Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull KVM updates from Paolo Bonzini:
 "Fixes and features for 3.18.

  Apart from the usual cleanups, here is the summary of new features:

   - s390 moves closer towards host large page support

   - PowerPC has improved support for debugging (both inside the guest
     and via gdbstub) and support for e6500 processors

   - ARM/ARM64 support read-only memory (which is necessary to put
     firmware in emulated NOR flash)

   - x86 has the usual emulator fixes and nested virtualization
     improvements (including improved Windows support on Intel and
     Jailhouse hypervisor support on AMD), adaptive PLE which helps
     overcommitting of huge guests.  Also included are some patches that
     make KVM more friendly to memory hot-unplug, and fixes for rare
     caching bugs.

  Two patches have trivial mm/ parts that were acked by Rik and Andrew.

  Note: I will soon switch to a subkey for signing purposes"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (157 commits)
  kvm: do not handle APIC access page if in-kernel irqchip is not in use
  KVM: s390: count vcpu wakeups in stat.halt_wakeup
  KVM: s390/facilities: allow TOD-CLOCK steering facility bit
  KVM: PPC: BOOK3S: HV: CMA: Reserve cma region only in hypervisor mode
  arm/arm64: KVM: Report correct FSC for unsupported fault types
  arm/arm64: KVM: Fix VTTBR_BADDR_MASK and pgd alloc
  kvm: Fix kvm_get_page_retry_io __gup retval check
  arm/arm64: KVM: Fix set_clear_sgi_pend_reg offset
  kvm: x86: Unpin and remove kvm_arch->apic_access_page
  kvm: vmx: Implement set_apic_access_page_addr
  kvm: x86: Add request bit to reload APIC access page address
  kvm: Add arch specific mmu notifier for page invalidation
  kvm: Rename make_all_cpus_request() to kvm_make_all_cpus_request() and make it non-static
  kvm: Fix page ageing bugs
  kvm/x86/mmu: Pass gfn and level to rmapp callback.
  x86: kvm: use alternatives for VMCALL vs. VMMCALL if kernel text is read-only
  kvm: x86: use macros to compute bank MSRs
  KVM: x86: Remove debug assertion of non-PAE reserved bits
  kvm: don't take vcpu mutex for obviously invalid vcpu ioctls
  kvm: Faults which trigger IO release the mmap_sem
  ...

Documentation/virtual/kvm/api.txt

@@ -1901,6 +1901,8 @@ registers, find a list below:
   PPC   | KVM_REG_PPC_ARCH_COMPAT       | 32
   PPC   | KVM_REG_PPC_DABRX             | 32
   PPC   | KVM_REG_PPC_WORT              | 64
+  PPC	| KVM_REG_PPC_SPRG9             | 64
+  PPC	| KVM_REG_PPC_DBSR              | 32
   PPC   | KVM_REG_PPC_TM_GPR0           | 64
           ...
   PPC   | KVM_REG_PPC_TM_GPR31          | 64
@@ -2565,6 +2567,120 @@ 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 +2977,12 @@ 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/devices/arm-vgic.txt

@@ -71,3 +71,13 @@ Groups:
   Errors:
     -ENODEV: Getting or setting this register is not yet supported
     -EBUSY: One or more VCPUs are running
+
+  KVM_DEV_ARM_VGIC_GRP_NR_IRQS
+  Attributes:
+    A value describing the number of interrupts (SGI, PPI and SPI) for
+    this GIC instance, ranging from 64 to 1024, in increments of 32.
+
+  Errors:
+    -EINVAL: Value set is out of the expected range
+    -EBUSY: Value has already be set, or GIC has already been initialized
+            with default values.

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_emulate.h

@@ -148,6 +148,11 @@ static inline bool kvm_vcpu_trap_is_iabt(struct kvm_vcpu *vcpu)
 }
 
 static inline u8 kvm_vcpu_trap_get_fault(struct kvm_vcpu *vcpu)
+{
+	return kvm_vcpu_get_hsr(vcpu) & HSR_FSC;
+}
+
+static inline u8 kvm_vcpu_trap_get_fault_type(struct kvm_vcpu *vcpu)
 {
 	return kvm_vcpu_get_hsr(vcpu) & HSR_FSC_TYPE;
 }

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>
@@ -40,9 +42,8 @@
 
 #include <kvm/arm_vgic.h>
 
-struct kvm_vcpu;
 u32 *kvm_vcpu_reg(struct kvm_vcpu *vcpu, u8 reg_num, u32 mode);
-int kvm_target_cpu(void);
+int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
 void kvm_reset_coprocs(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);
@@ -172,7 +170,8 @@ unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
 int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
 
 /* We do not have shadow page tables, hence the empty hooks */
-static inline int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+static inline int kvm_age_hva(struct kvm *kvm, unsigned long start,
+			      unsigned long end)
 {
 	return 0;
 }
@@ -182,12 +181,16 @@ static inline int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
 	return 0;
 }
 
+static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
+							 unsigned long address)
+{
+}
+
 struct kvm_vcpu *kvm_arm_get_running_vcpu(void);
 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 +236,10 @@ 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/include/asm/kvm_mmu.h

@@ -78,17 +78,6 @@ static inline void kvm_set_pte(pte_t *pte, pte_t new_pte)
 	flush_pmd_entry(pte);
 }
 
-static inline bool kvm_is_write_fault(unsigned long hsr)
-{
-	unsigned long hsr_ec = hsr >> HSR_EC_SHIFT;
-	if (hsr_ec == HSR_EC_IABT)
-		return false;
-	else if ((hsr & HSR_ISV) && !(hsr & HSR_WNR))
-		return false;
-	else
-		return true;
-}
-
 static inline void kvm_clean_pgd(pgd_t *pgd)
 {
 	clean_dcache_area(pgd, PTRS_PER_S2_PGD * sizeof(pgd_t));

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

@@ -25,6 +25,7 @@
 
 #define __KVM_HAVE_GUEST_DEBUG
 #define __KVM_HAVE_IRQ_LINE
+#define __KVM_HAVE_READONLY_MEM
 
 #define KVM_REG_SIZE(id)						\
 	(1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
@@ -173,6 +174,7 @@ struct kvm_arch_memory_slot {
 #define   KVM_DEV_ARM_VGIC_CPUID_MASK	(0xffULL << KVM_DEV_ARM_VGIC_CPUID_SHIFT)
 #define   KVM_DEV_ARM_VGIC_OFFSET_SHIFT	0
 #define   KVM_DEV_ARM_VGIC_OFFSET_MASK	(0xffffffffULL << KVM_DEV_ARM_VGIC_OFFSET_SHIFT)
+#define KVM_DEV_ARM_VGIC_GRP_NR_IRQS	3
 
 /* KVM_IRQ_LINE irq field index values */
 #define KVM_ARM_IRQ_TYPE_SHIFT		24

arch/arm/kvm/arm.c

@@ -82,12 +82,12 @@ struct kvm_vcpu *kvm_arm_get_running_vcpu(void)
 /**
  * kvm_arm_get_running_vcpus - get the per-CPU array of currently running vcpus.
  */
-struct kvm_vcpu __percpu **kvm_get_running_vcpus(void)
+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
@@ -172,6 +161,8 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 			kvm->vcpus[i] = NULL;
 		}
 	}
+
+	kvm_vgic_destroy(kvm);
 }
 
 int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
@@ -188,6 +179,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_ONE_REG:
 	case KVM_CAP_ARM_PSCI:
 	case KVM_CAP_ARM_PSCI_0_2:
+	case KVM_CAP_READONLY_MEM:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -253,6 +245,7 @@ void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
 {
 	kvm_mmu_free_memory_caches(vcpu);
 	kvm_timer_vcpu_terminate(vcpu);
+	kvm_vgic_vcpu_destroy(vcpu);
 	kmem_cache_free(kvm_vcpu_cache, vcpu);
 }
 
@@ -268,26 +261,15 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 
 int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 {
-	int ret;
-
 	/* Force users to call KVM_ARM_VCPU_INIT */
 	vcpu->arch.target = -1;
 
-	/* Set up VGIC */
-	ret = kvm_vgic_vcpu_init(vcpu);
-	if (ret)
-		return ret;
-
 	/* Set up the timer */
 	kvm_timer_vcpu_init(vcpu);
 
 	return 0;
 }
 
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
-{
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	vcpu->cpu = cpu;
@@ -428,9 +410,9 @@ static void update_vttbr(struct kvm *kvm)
 
 	/* update vttbr to be used with the new vmid */
 	pgd_phys = virt_to_phys(kvm->arch.pgd);
+	BUG_ON(pgd_phys & ~VTTBR_BADDR_MASK);
 	vmid = ((u64)(kvm->arch.vmid) << VTTBR_VMID_SHIFT) & VTTBR_VMID_MASK;
-	kvm->arch.vttbr = pgd_phys & VTTBR_BADDR_MASK;
-	kvm->arch.vttbr |= vmid;
+	kvm->arch.vttbr = pgd_phys | vmid;
 
 	spin_unlock(&kvm_vmid_lock);
 }

arch/arm/kvm/coproc.c

@@ -791,7 +791,7 @@ static bool is_valid_cache(u32 val)
 	u32 level, ctype;
 
 	if (val >= CSSELR_MAX)
-		return -ENOENT;
+		return false;
 
 	/* Bottom bit is Instruction or Data bit.  Next 3 bits are level. */
         level = (val >> 1);

arch/arm/kvm/guest.c

@@ -163,7 +163,7 @@ static int set_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
 
 	ret = copy_from_user(&val, uaddr, KVM_REG_SIZE(reg->id));
 	if (ret != 0)
-		return ret;
+		return -EFAULT;
 
 	return kvm_arm_timer_set_reg(vcpu, reg->id, val);
 }

arch/arm/kvm/mmu.c

@@ -746,22 +746,29 @@ static bool transparent_hugepage_adjust(pfn_t *pfnp, phys_addr_t *ipap)
 	return false;
 }
 
+static bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
+{
+	if (kvm_vcpu_trap_is_iabt(vcpu))
+		return false;
+
+	return kvm_vcpu_dabt_iswrite(vcpu);
+}
+
 static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
-			  struct kvm_memory_slot *memslot,
+			  struct kvm_memory_slot *memslot, unsigned long hva,
 			  unsigned long fault_status)
 {
 	int ret;
 	bool write_fault, writable, hugetlb = false, force_pte = false;
 	unsigned long mmu_seq;
 	gfn_t gfn = fault_ipa >> PAGE_SHIFT;
-	unsigned long hva = gfn_to_hva(vcpu->kvm, gfn);
 	struct kvm *kvm = vcpu->kvm;
 	struct kvm_mmu_memory_cache *memcache = &vcpu->arch.mmu_page_cache;
 	struct vm_area_struct *vma;
 	pfn_t pfn;
 	pgprot_t mem_type = PAGE_S2;
 
-	write_fault = kvm_is_write_fault(kvm_vcpu_get_hsr(vcpu));
+	write_fault = kvm_is_write_fault(vcpu);
 	if (fault_status == FSC_PERM && !write_fault) {
 		kvm_err("Unexpected L2 read permission error\n");
 		return -EFAULT;
@@ -863,7 +870,8 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	unsigned long fault_status;
 	phys_addr_t fault_ipa;
 	struct kvm_memory_slot *memslot;
-	bool is_iabt;
+	unsigned long hva;
+	bool is_iabt, write_fault, writable;
 	gfn_t gfn;
 	int ret, idx;
 
@@ -874,17 +882,22 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 			      kvm_vcpu_get_hfar(vcpu), fault_ipa);
 
 	/* Check the stage-2 fault is trans. fault or write fault */
-	fault_status = kvm_vcpu_trap_get_fault(vcpu);
+	fault_status = kvm_vcpu_trap_get_fault_type(vcpu);
 	if (fault_status != FSC_FAULT && fault_status != FSC_PERM) {
-		kvm_err("Unsupported fault status: EC=%#x DFCS=%#lx\n",
-			kvm_vcpu_trap_get_class(vcpu), fault_status);
+		kvm_err("Unsupported FSC: EC=%#x xFSC=%#lx ESR_EL2=%#lx\n",
+			kvm_vcpu_trap_get_class(vcpu),
+			(unsigned long)kvm_vcpu_trap_get_fault(vcpu),
+			(unsigned long)kvm_vcpu_get_hsr(vcpu));
 		return -EFAULT;
 	}
 
 	idx = srcu_read_lock(&vcpu->kvm->srcu);
 
 	gfn = fault_ipa >> PAGE_SHIFT;
-	if (!kvm_is_visible_gfn(vcpu->kvm, gfn)) {
+	memslot = gfn_to_memslot(vcpu->kvm, gfn);
+	hva = gfn_to_hva_memslot_prot(memslot, gfn, &writable);
+	write_fault = kvm_is_write_fault(vcpu);
+	if (kvm_is_error_hva(hva) || (write_fault && !writable)) {
 		if (is_iabt) {
 			/* Prefetch Abort on I/O address */
 			kvm_inject_pabt(vcpu, kvm_vcpu_get_hfar(vcpu));
@@ -892,13 +905,6 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 			goto out_unlock;
 		}
 
-		if (fault_status != FSC_FAULT) {
-			kvm_err("Unsupported fault status on io memory: %#lx\n",
-				fault_status);
-			ret = -EFAULT;
-			goto out_unlock;
-		}
-
 		/*
 		 * The IPA is reported as [MAX:12], so we need to
 		 * complement it with the bottom 12 bits from the
@@ -910,9 +916,7 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		goto out_unlock;
 	}
 
-	memslot = gfn_to_memslot(vcpu->kvm, gfn);
-
-	ret = user_mem_abort(vcpu, fault_ipa, memslot, fault_status);
+	ret = user_mem_abort(vcpu, fault_ipa, memslot, hva, fault_status);
 	if (ret == 0)
 		ret = 1;
 out_unlock:

arch/arm64/include/asm/kvm_arm.h

@@ -122,6 +122,17 @@
 #define VTCR_EL2_T0SZ_MASK	0x3f
 #define VTCR_EL2_T0SZ_40B	24
 
+/*
+ * We configure the Stage-2 page tables to always restrict the IPA space to be
+ * 40 bits wide (T0SZ = 24).  Systems with a PARange smaller than 40 bits are
+ * not known to exist and will break with this configuration.
+ *
+ * Note that when using 4K pages, we concatenate two first level page tables
+ * together.
+ *
+ * The magic numbers used for VTTBR_X in this patch can be found in Tables
+ * D4-23 and D4-25 in ARM DDI 0487A.b.
+ */
 #ifdef CONFIG_ARM64_64K_PAGES
 /*
  * Stage2 translation configuration:
@@ -149,7 +160,7 @@
 #endif
 
 #define VTTBR_BADDR_SHIFT (VTTBR_X - 1)
-#define VTTBR_BADDR_MASK  (((1LLU << (40 - VTTBR_X)) - 1) << VTTBR_BADDR_SHIFT)
+#define VTTBR_BADDR_MASK  (((1LLU << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_BADDR_SHIFT)
 #define VTTBR_VMID_SHIFT  (48LLU)
 #define VTTBR_VMID_MASK	  (0xffLLU << VTTBR_VMID_SHIFT)
 

arch/arm64/include/asm/kvm_emulate.h

@@ -173,6 +173,11 @@ static inline bool kvm_vcpu_trap_is_iabt(const struct kvm_vcpu *vcpu)
 }
 
 static inline u8 kvm_vcpu_trap_get_fault(const struct kvm_vcpu *vcpu)
+{
+	return kvm_vcpu_get_hsr(vcpu) & ESR_EL2_FSC;
+}
+
+static inline u8 kvm_vcpu_trap_get_fault_type(const struct kvm_vcpu *vcpu)
 {
 	return kvm_vcpu_get_hsr(vcpu) & ESR_EL2_FSC_TYPE;
 }

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>
@@ -41,8 +43,7 @@
 
 #define KVM_VCPU_MAX_FEATURES 3
 
-struct kvm_vcpu;
-int kvm_target_cpu(void);
+int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
 int kvm_arch_dev_ioctl_check_extension(long ext);
 
@@ -164,25 +165,23 @@ 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);
 void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 
 /* We do not have shadow page tables, hence the empty hooks */
-static inline int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+static inline int kvm_age_hva(struct kvm *kvm, unsigned long start,
+			      unsigned long end)
 {
 	return 0;
 }
@@ -192,8 +191,13 @@ static inline int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
 	return 0;
 }
 
+static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
+							 unsigned long address)
+{
+}
+
 struct kvm_vcpu *kvm_arm_get_running_vcpu(void);
-struct kvm_vcpu __percpu **kvm_get_running_vcpus(void);
+struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void);
 
 u64 kvm_call_hyp(void *hypfn, ...);
 
@@ -244,4 +248,10 @@ 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/arm64/include/asm/kvm_mmu.h

@@ -59,10 +59,9 @@
 #define KERN_TO_HYP(kva)	((unsigned long)kva - PAGE_OFFSET + HYP_PAGE_OFFSET)
 
 /*
- * Align KVM with the kernel's view of physical memory. Should be
- * 40bit IPA, with PGD being 8kB aligned in the 4KB page configuration.
+ * We currently only support a 40bit IPA.
  */
-#define KVM_PHYS_SHIFT	PHYS_MASK_SHIFT
+#define KVM_PHYS_SHIFT	(40)
 #define KVM_PHYS_SIZE	(1UL << KVM_PHYS_SHIFT)
 #define KVM_PHYS_MASK	(KVM_PHYS_SIZE - 1UL)
 
@@ -93,19 +92,6 @@ void kvm_clear_hyp_idmap(void);
 #define	kvm_set_pte(ptep, pte)		set_pte(ptep, pte)
 #define	kvm_set_pmd(pmdp, pmd)		set_pmd(pmdp, pmd)
 
-static inline bool kvm_is_write_fault(unsigned long esr)
-{
-	unsigned long esr_ec = esr >> ESR_EL2_EC_SHIFT;
-
-	if (esr_ec == ESR_EL2_EC_IABT)
-		return false;
-
-	if ((esr & ESR_EL2_ISV) && !(esr & ESR_EL2_WNR))
-		return false;
-
-	return true;
-}
-
 static inline void kvm_clean_pgd(pgd_t *pgd) {}
 static inline void kvm_clean_pmd_entry(pmd_t *pmd) {}
 static inline void kvm_clean_pte(pte_t *pte) {}

arch/arm64/include/uapi/asm/kvm.h

@@ -37,6 +37,7 @@
 
 #define __KVM_HAVE_GUEST_DEBUG
 #define __KVM_HAVE_IRQ_LINE
+#define __KVM_HAVE_READONLY_MEM
 
 #define KVM_REG_SIZE(id)						\
 	(1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
@@ -159,6 +160,7 @@ struct kvm_arch_memory_slot {
 #define   KVM_DEV_ARM_VGIC_CPUID_MASK	(0xffULL << KVM_DEV_ARM_VGIC_CPUID_SHIFT)
 #define   KVM_DEV_ARM_VGIC_OFFSET_SHIFT	0
 #define   KVM_DEV_ARM_VGIC_OFFSET_MASK	(0xffffffffULL << KVM_DEV_ARM_VGIC_OFFSET_SHIFT)
+#define KVM_DEV_ARM_VGIC_GRP_NR_IRQS	3
 
 /* KVM_IRQ_LINE irq field index values */
 #define KVM_ARM_IRQ_TYPE_SHIFT		24

arch/arm64/kvm/guest.c

@@ -174,7 +174,7 @@ static int set_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
 
 	ret = copy_from_user(&val, uaddr, KVM_REG_SIZE(reg->id));
 	if (ret != 0)
-		return ret;
+		return -EFAULT;
 
 	return kvm_arm_timer_set_reg(vcpu, reg->id, val);
 }

arch/arm64/kvm/sys_regs.c

@@ -1218,7 +1218,7 @@ static bool is_valid_cache(u32 val)
 	u32 level, ctype;
 
 	if (val >= CSSELR_MAX)
-		return -ENOENT;
+		return false;
 
 	/* Bottom bit is Instruction or Data bit.  Next 3 bits are level. */
 	level = (val >> 1);

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;
@@ -998,10 +964,6 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
-{
-}
-
 int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
 				  struct kvm_translation *tr)
 {

arch/powerpc/include/asm/kvm_asm.h

@@ -53,17 +53,17 @@
 #define BOOKE_INTERRUPT_DEBUG 15
 
 /* E500 */
-#define BOOKE_INTERRUPT_SPE_ALTIVEC_UNAVAIL 32
-#define BOOKE_INTERRUPT_SPE_FP_DATA_ALTIVEC_ASSIST 33
-/*
- * TODO: Unify 32-bit and 64-bit kernel exception handlers to use same defines
- */
-#define BOOKE_INTERRUPT_SPE_UNAVAIL BOOKE_INTERRUPT_SPE_ALTIVEC_UNAVAIL
-#define BOOKE_INTERRUPT_SPE_FP_DATA BOOKE_INTERRUPT_SPE_FP_DATA_ALTIVEC_ASSIST
-#define BOOKE_INTERRUPT_ALTIVEC_UNAVAIL BOOKE_INTERRUPT_SPE_ALTIVEC_UNAVAIL
-#define BOOKE_INTERRUPT_ALTIVEC_ASSIST \
-				BOOKE_INTERRUPT_SPE_FP_DATA_ALTIVEC_ASSIST
+#ifdef CONFIG_SPE_POSSIBLE
+#define BOOKE_INTERRUPT_SPE_UNAVAIL 32
+#define BOOKE_INTERRUPT_SPE_FP_DATA 33
 #define BOOKE_INTERRUPT_SPE_FP_ROUND 34
+#endif
+
+#ifdef CONFIG_PPC_E500MC
+#define BOOKE_INTERRUPT_ALTIVEC_UNAVAIL 32
+#define BOOKE_INTERRUPT_ALTIVEC_ASSIST 33
+#endif
+
 #define BOOKE_INTERRUPT_PERFORMANCE_MONITOR 35
 #define BOOKE_INTERRUPT_DOORBELL 36
 #define BOOKE_INTERRUPT_DOORBELL_CRITICAL 37

arch/powerpc/include/asm/kvm_booke.h

@@ -23,15 +23,16 @@
 #include <linux/types.h>
 #include <linux/kvm_host.h>
 
-/* LPIDs we support with this build -- runtime limit may be lower */
+/*
+ * Number of available lpids. Only the low-order 6 bits of LPID rgister are
+ * implemented on e500mc+ cores.
+ */
 #define KVMPPC_NR_LPIDS                        64
 
 #define KVMPPC_INST_EHPRIV		0x7c00021c
 #define EHPRIV_OC_SHIFT			11
 /* "ehpriv 1" : ehpriv with OC = 1 is used for debug emulation */
 #define EHPRIV_OC_DEBUG			1
-#define KVMPPC_INST_EHPRIV_DEBUG	(KVMPPC_INST_EHPRIV | \
-					 (EHPRIV_OC_DEBUG << EHPRIV_OC_SHIFT))
 
 static inline void kvmppc_set_gpr(struct kvm_vcpu *vcpu, int num, ulong val)
 {

arch/powerpc/include/asm/kvm_host.h

@@ -53,14 +53,18 @@
 
 #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);
-extern int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+extern int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
 extern int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 extern void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 
+static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
+							 unsigned long address)
+{
+}
+
 #define HPTEG_CACHE_NUM			(1 << 15)
 #define HPTEG_HASH_BITS_PTE		13
 #define HPTEG_HASH_BITS_PTE_LONG	12
@@ -76,10 +80,6 @@ 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;
@@ -144,6 +144,7 @@ enum kvm_exit_types {
 	EMULATED_TLBWE_EXITS,
 	EMULATED_RFI_EXITS,
 	EMULATED_RFCI_EXITS,
+	EMULATED_RFDI_EXITS,
 	DEC_EXITS,
 	EXT_INTR_EXITS,
 	HALT_WAKEUP,
@@ -589,8 +590,6 @@ struct kvm_vcpu_arch {
 	u32 crit_save;
 	/* guest debug registers*/
 	struct debug_reg dbg_reg;
-	/* hardware visible debug registers when in guest state */
-	struct debug_reg shadow_dbg_reg;
 #endif
 	gpa_t paddr_accessed;
 	gva_t vaddr_accessed;
@@ -612,7 +611,6 @@ struct kvm_vcpu_arch {
 	u32 cpr0_cfgaddr; /* holds the last set cpr0_cfgaddr */
 
 	struct hrtimer dec_timer;
-	struct tasklet_struct tasklet;
 	u64 dec_jiffies;
 	u64 dec_expires;
 	unsigned long pending_exceptions;
@@ -687,4 +685,12 @@ 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/include/asm/kvm_ppc.h

@@ -38,6 +38,12 @@
 #include <asm/paca.h>
 #endif
 
+/*
+ * KVMPPC_INST_SW_BREAKPOINT is debug Instruction
+ * for supporting software breakpoint.
+ */
+#define KVMPPC_INST_SW_BREAKPOINT	0x00dddd00
+
 enum emulation_result {
 	EMULATE_DONE,         /* no further processing */
 	EMULATE_DO_MMIO,      /* kvm_run filled with MMIO request */
@@ -89,7 +95,7 @@ extern int kvmppc_emulate_loadstore(struct kvm_vcpu *vcpu);
 extern int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu);
 extern void kvmppc_emulate_dec(struct kvm_vcpu *vcpu);
 extern u32 kvmppc_get_dec(struct kvm_vcpu *vcpu, u64 tb);
-extern void kvmppc_decrementer_func(unsigned long data);
+extern void kvmppc_decrementer_func(struct kvm_vcpu *vcpu);
 extern int kvmppc_sanity_check(struct kvm_vcpu *vcpu);
 extern int kvmppc_subarch_vcpu_init(struct kvm_vcpu *vcpu);
 extern void kvmppc_subarch_vcpu_uninit(struct kvm_vcpu *vcpu);
@@ -206,6 +212,9 @@ extern int kvmppc_xics_get_xive(struct kvm *kvm, u32 irq, u32 *server,
 extern int kvmppc_xics_int_on(struct kvm *kvm, u32 irq);
 extern int kvmppc_xics_int_off(struct kvm *kvm, u32 irq);
 
+void kvmppc_core_dequeue_debug(struct kvm_vcpu *vcpu);
+void kvmppc_core_queue_debug(struct kvm_vcpu *vcpu);
+
 union kvmppc_one_reg {
 	u32	wval;
 	u64	dval;
@@ -243,7 +252,7 @@ struct kvmppc_ops {
 	int (*unmap_hva)(struct kvm *kvm, unsigned long hva);
 	int (*unmap_hva_range)(struct kvm *kvm, unsigned long start,
 			   unsigned long end);
-	int (*age_hva)(struct kvm *kvm, unsigned long hva);
+	int (*age_hva)(struct kvm *kvm, unsigned long start, unsigned long end);
 	int (*test_age_hva)(struct kvm *kvm, unsigned long hva);
 	void (*set_spte_hva)(struct kvm *kvm, unsigned long hva, pte_t pte);
 	void (*mmu_destroy)(struct kvm_vcpu *vcpu);

arch/powerpc/include/asm/reg_booke.h

@@ -319,6 +319,8 @@
  * DBSR bits which have conflicting definitions on true Book E versus IBM 40x.
  */
 #ifdef CONFIG_BOOKE
+#define DBSR_IDE	0x80000000	/* Imprecise Debug Event */
+#define DBSR_MRR	0x30000000	/* Most Recent Reset */
 #define DBSR_IC		0x08000000	/* Instruction Completion */
 #define DBSR_BT		0x04000000	/* Branch Taken */
 #define DBSR_IRPT	0x02000000	/* Exception Debug Event */

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

@@ -476,6 +476,11 @@ struct kvm_get_htab_header {
 
 /* FP and vector status/control registers */
 #define KVM_REG_PPC_FPSCR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x80)
+/*
+ * VSCR register is documented as a 32-bit register in the ISA, but it can
+ * only be accesses via a vector register. Expose VSCR as a 32-bit register
+ * even though the kernel represents it as a 128-bit vector.
+ */
 #define KVM_REG_PPC_VSCR	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x81)
 
 /* Virtual processor areas */
@@ -557,6 +562,7 @@ struct kvm_get_htab_header {
 #define KVM_REG_PPC_DABRX	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0xb8)
 #define KVM_REG_PPC_WORT	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xb9)
 #define KVM_REG_PPC_SPRG9	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xba)
+#define KVM_REG_PPC_DBSR	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0xbb)
 
 /* Transactional Memory checkpointed state:
  * This is all GPRs, all VSX regs and a subset of SPRs

arch/powerpc/kernel/cpu_setup_fsl_booke.S

@@ -91,6 +91,7 @@ _GLOBAL(setup_altivec_idle)
 
 	blr
 
+#ifdef CONFIG_PPC_E500MC
 _GLOBAL(__setup_cpu_e6500)
 	mflr	r6
 #ifdef CONFIG_PPC64
@@ -107,14 +108,20 @@ _GLOBAL(__setup_cpu_e6500)
 	bl	__setup_cpu_e5500
 	mtlr	r6
 	blr
+#endif /* CONFIG_PPC_E500MC */
 
 #ifdef CONFIG_PPC32
+#ifdef CONFIG_E200
 _GLOBAL(__setup_cpu_e200)
 	/* enable dedicated debug exception handling resources (Debug APU) */
 	mfspr	r3,SPRN_HID0
 	ori	r3,r3,HID0_DAPUEN@l
 	mtspr	SPRN_HID0,r3
 	b	__setup_e200_ivors
+#endif /* CONFIG_E200 */
+
+#ifdef CONFIG_E500
+#ifndef CONFIG_PPC_E500MC
 _GLOBAL(__setup_cpu_e500v1)
 _GLOBAL(__setup_cpu_e500v2)
 	mflr	r4
@@ -129,6 +136,7 @@ _GLOBAL(__setup_cpu_e500v2)
 #endif
 	mtlr	r4
 	blr
+#else /* CONFIG_PPC_E500MC */
 _GLOBAL(__setup_cpu_e500mc)
 _GLOBAL(__setup_cpu_e5500)
 	mflr	r5
@@ -159,7 +167,9 @@ _GLOBAL(__setup_cpu_e5500)
 2:
 	mtlr	r5
 	blr
-#endif
+#endif /* CONFIG_PPC_E500MC */
+#endif /* CONFIG_E500 */
+#endif /* CONFIG_PPC32 */
 
