
-	INIT_LIST_HEAD(&mp->mapper);
-	page->index = (unsigned long) mp;
 	atomic_set(&page->_mapcount, 0);
 	table = (unsigned long *) page_to_phys(page);
 	clear_table(table, _PAGE_INVALID, PAGE_SIZE/2);
@@ -874,14 +827,10 @@ static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
 static inline void page_table_free_pgste(unsigned long *table)
 {
 	struct page *page;
-	struct gmap_pgtable *mp;
 
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
-	mp = (struct gmap_pgtable *) page->index;
-	BUG_ON(!list_empty(&mp->mapper));
 	pgtable_page_dtor(page);
 	atomic_set(&page->_mapcount, -1);
-	kfree(mp);
 	__free_page(page);
 }
 
@@ -986,11 +935,21 @@ int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 	pte_t *ptep;
 
 	down_read(&mm->mmap_sem);
+retry:
 	ptep = get_locked_pte(current->mm, addr, &ptl);
 	if (unlikely(!ptep)) {
 		up_read(&mm->mmap_sem);
 		return -EFAULT;
 	}
+	if (!(pte_val(*ptep) & _PAGE_INVALID) &&
+	     (pte_val(*ptep) & _PAGE_PROTECT)) {
+		pte_unmap_unlock(*ptep, ptl);
+		if (fixup_user_fault(current, mm, addr, FAULT_FLAG_WRITE)) {
+			up_read(&mm->mmap_sem);
+			return -EFAULT;
+		}
+		goto retry;
+	}
 
 	new = old = pgste_get_lock(ptep);
 	pgste_val(new) &= ~(PGSTE_GR_BIT | PGSTE_GC_BIT |
@@ -1028,8 +987,7 @@ static inline int page_table_with_pgste(struct page *page)
 	return 0;
 }
 
-static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
-						    unsigned long vmaddr)
+static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm)
 {
 	return NULL;
 }
@@ -1043,8 +1001,8 @@ static inline void page_table_free_pgste(unsigned long *table)
 {
 }
 
-static inline void gmap_disconnect_pgtable(struct mm_struct *mm,
-					   unsigned long *table)
+static inline void gmap_unlink(struct mm_struct *mm, unsigned long *table,
+			unsigned long vmaddr)
 {
 }
 
@@ -1064,14 +1022,14 @@ static inline unsigned int atomic_xor_bits(atomic_t *v, unsigned int bits)
 /*
  * page table entry allocation/free routines.
  */
-unsigned long *page_table_alloc(struct mm_struct *mm, unsigned long vmaddr)
+unsigned long *page_table_alloc(struct mm_struct *mm)
 {
 	unsigned long *uninitialized_var(table);
 	struct page *uninitialized_var(page);
 	unsigned int mask, bit;
 
 	if (mm_has_pgste(mm))
-		return page_table_alloc_pgste(mm, vmaddr);
+		return page_table_alloc_pgste(mm);
 	/* Allocate fragments of a 4K page as 1K/2K page table */
 	spin_lock_bh(&mm->context.list_lock);
 	mask = FRAG_MASK;
@@ -1113,10 +1071,8 @@ void page_table_free(struct mm_struct *mm, unsigned long *table)
 	unsigned int bit, mask;
 
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
-	if (page_table_with_pgste(page)) {
-		gmap_disconnect_pgtable(mm, table);
+	if (page_table_with_pgste(page))
 		return page_table_free_pgste(table);
-	}
 	/* Free 1K/2K page table fragment of a 4K page */
 	bit = 1 << ((__pa(table) & ~PAGE_MASK)/(PTRS_PER_PTE*sizeof(pte_t)));
 	spin_lock_bh(&mm->context.list_lock);
@@ -1148,7 +1104,8 @@ static void __page_table_free_rcu(void *table, unsigned bit)
 	}
 }
 
-void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table)
+void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table,
+			 unsigned long vmaddr)
 {
 	struct mm_struct *mm;
 	struct page *page;
@@ -1157,7 +1114,7 @@ void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table)
 	mm = tlb->mm;
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
 	if (page_table_with_pgste(page)) {
-		gmap_disconnect_pgtable(mm, table);
+		gmap_unlink(mm, table, vmaddr);
 		table = (unsigned long *) (__pa(table) | FRAG_MASK);
 		tlb_remove_table(tlb, table);
 		return;
@@ -1293,7 +1250,7 @@ again:
 		if (page_table_with_pgste(page))
 			continue;
 		/* Allocate new page table with pgstes */
-		new = page_table_alloc_pgste(mm, addr);
+		new = page_table_alloc_pgste(mm);
 		if (!new)
 			return -ENOMEM;
 
@@ -1308,7 +1265,7 @@ again:
 			/* Establish new table */
 			pmd_populate(mm, pmd, (pte_t *) new);
 			/* Free old table with rcu, there might be a walker! */
-			page_table_free_rcu(tlb, table);
+			page_table_free_rcu(tlb, table, addr);
 			new = NULL;
 		}
 		spin_unlock(ptl);

arch/s390/mm/vmem.c

@@ -65,7 +65,7 @@ static pte_t __ref *vmem_pte_alloc(unsigned long address)
 	pte_t *pte;
 
 	if (slab_is_available())
-		pte = (pte_t *) page_table_alloc(&init_mm, address);
+		pte = (pte_t *) page_table_alloc(&init_mm);
 	else
 		pte = alloc_bootmem_align(PTRS_PER_PTE * sizeof(pte_t),
 					  PTRS_PER_PTE * sizeof(pte_t));

arch/x86/include/asm/kvm_host.h

@@ -99,10 +99,6 @@ static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
 
 #define ASYNC_PF_PER_VCPU 64
 
-struct kvm_vcpu;
-struct kvm;
-struct kvm_async_pf;
-
 enum kvm_reg {
 	VCPU_REGS_RAX = 0,
 	VCPU_REGS_RCX = 1,
@@ -266,7 +262,8 @@ struct kvm_mmu {
 				  struct x86_exception *fault);
 	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva, u32 access,
 			    struct x86_exception *exception);
-	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access);
+	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
+			       struct x86_exception *exception);
 	int (*sync_page)(struct kvm_vcpu *vcpu,
 			 struct kvm_mmu_page *sp);
 	void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva);
@@ -481,6 +478,7 @@ struct kvm_vcpu_arch {
 	u64 mmio_gva;
 	unsigned access;
 	gfn_t mmio_gfn;
+	u64 mmio_gen;
 
 	struct kvm_pmu pmu;
 
@@ -580,7 +578,6 @@ struct kvm_arch {
 
 	gpa_t wall_clock;
 
-	struct page *ept_identity_pagetable;
 	bool ept_identity_pagetable_done;
 	gpa_t ept_identity_map_addr;
 
@@ -665,8 +662,8 @@ struct msr_data {
 struct kvm_x86_ops {
 	int (*cpu_has_kvm_support)(void);          /* __init */
 	int (*disabled_by_bios)(void);             /* __init */
-	int (*hardware_enable)(void *dummy);
-	void (*hardware_disable)(void *dummy);
+	int (*hardware_enable)(void);
+	void (*hardware_disable)(void);
 	void (*check_processor_compatibility)(void *rtn);
 	int (*hardware_setup)(void);               /* __init */
 	void (*hardware_unsetup)(void);            /* __exit */
@@ -896,7 +893,6 @@ void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
 int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
 			    gfn_t gfn, void *data, int offset, int len,
 			    u32 access);
-void kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
 bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl);
 
 static inline int __kvm_irq_line_state(unsigned long *irq_state,
@@ -927,7 +923,8 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
 int kvm_mmu_load(struct kvm_vcpu *vcpu);
 void kvm_mmu_unload(struct kvm_vcpu *vcpu);
 void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
-gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access);
+gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
+			   struct x86_exception *exception);
 gpa_t kvm_mmu_gva_to_gpa_read(struct kvm_vcpu *vcpu, gva_t gva,
 			      struct x86_exception *exception);
 gpa_t kvm_mmu_gva_to_gpa_fetch(struct kvm_vcpu *vcpu, gva_t gva,
@@ -947,7 +944,8 @@ void kvm_mmu_new_cr3(struct kvm_vcpu *vcpu);
 void kvm_enable_tdp(void);
 void kvm_disable_tdp(void);
 
-static inline gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access)
+static inline gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
+				  struct x86_exception *exception)
 {
 	return gpa;
 }

arch/x86/kvm/cpuid.h

@@ -88,6 +88,14 @@ static inline bool guest_cpuid_has_x2apic(struct kvm_vcpu *vcpu)
 	return best && (best->ecx & bit(X86_FEATURE_X2APIC));
 }
 
+static inline bool guest_cpuid_is_amd(struct kvm_vcpu *vcpu)
+{
+	struct kvm_cpuid_entry2 *best;
+
+	best = kvm_find_cpuid_entry(vcpu, 0, 0);
+	return best && best->ebx == X86EMUL_CPUID_VENDOR_AuthenticAMD_ebx;
+}
+
 static inline bool guest_cpuid_has_gbpages(struct kvm_vcpu *vcpu)
 {
 	struct kvm_cpuid_entry2 *best;

arch/x86/kvm/emulate.c

@@ -3139,12 +3139,8 @@ static int em_clts(struct x86_emulate_ctxt *ctxt)
 
 static int em_vmcall(struct x86_emulate_ctxt *ctxt)
 {
-	int rc;
-
-	if (ctxt->modrm_mod != 3 || ctxt->modrm_rm != 1)
-		return X86EMUL_UNHANDLEABLE;
+	int rc = ctxt->ops->fix_hypercall(ctxt);
 
-	rc = ctxt->ops->fix_hypercall(ctxt);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
@@ -3562,6 +3558,12 @@ static int check_perm_out(struct x86_emulate_ctxt *ctxt)
 		F2bv(((_f) | DstReg | SrcMem | ModRM) & ~Lock, _e),	\
 		F2bv(((_f) & ~Lock) | DstAcc | SrcImm, _e)
 
