KVM/arm64 updates for Linux 5.8:

- Move the arch-specific code into arch/arm64/kvm
 - Start the post-32bit cleanup
 - Cherry-pick a few non-invasive pre-NV patches
 -----BEGIN PGP SIGNATURE-----
 
 iQJDBAABCgAtFiEEn9UcU+C1Yxj9lZw9I9DQutE9ekMFAl7RLp8PHG1hekBrZXJu
 ZWwub3JnAAoJECPQ0LrRPXpD/iAQAJOHsS1PT9y/Gefam5os9FqKpogj68e3rx9k
 XfPcweexBVqmDWSI4vmL9xHW2F7z4EwAE4dIDsTCKHpihK30+jH8l12tOJBz35yp
 MR1hYjv43F54xzKkkuP4F4wo3Ygg4ipjHZPReGkaGj1QOQs6N/YKa1aSSYfzkzCz
 VLCSqPQz45CkGPYEGwuPn13AjHqGQAwPhteJNAoCxViw1KAldmoqDk6kbKB+b+7a
 2oIvxiTZejICsgSX6UvqQYNG52AyZ/5Daq8iraaigQ8sGyKr+/2Yi+3RUUH6p7ns
 aCsictk+RS3BzMAKDw6MPYc7OhJBhxQEV1pdiPpt0tpS4L9LNmBagKzlaBKZhwdr
 dYDAjOlbgZZUJpKnlBAipuVlQySHdm2WjXr4msdY69D7OGxmkzU/zkSIokqdA2hr
 MuL5W1v2Z1UpxyVltb+c/4lPcFZNnRI0Mz1WcvliEojlf2zzKYMcBAl3bTiAuil5
 aTT2+1G0OSCfUfr8Zart4LoAHeczw4zG/Pern+hl92eMXUlX3pIcqzQaLtVmmEE/
 ecPShMowKsXOOGGp/T8Q04N1fr6KzmufP5+kgJDFZfo6iJ6r5uQ9G8nuLmp3wQOX
 c9mNCwdSxrFBTJ10KfLHquKqwfl18VXzKDx1pzO5nSupmKWfWZ5YFO8j2709e83x
 R42MqKEG
 =aD+9
 -----END PGP SIGNATURE-----

Merge tag 'kvmarm-5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 updates for Linux 5.8:

- Move the arch-specific code into arch/arm64/kvm
- Start the post-32bit cleanup
- Cherry-pick a few non-invasive pre-NV patches

Documentation/virt/kvm/api.rst

@@ -5799,7 +5799,7 @@ will be initialized to 1 when created.  This also improves performance because
 dirty logging can be enabled gradually in small chunks on the first call
 to KVM_CLEAR_DIRTY_LOG.  KVM_DIRTY_LOG_INITIALLY_SET depends on
 KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on
-x86 for now).
+x86 and arm64 for now).
 
 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name
 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make

MAINTAINERS

@@ -9295,7 +9295,6 @@ F:	arch/arm64/include/asm/kvm*
 F:	arch/arm64/include/uapi/asm/kvm*
 F:	arch/arm64/kvm/
 F:	include/kvm/arm_*
-F:	virt/kvm/arm/
 
 KERNEL VIRTUAL MACHINE FOR MIPS (KVM/mips)
 L:	linux-mips@vger.kernel.org

arch/arm64/include/asm/kvm_asm.h

@@ -64,12 +64,14 @@ extern void __kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa);
 extern void __kvm_tlb_flush_vmid(struct kvm *kvm);
 extern void __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu);
 
-extern void __kvm_timer_set_cntvoff(u32 cntvoff_low, u32 cntvoff_high);
+extern void __kvm_timer_set_cntvoff(u64 cntvoff);
 
 extern int kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu);
 
 extern int __kvm_vcpu_run_nvhe(struct kvm_vcpu *vcpu);
 
+extern void __kvm_enable_ssbs(void);
+
 extern u64 __vgic_v3_get_ich_vtr_el2(void);
 extern u64 __vgic_v3_read_vmcr(void);
 extern void __vgic_v3_write_vmcr(u32 vmcr);

arch/arm64/include/asm/kvm_host.h

@@ -46,6 +46,9 @@
 #define KVM_REQ_RECORD_STEAL	KVM_ARCH_REQ(3)
 #define KVM_REQ_RELOAD_GICv4	KVM_ARCH_REQ(4)
 
+#define KVM_DIRTY_LOG_MANUAL_CAPS   (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \
+				     KVM_DIRTY_LOG_INITIALLY_SET)
+
 DECLARE_STATIC_KEY_FALSE(userspace_irqchip_in_use);
 
 extern unsigned int kvm_sve_max_vl;
@@ -112,12 +115,8 @@ struct kvm_vcpu_fault_info {
 	u64 disr_el1;		/* Deferred [SError] Status Register */
 };
 
-/*
- * 0 is reserved as an invalid value.
- * Order should be kept in sync with the save/restore code.
- */
 enum vcpu_sysreg {
-	__INVALID_SYSREG__,
+	__INVALID_SYSREG__,   /* 0 is reserved as an invalid value */
 	MPIDR_EL1,	/* MultiProcessor Affinity Register */
 	CSSELR_EL1,	/* Cache Size Selection Register */
 	SCTLR_EL1,	/* System Control Register */
@@ -532,39 +531,6 @@ static inline void kvm_init_host_cpu_context(struct kvm_cpu_context *cpu_ctxt)
 	cpu_ctxt->sys_regs[MPIDR_EL1] = read_cpuid_mpidr();
 }
 
-void __kvm_enable_ssbs(void);
-
-static inline void __cpu_init_hyp_mode(phys_addr_t pgd_ptr,
-				       unsigned long hyp_stack_ptr,
-				       unsigned long vector_ptr)
-{
-	/*
-	 * Calculate the raw per-cpu offset without a translation from the
-	 * kernel's mapping to the linear mapping, and store it in tpidr_el2
-	 * so that we can use adr_l to access per-cpu variables in EL2.
-	 */
-	u64 tpidr_el2 = ((u64)this_cpu_ptr(&kvm_host_data) -
-			 (u64)kvm_ksym_ref(kvm_host_data));
-
-	/*
-	 * Call initialization code, and switch to the full blown HYP code.
-	 * If the cpucaps haven't been finalized yet, something has gone very
-	 * wrong, and hyp will crash and burn when it uses any
-	 * cpus_have_const_cap() wrapper.
-	 */
-	BUG_ON(!system_capabilities_finalized());
-	__kvm_call_hyp((void *)pgd_ptr, hyp_stack_ptr, vector_ptr, tpidr_el2);
-
-	/*
-	 * Disabling SSBD on a non-VHE system requires us to enable SSBS
-	 * at EL2.
-	 */
-	if (!has_vhe() && this_cpu_has_cap(ARM64_SSBS) &&
-	    arm64_get_ssbd_state() == ARM64_SSBD_FORCE_DISABLE) {
-		kvm_call_hyp(__kvm_enable_ssbs);
-	}
-}
-
 static inline bool kvm_arch_requires_vhe(void)
 {
 	/*
@@ -600,8 +566,6 @@ int kvm_arm_vcpu_arch_get_attr(struct kvm_vcpu *vcpu,
 int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
 			       struct kvm_device_attr *attr);
 
-static inline void __cpu_init_stage2(void) {}
-
 /* Guest/host FPSIMD coordination helpers */
 int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);

arch/arm64/include/asm/kvm_hyp.h

@@ -56,12 +56,12 @@
 
 int __vgic_v2_perform_cpuif_access(struct kvm_vcpu *vcpu);
 
-void __vgic_v3_save_state(struct kvm_vcpu *vcpu);
+void __vgic_v3_save_state(struct vgic_v3_cpu_if *cpu_if);
-void __vgic_v3_restore_state(struct kvm_vcpu *vcpu);
+void __vgic_v3_restore_state(struct vgic_v3_cpu_if *cpu_if);
-void __vgic_v3_activate_traps(struct kvm_vcpu *vcpu);
+void __vgic_v3_activate_traps(struct vgic_v3_cpu_if *cpu_if);
-void __vgic_v3_deactivate_traps(struct kvm_vcpu *vcpu);
+void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if);
-void __vgic_v3_save_aprs(struct kvm_vcpu *vcpu);
+void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if);
-void __vgic_v3_restore_aprs(struct kvm_vcpu *vcpu);
+void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if);
 int __vgic_v3_perform_cpuif_access(struct kvm_vcpu *vcpu);
 
 void __timer_enable_traps(struct kvm_vcpu *vcpu);

arch/arm64/include/asm/kvm_mmu.h

@@ -363,8 +363,6 @@ static inline void __kvm_flush_dcache_pud(pud_t pud)
 	}
 }
 
-#define kvm_virt_to_phys(x)		__pa_symbol(x)
-
 void kvm_set_way_flush(struct kvm_vcpu *vcpu);
 void kvm_toggle_cache(struct kvm_vcpu *vcpu, bool was_enabled);
 
@@ -473,7 +471,7 @@ static inline int kvm_write_guest_lock(struct kvm *kvm, gpa_t gpa,
 extern void *__kvm_bp_vect_base;
 extern int __kvm_harden_el2_vector_slot;
 
-/*  This is only called on a VHE system */
+/*  This is called on both VHE and !VHE systems */
 static inline void *kvm_get_hyp_vector(void)
 {
 	struct bp_hardening_data *data = arm64_get_bp_hardening_data();

arch/arm64/include/asm/ptrace.h

@@ -35,6 +35,7 @@
 #define GIC_PRIO_PSR_I_SET		(1 << 4)
 
 /* Additional SPSR bits not exposed in the UABI */
+#define PSR_MODE_THREAD_BIT	(1 << 0)
 #define PSR_IL_BIT		(1 << 20)
 
 /* AArch32-specific ptrace requests */

arch/arm64/include/asm/virt.h

@@ -85,7 +85,7 @@ static inline bool is_kernel_in_hyp_mode(void)
 
 static __always_inline bool has_vhe(void)
 {
-	if (cpus_have_const_cap(ARM64_HAS_VIRT_HOST_EXTN))
+	if (cpus_have_final_cap(ARM64_HAS_VIRT_HOST_EXTN))
 		return true;
 
 	return false;

arch/arm64/kernel/asm-offsets.c

@@ -96,7 +96,7 @@ int main(void)
   DEFINE(CPU_BOOT_PTRAUTH_KEY,	offsetof(struct secondary_data, ptrauth_key));
 #endif
   BLANK();
-#ifdef CONFIG_KVM_ARM_HOST
+#ifdef CONFIG_KVM
   DEFINE(VCPU_CONTEXT,		offsetof(struct kvm_vcpu, arch.ctxt));
   DEFINE(VCPU_FAULT_DISR,	offsetof(struct kvm_vcpu, arch.fault.disr_el1));
   DEFINE(VCPU_WORKAROUND_FLAGS,	offsetof(struct kvm_vcpu, arch.workaround_flags));

arch/arm64/kernel/cpu_errata.c

@@ -234,7 +234,7 @@ static int detect_harden_bp_fw(void)
 		smccc_end = NULL;
 		break;
 
-#if IS_ENABLED(CONFIG_KVM_ARM_HOST)
+#if IS_ENABLED(CONFIG_KVM)
 	case SMCCC_CONDUIT_SMC:
 		cb = call_smc_arch_workaround_1;
 		smccc_start = __smccc_workaround_1_smc;

arch/arm64/kernel/smp.c

@@ -430,7 +430,7 @@ static void __init hyp_mode_check(void)
 			   "CPU: CPUs started in inconsistent modes");
 	else
 		pr_info("CPU: All CPU(s) started at EL1\n");
-	if (IS_ENABLED(CONFIG_KVM_ARM_HOST))
+	if (IS_ENABLED(CONFIG_KVM))
 		kvm_compute_layout();
 }
 

arch/arm64/kvm/Kconfig

@@ -3,7 +3,6 @@
 # KVM configuration
 #
 
-source "virt/kvm/Kconfig"
 source "virt/lib/Kconfig"
 
 menuconfig VIRTUALIZATION
@@ -18,7 +17,7 @@ menuconfig VIRTUALIZATION
 
 if VIRTUALIZATION
 
-config KVM
+menuconfig KVM
 	bool "Kernel-based Virtual Machine (KVM) support"
 	depends on OF
 	# for TASKSTATS/TASK_DELAY_ACCT:
@@ -28,13 +27,11 @@ config KVM
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
 	select HAVE_KVM_ARCH_TLB_FLUSH_ALL
 	select KVM_MMIO
-	select KVM_ARM_HOST
 	select KVM_GENERIC_DIRTYLOG_READ_PROTECT
 	select SRCU
 	select KVM_VFIO
 	select HAVE_KVM_EVENTFD
 	select HAVE_KVM_IRQFD
-	select KVM_ARM_PMU if HW_PERF_EVENTS
 	select HAVE_KVM_MSI
 	select HAVE_KVM_IRQCHIP
 	select HAVE_KVM_IRQ_ROUTING
@@ -45,23 +42,24 @@ config KVM
 	select TASK_DELAY_ACCT
 	---help---
 	  Support hosting virtualized guest machines.
-	  We don't support KVM with 16K page tables yet, due to the multiple
-	  levels of fake page tables.
 