 #ifdef CONFIG_PPC_BOOK3E_64
 _GLOBAL(__restore_cpu_e6500)

arch/powerpc/kernel/cputable.c

@@ -1961,6 +1961,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 #endif /* CONFIG_PPC32 */
 #ifdef CONFIG_E500
 #ifdef CONFIG_PPC32
+#ifndef CONFIG_PPC_E500MC
 	{	/* e500 */
 		.pvr_mask		= 0xffff0000,
 		.pvr_value		= 0x80200000,
@@ -2000,6 +2001,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.machine_check		= machine_check_e500,
 		.platform		= "ppc8548",
 	},
+#else
 	{	/* e500mc */
 		.pvr_mask		= 0xffff0000,
 		.pvr_value		= 0x80230000,
@@ -2018,7 +2020,9 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.machine_check		= machine_check_e500mc,
 		.platform		= "ppce500mc",
 	},
+#endif /* CONFIG_PPC_E500MC */
 #endif /* CONFIG_PPC32 */
+#ifdef CONFIG_PPC_E500MC
 	{	/* e5500 */
 		.pvr_mask		= 0xffff0000,
 		.pvr_value		= 0x80240000,
@@ -2062,6 +2066,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.machine_check		= machine_check_e500mc,
 		.platform		= "ppce6500",
 	},
+#endif /* CONFIG_PPC_E500MC */
 #ifdef CONFIG_PPC32
 	{	/* default match */
 		.pvr_mask		= 0x00000000,

arch/powerpc/kernel/exceptions-64e.S

@@ -635,7 +635,7 @@ interrupt_end_book3e:
 
 /* Altivec Unavailable Interrupt */
 	START_EXCEPTION(altivec_unavailable);
-	NORMAL_EXCEPTION_PROLOG(0x200, BOOKE_INTERRUPT_SPE_ALTIVEC_UNAVAIL,
+	NORMAL_EXCEPTION_PROLOG(0x200, BOOKE_INTERRUPT_ALTIVEC_UNAVAIL,
 				PROLOG_ADDITION_NONE)
 	/* we can probably do a shorter exception entry for that one... */
 	EXCEPTION_COMMON(0x200)
@@ -658,7 +658,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
 /* AltiVec Assist */
 	START_EXCEPTION(altivec_assist);
 	NORMAL_EXCEPTION_PROLOG(0x220,
-				BOOKE_INTERRUPT_SPE_FP_DATA_ALTIVEC_ASSIST,
+				BOOKE_INTERRUPT_ALTIVEC_ASSIST,
 				PROLOG_ADDITION_NONE)
 	EXCEPTION_COMMON(0x220)
 	INTS_DISABLE

arch/powerpc/kernel/head_fsl_booke.S

@@ -613,34 +613,36 @@ END_FTR_SECTION_IFSET(CPU_FTR_EMB_HV)
 	mfspr	r10, SPRN_SPRG_RSCRATCH0
 	b	InstructionStorage
 
+/* Define SPE handlers for e200 and e500v2 */
 #ifdef CONFIG_SPE
 	/* SPE Unavailable */
 	START_EXCEPTION(SPEUnavailable)
-	NORMAL_EXCEPTION_PROLOG(SPE_ALTIVEC_UNAVAIL)
+	NORMAL_EXCEPTION_PROLOG(SPE_UNAVAIL)
 	beq	1f
 	bl	load_up_spe
 	b	fast_exception_return
 1:	addi	r3,r1,STACK_FRAME_OVERHEAD
 	EXC_XFER_EE_LITE(0x2010, KernelSPE)
-#else
-	EXCEPTION(0x2020, SPE_ALTIVEC_UNAVAIL, SPEUnavailable, \
+#elif defined(CONFIG_SPE_POSSIBLE)
+	EXCEPTION(0x2020, SPE_UNAVAIL, SPEUnavailable, \
 		  unknown_exception, EXC_XFER_EE)
-#endif /* CONFIG_SPE */
+#endif /* CONFIG_SPE_POSSIBLE */
 
 	/* SPE Floating Point Data */
 #ifdef CONFIG_SPE
-	EXCEPTION(0x2030, SPE_FP_DATA_ALTIVEC_ASSIST, SPEFloatingPointData,
+	EXCEPTION(0x2030, SPE_FP_DATA, SPEFloatingPointData,
 		  SPEFloatingPointException, EXC_XFER_EE)
 
 	/* SPE Floating Point Round */
 	EXCEPTION(0x2050, SPE_FP_ROUND, SPEFloatingPointRound, \
 		  SPEFloatingPointRoundException, EXC_XFER_EE)
-#else
-	EXCEPTION(0x2040, SPE_FP_DATA_ALTIVEC_ASSIST, SPEFloatingPointData,
+#elif defined(CONFIG_SPE_POSSIBLE)
+	EXCEPTION(0x2040, SPE_FP_DATA, SPEFloatingPointData,
 		  unknown_exception, EXC_XFER_EE)
 	EXCEPTION(0x2050, SPE_FP_ROUND, SPEFloatingPointRound, \
 		  unknown_exception, EXC_XFER_EE)
-#endif /* CONFIG_SPE */
+#endif /* CONFIG_SPE_POSSIBLE */
+
 
 	/* Performance Monitor */
 	EXCEPTION(0x2060, PERFORMANCE_MONITOR, PerformanceMonitor, \
@@ -947,6 +949,7 @@ get_phys_addr:
  * Global functions
  */
 
+#ifdef CONFIG_E200
 /* Adjust or setup IVORs for e200 */
 _GLOBAL(__setup_e200_ivors)
 	li	r3,DebugDebug@l
@@ -959,7 +962,10 @@ _GLOBAL(__setup_e200_ivors)
 	mtspr	SPRN_IVOR34,r3
 	sync
 	blr
+#endif
 
+#ifdef CONFIG_E500
+#ifndef CONFIG_PPC_E500MC
 /* Adjust or setup IVORs for e500v1/v2 */
 _GLOBAL(__setup_e500_ivors)
 	li	r3,DebugCrit@l
@@ -974,7 +980,7 @@ _GLOBAL(__setup_e500_ivors)
 	mtspr	SPRN_IVOR35,r3
 	sync
 	blr
-
+#else
 /* Adjust or setup IVORs for e500mc */
 _GLOBAL(__setup_e500mc_ivors)
 	li	r3,DebugDebug@l
@@ -1000,6 +1006,8 @@ _GLOBAL(__setup_ehv_ivors)
 	mtspr	SPRN_IVOR41,r3
 	sync
 	blr
+#endif /* CONFIG_PPC_E500MC */
+#endif /* CONFIG_E500 */
 
 #ifdef CONFIG_SPE
 /*

arch/powerpc/kvm/book3s.c

@@ -535,174 +535,111 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 	return -ENOTSUPP;
 }
 
-int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
+int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id,
+			union kvmppc_one_reg *val)
 {
-	int r;
-	union kvmppc_one_reg val;
-	int size;
+	int r = 0;
 	long int i;
 
-	size = one_reg_size(reg->id);
-	if (size > sizeof(val))
-		return -EINVAL;
-
-	r = vcpu->kvm->arch.kvm_ops->get_one_reg(vcpu, reg->id, &val);
+	r = vcpu->kvm->arch.kvm_ops->get_one_reg(vcpu, id, val);
 	if (r == -EINVAL) {
 		r = 0;
-		switch (reg->id) {
+		switch (id) {
 		case KVM_REG_PPC_DAR:
-			val = get_reg_val(reg->id, kvmppc_get_dar(vcpu));
+			*val = get_reg_val(id, kvmppc_get_dar(vcpu));
 			break;
 		case KVM_REG_PPC_DSISR:
-			val = get_reg_val(reg->id, kvmppc_get_dsisr(vcpu));
+			*val = get_reg_val(id, kvmppc_get_dsisr(vcpu));
 			break;
 		case KVM_REG_PPC_FPR0 ... KVM_REG_PPC_FPR31:
-			i = reg->id - KVM_REG_PPC_FPR0;
-			val = get_reg_val(reg->id, VCPU_FPR(vcpu, i));
+			i = id - KVM_REG_PPC_FPR0;
+			*val = get_reg_val(id, VCPU_FPR(vcpu, i));
 			break;
 		case KVM_REG_PPC_FPSCR:
-			val = get_reg_val(reg->id, vcpu->arch.fp.fpscr);
-			break;
-#ifdef CONFIG_ALTIVEC
-		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
-			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
-				r = -ENXIO;
-				break;
-			}
-			val.vval = vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0];
-			break;
-		case KVM_REG_PPC_VSCR:
-			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
-				r = -ENXIO;
-				break;
-			}
-			val = get_reg_val(reg->id, vcpu->arch.vr.vscr.u[3]);
+			*val = get_reg_val(id, vcpu->arch.fp.fpscr);
 			break;
-		case KVM_REG_PPC_VRSAVE:
-			val = get_reg_val(reg->id, vcpu->arch.vrsave);
-			break;
-#endif /* CONFIG_ALTIVEC */
 #ifdef CONFIG_VSX
 		case KVM_REG_PPC_VSR0 ... KVM_REG_PPC_VSR31:
 			if (cpu_has_feature(CPU_FTR_VSX)) {
-				long int i = reg->id - KVM_REG_PPC_VSR0;
-				val.vsxval[0] = vcpu->arch.fp.fpr[i][0];
-				val.vsxval[1] = vcpu->arch.fp.fpr[i][1];
+				i = id - KVM_REG_PPC_VSR0;
+				val->vsxval[0] = vcpu->arch.fp.fpr[i][0];
+				val->vsxval[1] = vcpu->arch.fp.fpr[i][1];
 			} else {
 				r = -ENXIO;
 			}
 			break;
 #endif /* CONFIG_VSX */
-		case KVM_REG_PPC_DEBUG_INST: {
-			u32 opcode = INS_TW;
-			r = copy_to_user((u32 __user *)(long)reg->addr,
-					 &opcode, sizeof(u32));
+		case KVM_REG_PPC_DEBUG_INST:
+			*val = get_reg_val(id, INS_TW);
 			break;
-		}
 #ifdef CONFIG_KVM_XICS
 		case KVM_REG_PPC_ICP_STATE:
 			if (!vcpu->arch.icp) {
 				r = -ENXIO;
 				break;
 			}
-			val = get_reg_val(reg->id, kvmppc_xics_get_icp(vcpu));
+			*val = get_reg_val(id, kvmppc_xics_get_icp(vcpu));
 			break;
 #endif /* CONFIG_KVM_XICS */
 		case KVM_REG_PPC_FSCR:
-			val = get_reg_val(reg->id, vcpu->arch.fscr);
+			*val = get_reg_val(id, vcpu->arch.fscr);
 			break;
 		case KVM_REG_PPC_TAR:
-			val = get_reg_val(reg->id, vcpu->arch.tar);
+			*val = get_reg_val(id, vcpu->arch.tar);
 			break;
 		case KVM_REG_PPC_EBBHR:
-			val = get_reg_val(reg->id, vcpu->arch.ebbhr);
+			*val = get_reg_val(id, vcpu->arch.ebbhr);
 			break;
 		case KVM_REG_PPC_EBBRR:
-			val = get_reg_val(reg->id, vcpu->arch.ebbrr);
+			*val = get_reg_val(id, vcpu->arch.ebbrr);
 			break;
 		case KVM_REG_PPC_BESCR:
-			val = get_reg_val(reg->id, vcpu->arch.bescr);
+			*val = get_reg_val(id, vcpu->arch.bescr);
 			break;
 		case KVM_REG_PPC_VTB:
-			val = get_reg_val(reg->id, vcpu->arch.vtb);
+			*val = get_reg_val(id, vcpu->arch.vtb);
 			break;
 		case KVM_REG_PPC_IC:
-			val = get_reg_val(reg->id, vcpu->arch.ic);
+			*val = get_reg_val(id, vcpu->arch.ic);
 			break;
 		default:
 			r = -EINVAL;
 			break;
 		}
 	}
-	if (r)
-		return r;
-
-	if (copy_to_user((char __user *)(unsigned long)reg->addr, &val, size))
-		r = -EFAULT;
 
 	return r;
 }
 
-int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
+int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id,
+			union kvmppc_one_reg *val)
 {
-	int r;
-	union kvmppc_one_reg val;
-	int size;
+	int r = 0;
 	long int i;
 
-	size = one_reg_size(reg->id);
-	if (size > sizeof(val))
-		return -EINVAL;
-
-	if (copy_from_user(&val, (char __user *)(unsigned long)reg->addr, size))
-		return -EFAULT;
-
-	r = vcpu->kvm->arch.kvm_ops->set_one_reg(vcpu, reg->id, &val);
+	r = vcpu->kvm->arch.kvm_ops->set_one_reg(vcpu, id, val);
 	if (r == -EINVAL) {
 		r = 0;
-		switch (reg->id) {
+		switch (id) {
 		case KVM_REG_PPC_DAR:
-			kvmppc_set_dar(vcpu, set_reg_val(reg->id, val));
+			kvmppc_set_dar(vcpu, set_reg_val(id, *val));
 			break;
 		case KVM_REG_PPC_DSISR:
-			kvmppc_set_dsisr(vcpu, set_reg_val(reg->id, val));
+			kvmppc_set_dsisr(vcpu, set_reg_val(id, *val));
 			break;
 		case KVM_REG_PPC_FPR0 ... KVM_REG_PPC_FPR31:
-			i = reg->id - KVM_REG_PPC_FPR0;
-			VCPU_FPR(vcpu, i) = set_reg_val(reg->id, val);
+			i = id - KVM_REG_PPC_FPR0;
+			VCPU_FPR(vcpu, i) = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_FPSCR:
-			vcpu->arch.fp.fpscr = set_reg_val(reg->id, val);
-			break;
-#ifdef CONFIG_ALTIVEC
-		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
-			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
-				r = -ENXIO;
-				break;
-			}
-			vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0] = val.vval;
-			break;
-		case KVM_REG_PPC_VSCR:
-			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
-				r = -ENXIO;
-				break;
-			}
-			vcpu->arch.vr.vscr.u[3] = set_reg_val(reg->id, val);
-			break;
-		case KVM_REG_PPC_VRSAVE:
-			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
-				r = -ENXIO;
-				break;
-			}
-			vcpu->arch.vrsave = set_reg_val(reg->id, val);
+			vcpu->arch.fp.fpscr = set_reg_val(id, *val);
 			break;
-#endif /* CONFIG_ALTIVEC */
 #ifdef CONFIG_VSX
 		case KVM_REG_PPC_VSR0 ... KVM_REG_PPC_VSR31:
 			if (cpu_has_feature(CPU_FTR_VSX)) {
-				long int i = reg->id - KVM_REG_PPC_VSR0;
-				vcpu->arch.fp.fpr[i][0] = val.vsxval[0];
-				vcpu->arch.fp.fpr[i][1] = val.vsxval[1];
+				i = id - KVM_REG_PPC_VSR0;
+				vcpu->arch.fp.fpr[i][0] = val->vsxval[0];
+				vcpu->arch.fp.fpr[i][1] = val->vsxval[1];
 			} else {
 				r = -ENXIO;
 			}
@@ -715,29 +652,29 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 				break;
 			}
 			r = kvmppc_xics_set_icp(vcpu,
-						set_reg_val(reg->id, val));
+						set_reg_val(id, *val));
 			break;
 #endif /* CONFIG_KVM_XICS */
 		case KVM_REG_PPC_FSCR:
-			vcpu->arch.fscr = set_reg_val(reg->id, val);
+			vcpu->arch.fscr = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_TAR:
-			vcpu->arch.tar = set_reg_val(reg->id, val);
+			vcpu->arch.tar = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_EBBHR:
-			vcpu->arch.ebbhr = set_reg_val(reg->id, val);
+			vcpu->arch.ebbhr = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_EBBRR:
-			vcpu->arch.ebbrr = set_reg_val(reg->id, val);
+			vcpu->arch.ebbrr = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_BESCR:
-			vcpu->arch.bescr = set_reg_val(reg->id, val);
+			vcpu->arch.bescr = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_VTB:
-			vcpu->arch.vtb = set_reg_val(reg->id, val);
+			vcpu->arch.vtb = set_reg_val(id, *val);
 			break;
 		case KVM_REG_PPC_IC:
-			vcpu->arch.ic = set_reg_val(reg->id, val);
+			vcpu->arch.ic = set_reg_val(id, *val);
 			break;
 		default:
 			r = -EINVAL;
@@ -778,13 +715,12 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
 int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 					struct kvm_guest_debug *dbg)
 {
-	return -EINVAL;
+	vcpu->guest_debug = dbg->control;
+	return 0;
 }
 
-void kvmppc_decrementer_func(unsigned long data)
+void kvmppc_decrementer_func(struct kvm_vcpu *vcpu)
 {
-	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data;
-
 	kvmppc_core_queue_dec(vcpu);
 	kvm_vcpu_kick(vcpu);
 }
@@ -851,9 +787,9 @@ int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
 	return kvm->arch.kvm_ops->unmap_hva_range(kvm, start, end);
 }
 
-int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end)
 {
-	return kvm->arch.kvm_ops->age_hva(kvm, hva);
+	return kvm->arch.kvm_ops->age_hva(kvm, start, end);
 }
 
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)

arch/powerpc/kvm/book3s.h

@@ -17,7 +17,8 @@ extern void kvmppc_core_flush_memslot_hv(struct kvm *kvm,
 extern int kvm_unmap_hva_hv(struct kvm *kvm, unsigned long hva);
 extern int kvm_unmap_hva_range_hv(struct kvm *kvm, unsigned long start,
 				  unsigned long end);
-extern int kvm_age_hva_hv(struct kvm *kvm, unsigned long hva);
+extern int kvm_age_hva_hv(struct kvm *kvm, unsigned long start,
+			  unsigned long end);
 extern int kvm_test_age_hva_hv(struct kvm *kvm, unsigned long hva);
 extern void kvm_set_spte_hva_hv(struct kvm *kvm, unsigned long hva, pte_t pte);
 

arch/powerpc/kvm/book3s_64_mmu_hv.c

@@ -1002,11 +1002,11 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 	return ret;
 }
 
-int kvm_age_hva_hv(struct kvm *kvm, unsigned long hva)
+int kvm_age_hva_hv(struct kvm *kvm, unsigned long start, unsigned long end)
 {
 	if (!kvm->arch.using_mmu_notifiers)
 		return 0;
-	return kvm_handle_hva(kvm, hva, kvm_age_rmapp);
+	return kvm_handle_hva_range(kvm, start, end, kvm_age_rmapp);
 }
 
 static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,

arch/powerpc/kvm/book3s_hv.c

@@ -725,6 +725,30 @@ static int kvmppc_hcall_impl_hv(unsigned long cmd)
 	return kvmppc_hcall_impl_hv_realmode(cmd);
 }
 
+static int kvmppc_emulate_debug_inst(struct kvm_run *run,
+					struct kvm_vcpu *vcpu)
+{
+	u32 last_inst;
+
+	if (kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst) !=
+					EMULATE_DONE) {
+		/*
+		 * Fetch failed, so return to guest and
+		 * try executing it again.
+		 */
+		return RESUME_GUEST;
+	}
+
+	if (last_inst == KVMPPC_INST_SW_BREAKPOINT) {
+		run->exit_reason = KVM_EXIT_DEBUG;
+		run->debug.arch.address = kvmppc_get_pc(vcpu);
+		return RESUME_HOST;
+	} else {
+		kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
+		return RESUME_GUEST;
+	}
+}
+
 static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
 				 struct task_struct *tsk)
 {
@@ -807,12 +831,18 @@ static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		break;
 	/*
 	 * This occurs if the guest executes an illegal instruction.
-	 * We just generate a program interrupt to the guest, since
-	 * we don't emulate any guest instructions at this stage.
+	 * If the guest debug is disabled, generate a program interrupt
+	 * to the guest. If guest debug is enabled, we need to check
+	 * whether the instruction is a software breakpoint instruction.
+	 * Accordingly return to Guest or Host.
 	 */
 	case BOOK3S_INTERRUPT_H_EMUL_ASSIST:
-		kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
-		r = RESUME_GUEST;
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) {
+			r = kvmppc_emulate_debug_inst(run, vcpu);
+		} else {
+			kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
+			r = RESUME_GUEST;
+		}
 		break;
 	/*
 	 * This occurs if the guest (kernel or userspace), does something that
@@ -856,7 +886,9 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
 {
 	int i, j;
 
-	kvmppc_set_pvr_hv(vcpu, sregs->pvr);
+	/* Only accept the same PVR as the host's, since we can't spoof it */
+	if (sregs->pvr != vcpu->arch.pvr)
+		return -EINVAL;
 
 	j = 0;
 	for (i = 0; i < vcpu->arch.slb_nr; i++) {
@@ -922,6 +954,9 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 	long int i;
 
 	switch (id) {
+	case KVM_REG_PPC_DEBUG_INST:
+		*val = get_reg_val(id, KVMPPC_INST_SW_BREAKPOINT);
+		break;
 	case KVM_REG_PPC_HIOR:
 		*val = get_reg_val(id, 0);
 		break;
@@ -1489,7 +1524,7 @@ static void kvmppc_remove_runnable(struct kvmppc_vcore *vc,
 static int kvmppc_grab_hwthread(int cpu)
 {
 	struct paca_struct *tpaca;
-	long timeout = 1000;
+	long timeout = 10000;
 
 	tpaca = &paca[cpu];
 

arch/powerpc/kvm/book3s_hv_builtin.c

@@ -163,6 +163,12 @@ void __init kvm_cma_reserve(void)
 	unsigned long align_size;
 	struct memblock_region *reg;
 	phys_addr_t selected_size = 0;
+
+	/*
+	 * We need CMA reservation only when we are in HV mode
+	 */
+	if (!cpu_has_feature(CPU_FTR_HVMODE))
+		return;
 	/*
 	 * We cannot use memblock_phys_mem_size() here, because
 	 * memblock_analyze() has not been called yet.

arch/powerpc/kvm/book3s_hv_rmhandlers.S

@@ -355,6 +355,7 @@ kvmppc_hv_entry:
 	 * MSR = ~IR|DR
 	 * R13 = PACA
 	 * R1 = host R1
+	 * R2 = TOC
 	 * all other volatile GPRS = free
 	 */
 	mflr	r0
@@ -503,7 +504,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 toc_tlbie_lock:
 	.tc	native_tlbie_lock[TC],native_tlbie_lock
 	.previous
-	ld	r3,toc_tlbie_lock@toc(2)
+	ld	r3,toc_tlbie_lock@toc(r2)
 #ifdef __BIG_ENDIAN__
 	lwz	r8,PACA_LOCK_TOKEN(r13)
 #else

arch/powerpc/kvm/book3s_pr.c

@@ -295,7 +295,8 @@ static int kvm_unmap_hva_range_pr(struct kvm *kvm, unsigned long start,
 	return 0;
 }
 
-static int kvm_age_hva_pr(struct kvm *kvm, unsigned long hva)
+static int kvm_age_hva_pr(struct kvm *kvm, unsigned long start,
+			  unsigned long end)
 {
 	/* XXX could be more clever ;) */
 	return 0;
@@ -1319,6 +1320,9 @@ static int kvmppc_get_one_reg_pr(struct kvm_vcpu *vcpu, u64 id,
 	int r = 0;
 
 	switch (id) {
+	case KVM_REG_PPC_DEBUG_INST:
+		*val = get_reg_val(id, KVMPPC_INST_SW_BREAKPOINT);
+		break;
 	case KVM_REG_PPC_HIOR:
 		*val = get_reg_val(id, to_book3s(vcpu)->hior);
 		break;

arch/powerpc/kvm/booke.c

@@ -124,6 +124,40 @@ static void kvmppc_vcpu_sync_spe(struct kvm_vcpu *vcpu)
 }
 #endif
 
+/*
+ * Load up guest vcpu FP state if it's needed.
+ * It also set the MSR_FP in thread so that host know
+ * we're holding FPU, and then host can help to save
+ * guest vcpu FP state if other threads require to use FPU.
+ * This simulates an FP unavailable fault.
+ *
+ * It requires to be called with preemption disabled.
+ */
+static inline void kvmppc_load_guest_fp(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_PPC_FPU
+	if (!(current->thread.regs->msr & MSR_FP)) {
+		enable_kernel_fp();
+		load_fp_state(&vcpu->arch.fp);
+		current->thread.fp_save_area = &vcpu->arch.fp;
+		current->thread.regs->msr |= MSR_FP;
+	}
+#endif
+}
+
+/*
+ * Save guest vcpu FP state into thread.
+ * It requires to be called with preemption disabled.
+ */
+static inline void kvmppc_save_guest_fp(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_PPC_FPU
+	if (current->thread.regs->msr & MSR_FP)
+		giveup_fpu(current);
+	current->thread.fp_save_area = NULL;
+#endif
+}
+
 static void kvmppc_vcpu_sync_fpu(struct kvm_vcpu *vcpu)
 {
 #if defined(CONFIG_PPC_FPU) && !defined(CONFIG_KVM_BOOKE_HV)
@@ -134,6 +168,40 @@ static void kvmppc_vcpu_sync_fpu(struct kvm_vcpu *vcpu)
 #endif
 }
 
+/*
+ * Simulate AltiVec unavailable fault to load guest state
+ * from thread to AltiVec unit.
+ * It requires to be called with preemption disabled.
+ */
+static inline void kvmppc_load_guest_altivec(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_ALTIVEC
+	if (cpu_has_feature(CPU_FTR_ALTIVEC)) {
+		if (!(current->thread.regs->msr & MSR_VEC)) {
+			enable_kernel_altivec();
+			load_vr_state(&vcpu->arch.vr);
+			current->thread.vr_save_area = &vcpu->arch.vr;
+			current->thread.regs->msr |= MSR_VEC;
+		}
+	}
+#endif
+}
+
+/*
+ * Save guest vcpu AltiVec state into thread.
+ * It requires to be called with preemption disabled.
+ */
+static inline void kvmppc_save_guest_altivec(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_ALTIVEC
+	if (cpu_has_feature(CPU_FTR_ALTIVEC)) {
+		if (current->thread.regs->msr & MSR_VEC)
+			giveup_altivec(current);
+		current->thread.vr_save_area = NULL;
+	}
+#endif
+}
+
 static void kvmppc_vcpu_sync_debug(struct kvm_vcpu *vcpu)
 {
 	/* Synchronize guest's desire to get debug interrupts into shadow MSR */
@@ -267,6 +335,16 @@ static void kvmppc_core_dequeue_watchdog(struct kvm_vcpu *vcpu)
 	clear_bit(BOOKE_IRQPRIO_WATCHDOG, &vcpu->arch.pending_exceptions);
 }
 
+void kvmppc_core_queue_debug(struct kvm_vcpu *vcpu)
+{
+	kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_DEBUG);
+}
+
+void kvmppc_core_dequeue_debug(struct kvm_vcpu *vcpu)
+{
+	clear_bit(BOOKE_IRQPRIO_DEBUG, &vcpu->arch.pending_exceptions);
+}
+
 static void set_guest_srr(struct kvm_vcpu *vcpu, unsigned long srr0, u32 srr1)
 {
 	kvmppc_set_srr0(vcpu, srr0);
@@ -341,9 +419,15 @@ static int kvmppc_booke_irqprio_deliver(struct kvm_vcpu *vcpu,
 	case BOOKE_IRQPRIO_ITLB_MISS:
 	case BOOKE_IRQPRIO_SYSCALL:
 	case BOOKE_IRQPRIO_FP_UNAVAIL:
+#ifdef CONFIG_SPE_POSSIBLE
 	case BOOKE_IRQPRIO_SPE_UNAVAIL:
 	case BOOKE_IRQPRIO_SPE_FP_DATA:
 	case BOOKE_IRQPRIO_SPE_FP_ROUND:
+#endif
+#ifdef CONFIG_ALTIVEC
+	case BOOKE_IRQPRIO_ALTIVEC_UNAVAIL:
+	case BOOKE_IRQPRIO_ALTIVEC_ASSIST:
+#endif
 	case BOOKE_IRQPRIO_AP_UNAVAIL:
 		allowed = 1;
 		msr_mask = MSR_CE | MSR_ME | MSR_DE;
@@ -377,7 +461,11 @@ static int kvmppc_booke_irqprio_deliver(struct kvm_vcpu *vcpu,
 		allowed = vcpu->arch.shared->msr & MSR_DE;
 		allowed = allowed && !crit;
 		msr_mask = MSR_ME;
-		int_class = INT_CLASS_CRIT;
+		if (cpu_has_feature(CPU_FTR_DEBUG_LVL_EXC))
+			int_class = INT_CLASS_DBG;
+		else
+			int_class = INT_CLASS_CRIT;
+
 		break;
 	}
 
@@ -654,20 +742,27 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 
 	/*
 	 * Since we can't trap on MSR_FP in GS-mode, we consider the guest
-	 * as always using the FPU.  Kernel usage of FP (via
-	 * enable_kernel_fp()) in this thread must not occur while
-	 * vcpu->fpu_active is set.
+	 * as always using the FPU.
 	 */
-	vcpu->fpu_active = 1;
-
 	kvmppc_load_guest_fp(vcpu);
 #endif
 
+#ifdef CONFIG_ALTIVEC
+	/* Save userspace AltiVec state in stack */
+	if (cpu_has_feature(CPU_FTR_ALTIVEC))
+		enable_kernel_altivec();
+	/*
+	 * Since we can't trap on MSR_VEC in GS-mode, we consider the guest
+	 * as always using the AltiVec.
+	 */
+	kvmppc_load_guest_altivec(vcpu);
+#endif
+
 	/* Switch to guest debug context */
-	debug = vcpu->arch.shadow_dbg_reg;
+	debug = vcpu->arch.dbg_reg;
 	switch_booke_debug_regs(&debug);
 	debug = current->thread.debug;
-	current->thread.debug = vcpu->arch.shadow_dbg_reg;
+	current->thread.debug = vcpu->arch.dbg_reg;
 
 	vcpu->arch.pgdir = current->mm->pgd;
 	kvmppc_fix_ee_before_entry();
@@ -683,8 +778,10 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 
 #ifdef CONFIG_PPC_FPU
 	kvmppc_save_guest_fp(vcpu);
+#endif
 
-	vcpu->fpu_active = 0;
+#ifdef CONFIG_ALTIVEC
+	kvmppc_save_guest_altivec(vcpu);
 #endif
 
 out:
@@ -728,9 +825,36 @@ static int emulation_exit(struct kvm_run *run, struct kvm_vcpu *vcpu)
 
 static int kvmppc_handle_debug(struct kvm_run *run, struct kvm_vcpu *vcpu)
 {
-	struct debug_reg *dbg_reg = &(vcpu->arch.shadow_dbg_reg);
+	struct debug_reg *dbg_reg = &(vcpu->arch.dbg_reg);
 	u32 dbsr = vcpu->arch.dbsr;
 