+static const struct opcode group7_rm0[] = {
+	N,
+	I(SrcNone | Priv | EmulateOnUD,	em_vmcall),
+	N, N, N, N, N, N,
+};
+
 static const struct opcode group7_rm1[] = {
 	DI(SrcNone | Priv, monitor),
 	DI(SrcNone | Priv, mwait),
@@ -3655,7 +3657,7 @@ static const struct group_dual group7 = { {
 	II(SrcMem16 | Mov | Priv,		em_lmsw, lmsw),
 	II(SrcMem | ByteOp | Priv | NoAccess,	em_invlpg, invlpg),
 }, {
-	I(SrcNone | Priv | EmulateOnUD,	em_vmcall),
+	EXT(0, group7_rm0),
 	EXT(0, group7_rm1),
 	N, EXT(0, group7_rm3),
 	II(SrcNone | DstMem | Mov,		em_smsw, smsw), N,
@@ -3686,14 +3688,18 @@ static const struct gprefix pfx_0f_6f_0f_7f = {
 	I(Mmx, em_mov), I(Sse | Aligned, em_mov), N, I(Sse | Unaligned, em_mov),
 };
 
-static const struct gprefix pfx_vmovntpx = {
-	I(0, em_mov), N, N, N,
+static const struct gprefix pfx_0f_2b = {
+	I(0, em_mov), I(0, em_mov), N, N,
 };
 
 static const struct gprefix pfx_0f_28_0f_29 = {
 	I(Aligned, em_mov), I(Aligned, em_mov), N, N,
 };
 
+static const struct gprefix pfx_0f_e7 = {
+	N, I(Sse, em_mov), N, N,
+};
+
 static const struct escape escape_d9 = { {
 	N, N, N, N, N, N, N, I(DstMem, em_fnstcw),
 }, {
@@ -3900,7 +3906,7 @@ static const struct opcode twobyte_table[256] = {
 	N, N, N, N,
 	GP(ModRM | DstReg | SrcMem | Mov | Sse, &pfx_0f_28_0f_29),
 	GP(ModRM | DstMem | SrcReg | Mov | Sse, &pfx_0f_28_0f_29),
-	N, GP(ModRM | DstMem | SrcReg | Sse | Mov | Aligned, &pfx_vmovntpx),
+	N, GP(ModRM | DstMem | SrcReg | Mov | Sse, &pfx_0f_2b),
 	N, N, N, N,
 	/* 0x30 - 0x3F */
 	II(ImplicitOps | Priv, em_wrmsr, wrmsr),
@@ -3964,7 +3970,8 @@ static const struct opcode twobyte_table[256] = {
 	/* 0xD0 - 0xDF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
 	/* 0xE0 - 0xEF */
-	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
+	N, N, N, N, N, N, N, GP(SrcReg | DstMem | ModRM | Mov, &pfx_0f_e7),
+	N, N, N, N, N, N, N, N,
 	/* 0xF0 - 0xFF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N
 };

arch/x86/kvm/lapic.c

@@ -709,6 +709,8 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
 	int result = 0;
 	struct kvm_vcpu *vcpu = apic->vcpu;
 
+	trace_kvm_apic_accept_irq(vcpu->vcpu_id, delivery_mode,
+				  trig_mode, vector);
 	switch (delivery_mode) {
 	case APIC_DM_LOWEST:
 		vcpu->arch.apic_arb_prio++;
@@ -730,8 +732,6 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
 			kvm_make_request(KVM_REQ_EVENT, vcpu);
 			kvm_vcpu_kick(vcpu);
 		}
-		trace_kvm_apic_accept_irq(vcpu->vcpu_id, delivery_mode,
-					  trig_mode, vector, false);
 		break;
 
 	case APIC_DM_REMRD:

arch/x86/kvm/mmu.c

@@ -199,16 +199,20 @@ void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask)
 EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask);
 
 /*
- * spte bits of bit 3 ~ bit 11 are used as low 9 bits of generation number,
- * the bits of bits 52 ~ bit 61 are used as high 10 bits of generation
- * number.
+ * the low bit of the generation number is always presumed to be zero.
+ * This disables mmio caching during memslot updates.  The concept is
+ * similar to a seqcount but instead of retrying the access we just punt
+ * and ignore the cache.
+ *
+ * spte bits 3-11 are used as bits 1-9 of the generation number,
+ * the bits 52-61 are used as bits 10-19 of the generation number.
  */
-#define MMIO_SPTE_GEN_LOW_SHIFT		3
+#define MMIO_SPTE_GEN_LOW_SHIFT		2
 #define MMIO_SPTE_GEN_HIGH_SHIFT	52
 
-#define MMIO_GEN_SHIFT			19
-#define MMIO_GEN_LOW_SHIFT		9
-#define MMIO_GEN_LOW_MASK		((1 << MMIO_GEN_LOW_SHIFT) - 1)
+#define MMIO_GEN_SHIFT			20
+#define MMIO_GEN_LOW_SHIFT		10
+#define MMIO_GEN_LOW_MASK		((1 << MMIO_GEN_LOW_SHIFT) - 2)
 #define MMIO_GEN_MASK			((1 << MMIO_GEN_SHIFT) - 1)
 #define MMIO_MAX_GEN			((1 << MMIO_GEN_SHIFT) - 1)
 
@@ -236,12 +240,7 @@ static unsigned int get_mmio_spte_generation(u64 spte)
 
 static unsigned int kvm_current_mmio_generation(struct kvm *kvm)
 {
-	/*
-	 * Init kvm generation close to MMIO_MAX_GEN to easily test the
-	 * code of handling generation number wrap-around.
-	 */
-	return (kvm_memslots(kvm)->generation +
-		      MMIO_MAX_GEN - 150) & MMIO_GEN_MASK;
+	return kvm_memslots(kvm)->generation & MMIO_GEN_MASK;
 }
 
 static void mark_mmio_spte(struct kvm *kvm, u64 *sptep, u64 gfn,
@@ -296,11 +295,6 @@ static bool check_mmio_spte(struct kvm *kvm, u64 spte)
 	return likely(kvm_gen == spte_gen);
 }
 
-static inline u64 rsvd_bits(int s, int e)
-{
-	return ((1ULL << (e - s + 1)) - 1) << s;
-}
-
 void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
 		u64 dirty_mask, u64 nx_mask, u64 x_mask)
 {
@@ -3163,7 +3157,7 @@ static void mmu_sync_roots(struct kvm_vcpu *vcpu)
 	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
 		return;
 
-	vcpu_clear_mmio_info(vcpu, ~0ul);
+	vcpu_clear_mmio_info(vcpu, MMIO_GVA_ANY);
 	kvm_mmu_audit(vcpu, AUDIT_PRE_SYNC);
 	if (vcpu->arch.mmu.root_level == PT64_ROOT_LEVEL) {
 		hpa_t root = vcpu->arch.mmu.root_hpa;
@@ -3206,7 +3200,7 @@ static gpa_t nonpaging_gva_to_gpa_nested(struct kvm_vcpu *vcpu, gva_t vaddr,
 {
 	if (exception)
 		exception->error_code = 0;
-	return vcpu->arch.nested_mmu.translate_gpa(vcpu, vaddr, access);
+	return vcpu->arch.nested_mmu.translate_gpa(vcpu, vaddr, access, exception);
 }
 
 static bool quickly_check_mmio_pf(struct kvm_vcpu *vcpu, u64 addr, bool direct)
@@ -3518,6 +3512,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu,
 	int maxphyaddr = cpuid_maxphyaddr(vcpu);
 	u64 exb_bit_rsvd = 0;
 	u64 gbpages_bit_rsvd = 0;
+	u64 nonleaf_bit8_rsvd = 0;
 
 	context->bad_mt_xwr = 0;
 
@@ -3525,6 +3520,14 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu,
 		exb_bit_rsvd = rsvd_bits(63, 63);
 	if (!guest_cpuid_has_gbpages(vcpu))
 		gbpages_bit_rsvd = rsvd_bits(7, 7);
+
+	/*
+	 * Non-leaf PML4Es and PDPEs reserve bit 8 (which would be the G bit for
+	 * leaf entries) on AMD CPUs only.
+	 */
+	if (guest_cpuid_is_amd(vcpu))
+		nonleaf_bit8_rsvd = rsvd_bits(8, 8);
+
 	switch (context->root_level) {
 	case PT32_ROOT_LEVEL:
 		/* no rsvd bits for 2 level 4K page table entries */
@@ -3559,9 +3562,9 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu,
 		break;
 	case PT64_ROOT_LEVEL:
 		context->rsvd_bits_mask[0][3] = exb_bit_rsvd |
-			rsvd_bits(maxphyaddr, 51) | rsvd_bits(7, 7);
+			nonleaf_bit8_rsvd | rsvd_bits(7, 7) | rsvd_bits(maxphyaddr, 51);
 		context->rsvd_bits_mask[0][2] = exb_bit_rsvd |
-			gbpages_bit_rsvd | rsvd_bits(maxphyaddr, 51);
+			nonleaf_bit8_rsvd | gbpages_bit_rsvd | rsvd_bits(maxphyaddr, 51);
 		context->rsvd_bits_mask[0][1] = exb_bit_rsvd |
 			rsvd_bits(maxphyaddr, 51);
 		context->rsvd_bits_mask[0][0] = exb_bit_rsvd |
@@ -4433,7 +4436,7 @@ void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm)
 	 * The very rare case: if the generation-number is round,
 	 * zap all shadow pages.
 	 */
-	if (unlikely(kvm_current_mmio_generation(kvm) >= MMIO_MAX_GEN)) {
+	if (unlikely(kvm_current_mmio_generation(kvm) == 0)) {
 		printk_ratelimited(KERN_INFO "kvm: zapping shadow pages for mmio generation wraparound\n");
 		kvm_mmu_invalidate_zap_all_pages(kvm);
 	}

arch/x86/kvm/mmu.h

@@ -56,6 +56,11 @@
 #define PFERR_RSVD_MASK (1U << PFERR_RSVD_BIT)
 #define PFERR_FETCH_MASK (1U << PFERR_FETCH_BIT)
 