 	  If unsure, say N.
 
-config KVM_ARM_HOST
+if KVM
-	bool
+
-	---help---
+source "virt/kvm/Kconfig"
-	  Provides host support for ARM processors.
 
 config KVM_ARM_PMU
-	bool
+	bool "Virtual Performance Monitoring Unit (PMU) support"
+	depends on HW_PERF_EVENTS
+	default y
 	---help---
 	  Adds support for a virtual Performance Monitoring Unit (PMU) in
 	  virtual machines.
 
 config KVM_INDIRECT_VECTORS
-       def_bool KVM && (HARDEN_BRANCH_PREDICTOR || HARDEN_EL2_VECTORS)
+	def_bool HARDEN_BRANCH_PREDICTOR || HARDEN_EL2_VECTORS
+
+endif # KVM
 
 endif # VIRTUALIZATION

arch/arm64/kvm/Makefile

@@ -3,37 +3,25 @@
 # Makefile for Kernel-based Virtual Machine module
 #
 
-ccflags-y += -I $(srctree)/$(src) -I $(srctree)/virt/kvm/arm/vgic
+ccflags-y += -I $(srctree)/$(src)
 
 KVM=../../../virt/kvm
 
-obj-$(CONFIG_KVM_ARM_HOST) += kvm.o
+obj-$(CONFIG_KVM) += kvm.o
-obj-$(CONFIG_KVM_ARM_HOST) += hyp/
+obj-$(CONFIG_KVM) += hyp/
 
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o $(KVM)/eventfd.o $(KVM)/vfio.o
+kvm-y := $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o $(KVM)/eventfd.o \
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/arm.o $(KVM)/arm/mmu.o $(KVM)/arm/mmio.o
+	 $(KVM)/vfio.o $(KVM)/irqchip.o \
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/psci.o $(KVM)/arm/perf.o
+	 arm.o mmu.o mmio.o psci.o perf.o hypercalls.o pvtime.o \
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/hypercalls.o
+	 inject_fault.o regmap.o va_layout.o hyp.o hyp-init.o handle_exit.o \
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/pvtime.o
+	 guest.o debug.o reset.o sys_regs.o sys_regs_generic_v8.o \
+	 vgic-sys-reg-v3.o fpsimd.o pmu.o \
+	 aarch32.o arch_timer.o \
+	 vgic/vgic.o vgic/vgic-init.o \
+	 vgic/vgic-irqfd.o vgic/vgic-v2.o \
+	 vgic/vgic-v3.o vgic/vgic-v4.o \
+	 vgic/vgic-mmio.o vgic/vgic-mmio-v2.o \
+	 vgic/vgic-mmio-v3.o vgic/vgic-kvm-device.o \
+	 vgic/vgic-its.o vgic/vgic-debug.o
 
-kvm-$(CONFIG_KVM_ARM_HOST) += inject_fault.o regmap.o va_layout.o
+kvm-$(CONFIG_KVM_ARM_PMU)  += pmu-emul.o
-kvm-$(CONFIG_KVM_ARM_HOST) += hyp.o hyp-init.o handle_exit.o
-kvm-$(CONFIG_KVM_ARM_HOST) += guest.o debug.o reset.o sys_regs.o sys_regs_generic_v8.o
-kvm-$(CONFIG_KVM_ARM_HOST) += vgic-sys-reg-v3.o fpsimd.o pmu.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/aarch32.o
-
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-init.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-irqfd.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-v2.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-v3.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-v4.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio-v2.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio-v3.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-kvm-device.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-its.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-debug.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/irqchip.o
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/arch_timer.o
-kvm-$(CONFIG_KVM_ARM_PMU) += $(KVM)/arm/pmu.o

arch/arm64/kvm/arch_timer.c

@@ -451,17 +451,7 @@ static void timer_restore_state(struct arch_timer_context *ctx)
 
 static void set_cntvoff(u64 cntvoff)
 {
-	u32 low = lower_32_bits(cntvoff);
+	kvm_call_hyp(__kvm_timer_set_cntvoff, cntvoff);
-	u32 high = upper_32_bits(cntvoff);
-
-	/*
-	 * Since kvm_call_hyp doesn't fully support the ARM PCS especially on
-	 * 32-bit systems, but rather passes register by register shifted one
-	 * place (we put the function address in r0/x0), we cannot simply pass
-	 * a 64-bit value as an argument, but have to split the value in two
-	 * 32-bit halves.
-	 */
-	kvm_call_hyp(__kvm_timer_set_cntvoff, low, high);
 }
 
 static inline void set_timer_irq_phys_active(struct arch_timer_context *ctx, bool active)

arch/arm64/kvm/arm.c

@@ -22,7 +22,7 @@
 #include <trace/events/kvm.h>
 
 #define CREATE_TRACE_POINTS
-#include "trace.h"
+#include "trace_arm.h"
 
 #include <linux/uaccess.h>
 #include <asm/ptrace.h>
@@ -95,6 +95,11 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 	return r;
 }
 
+static int kvm_arm_default_max_vcpus(void)
+{
+	return vgic_present ? kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS;
+}
+
 /**
  * kvm_arch_init_vm - initializes a VM data structure
  * @kvm:	pointer to the KVM struct
@@ -128,8 +133,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	kvm->arch.vmid.vmid_gen = 0;
 
 	/* The maximum number of VCPUs is limited by the host's GIC model */
-	kvm->arch.max_vcpus = vgic_present ?
+	kvm->arch.max_vcpus = kvm_arm_default_max_vcpus();
-				kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS;
 
 	return ret;
 out_free_stage2_pgd:
@@ -204,10 +208,11 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		r = num_online_cpus();
 		break;
 	case KVM_CAP_MAX_VCPUS:
-		r = KVM_MAX_VCPUS;
-		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		if (kvm)
+			r = kvm->arch.max_vcpus;
+		else
+			r = kvm_arm_default_max_vcpus();
 		break;
 	case KVM_CAP_MSI_DEVID:
 		if (!kvm)
@@ -455,9 +460,9 @@ void force_vm_exit(const cpumask_t *mask)
  *
  * The hardware supports a limited set of values with the value zero reserved
  * for the host, so we check if an assigned value belongs to a previous
- * generation, which which requires us to assign a new value. If we're the
+ * generation, which requires us to assign a new value. If we're the first to
- * first to use a VMID for the new generation, we must flush necessary caches
+ * use a VMID for the new generation, we must flush necessary caches and TLBs
- * and TLBs on all CPUs.
+ * on all CPUs.
  */
 static bool need_new_vmid_gen(struct kvm_vmid *vmid)
 {
@@ -984,8 +989,11 @@ static int kvm_arch_vcpu_ioctl_vcpu_init(struct kvm_vcpu *vcpu,
 	/*
 	 * Ensure a rebooted VM will fault in RAM pages and detect if the
 	 * guest MMU is turned off and flush the caches as needed.
+	 *
+	 * S2FWB enforces all memory accesses to RAM being cacheable, we
+	 * ensure that the cache is always coherent.
 	 */
-	if (vcpu->arch.has_run_once)
+	if (vcpu->arch.has_run_once && !cpus_have_const_cap(ARM64_HAS_STAGE2_FWB))
 		stage2_unmap_vm(vcpu->kvm);
 
 	vcpu_reset_hcr(vcpu);
@@ -1266,19 +1274,41 @@ static void cpu_init_hyp_mode(void)
 {
 	phys_addr_t pgd_ptr;
 	unsigned long hyp_stack_ptr;
-	unsigned long stack_page;
 	unsigned long vector_ptr;
+	unsigned long tpidr_el2;
 
 	/* Switch from the HYP stub to our own HYP init vector */
 	__hyp_set_vectors(kvm_get_idmap_vector());
 
+	/*
+	 * Calculate the raw per-cpu offset without a translation from the
+	 * kernel's mapping to the linear mapping, and store it in tpidr_el2
+	 * so that we can use adr_l to access per-cpu variables in EL2.
+	 */
+	tpidr_el2 = ((unsigned long)this_cpu_ptr(&kvm_host_data) -
+		     (unsigned long)kvm_ksym_ref(kvm_host_data));
+
 	pgd_ptr = kvm_mmu_get_httbr();
-	stack_page = __this_cpu_read(kvm_arm_hyp_stack_page);
+	hyp_stack_ptr = __this_cpu_read(kvm_arm_hyp_stack_page) + PAGE_SIZE;
-	hyp_stack_ptr = stack_page + PAGE_SIZE;
 	vector_ptr = (unsigned long)kvm_get_hyp_vector();
 
-	__cpu_init_hyp_mode(pgd_ptr, hyp_stack_ptr, vector_ptr);
+	/*
-	__cpu_init_stage2();
+	 * Call initialization code, and switch to the full blown HYP code.
+	 * If the cpucaps haven't been finalized yet, something has gone very
+	 * wrong, and hyp will crash and burn when it uses any
+	 * cpus_have_const_cap() wrapper.
+	 */
+	BUG_ON(!system_capabilities_finalized());
+	__kvm_call_hyp((void *)pgd_ptr, hyp_stack_ptr, vector_ptr, tpidr_el2);
+
+	/*
+	 * Disabling SSBD on a non-VHE system requires us to enable SSBS
+	 * at EL2.
+	 */
+	if (this_cpu_has_cap(ARM64_SSBS) &&
+	    arm64_get_ssbd_state() == ARM64_SSBD_FORCE_DISABLE) {
+		kvm_call_hyp(__kvm_enable_ssbs);
+	}
 }
 
 static void cpu_hyp_reset(void)

arch/arm64/kvm/guest.c

@@ -266,7 +266,7 @@ static int set_sve_vls(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
 	/*
 	 * Vector lengths supported by the host can't currently be
 	 * hidden from the guest individually: instead we can only set a
-	 * maxmium via ZCR_EL2.LEN.  So, make sure the available vector
+	 * maximum via ZCR_EL2.LEN.  So, make sure the available vector
 	 * lengths match the set requested exactly up to the requested
 	 * maximum:
 	 */
@@ -336,7 +336,7 @@ static int sve_reg_to_region(struct sve_state_reg_region *region,
 	unsigned int reg_num;
 
 	unsigned int reqoffset, reqlen; /* User-requested offset and length */
-	unsigned int maxlen; /* Maxmimum permitted length */
+	unsigned int maxlen; /* Maximum permitted length */
 
 	size_t sve_state_size;
 

arch/arm64/kvm/handle_exit.c

@@ -23,7 +23,7 @@
 #include <kvm/arm_hypercalls.h>
 
 #define CREATE_TRACE_POINTS
-#include "trace.h"
+#include "trace_handle_exit.h"
 
 typedef int (*exit_handle_fn)(struct kvm_vcpu *, struct kvm_run *);
 

arch/arm64/kvm/hyp/Makefile

@@ -6,20 +6,10 @@
 ccflags-y += -fno-stack-protector -DDISABLE_BRANCH_PROFILING \
 		$(DISABLE_STACKLEAK_PLUGIN)
 
-KVM=../../../../virt/kvm
+obj-$(CONFIG_KVM) += hyp.o
 
-obj-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/hyp/vgic-v3-sr.o
+hyp-y := vgic-v3-sr.o timer-sr.o aarch32.o vgic-v2-cpuif-proxy.o sysreg-sr.o \
-obj-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/hyp/timer-sr.o
+	 debug-sr.o entry.o switch.o fpsimd.o tlb.o hyp-entry.o
-obj-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/hyp/aarch32.o
-
-obj-$(CONFIG_KVM_ARM_HOST) += vgic-v2-cpuif-proxy.o
-obj-$(CONFIG_KVM_ARM_HOST) += sysreg-sr.o
-obj-$(CONFIG_KVM_ARM_HOST) += debug-sr.o
-obj-$(CONFIG_KVM_ARM_HOST) += entry.o
-obj-$(CONFIG_KVM_ARM_HOST) += switch.o
-obj-$(CONFIG_KVM_ARM_HOST) += fpsimd.o
-obj-$(CONFIG_KVM_ARM_HOST) += tlb.o
-obj-$(CONFIG_KVM_ARM_HOST) += hyp-entry.o
 
