1 files changed, 70 insertions, 65 deletions

diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 3b3ecfed294f..1a479df5d78e 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -61,7 +61,7 @@ static enum kvm_wfx_trap_policy kvm_wfe_trap_policy __read_mostly = KVM_WFX_NOTR
 
 DECLARE_KVM_HYP_PER_CPU(unsigned long, kvm_hyp_vector);
 
-DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);
+DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_base);
 DECLARE_KVM_NVHE_PER_CPU(struct kvm_nvhe_init_params, kvm_init_params);
 
 DECLARE_KVM_NVHE_PER_CPU(struct kvm_cpu_context, kvm_hyp_ctxt);
@@ -80,31 +80,6 @@ int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
 	return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE;
 }
 
-/*
- * This functions as an allow-list of protected VM capabilities.
- * Features not explicitly allowed by this function are denied.
- */
-static bool pkvm_ext_allowed(struct kvm *kvm, long ext)
-{
-	switch (ext) {
-	case KVM_CAP_IRQCHIP:
-	case KVM_CAP_ARM_PSCI:
-	case KVM_CAP_ARM_PSCI_0_2:
-	case KVM_CAP_NR_VCPUS:
-	case KVM_CAP_MAX_VCPUS:
-	case KVM_CAP_MAX_VCPU_ID:
-	case KVM_CAP_MSI_DEVID:
-	case KVM_CAP_ARM_VM_IPA_SIZE:
-	case KVM_CAP_ARM_PMU_V3:
-	case KVM_CAP_ARM_SVE:
-	case KVM_CAP_ARM_PTRAUTH_ADDRESS:
-	case KVM_CAP_ARM_PTRAUTH_GENERIC:
-		return true;
-	default:
-		return false;
-	}
-}
-
 int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 			    struct kvm_enable_cap *cap)
 {
@@ -113,7 +88,7 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 	if (cap->flags)
 		return -EINVAL;
 
-	if (kvm_vm_is_protected(kvm) && !pkvm_ext_allowed(kvm, cap->cap))
+	if (kvm_vm_is_protected(kvm) && !kvm_pvm_ext_allowed(cap->cap))
 		return -EINVAL;
 
 	switch (cap->cap) {
@@ -311,7 +286,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 {
 	int r;
 
-	if (kvm && kvm_vm_is_protected(kvm) && !pkvm_ext_allowed(kvm, ext))
+	if (kvm && kvm_vm_is_protected(kvm) && !kvm_pvm_ext_allowed(ext))
 		return 0;
 
 	switch (ext) {
@@ -476,8 +451,6 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 
 	kvm_pmu_vcpu_init(vcpu);
 
-	kvm_arm_reset_debug_ptr(vcpu);
-
 	kvm_arm_pvtime_vcpu_init(&vcpu->arch);
 
 	vcpu->arch.hw_mmu = &vcpu->kvm->arch.mmu;
@@ -493,7 +466,11 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 	if (err)
 		return err;
 
-	return kvm_share_hyp(vcpu, vcpu + 1);
+	err = kvm_share_hyp(vcpu, vcpu + 1);
+	if (err)
+		kvm_vgic_vcpu_destroy(vcpu);
+
+	return err;
 }
 
 void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
@@ -502,7 +479,10 @@ void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
 
 void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 {
-	kvm_mmu_free_memory_cache(&vcpu->arch.mmu_page_cache);
+	if (!is_protected_kvm_enabled())
+		kvm_mmu_free_memory_cache(&vcpu->arch.mmu_page_cache);
+	else
+		free_hyp_memcache(&vcpu->arch.pkvm_memcache);
 	kvm_timer_vcpu_terminate(vcpu);
 	kvm_pmu_vcpu_destroy(vcpu);
 	kvm_vgic_vcpu_destroy(vcpu);
@@ -574,6 +554,9 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	struct kvm_s2_mmu *mmu;
 	int *last_ran;
 
+	if (is_protected_kvm_enabled())
+		goto nommu;
+
 	if (vcpu_has_nv(vcpu))
 		kvm_vcpu_load_hw_mmu(vcpu);
 
@@ -604,10 +587,12 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 		*last_ran = vcpu->vcpu_idx;
 	}
 
+nommu:
 	vcpu->cpu = cpu;
 
 	kvm_vgic_load(vcpu);
 	kvm_timer_vcpu_load(vcpu);
+	kvm_vcpu_load_debug(vcpu);
 	if (has_vhe())
 		kvm_vcpu_load_vhe(vcpu);
 	kvm_arch_vcpu_load_fp(vcpu);
@@ -627,7 +612,13 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 
 	vcpu_set_pauth_traps(vcpu);
 