+static inline u64 rsvd_bits(int s, int e)
+{
+	return ((1ULL << (e - s + 1)) - 1) << s;
+}
+
 int kvm_mmu_get_spte_hierarchy(struct kvm_vcpu *vcpu, u64 addr, u64 sptes[4]);
 void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask);
 

arch/x86/kvm/paging_tmpl.h

@@ -321,9 +321,22 @@ retry_walk:
 		walker->pte_gpa[walker->level - 1] = pte_gpa;
 
 		real_gfn = mmu->translate_gpa(vcpu, gfn_to_gpa(table_gfn),
-					      PFERR_USER_MASK|PFERR_WRITE_MASK);
+					      PFERR_USER_MASK|PFERR_WRITE_MASK,
+					      &walker->fault);
+
+		/*
+		 * FIXME: This can happen if emulation (for of an INS/OUTS
+		 * instruction) triggers a nested page fault.  The exit
+		 * qualification / exit info field will incorrectly have
+		 * "guest page access" as the nested page fault's cause,
+		 * instead of "guest page structure access".  To fix this,
+		 * the x86_exception struct should be augmented with enough
+		 * information to fix the exit_qualification or exit_info_1
+		 * fields.
+		 */
 		if (unlikely(real_gfn == UNMAPPED_GVA))
-			goto error;
+			return 0;
+
 		real_gfn = gpa_to_gfn(real_gfn);
 
 		host_addr = gfn_to_hva_prot(vcpu->kvm, real_gfn,
@@ -364,7 +377,7 @@ retry_walk:
 	if (PTTYPE == 32 && walker->level == PT_DIRECTORY_LEVEL && is_cpuid_PSE36())
 		gfn += pse36_gfn_delta(pte);
 
-	real_gpa = mmu->translate_gpa(vcpu, gfn_to_gpa(gfn), access);
+	real_gpa = mmu->translate_gpa(vcpu, gfn_to_gpa(gfn), access, &walker->fault);
 	if (real_gpa == UNMAPPED_GVA)
 		return 0;
 

arch/x86/kvm/svm.c

@@ -622,7 +622,7 @@ static int has_svm(void)
 	return 1;
 }
 
-static void svm_hardware_disable(void *garbage)
+static void svm_hardware_disable(void)
 {
 	/* Make sure we clean up behind us */
 	if (static_cpu_has(X86_FEATURE_TSCRATEMSR))
@@ -633,7 +633,7 @@ static void svm_hardware_disable(void *garbage)
 	amd_pmu_disable_virt();
 }
 
-static int svm_hardware_enable(void *garbage)
+static int svm_hardware_enable(void)
 {
 
 	struct svm_cpu_data *sd;
@@ -1257,7 +1257,8 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
 	svm->asid_generation = 0;
 	init_vmcb(svm);
 
-	svm->vcpu.arch.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
+	svm->vcpu.arch.apic_base = APIC_DEFAULT_PHYS_BASE |
+				   MSR_IA32_APICBASE_ENABLE;
 	if (kvm_vcpu_is_bsp(&svm->vcpu))
 		svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP;
 
@@ -1974,10 +1975,26 @@ static void nested_svm_inject_npf_exit(struct kvm_vcpu *vcpu,
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
 
-	svm->vmcb->control.exit_code = SVM_EXIT_NPF;
-	svm->vmcb->control.exit_code_hi = 0;
-	svm->vmcb->control.exit_info_1 = fault->error_code;
-	svm->vmcb->control.exit_info_2 = fault->address;
+	if (svm->vmcb->control.exit_code != SVM_EXIT_NPF) {
+		/*
+		 * TODO: track the cause of the nested page fault, and
+		 * correctly fill in the high bits of exit_info_1.
+		 */
+		svm->vmcb->control.exit_code = SVM_EXIT_NPF;
+		svm->vmcb->control.exit_code_hi = 0;
+		svm->vmcb->control.exit_info_1 = (1ULL << 32);
+		svm->vmcb->control.exit_info_2 = fault->address;
+	}
+
+	svm->vmcb->control.exit_info_1 &= ~0xffffffffULL;
+	svm->vmcb->control.exit_info_1 |= fault->error_code;
+
+	/*
+	 * The present bit is always zero for page structure faults on real
+	 * hardware.
+	 */
+	if (svm->vmcb->control.exit_info_1 & (2ULL << 32))
+		svm->vmcb->control.exit_info_1 &= ~1;
 
 	nested_svm_vmexit(svm);
 }
@@ -3031,7 +3048,7 @@ static int cr8_write_interception(struct vcpu_svm *svm)
 	return 0;
 }
 
-u64 svm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc)
+static u64 svm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc)
 {
 	struct vmcb *vmcb = get_host_vmcb(to_svm(vcpu));
 	return vmcb->control.tsc_offset +

arch/x86/kvm/trace.h

@@ -415,15 +415,14 @@ TRACE_EVENT(kvm_apic_ipi,
 );
 
 TRACE_EVENT(kvm_apic_accept_irq,
-	    TP_PROTO(__u32 apicid, __u16 dm, __u8 tm, __u8 vec, bool coalesced),
-	    TP_ARGS(apicid, dm, tm, vec, coalesced),
+	    TP_PROTO(__u32 apicid, __u16 dm, __u8 tm, __u8 vec),
+	    TP_ARGS(apicid, dm, tm, vec),
 
 	TP_STRUCT__entry(
 		__field(	__u32,		apicid		)
 		__field(	__u16,		dm		)
 		__field(	__u8,		tm		)
 		__field(	__u8,		vec		)
-		__field(	bool,		coalesced	)
 	),
 
 	TP_fast_assign(
@@ -431,14 +430,12 @@ TRACE_EVENT(kvm_apic_accept_irq,
 		__entry->dm		= dm;
 		__entry->tm		= tm;
 		__entry->vec		= vec;
-		__entry->coalesced	= coalesced;
 	),
 
-	TP_printk("apicid %x vec %u (%s|%s)%s",
+	TP_printk("apicid %x vec %u (%s|%s)",
 		  __entry->apicid, __entry->vec,
 		  __print_symbolic((__entry->dm >> 8 & 0x7), kvm_deliver_mode),
-		  __entry->tm ? "level" : "edge",
-		  __entry->coalesced ? " (coalesced)" : "")
+		  __entry->tm ? "level" : "edge")
 );
 
 TRACE_EVENT(kvm_eoi,
@@ -848,6 +845,8 @@ TRACE_EVENT(kvm_track_tsc,
 		  __print_symbolic(__entry->host_clock, host_clocks))
 );
 
+#endif /* CONFIG_X86_64 */
+
 TRACE_EVENT(kvm_ple_window,
 	TP_PROTO(bool grow, unsigned int vcpu_id, int new, int old),
 	TP_ARGS(grow, vcpu_id, new, old),
@@ -878,8 +877,6 @@ TRACE_EVENT(kvm_ple_window,
 #define trace_kvm_ple_window_shrink(vcpu_id, new, old) \
 	trace_kvm_ple_window(false, vcpu_id, new, old)
 
-#endif /* CONFIG_X86_64 */
-
 #endif /* _TRACE_KVM_H */
 
 #undef TRACE_INCLUDE_PATH

arch/x86/kvm/vmx.c

@@ -397,6 +397,7 @@ struct nested_vmx {
 	 * we must keep them pinned while L2 runs.
 	 */
 	struct page *apic_access_page;
+	struct page *virtual_apic_page;
 	u64 msr_ia32_feature_control;
 
 	struct hrtimer preemption_timer;
@@ -555,6 +556,7 @@ static int max_shadow_read_only_fields =
 	ARRAY_SIZE(shadow_read_only_fields);
 
 static unsigned long shadow_read_write_fields[] = {
+	TPR_THRESHOLD,
 	GUEST_RIP,
 	GUEST_RSP,
 	GUEST_CR0,
@@ -765,6 +767,7 @@ static u32 vmx_segment_access_rights(struct kvm_segment *var);
 static void vmx_sync_pir_to_irr_dummy(struct kvm_vcpu *vcpu);
 static void copy_vmcs12_to_shadow(struct vcpu_vmx *vmx);
 static void copy_shadow_to_vmcs12(struct vcpu_vmx *vmx);
+static int alloc_identity_pagetable(struct kvm *kvm);
 
 static DEFINE_PER_CPU(struct vmcs *, vmxarea);
 static DEFINE_PER_CPU(struct vmcs *, current_vmcs);
@@ -2157,7 +2160,7 @@ static u64 guest_read_tsc(void)
  * Like guest_read_tsc, but always returns L1's notion of the timestamp
  * counter, even if a nested guest (L2) is currently running.
  */
-u64 vmx_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc)
+static u64 vmx_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc)
 {
 	u64 tsc_offset;
 
@@ -2352,7 +2355,7 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 		CPU_BASED_MOV_DR_EXITING | CPU_BASED_UNCOND_IO_EXITING |
 		CPU_BASED_USE_IO_BITMAPS | CPU_BASED_MONITOR_EXITING |
 		CPU_BASED_RDPMC_EXITING | CPU_BASED_RDTSC_EXITING |
-		CPU_BASED_PAUSE_EXITING |
+		CPU_BASED_PAUSE_EXITING | CPU_BASED_TPR_SHADOW |
 		CPU_BASED_ACTIVATE_SECONDARY_CONTROLS;
 	/*
 	 * We can allow some features even when not supported by the
@@ -2726,7 +2729,7 @@ static void kvm_cpu_vmxon(u64 addr)
 			: "memory", "cc");
 }
 
-static int hardware_enable(void *garbage)
+static int hardware_enable(void)
 {
 	int cpu = raw_smp_processor_id();
 	u64 phys_addr = __pa(per_cpu(vmxarea, cpu));
@@ -2790,7 +2793,7 @@ static void kvm_cpu_vmxoff(void)
 	asm volatile (__ex(ASM_VMX_VMXOFF) : : : "cc");
 }
 