 # KVM code is run at a different exception code with a different map, so
 # compiler instrumentation that inserts callbacks or checks into the code may

arch/arm64/kvm/hyp/switch.c

@@ -270,8 +270,8 @@ static void __hyp_text __deactivate_vm(struct kvm_vcpu *vcpu)
 static void __hyp_text __hyp_vgic_save_state(struct kvm_vcpu *vcpu)
 {
 	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif)) {
-		__vgic_v3_save_state(vcpu);
+		__vgic_v3_save_state(&vcpu->arch.vgic_cpu.vgic_v3);
-		__vgic_v3_deactivate_traps(vcpu);
+		__vgic_v3_deactivate_traps(&vcpu->arch.vgic_cpu.vgic_v3);
 	}
 }
 
@@ -279,8 +279,8 @@ static void __hyp_text __hyp_vgic_save_state(struct kvm_vcpu *vcpu)
 static void __hyp_text __hyp_vgic_restore_state(struct kvm_vcpu *vcpu)
 {
 	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif)) {
-		__vgic_v3_activate_traps(vcpu);
+		__vgic_v3_activate_traps(&vcpu->arch.vgic_cpu.vgic_v3);
-		__vgic_v3_restore_state(vcpu);
+		__vgic_v3_restore_state(&vcpu->arch.vgic_cpu.vgic_v3);
 	}
 }
 

arch/arm64/kvm/hyp/timer-sr.c

@@ -10,9 +10,8 @@
 
 #include <asm/kvm_hyp.h>
 
-void __hyp_text __kvm_timer_set_cntvoff(u32 cntvoff_low, u32 cntvoff_high)
+void __hyp_text __kvm_timer_set_cntvoff(u64 cntvoff)
 {
-	u64 cntvoff = (u64)cntvoff_high << 32 | cntvoff_low;
 	write_sysreg(cntvoff, cntvoff_el2);
 }
 

arch/arm64/kvm/hyp/vgic-v3-sr.c

@@ -194,10 +194,9 @@ static u32 __hyp_text __vgic_v3_read_ap1rn(int n)
 	return val;
 }
 
-void __hyp_text __vgic_v3_save_state(struct kvm_vcpu *vcpu)
+void __hyp_text __vgic_v3_save_state(struct vgic_v3_cpu_if *cpu_if)
 {
-	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+	u64 used_lrs = cpu_if->used_lrs;
-	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
 
 	/*
 	 * Make sure stores to the GIC via the memory mapped interface
@@ -230,10 +229,9 @@ void __hyp_text __vgic_v3_save_state(struct kvm_vcpu *vcpu)
 	}
 }
 
-void __hyp_text __vgic_v3_restore_state(struct kvm_vcpu *vcpu)
+void __hyp_text __vgic_v3_restore_state(struct vgic_v3_cpu_if *cpu_if)
 {
-	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+	u64 used_lrs = cpu_if->used_lrs;
-	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
 	int i;
 
 	if (used_lrs || cpu_if->its_vpe.its_vm) {
@@ -257,10 +255,8 @@ void __hyp_text __vgic_v3_restore_state(struct kvm_vcpu *vcpu)
 	}
 }
 
-void __hyp_text __vgic_v3_activate_traps(struct kvm_vcpu *vcpu)
+void __hyp_text __vgic_v3_activate_traps(struct vgic_v3_cpu_if *cpu_if)
 {
-	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
-
 	/*
 	 * VFIQEn is RES1 if ICC_SRE_EL1.SRE is 1. This causes a
 	 * Group0 interrupt (as generated in GICv2 mode) to be
@@ -306,9 +302,8 @@ void __hyp_text __vgic_v3_activate_traps(struct kvm_vcpu *vcpu)
 		write_gicreg(cpu_if->vgic_hcr, ICH_HCR_EL2);
 }
 
-void __hyp_text __vgic_v3_deactivate_traps(struct kvm_vcpu *vcpu)
+void __hyp_text __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if)
 {
-	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
 	u64 val;
 
 	if (!cpu_if->vgic_sre) {
@@ -333,15 +328,11 @@ void __hyp_text __vgic_v3_deactivate_traps(struct kvm_vcpu *vcpu)
 		write_gicreg(0, ICH_HCR_EL2);
 }
 
-void __hyp_text __vgic_v3_save_aprs(struct kvm_vcpu *vcpu)
+void __hyp_text __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if)
 {
-	struct vgic_v3_cpu_if *cpu_if;
 	u64 val;
 	u32 nr_pre_bits;
 
-	vcpu = kern_hyp_va(vcpu);
-	cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
-
 	val = read_gicreg(ICH_VTR_EL2);
 	nr_pre_bits = vtr_to_nr_pre_bits(val);
 
@@ -370,15 +361,11 @@ void __hyp_text __vgic_v3_save_aprs(struct kvm_vcpu *vcpu)
 	}
 }
 
-void __hyp_text __vgic_v3_restore_aprs(struct kvm_vcpu *vcpu)
+void __hyp_text __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if)
 {
-	struct vgic_v3_cpu_if *cpu_if;
 	u64 val;
 	u32 nr_pre_bits;
 
-	vcpu = kern_hyp_va(vcpu);
-	cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
-
 	val = read_gicreg(ICH_VTR_EL2);
 	nr_pre_bits = vtr_to_nr_pre_bits(val);
 
@@ -431,8 +418,6 @@ void __hyp_text __vgic_v3_write_vmcr(u32 vmcr)
 	write_gicreg(vmcr, ICH_VMCR_EL2);
 }
 
-#ifdef CONFIG_ARM64
-
 static int __hyp_text __vgic_v3_bpr_min(void)
 {
 	/* See Pseudocode for VPriorityGroup */
@@ -453,7 +438,7 @@ static int __hyp_text __vgic_v3_highest_priority_lr(struct kvm_vcpu *vcpu,
 						    u32 vmcr,
 						    u64 *lr_val)
 {
-	unsigned int used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	unsigned int used_lrs = vcpu->arch.vgic_cpu.vgic_v3.used_lrs;
 	u8 priority = GICv3_IDLE_PRIORITY;
 	int i, lr = -1;
 
@@ -492,7 +477,7 @@ static int __hyp_text __vgic_v3_highest_priority_lr(struct kvm_vcpu *vcpu,
 static int __hyp_text __vgic_v3_find_active_lr(struct kvm_vcpu *vcpu,
 					       int intid, u64 *lr_val)
 {
-	unsigned int used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	unsigned int used_lrs = vcpu->arch.vgic_cpu.vgic_v3.used_lrs;
 	int i;
 
 	for (i = 0; i < used_lrs; i++) {
@@ -579,7 +564,7 @@ static u8 __hyp_text __vgic_v3_pri_to_pre(u8 pri, u32 vmcr, int grp)
 
 /*
  * The priority value is independent of any of the BPR values, so we
- * normalize it using the minumal BPR value. This guarantees that no
+ * normalize it using the minimal BPR value. This guarantees that no
  * matter what the guest does with its BPR, we can always set/get the
  * same value of a priority.
  */
@@ -1126,5 +1111,3 @@ int __hyp_text __vgic_v3_perform_cpuif_access(struct kvm_vcpu *vcpu)
 
 	return 1;
 }
-
-#endif

arch/arm64/kvm/inject_fault.c

@@ -26,28 +26,12 @@ enum exception_type {
 	except_type_serror	= 0x180,
 };
 
-static u64 get_except_vector(struct kvm_vcpu *vcpu, enum exception_type type)
-{
-	u64 exc_offset;
-
-	switch (*vcpu_cpsr(vcpu) & (PSR_MODE_MASK | PSR_MODE32_BIT)) {
-	case PSR_MODE_EL1t:
-		exc_offset = CURRENT_EL_SP_EL0_VECTOR;
-		break;
-	case PSR_MODE_EL1h:
-		exc_offset = CURRENT_EL_SP_ELx_VECTOR;
-		break;
-	case PSR_MODE_EL0t:
-		exc_offset = LOWER_EL_AArch64_VECTOR;
-		break;
-	default:
-		exc_offset = LOWER_EL_AArch32_VECTOR;
-	}
-
-	return vcpu_read_sys_reg(vcpu, VBAR_EL1) + exc_offset + type;
-}
-
 /*
+ * This performs the exception entry at a given EL (@target_mode), stashing PC
+ * and PSTATE into ELR and SPSR respectively, and compute the new PC/PSTATE.
+ * The EL passed to this function *must* be a non-secure, privileged mode with
+ * bit 0 being set (PSTATE.SP == 1).
+ *
  * When an exception is taken, most PSTATE fields are left unchanged in the
  * handler. However, some are explicitly overridden (e.g. M[4:0]). Luckily all
  * of the inherited bits have the same position in the AArch64/AArch32 SPSR_ELx
@@ -59,10 +43,35 @@ static u64 get_except_vector(struct kvm_vcpu *vcpu, enum exception_type type)
  * Here we manipulate the fields in order of the AArch64 SPSR_ELx layout, from
  * MSB to LSB.
  */
-static unsigned long get_except64_pstate(struct kvm_vcpu *vcpu)
+static void enter_exception64(struct kvm_vcpu *vcpu, unsigned long target_mode,
+			      enum exception_type type)
 {
-	unsigned long sctlr = vcpu_read_sys_reg(vcpu, SCTLR_EL1);
+	unsigned long sctlr, vbar, old, new, mode;
-	unsigned long old, new;
+	u64 exc_offset;
+
+	mode = *vcpu_cpsr(vcpu) & (PSR_MODE_MASK | PSR_MODE32_BIT);
+
+	if      (mode == target_mode)
+		exc_offset = CURRENT_EL_SP_ELx_VECTOR;
+	else if ((mode | PSR_MODE_THREAD_BIT) == target_mode)
+		exc_offset = CURRENT_EL_SP_EL0_VECTOR;
+	else if (!(mode & PSR_MODE32_BIT))
+		exc_offset = LOWER_EL_AArch64_VECTOR;
+	else
+		exc_offset = LOWER_EL_AArch32_VECTOR;
+
+	switch (target_mode) {
+	case PSR_MODE_EL1h:
+		vbar = vcpu_read_sys_reg(vcpu, VBAR_EL1);
+		sctlr = vcpu_read_sys_reg(vcpu, SCTLR_EL1);
+		vcpu_write_elr_el1(vcpu, *vcpu_pc(vcpu));
+		break;
+	default:
+		/* Don't do that */
+		BUG();
+	}
+
+	*vcpu_pc(vcpu) = vbar + exc_offset + type;
 
 	old = *vcpu_cpsr(vcpu);
 	new = 0;
@@ -105,9 +114,10 @@ static unsigned long get_except64_pstate(struct kvm_vcpu *vcpu)
 	new |= PSR_I_BIT;
 	new |= PSR_F_BIT;
 
-	new |= PSR_MODE_EL1h;
+	new |= target_mode;
 
-	return new;
+	*vcpu_cpsr(vcpu) = new;
+	vcpu_write_spsr(vcpu, old);
 }
 
 static void inject_abt64(struct kvm_vcpu *vcpu, bool is_iabt, unsigned long addr)
@@ -116,11 +126,7 @@ static void inject_abt64(struct kvm_vcpu *vcpu, bool is_iabt, unsigned long addr
 	bool is_aarch32 = vcpu_mode_is_32bit(vcpu);
 	u32 esr = 0;
 
-	vcpu_write_elr_el1(vcpu, *vcpu_pc(vcpu));
+	enter_exception64(vcpu, PSR_MODE_EL1h, except_type_sync);
-	*vcpu_pc(vcpu) = get_except_vector(vcpu, except_type_sync);
-
-	*vcpu_cpsr(vcpu) = get_except64_pstate(vcpu);
-	vcpu_write_spsr(vcpu, cpsr);
 
 	vcpu_write_sys_reg(vcpu, addr, FAR_EL1);
 
@@ -148,14 +154,9 @@ static void inject_abt64(struct kvm_vcpu *vcpu, bool is_iabt, unsigned long addr
 
 static void inject_undef64(struct kvm_vcpu *vcpu)
 {
-	unsigned long cpsr = *vcpu_cpsr(vcpu);
 	u32 esr = (ESR_ELx_EC_UNKNOWN << ESR_ELx_EC_SHIFT);
 