+	if (vcpu->guest_debug == 0) {
+		/*
+		 * Debug resources belong to Guest.
+		 * Imprecise debug event is not injected
+		 */
+		if (dbsr & DBSR_IDE) {
+			dbsr &= ~DBSR_IDE;
+			if (!dbsr)
+				return RESUME_GUEST;
+		}
+
+		if (dbsr && (vcpu->arch.shared->msr & MSR_DE) &&
+			    (vcpu->arch.dbg_reg.dbcr0 & DBCR0_IDM))
+			kvmppc_core_queue_debug(vcpu);
+
+		/* Inject a program interrupt if trap debug is not allowed */
+		if ((dbsr & DBSR_TIE) && !(vcpu->arch.shared->msr & MSR_DE))
+			kvmppc_core_queue_program(vcpu, ESR_PTR);
+
+		return RESUME_GUEST;
+	}
+
+	/*
+	 * Debug resource owned by userspace.
+	 * Clear guest dbsr (vcpu->arch.dbsr)
+	 */
+	vcpu->arch.dbsr = 0;
 	run->debug.arch.status = 0;
 	run->debug.arch.address = vcpu->arch.pc;
 
@@ -868,7 +992,12 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	case BOOKE_INTERRUPT_DATA_STORAGE:
 	case BOOKE_INTERRUPT_DTLB_MISS:
 	case BOOKE_INTERRUPT_HV_PRIV:
-		emulated = kvmppc_get_last_inst(vcpu, false, &last_inst);
+		emulated = kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst);
+		break;
+	case BOOKE_INTERRUPT_PROGRAM:
+		/* SW breakpoints arrive as illegal instructions on HV */
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			emulated = kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst);
 		break;
 	default:
 		break;
@@ -947,6 +1076,18 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		break;
 
 	case BOOKE_INTERRUPT_PROGRAM:
+		if ((vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) &&
+			(last_inst == KVMPPC_INST_SW_BREAKPOINT)) {
+			/*
+			 * We are here because of an SW breakpoint instr,
+			 * so lets return to host to handle.
+			 */
+			r = kvmppc_handle_debug(run, vcpu);
+			run->exit_reason = KVM_EXIT_DEBUG;
+			kvmppc_account_exit(vcpu, DEBUG_EXITS);
+			break;
+		}
+
 		if (vcpu->arch.shared->msr & (MSR_PR | MSR_GS)) {
 			/*
 			 * Program traps generated by user-level software must
@@ -991,7 +1132,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_SPE_FP_ROUND);
 		r = RESUME_GUEST;
 		break;
-#else
+#elif defined(CONFIG_SPE_POSSIBLE)
 	case BOOKE_INTERRUPT_SPE_UNAVAIL:
 		/*
 		 * Guest wants SPE, but host kernel doesn't support it.  Send
@@ -1012,6 +1153,22 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		run->hw.hardware_exit_reason = exit_nr;
 		r = RESUME_HOST;
 		break;
+#endif /* CONFIG_SPE_POSSIBLE */
+
+/*
+ * On cores with Vector category, KVM is loaded only if CONFIG_ALTIVEC,
+ * see kvmppc_core_check_processor_compat().
+ */
+#ifdef CONFIG_ALTIVEC
+	case BOOKE_INTERRUPT_ALTIVEC_UNAVAIL:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_ALTIVEC_UNAVAIL);
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_ALTIVEC_ASSIST:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_ALTIVEC_ASSIST);
+		r = RESUME_GUEST;
+		break;
 #endif
 
 	case BOOKE_INTERRUPT_DATA_STORAGE:
@@ -1188,6 +1345,8 @@ out:
 		else {
 			/* interrupts now hard-disabled */
 			kvmppc_fix_ee_before_entry();
+			kvmppc_load_guest_fp(vcpu);
+			kvmppc_load_guest_altivec(vcpu);
 		}
 	}
 
@@ -1243,6 +1402,11 @@ int kvmppc_subarch_vcpu_init(struct kvm_vcpu *vcpu)
 	setup_timer(&vcpu->arch.wdt_timer, kvmppc_watchdog_func,
 		    (unsigned long)vcpu);
 
+	/*
+	 * Clear DBSR.MRR to avoid guest debug interrupt as
+	 * this is of host interest
+	 */
+	mtspr(SPRN_DBSR, DBSR_MRR);
 	return 0;
 }
 
@@ -1457,144 +1621,125 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 	return vcpu->kvm->arch.kvm_ops->set_sregs(vcpu, sregs);
 }
 
-int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
+int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id,
+			union kvmppc_one_reg *val)
 {
 	int r = 0;
-	union kvmppc_one_reg val;
-	int size;
-
-	size = one_reg_size(reg->id);
-	if (size > sizeof(val))
-		return -EINVAL;
 
-	switch (reg->id) {
+	switch (id) {
 	case KVM_REG_PPC_IAC1:
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac1);
+		*val = get_reg_val(id, vcpu->arch.dbg_reg.iac1);
 		break;
 	case KVM_REG_PPC_IAC2:
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac2);
+		*val = get_reg_val(id, vcpu->arch.dbg_reg.iac2);
 		break;
 #if CONFIG_PPC_ADV_DEBUG_IACS > 2
 	case KVM_REG_PPC_IAC3:
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac3);
+		*val = get_reg_val(id, vcpu->arch.dbg_reg.iac3);
 		break;
 	case KVM_REG_PPC_IAC4:
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac4);
+		*val = get_reg_val(id, vcpu->arch.dbg_reg.iac4);
 		break;
 #endif
 	case KVM_REG_PPC_DAC1:
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.dac1);
+		*val = get_reg_val(id, vcpu->arch.dbg_reg.dac1);
 		break;
 	case KVM_REG_PPC_DAC2:
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.dac2);
+		*val = get_reg_val(id, vcpu->arch.dbg_reg.dac2);
 		break;
 	case KVM_REG_PPC_EPR: {
 		u32 epr = kvmppc_get_epr(vcpu);
-		val = get_reg_val(reg->id, epr);
+		*val = get_reg_val(id, epr);
 		break;
 	}
 #if defined(CONFIG_64BIT)
 	case KVM_REG_PPC_EPCR:
-		val = get_reg_val(reg->id, vcpu->arch.epcr);
+		*val = get_reg_val(id, vcpu->arch.epcr);
 		break;
 #endif
 	case KVM_REG_PPC_TCR:
-		val = get_reg_val(reg->id, vcpu->arch.tcr);
+		*val = get_reg_val(id, vcpu->arch.tcr);
 		break;
 	case KVM_REG_PPC_TSR:
-		val = get_reg_val(reg->id, vcpu->arch.tsr);
+		*val = get_reg_val(id, vcpu->arch.tsr);
 		break;
 	case KVM_REG_PPC_DEBUG_INST:
-		val = get_reg_val(reg->id, KVMPPC_INST_EHPRIV_DEBUG);
+		*val = get_reg_val(id, KVMPPC_INST_SW_BREAKPOINT);
 		break;
 	case KVM_REG_PPC_VRSAVE:
-		val = get_reg_val(reg->id, vcpu->arch.vrsave);
+		*val = get_reg_val(id, vcpu->arch.vrsave);
 		break;
 	default:
-		r = vcpu->kvm->arch.kvm_ops->get_one_reg(vcpu, reg->id, &val);
+		r = vcpu->kvm->arch.kvm_ops->get_one_reg(vcpu, id, val);
 		break;
 	}
 
-	if (r)
-		return r;
-
-	if (copy_to_user((char __user *)(unsigned long)reg->addr, &val, size))
-		r = -EFAULT;
-
 	return r;
 }
 
-int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
+int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id,
+			union kvmppc_one_reg *val)
 {
 	int r = 0;
-	union kvmppc_one_reg val;
-	int size;
 
-	size = one_reg_size(reg->id);
-	if (size > sizeof(val))
-		return -EINVAL;
-
-	if (copy_from_user(&val, (char __user *)(unsigned long)reg->addr, size))
-		return -EFAULT;
-
-	switch (reg->id) {
+	switch (id) {
 	case KVM_REG_PPC_IAC1:
-		vcpu->arch.dbg_reg.iac1 = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.iac1 = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_IAC2:
-		vcpu->arch.dbg_reg.iac2 = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.iac2 = set_reg_val(id, *val);
 		break;
 #if CONFIG_PPC_ADV_DEBUG_IACS > 2
 	case KVM_REG_PPC_IAC3:
-		vcpu->arch.dbg_reg.iac3 = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.iac3 = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_IAC4:
-		vcpu->arch.dbg_reg.iac4 = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.iac4 = set_reg_val(id, *val);
 		break;
 #endif
 	case KVM_REG_PPC_DAC1:
-		vcpu->arch.dbg_reg.dac1 = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.dac1 = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_DAC2:
-		vcpu->arch.dbg_reg.dac2 = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.dac2 = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_EPR: {
-		u32 new_epr = set_reg_val(reg->id, val);
+		u32 new_epr = set_reg_val(id, *val);
 		kvmppc_set_epr(vcpu, new_epr);
 		break;
 	}
 #if defined(CONFIG_64BIT)
 	case KVM_REG_PPC_EPCR: {
-		u32 new_epcr = set_reg_val(reg->id, val);
+		u32 new_epcr = set_reg_val(id, *val);
 		kvmppc_set_epcr(vcpu, new_epcr);
 		break;
 	}
 #endif
 	case KVM_REG_PPC_OR_TSR: {
-		u32 tsr_bits = set_reg_val(reg->id, val);
+		u32 tsr_bits = set_reg_val(id, *val);
 		kvmppc_set_tsr_bits(vcpu, tsr_bits);
 		break;
 	}
 	case KVM_REG_PPC_CLEAR_TSR: {
-		u32 tsr_bits = set_reg_val(reg->id, val);
+		u32 tsr_bits = set_reg_val(id, *val);
 		kvmppc_clr_tsr_bits(vcpu, tsr_bits);
 		break;
 	}
 	case KVM_REG_PPC_TSR: {
-		u32 tsr = set_reg_val(reg->id, val);
+		u32 tsr = set_reg_val(id, *val);
 		kvmppc_set_tsr(vcpu, tsr);
 		break;
 	}
 	case KVM_REG_PPC_TCR: {
-		u32 tcr = set_reg_val(reg->id, val);
+		u32 tcr = set_reg_val(id, *val);
 		kvmppc_set_tcr(vcpu, tcr);
 		break;
 	}
 	case KVM_REG_PPC_VRSAVE:
-		vcpu->arch.vrsave = set_reg_val(reg->id, val);
+		vcpu->arch.vrsave = set_reg_val(id, *val);
 		break;
 	default:
-		r = vcpu->kvm->arch.kvm_ops->set_one_reg(vcpu, reg->id, &val);
+		r = vcpu->kvm->arch.kvm_ops->set_one_reg(vcpu, id, val);
 		break;
 	}
 
@@ -1694,10 +1839,8 @@ void kvmppc_clr_tsr_bits(struct kvm_vcpu *vcpu, u32 tsr_bits)
 	update_timer_ints(vcpu);
 }
 
-void kvmppc_decrementer_func(unsigned long data)
+void kvmppc_decrementer_func(struct kvm_vcpu *vcpu)
 {
-	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data;
-
 	if (vcpu->arch.tcr & TCR_ARE) {
 		vcpu->arch.dec = vcpu->arch.decar;
 		kvmppc_emulate_dec(vcpu);
@@ -1842,7 +1985,7 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 	int n, b = 0, w = 0;
 
 	if (!(dbg->control & KVM_GUESTDBG_ENABLE)) {
-		vcpu->arch.shadow_dbg_reg.dbcr0 = 0;
+		vcpu->arch.dbg_reg.dbcr0 = 0;
 		vcpu->guest_debug = 0;
 		kvm_guest_protect_msr(vcpu, MSR_DE, false);
 		return 0;
@@ -1850,15 +1993,13 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 
 	kvm_guest_protect_msr(vcpu, MSR_DE, true);
 	vcpu->guest_debug = dbg->control;
-	vcpu->arch.shadow_dbg_reg.dbcr0 = 0;
-	/* Set DBCR0_EDM in guest visible DBCR0 register. */
-	vcpu->arch.dbg_reg.dbcr0 = DBCR0_EDM;
+	vcpu->arch.dbg_reg.dbcr0 = 0;
 
 	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
-		vcpu->arch.shadow_dbg_reg.dbcr0 |= DBCR0_IDM | DBCR0_IC;
+		vcpu->arch.dbg_reg.dbcr0 |= DBCR0_IDM | DBCR0_IC;
 
 	/* Code below handles only HW breakpoints */
-	dbg_reg = &(vcpu->arch.shadow_dbg_reg);
+	dbg_reg = &(vcpu->arch.dbg_reg);
 
 #ifdef CONFIG_KVM_BOOKE_HV
 	/*

arch/powerpc/kvm/booke.h

@@ -32,9 +32,15 @@
 #define BOOKE_IRQPRIO_ALIGNMENT 2
 #define BOOKE_IRQPRIO_PROGRAM 3
 #define BOOKE_IRQPRIO_FP_UNAVAIL 4
+#ifdef CONFIG_SPE_POSSIBLE
 #define BOOKE_IRQPRIO_SPE_UNAVAIL 5
 #define BOOKE_IRQPRIO_SPE_FP_DATA 6
 #define BOOKE_IRQPRIO_SPE_FP_ROUND 7
+#endif
+#ifdef CONFIG_PPC_E500MC
+#define BOOKE_IRQPRIO_ALTIVEC_UNAVAIL 5
+#define BOOKE_IRQPRIO_ALTIVEC_ASSIST 6
+#endif
 #define BOOKE_IRQPRIO_SYSCALL 8
 #define BOOKE_IRQPRIO_AP_UNAVAIL 9
 #define BOOKE_IRQPRIO_DTLB_MISS 10
@@ -116,40 +122,6 @@ extern int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn,
 extern int kvmppc_core_emulate_mfspr_e500(struct kvm_vcpu *vcpu, int sprn,
 					  ulong *spr_val);
 
-/*
- * Load up guest vcpu FP state if it's needed.
- * It also set the MSR_FP in thread so that host know
- * we're holding FPU, and then host can help to save
- * guest vcpu FP state if other threads require to use FPU.
- * This simulates an FP unavailable fault.
- *
- * It requires to be called with preemption disabled.
- */
-static inline void kvmppc_load_guest_fp(struct kvm_vcpu *vcpu)
-{
-#ifdef CONFIG_PPC_FPU
-	if (vcpu->fpu_active && !(current->thread.regs->msr & MSR_FP)) {
-		enable_kernel_fp();
-		load_fp_state(&vcpu->arch.fp);
-		current->thread.fp_save_area = &vcpu->arch.fp;
-		current->thread.regs->msr |= MSR_FP;
-	}
-#endif
-}
-
-/*
- * Save guest vcpu FP state into thread.
- * It requires to be called with preemption disabled.
- */
-static inline void kvmppc_save_guest_fp(struct kvm_vcpu *vcpu)
-{
-#ifdef CONFIG_PPC_FPU
-	if (vcpu->fpu_active && (current->thread.regs->msr & MSR_FP))
-		giveup_fpu(current);
-	current->thread.fp_save_area = NULL;
-#endif
-}
-
 static inline void kvmppc_clear_dbsr(void)
 {
 	mtspr(SPRN_DBSR, mfspr(SPRN_DBSR));

arch/powerpc/kvm/booke_emulate.c

@@ -25,6 +25,7 @@
 
 #define OP_19_XOP_RFI     50
 #define OP_19_XOP_RFCI    51
+#define OP_19_XOP_RFDI    39
 
 #define OP_31_XOP_MFMSR   83
 #define OP_31_XOP_WRTEE   131
@@ -37,6 +38,12 @@ static void kvmppc_emul_rfi(struct kvm_vcpu *vcpu)
 	kvmppc_set_msr(vcpu, vcpu->arch.shared->srr1);
 }
 
+static void kvmppc_emul_rfdi(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.pc = vcpu->arch.dsrr0;
+	kvmppc_set_msr(vcpu, vcpu->arch.dsrr1);
+}
+
 static void kvmppc_emul_rfci(struct kvm_vcpu *vcpu)
 {
 	vcpu->arch.pc = vcpu->arch.csrr0;
@@ -65,6 +72,12 @@ int kvmppc_booke_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			*advance = 0;
 			break;
 
+		case OP_19_XOP_RFDI:
+			kvmppc_emul_rfdi(vcpu);
+			kvmppc_set_exit_type(vcpu, EMULATED_RFDI_EXITS);
+			*advance = 0;
+			break;
+
 		default:
 			emulated = EMULATE_FAIL;
 			break;
@@ -118,6 +131,7 @@ int kvmppc_booke_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 int kvmppc_booke_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 {
 	int emulated = EMULATE_DONE;
+	bool debug_inst = false;
 
 	switch (sprn) {
 	case SPRN_DEAR:
@@ -132,14 +146,128 @@ int kvmppc_booke_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 	case SPRN_CSRR1:
 		vcpu->arch.csrr1 = spr_val;
 		break;
+	case SPRN_DSRR0:
+		vcpu->arch.dsrr0 = spr_val;
+		break;
+	case SPRN_DSRR1:
+		vcpu->arch.dsrr1 = spr_val;
+		break;
+	case SPRN_IAC1:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.iac1 = spr_val;
+		break;
+	case SPRN_IAC2:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.iac2 = spr_val;
+		break;
+#if CONFIG_PPC_ADV_DEBUG_IACS > 2
+	case SPRN_IAC3:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.iac3 = spr_val;
+		break;
+	case SPRN_IAC4:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.iac4 = spr_val;
+		break;
+#endif
+	case SPRN_DAC1:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.dac1 = spr_val;
+		break;
+	case SPRN_DAC2:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.dac2 = spr_val;
+		break;
 	case SPRN_DBCR0:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		spr_val &= (DBCR0_IDM | DBCR0_IC | DBCR0_BT | DBCR0_TIE |
+			DBCR0_IAC1 | DBCR0_IAC2 | DBCR0_IAC3 | DBCR0_IAC4  |
+			DBCR0_DAC1R | DBCR0_DAC1W | DBCR0_DAC2R | DBCR0_DAC2W);
+
 		vcpu->arch.dbg_reg.dbcr0 = spr_val;
 		break;
 	case SPRN_DBCR1:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
 		vcpu->arch.dbg_reg.dbcr1 = spr_val;
 		break;
+	case SPRN_DBCR2:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
+		debug_inst = true;
+		vcpu->arch.dbg_reg.dbcr2 = spr_val;
+		break;
 	case SPRN_DBSR:
+		/*
+		 * If userspace is debugging guest then guest
+		 * can not access debug registers.
+		 */
+		if (vcpu->guest_debug)
+			break;
+
 		vcpu->arch.dbsr &= ~spr_val;
+		if (!(vcpu->arch.dbsr & ~DBSR_IDE))
+			kvmppc_core_dequeue_debug(vcpu);
 		break;
 	case SPRN_TSR:
 		kvmppc_clr_tsr_bits(vcpu, spr_val);
@@ -252,6 +380,10 @@ int kvmppc_booke_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 		emulated = EMULATE_FAIL;
 	}
 
+	if (debug_inst) {
+		current->thread.debug = vcpu->arch.dbg_reg;
+		switch_booke_debug_regs(&vcpu->arch.dbg_reg);
+	}
 	return emulated;
 }
 
@@ -278,12 +410,43 @@ int kvmppc_booke_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
 	case SPRN_CSRR1:
 		*spr_val = vcpu->arch.csrr1;
 		break;
+	case SPRN_DSRR0:
+		*spr_val = vcpu->arch.dsrr0;
+		break;
+	case SPRN_DSRR1:
+		*spr_val = vcpu->arch.dsrr1;
+		break;
+	case SPRN_IAC1:
+		*spr_val = vcpu->arch.dbg_reg.iac1;
+		break;
+	case SPRN_IAC2:
+		*spr_val = vcpu->arch.dbg_reg.iac2;
+		break;
+#if CONFIG_PPC_ADV_DEBUG_IACS > 2
+	case SPRN_IAC3:
+		*spr_val = vcpu->arch.dbg_reg.iac3;
+		break;
+	case SPRN_IAC4:
+		*spr_val = vcpu->arch.dbg_reg.iac4;
+		break;
+#endif
+	case SPRN_DAC1:
+		*spr_val = vcpu->arch.dbg_reg.dac1;
+		break;
+	case SPRN_DAC2:
+		*spr_val = vcpu->arch.dbg_reg.dac2;
+		break;
 	case SPRN_DBCR0:
 		*spr_val = vcpu->arch.dbg_reg.dbcr0;
+		if (vcpu->guest_debug)
+			*spr_val = *spr_val | DBCR0_EDM;
 		break;
 	case SPRN_DBCR1:
 		*spr_val = vcpu->arch.dbg_reg.dbcr1;
 		break;
+	case SPRN_DBCR2:
+		*spr_val = vcpu->arch.dbg_reg.dbcr2;
+		break;
 	case SPRN_DBSR:
 		*spr_val = vcpu->arch.dbsr;
 		break;

arch/powerpc/kvm/bookehv_interrupts.S

@@ -238,7 +238,7 @@ kvm_handler BOOKE_INTERRUPT_EXTERNAL, EX_PARAMS(GEN), \
 kvm_handler BOOKE_INTERRUPT_ALIGNMENT, EX_PARAMS(GEN), \
 	SPRN_SRR0, SPRN_SRR1,(NEED_DEAR | NEED_ESR)
 kvm_handler BOOKE_INTERRUPT_PROGRAM, EX_PARAMS(GEN), \
-	SPRN_SRR0, SPRN_SRR1,NEED_ESR
+	SPRN_SRR0, SPRN_SRR1, (NEED_ESR | NEED_EMU)
 kvm_handler BOOKE_INTERRUPT_FP_UNAVAIL, EX_PARAMS(GEN), \
 	SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_AP_UNAVAIL, EX_PARAMS(GEN), \
@@ -256,11 +256,9 @@ kvm_handler BOOKE_INTERRUPT_DTLB_MISS, EX_PARAMS_TLB, \
 	SPRN_SRR0, SPRN_SRR1, (NEED_EMU | NEED_DEAR | NEED_ESR)
 kvm_handler BOOKE_INTERRUPT_ITLB_MISS, EX_PARAMS_TLB, \
 	SPRN_SRR0, SPRN_SRR1, 0
-kvm_handler BOOKE_INTERRUPT_SPE_UNAVAIL, EX_PARAMS(GEN), \
+kvm_handler BOOKE_INTERRUPT_ALTIVEC_UNAVAIL, EX_PARAMS(GEN), \
 	SPRN_SRR0, SPRN_SRR1, 0
-kvm_handler BOOKE_INTERRUPT_SPE_FP_DATA, EX_PARAMS(GEN), \
-	SPRN_SRR0, SPRN_SRR1, 0
-kvm_handler BOOKE_INTERRUPT_SPE_FP_ROUND, EX_PARAMS(GEN), \
+kvm_handler BOOKE_INTERRUPT_ALTIVEC_ASSIST, EX_PARAMS(GEN), \
 	SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_PERFORMANCE_MONITOR, EX_PARAMS(GEN), \
 	SPRN_SRR0, SPRN_SRR1, 0
@@ -350,7 +348,7 @@ kvm_handler BOOKE_INTERRUPT_INST_STORAGE, SPRN_SRR0, SPRN_SRR1, NEED_ESR
 kvm_handler BOOKE_INTERRUPT_EXTERNAL, SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_ALIGNMENT, \
 	SPRN_SRR0, SPRN_SRR1, (NEED_DEAR | NEED_ESR)
-kvm_handler BOOKE_INTERRUPT_PROGRAM, SPRN_SRR0, SPRN_SRR1, NEED_ESR
+kvm_handler BOOKE_INTERRUPT_PROGRAM, SPRN_SRR0, SPRN_SRR1, (NEED_ESR | NEED_EMU)
 kvm_handler BOOKE_INTERRUPT_FP_UNAVAIL, SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_SYSCALL, SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_AP_UNAVAIL, SPRN_SRR0, SPRN_SRR1, 0
@@ -361,9 +359,6 @@ kvm_lvl_handler BOOKE_INTERRUPT_WATCHDOG, \
 kvm_handler BOOKE_INTERRUPT_DTLB_MISS, \
 	SPRN_SRR0, SPRN_SRR1, (NEED_EMU | NEED_DEAR | NEED_ESR)
 kvm_handler BOOKE_INTERRUPT_ITLB_MISS, SPRN_SRR0, SPRN_SRR1, 0
-kvm_handler BOOKE_INTERRUPT_SPE_UNAVAIL, SPRN_SRR0, SPRN_SRR1, 0
-kvm_handler BOOKE_INTERRUPT_SPE_FP_DATA, SPRN_SRR0, SPRN_SRR1, 0
-kvm_handler BOOKE_INTERRUPT_SPE_FP_ROUND, SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_PERFORMANCE_MONITOR, SPRN_SRR0, SPRN_SRR1, 0
 kvm_handler BOOKE_INTERRUPT_DOORBELL, SPRN_SRR0, SPRN_SRR1, 0
 kvm_lvl_handler BOOKE_INTERRUPT_DOORBELL_CRITICAL, \

arch/powerpc/kvm/e500.h

@@ -22,6 +22,7 @@
 #include <linux/kvm_host.h>
 #include <asm/mmu-book3e.h>
 #include <asm/tlb.h>
+#include <asm/cputhreads.h>
 
 enum vcpu_ftr {
 	VCPU_FTR_MMU_V2
@@ -289,6 +290,25 @@ void kvmppc_e500_tlbil_all(struct kvmppc_vcpu_e500 *vcpu_e500);
 #define kvmppc_e500_get_tlb_stid(vcpu, gtlbe)       get_tlb_tid(gtlbe)
 #define get_tlbmiss_tid(vcpu)           get_cur_pid(vcpu)
 #define get_tlb_sts(gtlbe)              (gtlbe->mas1 & MAS1_TS)
+
+/*
+ * These functions should be called with preemption disabled
+ * and the returned value is valid only in that context
+ */
+static inline int get_thread_specific_lpid(int vm_lpid)
+{
+	int vcpu_lpid = vm_lpid;
+
+	if (threads_per_core == 2)
+		vcpu_lpid |= smp_processor_id() & 1;
+
+	return vcpu_lpid;
+}
+
+static inline int get_lpid(struct kvm_vcpu *vcpu)
+{
+	return get_thread_specific_lpid(vcpu->kvm->arch.lpid);
+}
 #else
 unsigned int kvmppc_e500_get_tlb_stid(struct kvm_vcpu *vcpu,
 				      struct kvm_book3e_206_tlb_entry *gtlbe);

arch/powerpc/kvm/e500_emulate.c

@@ -259,6 +259,7 @@ int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong spr_va
 		break;
 
 	/* extra exceptions */
+#ifdef CONFIG_SPE_POSSIBLE
 	case SPRN_IVOR32:
 		vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_UNAVAIL] = spr_val;
 		break;
@@ -268,6 +269,15 @@ int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong spr_va
 	case SPRN_IVOR34:
 		vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_FP_ROUND] = spr_val;
 		break;
+#endif
+#ifdef CONFIG_ALTIVEC
+	case SPRN_IVOR32:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_ALTIVEC_UNAVAIL] = spr_val;
+		break;
+	case SPRN_IVOR33:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_ALTIVEC_ASSIST] = spr_val;
+		break;
+#endif
 	case SPRN_IVOR35:
 		vcpu->arch.ivor[BOOKE_IRQPRIO_PERFORMANCE_MONITOR] = spr_val;
 		break;
@@ -381,6 +391,7 @@ int kvmppc_core_emulate_mfspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong *spr_v
 		break;
 
 	/* extra exceptions */
+#ifdef CONFIG_SPE_POSSIBLE
 	case SPRN_IVOR32:
 		*spr_val = vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_UNAVAIL];
 		break;
@@ -390,6 +401,15 @@ int kvmppc_core_emulate_mfspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong *spr_v
 	case SPRN_IVOR34:
 		*spr_val = vcpu->arch.ivor[BOOKE_IRQPRIO_SPE_FP_ROUND];
 		break;
+#endif
+#ifdef CONFIG_ALTIVEC
+	case SPRN_IVOR32:
+		*spr_val = vcpu->arch.ivor[BOOKE_IRQPRIO_ALTIVEC_UNAVAIL];
+		break;
+	case SPRN_IVOR33:
+		*spr_val = vcpu->arch.ivor[BOOKE_IRQPRIO_ALTIVEC_ASSIST];
+		break;
+#endif
 	case SPRN_IVOR35:
 		*spr_val = vcpu->arch.ivor[BOOKE_IRQPRIO_PERFORMANCE_MONITOR];
 		break;

arch/powerpc/kvm/e500_mmu_host.c

@@ -69,7 +69,8 @@ static inline u32 e500_shadow_mas3_attrib(u32 mas3, int usermode)
  * writing shadow tlb entry to host TLB
  */
 static inline void __write_host_tlbe(struct kvm_book3e_206_tlb_entry *stlbe,
-				     uint32_t mas0)
+				     uint32_t mas0,
+				     uint32_t lpid)
 {
 	unsigned long flags;
 
@@ -80,7 +81,7 @@ static inline void __write_host_tlbe(struct kvm_book3e_206_tlb_entry *stlbe,
 	mtspr(SPRN_MAS3, (u32)stlbe->mas7_3);
 	mtspr(SPRN_MAS7, (u32)(stlbe->mas7_3 >> 32));
 #ifdef CONFIG_KVM_BOOKE_HV
-	mtspr(SPRN_MAS8, stlbe->mas8);
+	mtspr(SPRN_MAS8, MAS8_TGS | get_thread_specific_lpid(lpid));
 #endif
 	asm volatile("isync; tlbwe" : : : "memory");
 
@@ -129,11 +130,12 @@ static inline void write_host_tlbe(struct kvmppc_vcpu_e500 *vcpu_e500,
 
 	if (tlbsel == 0) {
 		mas0 = get_host_mas0(stlbe->mas2);
-		__write_host_tlbe(stlbe, mas0);
+		__write_host_tlbe(stlbe, mas0, vcpu_e500->vcpu.kvm->arch.lpid);
 	} else {
 		__write_host_tlbe(stlbe,
 				  MAS0_TLBSEL(1) |
-				  MAS0_ESEL(to_htlb1_esel(sesel)));
+				  MAS0_ESEL(to_htlb1_esel(sesel)),
+				  vcpu_e500->vcpu.kvm->arch.lpid);
 	}
 }
 
@@ -176,7 +178,7 @@ void kvmppc_map_magic(struct kvm_vcpu *vcpu)
 		       MAS3_SW | MAS3_SR | MAS3_UW | MAS3_UR;
 	magic.mas8 = 0;
 