-static void hardware_disable(void *garbage)
+static void hardware_disable(void)
 {
 	if (vmm_exclusive) {
 		vmclear_local_loaded_vmcss();
@@ -3960,21 +3963,25 @@ out:
 
 static int init_rmode_identity_map(struct kvm *kvm)
 {
-	int i, idx, r, ret;
+	int i, idx, r = 0;
 	pfn_t identity_map_pfn;
 	u32 tmp;
 
 	if (!enable_ept)
-		return 1;
-	if (unlikely(!kvm->arch.ept_identity_pagetable)) {
-		printk(KERN_ERR "EPT: identity-mapping pagetable "
-			"haven't been allocated!\n");
 		return 0;
-	}
+
+	/* Protect kvm->arch.ept_identity_pagetable_done. */
+	mutex_lock(&kvm->slots_lock);
+
 	if (likely(kvm->arch.ept_identity_pagetable_done))
-		return 1;
-	ret = 0;
+		goto out2;
+
 	identity_map_pfn = kvm->arch.ept_identity_map_addr >> PAGE_SHIFT;
+
+	r = alloc_identity_pagetable(kvm);
+	if (r < 0)
+		goto out2;
+
 	idx = srcu_read_lock(&kvm->srcu);
 	r = kvm_clear_guest_page(kvm, identity_map_pfn, 0, PAGE_SIZE);
 	if (r < 0)
@@ -3989,10 +3996,13 @@ static int init_rmode_identity_map(struct kvm *kvm)
 			goto out;
 	}
 	kvm->arch.ept_identity_pagetable_done = true;
-	ret = 1;
+
 out:
 	srcu_read_unlock(&kvm->srcu, idx);
-	return ret;
+
+out2:
+	mutex_unlock(&kvm->slots_lock);
+	return r;
 }
 
 static void seg_setup(int seg)
@@ -4021,13 +4031,13 @@ static int alloc_apic_access_page(struct kvm *kvm)
 		goto out;
 	kvm_userspace_mem.slot = APIC_ACCESS_PAGE_PRIVATE_MEMSLOT;
 	kvm_userspace_mem.flags = 0;
-	kvm_userspace_mem.guest_phys_addr = 0xfee00000ULL;
+	kvm_userspace_mem.guest_phys_addr = APIC_DEFAULT_PHYS_BASE;
 	kvm_userspace_mem.memory_size = PAGE_SIZE;
 	r = __kvm_set_memory_region(kvm, &kvm_userspace_mem);
 	if (r)
 		goto out;
 
-	page = gfn_to_page(kvm, 0xfee00);
+	page = gfn_to_page(kvm, APIC_DEFAULT_PHYS_BASE >> PAGE_SHIFT);
 	if (is_error_page(page)) {
 		r = -EFAULT;
 		goto out;
@@ -4041,31 +4051,20 @@ out:
 
 static int alloc_identity_pagetable(struct kvm *kvm)
 {
-	struct page *page;
+	/* Called with kvm->slots_lock held. */
+
 	struct kvm_userspace_memory_region kvm_userspace_mem;
 	int r = 0;
 
-	mutex_lock(&kvm->slots_lock);
-	if (kvm->arch.ept_identity_pagetable)
-		goto out;
+	BUG_ON(kvm->arch.ept_identity_pagetable_done);
+
 	kvm_userspace_mem.slot = IDENTITY_PAGETABLE_PRIVATE_MEMSLOT;
 	kvm_userspace_mem.flags = 0;
 	kvm_userspace_mem.guest_phys_addr =
 		kvm->arch.ept_identity_map_addr;
 	kvm_userspace_mem.memory_size = PAGE_SIZE;
 	r = __kvm_set_memory_region(kvm, &kvm_userspace_mem);
-	if (r)
-		goto out;
 
-	page = gfn_to_page(kvm, kvm->arch.ept_identity_map_addr >> PAGE_SHIFT);
-	if (is_error_page(page)) {
-		r = -EFAULT;
-		goto out;
-	}
-
-	kvm->arch.ept_identity_pagetable = page;
-out:
-	mutex_unlock(&kvm->slots_lock);
 	return r;
 }
 
@@ -4500,7 +4499,7 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu)
 
 	vmx->vcpu.arch.regs[VCPU_REGS_RDX] = get_rdx_init_val();
 	kvm_set_cr8(&vmx->vcpu, 0);
-	apic_base_msr.data = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
+	apic_base_msr.data = APIC_DEFAULT_PHYS_BASE | MSR_IA32_APICBASE_ENABLE;
 	if (kvm_vcpu_is_bsp(&vmx->vcpu))
 		apic_base_msr.data |= MSR_IA32_APICBASE_BSP;
 	apic_base_msr.host_initiated = true;
@@ -6244,7 +6243,11 @@ static void free_nested(struct vcpu_vmx *vmx)
 	/* Unpin physical memory we referred to in current vmcs02 */
 	if (vmx->nested.apic_access_page) {
 		nested_release_page(vmx->nested.apic_access_page);
-		vmx->nested.apic_access_page = 0;
+		vmx->nested.apic_access_page = NULL;
+	}
+	if (vmx->nested.virtual_apic_page) {
+		nested_release_page(vmx->nested.virtual_apic_page);
+		vmx->nested.virtual_apic_page = NULL;
 	}
 
 	nested_free_all_saved_vmcss(vmx);
@@ -7034,7 +7037,7 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu)
 	case EXIT_REASON_MCE_DURING_VMENTRY:
 		return 0;
 	case EXIT_REASON_TPR_BELOW_THRESHOLD:
-		return 1;
+		return nested_cpu_has(vmcs12, CPU_BASED_TPR_SHADOW);
 	case EXIT_REASON_APIC_ACCESS:
 		return nested_cpu_has2(vmcs12,
 			SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES);
@@ -7155,6 +7158,12 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu)
 
 static void update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr)
 {
+	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
+
+	if (is_guest_mode(vcpu) &&
+		nested_cpu_has(vmcs12, CPU_BASED_TPR_SHADOW))
+		return;
+
 	if (irr == -1 || tpr < irr) {
 		vmcs_write32(TPR_THRESHOLD, 0);
 		return;
@@ -7745,10 +7754,8 @@ static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id)
 		if (!kvm->arch.ept_identity_map_addr)
 			kvm->arch.ept_identity_map_addr =
 				VMX_EPT_IDENTITY_PAGETABLE_ADDR;
-		err = -ENOMEM;
-		if (alloc_identity_pagetable(kvm) != 0)
-			goto free_vmcs;
-		if (!init_rmode_identity_map(kvm))
+		err = init_rmode_identity_map(kvm);
+		if (err)
 			goto free_vmcs;
 	}
 
@@ -7927,6 +7934,55 @@ static void vmx_inject_page_fault_nested(struct kvm_vcpu *vcpu,
 		kvm_inject_page_fault(vcpu, fault);
 }
 
+static bool nested_get_vmcs12_pages(struct kvm_vcpu *vcpu,
+					struct vmcs12 *vmcs12)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (nested_cpu_has2(vmcs12, SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES)) {
+		/* TODO: Also verify bits beyond physical address width are 0 */
+		if (!PAGE_ALIGNED(vmcs12->apic_access_addr))
+			return false;
+
+		/*
+		 * Translate L1 physical address to host physical
+		 * address for vmcs02. Keep the page pinned, so this
+		 * physical address remains valid. We keep a reference
+		 * to it so we can release it later.
+		 */
+		if (vmx->nested.apic_access_page) /* shouldn't happen */
+			nested_release_page(vmx->nested.apic_access_page);
+		vmx->nested.apic_access_page =
+			nested_get_page(vcpu, vmcs12->apic_access_addr);
+	}
+
+	if (nested_cpu_has(vmcs12, CPU_BASED_TPR_SHADOW)) {
+		/* TODO: Also verify bits beyond physical address width are 0 */
+		if (!PAGE_ALIGNED(vmcs12->virtual_apic_page_addr))
+			return false;
+
+		if (vmx->nested.virtual_apic_page) /* shouldn't happen */
+			nested_release_page(vmx->nested.virtual_apic_page);
+		vmx->nested.virtual_apic_page =
+			nested_get_page(vcpu, vmcs12->virtual_apic_page_addr);
+
+		/*
+		 * Failing the vm entry is _not_ what the processor does
+		 * but it's basically the only possibility we have.
+		 * We could still enter the guest if CR8 load exits are
+		 * enabled, CR8 store exits are enabled, and virtualize APIC
+		 * access is disabled; in this case the processor would never
+		 * use the TPR shadow and we could simply clear the bit from
+		 * the execution control.  But such a configuration is useless,
+		 * so let's keep the code simple.
+		 */
+		if (!vmx->nested.virtual_apic_page)
+			return false;
+	}
+
+	return true;
+}
+
 static void vmx_start_preemption_timer(struct kvm_vcpu *vcpu)
 {
 	u64 preemption_timeout = get_vmcs12(vcpu)->vmx_preemption_timer_value;
@@ -8072,16 +8128,6 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 			exec_control |= vmcs12->secondary_vm_exec_control;
 
 		if (exec_control & SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES) {
-			/*
-			 * Translate L1 physical address to host physical
-			 * address for vmcs02. Keep the page pinned, so this
-			 * physical address remains valid. We keep a reference
-			 * to it so we can release it later.
-			 */
-			if (vmx->nested.apic_access_page) /* shouldn't happen */
-				nested_release_page(vmx->nested.apic_access_page);
-			vmx->nested.apic_access_page =
-				nested_get_page(vcpu, vmcs12->apic_access_addr);
 			/*
 			 * If translation failed, no matter: This feature asks
 			 * to exit when accessing the given address, and if it
@@ -8127,6 +8173,13 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	exec_control &= ~CPU_BASED_VIRTUAL_NMI_PENDING;
 	exec_control &= ~CPU_BASED_TPR_SHADOW;
 	exec_control |= vmcs12->cpu_based_vm_exec_control;
+
+	if (exec_control & CPU_BASED_TPR_SHADOW) {
+		vmcs_write64(VIRTUAL_APIC_PAGE_ADDR,
+				page_to_phys(vmx->nested.virtual_apic_page));
+		vmcs_write32(TPR_THRESHOLD, vmcs12->tpr_threshold);
+	}
+
 	/*
 	 * Merging of IO and MSR bitmaps not currently supported.
 	 * Rather, exit every time.
@@ -8288,8 +8341,7 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch)
 		return 1;
 	}
 
-	if (nested_cpu_has2(vmcs12, SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES) &&
-			!PAGE_ALIGNED(vmcs12->apic_access_addr)) {
+	if (!nested_get_vmcs12_pages(vcpu, vmcs12)) {
 		/*TODO: Also verify bits beyond physical address width are 0*/
 		nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD);
 		return 1;
@@ -8893,7 +8945,11 @@ static void nested_vmx_vmexit(struct kvm_vcpu *vcpu, u32 exit_reason,
 	/* Unpin physical memory we referred to in vmcs02 */
 	if (vmx->nested.apic_access_page) {
 		nested_release_page(vmx->nested.apic_access_page);
-		vmx->nested.apic_access_page = 0;
+		vmx->nested.apic_access_page = NULL;
+	}
+	if (vmx->nested.virtual_apic_page) {
+		nested_release_page(vmx->nested.virtual_apic_page);
+		vmx->nested.virtual_apic_page = NULL;
 	}
 
