diff options
Diffstat (limited to 'arch/x86/kvm/svm')
| -rw-r--r-- | arch/x86/kvm/svm/avic.c | 1427 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/hyperv.c | 18 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/hyperv.h | 54 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/nested.c | 1261 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/pmu.c | 280 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/sev.c | 3536 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm.c | 3556 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm.h | 758 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm_onhyperv.c | 46 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm_onhyperv.h | 101 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm_ops.h | 15 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/vmenter.S | 361 |
12 files changed, 8055 insertions, 3358 deletions
diff --git a/arch/x86/kvm/svm/avic.c b/arch/x86/kvm/svm/avic.c index 1d01da64c333..6b77b2033208 100644 --- a/arch/x86/kvm/svm/avic.c +++ b/arch/x86/kvm/svm/avic.c @@ -12,14 +12,16 @@ * Avi Kivity <avi@qumranet.com> */ -#define pr_fmt(fmt) "SVM: " fmt +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kvm_types.h> #include <linux/hashtable.h> #include <linux/amd-iommu.h> #include <linux/kvm_host.h> +#include <linux/kvm_irqfd.h> #include <asm/irq_remapping.h> +#include <asm/msr.h> #include "trace.h" #include "lapic.h" @@ -27,32 +29,71 @@ #include "irq.h" #include "svm.h" -#define SVM_AVIC_DOORBELL 0xc001011b - -#define AVIC_HPA_MASK ~((0xFFFULL << 52) | 0xFFF) - /* - * 0xff is broadcast, so the max index allowed for physical APIC ID - * table is 0xfe. APIC IDs above 0xff are reserved. + * Encode the arbitrary VM ID and the vCPU's _index_ into the GATag so that + * KVM can retrieve the correct vCPU from a GALog entry if an interrupt can't + * be delivered, e.g. because the vCPU isn't running. Use the vCPU's index + * instead of its ID (a.k.a. its default APIC ID), as KVM is guaranteed a fast + * lookup on the index, where as vCPUs whose index doesn't match their ID need + * to walk the entire xarray of vCPUs in the worst case scenario. + * + * For the vCPU index, use however many bits are currently allowed for the max + * guest physical APIC ID (limited by the size of the physical ID table), and + * use whatever bits remain to assign arbitrary AVIC IDs to VMs. Note, the + * size of the GATag is defined by hardware (32 bits), but is an opaque value + * as far as hardware is concerned. */ -#define AVIC_MAX_PHYSICAL_ID_COUNT 255 +#define AVIC_VCPU_IDX_MASK AVIC_PHYSICAL_MAX_INDEX_MASK + +#define AVIC_VM_ID_SHIFT HWEIGHT32(AVIC_PHYSICAL_MAX_INDEX_MASK) +#define AVIC_VM_ID_MASK (GENMASK(31, AVIC_VM_ID_SHIFT) >> AVIC_VM_ID_SHIFT) + +#define AVIC_GATAG_TO_VMID(x) ((x >> AVIC_VM_ID_SHIFT) & AVIC_VM_ID_MASK) +#define AVIC_GATAG_TO_VCPUIDX(x) (x & AVIC_VCPU_IDX_MASK) + +#define __AVIC_GATAG(vm_id, vcpu_idx) ((((vm_id) & AVIC_VM_ID_MASK) << AVIC_VM_ID_SHIFT) | \ + ((vcpu_idx) & AVIC_VCPU_IDX_MASK)) +#define AVIC_GATAG(vm_id, vcpu_idx) \ +({ \ + u32 ga_tag = __AVIC_GATAG(vm_id, vcpu_idx); \ + \ + WARN_ON_ONCE(AVIC_GATAG_TO_VCPUIDX(ga_tag) != (vcpu_idx)); \ + WARN_ON_ONCE(AVIC_GATAG_TO_VMID(ga_tag) != (vm_id)); \ + ga_tag; \ +}) + +static_assert(__AVIC_GATAG(AVIC_VM_ID_MASK, AVIC_VCPU_IDX_MASK) == -1u); + +#define AVIC_AUTO_MODE -1 + +static int avic_param_set(const char *val, const struct kernel_param *kp) +{ + if (val && sysfs_streq(val, "auto")) { + *(int *)kp->arg = AVIC_AUTO_MODE; + return 0; + } + + return param_set_bint(val, kp); +} -#define AVIC_UNACCEL_ACCESS_WRITE_MASK 1 -#define AVIC_UNACCEL_ACCESS_OFFSET_MASK 0xFF0 -#define AVIC_UNACCEL_ACCESS_VECTOR_MASK 0xFFFFFFFF +static const struct kernel_param_ops avic_ops = { + .flags = KERNEL_PARAM_OPS_FL_NOARG, + .set = avic_param_set, + .get = param_get_bool, +}; -/* AVIC GATAG is encoded using VM and VCPU IDs */ -#define AVIC_VCPU_ID_BITS 8 -#define AVIC_VCPU_ID_MASK ((1 << AVIC_VCPU_ID_BITS) - 1) +/* + * Enable / disable AVIC. In "auto" mode (default behavior), AVIC is enabled + * for Zen4+ CPUs with x2AVIC (and all other criteria for enablement are met). + */ +static int avic = AVIC_AUTO_MODE; +module_param_cb(avic, &avic_ops, &avic, 0444); +__MODULE_PARM_TYPE(avic, "bool"); -#define AVIC_VM_ID_BITS 24 -#define AVIC_VM_ID_NR (1 << AVIC_VM_ID_BITS) -#define AVIC_VM_ID_MASK ((1 << AVIC_VM_ID_BITS) - 1) +module_param(enable_ipiv, bool, 0444); -#define AVIC_GATAG(x, y) (((x & AVIC_VM_ID_MASK) << AVIC_VCPU_ID_BITS) | \ - (y & AVIC_VCPU_ID_MASK)) -#define AVIC_GATAG_TO_VMID(x) ((x >> AVIC_VCPU_ID_BITS) & AVIC_VM_ID_MASK) -#define AVIC_GATAG_TO_VCPUID(x) (x & AVIC_VCPU_ID_MASK) +static bool force_avic; +module_param_unsafe(force_avic, bool, 0444); /* Note: * This hash table is used to map VM_ID to a struct kvm_svm, @@ -64,42 +105,159 @@ static DEFINE_HASHTABLE(svm_vm_data_hash, SVM_VM_DATA_HASH_BITS); static u32 next_vm_id = 0; static bool next_vm_id_wrapped = 0; static DEFINE_SPINLOCK(svm_vm_data_hash_lock); +static bool x2avic_enabled; +static u32 x2avic_max_physical_id; -/* - * This is a wrapper of struct amd_iommu_ir_data. - */ -struct amd_svm_iommu_ir { - struct list_head node; /* Used by SVM for per-vcpu ir_list */ - void *data; /* Storing pointer to struct amd_ir_data */ -}; +static void avic_set_x2apic_msr_interception(struct vcpu_svm *svm, + bool intercept) +{ + static const u32 x2avic_passthrough_msrs[] = { + X2APIC_MSR(APIC_ID), + X2APIC_MSR(APIC_LVR), + X2APIC_MSR(APIC_TASKPRI), + X2APIC_MSR(APIC_ARBPRI), + X2APIC_MSR(APIC_PROCPRI), + X2APIC_MSR(APIC_EOI), + X2APIC_MSR(APIC_RRR), + X2APIC_MSR(APIC_LDR), + X2APIC_MSR(APIC_DFR), + X2APIC_MSR(APIC_SPIV), + X2APIC_MSR(APIC_ISR), + X2APIC_MSR(APIC_TMR), + X2APIC_MSR(APIC_IRR), + X2APIC_MSR(APIC_ESR), + X2APIC_MSR(APIC_ICR), + X2APIC_MSR(APIC_ICR2), -enum avic_ipi_failure_cause { - AVIC_IPI_FAILURE_INVALID_INT_TYPE, - AVIC_IPI_FAILURE_TARGET_NOT_RUNNING, - AVIC_IPI_FAILURE_INVALID_TARGET, - AVIC_IPI_FAILURE_INVALID_BACKING_PAGE, -}; + /* + * Note! Always intercept LVTT, as TSC-deadline timer mode + * isn't virtualized by hardware, and the CPU will generate a + * #GP instead of a #VMEXIT. + */ + X2APIC_MSR(APIC_LVTTHMR), + X2APIC_MSR(APIC_LVTPC), + X2APIC_MSR(APIC_LVT0), + X2APIC_MSR(APIC_LVT1), + X2APIC_MSR(APIC_LVTERR), + X2APIC_MSR(APIC_TMICT), + X2APIC_MSR(APIC_TMCCT), + X2APIC_MSR(APIC_TDCR), + }; + int i; + + if (intercept == svm->x2avic_msrs_intercepted) + return; + + if (!x2avic_enabled) + return; + + for (i = 0; i < ARRAY_SIZE(x2avic_passthrough_msrs); i++) + svm_set_intercept_for_msr(&svm->vcpu, x2avic_passthrough_msrs[i], + MSR_TYPE_RW, intercept); + + svm->x2avic_msrs_intercepted = intercept; +} + +static u32 __avic_get_max_physical_id(struct kvm *kvm, struct kvm_vcpu *vcpu) +{ + u32 arch_max; + + /* + * Return the largest size (x2APIC) when querying without a vCPU, e.g. + * to allocate the per-VM table.. + */ + if (x2avic_enabled && (!vcpu || apic_x2apic_mode(vcpu->arch.apic))) + arch_max = x2avic_max_physical_id; + else + arch_max = AVIC_MAX_PHYSICAL_ID; + + /* + * Despite its name, KVM_CAP_MAX_VCPU_ID represents the maximum APIC ID + * plus one, so the max possible APIC ID is one less than that. + */ + return min(kvm->arch.max_vcpu_ids - 1, arch_max); +} + +static u32 avic_get_max_physical_id(struct kvm_vcpu *vcpu) +{ + return __avic_get_max_physical_id(vcpu->kvm, vcpu); +} + +static void avic_activate_vmcb(struct vcpu_svm *svm) +{ + struct vmcb *vmcb = svm->vmcb01.ptr; + struct kvm_vcpu *vcpu = &svm->vcpu; + + vmcb->control.int_ctl &= ~(AVIC_ENABLE_MASK | X2APIC_MODE_MASK); + + vmcb->control.avic_physical_id &= ~AVIC_PHYSICAL_MAX_INDEX_MASK; + vmcb->control.avic_physical_id |= avic_get_max_physical_id(vcpu); + + vmcb->control.int_ctl |= AVIC_ENABLE_MASK; + + /* + * Note: KVM supports hybrid-AVIC mode, where KVM emulates x2APIC MSR + * accesses, while interrupt injection to a running vCPU can be + * achieved using AVIC doorbell. KVM disables the APIC access page + * (deletes the memslot) if any vCPU has x2APIC enabled, thus enabling + * AVIC in hybrid mode activates only the doorbell mechanism. + */ + if (x2avic_enabled && apic_x2apic_mode(svm->vcpu.arch.apic)) { + vmcb->control.int_ctl |= X2APIC_MODE_MASK; + + /* Disabling MSR intercept for x2APIC registers */ + avic_set_x2apic_msr_interception(svm, false); + } else { + /* + * Flush the TLB, the guest may have inserted a non-APIC + * mapping into the TLB while AVIC was disabled. + */ + kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, &svm->vcpu); + + /* Enabling MSR intercept for x2APIC registers */ + avic_set_x2apic_msr_interception(svm, true); + } +} + +static void avic_deactivate_vmcb(struct vcpu_svm *svm) +{ + struct vmcb *vmcb = svm->vmcb01.ptr; + + vmcb->control.int_ctl &= ~(AVIC_ENABLE_MASK | X2APIC_MODE_MASK); + vmcb->control.avic_physical_id &= ~AVIC_PHYSICAL_MAX_INDEX_MASK; + + /* + * If running nested and the guest uses its own MSR bitmap, there + * is no need to update L0's msr bitmap + */ + if (is_guest_mode(&svm->vcpu) && + vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_MSR_PROT)) + return; + + /* Enabling MSR intercept for x2APIC registers */ + avic_set_x2apic_msr_interception(svm, true); +} /* Note: * This function is called from IOMMU driver to notify * SVM to schedule in a particular vCPU of a particular VM. */ -int avic_ga_log_notifier(u32 ga_tag) +static int avic_ga_log_notifier(u32 ga_tag) { unsigned long flags; struct kvm_svm *kvm_svm; struct kvm_vcpu *vcpu = NULL; u32 vm_id = AVIC_GATAG_TO_VMID(ga_tag); - u32 vcpu_id = AVIC_GATAG_TO_VCPUID(ga_tag); + u32 vcpu_idx = AVIC_GATAG_TO_VCPUIDX(ga_tag); - pr_debug("SVM: %s: vm_id=%#x, vcpu_id=%#x\n", __func__, vm_id, vcpu_id); - trace_kvm_avic_ga_log(vm_id, vcpu_id); + pr_debug("SVM: %s: vm_id=%#x, vcpu_idx=%#x\n", __func__, vm_id, vcpu_idx); + trace_kvm_avic_ga_log(vm_id, vcpu_idx); spin_lock_irqsave(&svm_vm_data_hash_lock, flags); hash_for_each_possible(svm_vm_data_hash, kvm_svm, hnode, vm_id) { if (kvm_svm->avic_vm_id != vm_id) continue; - vcpu = kvm_get_vcpu_by_id(&kvm_svm->kvm, vcpu_id); + vcpu = kvm_get_vcpu(&kvm_svm->kvm, vcpu_idx); break; } spin_unlock_irqrestore(&svm_vm_data_hash_lock, flags); @@ -115,6 +273,30 @@ int avic_ga_log_notifier(u32 ga_tag) return 0; } +static int avic_get_physical_id_table_order(struct kvm *kvm) +{ + /* Provision for the maximum physical ID supported in x2avic mode */ + return get_order((__avic_get_max_physical_id(kvm, NULL) + 1) * sizeof(u64)); +} + +int avic_alloc_physical_id_table(struct kvm *kvm) +{ + struct kvm_svm *kvm_svm = to_kvm_svm(kvm); + + if (!irqchip_in_kernel(kvm) || !enable_apicv) + return 0; + + if (kvm_svm->avic_physical_id_table) + return 0; + + kvm_svm->avic_physical_id_table = (void *)__get_free_pages(GFP_KERNEL_ACCOUNT | __GFP_ZERO, + avic_get_physical_id_table_order(kvm)); + if (!kvm_svm->avic_physical_id_table) + return -ENOMEM; + + return 0; +} + void avic_vm_destroy(struct kvm *kvm) { unsigned long flags; @@ -123,10 +305,9 @@ void avic_vm_destroy(struct kvm *kvm) if (!enable_apicv) return; - if (kvm_svm->avic_logical_id_table_page) - __free_page(kvm_svm->avic_logical_id_table_page); - if (kvm_svm->avic_physical_id_table_page) - __free_page(kvm_svm->avic_physical_id_table_page); + free_page((unsigned long)kvm_svm->avic_logical_id_table); + free_pages((unsigned long)kvm_svm->avic_physical_id_table, + avic_get_physical_id_table_order(kvm)); spin_lock_irqsave(&svm_vm_data_hash_lock, flags); hash_del(&kvm_svm->hnode); @@ -139,27 +320,15 @@ int avic_vm_init(struct kvm *kvm) int err = -ENOMEM; struct kvm_svm *kvm_svm = to_kvm_svm(kvm); struct kvm_svm *k2; - struct page *p_page; - struct page *l_page; u32 vm_id; if (!enable_apicv) return 0; - /* Allocating physical APIC ID table (4KB) */ - p_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); - if (!p_page) + kvm_svm->avic_logical_id_table = (void *)get_zeroed_page(GFP_KERNEL_ACCOUNT); + if (!kvm_svm->avic_logical_id_table) goto free_avic; - kvm_svm->avic_physical_id_table_page = p_page; - - /* Allocating logical APIC ID table (4KB) */ - l_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); - if (!l_page) - goto free_avic; - - kvm_svm->avic_logical_id_table_page = l_page; - spin_lock_irqsave(&svm_vm_data_hash_lock, flags); again: vm_id = next_vm_id = (next_vm_id + 1) & AVIC_VM_ID_MASK; @@ -185,127 +354,252 @@ free_avic: return err; } -void avic_init_vmcb(struct vcpu_svm *svm) +static phys_addr_t avic_get_backing_page_address(struct vcpu_svm *svm) +{ + return __sme_set(__pa(svm->vcpu.arch.apic->regs)); +} + +void avic_init_vmcb(struct vcpu_svm *svm, struct vmcb *vmcb) { - struct vmcb *vmcb = svm->vmcb; struct kvm_svm *kvm_svm = to_kvm_svm(svm->vcpu.kvm); - phys_addr_t bpa = __sme_set(page_to_phys(svm->avic_backing_page)); - phys_addr_t lpa = __sme_set(page_to_phys(kvm_svm->avic_logical_id_table_page)); - phys_addr_t ppa = __sme_set(page_to_phys(kvm_svm->avic_physical_id_table_page)); - - vmcb->control.avic_backing_page = bpa & AVIC_HPA_MASK; - vmcb->control.avic_logical_id = lpa & AVIC_HPA_MASK; - vmcb->control.avic_physical_id = ppa & AVIC_HPA_MASK; - vmcb->control.avic_physical_id |= AVIC_MAX_PHYSICAL_ID_COUNT; + + vmcb->control.avic_backing_page = avic_get_backing_page_address(svm); + vmcb->control.avic_logical_id = __sme_set(__pa(kvm_svm->avic_logical_id_table)); + vmcb->control.avic_physical_id = __sme_set(__pa(kvm_svm->avic_physical_id_table)); + vmcb->control.avic_vapic_bar = APIC_DEFAULT_PHYS_BASE; + if (kvm_apicv_activated(svm->vcpu.kvm)) - vmcb->control.int_ctl |= AVIC_ENABLE_MASK; + avic_activate_vmcb(svm); else - vmcb->control.int_ctl &= ~AVIC_ENABLE_MASK; + avic_deactivate_vmcb(svm); } -static u64 *avic_get_physical_id_entry(struct kvm_vcpu *vcpu, - unsigned int index) +static int avic_init_backing_page(struct kvm_vcpu *vcpu) { - u64 *avic_physical_id_table; struct kvm_svm *kvm_svm = to_kvm_svm(vcpu->kvm); + struct vcpu_svm *svm = to_svm(vcpu); + u32 id = vcpu->vcpu_id; + u64 new_entry; - if (index >= AVIC_MAX_PHYSICAL_ID_COUNT) - return NULL; + /* + * Inhibit AVIC if the vCPU ID is bigger than what is supported by AVIC + * hardware. Immediately clear apicv_active, i.e. don't wait until the + * KVM_REQ_APICV_UPDATE request is processed on the first KVM_RUN, as + * avic_vcpu_load() expects to be called if and only if the vCPU has + * fully initialized AVIC. + */ + if ((!x2avic_enabled && id > AVIC_MAX_PHYSICAL_ID) || + (id > x2avic_max_physical_id)) { + kvm_set_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_PHYSICAL_ID_TOO_BIG); + vcpu->arch.apic->apicv_active = false; + return 0; + } - avic_physical_id_table = page_address(kvm_svm->avic_physical_id_table_page); + BUILD_BUG_ON((AVIC_MAX_PHYSICAL_ID + 1) * sizeof(new_entry) > PAGE_SIZE || + (X2AVIC_MAX_PHYSICAL_ID + 1) * sizeof(new_entry) > PAGE_SIZE); - return &avic_physical_id_table[index]; + if (WARN_ON_ONCE(!vcpu->arch.apic->regs)) + return -EINVAL; + + if (kvm_apicv_activated(vcpu->kvm)) { + int ret; + + /* + * Note, AVIC hardware walks the nested page table to check + * permissions, but does not use the SPA address specified in + * the leaf SPTE since it uses address in the AVIC_BACKING_PAGE + * pointer field of the VMCB. + */ + ret = kvm_alloc_apic_access_page(vcpu->kvm); + if (ret) + return ret; + } + + /* Note, fls64() returns the bit position, +1. */ + BUILD_BUG_ON(__PHYSICAL_MASK_SHIFT > + fls64(AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK)); + + /* Setting AVIC backing page address in the phy APIC ID table */ + new_entry = avic_get_backing_page_address(svm) | + AVIC_PHYSICAL_ID_ENTRY_VALID_MASK; + svm->avic_physical_id_entry = new_entry; + + /* + * Initialize the real table, as vCPUs must have a valid entry in order + * for broadcast IPIs to function correctly (broadcast IPIs ignore + * invalid entries, i.e. aren't guaranteed to generate a VM-Exit). + */ + WRITE_ONCE(kvm_svm->avic_physical_id_table[id], new_entry); + + return 0; } -/* - * Note: - * AVIC hardware walks the nested page table to check permissions, - * but does not use the SPA address specified in the leaf page - * table entry since it uses address in the AVIC_BACKING_PAGE pointer - * field of the VMCB. Therefore, we set up the - * APIC_ACCESS_PAGE_PRIVATE_MEMSLOT (4KB) here. - */ -static int avic_update_access_page(struct kvm *kvm, bool activate) +void avic_ring_doorbell(struct kvm_vcpu *vcpu) { - void __user *ret; - int r = 0; - - mutex_lock(&kvm->slots_lock); /* - * During kvm_destroy_vm(), kvm_pit_set_reinject() could trigger - * APICv mode change, which update APIC_ACCESS_PAGE_PRIVATE_MEMSLOT - * memory region. So, we need to ensure that kvm->mm == current->mm. + * Note, the vCPU could get migrated to a different pCPU at any point, + * which could result in signalling the wrong/previous pCPU. But if + * that happens the vCPU is guaranteed to do a VMRUN (after being + * migrated) and thus will process pending interrupts, i.e. a doorbell + * is not needed (and the spurious one is harmless). */ - if ((kvm->arch.apic_access_memslot_enabled == activate) || - (kvm->mm != current->mm)) - goto out; - - ret = __x86_set_memory_region(kvm, - APIC_ACCESS_PAGE_PRIVATE_MEMSLOT, - APIC_DEFAULT_PHYS_BASE, - activate ? PAGE_SIZE : 0); - if (IS_ERR(ret)) { - r = PTR_ERR(ret); - goto out; + int cpu = READ_ONCE(vcpu->cpu); + + if (cpu != get_cpu()) { + wrmsrq(MSR_AMD64_SVM_AVIC_DOORBELL, kvm_cpu_get_apicid(cpu)); + trace_kvm_avic_doorbell(vcpu->vcpu_id, kvm_cpu_get_apicid(cpu)); } + put_cpu(); +} + - kvm->arch.apic_access_memslot_enabled = activate; -out: - mutex_unlock(&kvm->slots_lock); - return r; +static void avic_kick_vcpu(struct kvm_vcpu *vcpu, u32 icrl) +{ + vcpu->arch.apic->irr_pending = true; + svm_complete_interrupt_delivery(vcpu, + icrl & APIC_MODE_MASK, + icrl & APIC_INT_LEVELTRIG, + icrl & APIC_VECTOR_MASK); } -static int avic_init_backing_page(struct kvm_vcpu *vcpu) +static void avic_kick_vcpu_by_physical_id(struct kvm *kvm, u32 physical_id, + u32 icrl) { - u64 *entry, new_entry; - int id = vcpu->vcpu_id; - struct vcpu_svm *svm = to_svm(vcpu); + /* + * KVM inhibits AVIC if any vCPU ID diverges from the vCPUs APIC ID, + * i.e. APIC ID == vCPU ID. + */ + struct kvm_vcpu *target_vcpu = kvm_get_vcpu_by_id(kvm, physical_id); - if (id >= AVIC_MAX_PHYSICAL_ID_COUNT) - return -EINVAL; + /* Once again, nothing to do if the target vCPU doesn't exist. */ + if (unlikely(!target_vcpu)) + return; - if (!vcpu->arch.apic->regs) - return -EINVAL; + avic_kick_vcpu(target_vcpu, icrl); +} - if (kvm_apicv_activated(vcpu->kvm)) { - int ret; +static void avic_kick_vcpu_by_logical_id(struct kvm *kvm, u32 *avic_logical_id_table, + u32 logid_index, u32 icrl) +{ + u32 physical_id; - ret = avic_update_access_page(vcpu->kvm, true); - if (ret) - return ret; + if (avic_logical_id_table) { + u32 logid_entry = avic_logical_id_table[logid_index]; + + /* Nothing to do if the logical destination is invalid. */ + if (unlikely(!