-	__write_host_tlbe(&magic, MAS0_TLBSEL(1) | MAS0_ESEL(tlbcam_index));
+	__write_host_tlbe(&magic, MAS0_TLBSEL(1) | MAS0_ESEL(tlbcam_index), 0);
 	preempt_enable();
 }
 #endif
@@ -317,10 +319,6 @@ static void kvmppc_e500_setup_stlbe(
 	stlbe->mas2 = (gvaddr & MAS2_EPN) | (ref->flags & E500_TLB_MAS2_ATTR);
 	stlbe->mas7_3 = ((u64)pfn << PAGE_SHIFT) |
 			e500_shadow_mas3_attrib(gtlbe->mas7_3, pr);
-
-#ifdef CONFIG_KVM_BOOKE_HV
-	stlbe->mas8 = MAS8_TGS | vcpu->kvm->arch.lpid;
-#endif
 }
 
 static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500,
@@ -633,7 +631,7 @@ int kvmppc_load_last_inst(struct kvm_vcpu *vcpu, enum instruction_type type,
 
 	local_irq_save(flags);
 	mtspr(SPRN_MAS6, (vcpu->arch.pid << MAS6_SPID_SHIFT) | addr_space);
-	mtspr(SPRN_MAS5, MAS5_SGS | vcpu->kvm->arch.lpid);
+	mtspr(SPRN_MAS5, MAS5_SGS | get_lpid(vcpu));
 	asm volatile("tlbsx 0, %[geaddr]\n" : :
 		     [geaddr] "r" (geaddr));
 	mtspr(SPRN_MAS5, 0);
@@ -732,7 +730,7 @@ int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
 	return 0;
 }
 
-int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end)
 {
 	/* XXX could be more clever ;) */
 	return 0;

arch/powerpc/kvm/e500mc.c

@@ -48,10 +48,11 @@ void kvmppc_set_pending_interrupt(struct kvm_vcpu *vcpu, enum int_class type)
 		return;
 	}
 
-
-	tag = PPC_DBELL_LPID(vcpu->kvm->arch.lpid) | vcpu->vcpu_id;
+	preempt_disable();
+	tag = PPC_DBELL_LPID(get_lpid(vcpu)) | vcpu->vcpu_id;
 	mb();
 	ppc_msgsnd(dbell_type, 0, tag);
+	preempt_enable();
 }
 
 /* gtlbe must not be mapped by more than one host tlb entry */
@@ -60,12 +61,11 @@ void kvmppc_e500_tlbil_one(struct kvmppc_vcpu_e500 *vcpu_e500,
 {
 	unsigned int tid, ts;
 	gva_t eaddr;
-	u32 val, lpid;
+	u32 val;
 	unsigned long flags;
 
 	ts = get_tlb_ts(gtlbe);
 	tid = get_tlb_tid(gtlbe);
-	lpid = vcpu_e500->vcpu.kvm->arch.lpid;
 
 	/* We search the host TLB to invalidate its shadow TLB entry */
 	val = (tid << 16) | ts;
@@ -74,7 +74,7 @@ void kvmppc_e500_tlbil_one(struct kvmppc_vcpu_e500 *vcpu_e500,
 	local_irq_save(flags);
 
 	mtspr(SPRN_MAS6, val);
-	mtspr(SPRN_MAS5, MAS5_SGS | lpid);
+	mtspr(SPRN_MAS5, MAS5_SGS | get_lpid(&vcpu_e500->vcpu));
 
 	asm volatile("tlbsx 0, %[eaddr]\n" : : [eaddr] "r" (eaddr));
 	val = mfspr(SPRN_MAS1);
@@ -95,7 +95,7 @@ void kvmppc_e500_tlbil_all(struct kvmppc_vcpu_e500 *vcpu_e500)
 	unsigned long flags;
 
 	local_irq_save(flags);
-	mtspr(SPRN_MAS5, MAS5_SGS | vcpu_e500->vcpu.kvm->arch.lpid);
+	mtspr(SPRN_MAS5, MAS5_SGS | get_lpid(&vcpu_e500->vcpu));
 	asm volatile("tlbilxlpid");
 	mtspr(SPRN_MAS5, 0);
 	local_irq_restore(flags);
@@ -110,6 +110,7 @@ void kvmppc_mmu_msr_notify(struct kvm_vcpu *vcpu, u32 old_msr)
 {
 }
 
+/* We use two lpids per VM */
 static DEFINE_PER_CPU(struct kvm_vcpu *[KVMPPC_NR_LPIDS], last_vcpu_of_lpid);
 
 static void kvmppc_core_vcpu_load_e500mc(struct kvm_vcpu *vcpu, int cpu)
@@ -118,10 +119,12 @@ static void kvmppc_core_vcpu_load_e500mc(struct kvm_vcpu *vcpu, int cpu)
 
 	kvmppc_booke_vcpu_load(vcpu, cpu);
 
-	mtspr(SPRN_LPID, vcpu->kvm->arch.lpid);
+	mtspr(SPRN_LPID, get_lpid(vcpu));
 	mtspr(SPRN_EPCR, vcpu->arch.shadow_epcr);
 	mtspr(SPRN_GPIR, vcpu->vcpu_id);
 	mtspr(SPRN_MSRP, vcpu->arch.shadow_msrp);
+	vcpu->arch.eplc = EPC_EGS | (get_lpid(vcpu) << EPC_ELPID_SHIFT);
+	vcpu->arch.epsc = vcpu->arch.eplc;
 	mtspr(SPRN_EPLC, vcpu->arch.eplc);
 	mtspr(SPRN_EPSC, vcpu->arch.epsc);
 
@@ -141,12 +144,10 @@ static void kvmppc_core_vcpu_load_e500mc(struct kvm_vcpu *vcpu, int cpu)
 	mtspr(SPRN_GESR, vcpu->arch.shared->esr);
 
 	if (vcpu->arch.oldpir != mfspr(SPRN_PIR) ||
-	    __get_cpu_var(last_vcpu_of_lpid)[vcpu->kvm->arch.lpid] != vcpu) {
+	    __get_cpu_var(last_vcpu_of_lpid)[get_lpid(vcpu)] != vcpu) {
 		kvmppc_e500_tlbil_all(vcpu_e500);
-		__get_cpu_var(last_vcpu_of_lpid)[vcpu->kvm->arch.lpid] = vcpu;
+		__get_cpu_var(last_vcpu_of_lpid)[get_lpid(vcpu)] = vcpu;
 	}
-
-	kvmppc_load_guest_fp(vcpu);
 }
 
 static void kvmppc_core_vcpu_put_e500mc(struct kvm_vcpu *vcpu)
@@ -179,6 +180,16 @@ int kvmppc_core_check_processor_compat(void)
 		r = 0;
 	else if (strcmp(cur_cpu_spec->cpu_name, "e5500") == 0)
 		r = 0;
+#ifdef CONFIG_ALTIVEC
+	/*
+	 * Since guests have the priviledge to enable AltiVec, we need AltiVec
+	 * support in the host to save/restore their context.
+	 * Don't use CPU_FTR_ALTIVEC to identify cores with AltiVec unit
+	 * because it's cleared in the absence of CONFIG_ALTIVEC!
+	 */
+	else if (strcmp(cur_cpu_spec->cpu_name, "e6500") == 0)
+		r = 0;
+#endif
 	else
 		r = -ENOTSUPP;
 
@@ -194,9 +205,7 @@ int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu)
 #ifdef CONFIG_64BIT
 	vcpu->arch.shadow_epcr |= SPRN_EPCR_ICM;
 #endif
-	vcpu->arch.shadow_msrp = MSRP_UCLEP | MSRP_DEP | MSRP_PMMP;
-	vcpu->arch.eplc = EPC_EGS | (vcpu->kvm->arch.lpid << EPC_ELPID_SHIFT);
-	vcpu->arch.epsc = vcpu->arch.eplc;
+	vcpu->arch.shadow_msrp = MSRP_UCLEP | MSRP_PMMP;
 
 	vcpu->arch.pvr = mfspr(SPRN_PVR);
 	vcpu_e500->svr = mfspr(SPRN_SVR);
@@ -356,13 +365,26 @@ static int kvmppc_core_init_vm_e500mc(struct kvm *kvm)
 	if (lpid < 0)
 		return lpid;
 
+	/*
+	 * Use two lpids per VM on cores with two threads like e6500. Use
+	 * even numbers to speedup vcpu lpid computation with consecutive lpids
+	 * per VM. vm1 will use lpids 2 and 3, vm2 lpids 4 and 5, and so on.
+	 */
+	if (threads_per_core == 2)
+		lpid <<= 1;
+
 	kvm->arch.lpid = lpid;
 	return 0;
 }
 
 static void kvmppc_core_destroy_vm_e500mc(struct kvm *kvm)
 {
-	kvmppc_free_lpid(kvm->arch.lpid);
+	int lpid = kvm->arch.lpid;
+
+	if (threads_per_core == 2)
+		lpid >>= 1;
+
+	kvmppc_free_lpid(lpid);
 }
 
 static struct kvmppc_ops kvm_ops_e500mc = {
@@ -390,7 +412,13 @@ static int __init kvmppc_e500mc_init(void)
 	if (r)
 		goto err_out;
 
-	kvmppc_init_lpid(64);
+	/*
+	 * Use two lpids per VM on dual threaded processors like e6500
+	 * to workarround the lack of tlb write conditional instruction.
+	 * Expose half the number of available hardware lpids to the lpid
+	 * allocator.
+	 */
+	kvmppc_init_lpid(KVMPPC_NR_LPIDS/threads_per_core);
 	kvmppc_claim_lpid(0); /* host */
 
 	r = kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), 0, THIS_MODULE);

arch/powerpc/kvm/emulate.c

@@ -219,7 +219,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	/* this default type might be overwritten by subcategories */
 	kvmppc_set_exit_type(vcpu, EMULATED_INST_EXITS);
 
-	emulated = kvmppc_get_last_inst(vcpu, false, &inst);
+	emulated = kvmppc_get_last_inst(vcpu, INST_GENERIC, &inst);
 	if (emulated != EMULATE_DONE)
 		return emulated;
 
@@ -274,6 +274,21 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		}
 		break;
 
+	case 0:
+		/*
+		 * Instruction with primary opcode 0. Based on PowerISA
+		 * these are illegal instructions.
+		 */
+		if (inst == KVMPPC_INST_SW_BREAKPOINT) {
+			run->exit_reason = KVM_EXIT_DEBUG;
+			run->debug.arch.address = kvmppc_get_pc(vcpu);
+			emulated = EMULATE_EXIT_USER;
+			advance = 0;
+		} else
+			emulated = EMULATE_FAIL;
+
+		break;
+
 	default:
 		emulated = EMULATE_FAIL;
 	}

arch/powerpc/kvm/emulate_loadstore.c

@@ -58,7 +58,7 @@ int kvmppc_emulate_loadstore(struct kvm_vcpu *vcpu)
 	/* this default type might be overwritten by subcategories */
 	kvmppc_set_exit_type(vcpu, EMULATED_INST_EXITS);
 
-	emulated = kvmppc_get_last_inst(vcpu, false, &inst);
+	emulated = kvmppc_get_last_inst(vcpu, INST_GENERIC, &inst);
 	if (emulated != EMULATE_DONE)
 		return emulated;
 

arch/powerpc/kvm/powerpc.c

@@ -294,7 +294,7 @@ int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	{
 		u32 last_inst;
 
-		kvmppc_get_last_inst(vcpu, false, &last_inst);
+		kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst);
 		/* XXX Deliver Program interrupt to guest. */
 		pr_emerg("%s: emulation failed (%08x)\n", __func__, last_inst);
 		r = RESUME_HOST;
@@ -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)
 {
@@ -658,7 +638,6 @@ void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
 {
 	/* Make sure we're not using the vcpu anymore */
 	hrtimer_cancel(&vcpu->arch.dec_timer);
-	tasklet_kill(&vcpu->arch.tasklet);
 
 	kvmppc_remove_vcpu_debugfs(vcpu);
 
@@ -684,16 +663,12 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 	return kvmppc_core_pending_dec(vcpu);
 }
 
-/*
- * low level hrtimer wake routine. Because this runs in hardirq context
- * we schedule a tasklet to do the real work.
- */
 enum hrtimer_restart kvmppc_decrementer_wakeup(struct hrtimer *timer)
 {
 	struct kvm_vcpu *vcpu;
 
 	vcpu = container_of(timer, struct kvm_vcpu, arch.dec_timer);
-	tasklet_schedule(&vcpu->arch.tasklet);
+	kvmppc_decrementer_func(vcpu);
 
 	return HRTIMER_NORESTART;
 }
@@ -703,7 +678,6 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	int ret;
 
 	hrtimer_init(&vcpu->arch.dec_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
-	tasklet_init(&vcpu->arch.tasklet, kvmppc_decrementer_func, (ulong)vcpu);
 	vcpu->arch.dec_timer.function = kvmppc_decrementer_wakeup;
 	vcpu->arch.dec_expires = ~(u64)0;
 
@@ -927,6 +901,103 @@ int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
 }
 EXPORT_SYMBOL_GPL(kvmppc_handle_store);
 
+int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
+{
+	int r = 0;
+	union kvmppc_one_reg val;
+	int size;
+
+	size = one_reg_size(reg->id);
+	if (size > sizeof(val))
+		return -EINVAL;
+
+	r = kvmppc_get_one_reg(vcpu, reg->id, &val);
+	if (r == -EINVAL) {
+		r = 0;
+		switch (reg->id) {
+#ifdef CONFIG_ALTIVEC
+		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
+			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
+				r = -ENXIO;
+				break;
+			}
+			vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0] = val.vval;
+			break;
+		case KVM_REG_PPC_VSCR:
+			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
+				r = -ENXIO;
+				break;
+			}
+			vcpu->arch.vr.vscr.u[3] = set_reg_val(reg->id, val);
+			break;
+		case KVM_REG_PPC_VRSAVE:
+			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
+				r = -ENXIO;
+				break;
+			}
+			vcpu->arch.vrsave = set_reg_val(reg->id, val);
+			break;
+#endif /* CONFIG_ALTIVEC */
+		default:
+			r = -EINVAL;
+			break;
+		}
+	}
+
+	if (r)
+		return r;
+
+	if (copy_to_user((char __user *)(unsigned long)reg->addr, &val, size))
+		r = -EFAULT;
+
+	return r;
+}
+
+int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
+{
+	int r;
+	union kvmppc_one_reg val;
+	int size;
+
+	size = one_reg_size(reg->id);
+	if (size > sizeof(val))
+		return -EINVAL;
+
+	if (copy_from_user(&val, (char __user *)(unsigned long)reg->addr, size))
+		return -EFAULT;
+
+	r = kvmppc_set_one_reg(vcpu, reg->id, &val);
+	if (r == -EINVAL) {
+		r = 0;
+		switch (reg->id) {
+#ifdef CONFIG_ALTIVEC
+		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
+			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
+				r = -ENXIO;
+				break;
+			}
+			val.vval = vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0];
+			break;
+		case KVM_REG_PPC_VSCR:
+			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
+				r = -ENXIO;
+				break;
+			}
+			val = get_reg_val(reg->id, vcpu->arch.vr.vscr.u[3]);
+			break;
+		case KVM_REG_PPC_VRSAVE:
+			val = get_reg_val(reg->id, vcpu->arch.vrsave);
+			break;
+#endif /* CONFIG_ALTIVEC */
+		default:
+			r = -EINVAL;
+			break;
+		}
+	}
+
+	return r;
+}
+
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
 	int r;
@@ -1343,9 +1414,4 @@ int kvm_arch_init(void *opaque)
 	return 0;
 }
 
-void kvm_arch_exit(void)
-{
-
-}
-
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ppc_instr);

arch/powerpc/platforms/Kconfig.cputype

@@ -303,9 +303,13 @@ config PPC_ICSWX_USE_SIGILL
 
 	  If in doubt, say N here.
 
+config SPE_POSSIBLE
+	def_bool y
+	depends on E200 || (E500 && !PPC_E500MC)
+
 config SPE
 	bool "SPE Support"
-	depends on E200 || (E500 && !PPC_E500MC)
+	depends on SPE_POSSIBLE
 	default y
 	---help---
 	  This option enables kernel support for the Signal Processing

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 */
@@ -187,6 +192,7 @@ struct kvm_vcpu_stat {
 	u32 exit_stop_request;
 	u32 exit_validity;
 	u32 exit_instruction;
+	u32 halt_wakeup;
 	u32 instruction_lctl;
 	u32 instruction_lctlg;
 	u32 instruction_stctl;
@@ -407,6 +413,15 @@ 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 +435,7 @@ 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 +447,6 @@ 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 +464,18 @@ 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);
 	}
 
 	oldpte = pte = *ptep;
@@ -1179,7 +1165,7 @@ 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;
@@ -1203,7 +1189,7 @@ 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;
@@ -1240,7 +1226,7 @@ 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;
@@ -1274,7 +1260,7 @@ 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;
@@ -1299,7 +1285,7 @@ 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);
@@ -1325,7 +1311,7 @@ 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 */
@@ -629,6 +626,7 @@ void kvm_s390_vcpu_wakeup(struct kvm_vcpu *vcpu)
 		 */
 		vcpu->preempted = true;
 		wake_up_interruptible(&vcpu->wq);
+		vcpu->stat.halt_wakeup++;
 	}
 }
 
@@ -661,12 +659,13 @@ 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 +684,16 @@ 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 +710,13 @@ 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 +993,6 @@ 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 +1004,6 @@ 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 +1243,7 @@ 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 +1353,6 @@ 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

@@ -50,6 +50,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "exit_instruction", VCPU_STAT(exit_instruction) },
 	{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
 	{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
+	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
 	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
 	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
 	{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
@@ -100,16 +101,12 @@ 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 +121,10 @@ 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 +394,22 @@ 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 +447,9 @@ 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 +460,7 @@ 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 +474,8 @@ 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 +525,12 @@ 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 +543,7 @@ 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;
@@ -546,15 +554,12 @@ 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_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	save_fp_ctl(&vcpu->arch.host_fpregs.fpc);
@@ -607,6 +612,14 @@ 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);
@@ -653,6 +666,9 @@ 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;
 }
 
@@ -1049,6 +1065,11 @@ 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);
@@ -1085,18 +1106,8 @@ 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,
@@ -1191,8 +1202,11 @@ 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)
@@ -1296,6 +1310,48 @@ 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;
@@ -1317,17 +1373,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		return -EINVAL;
 	}
 
-	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);
@@ -1357,10 +1403,7 @@ 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);
@@ -1489,7 +1532,7 @@ 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;
 }
@@ -1644,9 +1687,7 @@ 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:
@@ -1677,21 +1718,12 @@ 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,
@@ -1737,15 +1769,6 @@ 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;
@@ -1764,7 +1787,7 @@ static int __init kvm_s390_init(void)
 		return -ENOMEM;
 	}
 	memcpy(vfacilities, S390_lowcore.stfle_fac_list, 16);
-	vfacilities[0] &= 0xff82fff3f4fc2000UL;
+	vfacilities[0] &= 0xff82fffbf47c2000UL;
 	vfacilities[1] &= 0x005c000000000000UL;
 	return 0;
 }

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);
 }
 
@@ -994,13 +943,13 @@ retry:
 	}
 	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;
+		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 |
@@ -1038,8 +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;
 }
@@ -1053,8 +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)
 {
 }
 
@@ -1074,14 +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;
@@ -1123,10 +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);
@@ -1158,7 +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;
@@ -1167,7 +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;
@@ -1303,7 +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;
 
@@ -1318,7 +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/cpufeature.h

@@ -202,6 +202,7 @@
 #define X86_FEATURE_DECODEASSISTS ( 8*32+12) /* AMD Decode Assists support */
 #define X86_FEATURE_PAUSEFILTER ( 8*32+13) /* AMD filtered pause intercept */
 #define X86_FEATURE_PFTHRESHOLD ( 8*32+14) /* AMD pause filter threshold */
+#define X86_FEATURE_VMMCALL     ( 8*32+15) /* Prefer vmmcall to vmcall */
 
 
 /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */

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;
 
@@ -576,11 +574,10 @@ struct kvm_arch {
 	struct kvm_apic_map *apic_map;
 
 	unsigned int tss_addr;
-	struct page *apic_access_page;
+	bool apic_access_page_done;
 
 	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 */
@@ -710,7 +707,6 @@ struct kvm_x86_ops {
 	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
 	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
 	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
-	void (*fpu_activate)(struct kvm_vcpu *vcpu);
 	void (*fpu_deactivate)(struct kvm_vcpu *vcpu);
 
 	void (*tlb_flush)(struct kvm_vcpu *vcpu);
@@ -740,6 +736,7 @@ struct kvm_x86_ops {
 	void (*hwapic_isr_update)(struct kvm *kvm, int isr);
 	void (*load_eoi_exitmap)(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap);
 	void (*set_virtual_x2apic_mode)(struct kvm_vcpu *vcpu, bool set);
+	void (*set_apic_access_page_addr)(struct kvm_vcpu *vcpu, hpa_t hpa);
 	void (*deliver_posted_interrupt)(struct kvm_vcpu *vcpu, int vector);
 	void (*sync_pir_to_irr)(struct kvm_vcpu *vcpu);
 	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
@@ -772,6 +769,8 @@ struct kvm_x86_ops {
 	bool (*mpx_supported)(void);
 
 	int (*check_nested_events)(struct kvm_vcpu *vcpu, bool external_intr);
+
+	void (*sched_in)(struct kvm_vcpu *kvm, int cpu);
 };
 
 struct kvm_arch_async_pf {
@@ -895,7 +894,6 @@ 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,
@@ -917,7 +915,6 @@ void kvm_inject_nmi(struct kvm_vcpu *vcpu);
 
 int fx_init(struct kvm_vcpu *vcpu);
 
-void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
 void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 		       const u8 *new, int bytes);
 int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn);
@@ -926,7 +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,
@@ -946,7 +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;
 }
@@ -1037,7 +1036,7 @@ asmlinkage void kvm_spurious_fault(void);
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end);
-int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
@@ -1046,6 +1045,9 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
 int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
 void kvm_vcpu_reset(struct kvm_vcpu *vcpu);
+void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu);
+void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
+					   unsigned long address);
 
 void kvm_define_shared_msr(unsigned index, u32 msr);
 void kvm_set_shared_msr(unsigned index, u64 val, u64 mask);

arch/x86/include/asm/kvm_para.h

@@ -2,6 +2,7 @@
 #define _ASM_X86_KVM_PARA_H
 
 #include <asm/processor.h>
+#include <asm/alternative.h>
 #include <uapi/asm/kvm_para.h>
 
 extern void kvmclock_init(void);
@@ -16,10 +17,15 @@ static inline bool kvm_check_and_clear_guest_paused(void)
 }
 #endif /* CONFIG_KVM_GUEST */
 
-/* This instruction is vmcall.  On non-VT architectures, it will generate a
- * trap that we will then rewrite to the appropriate instruction.
+#ifdef CONFIG_DEBUG_RODATA
+#define KVM_HYPERCALL \
+        ALTERNATIVE(".byte 0x0f,0x01,0xc1", ".byte 0x0f,0x01,0xd9", X86_FEATURE_VMMCALL)
+#else
+/* On AMD processors, vmcall will generate a trap that we will
+ * then rewrite to the appropriate instruction.
  */
 #define KVM_HYPERCALL ".byte 0x0f,0x01,0xc1"
+#endif
 
 /* For KVM hypercalls, a three-byte sequence of either the vmcall or the vmmcall
  * instruction.  The hypervisor may replace it with something else but only the

arch/x86/kernel/cpu/amd.c

@@ -525,6 +525,13 @@ static void early_init_amd(struct cpuinfo_x86 *c)
 	}
 #endif
 
+	/*
+	 * This is only needed to tell the kernel whether to use VMCALL
+	 * and VMMCALL.  VMMCALL is never executed except under virt, so
+	 * we can set it unconditionally.
+	 */
+	set_cpu_cap(c, X86_FEATURE_VMMCALL);
+
 	/* F16h erratum 793, CVE-2013-6885 */
 	if (c->x86 == 0x16 && c->x86_model <= 0xf)
 		msr_set_bit(MSR_AMD64_LS_CFG, 15);

arch/x86/kvm/cpuid.c

@@ -53,14 +53,14 @@ u64 kvm_supported_xcr0(void)
 	return xcr0;
 }
 
-void kvm_update_cpuid(struct kvm_vcpu *vcpu)
+int kvm_update_cpuid(struct kvm_vcpu *vcpu)
 {
 	struct kvm_cpuid_entry2 *best;
 	struct kvm_lapic *apic = vcpu->arch.apic;
 
 	best = kvm_find_cpuid_entry(vcpu, 1, 0);
 	if (!best)
-		return;
+		return 0;
 
 	/* Update OSXSAVE bit */
 	if (cpu_has_xsave && best->function == 0x1) {
@@ -88,7 +88,17 @@ void kvm_update_cpuid(struct kvm_vcpu *vcpu)
 			xstate_required_size(vcpu->arch.xcr0);
 	}
 
+	/*
+	 * The existing code assumes virtual address is 48-bit in the canonical
+	 * address checks; exit if it is ever changed.
+	 */
+	best = kvm_find_cpuid_entry(vcpu, 0x80000008, 0);
+	if (best && ((best->eax & 0xff00) >> 8) != 48 &&
+		((best->eax & 0xff00) >> 8) != 0)
+		return -EINVAL;
+
 	kvm_pmu_cpuid_update(vcpu);
+	return 0;
 }
 
 static int is_efer_nx(void)
@@ -112,8 +122,8 @@ static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
 			break;
 		}
 	}
-	if (entry && (entry->edx & (1 << 20)) && !is_efer_nx()) {
-		entry->edx &= ~(1 << 20);
+	if (entry && (entry->edx & bit(X86_FEATURE_NX)) && !is_efer_nx()) {
+		entry->edx &= ~bit(X86_FEATURE_NX);
 		printk(KERN_INFO "kvm: guest NX capability removed\n");
 	}
 }
@@ -151,10 +161,9 @@ int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
 	}
 	vcpu->arch.cpuid_nent = cpuid->nent;
 	cpuid_fix_nx_cap(vcpu);
-	r = 0;
 	kvm_apic_set_version(vcpu);
 	kvm_x86_ops->cpuid_update(vcpu);
-	kvm_update_cpuid(vcpu);
+	r = kvm_update_cpuid(vcpu);
 
 out_free:
 	vfree(cpuid_entries);
@@ -178,9 +187,7 @@ int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
 	vcpu->arch.cpuid_nent = cpuid->nent;
 	kvm_apic_set_version(vcpu);
 	kvm_x86_ops->cpuid_update(vcpu);
-	kvm_update_cpuid(vcpu);
-	return 0;
-
+	r = kvm_update_cpuid(vcpu);
 out:
 	return r;
 }
@@ -767,6 +774,12 @@ void kvm_cpuid(struct kvm_vcpu *vcpu, u32 *eax, u32 *ebx, u32 *ecx, u32 *edx)
 	if (!best)
 		best = check_cpuid_limit(vcpu, function, index);
 
+	/*
+	 * Perfmon not yet supported for L2 guest.
+	 */
+	if (is_guest_mode(vcpu) && function == 0xa)
+		best = NULL;
+
 	if (best) {
 		*eax = best->eax;
 		*ebx = best->ebx;

arch/x86/kvm/cpuid.h

@@ -3,7 +3,7 @@
 
 #include "x86.h"
 
-void kvm_update_cpuid(struct kvm_vcpu *vcpu);
+int kvm_update_cpuid(struct kvm_vcpu *vcpu);
 struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu,
 					      u32 function, u32 index);
 int kvm_dev_ioctl_get_cpuid(struct kvm_cpuid2 *cpuid,
@@ -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

@@ -527,6 +527,7 @@ static unsigned long seg_base(struct x86_emulate_ctxt *ctxt, int seg)
 static int emulate_exception(struct x86_emulate_ctxt *ctxt, int vec,
 			     u32 error, bool valid)
 {
+	WARN_ON(vec > 0x1f);
 	ctxt->exception.vector = vec;
 	ctxt->exception.error_code = error;
 	ctxt->exception.error_code_valid = valid;
@@ -1468,7 +1469,7 @@ static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 		return ret;
 
 	err_code = selector & 0xfffc;
-	err_vec = GP_VECTOR;
+	err_vec = in_task_switch ? TS_VECTOR : GP_VECTOR;
 
 	/* can't load system descriptor into segment selector */
 	if (seg <= VCPU_SREG_GS && !seg_desc.s)
@@ -1503,6 +1504,15 @@ static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 			if (rpl > cpl || dpl != cpl)
 				goto exception;
 		}
+		/* in long-mode d/b must be clear if l is set */
+		if (seg_desc.d && seg_desc.l) {
+			u64 efer = 0;
+
+			ctxt->ops->get_msr(ctxt, MSR_EFER, &efer);
+			if (efer & EFER_LMA)
+				goto exception;
+		}
+
 		/* CS(RPL) <- CPL */
 		selector = (selector & 0xfffc) | cpl;
 		break;
@@ -1549,8 +1559,7 @@ load:
 	ctxt->ops->set_segment(ctxt, selector, &seg_desc, base3, seg);
 	return X86EMUL_CONTINUE;
 exception:
-	emulate_exception(ctxt, err_vec, err_code, true);
-	return X86EMUL_PROPAGATE_FAULT;
+	return emulate_exception(ctxt, err_vec, err_code, true);
 }
 
 static int load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
@@ -2723,8 +2732,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
 	if (!next_tss_desc.p ||
 	    ((desc_limit < 0x67 && (next_tss_desc.type & 8)) ||
 	     desc_limit < 0x2b)) {
-		emulate_ts(ctxt, tss_selector & 0xfffc);
-		return X86EMUL_PROPAGATE_FAULT;
+		return emulate_ts(ctxt, tss_selector & 0xfffc);
 	}
 
 	if (reason == TASK_SWITCH_IRET || reason == TASK_SWITCH_JMP) {
@@ -3016,7 +3024,7 @@ static int em_movbe(struct x86_emulate_ctxt *ctxt)
 		ctxt->dst.val = swab64(ctxt->src.val);
 		break;
 	default:
-		return X86EMUL_PROPAGATE_FAULT;
+		BUG();
 	}
 	return X86EMUL_CONTINUE;
 }
@@ -3140,12 +3148,8 @@ 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;
 