 	/*
@@ -8949,7 +9005,7 @@ static int vmx_check_intercept(struct kvm_vcpu *vcpu,
 	return X86EMUL_CONTINUE;
 }
 
-void vmx_sched_in(struct kvm_vcpu *vcpu, int cpu)
+static void vmx_sched_in(struct kvm_vcpu *vcpu, int cpu)
 {
 	if (ple_gap)
 		shrink_ple_window(vcpu);

arch/x86/kvm/x86.c

@@ -246,7 +246,7 @@ void kvm_set_shared_msr(unsigned slot, u64 value, u64 mask)
 }
 EXPORT_SYMBOL_GPL(kvm_set_shared_msr);
 
-static void drop_user_return_notifiers(void *ignore)
+static void drop_user_return_notifiers(void)
 {
 	unsigned int cpu = smp_processor_id();
 	struct kvm_shared_msrs *smsr = per_cpu_ptr(shared_msrs, cpu);
@@ -408,12 +408,14 @@ void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault)
 }
 EXPORT_SYMBOL_GPL(kvm_inject_page_fault);
 
-void kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault)
+static bool kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault)
 {
 	if (mmu_is_nested(vcpu) && !fault->nested_page_fault)
 		vcpu->arch.nested_mmu.inject_page_fault(vcpu, fault);
 	else
 		vcpu->arch.mmu.inject_page_fault(vcpu, fault);
+
+	return fault->nested_page_fault;
 }
 
 void kvm_inject_nmi(struct kvm_vcpu *vcpu)
@@ -457,11 +459,12 @@ int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
 			    gfn_t ngfn, void *data, int offset, int len,
 			    u32 access)
 {
+	struct x86_exception exception;
 	gfn_t real_gfn;
 	gpa_t ngpa;
 
 	ngpa     = gfn_to_gpa(ngfn);
-	real_gfn = mmu->translate_gpa(vcpu, ngpa, access);
+	real_gfn = mmu->translate_gpa(vcpu, ngpa, access, &exception);
 	if (real_gfn == UNMAPPED_GVA)
 		return -EFAULT;
 
@@ -1518,7 +1521,7 @@ static void kvm_gen_update_masterclock(struct kvm *kvm)
 	pvclock_update_vm_gtod_copy(kvm);
 
 	kvm_for_each_vcpu(i, vcpu, kvm)
-		set_bit(KVM_REQ_CLOCK_UPDATE, &vcpu->requests);
+		kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 
 	/* guest entries allowed */
 	kvm_for_each_vcpu(i, vcpu, kvm)
@@ -1661,7 +1664,7 @@ static void kvmclock_update_fn(struct work_struct *work)
 	struct kvm_vcpu *vcpu;
 
 	kvm_for_each_vcpu(i, vcpu, kvm) {
-		set_bit(KVM_REQ_CLOCK_UPDATE, &vcpu->requests);
+		kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 		kvm_vcpu_kick(vcpu);
 	}
 }
@@ -1670,7 +1673,7 @@ static void kvm_gen_kvmclock_update(struct kvm_vcpu *v)
 {
 	struct kvm *kvm = v->kvm;
 
-	set_bit(KVM_REQ_CLOCK_UPDATE, &v->requests);
+	kvm_make_request(KVM_REQ_CLOCK_UPDATE, v);
 	schedule_delayed_work(&kvm->arch.kvmclock_update_work,
 					KVMCLOCK_UPDATE_DELAY);
 }
@@ -1726,7 +1729,7 @@ static bool valid_mtrr_type(unsigned t)
 static bool mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 {
 	int i;
-	u64 mask = 0;
+	u64 mask;
 
 	if (!msr_mtrr_valid(msr))
 		return false;
@@ -1750,8 +1753,7 @@ static bool mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 	/* variable MTRRs */
 	WARN_ON(!(msr >= 0x200 && msr < 0x200 + 2 * KVM_NR_VAR_MTRR));
 
-	for (i = 63; i > boot_cpu_data.x86_phys_bits; i--)
-		mask |= (1ULL << i);
+	mask = (~0ULL) << cpuid_maxphyaddr(vcpu);
 	if ((msr & 1) == 0) {
 		/* MTRR base */
 		if (!valid_mtrr_type(data & 0xff))
@@ -2847,7 +2849,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	if (unlikely(vcpu->arch.tsc_offset_adjustment)) {
 		adjust_tsc_offset_host(vcpu, vcpu->arch.tsc_offset_adjustment);
 		vcpu->arch.tsc_offset_adjustment = 0;
-		set_bit(KVM_REQ_CLOCK_UPDATE, &vcpu->requests);
+		kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 	}
 
 	if (unlikely(vcpu->cpu != cpu) || check_tsc_unstable()) {
@@ -4064,16 +4066,16 @@ void kvm_get_segment(struct kvm_vcpu *vcpu,
 	kvm_x86_ops->get_segment(vcpu, var, seg);
 }
 
-gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access)
+gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
+			   struct x86_exception *exception)
 {
 	gpa_t t_gpa;
-	struct x86_exception exception;
 
 	BUG_ON(!mmu_is_nested(vcpu));
 
 	/* NPT walks are always user-walks */
 	access |= PFERR_USER_MASK;
-	t_gpa  = vcpu->arch.mmu.gva_to_gpa(vcpu, gpa, access, &exception);
+	t_gpa  = vcpu->arch.mmu.gva_to_gpa(vcpu, gpa, access, exception);
 
 	return t_gpa;
 }
@@ -4930,16 +4932,18 @@ static void toggle_interruptibility(struct kvm_vcpu *vcpu, u32 mask)
 	}
 }
 
-static void inject_emulated_exception(struct kvm_vcpu *vcpu)
+static bool inject_emulated_exception(struct kvm_vcpu *vcpu)
 {
 	struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt;
 	if (ctxt->exception.vector == PF_VECTOR)
-		kvm_propagate_fault(vcpu, &ctxt->exception);
-	else if (ctxt->exception.error_code_valid)
+		return kvm_propagate_fault(vcpu, &ctxt->exception);
+
+	if (ctxt->exception.error_code_valid)
 		kvm_queue_exception_e(vcpu, ctxt->exception.vector,
 				      ctxt->exception.error_code);
 	else
 		kvm_queue_exception(vcpu, ctxt->exception.vector);
+	return false;
 }
 
 static void init_emulate_ctxt(struct kvm_vcpu *vcpu)
@@ -5301,8 +5305,9 @@ restart:
 	}
 
 	if (ctxt->have_exception) {
-		inject_emulated_exception(vcpu);
 		r = EMULATE_DONE;
+		if (inject_emulated_exception(vcpu))
+			return r;
 	} else if (vcpu->arch.pio.count) {
 		if (!vcpu->arch.pio.in) {
 			/* FIXME: return into emulator if single-stepping.  */
@@ -5570,7 +5575,7 @@ static void kvm_set_mmio_spte_mask(void)
 	 * entry to generate page fault with PFER.RSV = 1.
 	 */
 	 /* Mask the reserved physical address bits. */
-	mask = ((1ull << (51 - maxphyaddr + 1)) - 1) << maxphyaddr;
+	mask = rsvd_bits(maxphyaddr, 51);
 
 	/* Bit 62 is always reserved for 32bit host. */
 	mask |= 0x3ull << 62;
@@ -5601,7 +5606,7 @@ static void pvclock_gtod_update_fn(struct work_struct *work)
 	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		kvm_for_each_vcpu(i, vcpu, kvm)
-			set_bit(KVM_REQ_MASTERCLOCK_UPDATE, &vcpu->requests);
+			kvm_make_request(KVM_REQ_MASTERCLOCK_UPDATE, vcpu);
 	atomic_set(&kvm_guest_has_master_clock, 0);
 	spin_unlock(&kvm_lock);
 }
@@ -6959,7 +6964,7 @@ void kvm_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, unsigned int vector)
 	kvm_rip_write(vcpu, 0);
 }
 
-int kvm_arch_hardware_enable(void *garbage)
+int kvm_arch_hardware_enable(void)
 {
 	struct kvm *kvm;
 	struct kvm_vcpu *vcpu;
@@ -6970,7 +6975,7 @@ int kvm_arch_hardware_enable(void *garbage)
 	bool stable, backwards_tsc = false;
 
 	kvm_shared_msr_cpu_online();
-	ret = kvm_x86_ops->hardware_enable(garbage);
+	ret = kvm_x86_ops->hardware_enable();
 	if (ret != 0)
 		return ret;
 
@@ -6979,7 +6984,7 @@ int kvm_arch_hardware_enable(void *garbage)
 	list_for_each_entry(kvm, &vm_list, vm_list) {
 		kvm_for_each_vcpu(i, vcpu, kvm) {
 			if (!stable && vcpu->cpu == smp_processor_id())
-				set_bit(KVM_REQ_CLOCK_UPDATE, &vcpu->requests);
+				kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 			if (stable && vcpu->arch.last_host_tsc > local_tsc) {
 				backwards_tsc = true;
 				if (vcpu->arch.last_host_tsc > max_tsc)
@@ -7033,8 +7038,7 @@ int kvm_arch_hardware_enable(void *garbage)
 			kvm_for_each_vcpu(i, vcpu, kvm) {
 				vcpu->arch.tsc_offset_adjustment += delta_cyc;
 				vcpu->arch.last_host_tsc = local_tsc;
-				set_bit(KVM_REQ_MASTERCLOCK_UPDATE,
-					&vcpu->requests);
+				kvm_make_request(KVM_REQ_MASTERCLOCK_UPDATE, vcpu);
 			}
 
 			/*
@@ -7051,10 +7055,10 @@ int kvm_arch_hardware_enable(void *garbage)
 	return 0;
 }
 