-	kvm_arch_vcpu_load_debug_state_flags(vcpu);
+	if (is_protected_kvm_enabled()) {
+		kvm_call_hyp_nvhe(__pkvm_vcpu_load,
+				  vcpu->kvm->arch.pkvm.handle,
+				  vcpu->vcpu_idx, vcpu->arch.hcr_el2);
+		kvm_call_hyp(__vgic_v3_restore_vmcr_aprs,
+			     &vcpu->arch.vgic_cpu.vgic_v3);
+	}
 
 	if (!cpumask_test_cpu(cpu, vcpu->kvm->arch.supported_cpus))
 		vcpu_set_on_unsupported_cpu(vcpu);
@@ -635,7 +626,13 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
-	kvm_arch_vcpu_put_debug_state_flags(vcpu);
+	if (is_protected_kvm_enabled()) {
+		kvm_call_hyp(__vgic_v3_save_vmcr_aprs,
+			     &vcpu->arch.vgic_cpu.vgic_v3);
+		kvm_call_hyp_nvhe(__pkvm_vcpu_put);
+	}
+
+	kvm_vcpu_put_debug(vcpu);
 	kvm_arch_vcpu_put_fp(vcpu);
 	if (has_vhe())
 		kvm_vcpu_put_vhe(vcpu);
@@ -818,8 +815,6 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu)
 
 	kvm_init_mpidr_data(kvm);
 
-	kvm_arm_vcpu_init_debug(vcpu);
-
 	if (likely(irqchip_in_kernel(kvm))) {
 		/*
 		 * Map the VGIC hardware resources before running a vcpu the
@@ -1185,7 +1180,6 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
 			continue;
 		}
 
-		kvm_arm_setup_debug(vcpu);
 		kvm_arch_vcpu_ctxflush_fp(vcpu);
 
 		/**************************************************************
@@ -1202,8 +1196,6 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
 		 * Back from guest
 		 *************************************************************/
 
-		kvm_arm_clear_debug(vcpu);
-
 		/*
 		 * We must sync the PMU state before the vgic state so
 		 * that the vgic can properly sample the updated state of the
@@ -1226,6 +1218,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
 		if (unlikely(!irqchip_in_kernel(vcpu->kvm)))
 			kvm_timer_sync_user(vcpu);
 
+		if (is_hyp_ctxt(vcpu))
+			kvm_timer_sync_nested(vcpu);
+
 		kvm_arch_vcpu_ctxsync_fp(vcpu);
 
 		/*
@@ -1569,7 +1564,6 @@ static int kvm_arch_vcpu_ioctl_vcpu_init(struct kvm_vcpu *vcpu,
 	}
 
 	vcpu_reset_hcr(vcpu);
-	vcpu->arch.cptr_el2 = kvm_get_reset_cptr_el2(vcpu);
 
 	/*
 	 * Handle the "start in power-off" case.
@@ -1988,7 +1982,7 @@ static int kvm_init_vector_slots(void)
 static void __init cpu_prepare_hyp_mode(int cpu, u32 hyp_va_bits)
 {
 	struct kvm_nvhe_init_params *params = per_cpu_ptr_nvhe_sym(kvm_init_params, cpu);
-	unsigned long tcr, ips;
+	unsigned long tcr;
 
 	/*
 	 * Calculate the raw per-cpu offset without a translation from the
@@ -2002,19 +1996,18 @@ static void __init cpu_prepare_hyp_mode(int cpu, u32 hyp_va_bits)
 	params->mair_el2 = read_sysreg(mair_el1);
 
 	tcr = read_sysreg(tcr_el1);
-	ips = FIELD_GET(TCR_IPS_MASK, tcr);
 	if (cpus_have_final_cap(ARM64_KVM_HVHE)) {
+		tcr &= ~(TCR_HD | TCR_HA | TCR_A1 | TCR_T0SZ_MASK);
 		tcr |= TCR_EPD1_MASK;
 	} else {
+		unsigned long ips = FIELD_GET(TCR_IPS_MASK, tcr);
+
 		tcr &= TCR_EL2_MASK;
-		tcr |= TCR_EL2_RES1;
+		tcr |= TCR_EL2_RES1 | FIELD_PREP(TCR_EL2_PS_MASK, ips);
+		if (lpa2_is_enabled())
+			tcr |= TCR_EL2_DS;
 	}
-	tcr &= ~TCR_T0SZ_MASK;
 	tcr |= TCR_T0SZ(hyp_va_bits);
-	tcr &= ~TCR_EL2_PS_MASK;
-	tcr |= FIELD_PREP(TCR_EL2_PS_MASK, ips);
-	if (lpa2_is_enabled())
-		tcr |= TCR_EL2_DS;
 	params->tcr_el2 = tcr;
 
 	params->pgd_pa = kvm_mmu_get_httbr();
@@ -2107,6 +2100,7 @@ static void cpu_set_hyp_vector(void)
 static void cpu_hyp_init_context(void)
 {
 	kvm_init_host_cpu_context(host_data_ptr(host_ctxt));
+	kvm_init_host_debug_data();
 