@@ -3563,6 +3567,12 @@ 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),
@@ -3656,7 +3666,7 @@ 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,
@@ -3687,14 +3697,18 @@ 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),
 }, {
@@ -3901,7 +3915,7 @@ 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),
@@ -3965,7 +3979,8 @@ 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
 };
@@ -4829,8 +4844,10 @@ writeback:
 	ctxt->eip = ctxt->_eip;
 
 done:
-	if (rc == X86EMUL_PROPAGATE_FAULT)
+	if (rc == X86EMUL_PROPAGATE_FAULT) {
+		WARN_ON(ctxt->exception.vector > 0x1f);
 		ctxt->have_exception = true;
+	}
 	if (rc == X86EMUL_INTERCEPTED)
 		return EMULATION_INTERCEPTED;
 

arch/x86/kvm/lapic.c

@@ -112,17 +112,6 @@ static inline int __apic_test_and_clear_vector(int vec, void *bitmap)
 struct static_key_deferred apic_hw_disabled __read_mostly;
 struct static_key_deferred apic_sw_disabled __read_mostly;
 
-static inline void apic_set_spiv(struct kvm_lapic *apic, u32 val)
-{
-	if ((kvm_apic_get_reg(apic, APIC_SPIV) ^ val) & APIC_SPIV_APIC_ENABLED) {
-		if (val & APIC_SPIV_APIC_ENABLED)
-			static_key_slow_dec_deferred(&apic_sw_disabled);
-		else
-			static_key_slow_inc(&apic_sw_disabled.key);
-	}
-	apic_set_reg(apic, APIC_SPIV, val);
-}
-
 static inline int apic_enabled(struct kvm_lapic *apic)
 {
 	return kvm_apic_sw_enabled(apic) &&	kvm_apic_hw_enabled(apic);
@@ -210,6 +199,20 @@ out:
 	kvm_vcpu_request_scan_ioapic(kvm);
 }
 
+static inline void apic_set_spiv(struct kvm_lapic *apic, u32 val)
+{
+	u32 prev = kvm_apic_get_reg(apic, APIC_SPIV);
+
+	apic_set_reg(apic, APIC_SPIV, val);
+	if ((prev ^ val) & APIC_SPIV_APIC_ENABLED) {
+		if (val & APIC_SPIV_APIC_ENABLED) {
+			static_key_slow_dec_deferred(&apic_sw_disabled);
+			recalculate_apic_map(apic->vcpu->kvm);
+		} else
+			static_key_slow_inc(&apic_sw_disabled.key);
+	}
+}
+
 static inline void kvm_apic_set_id(struct kvm_lapic *apic, u8 id)
 {
 	apic_set_reg(apic, APIC_ID, id << 24);
@@ -706,6 +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++;
@@ -727,8 +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:
@@ -1352,6 +1355,9 @@ void kvm_set_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu, u64 data)
 		return;
 
 	hrtimer_cancel(&apic->lapic_timer.timer);
+	/* Inject here so clearing tscdeadline won't override new value */
+	if (apic_has_pending_timer(vcpu))
+		kvm_inject_apic_timer_irqs(vcpu);
 	apic->lapic_timer.tscdeadline = data;
 	start_apic_timer(apic);
 }
@@ -1639,6 +1645,8 @@ void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu)
 
 	if (atomic_read(&apic->lapic_timer.pending) > 0) {
 		kvm_apic_local_deliver(apic, APIC_LVTT);
+		if (apic_lvtt_tscdeadline(apic))
+			apic->lapic_timer.tscdeadline = 0;
 		atomic_set(&apic->lapic_timer.pending, 0);
 	}
 }

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)
 {
@@ -1180,7 +1174,7 @@ static void drop_large_spte(struct kvm_vcpu *vcpu, u64 *sptep)
  * Write-protect on the specified @sptep, @pt_protect indicates whether
  * spte write-protection is caused by protecting shadow page table.
  *
- * Note: write protection is difference between drity logging and spte
+ * Note: write protection is difference between dirty logging and spte
  * protection:
  * - for dirty logging, the spte can be set to writable at anytime if
  *   its dirty bitmap is properly set.
@@ -1268,7 +1262,8 @@ static bool rmap_write_protect(struct kvm *kvm, u64 gfn)
 }
 
 static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			   struct kvm_memory_slot *slot, unsigned long data)
+			   struct kvm_memory_slot *slot, gfn_t gfn, int level,
+			   unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator iter;
@@ -1276,7 +1271,8 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
 
 	while ((sptep = rmap_get_first(*rmapp, &iter))) {
 		BUG_ON(!(*sptep & PT_PRESENT_MASK));
-		rmap_printk("kvm_rmap_unmap_hva: spte %p %llx\n", sptep, *sptep);
+		rmap_printk("kvm_rmap_unmap_hva: spte %p %llx gfn %llx (%d)\n",
+			     sptep, *sptep, gfn, level);
 
 		drop_spte(kvm, sptep);
 		need_tlb_flush = 1;
@@ -1286,7 +1282,8 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
 }
 
 static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			     struct kvm_memory_slot *slot, unsigned long data)
+			     struct kvm_memory_slot *slot, gfn_t gfn, int level,
+			     unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator iter;
@@ -1300,7 +1297,8 @@ static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
 
 	for (sptep = rmap_get_first(*rmapp, &iter); sptep;) {
 		BUG_ON(!is_shadow_present_pte(*sptep));
-		rmap_printk("kvm_set_pte_rmapp: spte %p %llx\n", sptep, *sptep);
+		rmap_printk("kvm_set_pte_rmapp: spte %p %llx gfn %llx (%d)\n",
+			     sptep, *sptep, gfn, level);
 
 		need_flush = 1;
 
@@ -1334,6 +1332,8 @@ static int kvm_handle_hva_range(struct kvm *kvm,
 				int (*handler)(struct kvm *kvm,
 					       unsigned long *rmapp,
 					       struct kvm_memory_slot *slot,
+					       gfn_t gfn,
+					       int level,
 					       unsigned long data))
 {
 	int j;
@@ -1363,6 +1363,7 @@ static int kvm_handle_hva_range(struct kvm *kvm,
 		     j < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++j) {
 			unsigned long idx, idx_end;
 			unsigned long *rmapp;
+			gfn_t gfn = gfn_start;
 
 			/*
 			 * {idx(page_j) | page_j intersects with
@@ -1373,8 +1374,10 @@ static int kvm_handle_hva_range(struct kvm *kvm,
 
 			rmapp = __gfn_to_rmap(gfn_start, j, memslot);
 
-			for (; idx <= idx_end; ++idx)
-				ret |= handler(kvm, rmapp++, memslot, data);
+			for (; idx <= idx_end;
+			       ++idx, gfn += (1UL << KVM_HPAGE_GFN_SHIFT(j)))
+				ret |= handler(kvm, rmapp++, memslot,
+					       gfn, j, data);
 		}
 	}
 
@@ -1385,6 +1388,7 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 			  unsigned long data,
 			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
 					 struct kvm_memory_slot *slot,
+					 gfn_t gfn, int level,
 					 unsigned long data))
 {
 	return kvm_handle_hva_range(kvm, hva, hva + 1, data, handler);
@@ -1406,24 +1410,14 @@ void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
 }
 
 static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			 struct kvm_memory_slot *slot, unsigned long data)
+			 struct kvm_memory_slot *slot, gfn_t gfn, int level,
+			 unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator uninitialized_var(iter);
 	int young = 0;
 
-	/*
-	 * In case of absence of EPT Access and Dirty Bits supports,
-	 * emulate the accessed bit for EPT, by checking if this page has
-	 * an EPT mapping, and clearing it if it does. On the next access,
-	 * a new EPT mapping will be established.
-	 * This has some overhead, but not as much as the cost of swapping
-	 * out actively used pages or breaking up actively used hugepages.
-	 */
-	if (!shadow_accessed_mask) {
-		young = kvm_unmap_rmapp(kvm, rmapp, slot, data);
-		goto out;
-	}
+	BUG_ON(!shadow_accessed_mask);
 
 	for (sptep = rmap_get_first(*rmapp, &iter); sptep;
 	     sptep = rmap_get_next(&iter)) {
@@ -1435,14 +1429,13 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 				 (unsigned long *)sptep);
 		}
 	}
-out:
-	/* @data has hva passed to kvm_age_hva(). */
-	trace_kvm_age_page(data, slot, young);
+	trace_kvm_age_page(gfn, level, slot, young);
 	return young;
 }
 
 static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			      struct kvm_memory_slot *slot, unsigned long data)
+			      struct kvm_memory_slot *slot, gfn_t gfn,
+			      int level, unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator iter;
@@ -1480,13 +1473,33 @@ static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
 
 	rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level);
 
-	kvm_unmap_rmapp(vcpu->kvm, rmapp, NULL, 0);
+	kvm_unmap_rmapp(vcpu->kvm, rmapp, NULL, gfn, sp->role.level, 0);
 	kvm_flush_remote_tlbs(vcpu->kvm);
 }
 
-int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end)
 {
-	return kvm_handle_hva(kvm, hva, hva, kvm_age_rmapp);
+	/*
+	 * In case of absence of EPT Access and Dirty Bits supports,
+	 * emulate the accessed bit for EPT, by checking if this page has
+	 * an EPT mapping, and clearing it if it does. On the next access,
+	 * a new EPT mapping will be established.
+	 * This has some overhead, but not as much as the cost of swapping
+	 * out actively used pages or breaking up actively used hugepages.
+	 */
+	if (!shadow_accessed_mask) {
+		/*
+		 * We are holding the kvm->mmu_lock, and we are blowing up
+		 * shadow PTEs. MMU notifier consumers need to be kept at bay.
+		 * This is correct as long as we don't decouple the mmu_lock
+		 * protected regions (like invalidate_range_start|end does).
+		 */
+		kvm->mmu_notifier_seq++;
+		return kvm_handle_hva_range(kvm, start, end, 0,
+					    kvm_unmap_rmapp);
+	}
+
+	return kvm_handle_hva_range(kvm, start, end, 0, kvm_age_rmapp);
 }
 
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
@@ -1749,7 +1762,7 @@ static int __kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 		return 1;
 	}
 
-	kvm_mmu_flush_tlb(vcpu);
+	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 	return 0;
 }
 
@@ -1802,7 +1815,7 @@ static void kvm_sync_pages(struct kvm_vcpu *vcpu,  gfn_t gfn)
 
 	kvm_mmu_commit_zap_page(vcpu->kvm, &invalid_list);
 	if (flush)
-		kvm_mmu_flush_tlb(vcpu);
+		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 }
 
 struct mmu_page_path {
@@ -2536,7 +2549,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	      true, host_writable)) {
 		if (write_fault)
 			*emulate = 1;
-		kvm_mmu_flush_tlb(vcpu);
+		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 	}
 
 	if (unlikely(is_mmio_spte(*sptep) && emulate))
@@ -3163,7 +3176,7 @@ 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 +3219,7 @@ 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)
@@ -3450,13 +3463,6 @@ static void nonpaging_init_context(struct kvm_vcpu *vcpu,
 	context->nx = false;
 }
 
-void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
-{
-	++vcpu->stat.tlb_flush;
-	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
-}
-EXPORT_SYMBOL_GPL(kvm_mmu_flush_tlb);
-
 void kvm_mmu_new_cr3(struct kvm_vcpu *vcpu)
 {
 	mmu_free_roots(vcpu);
@@ -3518,6 +3524,7 @@ 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 +3532,14 @@ 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 +3574,9 @@ 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 |
@@ -3962,7 +3977,7 @@ static void mmu_pte_write_flush_tlb(struct kvm_vcpu *vcpu, bool zap_page,
 	if (remote_flush)
 		kvm_flush_remote_tlbs(vcpu->kvm);
 	else if (local_flush)
-		kvm_mmu_flush_tlb(vcpu);
+		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 }
 
 static u64 mmu_pte_write_fetch_gpte(struct kvm_vcpu *vcpu, gpa_t *gpa,
@@ -4223,7 +4238,7 @@ EXPORT_SYMBOL_GPL(kvm_mmu_page_fault);
 void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva)
 {
 	vcpu->arch.mmu.invlpg(vcpu, gva);
-	kvm_mmu_flush_tlb(vcpu);
+	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 	++vcpu->stat.invlpg;
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_invlpg);
@@ -4433,7 +4448,7 @@ 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

@@ -298,8 +298,7 @@ retry_walk:
 	}
 #endif
 	walker->max_level = walker->level;
-	ASSERT((!is_long_mode(vcpu) && is_pae(vcpu)) ||
-	       (mmu->get_cr3(vcpu) & CR3_NONPAE_RESERVED_BITS) == 0);
+	ASSERT(!is_long_mode(vcpu) && is_pae(vcpu));
 
 	accessed_dirty = PT_GUEST_ACCESSED_MASK;
 	pt_access = pte_access = ACC_ALL;
@@ -321,9 +320,22 @@ 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 +376,7 @@ 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/pmu.c

@@ -15,6 +15,7 @@
 #include <linux/types.h>
 #include <linux/kvm_host.h>
 #include <linux/perf_event.h>
+#include <asm/perf_event.h>
 #include "x86.h"
 #include "cpuid.h"
 #include "lapic.h"
@@ -463,7 +464,8 @@ void kvm_pmu_cpuid_update(struct kvm_vcpu *vcpu)
 {
 	struct kvm_pmu *pmu = &vcpu->arch.pmu;
 	struct kvm_cpuid_entry2 *entry;
-	unsigned bitmap_len;
+	union cpuid10_eax eax;
+	union cpuid10_edx edx;
 
 	pmu->nr_arch_gp_counters = 0;
 	pmu->nr_arch_fixed_counters = 0;
@@ -475,25 +477,27 @@ void kvm_pmu_cpuid_update(struct kvm_vcpu *vcpu)
 	entry = kvm_find_cpuid_entry(vcpu, 0xa, 0);
 	if (!entry)
 		return;
+	eax.full = entry->eax;
+	edx.full = entry->edx;
 
-	pmu->version = entry->eax & 0xff;
+	pmu->version = eax.split.version_id;
 	if (!pmu->version)
 		return;
 
-	pmu->nr_arch_gp_counters = min((int)(entry->eax >> 8) & 0xff,
-			INTEL_PMC_MAX_GENERIC);
-	pmu->counter_bitmask[KVM_PMC_GP] =
-		((u64)1 << ((entry->eax >> 16) & 0xff)) - 1;
-	bitmap_len = (entry->eax >> 24) & 0xff;
-	pmu->available_event_types = ~entry->ebx & ((1ull << bitmap_len) - 1);
+	pmu->nr_arch_gp_counters = min_t(int, eax.split.num_counters,
+					INTEL_PMC_MAX_GENERIC);
+	pmu->counter_bitmask[KVM_PMC_GP] = ((u64)1 << eax.split.bit_width) - 1;
+	pmu->available_event_types = ~entry->ebx &
+					((1ull << eax.split.mask_length) - 1);
 
 	if (pmu->version == 1) {
 		pmu->nr_arch_fixed_counters = 0;
 	} else {
-		pmu->nr_arch_fixed_counters = min((int)(entry->edx & 0x1f),
+		pmu->nr_arch_fixed_counters =
+			min_t(int, edx.split.num_counters_fixed,
 				INTEL_PMC_MAX_FIXED);
 		pmu->counter_bitmask[KVM_PMC_FIXED] =
-			((u64)1 << ((entry->edx >> 5) & 0xff)) - 1;
+			((u64)1 << edx.split.bit_width_fixed) - 1;
 	}
 
 	pmu->global_ctrl = ((1 << pmu->nr_arch_gp_counters) - 1) |

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 +
@@ -4305,6 +4322,10 @@ static void svm_handle_external_intr(struct kvm_vcpu *vcpu)
 	local_irq_enable();
 }
 
+static void svm_sched_in(struct kvm_vcpu *vcpu, int cpu)
+{
+}
+
 static struct kvm_x86_ops svm_x86_ops = {
 	.cpu_has_kvm_support = has_svm,
 	.disabled_by_bios = is_disabled,
@@ -4349,7 +4370,6 @@ static struct kvm_x86_ops svm_x86_ops = {
 	.cache_reg = svm_cache_reg,
 	.get_rflags = svm_get_rflags,
 	.set_rflags = svm_set_rflags,
-	.fpu_activate = svm_fpu_activate,
 	.fpu_deactivate = svm_fpu_deactivate,
 
 	.tlb_flush = svm_flush_tlb,
@@ -4406,6 +4426,8 @@ static struct kvm_x86_ops svm_x86_ops = {
 
 	.check_intercept = svm_check_intercept,
 	.handle_external_intr = svm_handle_external_intr,
+
+	.sched_in = svm_sched_in,
 };
 
 static int __init svm_init(void)

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,
@@ -850,6 +847,36 @@ TRACE_EVENT(kvm_track_tsc,
 
 #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),
+
+	TP_STRUCT__entry(
+		__field(                bool,      grow         )
+		__field(        unsigned int,   vcpu_id         )
+		__field(                 int,       new         )
+		__field(                 int,       old         )
+	),
+
+	TP_fast_assign(
+		__entry->grow           = grow;
+		__entry->vcpu_id        = vcpu_id;
+		__entry->new            = new;
+		__entry->old            = old;
+	),
+
+	TP_printk("vcpu %u: ple_window %d (%s %d)",
+	          __entry->vcpu_id,
+	          __entry->new,
+	          __entry->grow ? "grow" : "shrink",
+	          __entry->old)
+);
+
+#define trace_kvm_ple_window_grow(vcpu_id, new, old) \
+	trace_kvm_ple_window(true, vcpu_id, new, old)
+#define trace_kvm_ple_window_shrink(vcpu_id, new, old) \
+	trace_kvm_ple_window(false, vcpu_id, new, old)
+
 #endif /* _TRACE_KVM_H */
 
 #undef TRACE_INCLUDE_PATH

arch/x86/kvm/vmx.c

@@ -125,14 +125,32 @@ module_param(nested, bool, S_IRUGO);
  * Time is measured based on a counter that runs at the same rate as the TSC,
  * refer SDM volume 3b section 21.6.13 & 22.1.3.
  */
-#define KVM_VMX_DEFAULT_PLE_GAP    128
-#define KVM_VMX_DEFAULT_PLE_WINDOW 4096
+#define KVM_VMX_DEFAULT_PLE_GAP           128
+#define KVM_VMX_DEFAULT_PLE_WINDOW        4096
+#define KVM_VMX_DEFAULT_PLE_WINDOW_GROW   2
+#define KVM_VMX_DEFAULT_PLE_WINDOW_SHRINK 0
+#define KVM_VMX_DEFAULT_PLE_WINDOW_MAX    \
+		INT_MAX / KVM_VMX_DEFAULT_PLE_WINDOW_GROW
+
 static int ple_gap = KVM_VMX_DEFAULT_PLE_GAP;
 module_param(ple_gap, int, S_IRUGO);
 
 static int ple_window = KVM_VMX_DEFAULT_PLE_WINDOW;
 module_param(ple_window, int, S_IRUGO);
 
+/* Default doubles per-vcpu window every exit. */
+static int ple_window_grow = KVM_VMX_DEFAULT_PLE_WINDOW_GROW;
+module_param(ple_window_grow, int, S_IRUGO);
+
+/* Default resets per-vcpu window every exit to ple_window. */
+static int ple_window_shrink = KVM_VMX_DEFAULT_PLE_WINDOW_SHRINK;
+module_param(ple_window_shrink, int, S_IRUGO);
+
+/* Default is to compute the maximum so we can never overflow. */
+static int ple_window_actual_max = KVM_VMX_DEFAULT_PLE_WINDOW_MAX;
+static int ple_window_max        = KVM_VMX_DEFAULT_PLE_WINDOW_MAX;
+module_param(ple_window_max, int, S_IRUGO);
+
 extern const ulong vmx_return;
 
 #define NR_AUTOLOAD_MSRS 8
@@ -379,6 +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;
@@ -484,6 +503,10 @@ struct vcpu_vmx {
 
 	/* Support for a guest hypervisor (nested VMX) */
 	struct nested_vmx nested;
+
+	/* Dynamic PLE window. */
+	int ple_window;
+	bool ple_window_dirty;
 };
 
 enum segment_cache_field {
@@ -533,6 +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,
@@ -743,6 +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);
@@ -2135,7 +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;
 
@@ -2330,7 +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
@@ -2601,6 +2626,8 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		break;
 	case MSR_IA32_CR_PAT:
 		if (vmcs_config.vmentry_ctrl & VM_ENTRY_LOAD_IA32_PAT) {
+			if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data))
+				return 1;
 			vmcs_write64(GUEST_IA32_PAT, data);
 			vcpu->arch.pat = data;
 			break;
@@ -2704,7 +2731,7 @@ 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));
@@ -2768,7 +2795,7 @@ 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();
@@ -3107,9 +3134,17 @@ static __init int hardware_setup(void)
 	if (!cpu_has_vmx_unrestricted_guest())
 		enable_unrestricted_guest = 0;
 
-	if (!cpu_has_vmx_flexpriority())
+	if (!cpu_has_vmx_flexpriority()) {
 		flexpriority_enabled = 0;
 
+		/*
+		 * set_apic_access_page_addr() is used to reload apic access
+		 * page upon invalidation.  No need to do anything if the
+		 * processor does not have the APIC_ACCESS_ADDR VMCS field.
+		 */
+		kvm_x86_ops->set_apic_access_page_addr = NULL;
+	}
+
 	if (!cpu_has_vmx_tpr_shadow())
 		kvm_x86_ops->update_cr8_intercept = NULL;
 
@@ -3905,7 +3940,7 @@ static int init_rmode_tss(struct kvm *kvm)
 {
 	gfn_t fn;
 	u16 data = 0;
-	int r, idx, ret = 0;
+	int idx, r;
 
 	idx = srcu_read_lock(&kvm->srcu);
 	fn = kvm->arch.tss_addr >> PAGE_SHIFT;
@@ -3927,32 +3962,32 @@ static int init_rmode_tss(struct kvm *kvm)
 	r = kvm_write_guest_page(kvm, fn, &data,
 				 RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1,
 				 sizeof(u8));
-	if (r < 0)
-		goto out;
-
-	ret = 1;
 out:
 	srcu_read_unlock(&kvm->srcu, idx);
-	return ret;
+	return r;
 }
 
 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)
@@ -3967,10 +4002,13 @@ 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)
@@ -3995,23 +4033,28 @@ static int alloc_apic_access_page(struct kvm *kvm)
 	int r = 0;
 
 	mutex_lock(&kvm->slots_lock);
-	if (kvm->arch.apic_access_page)
+	if (kvm->arch.apic_access_page_done)
 		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;
 	}
 
-	kvm->arch.apic_access_page = page;
+	/*
+	 * Do not pin the page in memory, so that memory hot-unplug
+	 * is able to migrate it.
+	 */
+	put_page(page);
+	kvm->arch.apic_access_page_done = true;
 out:
 	mutex_unlock(&kvm->slots_lock);
 	return r;
@@ -4019,31 +4062,20 @@ 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;
 }
 
@@ -4402,7 +4434,8 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
 
 	if (ple_gap) {
 		vmcs_write32(PLE_GAP, ple_gap);
-		vmcs_write32(PLE_WINDOW, ple_window);
+		vmx->ple_window = ple_window;
+		vmx->ple_window_dirty = true;
 	}
 
 	vmcs_write32(PAGE_FAULT_ERROR_CODE_MASK, 0);
@@ -4477,7 +4510,7 @@ 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;
@@ -4537,9 +4570,7 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu)
 		vmcs_write32(TPR_THRESHOLD, 0);
 	}
 
-	if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
-		vmcs_write64(APIC_ACCESS_ADDR,
-			     page_to_phys(vmx->vcpu.kvm->arch.apic_access_page));
+	kvm_vcpu_reload_apic_access_page(vcpu);
 
 	if (vmx_vm_has_apicv(vcpu->kvm))
 		memset(&vmx->pi_desc, 0, sizeof(struct pi_desc));
@@ -4729,10 +4760,7 @@ static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
 	if (ret)
 		return ret;
 	kvm->arch.tss_addr = addr;
-	if (!init_rmode_tss(kvm))
-		return  -ENOMEM;
-
-	return 0;
+	return init_rmode_tss(kvm);
 }
 
 static bool rmode_exception(struct kvm_vcpu *vcpu, int vec)
@@ -5521,17 +5549,18 @@ static u64 ept_rsvd_mask(u64 spte, int level)
 	for (i = 51; i > boot_cpu_data.x86_phys_bits; i--)
 		mask |= (1ULL << i);
 
-	if (level > 2)
+	if (level == 4)
 		/* bits 7:3 reserved */
 		mask |= 0xf8;
-	else if (level == 2) {
-		if (spte & (1ULL << 7))
-			/* 2MB ref, bits 20:12 reserved */
-			mask |= 0x1ff000;
-		else
-			/* bits 6:3 reserved */
-			mask |= 0x78;
-	}
+	else if (spte & (1ULL << 7))
+		/*
+		 * 1GB/2MB page, bits 29:12 or 20:12 reserved respectively,
+		 * level == 1 if the hypervisor is using the ignored bit 7.
+		 */
+		mask |= (PAGE_SIZE << ((level - 1) * 9)) - PAGE_SIZE;
+	else if (level > 1)
+		/* bits 6:3 reserved */
+		mask |= 0x78;
 
 	return mask;
 }
@@ -5561,7 +5590,8 @@ static void ept_misconfig_inspect_spte(struct kvm_vcpu *vcpu, u64 spte,
 			WARN_ON(1);
 		}
 
-		if (level == 1 || (level == 2 && (spte & (1ULL << 7)))) {
+		/* bits 5:3 are _not_ reserved for large page or leaf page */
+		if ((rsvd_bits & 0x38) == 0) {
 			u64 ept_mem_type = (spte & 0x38) >> 3;
 
 			if (ept_mem_type == 2 || ept_mem_type == 3 ||
@@ -5676,12 +5706,85 @@ out:
 	return ret;
 }
 
+static int __grow_ple_window(int val)
+{
+	if (ple_window_grow < 1)
+		return ple_window;
+
+	val = min(val, ple_window_actual_max);
+
+	if (ple_window_grow < ple_window)
+		val *= ple_window_grow;
+	else
+		val += ple_window_grow;
+
+	return val;
+}
+
+static int __shrink_ple_window(int val, int modifier, int minimum)
+{
+	if (modifier < 1)
+		return ple_window;
+
+	if (modifier < ple_window)
+		val /= modifier;
+	else
+		val -= modifier;
+
+	return max(val, minimum);
+}
+
+static void grow_ple_window(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	int old = vmx->ple_window;
+
+	vmx->ple_window = __grow_ple_window(old);
+
+	if (vmx->ple_window != old)
+		vmx->ple_window_dirty = true;
+
+	trace_kvm_ple_window_grow(vcpu->vcpu_id, vmx->ple_window, old);
+}
+
+static void shrink_ple_window(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+	int old = vmx->ple_window;
+
+	vmx->ple_window = __shrink_ple_window(old,
+	                                      ple_window_shrink, ple_window);
+
+	if (vmx->ple_window != old)
+		vmx->ple_window_dirty = true;
+
+	trace_kvm_ple_window_shrink(vcpu->vcpu_id, vmx->ple_window, old);
+}
+
+/*
+ * ple_window_actual_max is computed to be one grow_ple_window() below
+ * ple_window_max. (See __grow_ple_window for the reason.)
+ * This prevents overflows, because ple_window_max is int.
+ * ple_window_max effectively rounded down to a multiple of ple_window_grow in
+ * this process.
+ * ple_window_max is also prevented from setting vmx->ple_window < ple_window.
+ */
+static void update_ple_window_actual_max(void)
+{
+	ple_window_actual_max =
+			__shrink_ple_window(max(ple_window_max, ple_window),
+			                    ple_window_grow, INT_MIN);
+}
+
 /*
  * Indicate a busy-waiting vcpu in spinlock. We do not enable the PAUSE
  * exiting, so only get here on cpu with PAUSE-Loop-Exiting.
  */
 static int handle_pause(struct kvm_vcpu *vcpu)
 {
+	if (ple_gap)
+		grow_ple_window(vcpu);
+
 	skip_emulated_instruction(vcpu);
 	kvm_vcpu_on_spin(vcpu);
 
@@ -6146,7 +6249,11 @@ 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);
@@ -6617,7 +6724,7 @@ static int handle_invept(struct kvm_vcpu *vcpu)
 	switch (type) {
 	case VMX_EPT_EXTENT_GLOBAL:
 		kvm_mmu_sync_roots(vcpu);
-		kvm_mmu_flush_tlb(vcpu);
+		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 		nested_vmx_succeed(vcpu);
 		break;
 	default:
@@ -6892,6 +6999,8 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu)
 	case EXIT_REASON_TASK_SWITCH:
 		return 1;
 	case EXIT_REASON_CPUID:
+		if (kvm_register_read(vcpu, VCPU_REGS_RAX) == 0xa)
+			return 0;
 		return 1;
 	case EXIT_REASON_HLT:
 		return nested_cpu_has(vmcs12, CPU_BASED_HLT_EXITING);
@@ -6936,7 +7045,7 @@ 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);
@@ -7057,6 +7166,12 @@ 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;
@@ -7094,6 +7209,29 @@ static void vmx_set_virtual_x2apic_mode(struct kvm_vcpu *vcpu, bool set)
 	vmx_set_msr_bitmap(vcpu);
 }
 
+static void vmx_set_apic_access_page_addr(struct kvm_vcpu *vcpu, hpa_t hpa)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	/*
+	 * Currently we do not handle the nested case where L2 has an
+	 * APIC access page of its own; that page is still pinned.
+	 * Hence, we skip the case where the VCPU is in guest mode _and_
+	 * L1 prepared an APIC access page for L2.
+	 *
+	 * For the case where L1 and L2 share the same APIC access page
+	 * (flexpriority=Y but SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES clear
+	 * in the vmcs12), this function will only update either the vmcs01
+	 * or the vmcs02.  If the former, the vmcs02 will be updated by
+	 * prepare_vmcs02.  If the latter, the vmcs01 will be updated in
+	 * the next L2->L1 exit.
+	 */
+	if (!is_guest_mode(vcpu) ||
+	    !nested_cpu_has2(vmx->nested.current_vmcs12,
+			     SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES))
+		vmcs_write64(APIC_ACCESS_ADDR, hpa);
+}
+
 static void vmx_hwapic_isr_update(struct kvm *kvm, int isr)
 {
 	u16 status;
@@ -7387,6 +7525,11 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	if (vmx->emulation_required)
 		return;
 
+	if (vmx->ple_window_dirty) {
+		vmx->ple_window_dirty = false;
+		vmcs_write32(PLE_WINDOW, vmx->ple_window);
+	}
+
 	if (vmx->nested.sync_shadow_vmcs) {
 		copy_vmcs12_to_shadow(vmx);
 		vmx->nested.sync_shadow_vmcs = false;
@@ -7642,10 +7785,8 @@ 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;
 	}
 