-void kvm_arch_hardware_disable(void *garbage)
+void kvm_arch_hardware_disable(void)
 {
-	kvm_x86_ops->hardware_disable(garbage);
-	drop_user_return_notifiers(garbage);
+	kvm_x86_ops->hardware_disable();
+	drop_user_return_notifiers();
 }
 
 int kvm_arch_hardware_setup(void)
@@ -7269,8 +7273,6 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kvm_free_vcpus(kvm);
 	if (kvm->arch.apic_access_page)
 		put_page(kvm->arch.apic_access_page);
-	if (kvm->arch.ept_identity_pagetable)
-		put_page(kvm->arch.ept_identity_pagetable);
 	kfree(rcu_dereference_check(kvm->arch.apic_map, 1));
 }
 

arch/x86/kvm/x86.h

@@ -88,15 +88,23 @@ static inline void vcpu_cache_mmio_info(struct kvm_vcpu *vcpu,
 	vcpu->arch.mmio_gva = gva & PAGE_MASK;
 	vcpu->arch.access = access;
 	vcpu->arch.mmio_gfn = gfn;
+	vcpu->arch.mmio_gen = kvm_memslots(vcpu->kvm)->generation;
+}
+
+static inline bool vcpu_match_mmio_gen(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.mmio_gen == kvm_memslots(vcpu->kvm)->generation;
 }
 
 /*
- * Clear the mmio cache info for the given gva,
- * specially, if gva is ~0ul, we clear all mmio cache info.
+ * Clear the mmio cache info for the given gva. If gva is MMIO_GVA_ANY, we
+ * clear all mmio cache info.
  */
+#define MMIO_GVA_ANY (~(gva_t)0)
+
 static inline void vcpu_clear_mmio_info(struct kvm_vcpu *vcpu, gva_t gva)
 {
-	if (gva != (~0ul) && vcpu->arch.mmio_gva != (gva & PAGE_MASK))
+	if (gva != MMIO_GVA_ANY && vcpu->arch.mmio_gva != (gva & PAGE_MASK))
 		return;
 
 	vcpu->arch.mmio_gva = 0;
@@ -104,7 +112,8 @@ static inline void vcpu_clear_mmio_info(struct kvm_vcpu *vcpu, gva_t gva)
 
 static inline bool vcpu_match_mmio_gva(struct kvm_vcpu *vcpu, unsigned long gva)
 {
-	if (vcpu->arch.mmio_gva && vcpu->arch.mmio_gva == (gva & PAGE_MASK))
+	if (vcpu_match_mmio_gen(vcpu) && vcpu->arch.mmio_gva &&
+	      vcpu->arch.mmio_gva == (gva & PAGE_MASK))
 		return true;
 
 	return false;
@@ -112,7 +121,8 @@ static inline bool vcpu_match_mmio_gva(struct kvm_vcpu *vcpu, unsigned long gva)
 
 static inline bool vcpu_match_mmio_gpa(struct kvm_vcpu *vcpu, gpa_t gpa)
 {
-	if (vcpu->arch.mmio_gfn && vcpu->arch.mmio_gfn == gpa >> PAGE_SHIFT)
+	if (vcpu_match_mmio_gen(vcpu) && vcpu->arch.mmio_gfn &&
+	      vcpu->arch.mmio_gfn == gpa >> PAGE_SHIFT)
 		return true;
 
 	return false;

include/linux/kvm_host.h

@@ -140,8 +140,6 @@ static inline bool is_error_page(struct page *page)
 #define KVM_USERSPACE_IRQ_SOURCE_ID		0
 #define KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID	1
 
-struct kvm;
-struct kvm_vcpu;
 extern struct kmem_cache *kvm_vcpu_cache;
 
 extern spinlock_t kvm_lock;
@@ -325,8 +323,6 @@ struct kvm_kernel_irq_routing_entry {
 	struct hlist_node link;
 };
 
-struct kvm_irq_routing_table;
-
 #ifndef KVM_PRIVATE_MEM_SLOTS
 #define KVM_PRIVATE_MEM_SLOTS 0
 #endif
@@ -636,8 +632,8 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu);
 int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
 
-int kvm_arch_hardware_enable(void *garbage);
-void kvm_arch_hardware_disable(void *garbage);
+int kvm_arch_hardware_enable(void);
+void kvm_arch_hardware_disable(void);
 int kvm_arch_hardware_setup(void);
 void kvm_arch_hardware_unsetup(void);
 void kvm_arch_check_processor_compat(void *rtn);
@@ -1038,8 +1034,6 @@ static inline bool kvm_check_request(int req, struct kvm_vcpu *vcpu)
 
 extern bool kvm_rebooting;
 
-struct kvm_device_ops;
-
 struct kvm_device {
 	struct kvm_device_ops *ops;
 	struct kvm *kvm;
@@ -1072,12 +1066,10 @@ struct kvm_device_ops {
 void kvm_device_get(struct kvm_device *dev);
 void kvm_device_put(struct kvm_device *dev);
 struct kvm_device *kvm_device_from_filp(struct file *filp);
+int kvm_register_device_ops(struct kvm_device_ops *ops, u32 type);
 
 extern struct kvm_device_ops kvm_mpic_ops;
 extern struct kvm_device_ops kvm_xics_ops;
-extern struct kvm_device_ops kvm_vfio_ops;
-extern struct kvm_device_ops kvm_arm_vgic_v2_ops;
-extern struct kvm_device_ops kvm_flic_ops;
 
 #ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
 

include/linux/kvm_types.h

@@ -17,6 +17,20 @@
 #ifndef __KVM_TYPES_H__
 #define __KVM_TYPES_H__
 
+struct kvm;
+struct kvm_async_pf;
+struct kvm_device_ops;
+struct kvm_interrupt;
+struct kvm_irq_routing_table;
+struct kvm_memory_slot;
+struct kvm_one_reg;
+struct kvm_run;
+struct kvm_userspace_memory_region;
+struct kvm_vcpu;
+struct kvm_vcpu_init;
+
+enum kvm_mr_change;
+
 #include <asm/types.h>
 
 /*

include/trace/events/kvm.h

@@ -95,6 +95,26 @@ TRACE_EVENT(kvm_ioapic_set_irq,
 		  __entry->coalesced ? " (coalesced)" : "")
 );
 
+TRACE_EVENT(kvm_ioapic_delayed_eoi_inj,
+	    TP_PROTO(__u64 e),
+	    TP_ARGS(e),
+
+	TP_STRUCT__entry(
+		__field(	__u64,		e		)
+	),
+
+	TP_fast_assign(
+		__entry->e		= e;
+	),
+
+	TP_printk("dst %x vec=%u (%s|%s|%s%s)",
+		  (u8)(__entry->e >> 56), (u8)__entry->e,
+		  __print_symbolic((__entry->e >> 8 & 0x7), kvm_deliver_mode),
+		  (__entry->e & (1<<11)) ? "logical" : "physical",
+		  (__entry->e & (1<<15)) ? "level" : "edge",
+		  (__entry->e & (1<<16)) ? "|masked" : "")
+);
+
 TRACE_EVENT(kvm_msi_set_irq,
 	    TP_PROTO(__u64 address, __u64 data),
 	    TP_ARGS(address, data),

include/uapi/linux/kvm.h

@@ -654,9 +654,7 @@ struct kvm_ppc_smmu_info {
 #endif
 /* Bug in KVM_SET_USER_MEMORY_REGION fixed: */
 #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
@@ -738,9 +736,7 @@ struct kvm_ppc_smmu_info {
 #define KVM_CAP_PPC_GET_SMMU_INFO 78
 #define KVM_CAP_S390_COW 79
 #define KVM_CAP_PPC_ALLOC_HTAB 80
-#ifdef __KVM_HAVE_READONLY_MEM
 #define KVM_CAP_READONLY_MEM 81
-#endif
 #define KVM_CAP_IRQFD_RESAMPLE 82
 #define KVM_CAP_PPC_BOOKE_WATCHDOG 83
 #define KVM_CAP_PPC_HTAB_FD 84
@@ -947,15 +943,25 @@ struct kvm_device_attr {
 	__u64	addr;		/* userspace address of attr data */
 };
 
-#define KVM_DEV_TYPE_FSL_MPIC_20	1
-#define KVM_DEV_TYPE_FSL_MPIC_42	2
-#define KVM_DEV_TYPE_XICS		3
-#define KVM_DEV_TYPE_VFIO		4
 #define  KVM_DEV_VFIO_GROUP			1
 #define   KVM_DEV_VFIO_GROUP_ADD			1
 #define   KVM_DEV_VFIO_GROUP_DEL			2
-#define KVM_DEV_TYPE_ARM_VGIC_V2	5
-#define KVM_DEV_TYPE_FLIC		6
+
+enum kvm_device_type {
+	KVM_DEV_TYPE_FSL_MPIC_20	= 1,
+#define KVM_DEV_TYPE_FSL_MPIC_20	KVM_DEV_TYPE_FSL_MPIC_20
+	KVM_DEV_TYPE_FSL_MPIC_42,
+#define KVM_DEV_TYPE_FSL_MPIC_42	KVM_DEV_TYPE_FSL_MPIC_42
+	KVM_DEV_TYPE_XICS,
+#define KVM_DEV_TYPE_XICS		KVM_DEV_TYPE_XICS
+	KVM_DEV_TYPE_VFIO,
+#define KVM_DEV_TYPE_VFIO		KVM_DEV_TYPE_VFIO
+	KVM_DEV_TYPE_ARM_VGIC_V2,
+#define KVM_DEV_TYPE_ARM_VGIC_V2	KVM_DEV_TYPE_ARM_VGIC_V2
+	KVM_DEV_TYPE_FLIC,
+#define KVM_DEV_TYPE_FLIC		KVM_DEV_TYPE_FLIC
+	KVM_DEV_TYPE_MAX,
+};
 