-	vcpu_write_elr_el1(vcpu, *vcpu_pc(vcpu));
+	enter_exception64(vcpu, PSR_MODE_EL1h, except_type_sync);
-	*vcpu_pc(vcpu) = get_except_vector(vcpu, except_type_sync);
-
-	*vcpu_cpsr(vcpu) = get_except64_pstate(vcpu);
-	vcpu_write_spsr(vcpu, cpsr);
 
 	/*
 	 * Build an unknown exception, depending on the instruction

arch/arm64/kvm/mmio.c

@@ -131,7 +131,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 
 	/*
 	 * No valid syndrome? Ask userspace for help if it has
-	 * voluntered to do so, and bail out otherwise.
+	 * volunteered to do so, and bail out otherwise.
 	 */
 	if (!kvm_vcpu_dabt_isvalid(vcpu)) {
 		if (vcpu->kvm->arch.return_nisv_io_abort_to_user) {

arch/arm64/kvm/mmu.c

@@ -422,6 +422,9 @@ static void stage2_flush_memslot(struct kvm *kvm,
 		next = stage2_pgd_addr_end(kvm, addr, end);
 		if (!stage2_pgd_none(kvm, *pgd))
 			stage2_flush_puds(kvm, pgd, addr, next);
+
+		if (next != end)
+			cond_resched_lock(&kvm->mmu_lock);
 	} while (pgd++, addr = next, addr != end);
 }
 
@@ -784,7 +787,7 @@ static int __create_hyp_private_mapping(phys_addr_t phys_addr, size_t size,
 	mutex_lock(&kvm_hyp_pgd_mutex);
 
 	/*
-	 * This assumes that we we have enough space below the idmap
+	 * This assumes that we have enough space below the idmap
 	 * page to allocate our VAs. If not, the check below will
 	 * kick. A potential alternative would be to detect that
 	 * overflow and switch to an allocation above the idmap.
@@ -964,7 +967,7 @@ static void stage2_unmap_memslot(struct kvm *kvm,
  * stage2_unmap_vm - Unmap Stage-2 RAM mappings
  * @kvm: The struct kvm pointer
  *
- * Go through the memregions and unmap any reguler RAM
+ * Go through the memregions and unmap any regular RAM
  * backing memory already mapped to the VM.
  */
 void stage2_unmap_vm(struct kvm *kvm)
@@ -1372,47 +1375,6 @@ int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa,
 	return ret;
 }
 
-static bool transparent_hugepage_adjust(kvm_pfn_t *pfnp, phys_addr_t *ipap)
-{
-	kvm_pfn_t pfn = *pfnp;
-	gfn_t gfn = *ipap >> PAGE_SHIFT;
-
-	if (kvm_is_transparent_hugepage(pfn)) {
-		unsigned long mask;
-		/*
-		 * The address we faulted on is backed by a transparent huge
-		 * page.  However, because we map the compound huge page and
-		 * not the individual tail page, we need to transfer the
-		 * refcount to the head page.  We have to be careful that the
-		 * THP doesn't start to split while we are adjusting the
-		 * refcounts.
-		 *
-		 * We are sure this doesn't happen, because mmu_notifier_retry
-		 * was successful and we are holding the mmu_lock, so if this
-		 * THP is trying to split, it will be blocked in the mmu
-		 * notifier before touching any of the pages, specifically
-		 * before being able to call __split_huge_page_refcount().
-		 *
-		 * We can therefore safely transfer the refcount from PG_tail
-		 * to PG_head and switch the pfn from a tail page to the head
-		 * page accordingly.
-		 */
-		mask = PTRS_PER_PMD - 1;
-		VM_BUG_ON((gfn & mask) != (pfn & mask));
-		if (pfn & mask) {
-			*ipap &= PMD_MASK;
-			kvm_release_pfn_clean(pfn);
-			pfn &= ~mask;
-			kvm_get_pfn(pfn);
-			*pfnp = pfn;
-		}
-
-		return true;
-	}
-
-	return false;
-}
-
 /**
  * stage2_wp_ptes - write protect PMD range
  * @pmd:	pointer to pmd entry
@@ -1607,6 +1569,10 @@ static bool fault_supports_stage2_huge_mapping(struct kvm_memory_slot *memslot,
 	hva_t uaddr_start, uaddr_end;
 	size_t size;
 
+	/* The memslot and the VMA are guaranteed to be aligned to PAGE_SIZE */
+	if (map_size == PAGE_SIZE)
+		return true;
+
 	size = memslot->npages * PAGE_SIZE;
 
 	gpa_start = memslot->base_gfn << PAGE_SHIFT;
@@ -1626,7 +1592,7 @@ static bool fault_supports_stage2_huge_mapping(struct kvm_memory_slot *memslot,
 	 *    |abcde|fgh  Stage-1 block  |    Stage-1 block tv|xyz|
 	 *    +-----+--------------------+--------------------+---+
 	 *
-	 *    memslot->base_gfn << PAGE_SIZE:
+	 *    memslot->base_gfn << PAGE_SHIFT:
 	 *      +---+--------------------+--------------------+-----+
 	 *      |abc|def  Stage-2 block  |    Stage-2 block   |tvxyz|
 	 *      +---+--------------------+--------------------+-----+
@@ -1656,6 +1622,59 @@ static bool fault_supports_stage2_huge_mapping(struct kvm_memory_slot *memslot,
 	       (hva & ~(map_size - 1)) + map_size <= uaddr_end;
 }
 
+/*
+ * Check if the given hva is backed by a transparent huge page (THP) and
+ * whether it can be mapped using block mapping in stage2. If so, adjust
+ * the stage2 PFN and IPA accordingly. Only PMD_SIZE THPs are currently
+ * supported. This will need to be updated to support other THP sizes.
+ *
+ * Returns the size of the mapping.
+ */
+static unsigned long
+transparent_hugepage_adjust(struct kvm_memory_slot *memslot,
+			    unsigned long hva, kvm_pfn_t *pfnp,
+			    phys_addr_t *ipap)
+{
+	kvm_pfn_t pfn = *pfnp;
+
+	/*
+	 * Make sure the adjustment is done only for THP pages. Also make
+	 * sure that the HVA and IPA are sufficiently aligned and that the
+	 * block map is contained within the memslot.
+	 */
+	if (kvm_is_transparent_hugepage(pfn) &&
+	    fault_supports_stage2_huge_mapping(memslot, hva, PMD_SIZE)) {
+		/*
+		 * The address we faulted on is backed by a transparent huge
+		 * page.  However, because we map the compound huge page and
+		 * not the individual tail page, we need to transfer the
+		 * refcount to the head page.  We have to be careful that the
+		 * THP doesn't start to split while we are adjusting the
+		 * refcounts.
+		 *
+		 * We are sure this doesn't happen, because mmu_notifier_retry
+		 * was successful and we are holding the mmu_lock, so if this
+		 * THP is trying to split, it will be blocked in the mmu
+		 * notifier before touching any of the pages, specifically
+		 * before being able to call __split_huge_page_refcount().
+		 *
+		 * We can therefore safely transfer the refcount from PG_tail
+		 * to PG_head and switch the pfn from a tail page to the head
+		 * page accordingly.
+		 */
+		*ipap &= PMD_MASK;
+		kvm_release_pfn_clean(pfn);
+		pfn &= ~(PTRS_PER_PMD - 1);
+		kvm_get_pfn(pfn);
+		*pfnp = pfn;
+
+		return PMD_SIZE;
+	}
+
+	/* Use page mapping if we cannot use block mapping. */
+	return PAGE_SIZE;
+}
+
 static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 			  struct kvm_memory_slot *memslot, unsigned long hva,
 			  unsigned long fault_status)
@@ -1769,20 +1788,13 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	if (mmu_notifier_retry(kvm, mmu_seq))
 		goto out_unlock;
 
-	if (vma_pagesize == PAGE_SIZE && !force_pte) {
+	/*
-		/*
+	 * If we are not forced to use page mapping, check if we are
-		 * Only PMD_SIZE transparent hugepages(THP) are
+	 * backed by a THP and thus use block mapping if possible.
-		 * currently supported. This code will need to be
+	 */
-		 * updated to support other THP sizes.
+	if (vma_pagesize == PAGE_SIZE && !force_pte)
-		 *
+		vma_pagesize = transparent_hugepage_adjust(memslot, hva,
-		 * Make sure the host VA and the guest IPA are sufficiently
+							   &pfn, &fault_ipa);
-		 * aligned and that the block is contained within the memslot.
-		 */
-		if (fault_supports_stage2_huge_mapping(memslot, hva, PMD_SIZE) &&
-		    transparent_hugepage_adjust(&pfn, &fault_ipa))
-			vma_pagesize = PMD_SIZE;
-	}
-
 	if (writable)
 		kvm_set_pfn_dirty(pfn);
 
@@ -2185,11 +2197,11 @@ int kvm_mmu_init(void)
 {
 	int err;
 
-	hyp_idmap_start = kvm_virt_to_phys(__hyp_idmap_text_start);
+	hyp_idmap_start = __pa_symbol(__hyp_idmap_text_start);
 	hyp_idmap_start = ALIGN_DOWN(hyp_idmap_start, PAGE_SIZE);
-	hyp_idmap_end = kvm_virt_to_phys(__hyp_idmap_text_end);
+	hyp_idmap_end = __pa_symbol(__hyp_idmap_text_end);
 	hyp_idmap_end = ALIGN(hyp_idmap_end, PAGE_SIZE);
-	hyp_idmap_vector = kvm_virt_to_phys(__kvm_hyp_init);
+	hyp_idmap_vector = __pa_symbol(__kvm_hyp_init);
 
 	/*
 	 * We rely on the linker script to ensure at build time that the HYP
@@ -2262,11 +2274,19 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 {
 	/*
 	 * At this point memslot has been committed and there is an
-	 * allocated dirty_bitmap[], dirty pages will be be tracked while the
+	 * allocated dirty_bitmap[], dirty pages will be tracked while the
 	 * memory slot is write protected.
 	 */
-	if (change != KVM_MR_DELETE && mem->flags & KVM_MEM_LOG_DIRTY_PAGES)
+	if (change != KVM_MR_DELETE && mem->flags & KVM_MEM_LOG_DIRTY_PAGES) {
-		kvm_mmu_wp_memory_region(kvm, mem->slot);
+		/*
+		 * If we're with initial-all-set, we don't need to write
+		 * protect any pages because they're all reported as dirty.
+		 * Huge pages and normal pages will be write protect gradually.
+		 */
+		if (!kvm_dirty_log_manual_protect_and_init_set(kvm)) {
+			kvm_mmu_wp_memory_region(kvm, mem->slot);
+		}
+	}
 }
 
 int kvm_arch_prepare_memory_region(struct kvm *kvm,

arch/arm64/kvm/psci.c

@@ -94,7 +94,7 @@ static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 
 	/*
 	 * NOTE: We always update r0 (or x0) because for PSCI v0.1
-	 * the general puspose registers are undefined upon CPU_ON.
+	 * the general purpose registers are undefined upon CPU_ON.
 	 */
 	reset_state->r0 = smccc_get_arg3(source_vcpu);
 
@@ -265,10 +265,10 @@ static int kvm_psci_0_2_call(struct kvm_vcpu *vcpu)
 	case PSCI_0_2_FN_SYSTEM_OFF:
 		kvm_psci_system_off(vcpu);
 		/*
-		 * We should'nt be going back to guest VCPU after
+		 * We shouldn't be going back to guest VCPU after
 		 * receiving SYSTEM_OFF request.
 		 *
-		 * If user space accidently/deliberately resumes
+		 * If user space accidentally/deliberately resumes
 		 * guest VCPU after SYSTEM_OFF request then guest
 		 * VCPU should see internal failure from PSCI return
 		 * value. To achieve this, we preload r0 (or x0) with

arch/arm64/kvm/reset.c

@@ -36,15 +36,11 @@ static u32 kvm_ipa_limit;
 /*
  * ARMv8 Reset Values
  */
-static const struct kvm_regs default_regs_reset = {
+#define VCPU_RESET_PSTATE_EL1	(PSR_MODE_EL1h | PSR_A_BIT | PSR_I_BIT | \
-	.regs.pstate = (PSR_MODE_EL1h | PSR_A_BIT | PSR_I_BIT |
+				 PSR_F_BIT | PSR_D_BIT)
-			PSR_F_BIT | PSR_D_BIT),
-};
 