(logid_entry & AVIC_LOGICAL_ID_ENTRY_VALID_MASK))) + return; + + physical_id = logid_entry & + AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK; + } else { + /* + * For x2APIC, the logical APIC ID is a read-only value that is + * derived from the x2APIC ID, thus the x2APIC ID can be found + * by reversing the calculation (stored in logid_index). Note, + * bits 31:20 of the x2APIC ID aren't propagated to the logical + * ID, but KVM limits the x2APIC ID limited to KVM_MAX_VCPU_IDS. + */ + physical_id = logid_index; } - svm->avic_backing_page = virt_to_page(vcpu->arch.apic->regs); + avic_kick_vcpu_by_physical_id(kvm, physical_id, icrl); +} - /* Setting AVIC backing page address in the phy APIC ID table */ - entry = avic_get_physical_id_entry(vcpu, id); - if (!entry) +/* + * A fast-path version of avic_kick_target_vcpus(), which attempts to match + * destination APIC ID to vCPU without looping through all vCPUs. + */ +static int avic_kick_target_vcpus_fast(struct kvm *kvm, struct kvm_lapic *source, + u32 icrl, u32 icrh, u32 index) +{ + int dest_mode = icrl & APIC_DEST_MASK; + int shorthand = icrl & APIC_SHORT_MASK; + struct kvm_svm *kvm_svm = to_kvm_svm(kvm); + u32 dest; + + if (shorthand != APIC_DEST_NOSHORT) return -EINVAL; - new_entry = __sme_set((page_to_phys(svm->avic_backing_page) & - AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK) | - AVIC_PHYSICAL_ID_ENTRY_VALID_MASK); - WRITE_ONCE(*entry, new_entry); + if (apic_x2apic_mode(source)) + dest = icrh; + else + dest = GET_XAPIC_DEST_FIELD(icrh); + + if (dest_mode == APIC_DEST_PHYSICAL) { + /* broadcast destination, use slow path */ + if (apic_x2apic_mode(source) && dest == X2APIC_BROADCAST) + return -EINVAL; + if (!apic_x2apic_mode(source) && dest == APIC_BROADCAST) + return -EINVAL; + + if (WARN_ON_ONCE(dest != index)) + return -EINVAL; + + avic_kick_vcpu_by_physical_id(kvm, dest, icrl); + } else { + u32 *avic_logical_id_table; + unsigned long bitmap, i; + u32 cluster; + + if (apic_x2apic_mode(source)) { + /* 16 bit dest mask, 16 bit cluster id */ + bitmap = dest & 0xFFFF; + cluster = (dest >> 16) << 4; + } else if (kvm_lapic_get_reg(source, APIC_DFR) == APIC_DFR_FLAT) { + /* 8 bit dest mask*/ + bitmap = dest; + cluster = 0; + } else { + /* 4 bit desk mask, 4 bit cluster id */ + bitmap = dest & 0xF; + cluster = (dest >> 4) << 2; + } + + /* Nothing to do if there are no destinations in the cluster. */ + if (unlikely(!bitmap)) + return 0; + + if (apic_x2apic_mode(source)) + avic_logical_id_table = NULL; + else + avic_logical_id_table = kvm_svm->avic_logical_id_table; - svm->avic_physical_id_cache = entry; + /* + * AVIC is inhibited if vCPUs aren't mapped 1:1 with logical + * IDs, thus each bit in the destination is guaranteed to map + * to at most one vCPU. + */ + for_each_set_bit(i, &bitmap, 16) + avic_kick_vcpu_by_logical_id(kvm, avic_logical_id_table, + cluster + i, icrl); + } return 0; } static void avic_kick_target_vcpus(struct kvm *kvm, struct kvm_lapic *source, - u32 icrl, u32 icrh) + u32 icrl, u32 icrh, u32 index) { + u32 dest = apic_x2apic_mode(source) ? icrh : GET_XAPIC_DEST_FIELD(icrh); + unsigned long i; struct kvm_vcpu *vcpu; - int i; - kvm_for_each_vcpu(i, vcpu, kvm) { - bool m = kvm_apic_match_dest(vcpu, source, - icrl & APIC_SHORT_MASK, - GET_APIC_DEST_FIELD(icrh), - icrl & APIC_DEST_MASK); + if (!avic_kick_target_vcpus_fast(kvm, source, icrl, icrh, index)) + return; - if (m && !avic_vcpu_is_running(vcpu)) - kvm_vcpu_wake_up(vcpu); + trace_kvm_avic_kick_vcpu_slowpath(icrh, icrl, index); + + /* + * Wake any target vCPUs that are blocking, i.e. waiting for a wake + * event. There's no need to signal doorbells, as hardware has handled + * vCPUs that were in guest at the time of the IPI, and vCPUs that have + * since entered the guest will have processed pending IRQs at VMRUN. + */ + kvm_for_each_vcpu(i, vcpu, kvm) { + if (kvm_apic_match_dest(vcpu, source, icrl & APIC_SHORT_MASK, + dest, icrl & APIC_DEST_MASK)) + avic_kick_vcpu(vcpu, icrl); } } @@ -315,26 +609,30 @@ int avic_incomplete_ipi_interception(struct kvm_vcpu *vcpu) u32 icrh = svm->vmcb->control.exit_info_1 >> 32; u32 icrl = svm->vmcb->control.exit_info_1; u32 id = svm->vmcb->control.exit_info_2 >> 32; - u32 index = svm->vmcb->control.exit_info_2 & 0xFF; + u32 index = svm->vmcb->control.exit_info_2 & AVIC_PHYSICAL_MAX_INDEX_MASK; struct kvm_lapic *apic = vcpu->arch.apic; trace_kvm_avic_incomplete_ipi(vcpu->vcpu_id, icrh, icrl, id, index); switch (id) { + case AVIC_IPI_FAILURE_INVALID_TARGET: case AVIC_IPI_FAILURE_INVALID_INT_TYPE: /* - * AVIC hardware handles the generation of - * IPIs when the specified Message Type is Fixed - * (also known as fixed delivery mode) and - * the Trigger Mode is edge-triggered. The hardware - * also supports self and broadcast delivery modes - * specified via the Destination Shorthand(DSH) - * field of the ICRL. Logical and physical APIC ID - * formats are supported. All other IPI types cause - * a #VMEXIT, which needs to emulated. + * Emulate IPIs that are not handled by AVIC hardware, which + * only virtualizes Fixed, Edge-Triggered INTRs, and falls over + * if _any_ targets are invalid, e.g. if the logical mode mask + * is a superset of running vCPUs. + * + * The exit is a trap, e.g. ICR holds the correct value and RIP + * has been advanced, KVM is responsible only for emulating the + * IPI. Sadly, hardware may sometimes leave the BUSY flag set, + * in which case KVM needs to emulate the ICR write as well in + * order to clear the BUSY flag. */ - kvm_lapic_reg_write(apic, APIC_ICR2, icrh); - kvm_lapic_reg_write(apic, APIC_ICR, icrl); + if (icrl & APIC_ICR_BUSY) + kvm_apic_write_nodecode(vcpu, APIC_ICR); + else + kvm_apic_send_ipi(apic, icrl, icrh); break; case AVIC_IPI_FAILURE_TARGET_NOT_RUNNING: /* @@ -342,52 +640,55 @@ int avic_incomplete_ipi_interception(struct kvm_vcpu *vcpu) * set the appropriate IRR bits on the valid target * vcpus. So, we just need to kick the appropriate vcpu. */ - avic_kick_target_vcpus(vcpu->kvm, apic, icrl, icrh); - break; - case AVIC_IPI_FAILURE_INVALID_TARGET: - WARN_ONCE(1, "Invalid IPI target: index=%u, vcpu=%d, icr=%#0x:%#0x\n", - index, vcpu->vcpu_id, icrh, icrl); + avic_kick_target_vcpus(vcpu->kvm, apic, icrl, icrh, index); break; case AVIC_IPI_FAILURE_INVALID_BACKING_PAGE: WARN_ONCE(1, "Invalid backing page\n"); break; + case AVIC_IPI_FAILURE_INVALID_IPI_VECTOR: + /* Invalid IPI with vector < 16 */ + break; default: - pr_err("Unknown IPI interception\n"); + vcpu_unimpl(vcpu, "Unknown avic incomplete IPI interception\n"); } return 1; } +unsigned long avic_vcpu_get_apicv_inhibit_reasons(struct kvm_vcpu *vcpu) +{ + if (is_guest_mode(vcpu)) + return APICV_INHIBIT_REASON_NESTED; + return 0; +} + static u32 *avic_get_logical_id_entry(struct kvm_vcpu *vcpu, u32 ldr, bool flat) { struct kvm_svm *kvm_svm = to_kvm_svm(vcpu->kvm); - int index; - u32 *logical_apic_id_table; - int dlid = GET_APIC_LOGICAL_ID(ldr); + u32 cluster, index; - if (!dlid) - return NULL; + ldr = GET_APIC_LOGICAL_ID(ldr); - if (flat) { /* flat */ - index = ffs(dlid) - 1; - if (index > 7) - return NULL; - } else { /* cluster */ - int cluster = (dlid & 0xf0) >> 4; - int apic = ffs(dlid & 0x0f) - 1; - - if ((apic < 0) || (apic > 7) || - (cluster >= 0xf)) + if (flat) { + cluster = 0; + } else { + cluster = (ldr >> 4); + if (cluster >= 0xf) return NULL; - index = (cluster << 2) + apic; + ldr &= 0xf; } + if (!ldr || !is_power_of_2(ldr)) + return NULL; - logical_apic_id_table = (u32 *) page_address(kvm_svm->avic_logical_id_table_page); + index = __ffs(ldr); + if (WARN_ON_ONCE(index > 7)) + return NULL; + index += (cluster << 2); - return &logical_apic_id_table[index]; + return &kvm_svm->avic_logical_id_table[index]; } -static