diff options
Diffstat (limited to 'arch/arm64/kvm/arm.c')
| -rw-r--r-- | arch/arm64/kvm/arm.c | 1414 |
1 files changed, 1038 insertions, 376 deletions
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index 83a7f61354d3..38a91bb5d4c7 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -16,7 +16,6 @@ #include <linux/fs.h> #include <linux/mman.h> #include <linux/sched.h> -#include <linux/kmemleak.h> #include <linux/kvm.h> #include <linux/kvm_irqfd.h> #include <linux/irqbypass.h> @@ -36,51 +35,62 @@ #include <asm/virt.h> #include <asm/kvm_arm.h> #include <asm/kvm_asm.h> -#include <asm/kvm_mmu.h> #include <asm/kvm_emulate.h> +#include <asm/kvm_mmu.h> +#include <asm/kvm_nested.h> +#include <asm/kvm_pkvm.h> +#include <asm/kvm_ptrauth.h> #include <asm/sections.h> #include <kvm/arm_hypercalls.h> #include <kvm/arm_pmu.h> #include <kvm/arm_psci.h> +#include "sys_regs.h" + static enum kvm_mode kvm_mode = KVM_MODE_DEFAULT; -DEFINE_STATIC_KEY_FALSE(kvm_protected_mode_initialized); + +enum kvm_wfx_trap_policy { + KVM_WFX_NOTRAP_SINGLE_TASK, /* Default option */ + KVM_WFX_NOTRAP, + KVM_WFX_TRAP, +}; + +static enum kvm_wfx_trap_policy kvm_wfi_trap_policy __read_mostly = KVM_WFX_NOTRAP_SINGLE_TASK; +static enum kvm_wfx_trap_policy kvm_wfe_trap_policy __read_mostly = KVM_WFX_NOTRAP_SINGLE_TASK; DECLARE_KVM_HYP_PER_CPU(unsigned long, kvm_hyp_vector); -static DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page); -unsigned long kvm_arm_hyp_percpu_base[NR_CPUS]; +DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_base); DECLARE_KVM_NVHE_PER_CPU(struct kvm_nvhe_init_params, kvm_init_params); -static bool vgic_present; +DECLARE_KVM_NVHE_PER_CPU(struct kvm_cpu_context, kvm_hyp_ctxt); -static DEFINE_PER_CPU(unsigned char, kvm_arm_hardware_enabled); -DEFINE_STATIC_KEY_FALSE(userspace_irqchip_in_use); +static bool vgic_present, kvm_arm_initialised; -int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu) -{ - return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE; -} +static DEFINE_PER_CPU(unsigned char, kvm_hyp_initialized); -int kvm_arch_hardware_setup(void *opaque) +bool is_kvm_arm_initialised(void) { - return 0; + return kvm_arm_initialised; } -int kvm_arch_check_processor_compat(void *opaque) +int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu) { - return 0; + return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE; } int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) { - int r; + int r = -EINVAL; if (cap->flags) return -EINVAL; + if (kvm_vm_is_protected(kvm) && !kvm_pvm_ext_allowed(cap->cap)) + return -EINVAL; + switch (cap->cap) { case KVM_CAP_ARM_NISV_TO_USER: r = 0; @@ -89,9 +99,7 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm, break; case KVM_CAP_ARM_MTE: mutex_lock(&kvm->lock); - if (!system_supports_mte() || kvm->created_vcpus) { - r = -EINVAL; - } else { + if (system_supports_mte() && !kvm->created_vcpus) { r = 0; set_bit(KVM_ARCH_FLAG_MTE_ENABLED, &kvm->arch.flags); } @@ -101,8 +109,31 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm, r = 0; set_bit(KVM_ARCH_FLAG_SYSTEM_SUSPEND_ENABLED, &kvm->arch.flags); break; + case KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE: + mutex_lock(&kvm->slots_lock); + /* + * To keep things simple, allow changing the chunk + * size only when no memory slots have been created. + */ + if (kvm_are_all_memslots_empty(kvm)) { + u64 new_cap = cap->args[0]; + + if (!new_cap || kvm_is_block_size_supported(new_cap)) { + r = 0; + kvm->arch.mmu.split_page_chunk_size = new_cap; + } + } + mutex_unlock(&kvm->slots_lock); + break; + case KVM_CAP_ARM_WRITABLE_IMP_ID_REGS: + mutex_lock(&kvm->lock); + if (!kvm->created_vcpus) { + r = 0; + set_bit(KVM_ARCH_FLAG_WRITABLE_IMP_ID_REGS, &kvm->arch.flags); + } + mutex_unlock(&kvm->lock); + break; default: - r = -EINVAL; break; } @@ -114,59 +145,62 @@ static int kvm_arm_default_max_vcpus(void) return vgic_present ? kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS; } -static void set_default_spectre(struct kvm *kvm) -{ - /* - * The default is to expose CSV2 == 1 if the HW isn't affected. - * Although this is a per-CPU feature, we make it global because - * asymmetric systems are just a nuisance. - * - * Userspace can override this as long as it doesn't promise - * the impossible. - */ - if (arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED) - kvm->arch.pfr0_csv2 = 1; - if (arm64_get_meltdown_state() == SPECTRE_UNAFFECTED) - kvm->arch.pfr0_csv3 = 1; -} - /** * kvm_arch_init_vm - initializes a VM data structure * @kvm: pointer to the KVM struct + * @type: kvm device type */ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) { int ret; - ret = kvm_arm_setup_stage2(kvm, type); - if (ret) - return ret; + mutex_init(&kvm->arch.config_lock); + +#ifdef CONFIG_LOCKDEP + /* Clue in lockdep that the config_lock must be taken inside kvm->lock */ + mutex_lock(&kvm->lock); + mutex_lock(&kvm->arch.config_lock); + mutex_unlock(&kvm->arch.config_lock); + mutex_unlock(&kvm->lock); +#endif + + kvm_init_nested(kvm); - ret = kvm_init_stage2_mmu(kvm, &kvm->arch.mmu); + ret = kvm_share_hyp(kvm, kvm + 1); if (ret) return ret; - ret = kvm_share_hyp(kvm, kvm + 1); + ret = pkvm_init_host_vm(kvm); if (ret) - goto out_free_stage2_pgd; + goto err_unshare_kvm; - if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL)) { + if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL_ACCOUNT)) { ret = -ENOMEM; - goto out_free_stage2_pgd; + goto err_unshare_kvm; } cpumask_copy(kvm->arch.supported_cpus, cpu_possible_mask); + ret = kvm_init_stage2_mmu(kvm, &kvm->arch.mmu, type); + if (ret) + goto err_free_cpumask; + kvm_vgic_early_init(kvm); + kvm_timer_init_vm(kvm); + /* The maximum number of VCPUs is limited by the host's GIC model */ kvm->max_vcpus = kvm_arm_default_max_vcpus(); - set_default_spectre(kvm); kvm_arm_init_hypercalls(kvm); - return ret; -out_free_stage2_pgd: - kvm_free_stage2_pgd(&kvm->arch.mmu); + bitmap_zero(kvm->arch.vcpu_features, KVM_VCPU_MAX_FEATURES); + + return 0; + +err_free_cpumask: + free_cpumask_var(kvm->arch.supported_cpus); +err_unshare_kvm: + kvm_unshare_hyp(kvm, kvm + 1); return ret; } @@ -175,6 +209,28 @@ vm_fault_t kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf) return VM_FAULT_SIGBUS; } +void kvm_arch_create_vm_debugfs(struct kvm *kvm) +{ + kvm_sys_regs_create_debugfs(kvm); + kvm_s2_ptdump_create_debugfs(kvm); +} + +static void kvm_destroy_mpidr_data(struct kvm *kvm) +{ + struct kvm_mpidr_data *data; + + mutex_lock(&kvm->arch.config_lock); + + data = rcu_dereference_protected(kvm->arch.mpidr_data, + lockdep_is_held(&kvm->arch.config_lock)); + if (data) { + rcu_assign_pointer(kvm->arch.mpidr_data, NULL); + synchronize_rcu(); + kfree(data); + } + + mutex_unlock(&kvm->arch.config_lock); +} /** * kvm_arch_destroy_vm - destroy the VM data structure @@ -187,20 +243,65 @@ void kvm_arch_destroy_vm(struct kvm *kvm) kvm_vgic_destroy(kvm); + if (is_protected_kvm_enabled()) + pkvm_destroy_hyp_vm(kvm); + + kvm_destroy_mpidr_data(kvm); + + kfree(kvm->arch.sysreg_masks); kvm_destroy_vcpus(kvm); kvm_unshare_hyp(kvm, kvm + 1); + + kvm_arm_teardown_hypercalls(kvm); +} + +static bool kvm_has_full_ptr_auth(void) +{ + bool apa, gpa, api, gpi, apa3, gpa3; + u64 isar1, isar2, val; + + /* + * Check that: + * + * - both Address and Generic auth are implemented for a given + * algorithm (Q5, IMPDEF or Q3) + * - only a single algorithm is implemented. + */ + if (!system_has_full_ptr_auth()) + return