-static const struct kvm_regs default_regs_reset32 = {
+#define VCPU_RESET_PSTATE_SVC	(PSR_AA32_MODE_SVC | PSR_AA32_A_BIT | \
-	.regs.pstate = (PSR_AA32_MODE_SVC | PSR_AA32_A_BIT |
+				 PSR_AA32_I_BIT | PSR_AA32_F_BIT)
-			PSR_AA32_I_BIT | PSR_AA32_F_BIT),
-};
 
 static bool cpu_has_32bit_el1(void)
 {
@@ -163,7 +159,7 @@ static int kvm_vcpu_finalize_sve(struct kvm_vcpu *vcpu)
 	vl = vcpu->arch.sve_max_vl;
 
 	/*
-	 * Resposibility for these properties is shared between
+	 * Responsibility for these properties is shared between
 	 * kvm_arm_init_arch_resources(), kvm_vcpu_enable_sve() and
 	 * set_sve_vls().  Double-check here just to be sure:
 	 */
@@ -249,7 +245,7 @@ static int kvm_vcpu_enable_ptrauth(struct kvm_vcpu *vcpu)
  * ioctl or as part of handling a request issued by another VCPU in the PSCI
  * handling code.  In the first case, the VCPU will not be loaded, and in the
  * second case the VCPU will be loaded.  Because this function operates purely
- * on the memory-backed valus of system registers, we want to do a full put if
+ * on the memory-backed values of system registers, we want to do a full put if
  * we were loaded (handling a request) and load the values back at the end of
  * the function.  Otherwise we leave the state alone.  In both cases, we
  * disable preemption around the vcpu reset as we would otherwise race with
@@ -257,9 +253,9 @@ static int kvm_vcpu_enable_ptrauth(struct kvm_vcpu *vcpu)
  */
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 {
-	const struct kvm_regs *cpu_reset;
 	int ret = -EINVAL;
 	bool loaded;
+	u32 pstate;
 
 	/* Reset PMU outside of the non-preemptible section */
 	kvm_pmu_vcpu_reset(vcpu);
@@ -290,16 +286,17 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 		if (test_bit(KVM_ARM_VCPU_EL1_32BIT, vcpu->arch.features)) {
 			if (!cpu_has_32bit_el1())
 				goto out;
-			cpu_reset = &default_regs_reset32;
+			pstate = VCPU_RESET_PSTATE_SVC;
 		} else {
-			cpu_reset = &default_regs_reset;
+			pstate = VCPU_RESET_PSTATE_EL1;
 		}
 
 		break;
 	}
 
 	/* Reset core registers */
-	memcpy(vcpu_gp_regs(vcpu), cpu_reset, sizeof(*cpu_reset));
+	memset(vcpu_gp_regs(vcpu), 0, sizeof(*vcpu_gp_regs(vcpu)));
+	vcpu_gp_regs(vcpu)->regs.pstate = pstate;
 
 	/* Reset system registers */
 	kvm_reset_sys_regs(vcpu);
@@ -357,7 +354,7 @@ void kvm_set_ipa_limit(void)
 	 *
 	 * So clamp the ipa limit further down to limit the number of levels.
 	 * Since we can concatenate upto 16 tables at entry level, we could
-	 * go upto 4bits above the maximum VA addressible with the current
+	 * go upto 4bits above the maximum VA addressable with the current
 	 * number of levels.
 	 */
 	va_max = PGDIR_SHIFT + PAGE_SHIFT - 3;

arch/arm64/kvm/sys_regs.c

@@ -34,7 +34,7 @@
 #include "trace.h"
 
 /*
- * All of this file is extremly similar to the ARM coproc.c, but the
+ * All of this file is extremely similar to the ARM coproc.c, but the
  * types are different. My gut feeling is that it should be pretty
  * easy to merge, but that would be an ABI breakage -- again. VFP
  * would also need to be abstracted.
@@ -64,11 +64,8 @@ static bool write_to_read_only(struct kvm_vcpu *vcpu,
 	return false;
 }
 
-u64 vcpu_read_sys_reg(const struct kvm_vcpu *vcpu, int reg)
+static bool __vcpu_read_sys_reg_from_cpu(int reg, u64 *val)
 {
-	if (!vcpu->arch.sysregs_loaded_on_cpu)
-		goto immediate_read;
-
 	/*
 	 * System registers listed in the switch are not saved on every
 	 * exit from the guest but are only saved on vcpu_put.
@@ -79,75 +76,92 @@ u64 vcpu_read_sys_reg(const struct kvm_vcpu *vcpu, int reg)
 	 * thread when emulating cross-VCPU communication.
 	 */
 	switch (reg) {
-	case CSSELR_EL1:	return read_sysreg_s(SYS_CSSELR_EL1);
+	case CSSELR_EL1:	*val = read_sysreg_s(SYS_CSSELR_EL1);	break;
-	case SCTLR_EL1:		return read_sysreg_s(SYS_SCTLR_EL12);
+	case SCTLR_EL1:		*val = read_sysreg_s(SYS_SCTLR_EL12);	break;
-	case ACTLR_EL1:		return read_sysreg_s(SYS_ACTLR_EL1);
+	case ACTLR_EL1:		*val = read_sysreg_s(SYS_ACTLR_EL1);	break;
-	case CPACR_EL1:		return read_sysreg_s(SYS_CPACR_EL12);
+	case CPACR_EL1:		*val = read_sysreg_s(SYS_CPACR_EL12);	break;
-	case TTBR0_EL1:		return read_sysreg_s(SYS_TTBR0_EL12);
+	case TTBR0_EL1:		*val = read_sysreg_s(SYS_TTBR0_EL12);	break;
-	case TTBR1_EL1:		return read_sysreg_s(SYS_TTBR1_EL12);
+	case TTBR1_EL1:		*val = read_sysreg_s(SYS_TTBR1_EL12);	break;
-	case TCR_EL1:		return read_sysreg_s(SYS_TCR_EL12);
+	case TCR_EL1:		*val = read_sysreg_s(SYS_TCR_EL12);	break;
-	case ESR_EL1:		return read_sysreg_s(SYS_ESR_EL12);
+	case ESR_EL1:		*val = read_sysreg_s(SYS_ESR_EL12);	break;
-	case AFSR0_EL1:		return read_sysreg_s(SYS_AFSR0_EL12);
+	case AFSR0_EL1:		*val = read_sysreg_s(SYS_AFSR0_EL12);	break;
-	case AFSR1_EL1:		return read_sysreg_s(SYS_AFSR1_EL12);
+	case AFSR1_EL1:		*val = read_sysreg_s(SYS_AFSR1_EL12);	break;
-	case FAR_EL1:		return read_sysreg_s(SYS_FAR_EL12);
+	case FAR_EL1:		*val = read_sysreg_s(SYS_FAR_EL12);	break;
-	case MAIR_EL1:		return read_sysreg_s(SYS_MAIR_EL12);
+	case MAIR_EL1:		*val = read_sysreg_s(SYS_MAIR_EL12);	break;
-	case VBAR_EL1:		return read_sysreg_s(SYS_VBAR_EL12);
+	case VBAR_EL1:		*val = read_sysreg_s(SYS_VBAR_EL12);	break;
-	case CONTEXTIDR_EL1:	return read_sysreg_s(SYS_CONTEXTIDR_EL12);
+	case CONTEXTIDR_EL1:	*val = read_sysreg_s(SYS_CONTEXTIDR_EL12);break;
-	case TPIDR_EL0:		return read_sysreg_s(SYS_TPIDR_EL0);
+	case TPIDR_EL0:		*val = read_sysreg_s(SYS_TPIDR_EL0);	break;
-	case TPIDRRO_EL0:	return read_sysreg_s(SYS_TPIDRRO_EL0);
+	case TPIDRRO_EL0:	*val = read_sysreg_s(SYS_TPIDRRO_EL0);	break;
-	case TPIDR_EL1:		return read_sysreg_s(SYS_TPIDR_EL1);
+	case TPIDR_EL1:		*val = read_sysreg_s(SYS_TPIDR_EL1);	break;
-	case AMAIR_EL1:		return read_sysreg_s(SYS_AMAIR_EL12);
+	case AMAIR_EL1:		*val = read_sysreg_s(SYS_AMAIR_EL12);	break;
-	case CNTKCTL_EL1:	return read_sysreg_s(SYS_CNTKCTL_EL12);
+	case CNTKCTL_EL1:	*val = read_sysreg_s(SYS_CNTKCTL_EL12);	break;
-	case PAR_EL1:		return read_sysreg_s(SYS_PAR_EL1);
+	case PAR_EL1:		*val = read_sysreg_s(SYS_PAR_EL1);	break;
-	case DACR32_EL2:	return read_sysreg_s(SYS_DACR32_EL2);
+	case DACR32_EL2:	*val = read_sysreg_s(SYS_DACR32_EL2);	break;
-	case IFSR32_EL2:	return read_sysreg_s(SYS_IFSR32_EL2);
+	case IFSR32_EL2:	*val = read_sysreg_s(SYS_IFSR32_EL2);	break;
-	case DBGVCR32_EL2:	return read_sysreg_s(SYS_DBGVCR32_EL2);
+	case DBGVCR32_EL2:	*val = read_sysreg_s(SYS_DBGVCR32_EL2);	break;
+	default:		return false;
 	}
 
-immediate_read:
+	return true;
-	return __vcpu_sys_reg(vcpu, reg);
 }
 
-void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg)
+static bool __vcpu_write_sys_reg_to_cpu(u64 val, int reg)
 {
-	if (!vcpu->arch.sysregs_loaded_on_cpu)
-		goto immediate_write;
-
 	/*
 	 * System registers listed in the switch are not restored on every
 	 * entry to the guest but are only restored on vcpu_load.
 	 *
 	 * Note that MPIDR_EL1 for the guest is set by KVM via VMPIDR_EL2 but
-	 * should never be listed below, because the the MPIDR should only be
+	 * should never be listed below, because the MPIDR should only be set
-	 * set once, before running the VCPU, and never changed later.
+	 * once, before running the VCPU, and never changed later.
 	 */
 	switch (reg) {
-	case CSSELR_EL1:	write_sysreg_s(val, SYS_CSSELR_EL1);	return;
+	case CSSELR_EL1:	write_sysreg_s(val, SYS_CSSELR_EL1);	break;
-	case SCTLR_EL1:		write_sysreg_s(val, SYS_SCTLR_EL12);	return;
+	case SCTLR_EL1:		write_sysreg_s(val, SYS_SCTLR_EL12);	break;
-	case ACTLR_EL1:		write_sysreg_s(val, SYS_ACTLR_EL1);	return;
+	case ACTLR_EL1:		write_sysreg_s(val, SYS_ACTLR_EL1);	break;
-	case CPACR_EL1:		write_sysreg_s(val, SYS_CPACR_EL12);	return;
+	case CPACR_EL1:		write_sysreg_s(val, SYS_CPACR_EL12);	break;
-	case TTBR0_EL1:		write_sysreg_s(val, SYS_TTBR0_EL12);	return;
+	case TTBR0_EL1:		write_sysreg_s(val, SYS_TTBR0_EL12);	break;
-	case TTBR1_EL1:		write_sysreg_s(val, SYS_TTBR1_EL12);	return;
+	case TTBR1_EL1:		write_sysreg_s(val, SYS_TTBR1_EL12);	break;
-	case TCR_EL1:		write_sysreg_s(val, SYS_TCR_EL12);	return;
+	case TCR_EL1:		write_sysreg_s(val, SYS_TCR_EL12);	break;
-	case ESR_EL1:		write_sysreg_s(val, SYS_ESR_EL12);	return;
+	case ESR_EL1:		write_sysreg_s(val, SYS_ESR_EL12);	break;
-	case AFSR0_EL1:		write_sysreg_s(val, SYS_AFSR0_EL12);	return;
+	case AFSR0_EL1:		write_sysreg_s(val, SYS_AFSR0_EL12);	break;
-	case AFSR1_EL1:		write_sysreg_s(val, SYS_AFSR1_EL12);	return;
+	case AFSR1_EL1:		write_sysreg_s(val, SYS_AFSR1_EL12);	break;
-	case FAR_EL1:		write_sysreg_s(val, SYS_FAR_EL12);	return;
+	case FAR_EL1:		write_sysreg_s(val, SYS_FAR_EL12);	break;
-	case MAIR_EL1:		write_sysreg_s(val, SYS_MAIR_EL12);	return;
+	case MAIR_EL1:		write_sysreg_s(val, SYS_MAIR_EL12);	break;
-	case VBAR_EL1:		write_sysreg_s(val, SYS_VBAR_EL12);	return;
+	case VBAR_EL1:		write_sysreg_s(val, SYS_VBAR_EL12);	break;
-	case CONTEXTIDR_EL1:	write_sysreg_s(val, SYS_CONTEXTIDR_EL12); return;
+	case CONTEXTIDR_EL1:	write_sysreg_s(val, SYS_CONTEXTIDR_EL12);break;
-	case TPIDR_EL0:		write_sysreg_s(val, SYS_TPIDR_EL0);	return;
+	case TPIDR_EL0:		write_sysreg_s(val, SYS_TPIDR_EL0);	break;
-	case TPIDRRO_EL0:	write_sysreg_s(val, SYS_TPIDRRO_EL0);	return;
+	case TPIDRRO_EL0:	write_sysreg_s(val, SYS_TPIDRRO_EL0);	break;
-	case TPIDR_EL1:		write_sysreg_s(val, SYS_TPIDR_EL1);	return;
+	case TPIDR_EL1:		write_sysreg_s(val, SYS_TPIDR_EL1);	break;
-	case AMAIR_EL1:		write_sysreg_s(val, SYS_AMAIR_EL12);	return;
+	case AMAIR_EL1:		write_sysreg_s(val, SYS_AMAIR_EL12);	break;
-	case CNTKCTL_EL1:	write_sysreg_s(val, SYS_CNTKCTL_EL12);	return;
+	case CNTKCTL_EL1:	write_sysreg_s(val, SYS_CNTKCTL_EL12);	break;
-	case PAR_EL1:		write_sysreg_s(val, SYS_PAR_EL1);	return;
+	case PAR_EL1:		write_sysreg_s(val, SYS_PAR_EL1);	break;
-	case DACR32_EL2:	write_sysreg_s(val, SYS_DACR32_EL2);	return;
+	case DACR32_EL2:	write_sysreg_s(val, SYS_DACR32_EL2);	break;
-	case IFSR32_EL2:	write_sysreg_s(val, SYS_IFSR32_EL2);	return;
+	case IFSR32_EL2:	write_sysreg_s(val, SYS_IFSR32_EL2);	break;
-	case DBGVCR32_EL2:	write_sysreg_s(val, SYS_DBGVCR32_EL2);	return;
+	case DBGVCR32_EL2:	write_sysreg_s(val, SYS_DBGVCR32_EL2);	break;
+	default:		return false;
 	}
 