int avic_ldr_write(struct kvm_vcpu *vcpu, u8 g_physical_id, u32 ldr) +static void avic_ldr_write(struct kvm_vcpu *vcpu, u8 g_physical_id, u32 ldr) { bool flat; u32 *entry, new_entry; @@ -395,75 +696,47 @@ static int avic_ldr_write(struct kvm_vcpu *vcpu, u8 g_physical_id, u32 ldr) flat = kvm_lapic_get_reg(vcpu->arch.apic, APIC_DFR) == APIC_DFR_FLAT; entry = avic_get_logical_id_entry(vcpu, ldr, flat); if (!entry) - return -EINVAL; + return; new_entry = READ_ONCE(*entry); new_entry &= ~AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK; new_entry |= (g_physical_id & AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK); new_entry |= AVIC_LOGICAL_ID_ENTRY_VALID_MASK; WRITE_ONCE(*entry, new_entry); - - return 0; } static void avic_invalidate_logical_id_entry(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); bool flat = svm->dfr_reg == APIC_DFR_FLAT; - u32 *entry = avic_get_logical_id_entry(vcpu, svm->ldr_reg, flat); + u32 *entry; + /* Note: x2AVIC does not use logical APIC ID table */ + if (apic_x2apic_mode(vcpu->arch.apic)) + return; + + entry = avic_get_logical_id_entry(vcpu, svm->ldr_reg, flat); if (entry) clear_bit(AVIC_LOGICAL_ID_ENTRY_VALID_BIT, (unsigned long *)entry); } -static int avic_handle_ldr_update(struct kvm_vcpu *vcpu) +static void avic_handle_ldr_update(struct kvm_vcpu *vcpu) { - int ret = 0; struct vcpu_svm *svm = to_svm(vcpu); u32 ldr = kvm_lapic_get_reg(vcpu->arch.apic, APIC_LDR); u32 id = kvm_xapic_id(vcpu->arch.apic); + /* AVIC does not support LDR update for x2APIC */ + if (apic_x2apic_mode(vcpu->arch.apic)) + return; + if (ldr == svm->ldr_reg) - return 0; + return; avic_invalidate_logical_id_entry(vcpu); - if (ldr) - ret = avic_ldr_write(vcpu, id, ldr); - - if (!ret) - svm->ldr_reg = ldr; - - return ret; -} - -static int avic_handle_apic_id_update(struct kvm_vcpu *vcpu) -{ - u64 *old, *new; - struct vcpu_svm *svm = to_svm(vcpu); - u32 id = kvm_xapic_id(vcpu->arch.apic); - - if (vcpu->vcpu_id == id) - return 0; - - old = avic_get_physical_id_entry(vcpu, vcpu->vcpu_id); - new = avic_get_physical_id_entry(vcpu, id); - if (!new || !old) - return 1; - - /* We need to move physical_id_entry to new offset */ - *new = *old; - *old = 0ULL; - to_svm(vcpu)->avic_physical_id_cache = new; - - /* - * Also update the guest physical APIC ID in the logical - * APIC ID table entry if already setup the LDR. - */ - if (svm->ldr_reg) - avic_handle_ldr_update(vcpu); - - return 0; + svm->ldr_reg = ldr; + avic_ldr_write(vcpu, id, ldr); } static void avic_handle_dfr_update(struct kvm_vcpu *vcpu) @@ -478,30 +751,26 @@ static void avic_handle_dfr_update(struct kvm_vcpu *vcpu) svm->dfr_reg = dfr; } -static int avic_unaccel_trap_write(struct vcpu_svm *svm) +static int avic_unaccel_trap_write(struct kvm_vcpu *vcpu) { - struct kvm_lapic *apic = svm->vcpu.arch.apic; - u32 offset = svm->vmcb->control.exit_info_1 & + u32 offset = to_svm(vcpu)->vmcb->control.exit_info_1 & AVIC_UNACCEL_ACCESS_OFFSET_MASK; switch (offset) { - case APIC_ID: - if (avic_handle_apic_id_update(&svm->vcpu)) - return 0; - break; case APIC_LDR: - if (avic_handle_ldr_update(&svm->vcpu)) - return 0; + avic_handle_ldr_update(vcpu); break; case APIC_DFR: - avic_handle_dfr_update(&svm->vcpu); + avic_handle_dfr_update(vcpu); break; + case APIC_RRR: + /* Ignore writes to Read Remote Data, it's read-only. */ + return 1; default: break; } - kvm_lapic_reg_write(apic, offset, kvm_lapic_get_reg(apic, offset)); - + kvm_apic_write_nodecode(vcpu, offset); return 1; } @@ -551,7 +820,7 @@ int avic_unaccelerated_access_interception(struct kvm_vcpu *vcpu) if (trap) { /* Handling Trap */ WARN_ONCE(!write, "svm: Handling trap read.\n"); - ret = avic_unaccel_trap_write(svm); + ret = avic_unaccel_trap_write(vcpu); } else { /* Handling Fault */ ret = kvm_emulate_instruction(vcpu, 0); @@ -565,6 +834,9 @@ int avic_init_vcpu(struct vcpu_svm *svm) int ret; struct kvm_vcpu *vcpu = &svm->vcpu; + INIT_LIST_HEAD(&svm->ir_list); + raw_spin_lock_init(&svm->ir_list_lock); + if (!enable_apicv || !irqchip_in_kernel(vcpu->kvm)) return 0; @@ -572,457 +844,462 @@ int avic_init_vcpu(struct vcpu_svm *svm) if (ret) return ret; - INIT_LIST_HEAD(&svm->ir_list); - spin_lock_init(&svm->ir_list_lock); svm->dfr_reg = APIC_DFR_FLAT; return ret; } -void avic_post_state_restore(struct kvm_vcpu *vcpu) +void avic_apicv_post_state_restore(struct kvm_vcpu *vcpu) { - if (avic_handle_apic_id_update(vcpu) != 0) - return; avic_handle_dfr_update(vcpu); avic_handle_ldr_update(vcpu); } -void svm_toggle_avic_for_irq_window(struct kvm_vcpu *vcpu, bool activate) +static void svm_ir_list_del(struct kvm_kernel_irqfd *irqfd) { - if (!enable_apicv || !lapic_in_kernel(vcpu)) + struct kvm_vcpu *vcpu = irqfd->irq_bypass_vcpu; + unsigned long flags; + + if (!vcpu) return; - srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); - kvm_request_apicv_update(vcpu->kvm, activate, - APICV_INHIBIT_REASON_IRQWIN); - vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + raw_spin_lock_irqsave(&to_svm(vcpu)->ir_list_lock, flags); + list_del(&irqfd->vcpu_list); + raw_spin_unlock_irqrestore(&to_svm(vcpu)->ir_list_lock, flags); } -void svm_set_virtual_apic_mode(struct kvm_vcpu *vcpu) +int avic_pi_update_irte(struct kvm_kernel_irqfd *irqfd, struct kvm *kvm, + unsigned int host_irq, uint32_t guest_irq, + struct kvm_vcpu *vcpu, u32 vector) { - return; -} + /* + * If the IRQ was affined to a different vCPU, remove the IRTE metadata + * from the *previous* vCPU's list. + */ + svm_ir_list_del(irqfd); -void svm_hwapic_irr_update(struct kvm_vcpu *vcpu, int max_irr) -{ -} + if (vcpu) { + /* + * Try to enable guest_mode in IRTE, unless AVIC is inhibited, + * in which case configure the IRTE for legacy mode, but track + * the IRTE metadata so that it can be converted to guest mode + * if AVIC is enabled/uninhibited in the future. + */ + struct amd_iommu_pi_data pi_data = { + .ga_tag = AVIC_GATAG(to_kvm_svm(kvm)->avic_vm_id, + vcpu->vcpu_idx), + .is_guest_mode = kvm_vcpu_apicv_active(vcpu), + .vapic_addr = avic_get_backing_page_address(to_svm(vcpu)), + .vector = vector, + }; + struct vcpu_svm *svm = to_svm(vcpu); + u64 entry; + int ret; -void svm_hwapic_isr_update(struct kvm_vcpu *vcpu, int max_isr) -{ + /* + * Prevent the vCPU from being scheduled out or migrated until + * the IRTE is updated and its metadata has been added to the + * list of IRQs being posted to the vCPU, to ensure the IRTE + * isn't programmed with stale pCPU/IsRunning information. + */ + guard(raw_spinlock_irqsave)(&svm->ir_list_lock); + + /* + * Update the target pCPU for IOMMU doorbells if the vCPU is + * running. If the vCPU is NOT running, i.e. is blocking or + * scheduled out, KVM will update the pCPU info when the vCPU + * is awakened and/or scheduled in. See also avic_vcpu_load(). + */ + entry = svm->avic_physical_id_entry; + if (entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK) { + pi_data.cpu = entry & AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK; + } else { + pi_data.cpu = -1; + pi_data.ga_log_intr = entry & AVIC_PHYSICAL_ID_ENTRY_GA_LOG_INTR; + } + + ret = irq_set_vcpu_affinity(host_irq, &pi_data); + if (ret) + return ret; + + /* + * Revert to legacy mode if the IOMMU didn't provide metadata + * for the IRTE, which KVM needs to keep the IRTE up-to-date, + * e.g. if the vCPU is migrated or AVIC is disabled. + */ + if (WARN_ON_ONCE(!pi_data.ir_data)) { + irq_set_vcpu_affinity(host_irq, NULL); + return -EIO; + } + + irqfd->irq_bypass_data = pi_data.ir_data; + list_add(&irqfd->vcpu_list, &svm->ir_list); + return 0; + } + return irq_set_vcpu_affinity(host_irq, NULL); } -static int svm_set_pi_irte_mode(struct kvm_vcpu *vcpu, bool activate) +enum avic_vcpu_action { + /* + * There is no need to differentiate between activate and deactivate, + * as KVM only refreshes AVIC state when the vCPU is scheduled in and + * isn't blocking, i.e. the pCPU must always be (in)valid when AVIC is + * being (de)activated. + */ + AVIC_TOGGLE_ON_OFF = BIT(0), + AVIC_ACTIVATE = AVIC_TOGGLE_ON_OFF, + AVIC_DEACTIVATE = AVIC_TOGGLE_ON_OFF, + + /* + * No unique action is required to deal with a vCPU that stops/starts + * running. A vCPU that starts running by definition stops blocking as + * well, and a vCPU that stops running can't have been blocking, i.e. + * doesn't need to toggle GALogIntr. + */ + AVIC_START_RUNNING = 0, + AVIC_STOP_RUNNING = 0, + + /* + * When a vCPU starts blocking, KVM needs to set the GALogIntr flag + * int all associated IRTEs so that KVM can wake the vCPU if an IRQ is + * sent to the vCPU. + */ + AVIC_START_BLOCKING = BIT(1), +}; + +static void avic_update_iommu_vcpu_affinity(struct kvm_vcpu *vcpu, int cpu, + enum avic_vcpu_action action) { - int ret = 0; - unsigned long flags; - struct amd_svm_iommu_ir *ir; + bool ga_log_intr = (action & AVIC_START_BLOCKING); struct vcpu_svm *svm = to_svm(vcpu); + struct kvm_kernel_irqfd *irqfd; - if (!kvm_arch_has_assigned_device(vcpu->kvm)) - return 0; + lockdep_assert_held(&svm->ir_list_lock); /* * Here, we go through the per-vcpu ir_list to update all existing * interrupt remapping table entry targeting this vcpu. */ - spin_lock_irqsave(&svm->ir_list_lock, flags); - if (list_empty(&svm->ir_list)) - goto out; + return; + + list_for_each_entry(irqfd, &svm->ir_list, vcpu_list) { + void *data = irqfd->irq_bypass_data; - list_for_each_entry(ir, &svm->ir_list, node) { - if (activate) - ret = amd_iommu_activate_guest_mode(ir->data); + if (!