@@ -7824,6 +7965,55 @@ 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;
@@ -7849,7 +8039,7 @@ static void vmx_start_preemption_timer(struct kvm_vcpu *vcpu)
 /*
  * prepare_vmcs02 is called when the L1 guest hypervisor runs its nested
  * L2 guest. L1 has a vmcs for L2 (vmcs12), and this function "merges" it
- * with L0's requirements for its guest (a.k.a. vmsc01), so we can run the L2
+ * with L0's requirements for its guest (a.k.a. vmcs01), so we can run the L2
  * guest in a way that will both be appropriate to L1's requests, and our
  * needs. In addition to modifying the active vmcs (which is vmcs02), this
  * function also has additional necessary side-effects, like setting various
@@ -7969,16 +8159,6 @@ 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
@@ -7994,8 +8174,7 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 		} else if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm)) {
 			exec_control |=
 				SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES;
-			vmcs_write64(APIC_ACCESS_ADDR,
-				page_to_phys(vcpu->kvm->arch.apic_access_page));
+			kvm_vcpu_reload_apic_access_page(vcpu);
 		}
 
 		vmcs_write32(SECONDARY_VM_EXEC_CONTROL, exec_control);
@@ -8024,6 +8203,13 @@ 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.
@@ -8185,8 +8371,7 @@ 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;
@@ -8790,9 +8975,19 @@ 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;
 	}
 
+	/*
+	 * We are now running in L2, mmu_notifier will force to reload the
+	 * page's hpa for L2 vmcs. Need to reload it for L1 before entering L1.
+	 */
+	kvm_vcpu_reload_apic_access_page(vcpu);
+
 	/*
 	 * Exiting from L2 to L1, we're now back to L1 which thinks it just
 	 * finished a VMLAUNCH or VMRESUME instruction, so we need to set the
@@ -8846,6 +9041,12 @@ static int vmx_check_intercept(struct kvm_vcpu *vcpu,
 	return X86EMUL_CONTINUE;
 }
 
+static void vmx_sched_in(struct kvm_vcpu *vcpu, int cpu)
+{
+	if (ple_gap)
+		shrink_ple_window(vcpu);
+}
+
 static struct kvm_x86_ops vmx_x86_ops = {
 	.cpu_has_kvm_support = cpu_has_kvm_support,
 	.disabled_by_bios = vmx_disabled_by_bios,
@@ -8890,7 +9091,6 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.cache_reg = vmx_cache_reg,
 	.get_rflags = vmx_get_rflags,
 	.set_rflags = vmx_set_rflags,
-	.fpu_activate = vmx_fpu_activate,
 	.fpu_deactivate = vmx_fpu_deactivate,
 
 	.tlb_flush = vmx_flush_tlb,
@@ -8913,6 +9113,7 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.enable_irq_window = enable_irq_window,
 	.update_cr8_intercept = update_cr8_intercept,
 	.set_virtual_x2apic_mode = vmx_set_virtual_x2apic_mode,
+	.set_apic_access_page_addr = vmx_set_apic_access_page_addr,
 	.vm_has_apicv = vmx_vm_has_apicv,
 	.load_eoi_exitmap = vmx_load_eoi_exitmap,
 	.hwapic_irr_update = vmx_hwapic_irr_update,
@@ -8951,6 +9152,8 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.mpx_supported = vmx_mpx_supported,
 
 	.check_nested_events = vmx_check_nested_events,
+
+	.sched_in = vmx_sched_in,
 };
 
 static int __init vmx_init(void)
@@ -9065,6 +9268,8 @@ static int __init vmx_init(void)
 	} else
 		kvm_disable_tdp();
 
+	update_ple_window_actual_max();
+
 	return 0;
 
 out7:
@@ -9098,7 +9303,7 @@ static void __exit vmx_exit(void)
 	free_page((unsigned long)vmx_vmread_bitmap);
 
 #ifdef CONFIG_KEXEC
-	rcu_assign_pointer(crash_vmclear_loaded_vmcss, NULL);
+	RCU_INIT_POINTER(crash_vmclear_loaded_vmcss, NULL);
 	synchronize_rcu();
 #endif
 

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;
 
@@ -726,7 +729,7 @@ int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
 {
 	if (cr3 == kvm_read_cr3(vcpu) && !pdptrs_changed(vcpu)) {
 		kvm_mmu_sync_roots(vcpu);
-		kvm_mmu_flush_tlb(vcpu);
+		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
 		return 0;
 	}
 
@@ -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);
 }
@@ -1723,9 +1726,10 @@ static bool valid_mtrr_type(unsigned t)
 	return t < 8 && (1 << t) & 0x73; /* 0, 1, 4, 5, 6 */
 }
 
-static bool mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data)
+bool kvm_mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 {
 	int i;
+	u64 mask;
 
 	if (!msr_mtrr_valid(msr))
 		return false;
@@ -1747,14 +1751,31 @@ static bool mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 	}
 
 	/* variable MTRRs */
-	return valid_mtrr_type(data & 0xff);
+	WARN_ON(!(msr >= 0x200 && msr < 0x200 + 2 * KVM_NR_VAR_MTRR));
+
+	mask = (~0ULL) << cpuid_maxphyaddr(vcpu);
+	if ((msr & 1) == 0) {
+		/* MTRR base */
+		if (!valid_mtrr_type(data & 0xff))
+			return false;
+		mask |= 0xf00;
+	} else
+		/* MTRR mask */
+		mask |= 0x7ff;
+	if (data & mask) {
+		kvm_inject_gp(vcpu, 0);
+		return false;
+	}
+
+	return true;
 }
+EXPORT_SYMBOL_GPL(kvm_mtrr_valid);
 
 static int set_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 {
 	u64 *p = (u64 *)&vcpu->arch.mtrr_state.fixed_ranges;
 
-	if (!mtrr_valid(vcpu, msr, data))
+	if (!kvm_mtrr_valid(vcpu, msr, data))
 		return 1;
 
 	if (msr == MSR_MTRRdefType) {
@@ -1805,7 +1826,7 @@ static int set_msr_mce(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 		break;
 	default:
 		if (msr >= MSR_IA32_MC0_CTL &&
-		    msr < MSR_IA32_MC0_CTL + 4 * bank_num) {
+		    msr < MSR_IA32_MCx_CTL(bank_num)) {
 			u32 offset = msr - MSR_IA32_MC0_CTL;
 			/* only 0 or all 1s can be written to IA32_MCi_CTL
 			 * some Linux kernels though clear bit 10 in bank 4 to
@@ -2164,7 +2185,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 
 	case MSR_IA32_MCG_CTL:
 	case MSR_IA32_MCG_STATUS:
-	case MSR_IA32_MC0_CTL ... MSR_IA32_MC0_CTL + 4 * KVM_MAX_MCE_BANKS - 1:
+	case MSR_IA32_MC0_CTL ... MSR_IA32_MCx_CTL(KVM_MAX_MCE_BANKS) - 1:
 		return set_msr_mce(vcpu, msr, data);
 
 	/* Performance counters are not protected by a CPUID bit,
@@ -2330,7 +2351,7 @@ static int get_msr_mce(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 		break;
 	default:
 		if (msr >= MSR_IA32_MC0_CTL &&
-		    msr < MSR_IA32_MC0_CTL + 4 * bank_num) {
+		    msr < MSR_IA32_MCx_CTL(bank_num)) {
 			u32 offset = msr - MSR_IA32_MC0_CTL;
 			data = vcpu->arch.mce_banks[offset];
 			break;
@@ -2419,7 +2440,13 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 	case MSR_K7_HWCR:
 	case MSR_VM_HSAVE_PA:
 	case MSR_K7_EVNTSEL0:
+	case MSR_K7_EVNTSEL1:
+	case MSR_K7_EVNTSEL2:
+	case MSR_K7_EVNTSEL3:
 	case MSR_K7_PERFCTR0:
+	case MSR_K7_PERFCTR1:
+	case MSR_K7_PERFCTR2:
+	case MSR_K7_PERFCTR3:
 	case MSR_K8_INT_PENDING_MSG:
 	case MSR_AMD64_NB_CFG:
 	case MSR_FAM10H_MMIO_CONF_BASE:
@@ -2505,7 +2532,7 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 	case MSR_IA32_MCG_CAP:
 	case MSR_IA32_MCG_CTL:
 	case MSR_IA32_MCG_STATUS:
-	case MSR_IA32_MC0_CTL ... MSR_IA32_MC0_CTL + 4 * KVM_MAX_MCE_BANKS - 1:
+	case MSR_IA32_MC0_CTL ... MSR_IA32_MCx_CTL(KVM_MAX_MCE_BANKS) - 1:
 		return get_msr_mce(vcpu, msr, pdata);
 	case MSR_K7_CLK_CTL:
 		/*
@@ -2823,7 +2850,7 @@ 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()) {
@@ -4040,16 +4067,16 @@ 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;
 }
@@ -4906,16 +4933,18 @@ 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)
@@ -4972,7 +5001,7 @@ static int handle_emulation_failure(struct kvm_vcpu *vcpu)
 
 	++vcpu->stat.insn_emulation_fail;
 	trace_kvm_emulate_insn_failed(vcpu);
-	if (!is_guest_mode(vcpu)) {
+	if (!is_guest_mode(vcpu) && kvm_x86_ops->get_cpl(vcpu) == 0) {
 		vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
 		vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
 		vcpu->run->internal.ndata = 0;
@@ -5224,6 +5253,7 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu,
 
 		ctxt->interruptibility = 0;
 		ctxt->have_exception = false;
+		ctxt->exception.vector = -1;
 		ctxt->perm_ok = false;
 
 		ctxt->ud = emulation_type & EMULTYPE_TRAP_UD;
@@ -5276,8 +5306,9 @@ 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.  */
@@ -5545,7 +5576,7 @@ 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;
@@ -5576,7 +5607,7 @@ 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);
 }
@@ -5989,6 +6020,44 @@ static void vcpu_scan_ioapic(struct kvm_vcpu *vcpu)
 	kvm_apic_update_tmr(vcpu, tmr);
 }
 
+static void kvm_vcpu_flush_tlb(struct kvm_vcpu *vcpu)
+{
+	++vcpu->stat.tlb_flush;
+	kvm_x86_ops->tlb_flush(vcpu);
+}
+
+void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu)
+{
+	struct page *page = NULL;
+
+	if (!irqchip_in_kernel(vcpu->kvm))
+		return;
+
+	if (!kvm_x86_ops->set_apic_access_page_addr)
+		return;
+
+	page = gfn_to_page(vcpu->kvm, APIC_DEFAULT_PHYS_BASE >> PAGE_SHIFT);
+	kvm_x86_ops->set_apic_access_page_addr(vcpu, page_to_phys(page));
+
+	/*
+	 * Do not pin apic access page in memory, the MMU notifier
+	 * will call us again if it is migrated or swapped out.
+	 */
+	put_page(page);
+}
+EXPORT_SYMBOL_GPL(kvm_vcpu_reload_apic_access_page);
+
+void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
+					   unsigned long address)
+{
+	/*
+	 * The physical address of apic access page is stored in the VMCS.
+	 * Update it when it becomes invalid.
+	 */
+	if (address == gfn_to_hva(kvm, APIC_DEFAULT_PHYS_BASE >> PAGE_SHIFT))
+		kvm_make_all_cpus_request(kvm, KVM_REQ_APIC_PAGE_RELOAD);
+}
+
 /*
  * Returns 1 to let __vcpu_run() continue the guest execution loop without
  * exiting to the userspace.  Otherwise, the value will be returned to the
@@ -6018,7 +6087,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 		if (kvm_check_request(KVM_REQ_MMU_SYNC, vcpu))
 			kvm_mmu_sync_roots(vcpu);
 		if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu))
-			kvm_x86_ops->tlb_flush(vcpu);
+			kvm_vcpu_flush_tlb(vcpu);
 		if (kvm_check_request(KVM_REQ_REPORT_TPR_ACCESS, vcpu)) {
 			vcpu->run->exit_reason = KVM_EXIT_TPR_ACCESS;
 			r = 0;
@@ -6049,6 +6118,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 			kvm_deliver_pmi(vcpu);
 		if (kvm_check_request(KVM_REQ_SCAN_IOAPIC, vcpu))
 			vcpu_scan_ioapic(vcpu);
+		if (kvm_check_request(KVM_REQ_APIC_PAGE_RELOAD, vcpu))
+			kvm_vcpu_reload_apic_access_page(vcpu);
 	}
 
 	if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) {
@@ -6934,7 +7005,7 @@ 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;
@@ -6945,7 +7016,7 @@ 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;
 
@@ -6954,7 +7025,7 @@ 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)
@@ -7008,8 +7079,7 @@ 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);
 			}
 
 			/*
@@ -7026,10 +7096,10 @@ 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)
@@ -7146,6 +7216,11 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 		static_key_slow_dec(&kvm_no_apic_vcpu);
 }
 
+void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu)
+{
+	kvm_x86_ops->sched_in(vcpu, cpu);
+}
+
 int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
 	if (type)
@@ -7237,10 +7312,6 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kfree(kvm->arch.vpic);
 	kfree(kvm->arch.vioapic);
 	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));
 }
 
@@ -7643,3 +7714,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_invlpga);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_skinit);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_write_tsc_offset);
+EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ple_window);

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;
@@ -149,6 +159,8 @@ int kvm_write_guest_virt_system(struct x86_emulate_ctxt *ctxt,
 	gva_t addr, void *val, unsigned int bytes,
 	struct x86_exception *exception);
 
+bool kvm_mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data);
+
 #define KVM_SUPPORTED_XCR0     (XSTATE_FP | XSTATE_SSE | XSTATE_YMM \
 				| XSTATE_BNDREGS | XSTATE_BNDCSR)
 extern u64 host_xcr0;

drivers/iommu/amd_iommu_v2.c

@@ -402,9 +402,11 @@ static void __mn_flush_page(struct mmu_notifier *mn,
 
 static int mn_clear_flush_young(struct mmu_notifier *mn,
 				struct mm_struct *mm,
-				unsigned long address)
+				unsigned long start,
+				unsigned long end)
 {
-	__mn_flush_page(mn, address);
+	for (; start < end; start += PAGE_SIZE)
+		__mn_flush_page(mn, start);
 
 	return 0;
 }

include/kvm/arm_vgic.h

@@ -25,26 +25,25 @@
 #include <linux/spinlock.h>
 #include <linux/types.h>
 
-#define VGIC_NR_IRQS		256
+#define VGIC_NR_IRQS_LEGACY	256
 #define VGIC_NR_SGIS		16
 #define VGIC_NR_PPIS		16
 #define VGIC_NR_PRIVATE_IRQS	(VGIC_NR_SGIS + VGIC_NR_PPIS)
-#define VGIC_NR_SHARED_IRQS	(VGIC_NR_IRQS - VGIC_NR_PRIVATE_IRQS)
-#define VGIC_MAX_CPUS		KVM_MAX_VCPUS
 
 #define VGIC_V2_MAX_LRS		(1 << 6)
 #define VGIC_V3_MAX_LRS		16
+#define VGIC_MAX_IRQS		1024
 
 /* Sanity checks... */
-#if (VGIC_MAX_CPUS > 8)
+#if (KVM_MAX_VCPUS > 8)
 #error	Invalid number of CPU interfaces
 #endif
 
-#if (VGIC_NR_IRQS & 31)
+#if (VGIC_NR_IRQS_LEGACY & 31)
 #error "VGIC_NR_IRQS must be a multiple of 32"
 #endif
 
-#if (VGIC_NR_IRQS > 1024)
+#if (VGIC_NR_IRQS_LEGACY > VGIC_MAX_IRQS)
 #error "VGIC_NR_IRQS must be <= 1024"
 #endif
 
@@ -54,19 +53,33 @@
  * - a bunch of shared interrupts (SPI)
  */
 struct vgic_bitmap {
-	union {
-		u32 reg[VGIC_NR_PRIVATE_IRQS / 32];
-		DECLARE_BITMAP(reg_ul, VGIC_NR_PRIVATE_IRQS);
-	} percpu[VGIC_MAX_CPUS];
-	union {
-		u32 reg[VGIC_NR_SHARED_IRQS / 32];
-		DECLARE_BITMAP(reg_ul, VGIC_NR_SHARED_IRQS);
-	} shared;
+	/*
+	 * - One UL per VCPU for private interrupts (assumes UL is at
+	 *   least 32 bits)
+	 * - As many UL as necessary for shared interrupts.
+	 *
+	 * The private interrupts are accessed via the "private"
+	 * field, one UL per vcpu (the state for vcpu n is in
+	 * private[n]). The shared interrupts are accessed via the
+	 * "shared" pointer (IRQn state is at bit n-32 in the bitmap).
+	 */
+	unsigned long *private;
+	unsigned long *shared;
 };
 
 struct vgic_bytemap {
-	u32 percpu[VGIC_MAX_CPUS][VGIC_NR_PRIVATE_IRQS / 4];
-	u32 shared[VGIC_NR_SHARED_IRQS  / 4];
+	/*
+	 * - 8 u32 per VCPU for private interrupts
+	 * - As many u32 as necessary for shared interrupts.
+	 *
+	 * The private interrupts are accessed via the "private"
+	 * field, (the state for vcpu n is in private[n*8] to
+	 * private[n*8 + 7]). The shared interrupts are accessed via
+	 * the "shared" pointer (IRQn state is at byte (n-32)%4 of the
+	 * shared[(n-32)/4] word).
+	 */
+	u32 *private;
+	u32 *shared;
 };
 
 struct kvm_vcpu;
@@ -127,6 +140,9 @@ struct vgic_dist {
 	bool			in_kernel;
 	bool			ready;
 
+	int			nr_cpus;
+	int			nr_irqs;
+
 	/* Virtual control interface mapping */
 	void __iomem		*vctrl_base;
 
@@ -140,11 +156,25 @@ struct vgic_dist {
 	/* Interrupt enabled (one bit per IRQ) */
 	struct vgic_bitmap	irq_enabled;
 
-	/* Interrupt 'pin' level */
-	struct vgic_bitmap	irq_state;
+	/* Level-triggered interrupt external input is asserted */
+	struct vgic_bitmap	irq_level;
 
-	/* Level-triggered interrupt in progress */
-	struct vgic_bitmap	irq_active;
+	/*
+	 * Interrupt state is pending on the distributor
+	 */
+	struct vgic_bitmap	irq_pending;
+
+	/*
+	 * Tracks writes to GICD_ISPENDRn and GICD_ICPENDRn for level-triggered
+	 * interrupts.  Essentially holds the state of the flip-flop in
+	 * Figure 4-10 on page 4-101 in ARM IHI 0048B.b.
+	 * Once set, it is only cleared for level-triggered interrupts on
+	 * guest ACKs (when we queue it) or writes to GICD_ICPENDRn.
+	 */
+	struct vgic_bitmap	irq_soft_pend;
+
+	/* Level-triggered interrupt queued on VCPU interface */
+	struct vgic_bitmap	irq_queued;
 
 	/* Interrupt priority. Not used yet. */
 	struct vgic_bytemap	irq_priority;
@@ -152,15 +182,36 @@ struct vgic_dist {
 	/* Level/edge triggered */
 	struct vgic_bitmap	irq_cfg;
 
-	/* Source CPU per SGI and target CPU */
-	u8			irq_sgi_sources[VGIC_MAX_CPUS][VGIC_NR_SGIS];
-
-	/* Target CPU for each IRQ */
-	u8			irq_spi_cpu[VGIC_NR_SHARED_IRQS];
-	struct vgic_bitmap	irq_spi_target[VGIC_MAX_CPUS];
+	/*
+	 * Source CPU per SGI and target CPU:
+	 *
+	 * Each byte represent a SGI observable on a VCPU, each bit of
+	 * this byte indicating if the corresponding VCPU has
+	 * generated this interrupt. This is a GICv2 feature only.
+	 *
+	 * For VCPUn (n < 8), irq_sgi_sources[n*16] to [n*16 + 15] are
+	 * the SGIs observable on VCPUn.
+	 */
+	u8			*irq_sgi_sources;
+
+	/*
+	 * Target CPU for each SPI:
+	 *
+	 * Array of available SPI, each byte indicating the target
+	 * VCPU for SPI. IRQn (n >=32) is at irq_spi_cpu[n-32].
+	 */
+	u8			*irq_spi_cpu;
+
+	/*
+	 * Reverse lookup of irq_spi_cpu for faster compute pending:
+	 *
+	 * Array of bitmaps, one per VCPU, describing if IRQn is
+	 * routed to a particular VCPU.
+	 */
+	struct vgic_bitmap	*irq_spi_target;
 
 	/* Bitmap indicating which CPU has something pending */
-	unsigned long		irq_pending_on_cpu;
+	unsigned long		*irq_pending_on_cpu;
 #endif
 };
 
@@ -190,11 +241,11 @@ struct vgic_v3_cpu_if {
 struct vgic_cpu {
 #ifdef CONFIG_KVM_ARM_VGIC
 	/* per IRQ to LR mapping */
-	u8		vgic_irq_lr_map[VGIC_NR_IRQS];
+	u8		*vgic_irq_lr_map;
 
 	/* Pending interrupts on this VCPU */
 	DECLARE_BITMAP(	pending_percpu, VGIC_NR_PRIVATE_IRQS);
-	DECLARE_BITMAP(	pending_shared, VGIC_NR_SHARED_IRQS);
+	unsigned long	*pending_shared;
 
 	/* Bitmap of used/free list registers */
 	DECLARE_BITMAP(	lr_used, VGIC_V2_MAX_LRS);
@@ -225,7 +276,8 @@ int kvm_vgic_addr(struct kvm *kvm, unsigned long type, u64 *addr, bool write);
 int kvm_vgic_hyp_init(void);
 int kvm_vgic_init(struct kvm *kvm);
 int kvm_vgic_create(struct kvm *kvm);
-int kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu);
+void kvm_vgic_destroy(struct kvm *kvm);
+void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu);
 void kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu);
 void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu);
 int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int irq_num,

include/linux/kvm_host.h

@@ -136,12 +136,11 @@ static inline bool is_error_page(struct page *page)
 #define KVM_REQ_GLOBAL_CLOCK_UPDATE 22
 #define KVM_REQ_ENABLE_IBS        23
 #define KVM_REQ_DISABLE_IBS       24
+#define KVM_REQ_APIC_PAGE_RELOAD  25
 
 #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;
@@ -200,6 +199,17 @@ int kvm_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, unsigned long hva,
 int kvm_async_pf_wakeup_all(struct kvm_vcpu *vcpu);
 #endif
 
+/*
+ * Carry out a gup that requires IO. Allow the mm to relinquish the mmap
+ * semaphore if the filemap/swap has to wait on a page lock. pagep == NULL
+ * controls whether we retry the gup one more time to completion in that case.
+ * Typically this is called after a FAULT_FLAG_RETRY_NOWAIT in the main tdp
+ * handler.
+ */
+int kvm_get_user_page_io(struct task_struct *tsk, struct mm_struct *mm,
+			 unsigned long addr, bool write_fault,
+			 struct page **pagep);
+
 enum {
 	OUTSIDE_GUEST_MODE,
 	IN_GUEST_MODE,
@@ -325,8 +335,6 @@ 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
@@ -528,6 +536,8 @@ struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn);
 unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn);
 unsigned long gfn_to_hva_prot(struct kvm *kvm, gfn_t gfn, bool *writable);
 unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot, gfn_t gfn);
+unsigned long gfn_to_hva_memslot_prot(struct kvm_memory_slot *slot, gfn_t gfn,
+				      bool *writable);
 void kvm_release_page_clean(struct page *page);
 void kvm_release_page_dirty(struct page *page);
 void kvm_set_page_accessed(struct page *page);
@@ -579,6 +589,7 @@ void kvm_flush_remote_tlbs(struct kvm *kvm);
 void kvm_reload_remote_mmus(struct kvm *kvm);
 void kvm_make_mclock_inprogress_request(struct kvm *kvm);
 void kvm_make_scan_ioapic_request(struct kvm *kvm);
+bool kvm_make_all_cpus_request(struct kvm *kvm, unsigned int req);
 
 long kvm_arch_dev_ioctl(struct file *filp,
 			unsigned int ioctl, unsigned long arg);
@@ -624,6 +635,8 @@ void kvm_arch_exit(void);
 int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu);
 
+void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu);
+
 void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
@@ -632,8 +645,8 @@ 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);
@@ -1034,8 +1047,6 @@ 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;
@@ -1068,12 +1079,10 @@ 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/linux/mm.h

@@ -1985,6 +1985,7 @@ static inline struct page *follow_page(struct vm_area_struct *vma,
 #define FOLL_HWPOISON	0x100	/* check page is hwpoisoned */
 #define FOLL_NUMA	0x200	/* force NUMA hinting page fault */
 #define FOLL_MIGRATION	0x400	/* wait for page to replace migration entry */
+#define FOLL_TRIED	0x800	/* a retry, previous pass started an IO */
 
 typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
 			void *data);

include/linux/mmu_notifier.h

@@ -57,10 +57,13 @@ struct mmu_notifier_ops {
 	 * pte. This way the VM will provide proper aging to the
 	 * accesses to the page through the secondary MMUs and not
 	 * only to the ones through the Linux pte.
+	 * Start-end is necessary in case the secondary MMU is mapping the page
+	 * at a smaller granularity than the primary MMU.
 	 */
 	int (*clear_flush_young)(struct mmu_notifier *mn,
 				 struct mm_struct *mm,
-				 unsigned long address);
+				 unsigned long start,
+				 unsigned long end);
 
 	/*
 	 * test_young is called to check the young/accessed bitflag in
@@ -175,7 +178,8 @@ extern void mmu_notifier_unregister_no_release(struct mmu_notifier *mn,
 extern void __mmu_notifier_mm_destroy(struct mm_struct *mm);
 extern void __mmu_notifier_release(struct mm_struct *mm);
 extern int __mmu_notifier_clear_flush_young(struct mm_struct *mm,
-					  unsigned long address);
+					  unsigned long start,
+					  unsigned long end);
 extern int __mmu_notifier_test_young(struct mm_struct *mm,
 				     unsigned long address);
 extern void __mmu_notifier_change_pte(struct mm_struct *mm,
@@ -194,10 +198,11 @@ static inline void mmu_notifier_release(struct mm_struct *mm)
 }
 
 static inline int mmu_notifier_clear_flush_young(struct mm_struct *mm,
-					  unsigned long address)
+					  unsigned long start,
+					  unsigned long end)
 {
 	if (mm_has_notifiers(mm))
-		return __mmu_notifier_clear_flush_young(mm, address);
+		return __mmu_notifier_clear_flush_young(mm, start, end);
 	return 0;
 }
 
@@ -255,7 +260,9 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm)
 	unsigned long ___address = __address;				\
 	__young = ptep_clear_flush_young(___vma, ___address, __ptep);	\
 	__young |= mmu_notifier_clear_flush_young(___vma->vm_mm,	\
-						  ___address);		\
+						  ___address,		\
+						  ___address +		\
+							PAGE_SIZE);	\
 	__young;							\
 })
 
@@ -266,7 +273,9 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm)
 	unsigned long ___address = __address;				\
 	__young = pmdp_clear_flush_young(___vma, ___address, __pmdp);	\
 	__young |= mmu_notifier_clear_flush_young(___vma->vm_mm,	\
-						  ___address);		\
+						  ___address,		\
+						  ___address +		\
+							PMD_SIZE);	\
 	__young;							\
 })
 
@@ -301,7 +310,8 @@ static inline void mmu_notifier_release(struct mm_struct *mm)
 }
 
 static inline int mmu_notifier_clear_flush_young(struct mm_struct *mm,
-					  unsigned long address)
+					  unsigned long start,
+					  unsigned long end)
 {
 	return 0;
 }

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),
@@ -205,24 +225,26 @@ TRACE_EVENT(kvm_fpu,
 );
 
 TRACE_EVENT(kvm_age_page,
-	TP_PROTO(ulong hva, struct kvm_memory_slot *slot, int ref),
-	TP_ARGS(hva, slot, ref),
+	TP_PROTO(ulong gfn, int level, struct kvm_memory_slot *slot, int ref),
+	TP_ARGS(gfn, level, slot, ref),
 
 	TP_STRUCT__entry(
 		__field(	u64,	hva		)
 		__field(	u64,	gfn		)
+		__field(	u8,	level		)
 		__field(	u8,	referenced	)
 	),
 
 	TP_fast_assign(
-		__entry->hva		= hva;
-		__entry->gfn		=
-		  slot->base_gfn + ((hva - slot->userspace_addr) >> PAGE_SHIFT);
+		__entry->gfn		= gfn;
+		__entry->level		= level;
+		__entry->hva		= ((gfn - slot->base_gfn) <<
+					    PAGE_SHIFT) + slot->userspace_addr;
 		__entry->referenced	= ref;
 	),
 
-	TP_printk("hva %llx gfn %llx %s",
-		  __entry->hva, __entry->gfn,
+	TP_printk("hva %llx gfn %llx level %u %s",
+		  __entry->hva, __entry->gfn, __entry->level,
 		  __entry->referenced ? "YOUNG" : "OLD")
 );
 

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)

mm/gup.c

@@ -281,6 +281,10 @@ static int faultin_page(struct task_struct *tsk, struct vm_area_struct *vma,
 		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
 	if (*flags & FOLL_NOWAIT)
 		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;
+	if (*flags & FOLL_TRIED) {
+		VM_WARN_ON_ONCE(fault_flags & FAULT_FLAG_ALLOW_RETRY);
+		fault_flags |= FAULT_FLAG_TRIED;
+	}
 
 	ret = handle_mm_fault(mm, vma, address, fault_flags);
 	if (ret & VM_FAULT_ERROR) {

mm/mmu_notifier.c

@@ -107,7 +107,8 @@ void __mmu_notifier_release(struct mm_struct *mm)
  * existed or not.
  */
 int __mmu_notifier_clear_flush_young(struct mm_struct *mm,
-					unsigned long address)
+					unsigned long start,
+					unsigned long end)
 {
 	struct mmu_notifier *mn;
 	int young = 0, id;
@@ -115,7 +116,7 @@ int __mmu_notifier_clear_flush_young(struct mm_struct *mm,
 	id = srcu_read_lock(&srcu);
 	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
 		if (mn->ops->clear_flush_young)
-			young |= mn->ops->clear_flush_young(mn, mm, address);
+			young |= mn->ops->clear_flush_young(mn, mm, start, end);
 	}
 	srcu_read_unlock(&srcu, id);
 

mm/rmap.c

@@ -1355,7 +1355,11 @@ static int try_to_unmap_cluster(unsigned long cursor, unsigned int *mapcount,
 			continue;	/* don't unmap */
 		}
 