 	if (!is_kernel_in_hyp_mode())
 		cpu_init_hyp_mode();
@@ -2115,7 +2109,6 @@ static void cpu_hyp_init_context(void)
 static void cpu_hyp_init_features(void)
 {
 	cpu_set_hyp_vector();
-	kvm_arm_init_debug();
 
 	if (is_kernel_in_hyp_mode())
 		kvm_timer_init_vhe();
@@ -2298,6 +2291,19 @@ static int __init init_subsystems(void)
 		break;
 	case -ENODEV:
 	case -ENXIO:
+		/*
+		 * No VGIC? No pKVM for you.
+		 *
+		 * Protected mode assumes that VGICv3 is present, so no point
+		 * in trying to hobble along if vgic initialization fails.
+		 */
+		if (is_protected_kvm_enabled())
+			goto out;
+
+		/*
+		 * Otherwise, userspace could choose to implement a GIC for its
+		 * guest on non-cooperative hardware.
+		 */
 		vgic_present = false;
 		err = 0;
 		break;
@@ -2337,7 +2343,7 @@ static void __init teardown_hyp_mode(void)
 
 	free_hyp_pgds();
 	for_each_possible_cpu(cpu) {
-		free_page(per_cpu(kvm_arm_hyp_stack_page, cpu));
+		free_pages(per_cpu(kvm_arm_hyp_stack_base, cpu), NVHE_STACK_SHIFT - PAGE_SHIFT);
 		free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order());
 
 		if (free_sve) {
@@ -2408,6 +2414,13 @@ static void kvm_hyp_init_symbols(void)
 	kvm_nvhe_sym(id_aa64smfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64SMFR0_EL1);
 	kvm_nvhe_sym(__icache_flags) = __icache_flags;
 	kvm_nvhe_sym(kvm_arm_vmid_bits) = kvm_arm_vmid_bits;
+
+	/*
+	 * Flush entire BSS since part of its data containing init symbols is read
+	 * while the MMU is off.
+	 */
+	kvm_flush_dcache_to_poc(kvm_ksym_ref(__hyp_bss_start),
+				kvm_ksym_ref(__hyp_bss_end) - kvm_ksym_ref(__hyp_bss_start));
 }
 
 static int __init kvm_hyp_init_protection(u32 hyp_va_bits)
@@ -2469,14 +2482,6 @@ static void finalize_init_hyp_mode(void)
 			per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state =
 				kern_hyp_va(sve_state);
 		}
-	} else {
-		for_each_possible_cpu(cpu) {
-			struct user_fpsimd_state *fpsimd_state;
-
-			fpsimd_state = &per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->host_ctxt.fp_regs;
-			per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->fpsimd_state =
-				kern_hyp_va(fpsimd_state);
-		}
 	}
 }
 
@@ -2525,15 +2530,15 @@ static int __init init_hyp_mode(void)
 	 * Allocate stack pages for Hypervisor-mode
 	 */
 	for_each_possible_cpu(cpu) {
-		unsigned long stack_page;
+		unsigned long stack_base;
 
-		stack_page = __get_free_page(GFP_KERNEL);
-		if (!stack_page) {
+		stack_base = __get_free_pages(GFP_KERNEL, NVHE_STACK_SHIFT - PAGE_SHIFT);
+		if (!stack_base) {
 			err = -ENOMEM;
 			goto out_err;
 		}
 
-		per_cpu(kvm_arm_hyp_stack_page, cpu) = stack_page;
+		per_cpu(kvm_arm_hyp_stack_base, cpu) = stack_base;
 	}
 
 	/*
@@ -2602,9 +2607,9 @@ static int __init init_hyp_mode(void)
 	 */
 	for_each_possible_cpu(cpu) {
 		struct kvm_nvhe_init_params *params = per_cpu_ptr_nvhe_sym(kvm_init_params, cpu);
-		char *stack_page = (char *)per_cpu(kvm_arm_hyp_stack_page, cpu);
+		char *stack_base = (char *)per_cpu(kvm_arm_hyp_stack_base, cpu);
 
-		err = create_hyp_stack(__pa(stack_page), &params->stack_hyp_va);
+		err = create_hyp_stack(__pa(stack_base), &params->stack_hyp_va);
 		if (err) {
 			kvm_err("Cannot map hyp stack\n");
 			goto out_err;
@@ -2616,7 +2621,7 @@ static int __init init_hyp_mode(void)
 		 * __hyp_pa() won't do the right thing there, since the stack
 		 * has been mapped in the flexible private VA space.
 		 */
-		params->stack_pa = __pa(stack_page);
+		params->stack_pa = __pa(stack_base);
 	}
 
 	for_each_possible_cpu(cpu) {