 /*
  * ioctls for VM fds
@@ -1093,7 +1099,7 @@ struct kvm_s390_ucas_mapping {
 #define KVM_S390_INITIAL_RESET    _IO(KVMIO,   0x97)
 #define KVM_GET_MP_STATE          _IOR(KVMIO,  0x98, struct kvm_mp_state)
 #define KVM_SET_MP_STATE          _IOW(KVMIO,  0x99, struct kvm_mp_state)
-/* Available with KVM_CAP_NMI */
+/* Available with KVM_CAP_USER_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)

virt/kvm/arm/vgic.c

@@ -1522,83 +1522,6 @@ int kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-static void vgic_init_maintenance_interrupt(void *info)
-{
-	enable_percpu_irq(vgic->maint_irq, 0);
-}
-
-static int vgic_cpu_notify(struct notifier_block *self,
-			   unsigned long action, void *cpu)
-{
-	switch (action) {
-	case CPU_STARTING:
-	case CPU_STARTING_FROZEN:
-		vgic_init_maintenance_interrupt(NULL);
-		break;
-	case CPU_DYING:
-	case CPU_DYING_FROZEN:
-		disable_percpu_irq(vgic->maint_irq);
-		break;
-	}
-
-	return NOTIFY_OK;
-}
-
-static struct notifier_block vgic_cpu_nb = {
-	.notifier_call = vgic_cpu_notify,
-};
-
-static const struct of_device_id vgic_ids[] = {
-	{ .compatible = "arm,cortex-a15-gic", .data = vgic_v2_probe, },
-	{ .compatible = "arm,gic-v3", .data = vgic_v3_probe, },
-	{},
-};
-
-int kvm_vgic_hyp_init(void)
-{
-	const struct of_device_id *matched_id;
-	const int (*vgic_probe)(struct device_node *,const struct vgic_ops **,
-				const struct vgic_params **);
-	struct device_node *vgic_node;
-	int ret;
-
-	vgic_node = of_find_matching_node_and_match(NULL,
-						    vgic_ids, &matched_id);
-	if (!vgic_node) {
-		kvm_err("error: no compatible GIC node found\n");
-		return -ENODEV;
-	}
-
-	vgic_probe = matched_id->data;
-	ret = vgic_probe(vgic_node, &vgic_ops, &vgic);
-	if (ret)
-		return ret;
-
-	ret = request_percpu_irq(vgic->maint_irq, vgic_maintenance_handler,
-				 "vgic", kvm_get_running_vcpus());
-	if (ret) {
-		kvm_err("Cannot register interrupt %d\n", vgic->maint_irq);
-		return ret;
-	}
-
-	ret = __register_cpu_notifier(&vgic_cpu_nb);
-	if (ret) {
-		kvm_err("Cannot register vgic CPU notifier\n");
-		goto out_free_irq;
-	}
-
-	/* Callback into for arch code for setup */
-	vgic_arch_setup(vgic);
-
-	on_each_cpu(vgic_init_maintenance_interrupt, NULL, 1);
-
-	return 0;
-
-out_free_irq:
-	free_percpu_irq(vgic->maint_irq, kvm_get_running_vcpus());
-	return ret;
-}
-
 /**
  * kvm_vgic_init - Initialize global VGIC state before running any VCPUs
  * @kvm: pointer to the kvm struct
@@ -2062,7 +1985,7 @@ static int vgic_create(struct kvm_device *dev, u32 type)
 	return kvm_vgic_create(dev->kvm);
 }
 
-struct kvm_device_ops kvm_arm_vgic_v2_ops = {
+static struct kvm_device_ops kvm_arm_vgic_v2_ops = {
 	.name = "kvm-arm-vgic",
 	.create = vgic_create,
 	.destroy = vgic_destroy,
@@ -2070,3 +1993,81 @@ struct kvm_device_ops kvm_arm_vgic_v2_ops = {
 	.get_attr = vgic_get_attr,
 	.has_attr = vgic_has_attr,
 };
+
+static void vgic_init_maintenance_interrupt(void *info)
+{
+	enable_percpu_irq(vgic->maint_irq, 0);
+}
+
+static int vgic_cpu_notify(struct notifier_block *self,
+			   unsigned long action, void *cpu)
+{
+	switch (action) {
+	case CPU_STARTING:
+	case CPU_STARTING_FROZEN:
+		vgic_init_maintenance_interrupt(NULL);
+		break;
+	case CPU_DYING:
+	case CPU_DYING_FROZEN:
+		disable_percpu_irq(vgic->maint_irq);
+		break;
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block vgic_cpu_nb = {
+	.notifier_call = vgic_cpu_notify,
+};
+
+static const struct of_device_id vgic_ids[] = {
+	{ .compatible = "arm,cortex-a15-gic", .data = vgic_v2_probe, },
+	{ .compatible = "arm,gic-v3", .data = vgic_v3_probe, },
+	{},
+};
+
+int kvm_vgic_hyp_init(void)
+{
+	const struct of_device_id *matched_id;
+	const int (*vgic_probe)(struct device_node *,const struct vgic_ops **,
+				const struct vgic_params **);
+	struct device_node *vgic_node;
+	int ret;
+
+	vgic_node = of_find_matching_node_and_match(NULL,
+						    vgic_ids, &matched_id);
+	if (!vgic_node) {
+		kvm_err("error: no compatible GIC node found\n");
+		return -ENODEV;
+	}
+
+	vgic_probe = matched_id->data;
+	ret = vgic_probe(vgic_node, &vgic_ops, &vgic);
+	if (ret)
+		return ret;
+
+	ret = request_percpu_irq(vgic->maint_irq, vgic_maintenance_handler,
+				 "vgic", kvm_get_running_vcpus());
+	if (ret) {
+		kvm_err("Cannot register interrupt %d\n", vgic->maint_irq);
+		return ret;
+	}
+
+	ret = __register_cpu_notifier(&vgic_cpu_nb);
+	if (ret) {
+		kvm_err("Cannot register vgic CPU notifier\n");
+		goto out_free_irq;
+	}
+
+	/* Callback into for arch code for setup */
+	vgic_arch_setup(vgic);
+
+	on_each_cpu(vgic_init_maintenance_interrupt, NULL, 1);
+
+	return kvm_register_device_ops(&kvm_arm_vgic_v2_ops,
+				       KVM_DEV_TYPE_ARM_VGIC_V2);
+
+out_free_irq:
+	free_percpu_irq(vgic->maint_irq, kvm_get_running_vcpus());
+	return ret;
+}

virt/kvm/ioapic.c

@@ -405,6 +405,26 @@ void kvm_ioapic_clear_all(struct kvm_ioapic *ioapic, int irq_source_id)
 	spin_unlock(&ioapic->lock);
 }
 
+static void kvm_ioapic_eoi_inject_work(struct work_struct *work)
+{
+	int i;
+	struct kvm_ioapic *ioapic = container_of(work, struct kvm_ioapic,
+						 eoi_inject.work);
+	spin_lock(&ioapic->lock);
+	for (i = 0; i < IOAPIC_NUM_PINS; i++) {
+		union kvm_ioapic_redirect_entry *ent = &ioapic->redirtbl[i];
+
+		if (ent->fields.trig_mode != IOAPIC_LEVEL_TRIG)
+			continue;
+
+		if (ioapic->irr & (1 << i) && !ent->fields.remote_irr)
+			ioapic_service(ioapic, i, false);
+	}
+	spin_unlock(&ioapic->lock);
+}
+
+#define IOAPIC_SUCCESSIVE_IRQ_MAX_COUNT 10000
+
 static void __kvm_ioapic_update_eoi(struct kvm_vcpu *vcpu,
 			struct kvm_ioapic *ioapic, int vector, int trigger_mode)
 {
@@ -435,8 +455,26 @@ static void __kvm_ioapic_update_eoi(struct kvm_vcpu *vcpu,
 
 		ASSERT(ent->fields.trig_mode == IOAPIC_LEVEL_TRIG);
 		ent->fields.remote_irr = 0;
-		if (ioapic->irr & (1 << i))
-			ioapic_service(ioapic, i, false);
+		if (!ent->fields.mask && (ioapic->irr & (1 << i))) {
+			++ioapic->irq_eoi[i];
+			if (ioapic->irq_eoi[i] == IOAPIC_SUCCESSIVE_IRQ_MAX_COUNT) {
+				/*
+				 * Real hardware does not deliver the interrupt
+				 * immediately during eoi broadcast, and this
+				 * lets a buggy guest make slow progress
+				 * even if it does not correctly handle a
+				 * level-triggered interrupt.  Emulate this
+				 * behavior if we detect an interrupt storm.
+				 */
+				schedule_delayed_work(&ioapic->eoi_inject, HZ / 100);
+				ioapic->irq_eoi[i] = 0;
+				trace_kvm_ioapic_delayed_eoi_inj(ent->bits);
+			} else {
+				ioapic_service(ioapic, i, false);
+			}
+		} else {
+			ioapic->irq_eoi[i] = 0;
+		}
 	}
 }
 
@@ -565,12 +603,14 @@ static void kvm_ioapic_reset(struct kvm_ioapic *ioapic)
 {
 	int i;
 
+	cancel_delayed_work_sync(&ioapic->eoi_inject);
 	for (i = 0; i < IOAPIC_NUM_PINS; i++)
 		ioapic->redirtbl[i].fields.mask = 1;
 	ioapic->base_address = IOAPIC_DEFAULT_BASE_ADDRESS;
 	ioapic->ioregsel = 0;
 	ioapic->irr = 0;
 	ioapic->id = 0;
+	memset(ioapic->irq_eoi, 0x00, IOAPIC_NUM_PINS);
 	rtc_irq_eoi_tracking_reset(ioapic);
 	update_handled_vectors(ioapic);
 }
@@ -589,6 +629,7 @@ int kvm_ioapic_init(struct kvm *kvm)
 	if (!ioapic)
 		return -ENOMEM;
 	spin_lock_init(&ioapic->lock);
+	INIT_DELAYED_WORK(&ioapic->eoi_inject, kvm_ioapic_eoi_inject_work);
 	kvm->arch.vioapic = ioapic;
 	kvm_ioapic_reset(ioapic);
 	kvm_iodevice_init(&ioapic->dev, &ioapic_mmio_ops);
@@ -609,6 +650,7 @@ void kvm_ioapic_destroy(struct kvm *kvm)
 {
 	struct kvm_ioapic *ioapic = kvm->arch.vioapic;
 