-		if (ptep_clear_flush_young_notify(vma, address, pte))
+		/*
+		 * No need for _notify because we're within an
+		 * mmu_notifier_invalidate_range_ {start|end} scope.
+		 */
+		if (ptep_clear_flush_young(vma, address, pte))
 			continue;
 
 		/* Nuke the page table entry. */

virt/kvm/arm/vgic.c

@@ -36,21 +36,22 @@
  * How the whole thing works (courtesy of Christoffer Dall):
  *
  * - At any time, the dist->irq_pending_on_cpu is the oracle that knows if
- *   something is pending
- * - VGIC pending interrupts are stored on the vgic.irq_state vgic
- *   bitmap (this bitmap is updated by both user land ioctls and guest
- *   mmio ops, and other in-kernel peripherals such as the
- *   arch. timers) and indicate the 'wire' state.
+ *   something is pending on the CPU interface.
+ * - Interrupts that are pending on the distributor are stored on the
+ *   vgic.irq_pending vgic bitmap (this bitmap is updated by both user land
+ *   ioctls and guest mmio ops, and other in-kernel peripherals such as the
+ *   arch. timers).
  * - Every time the bitmap changes, the irq_pending_on_cpu oracle is
  *   recalculated
  * - To calculate the oracle, we need info for each cpu from
  *   compute_pending_for_cpu, which considers:
- *   - PPI: dist->irq_state & dist->irq_enable
- *   - SPI: dist->irq_state & dist->irq_enable & dist->irq_spi_target
- *   - irq_spi_target is a 'formatted' version of the GICD_ICFGR
+ *   - PPI: dist->irq_pending & dist->irq_enable
+ *   - SPI: dist->irq_pending & dist->irq_enable & dist->irq_spi_target
+ *   - irq_spi_target is a 'formatted' version of the GICD_ITARGETSRn
  *     registers, stored on each vcpu. We only keep one bit of
  *     information per interrupt, making sure that only one vcpu can
  *     accept the interrupt.
+ * - If any of the above state changes, we must recalculate the oracle.
  * - The same is true when injecting an interrupt, except that we only
  *   consider a single interrupt at a time. The irq_spi_cpu array
  *   contains the target CPU for each SPI.
@@ -60,13 +61,18 @@
  * the 'line' again. This is achieved as such:
  *
  * - When a level interrupt is moved onto a vcpu, the corresponding
- *   bit in irq_active is set. As long as this bit is set, the line
+ *   bit in irq_queued is set. As long as this bit is set, the line
  *   will be ignored for further interrupts. The interrupt is injected
  *   into the vcpu with the GICH_LR_EOI bit set (generate a
  *   maintenance interrupt on EOI).
  * - When the interrupt is EOIed, the maintenance interrupt fires,
- *   and clears the corresponding bit in irq_active. This allow the
+ *   and clears the corresponding bit in irq_queued. This allows the
  *   interrupt line to be sampled again.
+ * - Note that level-triggered interrupts can also be set to pending from
+ *   writes to GICD_ISPENDRn and lowering the external input line does not
+ *   cause the interrupt to become inactive in such a situation.
+ *   Conversely, writes to GICD_ICPENDRn do not cause the interrupt to become
+ *   inactive as long as the external input line is held high.
  */
 
 #define VGIC_ADDR_UNDEF		(-1)
@@ -89,6 +95,7 @@ static void vgic_retire_disabled_irqs(struct kvm_vcpu *vcpu);
 static void vgic_retire_lr(int lr_nr, int irq, struct kvm_vcpu *vcpu);
 static void vgic_update_state(struct kvm *kvm);
 static void vgic_kick_vcpus(struct kvm *kvm);
+static u8 *vgic_get_sgi_sources(struct vgic_dist *dist, int vcpu_id, int sgi);
 static void vgic_dispatch_sgi(struct kvm_vcpu *vcpu, u32 reg);
 static struct vgic_lr vgic_get_lr(const struct kvm_vcpu *vcpu, int lr);
 static void vgic_set_lr(struct kvm_vcpu *vcpu, int lr, struct vgic_lr lr_desc);
@@ -99,10 +106,8 @@ static const struct vgic_ops *vgic_ops;
 static const struct vgic_params *vgic;
 
 /*
- * struct vgic_bitmap contains unions that provide two views of
- * the same data. In one case it is an array of registers of
- * u32's, and in the other case it is a bitmap of unsigned
- * longs.
+ * struct vgic_bitmap contains a bitmap made of unsigned longs, but
+ * extracts u32s out of them.
  *
  * This does not work on 64-bit BE systems, because the bitmap access
  * will store two consecutive 32-bit words with the higher-addressed
@@ -118,23 +123,45 @@ static const struct vgic_params *vgic;
 #define REG_OFFSET_SWIZZLE	0
 #endif
 
+static int vgic_init_bitmap(struct vgic_bitmap *b, int nr_cpus, int nr_irqs)
+{
+	int nr_longs;
+
+	nr_longs = nr_cpus + BITS_TO_LONGS(nr_irqs - VGIC_NR_PRIVATE_IRQS);
+
+	b->private = kzalloc(sizeof(unsigned long) * nr_longs, GFP_KERNEL);
+	if (!b->private)
+		return -ENOMEM;
+
+	b->shared = b->private + nr_cpus;
+
+	return 0;
+}
+
+static void vgic_free_bitmap(struct vgic_bitmap *b)
+{
+	kfree(b->private);
+	b->private = NULL;
+	b->shared = NULL;
+}
+
 static u32 *vgic_bitmap_get_reg(struct vgic_bitmap *x,
 				int cpuid, u32 offset)
 {
 	offset >>= 2;
 	if (!offset)
-		return x->percpu[cpuid].reg + (offset ^ REG_OFFSET_SWIZZLE);
+		return (u32 *)(x->private + cpuid) + REG_OFFSET_SWIZZLE;
 	else
-		return x->shared.reg + ((offset - 1) ^ REG_OFFSET_SWIZZLE);
+		return (u32 *)(x->shared) + ((offset - 1) ^ REG_OFFSET_SWIZZLE);
 }
 
 static int vgic_bitmap_get_irq_val(struct vgic_bitmap *x,
 				   int cpuid, int irq)
 {
 	if (irq < VGIC_NR_PRIVATE_IRQS)
-		return test_bit(irq, x->percpu[cpuid].reg_ul);
+		return test_bit(irq, x->private + cpuid);
 
-	return test_bit(irq - VGIC_NR_PRIVATE_IRQS, x->shared.reg_ul);
+	return test_bit(irq - VGIC_NR_PRIVATE_IRQS, x->shared);
 }
 
 static void vgic_bitmap_set_irq_val(struct vgic_bitmap *x, int cpuid,
@@ -143,9 +170,9 @@ static void vgic_bitmap_set_irq_val(struct vgic_bitmap *x, int cpuid,
 	unsigned long *reg;
 
 	if (irq < VGIC_NR_PRIVATE_IRQS) {
-		reg = x->percpu[cpuid].reg_ul;
+		reg = x->private + cpuid;
 	} else {
-		reg =  x->shared.reg_ul;
+		reg = x->shared;
 		irq -= VGIC_NR_PRIVATE_IRQS;
 	}
 
@@ -157,24 +184,49 @@ static void vgic_bitmap_set_irq_val(struct vgic_bitmap *x, int cpuid,
 
 static unsigned long *vgic_bitmap_get_cpu_map(struct vgic_bitmap *x, int cpuid)
 {
-	if (unlikely(cpuid >= VGIC_MAX_CPUS))
-		return NULL;
-	return x->percpu[cpuid].reg_ul;
+	return x->private + cpuid;
 }
 
 static unsigned long *vgic_bitmap_get_shared_map(struct vgic_bitmap *x)
 {
-	return x->shared.reg_ul;
+	return x->shared;
+}
+
+static int vgic_init_bytemap(struct vgic_bytemap *x, int nr_cpus, int nr_irqs)
+{
+	int size;
+
+	size  = nr_cpus * VGIC_NR_PRIVATE_IRQS;
+	size += nr_irqs - VGIC_NR_PRIVATE_IRQS;
+
+	x->private = kzalloc(size, GFP_KERNEL);
+	if (!x->private)
+		return -ENOMEM;
+
+	x->shared = x->private + nr_cpus * VGIC_NR_PRIVATE_IRQS / sizeof(u32);
+	return 0;
+}
+
+static void vgic_free_bytemap(struct vgic_bytemap *b)
+{
+	kfree(b->private);
+	b->private = NULL;
+	b->shared = NULL;
 }
 
 static u32 *vgic_bytemap_get_reg(struct vgic_bytemap *x, int cpuid, u32 offset)
 {
-	offset >>= 2;
-	BUG_ON(offset > (VGIC_NR_IRQS / 4));
-	if (offset < 8)
-		return x->percpu[cpuid] + offset;
-	else
-		return x->shared + offset - 8;
+	u32 *reg;
+
+	if (offset < VGIC_NR_PRIVATE_IRQS) {
+		reg = x->private;
+		offset += cpuid * VGIC_NR_PRIVATE_IRQS;
+	} else {
+		reg = x->shared;
+		offset -= VGIC_NR_PRIVATE_IRQS;
+	}
+
+	return reg + (offset / sizeof(u32));
 }
 
 #define VGIC_CFG_LEVEL	0
@@ -196,46 +248,81 @@ static int vgic_irq_is_enabled(struct kvm_vcpu *vcpu, int irq)
 	return vgic_bitmap_get_irq_val(&dist->irq_enabled, vcpu->vcpu_id, irq);
 }
 
-static int vgic_irq_is_active(struct kvm_vcpu *vcpu, int irq)
+static int vgic_irq_is_queued(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	return vgic_bitmap_get_irq_val(&dist->irq_queued, vcpu->vcpu_id, irq);
+}
+
+static void vgic_irq_set_queued(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	vgic_bitmap_set_irq_val(&dist->irq_queued, vcpu->vcpu_id, irq, 1);
+}
+
+static void vgic_irq_clear_queued(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	vgic_bitmap_set_irq_val(&dist->irq_queued, vcpu->vcpu_id, irq, 0);
+}
+
+static int vgic_dist_irq_get_level(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 
-	return vgic_bitmap_get_irq_val(&dist->irq_active, vcpu->vcpu_id, irq);
+	return vgic_bitmap_get_irq_val(&dist->irq_level, vcpu->vcpu_id, irq);
 }
 
-static void vgic_irq_set_active(struct kvm_vcpu *vcpu, int irq)
+static void vgic_dist_irq_set_level(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 
-	vgic_bitmap_set_irq_val(&dist->irq_active, vcpu->vcpu_id, irq, 1);
+	vgic_bitmap_set_irq_val(&dist->irq_level, vcpu->vcpu_id, irq, 1);
 }
 
-static void vgic_irq_clear_active(struct kvm_vcpu *vcpu, int irq)
+static void vgic_dist_irq_clear_level(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 
-	vgic_bitmap_set_irq_val(&dist->irq_active, vcpu->vcpu_id, irq, 0);
+	vgic_bitmap_set_irq_val(&dist->irq_level, vcpu->vcpu_id, irq, 0);
+}
+
+static int vgic_dist_irq_soft_pend(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	return vgic_bitmap_get_irq_val(&dist->irq_soft_pend, vcpu->vcpu_id, irq);
+}
+
+static void vgic_dist_irq_clear_soft_pend(struct kvm_vcpu *vcpu, int irq)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	vgic_bitmap_set_irq_val(&dist->irq_soft_pend, vcpu->vcpu_id, irq, 0);
 }
 
 static int vgic_dist_irq_is_pending(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 
-	return vgic_bitmap_get_irq_val(&dist->irq_state, vcpu->vcpu_id, irq);
+	return vgic_bitmap_get_irq_val(&dist->irq_pending, vcpu->vcpu_id, irq);
 }
 
-static void vgic_dist_irq_set(struct kvm_vcpu *vcpu, int irq)
+static void vgic_dist_irq_set_pending(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 
-	vgic_bitmap_set_irq_val(&dist->irq_state, vcpu->vcpu_id, irq, 1);
+	vgic_bitmap_set_irq_val(&dist->irq_pending, vcpu->vcpu_id, irq, 1);
 }
 
-static void vgic_dist_irq_clear(struct kvm_vcpu *vcpu, int irq)
+static void vgic_dist_irq_clear_pending(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 
-	vgic_bitmap_set_irq_val(&dist->irq_state, vcpu->vcpu_id, irq, 0);
+	vgic_bitmap_set_irq_val(&dist->irq_pending, vcpu->vcpu_id, irq, 0);
 }
 
 static void vgic_cpu_irq_set(struct kvm_vcpu *vcpu, int irq)
@@ -256,6 +343,11 @@ static void vgic_cpu_irq_clear(struct kvm_vcpu *vcpu, int irq)
 			  vcpu->arch.vgic_cpu.pending_shared);
 }
 
+static bool vgic_can_sample_irq(struct kvm_vcpu *vcpu, int irq)
+{
+	return vgic_irq_is_edge(vcpu, irq) || !vgic_irq_is_queued(vcpu, irq);
+}
+
 static u32 mmio_data_read(struct kvm_exit_mmio *mmio, u32 mask)
 {
 	return le32_to_cpu(*((u32 *)mmio->data)) & mask;
@@ -347,7 +439,7 @@ static bool handle_mmio_misc(struct kvm_vcpu *vcpu,
 
 	case 4:			/* GICD_TYPER */
 		reg  = (atomic_read(&vcpu->kvm->online_vcpus) - 1) << 5;
-		reg |= (VGIC_NR_IRQS >> 5) - 1;
+		reg |= (vcpu->kvm->arch.vgic.nr_irqs >> 5) - 1;
 		vgic_reg_access(mmio, &reg, word_offset,
 				ACCESS_READ_VALUE | ACCESS_WRITE_IGNORED);
 		break;
@@ -409,11 +501,33 @@ static bool handle_mmio_set_pending_reg(struct kvm_vcpu *vcpu,
 					struct kvm_exit_mmio *mmio,
 					phys_addr_t offset)
 {
-	u32 *reg = vgic_bitmap_get_reg(&vcpu->kvm->arch.vgic.irq_state,
-				       vcpu->vcpu_id, offset);
+	u32 *reg, orig;
+	u32 level_mask;
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	reg = vgic_bitmap_get_reg(&dist->irq_cfg, vcpu->vcpu_id, offset);
+	level_mask = (~(*reg));
+
+	/* Mark both level and edge triggered irqs as pending */
+	reg = vgic_bitmap_get_reg(&dist->irq_pending, vcpu->vcpu_id, offset);
+	orig = *reg;
 	vgic_reg_access(mmio, reg, offset,
 			ACCESS_READ_VALUE | ACCESS_WRITE_SETBIT);
+
 	if (mmio->is_write) {
+		/* Set the soft-pending flag only for level-triggered irqs */
+		reg = vgic_bitmap_get_reg(&dist->irq_soft_pend,
+					  vcpu->vcpu_id, offset);
+		vgic_reg_access(mmio, reg, offset,
+				ACCESS_READ_VALUE | ACCESS_WRITE_SETBIT);
+		*reg &= level_mask;
+
+		/* Ignore writes to SGIs */
+		if (offset < 2) {
+			*reg &= ~0xffff;
+			*reg |= orig & 0xffff;
+		}
+
 		vgic_update_state(vcpu->kvm);
 		return true;
 	}
@@ -425,11 +539,34 @@ static bool handle_mmio_clear_pending_reg(struct kvm_vcpu *vcpu,
 					  struct kvm_exit_mmio *mmio,
 					  phys_addr_t offset)
 {
-	u32 *reg = vgic_bitmap_get_reg(&vcpu->kvm->arch.vgic.irq_state,
-				       vcpu->vcpu_id, offset);
+	u32 *level_active;
+	u32 *reg, orig;
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	reg = vgic_bitmap_get_reg(&dist->irq_pending, vcpu->vcpu_id, offset);
+	orig = *reg;
 	vgic_reg_access(mmio, reg, offset,
 			ACCESS_READ_VALUE | ACCESS_WRITE_CLEARBIT);
 	if (mmio->is_write) {
+		/* Re-set level triggered level-active interrupts */
+		level_active = vgic_bitmap_get_reg(&dist->irq_level,
+					  vcpu->vcpu_id, offset);
+		reg = vgic_bitmap_get_reg(&dist->irq_pending,
+					  vcpu->vcpu_id, offset);
+		*reg |= *level_active;
+
+		/* Ignore writes to SGIs */
+		if (offset < 2) {
+			*reg &= ~0xffff;
+			*reg |= orig & 0xffff;
+		}
+
+		/* Clear soft-pending flags */
+		reg = vgic_bitmap_get_reg(&dist->irq_soft_pend,
+					  vcpu->vcpu_id, offset);
+		vgic_reg_access(mmio, reg, offset,
+				ACCESS_READ_VALUE | ACCESS_WRITE_CLEARBIT);
+
 		vgic_update_state(vcpu->kvm);
 		return true;
 	}
@@ -651,9 +788,9 @@ static void vgic_unqueue_irqs(struct kvm_vcpu *vcpu)
 		 * is fine, then we are only setting a few bits that were
 		 * already set.
 		 */
-		vgic_dist_irq_set(vcpu, lr.irq);
+		vgic_dist_irq_set_pending(vcpu, lr.irq);
 		if (lr.irq < VGIC_NR_SGIS)
-			dist->irq_sgi_sources[vcpu_id][lr.irq] |= 1 << lr.source;
+			*vgic_get_sgi_sources(dist, vcpu_id, lr.irq) |= 1 << lr.source;
 		lr.state &= ~LR_STATE_PENDING;
 		vgic_set_lr(vcpu, i, lr);
 
@@ -662,8 +799,10 @@ static void vgic_unqueue_irqs(struct kvm_vcpu *vcpu)
 		 * active), then the LR does not hold any useful info and can
 		 * be marked as free for other use.
 		 */
-		if (!(lr.state & LR_STATE_MASK))
+		if (!(lr.state & LR_STATE_MASK)) {
 			vgic_retire_lr(i, lr.irq, vcpu);
+			vgic_irq_clear_queued(vcpu, lr.irq);
+		}
 
 		/* Finally update the VGIC state. */
 		vgic_update_state(vcpu->kvm);
@@ -677,7 +816,7 @@ static bool read_set_clear_sgi_pend_reg(struct kvm_vcpu *vcpu,
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 	int sgi;
-	int min_sgi = (offset & ~0x3) * 4;
+	int min_sgi = (offset & ~0x3);
 	int max_sgi = min_sgi + 3;
 	int vcpu_id = vcpu->vcpu_id;
 	u32 reg = 0;
@@ -685,7 +824,7 @@ static bool read_set_clear_sgi_pend_reg(struct kvm_vcpu *vcpu,
 	/* Copy source SGIs from distributor side */
 	for (sgi = min_sgi; sgi <= max_sgi; sgi++) {
 		int shift = 8 * (sgi - min_sgi);
-		reg |= (u32)dist->irq_sgi_sources[vcpu_id][sgi] << shift;
+		reg |= ((u32)*vgic_get_sgi_sources(dist, vcpu_id, sgi)) << shift;
 	}
 
 	mmio_data_write(mmio, ~0, reg);
@@ -698,7 +837,7 @@ static bool write_set_clear_sgi_pend_reg(struct kvm_vcpu *vcpu,
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 	int sgi;
-	int min_sgi = (offset & ~0x3) * 4;
+	int min_sgi = (offset & ~0x3);
 	int max_sgi = min_sgi + 3;
 	int vcpu_id = vcpu->vcpu_id;
 	u32 reg;
@@ -709,14 +848,15 @@ static bool write_set_clear_sgi_pend_reg(struct kvm_vcpu *vcpu,
 	/* Clear pending SGIs on the distributor */
 	for (sgi = min_sgi; sgi <= max_sgi; sgi++) {
 		u8 mask = reg >> (8 * (sgi - min_sgi));
+		u8 *src = vgic_get_sgi_sources(dist, vcpu_id, sgi);
 		if (set) {
-			if ((dist->irq_sgi_sources[vcpu_id][sgi] & mask) != mask)
+			if ((*src & mask) != mask)
 				updated = true;
-			dist->irq_sgi_sources[vcpu_id][sgi] |= mask;
+			*src |= mask;
 		} else {
-			if (dist->irq_sgi_sources[vcpu_id][sgi] & mask)
+			if (*src & mask)
 				updated = true;
-			dist->irq_sgi_sources[vcpu_id][sgi] &= ~mask;
+			*src &= ~mask;
 		}
 	}
 
@@ -755,6 +895,7 @@ static bool handle_mmio_sgi_clear(struct kvm_vcpu *vcpu,
 struct mmio_range {
 	phys_addr_t base;
 	unsigned long len;
+	int bits_per_irq;
 	bool (*handle_mmio)(struct kvm_vcpu *vcpu, struct kvm_exit_mmio *mmio,
 			    phys_addr_t offset);
 };
@@ -763,56 +904,67 @@ static const struct mmio_range vgic_dist_ranges[] = {
 	{
 		.base		= GIC_DIST_CTRL,
 		.len		= 12,
+		.bits_per_irq	= 0,
 		.handle_mmio	= handle_mmio_misc,
 	},
 	{
 		.base		= GIC_DIST_IGROUP,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_raz_wi,
 	},
 	{
 		.base		= GIC_DIST_ENABLE_SET,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_set_enable_reg,
 	},
 	{
 		.base		= GIC_DIST_ENABLE_CLEAR,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_clear_enable_reg,
 	},
 	{
 		.base		= GIC_DIST_PENDING_SET,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_set_pending_reg,
 	},
 	{
 		.base		= GIC_DIST_PENDING_CLEAR,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_clear_pending_reg,
 	},
 	{
 		.base		= GIC_DIST_ACTIVE_SET,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_raz_wi,
 	},
 	{
 		.base		= GIC_DIST_ACTIVE_CLEAR,
-		.len		= VGIC_NR_IRQS / 8,
+		.len		= VGIC_MAX_IRQS / 8,
+		.bits_per_irq	= 1,
 		.handle_mmio	= handle_mmio_raz_wi,
 	},
 	{
 		.base		= GIC_DIST_PRI,
-		.len		= VGIC_NR_IRQS,
+		.len		= VGIC_MAX_IRQS,
+		.bits_per_irq	= 8,
 		.handle_mmio	= handle_mmio_priority_reg,
 	},
 	{
 		.base		= GIC_DIST_TARGET,
-		.len		= VGIC_NR_IRQS,
+		.len		= VGIC_MAX_IRQS,
+		.bits_per_irq	= 8,
 		.handle_mmio	= handle_mmio_target_reg,
 	},
 	{
 		.base		= GIC_DIST_CONFIG,
-		.len		= VGIC_NR_IRQS / 4,
+		.len		= VGIC_MAX_IRQS / 4,
+		.bits_per_irq	= 2,
 		.handle_mmio	= handle_mmio_cfg_reg,
 	},
 	{
@@ -850,6 +1002,22 @@ struct mmio_range *find_matching_range(const struct mmio_range *ranges,
 	return NULL;
 }
 
+static bool vgic_validate_access(const struct vgic_dist *dist,
+				 const struct mmio_range *range,
+				 unsigned long offset)
+{
+	int irq;
+
+	if (!range->bits_per_irq)
+		return true;	/* Not an irq-based access */
+
+	irq = offset * 8 / range->bits_per_irq;
+	if (irq >= dist->nr_irqs)
+		return false;
+
+	return true;
+}
+
 /**
  * vgic_handle_mmio - handle an in-kernel MMIO access
  * @vcpu:	pointer to the vcpu performing the access
@@ -889,7 +1057,13 @@ bool vgic_handle_mmio(struct kvm_vcpu *vcpu, struct kvm_run *run,
 
 	spin_lock(&vcpu->kvm->arch.vgic.lock);
 	offset = mmio->phys_addr - range->base - base;
-	updated_state = range->handle_mmio(vcpu, mmio, offset);
+	if (vgic_validate_access(dist, range, offset)) {
+		updated_state = range->handle_mmio(vcpu, mmio, offset);
+	} else {
+		vgic_reg_access(mmio, NULL, offset,
+				ACCESS_READ_RAZ | ACCESS_WRITE_IGNORED);
+		updated_state = false;
+	}
 	spin_unlock(&vcpu->kvm->arch.vgic.lock);
 	kvm_prepare_mmio(run, mmio);
 	kvm_handle_mmio_return(vcpu, run);
@@ -900,6 +1074,11 @@ bool vgic_handle_mmio(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	return true;
 }
 
+static u8 *vgic_get_sgi_sources(struct vgic_dist *dist, int vcpu_id, int sgi)
+{
+	return dist->irq_sgi_sources + vcpu_id * VGIC_NR_SGIS + sgi;
+}
+
 static void vgic_dispatch_sgi(struct kvm_vcpu *vcpu, u32 reg)
 {
 	struct kvm *kvm = vcpu->kvm;
@@ -932,8 +1111,8 @@ static void vgic_dispatch_sgi(struct kvm_vcpu *vcpu, u32 reg)
 	kvm_for_each_vcpu(c, vcpu, kvm) {
 		if (target_cpus & 1) {
 			/* Flag the SGI as pending */
-			vgic_dist_irq_set(vcpu, sgi);
-			dist->irq_sgi_sources[c][sgi] |= 1 << vcpu_id;
+			vgic_dist_irq_set_pending(vcpu, sgi);
+			*vgic_get_sgi_sources(dist, c, sgi) |= 1 << vcpu_id;
 			kvm_debug("SGI%d from CPU%d to CPU%d\n", sgi, vcpu_id, c);
 		}
 
@@ -941,32 +1120,38 @@ static void vgic_dispatch_sgi(struct kvm_vcpu *vcpu, u32 reg)
 	}
 }
 
+static int vgic_nr_shared_irqs(struct vgic_dist *dist)
+{
+	return dist->nr_irqs - VGIC_NR_PRIVATE_IRQS;
+}
+
 static int compute_pending_for_cpu(struct kvm_vcpu *vcpu)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 	unsigned long *pending, *enabled, *pend_percpu, *pend_shared;
 	unsigned long pending_private, pending_shared;
+	int nr_shared = vgic_nr_shared_irqs(dist);
 	int vcpu_id;
 
 	vcpu_id = vcpu->vcpu_id;
 	pend_percpu = vcpu->arch.vgic_cpu.pending_percpu;
 	pend_shared = vcpu->arch.vgic_cpu.pending_shared;
 
-	pending = vgic_bitmap_get_cpu_map(&dist->irq_state, vcpu_id);
+	pending = vgic_bitmap_get_cpu_map(&dist->irq_pending, vcpu_id);
 	enabled = vgic_bitmap_get_cpu_map(&dist->irq_enabled, vcpu_id);
 	bitmap_and(pend_percpu, pending, enabled, VGIC_NR_PRIVATE_IRQS);
 
-	pending = vgic_bitmap_get_shared_map(&dist->irq_state);
+	pending = vgic_bitmap_get_shared_map(&dist->irq_pending);
 	enabled = vgic_bitmap_get_shared_map(&dist->irq_enabled);
-	bitmap_and(pend_shared, pending, enabled, VGIC_NR_SHARED_IRQS);
+	bitmap_and(pend_shared, pending, enabled, nr_shared);
 	bitmap_and(pend_shared, pend_shared,
 		   vgic_bitmap_get_shared_map(&dist->irq_spi_target[vcpu_id]),
-		   VGIC_NR_SHARED_IRQS);
+		   nr_shared);
 
 	pending_private = find_first_bit(pend_percpu, VGIC_NR_PRIVATE_IRQS);
-	pending_shared = find_first_bit(pend_shared, VGIC_NR_SHARED_IRQS);
+	pending_shared = find_first_bit(pend_shared, nr_shared);
 	return (pending_private < VGIC_NR_PRIVATE_IRQS ||
-		pending_shared < VGIC_NR_SHARED_IRQS);
+		pending_shared < vgic_nr_shared_irqs(dist));
 }
 
 /*
@@ -980,14 +1165,14 @@ static void vgic_update_state(struct kvm *kvm)
 	int c;
 
 	if (!dist->enabled) {
-		set_bit(0, &dist->irq_pending_on_cpu);
+		set_bit(0, dist->irq_pending_on_cpu);
 		return;
 	}
 
 	kvm_for_each_vcpu(c, vcpu, kvm) {
 		if (compute_pending_for_cpu(vcpu)) {
 			pr_debug("CPU%d has pending interrupts\n", c);
-			set_bit(c, &dist->irq_pending_on_cpu);
+			set_bit(c, dist->irq_pending_on_cpu);
 		}
 	}
 }
@@ -1079,8 +1264,8 @@ static void vgic_retire_disabled_irqs(struct kvm_vcpu *vcpu)
 
 		if (!vgic_irq_is_enabled(vcpu, vlr.irq)) {
 			vgic_retire_lr(lr, vlr.irq, vcpu);
-			if (vgic_irq_is_active(vcpu, vlr.irq))
-				vgic_irq_clear_active(vcpu, vlr.irq);
+			if (vgic_irq_is_queued(vcpu, vlr.irq))
+				vgic_irq_clear_queued(vcpu, vlr.irq);
 		}
 	}
 }
@@ -1092,13 +1277,14 @@ static void vgic_retire_disabled_irqs(struct kvm_vcpu *vcpu)
 static bool vgic_queue_irq(struct kvm_vcpu *vcpu, u8 sgi_source_id, int irq)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 	struct vgic_lr vlr;
 	int lr;
 
 	/* Sanitize the input... */
 	BUG_ON(sgi_source_id & ~7);
 	BUG_ON(sgi_source_id && irq >= VGIC_NR_SGIS);
-	BUG_ON(irq >= VGIC_NR_IRQS);
+	BUG_ON(irq >= dist->nr_irqs);
 
 	kvm_debug("Queue IRQ%d\n", irq);
 
@@ -1144,14 +1330,14 @@ static bool vgic_queue_sgi(struct kvm_vcpu *vcpu, int irq)
 	int vcpu_id = vcpu->vcpu_id;
 	int c;
 
-	sources = dist->irq_sgi_sources[vcpu_id][irq];
+	sources = *vgic_get_sgi_sources(dist, vcpu_id, irq);
 
-	for_each_set_bit(c, &sources, VGIC_MAX_CPUS) {
+	for_each_set_bit(c, &sources, dist->nr_cpus) {
 		if (vgic_queue_irq(vcpu, c, irq))
 			clear_bit(c, &sources);
 	}
 