false; + + isar1 = read_sanitised_ftr_reg(SYS_ID_AA64ISAR1_EL1); + isar2 = read_sanitised_ftr_reg(SYS_ID_AA64ISAR2_EL1); + + apa = !!FIELD_GET(ID_AA64ISAR1_EL1_APA_MASK, isar1); + val = FIELD_GET(ID_AA64ISAR1_EL1_GPA_MASK, isar1); + gpa = (val == ID_AA64ISAR1_EL1_GPA_IMP); + + api = !!FIELD_GET(ID_AA64ISAR1_EL1_API_MASK, isar1); + val = FIELD_GET(ID_AA64ISAR1_EL1_GPI_MASK, isar1); + gpi = (val == ID_AA64ISAR1_EL1_GPI_IMP); + + apa3 = !!FIELD_GET(ID_AA64ISAR2_EL1_APA3_MASK, isar2); + val = FIELD_GET(ID_AA64ISAR2_EL1_GPA3_MASK, isar2); + gpa3 = (val == ID_AA64ISAR2_EL1_GPA3_IMP); + + return (apa == gpa && api == gpi && apa3 == gpa3 && + (apa + api + apa3) == 1); } int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) { int r; + + if (kvm && kvm_vm_is_protected(kvm) && !kvm_pvm_ext_allowed(ext)) + return 0; + switch (ext) { case KVM_CAP_IRQCHIP: r = vgic_present; break; case KVM_CAP_IOEVENTFD: - case KVM_CAP_DEVICE_CTRL: case KVM_CAP_USER_MEMORY: case KVM_CAP_SYNC_MMU: case KVM_CAP_DESTROY_MEMORY_REGION_WORKS: @@ -218,6 +319,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_VCPU_ATTRIBUTES: case KVM_CAP_PTP_KVM: case KVM_CAP_ARM_SYSTEM_SUSPEND: + case KVM_CAP_IRQFD_RESAMPLE: + case KVM_CAP_COUNTER_OFFSET: + case KVM_CAP_ARM_WRITABLE_IMP_ID_REGS: r = 1; break; case KVM_CAP_SET_GUEST_DEBUG2: @@ -262,7 +366,13 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = kvm_arm_pvtime_supported(); break; case KVM_CAP_ARM_EL1_32BIT: - r = cpus_have_const_cap(ARM64_HAS_32BIT_EL1); + r = cpus_have_final_cap(ARM64_HAS_32BIT_EL1); + break; + case KVM_CAP_ARM_EL2: + r = cpus_have_final_cap(ARM64_HAS_NESTED_VIRT); + break; + case KVM_CAP_ARM_EL2_E2H0: + r = cpus_have_final_cap(ARM64_HAS_HCR_NV1); break; case KVM_CAP_GUEST_DEBUG_HW_BPS: r = get_num_brps(); @@ -271,10 +381,10 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = get_num_wrps(); break; case KVM_CAP_ARM_PMU_V3: - r = kvm_arm_support_pmu_v3(); + r = kvm_supports_guest_pmuv3(); break; case KVM_CAP_ARM_INJECT_SERROR_ESR: - r = cpus_have_const_cap(ARM64_HAS_RAS_EXTN); + r = cpus_have_final_cap(ARM64_HAS_RAS_EXTN); break; case KVM_CAP_ARM_VM_IPA_SIZE: r = get_kvm_ipa_limit(); @@ -284,7 +394,19 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) break; case KVM_CAP_ARM_PTRAUTH_ADDRESS: case KVM_CAP_ARM_PTRAUTH_GENERIC: - r = system_has_full_ptr_auth(); + r = kvm_has_full_ptr_auth(); + break; + case KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE: + if (kvm) + r = kvm->arch.mmu.split_page_chunk_size; + else + r = KVM_ARM_EAGER_SPLIT_CHUNK_SIZE_DEFAULT; + break; + case KVM_CAP_ARM_SUPPORTED_BLOCK_SIZES: + r = kvm_supported_block_sizes(); + break; + case KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES: + r = BIT(0); break; default: r = 0; @@ -324,9 +446,18 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu) { int err; + spin_lock_init(&vcpu->arch.mp_state_lock); + +#ifdef CONFIG_LOCKDEP + /* Inform lockdep that the config_lock is acquired after vcpu->mutex */ + mutex_lock(&vcpu->mutex); + mutex_lock(&vcpu->kvm->arch.config_lock); + mutex_unlock(&vcpu->kvm->arch.config_lock); + mutex_unlock(&vcpu->mutex); +#endif + /* Force users to call KVM_ARM_VCPU_INIT */ - vcpu->arch.target = -1; - bitmap_zero(vcpu->arch.features, KVM_VCPU_MAX_FEATURES); + vcpu_clear_flag(vcpu, VCPU_INITIALIZED); vcpu->arch.mmu_page_cache.gfp_zero = __GFP_ZERO; @@ -335,17 +466,26 @@ 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; + /* + * This vCPU may have been created after mpidr_data was initialized. + * Throw out the pre-computed mappings if that is the case which forces + * KVM to fall back to iteratively searching the vCPUs. + */ + kvm_destroy_mpidr_data(vcpu->kvm); + err = kvm_vgic_vcpu_init(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) @@ -354,13 +494,13 @@ void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu) void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) { - if (vcpu_has_run_once(vcpu) && unlikely(!irqchip_in_kernel(vcpu->kvm))) - static_branch_dec(&userspace_irqchip_in_use); - - 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); kvm_arm_vcpu_destroy(vcpu); } @@ -374,15 +514,81 @@ void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu) } +static void vcpu_set_pauth_traps(struct kvm_vcpu *vcpu) +{ + if (vcpu_has_ptrauth(vcpu) && !is_protected_kvm_enabled()) { + /* + * Either we're running an L2 guest, and the API/APK bits come + * from L1's HCR_EL2, or API/APK are both set. + */ + if (unlikely(vcpu_has_nv(vcpu) && !is_hyp_ctxt(vcpu))) { + u64 val; + + val = __vcpu_sys_reg(vcpu, HCR_EL2); + val &= (HCR_API | HCR_APK); + vcpu->arch.hcr_el2 &= ~(HCR_API | HCR_APK); + vcpu->arch.hcr_el2 |= val; + } else { + vcpu->arch.hcr_el2 |= (HCR_API | HCR_APK); + } + + /* + * Save the host keys if there is any chance for the guest + * to use pauth, as the entry code will reload the guest + * keys in that case. + */ + if (vcpu->arch.hcr_el2 & (HCR_API | HCR_APK)) { + struct kvm_cpu_context *ctxt; + + ctxt = this_cpu_ptr_hyp_sym(kvm_hyp_ctxt); + ptrauth_save_keys(ctxt); + } + } +} + +static bool kvm_vcpu_should_clear_twi(struct kvm_vcpu *vcpu) +{ + if (unlikely(kvm_wfi_trap_policy != KVM_WFX_NOTRAP_SINGLE_TASK)) + return kvm_wfi_trap_policy == KVM_WFX_NOTRAP; + + return single_task_running() && + (atomic_read(&vcpu->arch.vgic_cpu.vgic_v3.its_vpe.vlpi_count) || + vcpu->kvm->arch.vgic.nassgireq); +} + +static bool kvm_vcpu_should_clear_twe(struct kvm_vcpu *vcpu) +{ + if (unlikely(kvm_wfe_trap_policy != KVM_WFX_NOTRAP_SINGLE_TASK)) + return kvm_wfe_trap_policy == KVM_WFX_NOTRAP; + + return single_task_running(); +} + 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); + mmu = vcpu->arch.hw_mmu; last_ran = this_cpu_ptr(mmu->last_vcpu_ran); /* + * Ensure a VMID is allocated for the MMU before programming VTTBR_EL2, + * which happens eagerly in VHE. + * + * Also, the VMID allocator only preserves VMIDs that are active at the + * time of rollover, so KVM might need to grab a new VMID for the MMU if + * this is called from kvm_sched_in(). + */ + kvm_arm_vmid_update(&mmu->vmid); + + /* * We guarantee that both TLBs and I-cache are private to each * vcpu. If detecting that a vcpu from the same VM has * previously run on the same physical CPU, call into the @@ -391,78 +597,110 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) * We might get preempted before the vCPU actually runs, but * over-invalidation doesn't affect correctness. */ - if (*last_ran != vcpu->vcpu_id) { + if (*last_ran != vcpu->vcpu_idx) { kvm_call_hyp(__kvm_flush_cpu_context, mmu); - *last_ran = vcpu->vcpu_id; + *last_ran = vcpu->vcpu_idx; } +nommu: vcpu->cpu = cpu; - kvm_vgic_load(vcpu); + /* + * The timer must be loaded before the vgic to correctly set up physical + * interrupt deactivation in nested state (e.g. timer interrupt). + */ kvm_timer_vcpu_load(vcpu); + kvm_vgic_load(vcpu); + kvm_vcpu_load_debug(vcpu); if (has_vhe()) - kvm_vcpu_load_sysregs_vhe(vcpu); + kvm_vcpu_load_vhe(vcpu); kvm_arch_vcpu_load_fp(vcpu); kvm_vcpu_pmu_restore_guest(vcpu); if (kvm_arm_is_pvtime_enabled(&vcpu->arch)) kvm_make_request(KVM_REQ_RECORD_STEAL, vcpu); - if (single_task_running()) - vcpu_clear_wfx_traps(vcpu); + if (kvm_vcpu_should_clear_twe(vcpu)) + vcpu->arch.hcr_el2 &= ~HCR_TWE; + else + vcpu->arch.hcr_el2 |= HCR_TWE; + + if (kvm_vcpu_should_clear_twi(vcpu)) + vcpu->arch.hcr_el2 &= ~HCR_TWI; else - vcpu_set_wfx_traps(vcpu); + vcpu->arch.hcr_el2 |= HCR_TWI; - if (vcpu_has_ptrauth(vcpu)) - vcpu_ptrauth_disable(vcpu); - kvm_arch_vcpu_load_debug_state_flags(vcpu); + vcpu_set_pauth_traps(vcpu); - if (!cpumask_test_cpu(smp_processor_id(), vcpu->kvm->arch.supported_cpus)) + 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); } 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_sysregs_vhe(vcpu); + kvm_vcpu_put_vhe(vcpu); kvm_timer_vcpu_put(vcpu); kvm_vgic_put(vcpu); kvm_vcpu_pmu_restore_host(vcpu); + if (vcpu_has_nv(vcpu)) + kvm_vcpu_put_hw_mmu(vcpu); kvm_arm_vmid_clear_active(); vcpu_clear_on_unsupported_cpu(vcpu); vcpu->cpu = -1; } -void kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu) +static void __kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu) { - vcpu->arch.mp_state.mp_state = KVM_MP_STATE_STOPPED; + WRITE_ONCE(vcpu->arch.mp_state.mp_state, KVM_MP_STATE_STOPPED); kvm_make_request(KVM_REQ_SLEEP, vcpu); kvm_vcpu_kick(vcpu); } +void kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu) +{ + spin_lock(&vcpu->arch.mp_state_lock); + __kvm_arm_vcpu_power_off(vcpu); + spin_unlock(&vcpu->arch.mp_state_lock); +} + bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu) { - return vcpu->arch.mp_state.mp_state == KVM_MP_STATE_STOPPED; + return READ_ONCE(vcpu->arch.mp_state.mp_state) == KVM_MP_STATE_STOPPED; } static void kvm_arm_vcpu_suspend(struct kvm_vcpu *vcpu) { - vcpu->arch.mp_state.mp_state = KVM_MP_STATE_SUSPENDED; + WRITE_ONCE(vcpu->arch.mp_state.mp_state, KVM_MP_STATE_SUSPENDED); kvm_make_request(KVM_REQ_SUSPEND, vcpu); kvm_vcpu_kick(vcpu); } static bool kvm_arm_vcpu_suspended(struct kvm_vcpu *vcpu) { - return vcpu->arch.mp_state.mp_state == KVM_MP_STATE_SUSPENDED; + return READ_ONCE(vcpu->arch.mp_state.mp_state) == KVM_MP_STATE_SUSPENDED; } int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, struct kvm_mp_state *mp_state) { - *mp_state = vcpu->arch.mp_state; + *mp_state = READ_ONCE(vcpu->arch.mp_state); return 0; } @@ -472,12 +710,14 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, { int ret = 0; + spin_lock(&vcpu->arch.mp_state_lock); + switch (mp_state->mp_state) { case KVM_MP_STATE_RUNNABLE: - vcpu->arch.mp_state = *mp_state; + WRITE_ONCE(vcpu->arch.mp_state, *mp_state); break; case KVM_MP_STATE_STOPPED: - kvm_arm_vcpu_power_off(vcpu); + __kvm_arm_vcpu_power_off(vcpu); break; case KVM_MP_STATE_SUSPENDED: kvm_arm_vcpu_suspend(vcpu); @@ -486,6 +726,8 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, ret = -EINVAL; } + spin_unlock(&vcpu->arch.mp_state_lock); + return ret; } @@ -515,9 +757,56 @@ unsigned long kvm_arch_vcpu_get_ip(struct kvm_vcpu *vcpu) } #endif -static int kvm_vcpu_initialized(struct kvm_vcpu *vcpu) +static void kvm_init_mpidr_data(struct kvm *kvm) { - return vcpu->arch.target >= 0; + struct kvm_mpidr_data *data = NULL; + unsigned long c, mask, nr_entries; + u64 aff_set = 0, aff_clr = ~0UL; + struct kvm_vcpu *vcpu; + + mutex_lock(&kvm->arch.config_lock); + + if (rcu_access_pointer(kvm->arch.mpidr_data) || + atomic_read(&kvm->online_vcpus) == 1) + goto out; + + kvm_for_each_vcpu(c, vcpu, kvm) { + u64 aff = kvm_vcpu_get_mpidr_aff(vcpu); + aff_set |= aff; + aff_clr &= aff; + } + + /* + * A significant bit can be either 0 or 1, and will only appear in + * aff_set. Use aff_clr to weed out the useless stuff. + */ + mask = aff_set ^ aff_clr; + nr_entries = BIT_ULL(hweight_long(mask)); + + /* + * Don't let userspace fool us. If we need more than a single page + * to describe the compressed MPIDR array, just fall back to the + * iterative method. Single vcpu VMs do not need this either. + */ + if (struct_size(data, cmpidr_to_idx, nr_entries) <= PAGE_SIZE) + data = kzalloc(struct_size(data, cmpidr_to_idx, nr_entries), + GFP_KERNEL_ACCOUNT); + + if (!data) + goto out; + + data->mpidr_mask = mask; + + kvm_for_each_vcpu(c, vcpu, kvm) { + u64 aff = kvm_vcpu_get_mpidr_aff(vcpu); + u16 index = kvm_mpidr_index(data, aff); + + data->cmpidr_to_idx[index] = c; + } + + rcu_assign_pointer(kvm->arch.mpidr_data, data); +out: + mutex_unlock(&kvm->arch.config_lock); } /* @@ -543,7 +832,7 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu) if (likely(vcpu_has_run_once(vcpu))) return 0; - kvm_arm_vcpu_init_debug(vcpu); + kvm_init_mpidr_data(kvm); if (likely(irqchip_in_kernel(kvm))) { /* @@ -555,33 +844,49 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu) return ret; } - ret = kvm_timer_enable(vcpu); + ret = kvm_finalize_sys_regs(vcpu); if (ret) return ret; - ret = kvm_arm_pmu_v3_enable(vcpu); - if (ret) - return ret; + if (vcpu_has_nv(vcpu)) { + ret = kvm_vcpu_allocate_vncr_tlb(vcpu); + if (ret) + return ret; - if (!irqchip_in_kernel(kvm)) { - /* - * Tell the rest of the code that there are userspace irqchip - * VMs in the wild. - */ - static_branch_inc(&userspace_irqchip_in_use); + ret = kvm_vgic_vcpu_nv_init(vcpu); + if (ret) + return ret; } /* - * Initialize traps for protected VMs. - * NOTE: Move to run in EL2 directly, rather than via a hypercall, once - * the code is in place for first run initialization at EL2. + * This needs to happen after any restriction has been applied + * to the feature set. */ - if (kvm_vm_is_protected(kvm)) - kvm_call_hyp_nvhe(__pkvm_vcpu_init_traps, vcpu); + kvm_calculate_traps(vcpu); - mutex_lock(&kvm->lock); + ret = kvm_timer_enable(vcpu); + if (ret) + return ret; + + if (kvm_vcpu_has_pmu(vcpu)) { + ret = kvm_arm_pmu_v3_enable(vcpu); + if (ret) + return ret; + } + + if (is_protected_kvm_enabled()) { + ret = pkvm_create_hyp_vm(kvm); + if (ret) + return ret; + + ret = pkvm_create_hyp_vcpu(vcpu); + if (ret) + return ret; + } + + mutex_lock(&kvm->arch.config_lock); set_bit(KVM_ARCH_FLAG_HAS_RAN_ONCE, &kvm->arch.flags); - mutex_unlock(&kvm->lock); + mutex_unlock(&kvm->arch.config_lock); return ret; } @@ -654,16 +959,16 @@ void kvm_vcpu_wfi(struct kvm_vcpu *vcpu) * doorbells to be signalled, should an interrupt become pending. */ preempt_disable(); - kvm_vgic_vmcr_sync(vcpu); - vgic_v4_put(vcpu, true); + vcpu_set_flag(vcpu, IN_WFI); + kvm_vgic_put(vcpu); preempt_enable(); kvm_vcpu_halt(vcpu); - vcpu->arch.flags &= ~KVM_ARM64_WFIT; - kvm_clear_request(KVM_REQ_UNHALT, vcpu); + vcpu_clear_flag(vcpu, IN_WFIT); preempt_disable(); - vgic_v4_load(vcpu); + vcpu_clear_flag(vcpu, IN_WFI); + kvm_vgic_load(vcpu); preempt_enable(); } @@ -712,6 +1017,9 @@ static int kvm_vcpu_suspend(struct kvm_vcpu *vcpu) static int check_vcpu_requests(struct kvm_vcpu *vcpu) { if (kvm_request_pending(vcpu)) { + if (kvm_check_request(KVM_REQ_VM_DEAD, vcpu)) + return -EIO; + if (kvm_check_request(KVM_REQ_SLEEP, vcpu)) kvm_vcpu_sleep(vcpu); @@ -730,17 +1038,24 @@ static int check_vcpu_requests(struct kvm_vcpu *vcpu) if (kvm_check_request(KVM_REQ_RELOAD_GICv4, vcpu)) { /* The distributor enable bits were changed */ preempt_disable(); - vgic_v4_put(vcpu, false); + vgic_v4_put(vcpu); vgic_v4_load(vcpu); preempt_enable(); } if (kvm_check_request(KVM_REQ_RELOAD_PMU, vcpu)) - kvm_pmu_handle_pmcr(vcpu, - __vcpu_sys_reg(vcpu, PMCR_EL0)); + kvm_vcpu_reload_pmu(vcpu); + + if (kvm_check_request(KVM_REQ_RESYNC_PMU_EL0, vcpu)) + kvm_vcpu_pmu_restore_guest(vcpu); if (kvm_check_request(KVM_REQ_SUSPEND, vcpu)) return kvm_vcpu_suspend(vcpu); + + if (kvm_dirty_ring_check_request(vcpu)) + return 0; + + check_nested_vcpu_requests(vcpu); } return 1; @@ -751,8 +1066,10 @@ static bool vcpu_mode_is_bad_32bit(struct kvm_vcpu *vcpu) if (likely(!vcpu_mode_is_32bit(vcpu))) return false; - return !system_supports_32bit_el0() || - static_branch_unlikely(&arm64_mismatched_32bit_el0); + if (vcpu_has_nv(vcpu)) + return true; + + return !kvm_supports_32bit_el0(); } /** @@ -780,7 +1097,7 @@ static bool kvm_vcpu_exit_request(struct kvm_vcpu *vcpu, int *ret) * state gets updated in kvm_timer_update_run and * kvm_pmu_update_run below). */ - if (static_branch_unlikely(&userspace_irqchip_in_use)) { + if (unlikely(!irqchip_in_kernel(vcpu->kvm))) { if (kvm_timer_should_notify_user(vcpu) || kvm_pmu_should_notify_user(vcpu)) { *ret = -EINTR; @@ -836,13 +1153,13 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) if (run->exit_reason == KVM_EXIT_MMIO) { ret = kvm_handle_mmio_return(vcpu); - if (ret) + if (ret <= 0) return ret; } vcpu_load(vcpu); - if (run->immediate_exit) { + if (!vcpu->wants_to_run) { ret = -EINTR; goto out; } @@ -870,16 +1187,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) */ preempt_disable(); - /* - * The VMID allocator only tracks active VMIDs per - * physical CPU, and therefore the VMID allocated may not be - * preserved on VMID roll-over if the task was preempted, - * making a thread's VMID inactive. So we need to call - * kvm_arm_vmid_update() in non-premptible context. - */ - kvm_arm_vmid_update(&vcpu->arch.hw_mmu->vmid); - - kvm_pmu_flush_hwstate(vcpu); + if (kvm_vcpu_has_pmu(vcpu)) + kvm_pmu_flush_hwstate(vcpu); local_irq_disable(); @@ -898,8 +1207,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) if (ret <= 0 || kvm_vcpu_exit_request(vcpu, &ret)) { vcpu->mode = OUTSIDE_GUEST_MODE; isb(); /* Ensure work in x_flush_hwstate is committed */ - kvm_pmu_sync_hwstate(vcpu); - if (static_branch_unlikely(&userspace_irqchip_in_use)) + if (kvm_vcpu_has_pmu(vcpu)) + kvm_pmu_sync_hwstate(vcpu); + if (unlikely(!irqchip_in_kernel(vcpu->kvm))) kvm_timer_sync_user(vcpu); kvm_vgic_sync_hwstate(vcpu); local_irq_enable(); @@ -907,7 +1217,6 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) continue; } - kvm_arm_setup_debug(vcpu); kvm_arch_vcpu_ctxflush_fp(vcpu); /************************************************************** @@ -924,14 +1233,13 @@ 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 * interrupt line. */ - kvm_pmu_sync_hwstate(vcpu); + if (kvm_vcpu_has_pmu(vcpu)) + kvm_pmu_sync_hwstate(vcpu); /* * Sync the vgic state before syncing the timer state because @@ -945,9 +1253,12 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) * we don't want vtimer interrupts to race with syncing the * timer virtual interrupt state. */ - if (static_branch_unlikely(&userspace_irqchip_in_use)) + 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); /* @@ -992,7 +1303,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) * invalid. The VMM can try and fix it by issuing a * KVM_ARM_VCPU_INIT if it really wants to. */ - vcpu->arch.target = -1; + vcpu_clear_flag(vcpu, VCPU_INITIALIZED); ret = ARM_EXCEPTION_IL; } @@ -1015,8 +1326,8 @@ out: * the vcpu state. Note that this relies on __kvm_adjust_pc() * being preempt-safe on VHE. */ - if (unlikely(vcpu->arch.flags & (KVM_ARM64_PENDING_EXCEPTION | - KVM_ARM64_INCREMENT_PC))) + if (unlikely(vcpu_get_flag(vcpu, PENDING_EXCEPTION) || + vcpu_get_flag(vcpu, INCREMENT_PC))) kvm_call_hyp(__kvm_adjust_pc, vcpu); vcpu_put(vcpu); @@ -1061,27 +1372,23 @@ int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_level, bool line_status) { u32 irq = irq_level->irq; - unsigned int irq_type, vcpu_idx, irq_num; - int nrcpus = atomic_read(&kvm->online_vcpus); + unsigned int irq_type, vcpu_id, irq_num; struct kvm_vcpu *vcpu = NULL; bool level = irq_level->level; irq_type = (irq >> KVM_ARM_IRQ_TYPE_SHIFT) & KVM_ARM_IRQ_TYPE_MASK; - vcpu_idx = (irq >> KVM_ARM_IRQ_VCPU_SHIFT) & KVM_ARM_IRQ_VCPU_MASK; - vcpu_idx += ((irq >> KVM_ARM_IRQ_VCPU2_SHIFT) & KVM_ARM_IRQ_VCPU2_MASK) * (KVM_ARM_IRQ_VCPU_MASK + 1); + vcpu_id = (irq >> KVM_ARM_IRQ_VCPU_SHIFT) & KVM_ARM_IRQ_VCPU_MASK; + vcpu_id += ((irq >> KVM_ARM_IRQ_VCPU2_SHIFT) & KVM_ARM_IRQ_VCPU2_MASK) * (KVM_ARM_IRQ_VCPU_MASK + 1); irq_num = (irq >> KVM_ARM_IRQ_NUM_SHIFT) & KVM_ARM_IRQ_NUM_MASK; - trace_kvm_irq_line(irq_type, vcpu_idx, irq_num, irq_level->level); + trace_kvm_irq_line(irq_type, vcpu_id, irq_num, irq_level->level); switch (irq_type) { case KVM_ARM_IRQ_TYPE_CPU: if (irqchip_in_kernel(kvm)) return -ENXIO; - if (vcpu_idx >= nrcpus) - return -EINVAL; - - vcpu = kvm_get_vcpu(kvm, vcpu_idx); + vcpu = kvm_get_vcpu_by_id(kvm, vcpu_id); if (!vcpu) return -EINVAL; @@ -1093,17 +1400,14 @@ int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_level, if (!irqchip_in_kernel(kvm)) return -ENXIO; - if (vcpu_idx >= nrcpus) - return -EINVAL; - - vcpu = kvm_get_vcpu(kvm, vcpu_idx); + vcpu = kvm_get_vcpu_by_id(kvm, vcpu_id); if (!vcpu) return -EINVAL; if (irq_num < VGIC_NR_SGIS || irq_num >= VGIC_NR_PRIVATE_IRQS) return -EINVAL; - return kvm_vgic_inject_irq(kvm, vcpu->vcpu_id, irq_num, level, NULL); + return kvm_vgic_inject_irq(kvm, vcpu, irq_num, level, NULL); case KVM_ARM_IRQ_TYPE_SPI: if (!irqchip_in_kernel(kvm)) return -ENXIO; @@ -1111,64 +1415,172 @@ int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_level, if (irq_num < VGIC_NR_PRIVATE_IRQS) return -EINVAL; - return kvm_vgic_inject_irq(kvm, 0, irq_num, level, NULL); + return kvm_vgic_inject_irq(kvm, NULL, irq_num, level, NULL); } return -EINVAL; } -static int kvm_vcpu_set_target(struct kvm_vcpu *vcpu, - const struct kvm_vcpu_init *init) +static unsigned long system_supported_vcpu_features(void) { - unsigned int i, ret; - u32 phys_target = kvm_target_cpu(); + unsigned long features = KVM_VCPU_VALID_FEATURES; + + if (!cpus_have_final_cap(ARM64_HAS_32BIT_EL1)) + clear_bit(KVM_ARM_VCPU_EL1_32BIT, &features); + + if (!kvm_supports_guest_pmuv3()) + clear_bit(KVM_ARM_VCPU_PMU_V3, &features); + + if (!system_supports_sve()) + clear_bit(KVM_ARM_VCPU_SVE, &features); + + if (!kvm_has_full_ptr_auth()) { + clear_bit(KVM_ARM_VCPU_PTRAUTH_ADDRESS, &features); + clear_bit(KVM_ARM_VCPU_PTRAUTH_GENERIC, &features); + } + + if (!cpus_have_final_cap(ARM64_HAS_NESTED_VIRT)) + clear_bit(KVM_ARM_VCPU_HAS_EL2, &features); + + return features; +} - if (init->target != phys_target) +static int kvm_vcpu_init_check_features(struct kvm_vcpu *vcpu, + const struct kvm_vcpu_init *init) +{ + unsigned long features = init->features[0]; + int i; + + if (features & ~KVM_VCPU_VALID_FEATURES) + return -ENOENT; + + for (i = 1; i < ARRAY_SIZE(init->features); i++) { + if (init->features[i]) + return -ENOENT; + } + + if (features & ~system_supported_vcpu_features()) return -EINVAL; /* - * Secondary and subsequent calls to KVM_ARM_VCPU_INIT must - * use the same target. + * For now make sure that both address/generic pointer authentication + * features are requested by the userspace together. */ - if (vcpu->arch.target != -1 && vcpu->arch.target != init->target) + if (test_bit(KVM_ARM_VCPU_PTRAUTH_ADDRESS, &features) != + test_bit(KVM_ARM_VCPU_PTRAUTH_GENERIC, &features)) return -EINVAL; - /* -ENOENT for unknown features, -EINVAL for invalid combinations. */ - for (i = 0; i < sizeof(init->features) * 8; i++) { - bool set = (init->features[i / 32] & (1 << (i % 32))); + if (!test_bit(KVM_ARM_VCPU_EL1_32BIT, &features)) + return 0; - if (set && i >= KVM_VCPU_MAX_FEATURES) - return -ENOENT; + /* MTE is incompatible with AArch32 */ + if (kvm_has_mte(vcpu->kvm)) + return -EINVAL; - /* - * Secondary and subsequent calls to KVM_ARM_VCPU_INIT must - * use the same feature set. - */ - if (vcpu->arch.target != -1 && i < KVM_VCPU_MAX_FEATURES && - test_bit(i, vcpu->arch.features) != set) - return -EINVAL; + /* NV is incompatible with AArch32 */ + if (test_bit(KVM_ARM_VCPU_HAS_EL2, &features)) + return -EINVAL; - if (set) - set_bit(i, vcpu->arch.features); - } + return 0; +} - vcpu->arch.target = phys_target; +static bool kvm_vcpu_init_changed(struct kvm_vcpu *vcpu, + const struct kvm_vcpu_init *init) +{ + unsigned long features = init->features[0]; + + return !bitmap_equal(vcpu->kvm->arch.vcpu_features, &features, + KVM_VCPU_MAX_FEATURES); +} + +static int kvm_setup_vcpu(struct kvm_vcpu *vcpu) +{ + struct kvm *kvm = vcpu->kvm; + int ret = 0; + + /* + * When the vCPU has a PMU, but no PMU is set for the guest + * yet, set the default one. + */ + if (kvm_vcpu_has_pmu(vcpu) && !kvm->arch.arm_pmu) + ret = kvm_arm_set_default_pmu(kvm); + + /* Prepare for nested if required */ + if (!ret && vcpu_has_nv(vcpu)) + ret = kvm_vcpu_init_nested(vcpu); + + return ret; +} + +static int __kvm_vcpu_set_target(struct kvm_vcpu *vcpu, + const struct kvm_vcpu_init *init) +{ + unsigned long features = init->features[0]; + struct kvm *kvm = vcpu->kvm; + int ret = -EINVAL; + + mutex_lock(&kvm->arch.config_lock); + + if (test_bit(KVM_ARCH_FLAG_VCPU_FEATURES_CONFIGURED, &kvm->arch.flags) && + kvm_vcpu_init_changed(vcpu, init)) + goto out_unlock; + + bitmap_copy(kvm->arch.vcpu_features, &features, KVM_VCPU_MAX_FEATURES); + + ret = kvm_setup_vcpu(vcpu); + if (ret) + goto out_unlock; /* Now we know what it is, we can reset it. */ - ret = kvm_reset_vcpu(vcpu); - if (ret) { - vcpu->arch.target = -1; - bitmap_zero(vcpu->arch.features, KVM_VCPU_MAX_FEATURES); - } + kvm_reset_vcpu(vcpu); + set_bit(KVM_ARCH_FLAG_VCPU_FEATURES_CONFIGURED, &kvm->arch.flags); + vcpu_set_flag(vcpu, VCPU_INITIALIZED); + ret = 0; +out_unlock: + mutex_unlock(&kvm->arch.config_lock); return ret; } +static int kvm_vcpu_set_target(struct kvm_vcpu *vcpu, + const struct kvm_vcpu_init *init) +{ + int ret; + + if (init->target != KVM_ARM_TARGET_GENERIC_V8 && + init->target != kvm_target_cpu()) + return -EINVAL; + + ret = kvm_vcpu_init_check_features(vcpu, init); + if (ret) + return ret; + + if (!kvm_vcpu_initialized(vcpu)) + return __kvm_vcpu_set_target(vcpu, init); + + if (kvm_vcpu_init_changed(vcpu, init)) + return -EINVAL; + + kvm_reset_vcpu(vcpu); + return 0; +} + static int kvm_arch_vcpu_ioctl_vcpu_init(struct kvm_vcpu *vcpu, struct kvm_vcpu_init *init) { + bool power_off = false; int ret; + /* + * Treat the power-off vCPU feature as ephemeral. Clear the bit to avoid + * reflecting it in the finalized feature set, thus limiting its scope + * to a single KVM_ARM_VCPU_INIT call. + */ + if (init->features[0] & BIT(KVM_ARM_VCPU_POWER_OFF)) { + init->features[0] &= ~BIT(KVM_ARM_VCPU_POWER_OFF); + power_off = true; + } + ret = kvm_vcpu_set_target(vcpu, init); if (ret) return ret; @@ -1190,15 +1602,18 @@ static int kvm_arch_vcpu_ioctl_vcpu_init(struct kvm_vcpu *vcpu, } vcpu_reset_hcr(vcpu); - vcpu->arch.cptr_el2 = CPTR_EL2_DEFAULT; /* * Handle the "start in power-off" case. */ - if (test_bit(KVM_ARM_VCPU_POWER_OFF, vcpu->arch.features)) - kvm_arm_vcpu_power_off(vcpu); + spin_lock(&vcpu->arch.mp_state_lock); + + if (power_off) + __kvm_arm_vcpu_power_off(vcpu); else - vcpu->arch.mp_state.mp_state = KVM_MP_STATE_RUNNABLE; + WRITE_ONCE(vcpu->arch.mp_state.mp_state, KVM_MP_STATE_RUNNABLE); + + spin_unlock(&vcpu->arch.mp_state_lock); return 0; } @@ -1405,37 +1820,44 @@ void kvm_arch_sync_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot) } -void kvm_arch_flush_remote_tlbs_memslot(struct kvm *kvm, - const struct kvm_memory_slot *memslot) -{ - kvm_flush_remote_tlbs(kvm); -} - static int kvm_vm_ioctl_set_device_addr(struct kvm *kvm, struct kvm_arm_device_addr *dev_addr) { - unsigned long dev_id, type; - - dev_id = (dev_addr->id & KVM_ARM_DEVICE_ID_MASK) >> - KVM_ARM_DEVICE_ID_SHIFT; - type = (dev_addr->id & KVM_ARM_DEVICE_TYPE_MASK) >> - KVM_ARM_DEVICE_TYPE_SHIFT; - - switch (dev_id) { + switch (FIELD_GET(KVM_ARM_DEVICE_ID_MASK, dev_addr->id)) { case KVM_ARM_DEVICE_VGIC_V2: if (!vgic_present) return -ENXIO; - return kvm_vgic_addr(kvm, type, &dev_addr->addr, true); + return kvm_set_legacy_vgic_v2_addr(kvm, dev_addr); default: return -ENODEV; } } -long kvm_arch_vm_ioctl(struct file *filp, - unsigned int ioctl, unsigned long arg) +static int kvm_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + switch (attr->group) { + case KVM_ARM_VM_SMCCC_CTRL: + return kvm_vm_smccc_has_attr(kvm, attr); + default: + return -ENXIO; + } +} + +static int kvm_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + switch (attr->group) { + case KVM_ARM_VM_SMCCC_CTRL: + return kvm_vm_smccc_set_attr(kvm, attr); + default: + return -ENXIO; + } +} + +int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { struct kvm *kvm = filp->private_data; void __user *argp = (void __user *)arg; + struct kvm_device_attr attr; switch (ioctl) { case KVM_CREATE_IRQCHIP: { @@ -1455,9 +1877,9 @@ long kvm_arch_vm_ioctl(struct file *filp, return kvm_vm_ioctl_set_device_addr(kvm, &dev_addr); } case KVM_ARM_PREFERRED_TARGET: { - struct kvm_vcpu_init init; - - kvm_vcpu_preferred_target(&init); + struct kvm_vcpu_init init = { + .target = KVM_ARM_TARGET_GENERIC_V8, + }; if (copy_to_user(argp, &init, sizeof(init))) return -EFAULT; @@ -1471,6 +1893,32 @@ long kvm_arch_vm_ioctl(struct file *filp, return -EFAULT; return kvm_vm_ioctl_mte_copy_tags(kvm, ©_tags); } + case KVM_ARM_SET_COUNTER_OFFSET: { + struct kvm_arm_counter_offset offset; + + if (copy_from_user(&offset, argp, sizeof(offset))) + return -EFAULT; + return kvm_vm_ioctl_set_counter_offset(kvm, &offset); + } + case KVM_HAS_DEVICE_ATTR: { + if (copy_from_user(&attr, argp, sizeof(attr))) + return -EFAULT; + + return kvm_vm_has_attr(kvm, &attr); + } + case KVM_SET_DEVICE_ATTR: { + if (copy_from_user(&attr, argp, sizeof(attr))) + return -EFAULT; + + return kvm_vm_set_attr(kvm, &attr); + } + case KVM_ARM_GET_REG_WRITABLE_MASKS: { + struct reg_mask_range range; + + if (copy_from_user(&range, argp, sizeof(range))) + return -EFAULT; + return kvm_vm_ioctl_get_reg_writable_masks(kvm, &range); + } default: return -EINVAL; } @@ -1489,6 +1937,11 @@ static unsigned long nvhe_percpu_order(void) return size ? get_order(size) : 0; } +static size_t pkvm_host_sve_state_order(void) +{ + return get_order(pkvm_host_sve_state_size()); +} + /* A lookup table holding the hypervisor VA for each vector slot */ static void *hyp_spectre_vector_selector[BP_HARDEN_EL2_SLOTS]; @@ -1521,7 +1974,7 @@ static int kvm_init_vector_slots(void) return 0; } -static void cpu_prepare_hyp_mode(int cpu) +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; @@ -1537,23 +1990,19 @@ static void cpu_prepare_hyp_mode(int cpu) params->mair_el2 = read_sysreg(mair_el1); - /* - * The ID map may be configured to use an extended virtual address - * range. This is only the case if system RAM is out of range for the - * currently configured page size and VA_BITS, in which case we will - * also need the extended virtual range for the HYP ID map, or we won't - * be able to enable the EL2 MMU. - * - * However, at EL2, there is only one TTBR register, and we can't switch - * between translation tables *and* update TCR_EL2.T0SZ at the same - * time. Bottom line: we need to use the extended range with *both* our - * translation tables. - * - * So use the same T0SZ value we use for the ID map. - */ - tcr = (read_sysreg(tcr_el1) & TCR_EL2_MASK) | TCR_EL2_RES1; - tcr &= ~TCR_T0SZ_MASK; - tcr |= (idmap_t0sz & GENMASK(TCR_TxSZ_WIDTH - 1, 0)) << TCR_T0SZ_OFFSET; + tcr = read_sysreg(tcr_el1); + 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 | FIELD_PREP(TCR_EL2_PS_MASK, ips); + if (lpa2_is_enabled()) + tcr |= TCR_EL2_DS; + } + tcr |= TCR_T0SZ(hyp_va_bits); params->tcr_el2 = tcr; params->pgd_pa = kvm_mmu_get_httbr(); @@ -1561,6 +2010,8 @@ static void cpu_prepare_hyp_mode(int cpu) params->hcr_el2 = HCR_HOST_NVHE_PROTECTED_FLAGS; else params->hcr_el2 = HCR_HOST_NVHE_FLAGS; + if (cpus_have_final_cap(ARM64_KVM_HVHE)) + params->hcr_el2 |= HCR_E2H; params->vttbr = params->vtcr = 0; /* @@ -1582,7 +2033,7 @@ static void hyp_install_host_vector(void) * 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. + * cpus_have_*_cap() wrapper. */ BUG_ON(!system_capabilities_finalized()); params = this_cpu_ptr_nvhe_sym(kvm_init_params); @@ -1643,7 +2094,8 @@ static void cpu_set_hyp_vector(void) static void cpu_hyp_init_context(void) { - kvm_init_host_cpu_context(&this_cpu_ptr_hyp_sym(kvm_host_data)->host_ctxt); + 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(); @@ -1652,7 +2104,6 @@ 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(); @@ -1668,32 +2119,49 @@ static void cpu_hyp_reinit(void) cpu_hyp_init_features(); } -static void _kvm_arch_hardware_enable(void *discard) +static void cpu_hyp_init(void *discard) { - if (!__this_cpu_read(kvm_arm_hardware_enabled)) { + if (!__this_cpu_read(kvm_hyp_initialized)) { cpu_hyp_reinit(); - __this_cpu_write(kvm_arm_hardware_enabled, 1); + __this_cpu_write(kvm_hyp_initialized, 1); } } -int kvm_arch_hardware_enable(void) +static void cpu_hyp_uninit(void *discard) { - _kvm_arch_hardware_enable(NULL); - return 0; + if (__this_cpu_read(kvm_hyp_initialized)) { + cpu_hyp_reset(); + __this_cpu_write(kvm_hyp_initialized, 0); + } } -static void _kvm_arch_hardware_disable(void *discard) +int kvm_arch_enable_virtualization_cpu(void) { - if (__this_cpu_read(kvm_arm_hardware_enabled)) { - cpu_hyp_reset(); - __this_cpu_write(kvm_arm_hardware_enabled, 0); - } + /* + * Most calls to this function are made with migration + * disabled, but not with preemption disabled. The former is + * enough to ensure correctness, but most of the helpers + * expect the later and will throw a tantrum otherwise. + */ + preempt_disable(); + + cpu_hyp_init(NULL); + + kvm_vgic_cpu_up(); + kvm_timer_cpu_up(); + + preempt_enable(); + + return 0; } -void kvm_arch_hardware_disable(void) +void kvm_arch_disable_virtualization_cpu(void) { + kvm_timer_cpu_down(); + kvm_vgic_cpu_down(); + if (!is_protected_kvm_enabled()) - _kvm_arch_hardware_disable(NULL); + cpu_hyp_uninit(NULL); } #ifdef CONFIG_CPU_PM @@ -1702,16 +2170,16 @@ static int hyp_init_cpu_pm_notifier(struct notifier_block *self, void *v) { /* - * kvm_arm_hardware_enabled is left with its old value over + * kvm_hyp_initialized is left with its old value over * PM_ENTER->PM_EXIT. It is used to indicate PM_EXIT should * re-enable hyp. */ switch (cmd) { case CPU_PM_ENTER: - if (__this_cpu_read(kvm_arm_hardware_enabled)) + if (__this_cpu_read(kvm_hyp_initialized)) /* - * don't update kvm_arm_hardware_enabled here - * so that the hardware will be re-enabled + * don't update kvm_hyp_initialized here + * so that the hyp will be re-enabled * when we resume. See below. */ cpu_hyp_reset(); @@ -1719,8 +2187,8 @@ static int hyp_init_cpu_pm_notifier(struct notifier_block *self, return NOTIFY_OK; case CPU_PM_ENTER_FAILED: case CPU_PM_EXIT: - if (__this_cpu_read(kvm_arm_hardware_enabled)) - /* The hardware was enabled before suspend. */ + if (__this_cpu_read(kvm_hyp_initialized)) + /* The hyp was enabled before suspend. */ cpu_hyp_reinit(); return NOTIFY_OK; @@ -1734,26 +2202,26 @@ static struct notifier_block hyp_init_cpu_pm_nb = { .notifier_call = hyp_init_cpu_pm_notifier, }; -static void hyp_cpu_pm_init(void) +static void __init hyp_cpu_pm_init(void) { if (!is_protected_kvm_enabled()) cpu_pm_register_notifier(&hyp_init_cpu_pm_nb); } -static void hyp_cpu_pm_exit(void) +static void __init hyp_cpu_pm_exit(void) { if (!is_protected_kvm_enabled()) cpu_pm_unregister_notifier(&hyp_init_cpu_pm_nb); } #else -static inline void hyp_cpu_pm_init(void) +static inline void __init hyp_cpu_pm_init(void) { } -static inline void hyp_cpu_pm_exit(void) +static inline void __init hyp_cpu_pm_exit(void) { } #endif -static void init_cpu_logical_map(void) +static void __init init_cpu_logical_map(void) { unsigned int cpu; @@ -1770,7 +2238,7 @@ static void init_cpu_logical_map(void) #define init_psci_0_1_impl_state(config, what) \ config.psci_0_1_ ## what ## _implemented = psci_ops.what -static bool init_psci_relay(void) +static bool __init init_psci_relay(void) { /* * If PSCI has not been initialized, protected KVM cannot install @@ -1782,6 +2250,7 @@ static bool init_psci_relay(void) } kvm_host_psci_config.version = psci_ops.get_version(); + kvm_host_psci_config.smccc_version = arm_smccc_get_version(); if (kvm_host_psci_config.version == PSCI_VERSION(0, 1)) { kvm_host_psci_config.function_ids_0_1 = get_psci_0_1_function_ids(); @@ -1793,14 +2262,14 @@ static bool init_psci_relay(void) return true; } -static int init_subsystems(void) +static int __init init_subsystems(void) { int err = 0; /* * Enable hardware so that subsystem initialisation can access EL2. */ - on_each_cpu(_kvm_arch_hardware_enable, NULL, 1); + on_each_cpu(cpu_hyp_init, NULL, 1); /* * Register CPU lower-power notifier @@ -1817,6 +2286,19 @@ static int 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; @@ -1824,6 +2306,13 @@ static int init_subsystems(void) goto out; } + if (kvm_mode == KVM_MODE_NV && + !(vgic_present && kvm_vgic_global_state.type == VGIC_V3)) { + kvm_err("NV support requires GICv3, giving up\n"); + err = -EINVAL; + goto out; + } + /* * Init HYP architected timer support */ @@ -1834,26 +2323,43 @@ static int init_subsystems(void) kvm_register_perf_callbacks(NULL); out: + if (err) + hyp_cpu_pm_exit(); + if (err || !is_protected_kvm_enabled()) - on_each_cpu(_kvm_arch_hardware_disable, NULL, 1); + on_each_cpu(cpu_hyp_uninit, NULL, 1); return err; } -static void teardown_hyp_mode(void) +static void __init teardown_subsystems(void) { + kvm_unregister_perf_callbacks(); + hyp_cpu_pm_exit(); +} + +static void __init teardown_hyp_mode(void) +{ + bool free_sve = system_supports_sve() && is_protected_kvm_enabled(); int cpu; free_hyp_pgds(); for_each_possible_cpu(cpu) { - free_page(per_cpu(kvm_arm_hyp_stack_page, cpu)); - free_pages(kvm_arm_hyp_percpu_base[cpu], nvhe_percpu_order()); + 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) { + struct cpu_sve_state *sve_state; + + sve_state = per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state; + free_pages((unsigned long) sve_state, pkvm_host_sve_state_order()); + } } } -static int do_pkvm_init(u32 hyp_va_bits) +static int __init do_pkvm_init(u32 hyp_va_bits) { - void *per_cpu_base = kvm_ksym_ref(kvm_arm_hyp_percpu_base); + void *per_cpu_base = kvm_ksym_ref(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)); int ret; preempt_disable(); @@ -1865,20 +2371,41 @@ static int do_pkvm_init(u32 hyp_va_bits) /* * The stub hypercalls are now disabled, so set our local flag to - * prevent a later re-init attempt in kvm_arch_hardware_enable(). + * prevent a later re-init attempt in kvm_arch_enable_virtualization_cpu(). */ - __this_cpu_write(kvm_arm_hardware_enabled, 1); + __this_cpu_write(kvm_hyp_initialized, 