+	cancel_delayed_work_sync(&ioapic->eoi_inject);
 	if (ioapic) {
 		kvm_io_bus_unregister_dev(kvm, KVM_MMIO_BUS, &ioapic->dev);
 		kvm->arch.vioapic = NULL;

virt/kvm/ioapic.h

@@ -59,6 +59,8 @@ struct kvm_ioapic {
 	spinlock_t lock;
 	DECLARE_BITMAP(handled_vectors, 256);
 	struct rtc_status rtc_status;
+	struct delayed_work eoi_inject;
+	u32 irq_eoi[IOAPIC_NUM_PINS];
 };
 
 #ifdef DEBUG

virt/kvm/kvm_main.c

@@ -95,8 +95,6 @@ static int hardware_enable_all(void);
 static void hardware_disable_all(void);
 
 static void kvm_io_bus_destroy(struct kvm_io_bus *bus);
-static void update_memslots(struct kvm_memslots *slots,
-			    struct kvm_memory_slot *new, u64 last_generation);
 
 static void kvm_release_pfn_dirty(pfn_t pfn);
 static void mark_page_dirty_in_slot(struct kvm *kvm,
@@ -477,6 +475,13 @@ static struct kvm *kvm_create_vm(unsigned long type)
 	kvm->memslots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL);
 	if (!kvm->memslots)
 		goto out_err_no_srcu;
+
+	/*
+	 * Init kvm generation close to the maximum to easily test the
+	 * code of handling generation number wrap-around.
+	 */
+	kvm->memslots->generation = -150;
+
 	kvm_init_memslots_id(kvm);
 	if (init_srcu_struct(&kvm->srcu))
 		goto out_err_no_srcu;
@@ -688,8 +693,7 @@ static void sort_memslots(struct kvm_memslots *slots)
 }
 
 static void update_memslots(struct kvm_memslots *slots,
-			    struct kvm_memory_slot *new,
-			    u64 last_generation)
+			    struct kvm_memory_slot *new)
 {
 	if (new) {
 		int id = new->id;
@@ -700,15 +704,13 @@ static void update_memslots(struct kvm_memslots *slots,
 		if (new->npages != npages)
 			sort_memslots(slots);
 	}
-
-	slots->generation = last_generation + 1;
 }
 
 static int check_memory_region_flags(struct kvm_userspace_memory_region *mem)
 {
 	u32 valid_flags = KVM_MEM_LOG_DIRTY_PAGES;
 
-#ifdef KVM_CAP_READONLY_MEM
+#ifdef __KVM_HAVE_READONLY_MEM
 	valid_flags |= KVM_MEM_READONLY;
 #endif
 
@@ -723,10 +725,24 @@ static struct kvm_memslots *install_new_memslots(struct kvm *kvm,
 {
 	struct kvm_memslots *old_memslots = kvm->memslots;
 
-	update_memslots(slots, new, kvm->memslots->generation);
+	/*
+	 * Set the low bit in the generation, which disables SPTE caching
+	 * until the end of synchronize_srcu_expedited.
+	 */
+	WARN_ON(old_memslots->generation & 1);
+	slots->generation = old_memslots->generation + 1;
+
+	update_memslots(slots, new);
 	rcu_assign_pointer(kvm->memslots, slots);
 	synchronize_srcu_expedited(&kvm->srcu);
 
+	/*
+	 * Increment the new memslot generation a second time. This prevents
+	 * vm exits that race with memslot updates from caching a memslot
+	 * generation that will (potentially) be valid forever.
+	 */
+	slots->generation++;
+
 	kvm_arch_memslots_updated(kvm);
 
 	return old_memslots;
@@ -777,7 +793,6 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	base_gfn = mem->guest_phys_addr >> PAGE_SHIFT;
 	npages = mem->memory_size >> PAGE_SHIFT;
 
-	r = -EINVAL;
 	if (npages > KVM_MEM_MAX_NR_PAGES)
 		goto out;
 
@@ -791,7 +806,6 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	new.npages = npages;
 	new.flags = mem->flags;
 
-	r = -EINVAL;
 	if (npages) {
 		if (!old.npages)
 			change = KVM_MR_CREATE;
@@ -847,7 +861,6 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	}
 
 	if ((change == KVM_MR_DELETE) || (change == KVM_MR_MOVE)) {
-		r = -ENOMEM;
 		slots = kmemdup(kvm->memslots, sizeof(struct kvm_memslots),
 				GFP_KERNEL);
 		if (!slots)
@@ -1776,8 +1789,7 @@ static bool kvm_vcpu_eligible_for_directed_yield(struct kvm_vcpu *vcpu)
 	bool eligible;
 
 	eligible = !vcpu->spin_loop.in_spin_loop ||
-			(vcpu->spin_loop.in_spin_loop &&
-			 vcpu->spin_loop.dy_eligible);
+		    vcpu->spin_loop.dy_eligible;
 
 	if (vcpu->spin_loop.in_spin_loop)
 		kvm_vcpu_set_dy_eligible(vcpu, !vcpu->spin_loop.dy_eligible);
@@ -2267,6 +2279,29 @@ struct kvm_device *kvm_device_from_filp(struct file *filp)
 	return filp->private_data;
 }
 
+static struct kvm_device_ops *kvm_device_ops_table[KVM_DEV_TYPE_MAX] = {
+#ifdef CONFIG_KVM_MPIC
+	[KVM_DEV_TYPE_FSL_MPIC_20]	= &kvm_mpic_ops,
+	[KVM_DEV_TYPE_FSL_MPIC_42]	= &kvm_mpic_ops,
+#endif
+
+#ifdef CONFIG_KVM_XICS
+	[KVM_DEV_TYPE_XICS]		= &kvm_xics_ops,
+#endif
+};
+
+int kvm_register_device_ops(struct kvm_device_ops *ops, u32 type)
+{
+	if (type >= ARRAY_SIZE(kvm_device_ops_table))
+		return -ENOSPC;
+
+	if (kvm_device_ops_table[type] != NULL)
+		return -EEXIST;
+
+	kvm_device_ops_table[type] = ops;
+	return 0;
+}
+
 static int kvm_ioctl_create_device(struct kvm *kvm,
 				   struct kvm_create_device *cd)
 {
@@ -2275,36 +2310,12 @@ static int kvm_ioctl_create_device(struct kvm *kvm,
 	bool test = cd->flags & KVM_CREATE_DEVICE_TEST;
 	int ret;
 
-	switch (cd->type) {
-#ifdef CONFIG_KVM_MPIC
-	case KVM_DEV_TYPE_FSL_MPIC_20:
-	case KVM_DEV_TYPE_FSL_MPIC_42:
-		ops = &kvm_mpic_ops;
-		break;
-#endif
-#ifdef CONFIG_KVM_XICS
-	case KVM_DEV_TYPE_XICS:
-		ops = &kvm_xics_ops;
-		break;
-#endif
-#ifdef CONFIG_KVM_VFIO
-	case KVM_DEV_TYPE_VFIO:
-		ops = &kvm_vfio_ops;
-		break;
-#endif
-#ifdef CONFIG_KVM_ARM_VGIC
-	case KVM_DEV_TYPE_ARM_VGIC_V2:
-		ops = &kvm_arm_vgic_v2_ops;
-		break;
-#endif
-#ifdef CONFIG_S390
-	case KVM_DEV_TYPE_FLIC:
-		ops = &kvm_flic_ops;
-		break;
-#endif
-	default:
+	if (cd->type >= ARRAY_SIZE(kvm_device_ops_table))
+		return -ENODEV;
+
+	ops = kvm_device_ops_table[cd->type];
+	if (ops == NULL)
 		return -ENODEV;
-	}
 
 	if (test)
 		return 0;
@@ -2619,7 +2630,6 @@ static long kvm_dev_ioctl(struct file *filp,
 
 	switch (ioctl) {
 	case KVM_GET_API_VERSION:
-		r = -EINVAL;
 		if (arg)
 			goto out;
 		r = KVM_API_VERSION;
@@ -2631,7 +2641,6 @@ static long kvm_dev_ioctl(struct file *filp,
 		r = kvm_vm_ioctl_check_extension_generic(NULL, arg);
 		break;
 	case KVM_GET_VCPU_MMAP_SIZE:
-		r = -EINVAL;
 		if (arg)
 			goto out;
 		r = PAGE_SIZE;     /* struct kvm_run */
@@ -2676,7 +2685,7 @@ static void hardware_enable_nolock(void *junk)
 
 	cpumask_set_cpu(cpu, cpus_hardware_enabled);
 
-	r = kvm_arch_hardware_enable(NULL);
+	r = kvm_arch_hardware_enable();
 
 	if (r) {
 		cpumask_clear_cpu(cpu, cpus_hardware_enabled);
@@ -2701,7 +2710,7 @@ static void hardware_disable_nolock(void *junk)
 	if (!cpumask_test_cpu(cpu, cpus_hardware_enabled))
 		return;
 	cpumask_clear_cpu(cpu, cpus_hardware_enabled);
-	kvm_arch_hardware_disable(NULL);
+	kvm_arch_hardware_disable();
 }
 
 static void hardware_disable(void)

virt/kvm/vfio.c

@@ -246,6 +246,16 @@ static void kvm_vfio_destroy(struct kvm_device *dev)
 	kfree(dev); /* alloc by kvm_ioctl_create_device, free by .destroy */
 }
 
+static int kvm_vfio_create(struct kvm_device *dev, u32 type);
+
+static struct kvm_device_ops kvm_vfio_ops = {
+	.name = "kvm-vfio",
+	.create = kvm_vfio_create,
+	.destroy = kvm_vfio_destroy,
+	.set_attr = kvm_vfio_set_attr,
+	.has_attr = kvm_vfio_has_attr,
+};
+
 static int kvm_vfio_create(struct kvm_device *dev, u32 type)
 {
 	struct kvm_device *tmp;
@@ -268,10 +278,8 @@ static int kvm_vfio_create(struct kvm_device *dev, u32 type)
 	return 0;
 }
 
-struct kvm_device_ops kvm_vfio_ops = {
-	.name = "kvm-vfio",
-	.create = kvm_vfio_create,
-	.destroy = kvm_vfio_destroy,
-	.set_attr = kvm_vfio_set_attr,
-	.has_attr = kvm_vfio_has_attr,
-};
+static int __init kvm_vfio_ops_init(void)
+{
+	return kvm_register_device_ops(&kvm_vfio_ops, KVM_DEV_TYPE_VFIO);
+}
+module_init(kvm_vfio_ops_init);