-	dist->irq_sgi_sources[vcpu_id][irq] = sources;
+	*vgic_get_sgi_sources(dist, vcpu_id, irq) = sources;
 
 	/*
 	 * If the sources bitmap has been cleared it means that we
@@ -1160,7 +1346,7 @@ static bool vgic_queue_sgi(struct kvm_vcpu *vcpu, int irq)
 	 * our emulated gic and can get rid of them.
 	 */
 	if (!sources) {
-		vgic_dist_irq_clear(vcpu, irq);
+		vgic_dist_irq_clear_pending(vcpu, irq);
 		vgic_cpu_irq_clear(vcpu, irq);
 		return true;
 	}
@@ -1170,15 +1356,15 @@ static bool vgic_queue_sgi(struct kvm_vcpu *vcpu, int irq)
 
 static bool vgic_queue_hwirq(struct kvm_vcpu *vcpu, int irq)
 {
-	if (vgic_irq_is_active(vcpu, irq))
+	if (!vgic_can_sample_irq(vcpu, irq))
 		return true; /* level interrupt, already queued */
 
 	if (vgic_queue_irq(vcpu, 0, irq)) {
 		if (vgic_irq_is_edge(vcpu, irq)) {
-			vgic_dist_irq_clear(vcpu, irq);
+			vgic_dist_irq_clear_pending(vcpu, irq);
 			vgic_cpu_irq_clear(vcpu, irq);
 		} else {
-			vgic_irq_set_active(vcpu, irq);
+			vgic_irq_set_queued(vcpu, irq);
 		}
 
 		return true;
@@ -1223,7 +1409,7 @@ static void __kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu)
 	}
 
 	/* SPIs */
-	for_each_set_bit(i, vgic_cpu->pending_shared, VGIC_NR_SHARED_IRQS) {
+	for_each_set_bit(i, vgic_cpu->pending_shared, vgic_nr_shared_irqs(dist)) {
 		if (!vgic_queue_hwirq(vcpu, i + VGIC_NR_PRIVATE_IRQS))
 			overflow = 1;
 	}
@@ -1239,7 +1425,7 @@ epilog:
 		 * us. Claim we don't have anything pending. We'll
 		 * adjust that if needed while exiting.
 		 */
-		clear_bit(vcpu_id, &dist->irq_pending_on_cpu);
+		clear_bit(vcpu_id, dist->irq_pending_on_cpu);
 	}
 }
 
@@ -1261,17 +1447,32 @@ static bool vgic_process_maintenance(struct kvm_vcpu *vcpu)
 
 		for_each_set_bit(lr, eisr_ptr, vgic->nr_lr) {
 			struct vgic_lr vlr = vgic_get_lr(vcpu, lr);
+			WARN_ON(vgic_irq_is_edge(vcpu, vlr.irq));
 
-			vgic_irq_clear_active(vcpu, vlr.irq);
+			vgic_irq_clear_queued(vcpu, vlr.irq);
 			WARN_ON(vlr.state & LR_STATE_MASK);
 			vlr.state = 0;
 			vgic_set_lr(vcpu, lr, vlr);
 
+			/*
+			 * If the IRQ was EOIed it was also ACKed and we we
+			 * therefore assume we can clear the soft pending
+			 * state (should it had been set) for this interrupt.
+			 *
+			 * Note: if the IRQ soft pending state was set after
+			 * the IRQ was acked, it actually shouldn't be
+			 * cleared, but we have no way of knowing that unless
+			 * we start trapping ACKs when the soft-pending state
+			 * is set.
+			 */
+			vgic_dist_irq_clear_soft_pend(vcpu, vlr.irq);
+
 			/* Any additional pending interrupt? */
-			if (vgic_dist_irq_is_pending(vcpu, vlr.irq)) {
+			if (vgic_dist_irq_get_level(vcpu, vlr.irq)) {
 				vgic_cpu_irq_set(vcpu, vlr.irq);
 				level_pending = true;
 			} else {
+				vgic_dist_irq_clear_pending(vcpu, vlr.irq);
 				vgic_cpu_irq_clear(vcpu, vlr.irq);
 			}
 
@@ -1315,14 +1516,14 @@ static void __kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu)
 
 		vlr = vgic_get_lr(vcpu, lr);
 
-		BUG_ON(vlr.irq >= VGIC_NR_IRQS);
+		BUG_ON(vlr.irq >= dist->nr_irqs);
 		vgic_cpu->vgic_irq_lr_map[vlr.irq] = LR_EMPTY;
 	}
 
 	/* Check if we still have something up our sleeve... */
 	pending = find_first_zero_bit(elrsr_ptr, vgic->nr_lr);
 	if (level_pending || pending < vgic->nr_lr)
-		set_bit(vcpu->vcpu_id, &dist->irq_pending_on_cpu);
+		set_bit(vcpu->vcpu_id, dist->irq_pending_on_cpu);
 }
 
 void kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu)
@@ -1356,7 +1557,7 @@ int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu)
 	if (!irqchip_in_kernel(vcpu->kvm))
 		return 0;
 
-	return test_bit(vcpu->vcpu_id, &dist->irq_pending_on_cpu);
+	return test_bit(vcpu->vcpu_id, dist->irq_pending_on_cpu);
 }
 
 static void vgic_kick_vcpus(struct kvm *kvm)
@@ -1376,34 +1577,36 @@ static void vgic_kick_vcpus(struct kvm *kvm)
 
 static int vgic_validate_injection(struct kvm_vcpu *vcpu, int irq, int level)
 {
-	int is_edge = vgic_irq_is_edge(vcpu, irq);
-	int state = vgic_dist_irq_is_pending(vcpu, irq);
+	int edge_triggered = vgic_irq_is_edge(vcpu, irq);
 
 	/*
 	 * Only inject an interrupt if:
 	 * - edge triggered and we have a rising edge
 	 * - level triggered and we change level
 	 */
-	if (is_edge)
+	if (edge_triggered) {
+		int state = vgic_dist_irq_is_pending(vcpu, irq);
 		return level > state;
-	else
+	} else {
+		int state = vgic_dist_irq_get_level(vcpu, irq);
 		return level != state;
+	}
 }
 
-static bool vgic_update_irq_state(struct kvm *kvm, int cpuid,
+static bool vgic_update_irq_pending(struct kvm *kvm, int cpuid,
 				  unsigned int irq_num, bool level)
 {
 	struct vgic_dist *dist = &kvm->arch.vgic;
 	struct kvm_vcpu *vcpu;
-	int is_edge, is_level;
+	int edge_triggered, level_triggered;
 	int enabled;
 	bool ret = true;
 
 	spin_lock(&dist->lock);
 
 	vcpu = kvm_get_vcpu(kvm, cpuid);
-	is_edge = vgic_irq_is_edge(vcpu, irq_num);
-	is_level = !is_edge;
+	edge_triggered = vgic_irq_is_edge(vcpu, irq_num);
+	level_triggered = !edge_triggered;
 
 	if (!vgic_validate_injection(vcpu, irq_num, level)) {
 		ret = false;
@@ -1417,10 +1620,19 @@ static bool vgic_update_irq_state(struct kvm *kvm, int cpuid,
 
 	kvm_debug("Inject IRQ%d level %d CPU%d\n", irq_num, level, cpuid);
 
-	if (level)
-		vgic_dist_irq_set(vcpu, irq_num);
-	else
-		vgic_dist_irq_clear(vcpu, irq_num);
+	if (level) {
+		if (level_triggered)
+			vgic_dist_irq_set_level(vcpu, irq_num);
+		vgic_dist_irq_set_pending(vcpu, irq_num);
+	} else {
+		if (level_triggered) {
+			vgic_dist_irq_clear_level(vcpu, irq_num);
+			if (!vgic_dist_irq_soft_pend(vcpu, irq_num))
+				vgic_dist_irq_clear_pending(vcpu, irq_num);
+		} else {
+			vgic_dist_irq_clear_pending(vcpu, irq_num);
+		}
+	}
 
 	enabled = vgic_irq_is_enabled(vcpu, irq_num);
 
@@ -1429,7 +1641,7 @@ static bool vgic_update_irq_state(struct kvm *kvm, int cpuid,
 		goto out;
 	}
 
-	if (is_level && vgic_irq_is_active(vcpu, irq_num)) {
+	if (!vgic_can_sample_irq(vcpu, irq_num)) {
 		/*
 		 * Level interrupt in progress, will be picked up
 		 * when EOId.
@@ -1440,7 +1652,7 @@ static bool vgic_update_irq_state(struct kvm *kvm, int cpuid,
 
 	if (level) {
 		vgic_cpu_irq_set(vcpu, irq_num);
-		set_bit(cpuid, &dist->irq_pending_on_cpu);
+		set_bit(cpuid, dist->irq_pending_on_cpu);
 	}
 
 out:
@@ -1466,7 +1678,8 @@ out:
 int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int irq_num,
 			bool level)
 {
-	if (vgic_update_irq_state(kvm, cpuid, irq_num, level))
+	if (likely(vgic_initialized(kvm)) &&
+	    vgic_update_irq_pending(kvm, cpuid, irq_num, level))
 		vgic_kick_vcpus(kvm);
 
 	return 0;
@@ -1483,6 +1696,32 @@ static irqreturn_t vgic_maintenance_handler(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
+void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	kfree(vgic_cpu->pending_shared);
+	kfree(vgic_cpu->vgic_irq_lr_map);
+	vgic_cpu->pending_shared = NULL;
+	vgic_cpu->vgic_irq_lr_map = NULL;
+}
+
+static int vgic_vcpu_init_maps(struct kvm_vcpu *vcpu, int nr_irqs)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	int sz = (nr_irqs - VGIC_NR_PRIVATE_IRQS) / 8;
+	vgic_cpu->pending_shared = kzalloc(sz, GFP_KERNEL);
+	vgic_cpu->vgic_irq_lr_map = kzalloc(nr_irqs, GFP_KERNEL);
+
+	if (!vgic_cpu->pending_shared || !vgic_cpu->vgic_irq_lr_map) {
+		kvm_vgic_vcpu_destroy(vcpu);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
 /**
  * kvm_vgic_vcpu_init - Initialize per-vcpu VGIC state
  * @vcpu: pointer to the vcpu struct
@@ -1490,16 +1729,13 @@ static irqreturn_t vgic_maintenance_handler(int irq, void *data)
  * Initialize the vgic_cpu struct and vgic_dist struct fields pertaining to
  * this vcpu and enable the VGIC for this VCPU
  */
-int kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu)
+static void kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
 	int i;
 
-	if (vcpu->vcpu_id >= VGIC_MAX_CPUS)
-		return -EBUSY;
-
-	for (i = 0; i < VGIC_NR_IRQS; i++) {
+	for (i = 0; i < dist->nr_irqs; i++) {
 		if (i < VGIC_NR_PPIS)
 			vgic_bitmap_set_irq_val(&dist->irq_enabled,
 						vcpu->vcpu_id, i, 1);
@@ -1518,84 +1754,112 @@ int kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu)
 	vgic_cpu->nr_lr = vgic->nr_lr;
 
 	vgic_enable(vcpu);
-
-	return 0;
 }
 
-static void vgic_init_maintenance_interrupt(void *info)
+void kvm_vgic_destroy(struct kvm *kvm)
 {
-	enable_percpu_irq(vgic->maint_irq, 0);
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct kvm_vcpu *vcpu;
+	int i;
+
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_vgic_vcpu_destroy(vcpu);
+
+	vgic_free_bitmap(&dist->irq_enabled);
+	vgic_free_bitmap(&dist->irq_level);
+	vgic_free_bitmap(&dist->irq_pending);
+	vgic_free_bitmap(&dist->irq_soft_pend);
+	vgic_free_bitmap(&dist->irq_queued);
+	vgic_free_bitmap(&dist->irq_cfg);
+	vgic_free_bytemap(&dist->irq_priority);
+	if (dist->irq_spi_target) {
+		for (i = 0; i < dist->nr_cpus; i++)
+			vgic_free_bitmap(&dist->irq_spi_target[i]);
+	}
+	kfree(dist->irq_sgi_sources);
+	kfree(dist->irq_spi_cpu);
+	kfree(dist->irq_spi_target);
+	kfree(dist->irq_pending_on_cpu);
+	dist->irq_sgi_sources = NULL;
+	dist->irq_spi_cpu = NULL;
+	dist->irq_spi_target = NULL;
+	dist->irq_pending_on_cpu = NULL;
 }
 
-static int vgic_cpu_notify(struct notifier_block *self,
-			   unsigned long action, void *cpu)
+/*
+ * Allocate and initialize the various data structures. Must be called
+ * with kvm->lock held!
+ */
+static int vgic_init_maps(struct kvm *kvm)
 {
-	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;
-	}
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct kvm_vcpu *vcpu;
+	int nr_cpus, nr_irqs;
+	int ret, i;
 
-	return NOTIFY_OK;
-}
+	if (dist->nr_cpus)	/* Already allocated */
+		return 0;
 
-static struct notifier_block vgic_cpu_nb = {
-	.notifier_call = vgic_cpu_notify,
-};
+	nr_cpus = dist->nr_cpus = atomic_read(&kvm->online_vcpus);
+	if (!nr_cpus)		/* No vcpus? Can't be good... */
+		return -EINVAL;
 
-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, },
-	{},
-};
+	/*
+	 * If nobody configured the number of interrupts, use the
+	 * legacy one.
+	 */
+	if (!dist->nr_irqs)
+		dist->nr_irqs = VGIC_NR_IRQS_LEGACY;
 
-int kvm_vgic_hyp_init(void)
-{
-	const struct of_device_id *matched_id;
-	int (*vgic_probe)(struct device_node *,const struct vgic_ops **,
-			  const struct vgic_params **);
-	struct device_node *vgic_node;
-	int ret;
+	nr_irqs = dist->nr_irqs;
 
-	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;
-	}
+	ret  = vgic_init_bitmap(&dist->irq_enabled, nr_cpus, nr_irqs);
+	ret |= vgic_init_bitmap(&dist->irq_level, nr_cpus, nr_irqs);
+	ret |= vgic_init_bitmap(&dist->irq_pending, nr_cpus, nr_irqs);
+	ret |= vgic_init_bitmap(&dist->irq_soft_pend, nr_cpus, nr_irqs);
+	ret |= vgic_init_bitmap(&dist->irq_queued, nr_cpus, nr_irqs);
+	ret |= vgic_init_bitmap(&dist->irq_cfg, nr_cpus, nr_irqs);
+	ret |= vgic_init_bytemap(&dist->irq_priority, nr_cpus, nr_irqs);
 
-	vgic_probe = matched_id->data;
-	ret = vgic_probe(vgic_node, &vgic_ops, &vgic);
 	if (ret)
-		return ret;
+		goto out;
 
-	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;
+	dist->irq_sgi_sources = kzalloc(nr_cpus * VGIC_NR_SGIS, GFP_KERNEL);
+	dist->irq_spi_cpu = kzalloc(nr_irqs - VGIC_NR_PRIVATE_IRQS, GFP_KERNEL);
+	dist->irq_spi_target = kzalloc(sizeof(*dist->irq_spi_target) * nr_cpus,
+				       GFP_KERNEL);
+	dist->irq_pending_on_cpu = kzalloc(BITS_TO_LONGS(nr_cpus) * sizeof(long),
+					   GFP_KERNEL);
+	if (!dist->irq_sgi_sources ||
+	    !dist->irq_spi_cpu ||
+	    !dist->irq_spi_target ||
+	    !dist->irq_pending_on_cpu) {
+		ret = -ENOMEM;
+		goto out;
 	}
 
-	ret = __register_cpu_notifier(&vgic_cpu_nb);
-	if (ret) {
-		kvm_err("Cannot register vgic CPU notifier\n");
-		goto out_free_irq;
-	}
+	for (i = 0; i < nr_cpus; i++)
+		ret |= vgic_init_bitmap(&dist->irq_spi_target[i],
+					nr_cpus, nr_irqs);
 
-	/* Callback into for arch code for setup */
-	vgic_arch_setup(vgic);
+	if (ret)
+		goto out;
 
-	on_each_cpu(vgic_init_maintenance_interrupt, NULL, 1);
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		ret = vgic_vcpu_init_maps(vcpu, nr_irqs);
+		if (ret) {
+			kvm_err("VGIC: Failed to allocate vcpu memory\n");
+			break;
+		}
+	}
 
-	return 0;
+	for (i = VGIC_NR_PRIVATE_IRQS; i < dist->nr_irqs; i += 4)
+		vgic_set_target_reg(kvm, 0, i);
+
+out:
+	if (ret)
+		kvm_vgic_destroy(kvm);
 
-out_free_irq:
-	free_percpu_irq(vgic->maint_irq, kvm_get_running_vcpus());
 	return ret;
 }
 
@@ -1610,6 +1874,7 @@ out_free_irq:
  */
 int kvm_vgic_init(struct kvm *kvm)
 {
+	struct kvm_vcpu *vcpu;
 	int ret = 0, i;
 
 	if (!irqchip_in_kernel(kvm))
@@ -1627,6 +1892,12 @@ int kvm_vgic_init(struct kvm *kvm)
 		goto out;
 	}
 
+	ret = vgic_init_maps(kvm);
+	if (ret) {
+		kvm_err("Unable to allocate maps\n");
+		goto out;
+	}
+
 	ret = kvm_phys_addr_ioremap(kvm, kvm->arch.vgic.vgic_cpu_base,
 				    vgic->vcpu_base, KVM_VGIC_V2_CPU_SIZE);
 	if (ret) {
@@ -1634,11 +1905,13 @@ int kvm_vgic_init(struct kvm *kvm)
 		goto out;
 	}
 
-	for (i = VGIC_NR_PRIVATE_IRQS; i < VGIC_NR_IRQS; i += 4)
-		vgic_set_target_reg(kvm, 0, i);
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_vgic_vcpu_init(vcpu);
 
 	kvm->arch.vgic.ready = true;
 out:
+	if (ret)
+		kvm_vgic_destroy(kvm);
 	mutex_unlock(&kvm->lock);
 	return ret;
 }
@@ -1690,7 +1963,7 @@ out:
 	return ret;
 }
 
-static bool vgic_ioaddr_overlap(struct kvm *kvm)
+static int vgic_ioaddr_overlap(struct kvm *kvm)
 {
 	phys_addr_t dist = kvm->arch.vgic.vgic_dist_base;
 	phys_addr_t cpu = kvm->arch.vgic.vgic_cpu_base;
@@ -1879,6 +2152,10 @@ static int vgic_attr_regs_access(struct kvm_device *dev,
 
 	mutex_lock(&dev->kvm->lock);
 
+	ret = vgic_init_maps(dev->kvm);
+	if (ret)
+		goto out;
+
 	if (cpuid >= atomic_read(&dev->kvm->online_vcpus)) {
 		ret = -EINVAL;
 		goto out;
@@ -1976,6 +2253,36 @@ static int vgic_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 
 		return vgic_attr_regs_access(dev, attr, &reg, true);
 	}
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u32 val;
+		int ret = 0;
+
+		if (get_user(val, uaddr))
+			return -EFAULT;
+
+		/*
+		 * We require:
+		 * - at least 32 SPIs on top of the 16 SGIs and 16 PPIs
+		 * - at most 1024 interrupts
+		 * - a multiple of 32 interrupts
+		 */
+		if (val < (VGIC_NR_PRIVATE_IRQS + 32) ||
+		    val > VGIC_MAX_IRQS ||
+		    (val & 31))
+			return -EINVAL;
+
+		mutex_lock(&dev->kvm->lock);
+
+		if (vgic_initialized(dev->kvm) || dev->kvm->arch.vgic.nr_irqs)
+			ret = -EBUSY;
+		else
+			dev->kvm->arch.vgic.nr_irqs = val;
+
+		mutex_unlock(&dev->kvm->lock);
+
+		return ret;
+	}
 
 	}
 
@@ -2012,6 +2319,11 @@ static int vgic_get_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 		r = put_user(reg, uaddr);
 		break;
 	}
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		r = put_user(dev->kvm->arch.vgic.nr_irqs, uaddr);
+		break;
+	}
 
 	}
 
@@ -2048,6 +2360,8 @@ static int vgic_has_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 	case KVM_DEV_ARM_VGIC_GRP_CPU_REGS:
 		offset = attr->attr & KVM_DEV_ARM_VGIC_OFFSET_MASK;
 		return vgic_has_attr_regs(vgic_cpu_ranges, offset);
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS:
+		return 0;
 	}
 	return -ENXIO;
 }
@@ -2062,7 +2376,7 @@ 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 +2384,81 @@ 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/async_pf.c

@@ -80,9 +80,7 @@ static void async_pf_execute(struct work_struct *work)
 
 	might_sleep();
 
-	down_read(&mm->mmap_sem);
-	get_user_pages(NULL, mm, addr, 1, 1, 0, NULL, NULL);
-	up_read(&mm->mmap_sem);
+	kvm_get_user_page_io(NULL, mm, addr, 1, NULL);
 	kvm_async_page_present_sync(vcpu, apf);
 
 	spin_lock(&vcpu->async_pf.lock);

virt/kvm/eventfd.c

@@ -36,7 +36,9 @@
 #include <linux/seqlock.h>
 #include <trace/events/kvm.h>
 
-#include "irq.h"
+#ifdef __KVM_HAVE_IOAPIC
+#include "ioapic.h"
+#endif
 #include "iodev.h"
 
 #ifdef CONFIG_HAVE_KVM_IRQFD

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

@@ -52,11 +52,13 @@
 
 #include <asm/processor.h>
 #include <asm/io.h>
+#include <asm/ioctl.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
 #include "coalesced_mmio.h"
 #include "async_pf.h"
+#include "vfio.h"
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/kvm.h>
@@ -95,8 +97,6 @@ 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,
@@ -129,7 +129,8 @@ int vcpu_load(struct kvm_vcpu *vcpu)
 		struct pid *oldpid = vcpu->pid;
 		struct pid *newpid = get_task_pid(current, PIDTYPE_PID);
 		rcu_assign_pointer(vcpu->pid, newpid);
-		synchronize_rcu();
+		if (oldpid)
+			synchronize_rcu();
 		put_pid(oldpid);
 	}
 	cpu = get_cpu();
@@ -152,7 +153,7 @@ static void ack_flush(void *_completed)
 {
 }
 
-static bool make_all_cpus_request(struct kvm *kvm, unsigned int req)
+bool kvm_make_all_cpus_request(struct kvm *kvm, unsigned int req)
 {
 	int i, cpu, me;
 	cpumask_var_t cpus;
@@ -189,7 +190,7 @@ void kvm_flush_remote_tlbs(struct kvm *kvm)
 	long dirty_count = kvm->tlbs_dirty;
 
 	smp_mb();
-	if (make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH))
+	if (kvm_make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH))
 		++kvm->stat.remote_tlb_flush;
 	cmpxchg(&kvm->tlbs_dirty, dirty_count, 0);
 }
@@ -197,17 +198,17 @@ EXPORT_SYMBOL_GPL(kvm_flush_remote_tlbs);
 
 void kvm_reload_remote_mmus(struct kvm *kvm)
 {
-	make_all_cpus_request(kvm, KVM_REQ_MMU_RELOAD);
+	kvm_make_all_cpus_request(kvm, KVM_REQ_MMU_RELOAD);
 }
 
 void kvm_make_mclock_inprogress_request(struct kvm *kvm)
 {
-	make_all_cpus_request(kvm, KVM_REQ_MCLOCK_INPROGRESS);
+	kvm_make_all_cpus_request(kvm, KVM_REQ_MCLOCK_INPROGRESS);
 }
 
 void kvm_make_scan_ioapic_request(struct kvm *kvm)
 {
-	make_all_cpus_request(kvm, KVM_REQ_SCAN_IOAPIC);
+	kvm_make_all_cpus_request(kvm, KVM_REQ_SCAN_IOAPIC);
 }
 
 int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
@@ -295,6 +296,9 @@ static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn,
 		kvm_flush_remote_tlbs(kvm);
 
 	spin_unlock(&kvm->mmu_lock);
+
+	kvm_arch_mmu_notifier_invalidate_page(kvm, address);
+
 	srcu_read_unlock(&kvm->srcu, idx);
 }
 
@@ -368,7 +372,8 @@ static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
 
 static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn,
 					      struct mm_struct *mm,
-					      unsigned long address)
+					      unsigned long start,
+					      unsigned long end)
 {
 	struct kvm *kvm = mmu_notifier_to_kvm(mn);
 	int young, idx;
@@ -376,7 +381,7 @@ static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn,
 	idx = srcu_read_lock(&kvm->srcu);
 	spin_lock(&kvm->mmu_lock);
 
-	young = kvm_age_hva(kvm, address);
+	young = kvm_age_hva(kvm, start, end);
 	if (young)
 		kvm_flush_remote_tlbs(kvm);
 
@@ -476,6 +481,13 @@ 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;
@@ -687,8 +699,7 @@ 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;
@@ -699,15 +710,13 @@ 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
 
@@ -722,10 +731,24 @@ 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;
@@ -776,7 +799,6 @@ 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;
 
@@ -790,7 +812,6 @@ 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;
@@ -846,7 +867,6 @@ 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)
@@ -1075,9 +1095,9 @@ EXPORT_SYMBOL_GPL(gfn_to_hva);
  * If writable is set to false, the hva returned by this function is only
  * allowed to be read.
  */
-unsigned long gfn_to_hva_prot(struct kvm *kvm, gfn_t gfn, bool *writable)
+unsigned long gfn_to_hva_memslot_prot(struct kvm_memory_slot *slot,
+				      gfn_t gfn, bool *writable)
 {
-	struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn);
 	unsigned long hva = __gfn_to_hva_many(slot, gfn, NULL, false);
 
 	if (!kvm_is_error_hva(hva) && writable)
@@ -1086,6 +1106,13 @@ unsigned long gfn_to_hva_prot(struct kvm *kvm, gfn_t gfn, bool *writable)
 	return hva;
 }
 
+unsigned long gfn_to_hva_prot(struct kvm *kvm, gfn_t gfn, bool *writable)
+{
+	struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn);
+
+	return gfn_to_hva_memslot_prot(slot, gfn, writable);
+}
+
 static int kvm_read_hva(void *data, void __user *hva, int len)
 {
 	return __copy_from_user(data, hva, len);
@@ -1107,6 +1134,43 @@ static int get_user_page_nowait(struct task_struct *tsk, struct mm_struct *mm,
 	return __get_user_pages(tsk, mm, start, 1, flags, page, NULL, NULL);
 }
 
+int kvm_get_user_page_io(struct task_struct *tsk, struct mm_struct *mm,
+			 unsigned long addr, bool write_fault,
+			 struct page **pagep)
+{
+	int npages;
+	int locked = 1;
+	int flags = FOLL_TOUCH | FOLL_HWPOISON |
+		    (pagep ? FOLL_GET : 0) |
+		    (write_fault ? FOLL_WRITE : 0);
+
+	/*
+	 * If retrying the fault, we get here *not* having allowed the filemap
+	 * to wait on the page lock. We should now allow waiting on the IO with
+	 * the mmap semaphore released.
+	 */
+	down_read(&mm->mmap_sem);
+	npages = __get_user_pages(tsk, mm, addr, 1, flags, pagep, NULL,
+				  &locked);
+	if (!locked) {
+		VM_BUG_ON(npages);
+
+		if (!pagep)
+			return 0;
+
+		/*
+		 * The previous call has now waited on the IO. Now we can
+		 * retry and complete. Pass TRIED to ensure we do not re
+		 * schedule async IO (see e.g. filemap_fault).
+		 */
+		down_read(&mm->mmap_sem);
+		npages = __get_user_pages(tsk, mm, addr, 1, flags | FOLL_TRIED,
+					  pagep, NULL, NULL);
+	}
+	up_read(&mm->mmap_sem);
+	return npages;
+}
+
 static inline int check_user_page_hwpoison(unsigned long addr)
 {
 	int rc, flags = FOLL_TOUCH | FOLL_HWPOISON | FOLL_WRITE;
@@ -1169,9 +1233,15 @@ static int hva_to_pfn_slow(unsigned long addr, bool *async, bool write_fault,
 		npages = get_user_page_nowait(current, current->mm,
 					      addr, write_fault, page);
 		up_read(&current->mm->mmap_sem);
-	} else
-		npages = get_user_pages_fast(addr, 1, write_fault,
-					     page);
+	} else {
+		/*
+		 * By now we have tried gup_fast, and possibly async_pf, and we
+		 * are certainly not atomic. Time to retry the gup, allowing
+		 * mmap semaphore to be relinquished in the case of IO.
+		 */
+		npages = kvm_get_user_page_io(current, current->mm, addr,
+					      write_fault, page);
+	}
 	if (npages != 1)
 		return npages;
 
@@ -1768,8 +1838,7 @@ 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);
@@ -1975,6 +2044,9 @@ static long kvm_vcpu_ioctl(struct file *filp,
 	if (vcpu->kvm->mm != current->mm)
 		return -EIO;
 
+	if (unlikely(_IOC_TYPE(ioctl) != KVMIO))
+		return -EINVAL;
+
 #if defined(CONFIG_S390) || defined(CONFIG_PPC) || defined(CONFIG_MIPS)
 	/*
 	 * Special cases: vcpu ioctls that are asynchronous to vcpu execution,
@@ -2259,6 +2331,29 @@ 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)
 {
@@ -2267,36 +2362,12 @@ 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;
@@ -2611,7 +2682,6 @@ 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;
@@ -2623,7 +2693,6 @@ 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 */
@@ -2668,7 +2737,7 @@ 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);
@@ -2693,7 +2762,7 @@ 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)
@@ -3123,6 +3192,8 @@ static void kvm_sched_in(struct preempt_notifier *pn, int cpu)
 	if (vcpu->preempted)
 		vcpu->preempted = false;
 
+	kvm_arch_sched_in(vcpu, cpu);
+
 	kvm_arch_vcpu_load(vcpu, cpu);
 }
 
@@ -3214,6 +3285,9 @@ int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
 		goto out_undebugfs;
 	}
 
+	r = kvm_vfio_ops_init();
+	WARN_ON(r);
+
 	return 0;
 
 out_undebugfs:

virt/kvm/vfio.c

@@ -18,6 +18,7 @@
 #include <linux/slab.h>
 #include <linux/uaccess.h>
 #include <linux/vfio.h>
+#include "vfio.h"
 
 struct kvm_vfio_group {
 	struct list_head node;
@@ -246,6 +247,16 @@ 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 +279,7 @@ 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,
-};
+int kvm_vfio_ops_init(void)
+{
+	return kvm_register_device_ops(&kvm_vfio_ops, KVM_DEV_TYPE_VFIO);
+}