1); preempt_enable(); return ret; } -static int kvm_hyp_init_protection(u32 hyp_va_bits) +static u64 get_hyp_id_aa64pfr0_el1(void) { - void *addr = phys_to_virt(hyp_mem_base); - int ret; + /* + * Track whether the system isn't affected by spectre/meltdown in the + * hypervisor's view of id_aa64pfr0_el1, used for protected VMs. + * Although this is per-CPU, we make it global for simplicity, e.g., not + * to have to worry about vcpu migration. + * + * Unlike for non-protected VMs, userspace cannot override this for + * protected VMs. + */ + u64 val = read_sanitised_ftr_reg(SYS_ID_AA64PFR0_EL1); + + val &= ~(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2) | + ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3)); + + val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2), + arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED); + val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3), + arm64_get_meltdown_state() == SPECTRE_UNAFFECTED); + + return val; +} - kvm_nvhe_sym(id_aa64pfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64PFR0_EL1); +static void kvm_hyp_init_symbols(void) +{ + kvm_nvhe_sym(id_aa64pfr0_el1_sys_val) = get_hyp_id_aa64pfr0_el1(); kvm_nvhe_sym(id_aa64pfr1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64PFR1_EL1); kvm_nvhe_sym(id_aa64isar0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64ISAR0_EL1); kvm_nvhe_sym(id_aa64isar1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64ISAR1_EL1); @@ -1886,6 +2413,35 @@ static int kvm_hyp_init_protection(u32 hyp_va_bits) kvm_nvhe_sym(id_aa64mmfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1); kvm_nvhe_sym(id_aa64mmfr1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1); kvm_nvhe_sym(id_aa64mmfr2_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR2_EL1); + 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; + + /* Propagate the FGT state to the the nVHE side */ + kvm_nvhe_sym(hfgrtr_masks) = hfgrtr_masks; + kvm_nvhe_sym(hfgwtr_masks) = hfgwtr_masks; + kvm_nvhe_sym(hfgitr_masks) = hfgitr_masks; + kvm_nvhe_sym(hdfgrtr_masks) = hdfgrtr_masks; + kvm_nvhe_sym(hdfgwtr_masks) = hdfgwtr_masks; + kvm_nvhe_sym(hafgrtr_masks) = hafgrtr_masks; + kvm_nvhe_sym(hfgrtr2_masks) = hfgrtr2_masks; + kvm_nvhe_sym(hfgwtr2_masks) = hfgwtr2_masks; + kvm_nvhe_sym(hfgitr2_masks) = hfgitr2_masks; + kvm_nvhe_sym(hdfgrtr2_masks)= hdfgrtr2_masks; + kvm_nvhe_sym(hdfgwtr2_masks)= hdfgwtr2_masks; + + /* + * 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) +{ + void *addr = phys_to_virt(hyp_mem_base); + int ret; ret = create_hyp_mappings(addr, addr + hyp_mem_size, PAGE_HYP); if (ret) @@ -1900,10 +2456,72 @@ static int kvm_hyp_init_protection(u32 hyp_va_bits) return 0; } -/** - * Inits Hyp-mode on all online CPUs +static int init_pkvm_host_sve_state(void) +{ + int cpu; + + if (!system_supports_sve()) + return 0; + + /* Allocate pages for host sve state in protected mode. */ + for_each_possible_cpu(cpu) { + struct page *page = alloc_pages(GFP_KERNEL, pkvm_host_sve_state_order()); + + if (!page) + return -ENOMEM; + + per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state = page_address(page); + } + + /* + * Don't map the pages in hyp since these are only used in protected + * mode, which will (re)create its own mapping when initialized. + */ + + return 0; +} + +/* + * Finalizes the initialization of hyp mode, once everything else is initialized + * and the initialziation process cannot fail. */ -static int init_hyp_mode(void) +static void finalize_init_hyp_mode(void) +{ + int cpu; + + if (system_supports_sve() && is_protected_kvm_enabled()) { + for_each_possible_cpu(cpu) { + struct cpu_sve_state *sve_state; + + sve_state = per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state; + per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state = + kern_hyp_va(sve_state); + } + } +} + +static void pkvm_hyp_init_ptrauth(void) +{ + struct kvm_cpu_context *hyp_ctxt; + int cpu; + + for_each_possible_cpu(cpu) { + hyp_ctxt = per_cpu_ptr_nvhe_sym(kvm_hyp_ctxt, cpu); + hyp_ctxt->sys_regs[APIAKEYLO_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APIAKEYHI_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APIBKEYLO_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APIBKEYHI_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APDAKEYLO_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APDAKEYHI_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APDBKEYLO_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APDBKEYHI_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APGAKEYLO_EL1] = get_random_long(); + hyp_ctxt->sys_regs[APGAKEYHI_EL1] = get_random_long(); + } +} + +/* Inits Hyp-mode on all online CPUs */ +static int __init init_hyp_mode(void) { u32 hyp_va_bits; int cpu; @@ -1927,15 +2545,15 @@ static int 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; } /* @@ -1953,7 +2571,7 @@ static int init_hyp_mode(void) page_addr = page_address(page); memcpy(page_addr, CHOOSE_NVHE_SYM(__per_cpu_start), nvhe_percpu_size()); - kvm_arm_hyp_percpu_base[cpu] = (unsigned long)page_addr; + kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu] = (unsigned long)page_addr; } /* @@ -1966,6 +2584,13 @@ static int init_hyp_mode(void) goto out_err; } + err = create_hyp_mappings(kvm_ksym_ref(__hyp_data_start), + kvm_ksym_ref(__hyp_data_end), PAGE_HYP); + if (err) { + kvm_err("Cannot map .hyp.data section\n"); + goto out_err; + } + err = create_hyp_mappings(kvm_ksym_ref(__hyp_rodata_start), kvm_ksym_ref(__hyp_rodata_end), PAGE_HYP_RO); if (err) { @@ -2004,31 +2629,9 @@ static int 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); - unsigned long hyp_addr; + char *stack_base = (char *)per_cpu(kvm_arm_hyp_stack_base, cpu); - /* - * Allocate a contiguous HYP private VA range for the stack - * and guard page. The allocation is also aligned based on - * the order of its size. - */ - err = hyp_alloc_private_va_range(PAGE_SIZE * 2, &hyp_addr); - if (err) { - kvm_err("Cannot allocate hyp stack guard page\n"); - goto out_err; - } - - /* - * Since the stack grows downwards, map the stack to the page - * at the higher address and leave the lower guard page - * unbacked. - * - * Any valid stack address now has the PAGE_SHIFT bit as 1 - * and addresses corresponding to the guard page have the - * PAGE_SHIFT bit as 0 - this is used for overflow detection. - */ - err = __create_hyp_mappings(hyp_addr + PAGE_SIZE, PAGE_SIZE, - __pa(stack_page), PAGE_HYP); + err = create_hyp_stack(__pa(stack_base), ¶ms->stack_hyp_va); if (err) { kvm_err("Cannot map hyp stack\n"); goto out_err; @@ -2040,13 +2643,11 @@ static int 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_hyp_va = hyp_addr + (2 * PAGE_SIZE); + params->stack_pa = __pa(stack_base); } for_each_possible_cpu(cpu) { - char *percpu_begin = (char *)kvm_arm_hyp_percpu_base[cpu]; + char *percpu_begin = (char *)kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu]; char *percpu_end = percpu_begin + nvhe_percpu_size(); /* Map Hyp percpu pages */ @@ -2057,19 +2658,27 @@ static int init_hyp_mode(void) } /* Prepare the CPU initialization parameters */ - cpu_prepare_hyp_mode(cpu); + cpu_prepare_hyp_mode(cpu, hyp_va_bits); } + kvm_hyp_init_symbols(); + if (is_protected_kvm_enabled()) { + if (IS_ENABLED(CONFIG_ARM64_PTR_AUTH_KERNEL) && + cpus_have_final_cap(ARM64_HAS_ADDRESS_AUTH)) + pkvm_hyp_init_ptrauth(); + init_cpu_logical_map(); if (!init_psci_relay()) { err = -ENODEV; goto out_err; } - } - if (is_protected_kvm_enabled()) { + err = init_pkvm_host_sve_state(); + if (err) + goto out_err; + err = kvm_hyp_init_protection(hyp_va_bits); if (err) { kvm_err("Failed to init hyp memory protection\n"); @@ -2085,47 +2694,30 @@ out_err: return err; } -static void _kvm_host_prot_finalize(void *arg) +struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr) { - int *err = arg; + struct kvm_vcpu *vcpu = NULL; + struct kvm_mpidr_data *data; + unsigned long i; - if (WARN_ON(kvm_call_hyp_nvhe(__pkvm_prot_finalize))) - WRITE_ONCE(*err, -EINVAL); -} + mpidr &= MPIDR_HWID_BITMASK; -static int pkvm_drop_host_privileges(void) -{ - int ret = 0; + rcu_read_lock(); + data = rcu_dereference(kvm->arch.mpidr_data); - /* - * Flip the static key upfront as that may no longer be possible - * once the host stage 2 is installed. - */ - static_branch_enable(&kvm_protected_mode_initialized); - on_each_cpu(_kvm_host_prot_finalize, &ret, 1); - return ret; -} + if (data) { + u16 idx = kvm_mpidr_index(data, mpidr); -static int finalize_hyp_mode(void) -{ - if (!is_protected_kvm_enabled()) - return 0; + vcpu = kvm_get_vcpu(kvm, data->cmpidr_to_idx[idx]); + if (mpidr != kvm_vcpu_get_mpidr_aff(vcpu)) + vcpu = NULL; + } - /* - * Exclude HYP sections from kmemleak so that they don't get peeked - * at, which would end badly once inaccessible. - */ - kmemleak_free_part(__hyp_bss_start, __hyp_bss_end - __hyp_bss_start); - kmemleak_free_part(__va(hyp_mem_base), hyp_mem_size); - return pkvm_drop_host_privileges(); -} + rcu_read_unlock(); -struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr) -{ - struct kvm_vcpu *vcpu; - unsigned long i; + if (vcpu) + return vcpu; - mpidr &= MPIDR_HWID_BITMASK; kvm_for_each_vcpu(i, vcpu, kvm) { if (mpidr == kvm_vcpu_get_mpidr_aff(vcpu)) return vcpu; @@ -2133,9 +2725,9 @@ struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr) return NULL; } -bool kvm_arch_has_irq_bypass(void) +bool kvm_arch_irqchip_in_kernel(struct kvm *kvm) { - return true; + return irqchip_in_kernel(kvm); } int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons, @@ -2143,18 +2735,53 @@ int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons, { struct kvm_kernel_irqfd *irqfd = container_of(cons, struct kvm_kernel_irqfd, consumer); + struct kvm_kernel_irq_routing_entry *irq_entry = &irqfd->irq_entry; + + /* + * The only thing we have a chance of directly-injecting is LPIs. Maybe + * one day... + */ + if (irq_entry->type != KVM_IRQ_ROUTING_MSI) + return 0; return kvm_vgic_v4_set_forwarding(irqfd->kvm, prod->irq, &irqfd->irq_entry); } + void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons, struct irq_bypass_producer *prod) { struct kvm_kernel_irqfd *irqfd = container_of(cons, struct kvm_kernel_irqfd, consumer); + struct kvm_kernel_irq_routing_entry *irq_entry = &irqfd->irq_entry; - kvm_vgic_v4_unset_forwarding(irqfd->kvm, prod->irq, - &irqfd->irq_entry); + if (irq_entry->type != KVM_IRQ_ROUTING_MSI) + return; + + kvm_vgic_v4_unset_forwarding(irqfd->kvm, prod->irq); +} + +bool kvm_arch_irqfd_route_changed(struct kvm_kernel_irq_routing_entry *old, + struct kvm_kernel_irq_routing_entry *new) +{ + if (old->type != KVM_IRQ_ROUTING_MSI || + new->type != KVM_IRQ_ROUTING_MSI) + return true; + + return memcmp(&old->msi, &new->msi, sizeof(new->msi)); +} + +int kvm_arch_update_irqfd_routing(struct kvm *kvm, unsigned int host_irq, + uint32_t guest_irq, bool set) +{ + /* + * Remapping the vLPI requires taking the its_lock mutex to resolve + * the new translation. We're in spinlock land at this point, so no + * chance of resolving the translation. + * + * Unmap the vLPI and fall back to software LPI injection. + */ + return kvm_vgic_v4_unset_forwarding(kvm, host_irq); } void kvm_arch_irq_bypass_stop(struct irq_bypass_consumer *cons) @@ -2173,10 +2800,8 @@ void kvm_arch_irq_bypass_start(struct irq_bypass_consumer *cons) kvm_arm_resume_guest(irqfd->kvm); } -/** - * Initialize Hyp-mode and memory mappings on all CPUs. - */ -int kvm_arch_init(void *opaque) +/* Initialize Hyp-mode and memory mappings on all CPUs */ +static __init int kvm_arm_init(void) { int err; bool in_hyp_mode; @@ -2227,33 +2852,42 @@ int kvm_arch_init(void *opaque) err = kvm_init_vector_slots(); if (err) { kvm_err("Cannot initialise vector slots\n"); - goto out_err; + goto out_hyp; } err = init_subsystems(); if (err) goto out_hyp; - if (!in_hyp_mode) { - err = finalize_hyp_mode(); - if (err) { - kvm_err("Failed to finalize Hyp protection\n"); - goto out_hyp; - } - } + kvm_info("%s%sVHE%s mode initialized successfully\n", + in_hyp_mode ? "" : (is_protected_kvm_enabled() ? + "Protected " : "Hyp "), + in_hyp_mode ? "" : (cpus_have_final_cap(ARM64_KVM_HVHE) ? + "h" : "n"), + cpus_have_final_cap(ARM64_HAS_NESTED_VIRT) ? "+NV2": ""); - if (is_protected_kvm_enabled()) { - kvm_info("Protected nVHE mode initialized successfully\n"); - } else if (in_hyp_mode) { - kvm_info("VHE mode initialized successfully\n"); - } else { - kvm_info("Hyp mode initialized successfully\n"); - } + /* + * FIXME: Do something reasonable if kvm_init() fails after pKVM + * hypervisor protection is finalized. + */ + err = kvm_init(sizeof(struct kvm_vcpu), 0, THIS_MODULE); + if (err) + goto out_subs; + + /* + * This should be called after initialization is done and failure isn't + * possible anymore. + */ + if (!in_hyp_mode) + finalize_init_hyp_mode(); + + kvm_arm_initialised = true; return 0; +out_subs: + teardown_subsystems(); out_hyp: - hyp_cpu_pm_exit(); if (!in_hyp_mode) teardown_hyp_mode(); out_err: @@ -2261,17 +2895,21 @@ out_err: return err; } -/* NOP: Compiling as a module not supported */ -void kvm_arch_exit(void) -{ - kvm_unregister_perf_callbacks(); -} - static int __init early_kvm_mode_cfg(char *arg) { if (!arg) return -EINVAL; + if (strcmp(arg, "none") == 0) { + kvm_mode = KVM_MODE_NONE; + return 0; + } + + if (!is_hyp_mode_available()) { + pr_warn_once("KVM is not available. Ignoring kvm-arm.mode\n"); + return 0; + } + if (strcmp(arg, "protected") == 0) { if (!is_kernel_in_hyp_mode()) kvm_mode = KVM_MODE_PROTECTED; @@ -2286,8 +2924,8 @@ static int __init early_kvm_mode_cfg(char *arg) return 0; } - if (strcmp(arg, "none") == 0) { - kvm_mode = KVM_MODE_NONE; + if (strcmp(arg, "nested") == 0 && !WARN_ON(!is_kernel_in_hyp_mode())) { + kvm_mode = KVM_MODE_NV; return 0; } @@ -2295,15 +2933,39 @@ static int __init early_kvm_mode_cfg(char *arg) } early_param("kvm-arm.mode", early_kvm_mode_cfg); -enum kvm_mode kvm_get_mode(void) +static int __init early_kvm_wfx_trap_policy_cfg(char *arg, enum kvm_wfx_trap_policy *p) { - return kvm_mode; + if (!arg) + return -EINVAL; + + if (strcmp(arg, "trap") == 0) { + *p = KVM_WFX_TRAP; + return 0; + } + + if (strcmp(arg, "notrap") == 0) { + *p = KVM_WFX_NOTRAP; + return 0; + } + + return -EINVAL; } -static int arm_init(void) +static int __init early_kvm_wfi_trap_policy_cfg(char *arg) { - int rc = kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE); - return rc; + return early_kvm_wfx_trap_policy_cfg(arg, &kvm_wfi_trap_policy); +} +early_param("kvm-arm.wfi_trap_policy", early_kvm_wfi_trap_policy_cfg); + +static int __init early_kvm_wfe_trap_policy_cfg(char *arg) +{ + return early_kvm_wfx_trap_policy_cfg(arg, &kvm_wfe_trap_policy); +} +early_param("kvm-arm.wfe_trap_policy", early_kvm_wfe_trap_policy_cfg); + +enum kvm_mode kvm_get_mode(void) +{ + return kvm_mode; } -module_init(arm_init); +module_init(kvm_arm_init); |