(action & AVIC_TOGGLE_ON_OFF)) + WARN_ON_ONCE(amd_iommu_update_ga(data, cpu, ga_log_intr)); + else if (cpu >= 0) + WARN_ON_ONCE(amd_iommu_activate_guest_mode(data, cpu, ga_log_intr)); else - ret = amd_iommu_deactivate_guest_mode(ir->data); - if (ret) - break; + WARN_ON_ONCE(amd_iommu_deactivate_guest_mode(data)); } -out: - spin_unlock_irqrestore(&svm->ir_list_lock, flags); - return ret; } -void svm_refresh_apicv_exec_ctrl(struct kvm_vcpu *vcpu) +static void __avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu, + enum avic_vcpu_action action) { + struct kvm_svm *kvm_svm = to_kvm_svm(vcpu->kvm); + int h_physical_id = kvm_cpu_get_apicid(cpu); struct vcpu_svm *svm = to_svm(vcpu); - struct vmcb *vmcb = svm->vmcb; - bool activated = kvm_vcpu_apicv_active(vcpu); + unsigned long flags; + u64 entry; - if (!enable_apicv) + lockdep_assert_preemption_disabled(); + + if (WARN_ON(h_physical_id & ~AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK)) return; - if (activated) { - /** - * During AVIC temporary deactivation, guest could update - * APIC ID, DFR and LDR registers, which would not be trapped - * by avic_unaccelerated_access_interception(). In this case, - * we need to check and update the AVIC logical APIC ID table - * accordingly before re-activating. - */ - avic_post_state_restore(vcpu); - vmcb->control.int_ctl |= AVIC_ENABLE_MASK; - } else { - vmcb->control.int_ctl &= ~AVIC_ENABLE_MASK; - } - vmcb_mark_dirty(vmcb, VMCB_AVIC); + if (WARN_ON_ONCE(vcpu->vcpu_id * sizeof(entry) >= + PAGE_SIZE << avic_get_physical_id_table_order(vcpu->kvm))) + return; - svm_set_pi_irte_mode(vcpu, activated); -} + /* + * Grab the per-vCPU interrupt remapping lock even if the VM doesn't + * _currently_ have assigned devices, as that can change. Holding + * ir_list_lock ensures that either svm_ir_list_add() will consume + * up-to-date entry information, or that this task will wait until + * svm_ir_list_add() completes to set the new target pCPU. + */ + raw_spin_lock_irqsave(&svm->ir_list_lock, flags); -void svm_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap) -{ - return; -} + entry = svm->avic_physical_id_entry; + WARN_ON_ONCE(entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK); -int svm_deliver_avic_intr(struct kvm_vcpu *vcpu, int vec) -{ - if (!vcpu->arch.apicv_active) - return -1; + entry &= ~(AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK | + AVIC_PHYSICAL_ID_ENTRY_GA_LOG_INTR); + entry |= (h_physical_id & AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK); + entry |= AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK; - kvm_lapic_set_irr(vec, vcpu->arch.apic); - smp_mb__after_atomic(); + svm->avic_physical_id_entry = entry; - if (avic_vcpu_is_running(vcpu)) { - int cpuid = vcpu->cpu; + /* + * If IPI virtualization is disabled, clear IsRunning when updating the + * actual Physical ID table, so that the CPU never sees IsRunning=1. + * Keep the APIC ID up-to-date in the entry to minimize the chances of + * things going sideways if hardware peeks at the ID. + */ + if (!enable_ipiv) + entry &= ~AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK; - if (cpuid != get_cpu()) - wrmsrl(SVM_AVIC_DOORBELL, kvm_cpu_get_apicid(cpuid)); - put_cpu(); - } else - kvm_vcpu_wake_up(vcpu); + WRITE_ONCE(kvm_svm->avic_physical_id_table[vcpu->vcpu_id], entry); - return 0; -} + avic_update_iommu_vcpu_affinity(vcpu, h_physical_id, action); -bool svm_dy_apicv_has_pending_interrupt(struct kvm_vcpu *vcpu) -{ - return false; + raw_spin_unlock_irqrestore(&svm->ir_list_lock, flags); } -static void svm_ir_list_del(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi) +void avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { - unsigned long flags; - struct amd_svm_iommu_ir *cur; + /* + * No need to update anything if the vCPU is blocking, i.e. if the vCPU + * is being scheduled in after being preempted. The CPU entries in the + * Physical APIC table and IRTE are consumed iff IsRun{ning} is '1'. + * If the vCPU was migrated, its new CPU value will be stuffed when the + * vCPU unblocks. + */ + if (kvm_vcpu_is_blocking(vcpu)) + return; - spin_lock_irqsave(&svm->ir_list_lock, flags); - list_for_each_entry(cur, &svm->ir_list, node) { - if (cur->data != pi->ir_data) - continue; - list_del(&cur->node); - kfree(cur); - break; - } - spin_unlock_irqrestore(&svm->ir_list_lock, flags); + __avic_vcpu_load(vcpu, cpu, AVIC_START_RUNNING); } -static int svm_ir_list_add(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi) +static void __avic_vcpu_put(struct kvm_vcpu *vcpu, enum avic_vcpu_action action) { - int ret = 0; + struct kvm_svm *kvm_svm = to_kvm_svm(vcpu->kvm); + struct vcpu_svm *svm = to_svm(vcpu); unsigned long flags; - struct amd_svm_iommu_ir *ir; + u64 entry = svm->avic_physical_id_entry; - /** - * In some cases, the existing irte is updated and re-set, - * so we need to check here if it's already been * added - * to the ir_list. - */ - if (pi->ir_data && (pi->prev_ga_tag != 0)) { - struct kvm *kvm = svm->vcpu.kvm; - u32 vcpu_id = AVIC_GATAG_TO_VCPUID(pi->prev_ga_tag); - struct kvm_vcpu *prev_vcpu = kvm_get_vcpu_by_id(kvm, vcpu_id); - struct vcpu_svm *prev_svm; - - if (!prev_vcpu) { - ret = -EINVAL; - goto out; - } + lockdep_assert_preemption_disabled(); - prev_svm = to_svm(prev_vcpu); - svm_ir_list_del(prev_svm, pi); - } + if (WARN_ON_ONCE(vcpu->vcpu_id * sizeof(entry) >= + PAGE_SIZE << avic_get_physical_id_table_order(vcpu->kvm))) + return; - /** - * Allocating new amd_iommu_pi_data, which will get - * add to the per-vcpu ir_list. + /* + * Take and hold the per-vCPU interrupt remapping lock while updating + * the Physical ID entry even though the lock doesn't protect against + * multiple writers (see above). Holding ir_list_lock ensures that + * either svm_ir_list_add() will consume up-to-date entry information, + * or that this task will wait until svm_ir_list_add() completes to + * mark the vCPU as not running. */ - ir = kzalloc(sizeof(struct amd_svm_iommu_ir), GFP_KERNEL_ACCOUNT); - if (!ir) { - ret = -ENOMEM; - goto out; - } - ir->data = pi->ir_data; + raw_spin_lock_irqsave(&svm->ir_list_lock, flags); - spin_lock_irqsave(&svm->ir_list_lock, flags); - list_add(&ir->node, &svm->ir_list); - spin_unlock_irqrestore(&svm->ir_list_lock, flags); -out: - return ret; -} + avic_update_iommu_vcpu_affinity(vcpu, -1, action); -/* - * Note: - * The HW cannot support posting multicast/broadcast - * interrupts to a vCPU. So, we still use legacy interrupt - * remapping for these kind of interrupts. - * - * For lowest-priority interrupts, we only support - * those with single CPU as the destination, e.g. user - * configures the interrupts via /proc/irq or uses - * irqbalance to make the interrupts single-CPU. - */ -static int -get_pi_vcpu_info(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e, - struct vcpu_data *vcpu_info, struct vcpu_svm **svm) -{ - struct kvm_lapic_irq irq; - struct kvm_vcpu *vcpu = NULL; + WARN_ON_ONCE(entry & AVIC_PHYSICAL_ID_ENTRY_GA_LOG_INTR); - kvm_set_msi_irq(kvm, e, &irq); + /* + * Keep the previous APIC ID in the entry so that a rogue doorbell from + * hardware is at least restricted to a CPU associated with the vCPU. + */ + entry &= ~AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK; - if (!kvm_intr_is_single_vcpu(kvm, &irq, &vcpu) || - !kvm_irq_is_postable(&irq)) { - pr_debug("SVM: %s: use legacy intr remap mode for irq %u\n", - __func__, irq.vector); - return -1; - } + if (enable_ipiv) + WRITE_ONCE(kvm_svm->avic_physical_id_table[vcpu->vcpu_id], entry); - pr_debug("SVM: %s: use GA mode for irq %u\n", __func__, - irq.vector); - *svm = to_svm(vcpu); - vcpu_info->pi_desc_addr = __sme_set(page_to_phys((*svm)->avic_backing_page)); - vcpu_info->vector = irq.vector; + /* + * Note! Don't set AVIC_PHYSICAL_ID_ENTRY_GA_LOG_INTR in the table as + * it's a synthetic flag that usurps an unused should-be-zero bit. + */ + if (action & AVIC_START_BLOCKING) + entry |= AVIC_PHYSICAL_ID_ENTRY_GA_LOG_INTR; - return 0; + svm->avic_physical_id_entry = entry; + + raw_spin_unlock_irqrestore(&svm->ir_list_lock, flags); } -/* - * svm_update_pi_irte - set IRTE for Posted-Interrupts - * - * @kvm: kvm - * @host_irq: host irq of the interrupt - * @guest_irq: gsi of the interrupt - * @set: set or unset PI - * returns 0 on success, < 0 on failure - */ -int svm_update_pi_irte(struct