-immediate_write:
+	return true;
+}
+
+u64 vcpu_read_sys_reg(const struct kvm_vcpu *vcpu, int reg)
+{
+	u64 val = 0x8badf00d8badf00d;
+
+	if (vcpu->arch.sysregs_loaded_on_cpu &&
+	    __vcpu_read_sys_reg_from_cpu(reg, &val))
+		return val;
+
+	return __vcpu_sys_reg(vcpu, reg);
+}
+
+void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg)
+{
+	if (vcpu->arch.sysregs_loaded_on_cpu &&
+	    __vcpu_write_sys_reg_to_cpu(val, reg))
+		return;
+
 	 __vcpu_sys_reg(vcpu, reg) = val;
 }
 
@@ -1532,7 +1546,7 @@ static const struct sys_reg_desc sys_reg_descs[] = {
 	{ SYS_DESC(SYS_PAR_EL1), NULL, reset_unknown, PAR_EL1 },
 
 	{ SYS_DESC(SYS_PMINTENSET_EL1), access_pminten, reset_unknown, PMINTENSET_EL1 },
-	{ SYS_DESC(SYS_PMINTENCLR_EL1), access_pminten, NULL, PMINTENSET_EL1 },
+	{ SYS_DESC(SYS_PMINTENCLR_EL1), access_pminten, reset_unknown, PMINTENSET_EL1 },
 
 	{ SYS_DESC(SYS_MAIR_EL1), access_vm_reg, reset_unknown, MAIR_EL1 },
 	{ SYS_DESC(SYS_AMAIR_EL1), access_vm_reg, reset_amair_el1, AMAIR_EL1 },
@@ -1571,8 +1585,8 @@ static const struct sys_reg_desc sys_reg_descs[] = {
 
 	{ SYS_DESC(SYS_PMCR_EL0), access_pmcr, reset_pmcr, PMCR_EL0 },
 	{ SYS_DESC(SYS_PMCNTENSET_EL0), access_pmcnten, reset_unknown, PMCNTENSET_EL0 },
-	{ SYS_DESC(SYS_PMCNTENCLR_EL0), access_pmcnten, NULL, PMCNTENSET_EL0 },
+	{ SYS_DESC(SYS_PMCNTENCLR_EL0), access_pmcnten, reset_unknown, PMCNTENSET_EL0 },
-	{ SYS_DESC(SYS_PMOVSCLR_EL0), access_pmovs, NULL, PMOVSSET_EL0 },
+	{ SYS_DESC(SYS_PMOVSCLR_EL0), access_pmovs, reset_unknown, PMOVSSET_EL0 },
 	{ SYS_DESC(SYS_PMSWINC_EL0), access_pmswinc, reset_unknown, PMSWINC_EL0 },
 	{ SYS_DESC(SYS_PMSELR_EL0), access_pmselr, reset_unknown, PMSELR_EL0 },
 	{ SYS_DESC(SYS_PMCEID0_EL0), access_pmceid },
@@ -2073,12 +2087,37 @@ static const struct sys_reg_desc cp15_64_regs[] = {
 	{ SYS_DESC(SYS_AARCH32_CNTP_CVAL),    access_arch_timer },
 };
 
+static int check_sysreg_table(const struct sys_reg_desc *table, unsigned int n,
+			      bool is_32)
+{
+	unsigned int i;
+
+	for (i = 0; i < n; i++) {
+		if (!is_32 && table[i].reg && !table[i].reset) {
+			kvm_err("sys_reg table %p entry %d has lacks reset\n",
+				table, i);
+			return 1;
+		}
+
+		if (i && cmp_sys_reg(&table[i-1], &table[i]) >= 0) {
+			kvm_err("sys_reg table %p out of order (%d)\n", table, i - 1);
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
 /* Target specific emulation tables */
 static struct kvm_sys_reg_target_table *target_tables[KVM_ARM_NUM_TARGETS];
 
 void kvm_register_target_sys_reg_table(unsigned int target,
 				       struct kvm_sys_reg_target_table *table)
 {
+	if (check_sysreg_table(table->table64.table, table->table64.num, false) ||
+	    check_sysreg_table(table->table32.table, table->table32.num, true))
+		return;
+
 	target_tables[target] = table;
 }
 
@@ -2364,19 +2403,13 @@ static int emulate_sys_reg(struct kvm_vcpu *vcpu,
 }
 
 static void reset_sys_reg_descs(struct kvm_vcpu *vcpu,
-				const struct sys_reg_desc *table, size_t num,
+				const struct sys_reg_desc *table, size_t num)
-				unsigned long *bmap)
 {
 	unsigned long i;
 
 	for (i = 0; i < num; i++)
-		if (table[i].reset) {
+		if (table[i].reset)
-			int reg = table[i].reg;
-
 			table[i].reset(vcpu, &table[i]);
-			if (reg > 0 && reg < NR_SYS_REGS)
-				set_bit(reg, bmap);
-		}
 }
 
 /**
@@ -2832,32 +2865,18 @@ int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
 	return write_demux_regids(uindices);
 }
 
-static int check_sysreg_table(const struct sys_reg_desc *table, unsigned int n)
-{
-	unsigned int i;
-
-	for (i = 1; i < n; i++) {
-		if (cmp_sys_reg(&table[i-1], &table[i]) >= 0) {
-			kvm_err("sys_reg table %p out of order (%d)\n", table, i - 1);
-			return 1;
-		}
-	}
-
-	return 0;
-}
-
 void kvm_sys_reg_table_init(void)
 {
 	unsigned int i;
 	struct sys_reg_desc clidr;
 
 	/* Make sure tables are unique and in order. */
-	BUG_ON(check_sysreg_table(sys_reg_descs, ARRAY_SIZE(sys_reg_descs)));
+	BUG_ON(check_sysreg_table(sys_reg_descs, ARRAY_SIZE(sys_reg_descs), false));
-	BUG_ON(check_sysreg_table(cp14_regs, ARRAY_SIZE(cp14_regs)));
+	BUG_ON(check_sysreg_table(cp14_regs, ARRAY_SIZE(cp14_regs), true));
-	BUG_ON(check_sysreg_table(cp14_64_regs, ARRAY_SIZE(cp14_64_regs)));
+	BUG_ON(check_sysreg_table(cp14_64_regs, ARRAY_SIZE(cp14_64_regs), true));
-	BUG_ON(check_sysreg_table(cp15_regs, ARRAY_SIZE(cp15_regs)));
+	BUG_ON(check_sysreg_table(cp15_regs, ARRAY_SIZE(cp15_regs), true));
-	BUG_ON(check_sysreg_table(cp15_64_regs, ARRAY_SIZE(cp15_64_regs)));
+	BUG_ON(check_sysreg_table(cp15_64_regs, ARRAY_SIZE(cp15_64_regs), true));
-	BUG_ON(check_sysreg_table(invariant_sys_regs, ARRAY_SIZE(invariant_sys_regs)));
+	BUG_ON(check_sysreg_table(invariant_sys_regs, ARRAY_SIZE(invariant_sys_regs), false));
 
 	/* We abuse the reset function to overwrite the table itself. */
 	for (i = 0; i < ARRAY_SIZE(invariant_sys_regs); i++)
@@ -2893,17 +2912,10 @@ void kvm_reset_sys_regs(struct kvm_vcpu *vcpu)
 {
 	size_t num;
 	const struct sys_reg_desc *table;
-	DECLARE_BITMAP(bmap, NR_SYS_REGS) = { 0, };
 
 	/* Generic chip reset first (so target could override). */
-	reset_sys_reg_descs(vcpu, sys_reg_descs, ARRAY_SIZE(sys_reg_descs), bmap);
+	reset_sys_reg_descs(vcpu, sys_reg_descs, ARRAY_SIZE(sys_reg_descs));
 
 	table = get_target_table(vcpu->arch.target, true, &num);
-	reset_sys_reg_descs(vcpu, table, num, bmap);
+	reset_sys_reg_descs(vcpu, table, num);
-
-	for (num = 1; num < NR_SYS_REGS; num++) {
-		if (WARN(!test_bit(num, bmap),
-			 "Didn't reset __vcpu_sys_reg(%zi)\n", num))
-			break;
-	}
 }

arch/arm64/kvm/trace.h

@@ -1,216 +1,8 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-#if !defined(_TRACE_ARM64_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
+#ifndef _TRACE_ARM64_KVM_H
 #define _TRACE_ARM64_KVM_H
 
-#include <linux/tracepoint.h>
+#include "trace_arm.h"
-#include "sys_regs.h"
+#include "trace_handle_exit.h"
 