kvm *kvm, unsigned int host_irq, - uint32_t guest_irq, bool set) +void avic_vcpu_put(struct kvm_vcpu *vcpu) { - struct kvm_kernel_irq_routing_entry *e; - struct kvm_irq_routing_table *irq_rt; - int idx, ret = -EINVAL; - - if (!kvm_arch_has_assigned_device(kvm) || - !irq_remapping_cap(IRQ_POSTING_CAP)) - return 0; - - pr_debug("SVM: %s: host_irq=%#x, guest_irq=%#x, set=%#x\n", - __func__, host_irq, guest_irq, set); - - idx = srcu_read_lock(&kvm->irq_srcu); - irq_rt = srcu_dereference(kvm->irq_routing, &kvm->irq_srcu); - WARN_ON(guest_irq >= irq_rt->nr_rt_entries); - - hlist_for_each_entry(e, &irq_rt->map[guest_irq], link) { - struct vcpu_data vcpu_info; - struct vcpu_svm *svm = NULL; + /* + * Note, reading the Physical ID entry outside of ir_list_lock is safe + * as only the pCPU that has loaded (or is loading) the vCPU is allowed + * to modify the entry, and preemption is disabled. I.e. the vCPU + * can't be scheduled out and thus avic_vcpu_{put,load}() can't run + * recursively. + */ + u64 entry = to_svm(vcpu)->avic_physical_id_entry; - if (e->type != KVM_IRQ_ROUTING_MSI) - continue; + /* + * Nothing to do if IsRunning == '0' due to vCPU blocking, i.e. if the + * vCPU is preempted while its in the process of blocking. WARN if the + * vCPU wasn't running and isn't blocking, KVM shouldn't attempt to put + * the AVIC if it wasn't previously loaded. + */ + if (!(entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK)) { + if (WARN_ON_ONCE(!kvm_vcpu_is_blocking(vcpu))) + return; - /** - * Here, we setup with legacy mode in the following cases: - * 1. When cannot target interrupt to a specific vcpu. - * 2. Unsetting posted interrupt. - * 3. APIC virtualization is disabled for the vcpu. - * 4. IRQ has incompatible delivery mode (SMI, INIT, etc) + /* + * The vCPU was preempted while blocking, ensure its IRTEs are + * configured to generate GA Log Interrupts. */ - if (!get_pi_vcpu_info(kvm, e, &vcpu_info, &svm) && set && - kvm_vcpu_apicv_active(&svm->vcpu)) { - struct amd_iommu_pi_data pi; - - /* Try to enable guest_mode in IRTE */ - pi.base = __sme_set(page_to_phys(svm->avic_backing_page) & - AVIC_HPA_MASK); - pi.ga_tag = AVIC_GATAG(to_kvm_svm(kvm)->avic_vm_id, - svm->vcpu.vcpu_id); - pi.is_guest_mode = true; - pi.vcpu_data = &vcpu_info; - ret = irq_set_vcpu_affinity(host_irq, &pi); - - /** - * Here, we successfully setting up vcpu affinity in - * IOMMU guest mode. Now, we need to store the posted - * interrupt information in a per-vcpu ir_list so that - * we can reference to them directly when we update vcpu - * scheduling information in IOMMU irte. - */ - if (!ret && pi.is_guest_mode) - svm_ir_list_add(svm, &pi); - } else { - /* Use legacy mode in IRTE */ - struct amd_iommu_pi_data pi; - - /** - * Here, pi is used to: - * - Tell IOMMU to use legacy mode for this interrupt. - * - Retrieve ga_tag of prior interrupt remapping data. - */ - pi.prev_ga_tag = 0; - pi.is_guest_mode = false; - ret = irq_set_vcpu_affinity(host_irq, &pi); - - /** - * Check if the posted interrupt was previously - * setup with the guest_mode by checking if the ga_tag - * was cached. If so, we need to clean up the per-vcpu - * ir_list. - */ - if (!ret && pi.prev_ga_tag) { - int id = AVIC_GATAG_TO_VCPUID(pi.prev_ga_tag); - struct kvm_vcpu *vcpu; - - vcpu = kvm_get_vcpu_by_id(kvm, id); - if (vcpu) - svm_ir_list_del(to_svm(vcpu), &pi); - } - } - - if (!ret && svm) { - trace_kvm_pi_irte_update(host_irq, svm->vcpu.vcpu_id, - e->gsi, vcpu_info.vector, - vcpu_info.pi_desc_addr, set); - } - - if (ret < 0) { - pr_err("%s: failed to update PI IRTE\n", __func__); - goto out; - } + if (!(WARN_ON_ONCE(!(entry & AVIC_PHYSICAL_ID_ENTRY_GA_LOG_INTR)))) + return; } - ret = 0; -out: - srcu_read_unlock(&kvm->irq_srcu, idx); - return ret; + __avic_vcpu_put(vcpu, kvm_vcpu_is_blocking(vcpu) ? AVIC_START_BLOCKING : + AVIC_STOP_RUNNING); } -bool svm_check_apicv_inhibit_reasons(ulong bit) +void avic_refresh_virtual_apic_mode(struct kvm_vcpu *vcpu) { - ulong supported = BIT(APICV_INHIBIT_REASON_DISABLE) | - BIT(APICV_INHIBIT_REASON_HYPERV) | - BIT(APICV_INHIBIT_REASON_NESTED) | - BIT(APICV_INHIBIT_REASON_IRQWIN) | - BIT(APICV_INHIBIT_REASON_PIT_REINJ) | - BIT(APICV_INHIBIT_REASON_X2APIC); - - return supported & BIT(bit); -} + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb *vmcb = svm->vmcb01.ptr; -void svm_pre_update_apicv_exec_ctrl(struct kvm *kvm, bool activate) -{ - avic_update_access_page(kvm, activate); + if (!lapic_in_kernel(vcpu) || !enable_apicv) + return; + + if (kvm_vcpu_apicv_active(vcpu)) { + /** + * During AVIC temporary deactivation, guest could update + * APIC ID, DFR and LDR registers, which would not be trapped + * by avic_unaccelerated_access_interception(). In this case, + * we need to check and update the AVIC logical APIC ID table + * accordingly before re-activating. + */ + avic_apicv_post_state_restore(vcpu); + avic_activate_vmcb(svm); + } else { + avic_deactivate_vmcb(svm); + } + vmcb_mark_dirty(vmcb, VMCB_AVIC); } -static inline int -avic_update_iommu_vcpu_affinity(struct kvm_vcpu *vcpu, int cpu, bool r) +void avic_refresh_apicv_exec_ctrl(struct kvm_vcpu *vcpu) { - int ret = 0; - unsigned long flags; - struct amd_svm_iommu_ir *ir; - struct vcpu_svm *svm = to_svm(vcpu); - - if (!kvm_arch_has_assigned_device(vcpu->kvm)) - return 0; + if (!enable_apicv) + return; - /* - * Here, we go through the per-vcpu ir_list to update all existing - * interrupt remapping table entry targeting this vcpu. - */ - spin_lock_irqsave(&svm->ir_list_lock, flags); + /* APICv should only be toggled on/off while the vCPU is running. */ + WARN_ON_ONCE(kvm_vcpu_is_blocking(vcpu)); - if (list_empty(&svm->ir_list)) - goto out; + avic_refresh_virtual_apic_mode(vcpu); - list_for_each_entry(ir, &svm->ir_list, node) { - ret = amd_iommu_update_ga(cpu, r, ir->data); - if (ret) - break; - } -out: - spin_unlock_irqrestore(&svm->ir_list_lock, flags); - return ret; + if (kvm_vcpu_apicv_active(vcpu)) + __avic_vcpu_load(vcpu, vcpu->cpu, AVIC_ACTIVATE); + else + __avic_vcpu_put(vcpu, AVIC_DEACTIVATE); } -void avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu) +void avic_vcpu_blocking(struct kvm_vcpu *vcpu) { - u64 entry; - /* ID = 0xff (broadcast), ID > 0xff (reserved) */ - int h_physical_id = kvm_cpu_get_apicid(cpu); - struct vcpu_svm *svm = to_svm(vcpu); - if (!kvm_vcpu_apicv_active(vcpu)) return; /* - * Since the host physical APIC id is 8 bits, - * we can support host APIC ID upto 255. + * Unload the AVIC when the vCPU is about to block, _before_ the vCPU + * actually blocks. + * + * Note, any IRQs that arrive before IsRunning=0 will not cause an + * incomplete IPI vmexit on the source; kvm_vcpu_check_block() handles + * this by checking vIRR one last time before blocking. The memory + * barrier implicit in set_current_state orders writing IsRunning=0 + * before reading the vIRR. The processor needs a matching memory + * barrier on interrupt delivery between writing IRR and reading + * IsRunning; the lack of this barrier might be the cause of errata #1235). + * + * Clear IsRunning=0 even if guest IRQs are disabled, i.e. even if KVM + * doesn't need to detect events for scheduling purposes. The doorbell + * used to signal running vCPUs cannot be blocked, i.e. will perturb the + * CPU and cause noisy neighbor problems if the VM is sending interrupts + * to the vCPU while it's scheduled out. */ - if (WARN_ON(h_physical_id > AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK)) - return; - - entry = READ_ONCE(*(svm->avic_physical_id_cache)); - WARN_ON(entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK); - - entry &= ~AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK; - entry |= (h_physical_id & AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK); + __avic_vcpu_put(vcpu, AVIC_START_BLOCKING); +} - entry &= ~AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK; - if (svm->avic_is_running) - entry |= AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK; +void avic_vcpu_unblocking(struct kvm_vcpu *vcpu) +{ + if (!kvm_vcpu_apicv_active(vcpu)) + return; - WRITE_ONCE(*(svm->avic_physical_id_cache), entry); - avic_update_iommu_vcpu_affinity(vcpu, h_physical_id, - svm->avic_is_running); + avic_vcpu_load(vcpu, vcpu->cpu); } -void avic_vcpu_put(struct kvm_vcpu *vcpu) +static bool __init avic_want_avic_enabled(void) { - u64 entry; - struct vcpu_svm *svm = to_svm(vcpu); + /* + * In "auto" mode, enable AVIC by default for Zen4+ if x2AVIC is + * supported (to avoid enabling partial support by default, and because + * x2AVIC should be supported by all Zen4+ CPUs). Explicitly check for + * family 0x19 and later (Zen5+), as the kernel's synthetic ZenX flags + * aren't inclusive of previous generations, i.e. the kernel will set + * at most one ZenX feature flag. + */ + if (avic == AVIC_AUTO_MODE) + avic = boot_cpu_has(X86_FEATURE_X2AVIC) && + (boot_cpu_data.x86 > 0x19 || cpu_feature_enabled(X86_FEATURE_ZEN4)); + + if (!avic || !npt_enabled) + return false; + + /* AVIC is a prerequisite for x2AVIC. */ + if (!boot_cpu_has(X86_FEATURE_AVIC) && !force_avic) { + if (boot_cpu_has(X86_FEATURE_X2AVIC)) + pr_warn(FW_BUG "Cannot enable x2AVIC, AVIC is unsupported\n"); + return false; + } - if (!kvm_vcpu_apicv_active(vcpu)) - return; + if (cc_platform_has(CC_ATTR_HOST_SEV_SNP) && + !boot_cpu_has(X86_FEATURE_HV_INUSE_WR_ALLOWED)) { + pr_warn("AVIC disabled: missing HvInUseWrAllowed on SNP-enabled system\n"); + return false; + } - entry = READ_ONCE(*(svm->avic_physical_id_cache)); - if (entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK) - avic_update_iommu_vcpu_affinity(vcpu, -1, 0); + /* + * Print a scary message if AVIC is force enabled to make it abundantly + * clear that ignoring CPUID could have repercussions. See Revision + * Guide for specific AMD processor for more details. + */ + if (!boot_cpu_has(X86_FEATURE_AVIC)) + pr_warn("AVIC unsupported in CPUID but force enabled, your system might crash and burn\n"); - entry &= ~AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK; - WRITE_ONCE(*(svm->avic_physical_id_cache), entry); + return true; } /* - * This function is called during VCPU halt/unhalt. + * Note: + * - The module param avic enable both xAPIC and x2APIC mode. + * - Hypervisor can support both xAVIC and x2AVIC in the same guest. + * - The mode can be switched at run-time. */ -static void avic_set_running(struct kvm_vcpu *vcpu, bool is_run) +bool __init avic_hardware_setup(void) { - struct vcpu_svm *svm = to_svm(vcpu); + avic = avic_want_avic_enabled(); + if (!avic) + return false; - svm->avic_is_running = is_run; - if (is_run) - avic_vcpu_load(vcpu, vcpu->cpu); - else - avic_vcpu_put(vcpu); -} + pr_info("AVIC enabled\n"); -void svm_vcpu_blocking(struct kvm_vcpu *vcpu) -{ - avic_set_running(vcpu, false); + /* AVIC is a prerequisite for x2AVIC. */ + x2avic_enabled = boot_cpu_has(X86_FEATURE_X2AVIC); + if (x2avic_enabled) { + if (cpu_feature_enabled(X86_FEATURE_X2AVIC_EXT)) + x2avic_max_physical_id = X2AVIC_4K_MAX_PHYSICAL_ID; + else + x2avic_max_physical_id = X2AVIC_MAX_PHYSICAL_ID; + pr_info("x2AVIC enabled (max %u vCPUs)\n", x2avic_max_physical_id + 1); + } else { + svm_x86_ops.allow_apicv_in_x2apic_without_x2apic_virtualization = true; + } + + /* + * Disable IPI virtualization for AMD Family 17h CPUs (Zen1 and Zen2) + * due to erratum 1235, which results in missed VM-Exits on the sender + * and thus missed wake events for blocking vCPUs due to the CPU + * failing to see a software update to clear IsRunning. + */ + enable_ipiv = enable_ipiv && boot_cpu_data.x86 != 0x17; + + amd_iommu_register_ga_log_notifier(&avic_ga_log_notifier); + + return true; } -void svm_vcpu_unblocking(struct kvm_vcpu *vcpu) +void avic_hardware_unsetup(void) { - if (kvm_check_request(KVM_REQ_APICV_UPDATE, vcpu)) - kvm_vcpu_update_apicv(vcpu); - avic_set_running(vcpu, true); + if (avic) + amd_iommu_register_ga_log_notifier(NULL); } diff --git a/arch/x86/kvm/svm/hyperv.c b/arch/x86/kvm/svm/hyperv.c new file mode 100644 index 000000000000..088f6429b24c --- /dev/null +++ b/arch/x86/kvm/svm/hyperv.c @@ -0,0 +1,18 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * AMD SVM specific code for Hyper-V on KVM. + * + * Copyright 2022 Red Hat, Inc. and/or its affiliates. + */ +#include "hyperv.h" + +void svm_hv_inject_synthetic_vmexit_post_tlb_flush(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + svm->vmcb->control.exit_code = HV_SVM_EXITCODE_ENL; + svm->vmcb->control.exit_code_hi = 0; + svm->vmcb->control.exit_info_1 = HV_SVM_ENL_EXITCODE_TRAP_AFTER_FLUSH; + svm->vmcb->control.exit_info_2 = 0; + nested_svm_vmexit(svm); +} diff --git a/arch/x86/kvm/svm/hyperv.h b/arch/x86/kvm/svm/hyperv.h new file mode 100644 index 000000000000..d3f8bfc05832 --- /dev/null +++ b/arch/x86/kvm/svm/hyperv.h @@ -0,0 +1,54 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Common Hyper-V on KVM and KVM on Hyper-V definitions (SVM). + */ + +#ifndef __ARCH_X86_KVM_SVM_HYPERV_H__ +#define __ARCH_X86_KVM_SVM_HYPERV_H__ + +#include <asm/mshyperv.h> + +#include "../hyperv.h" +#include "svm.h" + +#ifdef CONFIG_KVM_HYPERV +static inline void nested_svm_hv_update_vm_vp_ids(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct hv_vmcb_enlightenments *hve = &svm->nested.ctl.hv_enlightenments; + struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu); + + if (!hv_vcpu) + return; + + hv_vcpu->nested.pa_page_gpa = hve->partition_assist_page; + hv_vcpu->nested.vm_id = hve->hv_vm_id; + hv_vcpu->nested.vp_id = hve->hv_vp_id; +} + +static inline bool nested_svm_l2_tlb_flush_enabled(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct hv_vmcb_enlightenments *hve = &svm->nested.ctl.hv_enlightenments; + struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu); + + if (!hv_vcpu) + return false; + + if (!hve->hv_enlightenments_control.nested_flush_hypercall) + return false; + + return hv_vcpu->vp_assist_page.nested_control.features.directhypercall; +} + +void svm_hv_inject_synthetic_vmexit_post_tlb_flush(struct kvm_vcpu *vcpu); +#else /* CONFIG_KVM_HYPERV */ +static inline void nested_svm_hv_update_vm_vp_ids(struct kvm_vcpu *vcpu) {} +static inline bool nested_svm_l2_tlb_flush_enabled(struct kvm_vcpu *vcpu) +{ + return false; +} +static inline void svm_hv_inject_synthetic_vmexit_post_tlb_flush(struct kvm_vcpu *vcpu) {} +#endif /* CONFIG_KVM_HYPERV */ + +#endif /* __ARCH_X86_KVM_SVM_HYPERV_H__ */ diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c index 21d03e3a5dfd..c81005b24522 100644 --- a/arch/x86/kvm/svm/nested.c +++ b/arch/x86/kvm/svm/nested.c @@ -12,7 +12,7 @@ * Avi Kivity <avi@qumranet.com> */ -#define pr_fmt(fmt) "SVM: " fmt +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kvm_types.h> #include <linux/kvm_host.h> @@ -25,9 +25,11 @@ #include "trace.h" #include "mmu.h" #include "x86.h" +#include "smm.h" #include "cpuid.h" #include "lapic.h" #include "svm.h" +#include "hyperv.h" #define CC KVM_NESTED_VMENTER_CONSISTENCY_CHECK @@ -35,41 +37,25 @@ static void nested_svm_inject_npf_exit(struct kvm_vcpu *vcpu, struct x86_exception *fault) { struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb *vmcb = svm->vmcb; - if (svm->vmcb->control.exit_code != SVM_EXIT_NPF) { + if (vmcb->control.exit_code != SVM_EXIT_NPF) { /* * TODO: track the cause of the nested page fault, and * correctly fill in the high bits of exit_info_1. */ - svm->vmcb->control.exit_code = SVM_EXIT_NPF; - svm->vmcb->control.exit_code_hi = 0; - svm->vmcb->control.exit_info_1 = (1ULL << 32); - svm->vmcb->control.exit_info_2 = fault->address; + vmcb->control.exit_code = SVM_EXIT_NPF; + vmcb->control.exit_code_hi = 0; + vmcb->control.exit_info_1 = (1ULL << 32); + vmcb->control.exit_info_2 = fault->address; } - svm->vmcb->control.exit_info_1 &= ~0xffffffffULL; - svm->vmcb->control.exit_info_1 |= fault->error_code; + vmcb->control.exit_info_1 &= ~0xffffffffULL; + vmcb->control.exit_info_1 |= fault->error_code; nested_svm_vmexit(svm); } -static void svm_inject_page_fault_nested(struct kvm_vcpu *vcpu, struct x86_exception *fault) -{ - struct vcpu_svm *svm = to_svm(vcpu); - WARN_ON(!is_guest_mode(vcpu)); - - if (vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_EXCEPTION_OFFSET + PF_VECTOR) && - !svm->nested.nested_run_pending) { - svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + PF_VECTOR; - svm->vmcb->control.exit_code_hi = 0; - svm->vmcb->control.exit_info_1 = fault->error_code; - svm->vmcb->control.exit_info_2 = fault->address; - nested_svm_vmexit(svm); - } else { - kvm_inject_page_fault(vcpu, fault); - } -} - static u64 nested_svm_get_tdp_pdptr(struct kvm_vcpu *vcpu, int index) { struct vcpu_svm *svm = to_svm(vcpu); @@ -77,8 +63,12 @@ static u64 nested_svm_get_tdp_pdptr(struct kvm_vcpu *vcpu, int index) u64 pdpte; int ret; + /* + * Note, nCR3 is "assumed" to be 32-byte aligned, i.e. the CPU ignores + * nCR3[4:0] when loading PDPTEs from memory. + */ ret = kvm_vcpu_read_guest_page(vcpu, gpa_to_gfn(cr3), &pdpte, - offset_in_page(cr3) + index * 8, 8); + (cr3 & GENMASK(11, 5)) + index * 8, 8); if (ret) return 0; return pdpte; @@ -119,9 +109,24 @@ static void nested_svm_uninit_mmu_context(struct kvm_vcpu *vcpu) vcpu->arch.walk_mmu = &vcpu->arch.root_mmu; } +static bool nested_vmcb_needs_vls_intercept(struct vcpu_svm *svm) +{ + if (!guest_cpu_cap_has(&svm->vcpu, X86_FEATURE_V_VMSAVE_VMLOAD)) + return true; + + if (!nested_npt_enabled(svm)) + return true; + + if (!