-#undef TRACE_SYSTEM
+#endif	/* _TRACE_ARM64_KVM_H */
-#define TRACE_SYSTEM kvm
-
-TRACE_EVENT(kvm_wfx_arm64,
-	TP_PROTO(unsigned long vcpu_pc, bool is_wfe),
-	TP_ARGS(vcpu_pc, is_wfe),
-
-	TP_STRUCT__entry(
-		__field(unsigned long,	vcpu_pc)
-		__field(bool,		is_wfe)
-	),
-
-	TP_fast_assign(
-		__entry->vcpu_pc = vcpu_pc;
-		__entry->is_wfe  = is_wfe;
-	),
-
-	TP_printk("guest executed wf%c at: 0x%08lx",
-		  __entry->is_wfe ? 'e' : 'i', __entry->vcpu_pc)
-);
-
-TRACE_EVENT(kvm_hvc_arm64,
-	TP_PROTO(unsigned long vcpu_pc, unsigned long r0, unsigned long imm),
-	TP_ARGS(vcpu_pc, r0, imm),
-
-	TP_STRUCT__entry(
-		__field(unsigned long, vcpu_pc)
-		__field(unsigned long, r0)
-		__field(unsigned long, imm)
-	),
-
-	TP_fast_assign(
-		__entry->vcpu_pc = vcpu_pc;
-		__entry->r0 = r0;
-		__entry->imm = imm;
-	),
-
-	TP_printk("HVC at 0x%08lx (r0: 0x%08lx, imm: 0x%lx)",
-		  __entry->vcpu_pc, __entry->r0, __entry->imm)
-);
-
-TRACE_EVENT(kvm_arm_setup_debug,
-	TP_PROTO(struct kvm_vcpu *vcpu, __u32 guest_debug),
-	TP_ARGS(vcpu, guest_debug),
-
-	TP_STRUCT__entry(
-		__field(struct kvm_vcpu *, vcpu)
-		__field(__u32, guest_debug)
-	),
-
-	TP_fast_assign(
-		__entry->vcpu = vcpu;
-		__entry->guest_debug = guest_debug;
-	),
-
-	TP_printk("vcpu: %p, flags: 0x%08x", __entry->vcpu, __entry->guest_debug)
-);
-
-TRACE_EVENT(kvm_arm_clear_debug,
-	TP_PROTO(__u32 guest_debug),
-	TP_ARGS(guest_debug),
-
-	TP_STRUCT__entry(
-		__field(__u32, guest_debug)
-	),
-
-	TP_fast_assign(
-		__entry->guest_debug = guest_debug;
-	),
-
-	TP_printk("flags: 0x%08x", __entry->guest_debug)
-);
-
-TRACE_EVENT(kvm_arm_set_dreg32,
-	TP_PROTO(const char *name, __u32 value),
-	TP_ARGS(name, value),
-
-	TP_STRUCT__entry(
-		__field(const char *, name)
-		__field(__u32, value)
-	),
-
-	TP_fast_assign(
-		__entry->name = name;
-		__entry->value = value;
-	),
-
-	TP_printk("%s: 0x%08x", __entry->name, __entry->value)
-);
-
-TRACE_DEFINE_SIZEOF(__u64);
-
-TRACE_EVENT(kvm_arm_set_regset,
-	TP_PROTO(const char *type, int len, __u64 *control, __u64 *value),
-	TP_ARGS(type, len, control, value),
-	TP_STRUCT__entry(
-		__field(const char *, name)
-		__field(int, len)
-		__array(u64, ctrls, 16)
-		__array(u64, values, 16)
-	),
-	TP_fast_assign(
-		__entry->name = type;
-		__entry->len = len;
-		memcpy(__entry->ctrls, control, len << 3);
-		memcpy(__entry->values, value, len << 3);
-	),
-	TP_printk("%d %s CTRL:%s VALUE:%s", __entry->len, __entry->name,
-		__print_array(__entry->ctrls, __entry->len, sizeof(__u64)),
-		__print_array(__entry->values, __entry->len, sizeof(__u64)))
-);
-
-TRACE_EVENT(trap_reg,
-	TP_PROTO(const char *fn, int reg, bool is_write, u64 write_value),
-	TP_ARGS(fn, reg, is_write, write_value),
-
-	TP_STRUCT__entry(
-		__field(const char *, fn)
-		__field(int, reg)
-		__field(bool, is_write)
-		__field(u64, write_value)
-	),
-
-	TP_fast_assign(
-		__entry->fn = fn;
-		__entry->reg = reg;
-		__entry->is_write = is_write;
-		__entry->write_value = write_value;
-	),
-
-	TP_printk("%s %s reg %d (0x%08llx)", __entry->fn,  __entry->is_write?"write to":"read from", __entry->reg, __entry->write_value)
-);
-
-TRACE_EVENT(kvm_handle_sys_reg,
-	TP_PROTO(unsigned long hsr),
-	TP_ARGS(hsr),
-
-	TP_STRUCT__entry(
-		__field(unsigned long,	hsr)
-	),
-
-	TP_fast_assign(
-		__entry->hsr = hsr;
-	),
-
-	TP_printk("HSR 0x%08lx", __entry->hsr)
-);
-
-TRACE_EVENT(kvm_sys_access,
-	TP_PROTO(unsigned long vcpu_pc, struct sys_reg_params *params, const struct sys_reg_desc *reg),
-	TP_ARGS(vcpu_pc, params, reg),
-
-	TP_STRUCT__entry(
-		__field(unsigned long,			vcpu_pc)
-		__field(bool,				is_write)
-		__field(const char *,			name)
-		__field(u8,				Op0)
-		__field(u8,				Op1)
-		__field(u8,				CRn)
-		__field(u8,				CRm)
-		__field(u8,				Op2)
-	),
-
-	TP_fast_assign(
-		__entry->vcpu_pc = vcpu_pc;
-		__entry->is_write = params->is_write;
-		__entry->name = reg->name;
-		__entry->Op0 = reg->Op0;
-		__entry->Op0 = reg->Op0;
-		__entry->Op1 = reg->Op1;
-		__entry->CRn = reg->CRn;
-		__entry->CRm = reg->CRm;
-		__entry->Op2 = reg->Op2;
-	),
-
-	TP_printk("PC: %lx %s (%d,%d,%d,%d,%d) %s",
-		  __entry->vcpu_pc, __entry->name ?: "UNKN",
-		  __entry->Op0, __entry->Op1, __entry->CRn,
-		  __entry->CRm, __entry->Op2,
-		  __entry->is_write ? "write" : "read")
-);
-
-TRACE_EVENT(kvm_set_guest_debug,
-	TP_PROTO(struct kvm_vcpu *vcpu, __u32 guest_debug),
-	TP_ARGS(vcpu, guest_debug),
-
-	TP_STRUCT__entry(
-		__field(struct kvm_vcpu *, vcpu)
-		__field(__u32, guest_debug)
-	),
-
-	TP_fast_assign(
-		__entry->vcpu = vcpu;
-		__entry->guest_debug = guest_debug;
-	),
-
-	TP_printk("vcpu: %p, flags: 0x%08x", __entry->vcpu, __entry->guest_debug)
-);
-
-
-#endif /* _TRACE_ARM64_KVM_H */
-
-#undef TRACE_INCLUDE_PATH
-#define TRACE_INCLUDE_PATH .
-#undef TRACE_INCLUDE_FILE
-#define TRACE_INCLUDE_FILE trace
-
-/* This part must be outside protection */
-#include <trace/define_trace.h>

arch/arm64/kvm/trace_arm.h

@@ -1,10 +1,9 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
+#if !defined(_TRACE_ARM_ARM64_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
-#define _TRACE_KVM_H
+#define _TRACE_ARM_ARM64_KVM_H
 
 #include <kvm/arm_arch_timer.h>
 #include <linux/tracepoint.h>
-#include <asm/kvm_arm.h>
 
 #undef TRACE_SYSTEM
 #define TRACE_SYSTEM kvm
@@ -368,12 +367,12 @@ TRACE_EVENT(kvm_timer_emulate,
 		  __entry->timer_idx, __entry->should_fire)
 );
 
-#endif /* _TRACE_KVM_H */
+#endif /* _TRACE_ARM_ARM64_KVM_H */
 
 #undef TRACE_INCLUDE_PATH
-#define TRACE_INCLUDE_PATH ../../virt/kvm/arm
+#define TRACE_INCLUDE_PATH .
 #undef TRACE_INCLUDE_FILE
-#define TRACE_INCLUDE_FILE trace
+#define TRACE_INCLUDE_FILE trace_arm
 
 /* This part must be outside protection */
 #include <trace/define_trace.h>

arch/arm64/kvm/trace_handle_exit.h

@@ -0,0 +1,215 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#if !defined(_TRACE_HANDLE_EXIT_ARM64_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_HANDLE_EXIT_ARM64_KVM_H
+
+#include <linux/tracepoint.h>
+#include "sys_regs.h"
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM kvm
+
+TRACE_EVENT(kvm_wfx_arm64,
+	TP_PROTO(unsigned long vcpu_pc, bool is_wfe),
+	TP_ARGS(vcpu_pc, is_wfe),
+
+	TP_STRUCT__entry(
+		__field(unsigned long,	vcpu_pc)
+		__field(bool,		is_wfe)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_pc = vcpu_pc;
+		__entry->is_wfe  = is_wfe;
+	),
+
+	TP_printk("guest executed wf%c at: 0x%08lx",
+		  __entry->is_wfe ? 'e' : 'i', __entry->vcpu_pc)
+);
+
+TRACE_EVENT(kvm_hvc_arm64,
+	TP_PROTO(unsigned long vcpu_pc, unsigned long r0, unsigned long imm),
+	TP_ARGS(vcpu_pc, r0, imm),
+
+	TP_STRUCT__entry(
+		__field(unsigned long, vcpu_pc)
+		__field(unsigned long, r0)
+		__field(unsigned long, imm)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_pc = vcpu_pc;
+		__entry->r0 = r0;
+		__entry->imm = imm;
+	),
+
+	TP_printk("HVC at 0x%08lx (r0: 0x%08lx, imm: 0x%lx)",
+		  __entry->vcpu_pc, __entry->r0, __entry->imm)
+);
+
+TRACE_EVENT(kvm_arm_setup_debug,
+	TP_PROTO(struct kvm_vcpu *vcpu, __u32 guest_debug),
+	TP_ARGS(vcpu, guest_debug),
+
+	TP_STRUCT__entry(
+		__field(struct kvm_vcpu *, vcpu)
+		__field(__u32, guest_debug)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu = vcpu;
+		__entry->guest_debug = guest_debug;
+	),
+
+	TP_printk("vcpu: %p, flags: 0x%08x", __entry->vcpu, __entry->guest_debug)
+);
+
+TRACE_EVENT(kvm_arm_clear_debug,
+	TP_PROTO(__u32 guest_debug),
+	TP_ARGS(guest_debug),
+
+	TP_STRUCT__entry(
+		__field(__u32, guest_debug)
+	),
+
+	TP_fast_assign(
+		__entry->guest_debug = guest_debug;
+	),
+
+	TP_printk("flags: 0x%08x", __entry->guest_debug)
+);
+
+TRACE_EVENT(kvm_arm_set_dreg32,
+	TP_PROTO(const char *name, __u32 value),
+	TP_ARGS(name, value),
+
+	TP_STRUCT__entry(
+		__field(const char *, name)
+		__field(__u32, value)
+	),
+
+	TP_fast_assign(
+		__entry->name = name;
+		__entry->value = value;
+	),
+
+	TP_printk("%s: 0x%08x", __entry->name, __entry->value)
+);
+
+TRACE_DEFINE_SIZEOF(__u64);
+
+TRACE_EVENT(kvm_arm_set_regset,
+	TP_PROTO(const char *type, int len, __u64 *control, __u64 *value),
+	TP_ARGS(type, len, control, value),
+	TP_STRUCT__entry(
+		__field(const char *, name)
+		__field(int, len)
+		__array(u64, ctrls, 16)
+		__array(u64, values, 16)
+	),
+	TP_fast_assign(
+		__entry->name = type;
+		__entry->len = len;
+		memcpy(__entry->ctrls, control, len << 3);
+		memcpy(__entry->values, value, len << 3);
+	),
+	TP_printk("%d %s CTRL:%s VALUE:%s", __entry->len, __entry->name,
+		__print_array(__entry->ctrls, __entry->len, sizeof(__u64)),
+		__print_array(__entry->values, __entry->len, sizeof(__u64)))
+);
+
+TRACE_EVENT(trap_reg,
+	TP_PROTO(const char *fn, int reg, bool is_write, u64 write_value),
+	TP_ARGS(fn, reg, is_write, write_value),
+
+	TP_STRUCT__entry(
+		__field(const char *, fn)
+		__field(int, reg)
+		__field(bool, is_write)
+		__field(u64, write_value)
+	),
+
+	TP_fast_assign(
+		__entry->fn = fn;
+		__entry->reg = reg;
+		__entry->is_write = is_write;
+		__entry->write_value = write_value;
+	),
+
+	TP_printk("%s %s reg %d (0x%08llx)", __entry->fn,  __entry->is_write?"write to":"read from", __entry->reg, __entry->write_value)
+);
+
+TRACE_EVENT(kvm_handle_sys_reg,
+	TP_PROTO(unsigned long hsr),
+	TP_ARGS(hsr),
+
+	TP_STRUCT__entry(
+		__field(unsigned long,	hsr)
+	),
+
+	TP_fast_assign(
+		__entry->hsr = hsr;
+	),
+
+	TP_printk("HSR 0x%08lx", __entry->hsr)
+);
+
+TRACE_EVENT(kvm_sys_access,
+	TP_PROTO(unsigned long vcpu_pc, struct sys_reg_params *params, const struct sys_reg_desc *reg),
+	TP_ARGS(vcpu_pc, params, reg),
+
+	TP_STRUCT__entry(
+		__field(unsigned long,			vcpu_pc)
+		__field(bool,				is_write)
+		__field(const char *,			name)
+		__field(u8,				Op0)
+		__field(u8,				Op1)
+		__field(u8,				CRn)
+		__field(u8,				CRm)
+		__field(u8,				Op2)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_pc = vcpu_pc;
+		__entry->is_write = params->is_write;
+		__entry->name = reg->name;
+		__entry->Op0 = reg->Op0;
+		__entry->Op0 = reg->Op0;
+		__entry->Op1 = reg->Op1;
+		__entry->CRn = reg->CRn;
+		__entry->CRm = reg->CRm;
+		__entry->Op2 = reg->Op2;
+	),
+
+	TP_printk("PC: %lx %s (%d,%d,%d,%d,%d) %s",
+		  __entry->vcpu_pc, __entry->name ?: "UNKN",
+		  __entry->Op0, __entry->Op1, __entry->CRn,
+		  __entry->CRm, __entry->Op2,
+		  __entry->is_write ? "write" : "read")
+);
+
+TRACE_EVENT(kvm_set_guest_debug,
+	TP_PROTO(struct kvm_vcpu *vcpu, __u32 guest_debug),
+	TP_ARGS(vcpu, guest_debug),
+
+	TP_STRUCT__entry(
+		__field(struct kvm_vcpu *, vcpu)
+		__field(__u32, guest_debug)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu = vcpu;
+		__entry->guest_debug = guest_debug;
+	),
+
+	TP_printk("vcpu: %p, flags: 0x%08x", __entry->vcpu, __entry->guest_debug)
+);
+
+#endif /* _TRACE_HANDLE_EXIT_ARM64_KVM_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE trace_handle_exit
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>

arch/arm64/kvm/vgic-sys-reg-v3.c

@@ -7,7 +7,7 @@
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 #include <asm/kvm_emulate.h>
-#include "vgic.h"
+#include "vgic/vgic.h"
 #include "sys_regs.h"
 
 static bool access_gic_ctlr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,

arch/arm64/kvm/vgic/trace.h

@@ -30,7 +30,7 @@ TRACE_EVENT(vgic_update_irq_pending,
 #endif /* _TRACE_VGIC_H */
 
 #undef TRACE_INCLUDE_PATH
-#define TRACE_INCLUDE_PATH ../../virt/kvm/arm/vgic
+#define TRACE_INCLUDE_PATH ../../arch/arm64/kvm/vgic
 #undef TRACE_INCLUDE_FILE
 #define TRACE_INCLUDE_FILE trace
 

arch/arm64/kvm/vgic/vgic-v2.c

@@ -56,7 +56,7 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu)
 
 	cpuif->vgic_hcr &= ~GICH_HCR_UIE;
 
-	for (lr = 0; lr < vgic_cpu->used_lrs; lr++) {
+	for (lr = 0; lr < vgic_cpu->vgic_v2.used_lrs; lr++) {
 		u32 val = cpuif->vgic_lr[lr];
 		u32 cpuid, intid = val & GICH_LR_VIRTUALID;
 		struct vgic_irq *irq;
@@ -120,7 +120,7 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu)
 		vgic_put_irq(vcpu->kvm, irq);
 	}
 
-	vgic_cpu->used_lrs = 0;
+	cpuif->used_lrs = 0;
 }
 
 /*
@@ -427,7 +427,7 @@ int vgic_v2_probe(const struct gic_kvm_info *info)
 static void save_lrs(struct kvm_vcpu *vcpu, void __iomem *base)
 {
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
-	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	u64 used_lrs = cpu_if->used_lrs;
 	u64 elrsr;
 	int i;
 
@@ -448,7 +448,7 @@ static void save_lrs(struct kvm_vcpu *vcpu, void __iomem *base)
 void vgic_v2_save_state(struct kvm_vcpu *vcpu)
 {
 	void __iomem *base = kvm_vgic_global_state.vctrl_base;
-	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	u64 used_lrs = vcpu->arch.vgic_cpu.vgic_v2.used_lrs;
 
 	if (!base)
 		return;
@@ -463,7 +463,7 @@ void vgic_v2_restore_state(struct kvm_vcpu *vcpu)
 {
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
 	void __iomem *base = kvm_vgic_global_state.vctrl_base;
-	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	u64 used_lrs = cpu_if->used_lrs;
 	int i;
 
 	if (!base)

arch/arm64/kvm/vgic/vgic-v3.c

@@ -39,7 +39,7 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
 
 	cpuif->vgic_hcr &= ~ICH_HCR_UIE;
 
-	for (lr = 0; lr < vgic_cpu->used_lrs; lr++) {
+	for (lr = 0; lr < cpuif->used_lrs; lr++) {
 		u64 val = cpuif->vgic_lr[lr];
 		u32 intid, cpuid;
 		struct vgic_irq *irq;
@@ -111,7 +111,7 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
 		vgic_put_irq(vcpu->kvm, irq);
 	}
 
-	vgic_cpu->used_lrs = 0;
+	cpuif->used_lrs = 0;
 }
 
 /* Requires the irq to be locked already */
@@ -587,7 +587,7 @@ int vgic_v3_probe(const struct gic_kvm_info *info)
 	int ret;
 
 	/*
-	 * The ListRegs field is 5 bits, but there is a architectural
+	 * The ListRegs field is 5 bits, but there is an architectural
 	 * maximum of 16 list registers. Just ignore bit 4...
 	 */
 	kvm_vgic_global_state.nr_lr = (ich_vtr_el2 & 0xf) + 1;
@@ -630,12 +630,10 @@ int vgic_v3_probe(const struct gic_kvm_info *info)
 	if (kvm_vgic_global_state.vcpu_base == 0)
 		kvm_info("disabling GICv2 emulation\n");
 
-#ifdef CONFIG_ARM64
 	if (cpus_have_const_cap(ARM64_WORKAROUND_CAVIUM_30115)) {
 		group0_trap = true;
 		group1_trap = true;
 	}
-#endif
 
 	if (group0_trap || group1_trap || common_trap) {
 		kvm_info("GICv3 sysreg trapping enabled ([%s%s%s], reduced performance)\n",
@@ -664,10 +662,10 @@ void vgic_v3_load(struct kvm_vcpu *vcpu)
 	if (likely(cpu_if->vgic_sre))
 		kvm_call_hyp(__vgic_v3_write_vmcr, cpu_if->vgic_vmcr);
 
-	kvm_call_hyp(__vgic_v3_restore_aprs, vcpu);
+	kvm_call_hyp(__vgic_v3_restore_aprs, kern_hyp_va(cpu_if));
 
 	if (has_vhe())
-		__vgic_v3_activate_traps(vcpu);
+		__vgic_v3_activate_traps(cpu_if);
 
 	WARN_ON(vgic_v4_load(vcpu));
 }
@@ -682,12 +680,14 @@ void vgic_v3_vmcr_sync(struct kvm_vcpu *vcpu)
 
 void vgic_v3_put(struct kvm_vcpu *vcpu)
 {
+	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+
 	WARN_ON(vgic_v4_put(vcpu, false));
 
 	vgic_v3_vmcr_sync(vcpu);
 
-	kvm_call_hyp(__vgic_v3_save_aprs, vcpu);
+	kvm_call_hyp(__vgic_v3_save_aprs, kern_hyp_va(cpu_if));
 
 	if (has_vhe())
-		__vgic_v3_deactivate_traps(vcpu);
+		__vgic_v3_deactivate_traps(cpu_if);
 }

arch/arm64/kvm/vgic/vgic.c

@@ -786,6 +786,7 @@ static void vgic_flush_lr_state(struct kvm_vcpu *vcpu)
 	int count;
 	bool multi_sgi;
 	u8 prio = 0xff;
+	int i = 0;
 
 	lockdep_assert_held(&vgic_cpu->ap_list_lock);
 
@@ -827,11 +828,14 @@ static void vgic_flush_lr_state(struct kvm_vcpu *vcpu)
 		}
 	}
 
-	vcpu->arch.vgic_cpu.used_lrs = count;
-
 	/* Nuke remaining LRs */
-	for ( ; count < kvm_vgic_global_state.nr_lr; count++)
+	for (i = count ; i < kvm_vgic_global_state.nr_lr; i++)
-		vgic_clear_lr(vcpu, count);
+		vgic_clear_lr(vcpu, i);
+
+	if (!static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
+		vcpu->arch.vgic_cpu.vgic_v2.used_lrs = count;
+	else
+		vcpu->arch.vgic_cpu.vgic_v3.used_lrs = count;
 }
 
 static inline bool can_access_vgic_from_kernel(void)
@@ -849,13 +853,13 @@ static inline void vgic_save_state(struct kvm_vcpu *vcpu)
 	if (!static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
 		vgic_v2_save_state(vcpu);
 	else
-		__vgic_v3_save_state(vcpu);
+		__vgic_v3_save_state(&vcpu->arch.vgic_cpu.vgic_v3);
 }
 
 /* Sync back the hardware VGIC state into our emulation after a guest's run. */
 void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu)
 {
-	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	int used_lrs;
 
 	/* An empty ap_list_head implies used_lrs == 0 */
 	if (list_empty(&vcpu->arch.vgic_cpu.ap_list_head))
@@ -864,7 +868,12 @@ void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu)
 	if (can_access_vgic_from_kernel())
 		vgic_save_state(vcpu);
 
-	if (vgic_cpu->used_lrs)
+	if (!static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
+		used_lrs = vcpu->arch.vgic_cpu.vgic_v2.used_lrs;
+	else
+		used_lrs = vcpu->arch.vgic_cpu.vgic_v3.used_lrs;
+
+	if (used_lrs)
 		vgic_fold_lr_state(vcpu);
 	vgic_prune_ap_list(vcpu);
 }
@@ -874,7 +883,7 @@ static inline void vgic_restore_state(struct kvm_vcpu *vcpu)
 	if (!static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
 		vgic_v2_restore_state(vcpu);
 	else
-		__vgic_v3_restore_state(vcpu);
+		__vgic_v3_restore_state(&vcpu->arch.vgic_cpu.vgic_v3);
 }
 
 /* Flush our emulation state into the GIC hardware before entering the guest. */

include/kvm/arm_vgic.h

@@ -274,6 +274,8 @@ struct vgic_v2_cpu_if {
 	u32		vgic_vmcr;
 	u32		vgic_apr;
 	u32		vgic_lr[VGIC_V2_MAX_LRS];
+
+	unsigned int used_lrs;
 };
 
 struct vgic_v3_cpu_if {
@@ -291,6 +293,8 @@ struct vgic_v3_cpu_if {
 	 * linking the Linux IRQ subsystem and the ITS together.
 	 */
 	struct its_vpe	its_vpe;
+
+	unsigned int used_lrs;
 };
 
 struct vgic_cpu {
@@ -300,7 +304,6 @@ struct vgic_cpu {
 		struct vgic_v3_cpu_if	vgic_v3;
 	};
 
-	unsigned int used_lrs;
 	struct vgic_irq private_irqs[VGIC_NR_PRIVATE_IRQS];
 
 	raw_spinlock_t ap_list_lock;	/* Protects the ap_list */

virt/kvm/coalesced_mmio.c

@@ -119,7 +119,7 @@ int kvm_coalesced_mmio_init(struct kvm *kvm)
 
 	/*
 	 * We're using this spinlock to sync access to the coalesced ring.
-	 * The list doesn't need it's own lock since device registration and
+	 * The list doesn't need its own lock since device registration and
 	 * unregistration should only happen when kvm->slots_lock is held.
 	 */
 	spin_lock_init(&kvm->ring_lock);

virt/kvm/eventfd.c

@@ -116,7 +116,7 @@ irqfd_shutdown(struct work_struct *work)
 	struct kvm *kvm = irqfd->kvm;
 	u64 cnt;
 
-	/* Make sure irqfd has been initalized in assign path. */
+	/* Make sure irqfd has been initialized in assign path. */
 	synchronize_srcu(&kvm->irq_srcu);
 
 	/*

virt/kvm/kvm_main.c

@@ -2825,7 +2825,7 @@ EXPORT_SYMBOL_GPL(kvm_vcpu_yield_to);
  *
  *  (a) VCPU which has not done pl-exit or cpu relax intercepted recently
  *  (preempted lock holder), indicated by @in_spin_loop.
- *  Set at the beiginning and cleared at the end of interception/PLE handler.
+ *  Set at the beginning and cleared at the end of interception/PLE handler.
  *
  *  (b) VCPU which has done pl-exit/ cpu relax intercepted but did not get
  *  chance last time (mostly it has become eligible now since we have probably