(svm->nested.ctl.virt_ext & VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK)) + return true; + + return false; +} + void recalc_intercepts(struct vcpu_svm *svm) { - struct vmcb_control_area *c, *h, *g; + struct vmcb_control_area *c, *h; + struct vmcb_ctrl_area_cached *g; unsigned int i; vmcb_mark_dirty(svm->vmcb, VMCB_INTERCEPTS); @@ -137,84 +142,172 @@ void recalc_intercepts(struct vcpu_svm *svm) c->intercepts[i] = h->intercepts[i]; if (g->int_ctl & V_INTR_MASKING_MASK) { - /* We only want the cr8 intercept bits of L1 */ - vmcb_clr_intercept(c, INTERCEPT_CR8_READ); - vmcb_clr_intercept(c, INTERCEPT_CR8_WRITE); - /* - * Once running L2 with HF_VINTR_MASK, EFLAGS.IF does not - * affect any interrupt we may want to inject; therefore, - * interrupt window vmexits are irrelevant to L0. + * If L2 is active and V_INTR_MASKING is enabled in vmcb12, + * disable intercept of CR8 writes as L2's CR8 does not affect + * any interrupt KVM may want to inject. + * + * Similarly, disable intercept of virtual interrupts (used to + * detect interrupt windows) if the saved RFLAGS.IF is '0', as + * the effective RFLAGS.IF for L1 interrupts will never be set + * while L2 is running (L2's RFLAGS.IF doesn't affect L1 IRQs). */ - vmcb_clr_intercept(c, INTERCEPT_VINTR); + vmcb_clr_intercept(c, INTERCEPT_CR8_WRITE); + if (!(svm->vmcb01.ptr->save.rflags & X86_EFLAGS_IF)) + vmcb_clr_intercept(c, INTERCEPT_VINTR); } - /* We don't want to see VMMCALLs from a nested guest */ - vmcb_clr_intercept(c, INTERCEPT_VMMCALL); + /* + * We want to see VMMCALLs from a nested guest only when Hyper-V L2 TLB + * flush feature is enabled. + */ + if (!nested_svm_l2_tlb_flush_enabled(&svm->vcpu)) + vmcb_clr_intercept(c, INTERCEPT_VMMCALL); for (i = 0; i < MAX_INTERCEPT; i++) c->intercepts[i] |= g->intercepts[i]; + + /* If SMI is not intercepted, ignore guest SMI intercept as well */ + if (!intercept_smi) + vmcb_clr_intercept(c, INTERCEPT_SMI); + + if (nested_vmcb_needs_vls_intercept(svm)) { + /* + * If the virtual VMLOAD/VMSAVE is not enabled for the L2, + * we must intercept these instructions to correctly + * emulate them in case L1 doesn't intercept them. + */ + vmcb_set_intercept(c, INTERCEPT_VMLOAD); + vmcb_set_intercept(c, INTERCEPT_VMSAVE); + } else { + WARN_ON(!(c->virt_ext & VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK)); + } } -static void copy_vmcb_control_area(struct vmcb_control_area *dst, - struct vmcb_control_area *from) +/* + * This array (and its actual size) holds the set of offsets (indexing by chunk + * size) to process when merging vmcb12's MSRPM with vmcb01's MSRPM. Note, the + * set of MSRs for which interception is disabled in vmcb01 is per-vCPU, e.g. + * based on CPUID features. This array only tracks MSRs that *might* be passed + * through to the guest. + * + * Hardcode the capacity of the array based on the maximum number of _offsets_. + * MSRs are batched together, so there are fewer offsets than MSRs. + */ +static int nested_svm_msrpm_merge_offsets[7] __ro_after_init; +static int nested_svm_nr_msrpm_merge_offsets __ro_after_init; +typedef unsigned long nsvm_msrpm_merge_t; + +int __init nested_svm_init_msrpm_merge_offsets(void) { - unsigned int i; + static const u32 merge_msrs[] __initconst = { + MSR_STAR, + MSR_IA32_SYSENTER_CS, + MSR_IA32_SYSENTER_EIP, + MSR_IA32_SYSENTER_ESP, + #ifdef CONFIG_X86_64 + MSR_GS_BASE, + MSR_FS_BASE, + MSR_KERNEL_GS_BASE, + MSR_LSTAR, + MSR_CSTAR, + MSR_SYSCALL_MASK, + #endif + MSR_IA32_SPEC_CTRL, + MSR_IA32_PRED_CMD, + MSR_IA32_FLUSH_CMD, + MSR_IA32_APERF, + MSR_IA32_MPERF, + MSR_IA32_LASTBRANCHFROMIP, + MSR_IA32_LASTBRANCHTOIP, + MSR_IA32_LASTINTFROMIP, + MSR_IA32_LASTINTTOIP, + }; + int i, j; - for (i = 0; i < MAX_INTERCEPT; i++) - dst->intercepts[i] = from->intercepts[i]; + for (i = 0; i < ARRAY_SIZE(merge_msrs); i++) { + int bit_nr = svm_msrpm_bit_nr(merge_msrs[i]); + u32 offset; - dst->iopm_base_pa = from->iopm_base_pa; - dst->msrpm_base_pa = from->msrpm_base_pa; - dst->tsc_offset = from->tsc_offset; - /* asid not copied, it is handled manually for svm->vmcb. */ - dst->tlb_ctl = from->tlb_ctl; - dst->int_ctl = from->int_ctl; - dst->int_vector = from->int_vector; - dst->int_state = from->int_state; - dst->exit_code = from->exit_code; - dst->exit_code_hi = from->exit_code_hi; - dst->exit_info_1 = from->exit_info_1; - dst->exit_info_2 = from->exit_info_2; - dst->exit_int_info = from->exit_int_info; - dst->exit_int_info_err = from->exit_int_info_err; - dst->nested_ctl = from->nested_ctl; - dst->event_inj = from->event_inj; - dst->event_inj_err = from->event_inj_err; - dst->nested_cr3 = from->nested_cr3; - dst->virt_ext = from->virt_ext; - dst->pause_filter_count = from->pause_filter_count; - dst->pause_filter_thresh = from->pause_filter_thresh; + if (WARN_ON(bit_nr < 0)) + return -EIO; + + /* + * Merging is done in chunks to reduce the number of accesses + * to L1's bitmap. + */ + offset = bit_nr / BITS_PER_BYTE / sizeof(nsvm_msrpm_merge_t); + + for (j = 0; j < nested_svm_nr_msrpm_merge_offsets; j++) { + if (nested_svm_msrpm_merge_offsets[j] == offset) + break; + } + + if (j < nested_svm_nr_msrpm_merge_offsets) + continue; + + if (WARN_ON(j >= ARRAY_SIZE(nested_svm_msrpm_merge_offsets))) + return -EIO; + + nested_svm_msrpm_merge_offsets[j] = offset; + nested_svm_nr_msrpm_merge_offsets++; + } + + return 0; } -static bool nested_svm_vmrun_msrpm(struct vcpu_svm *svm) +/* + * Merge L0's (KVM) and L1's (Nested VMCB) MSR permission bitmaps. The function + * is optimized in that it only merges the parts where KVM MSR permission bitmap + * may contain zero bits. + */ +static bool nested_svm_merge_msrpm(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); + nsvm_msrpm_merge_t *msrpm02 = svm->nested.msrpm; + nsvm_msrpm_merge_t *msrpm01 = svm->msrpm; + int i; + /* - * This function merges the msr permission bitmaps of kvm and the - * nested vmcb. It is optimized in that it only merges the parts where - * the kvm msr permission bitmap may contain zero bits + * MSR bitmap update can be skipped when: + * - MSR bitmap for L1 hasn't changed. + * - Nested hypervisor (L1) is attempting to launch the same L2 as + * before. + * - Nested hypervisor (L1) is using Hyper-V emulation interface and + * tells KVM (L0) there were no changes in MSR bitmap for L2. */ - int i; +#ifdef CONFIG_KVM_HYPERV + if (!svm->nested.force_msr_bitmap_recalc) { + struct hv_vmcb_enlightenments *hve = &svm->nested.ctl.hv_enlightenments; + + if (kvm_hv_hypercall_enabled(vcpu) && + hve->hv_enlightenments_control.msr_bitmap && + (svm->nested.ctl.clean & BIT(HV_VMCB_NESTED_ENLIGHTENMENTS))) + goto set_msrpm_base_pa; + } +#endif - if (!(vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_MSR_PROT))) + if (!(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_MSR_PROT))) return true; - for (i = 0; i < MSRPM_OFFSETS; i++) { - u32 value, p; - u64 offset; + for (i = 0; i < nested_svm_nr_msrpm_merge_offsets; i++) { + const int p = nested_svm_msrpm_merge_offsets[i]; + nsvm_msrpm_merge_t l1_val; + gpa_t gpa; - if (msrpm_offsets[i] == 0xffffffff) - break; + gpa = svm->nested.ctl.msrpm_base_pa + (p * sizeof(l1_val)); - p = msrpm_offsets[i]; - offset = svm->nested.ctl.msrpm_base_pa + (p * 4); - - if (kvm_vcpu_read_guest(&svm->vcpu, offset, &value, 4)) + if (kvm_vcpu_read_guest(vcpu, gpa, &l1_val, sizeof(l1_val))) return false; - svm->nested.msrpm[p] = svm->msrpm[p] | value; + msrpm02[p] = msrpm01[p] | l1_val; } + svm->nested.force_msr_bitmap_recalc = false; + +#ifdef CONFIG_KVM_HYPERV +set_msrpm_base_pa: +#endif svm->vmcb->control.msrpm_base_pa = __sme_set(__pa(svm->nested.msrpm)); return true; @@ -231,10 +324,10 @@ static bool nested_svm_check_bitmap_pa(struct kvm_vcpu *vcpu, u64 pa, u32 size) kvm_vcpu_is_legal_gpa(vcpu, addr + size - 1); } -static bool nested_vmcb_check_controls(struct kvm_vcpu *vcpu, - struct vmcb_control_area *control) +static bool __nested_vmcb_check_controls(struct kvm_vcpu *vcpu, + struct vmcb_ctrl_area_cached *control) { - if (CC(!vmcb_is_intercept(control, INTERCEPT_VMRUN))) + if (CC(!vmcb12_is_intercept(control, INTERCEPT_VMRUN))) return false; if (CC(control->asid == 0)) @@ -250,12 +343,28 @@ static bool nested_vmcb_check_controls(struct kvm_vcpu *vcpu, IOPM_SIZE))) return false; + if (CC((control->int_ctl & V_NMI_ENABLE_MASK) && + !vmcb12_is_intercept(control, INTERCEPT_NMI))) { + return false; + } + return true; } -static bool nested_vmcb_check_cr3_cr4(struct kvm_vcpu *vcpu, - struct vmcb_save_area *save) +/* Common checks that apply to both L1 and L2 state. */ +static bool __nested_vmcb_check_save(struct kvm_vcpu *vcpu, + struct vmcb_save_area_cached *save) { + if (CC(!(save->efer & EFER_SVME))) + return false; + + if (CC((save->cr0 & X86_CR0_CD) == 0 && (save->cr0 & X86_CR0_NW)) || + CC(save->cr0 & ~0xffffffffULL)) + return false; + + if (CC(!kvm_dr6_valid(save->dr6)) || CC(!kvm_dr7_valid(save->dr7))) + return false; + /* * These checks are also performed by KVM_SET_SREGS, * except that EFER.LMA is not checked by SVM against @@ -264,55 +373,109 @@ static bool nested_vmcb_check_cr3_cr4(struct kvm_vcpu *vcpu, if ((save->efer & EFER_LME) && (save->cr0 & X86_CR0_PG)) { if (CC(!(save->cr4 & X86_CR4_PAE)) || CC(!(save->cr0 & X86_CR0_PE)) || - CC(kvm_vcpu_is_illegal_gpa(vcpu, save->cr3))) + CC(!kvm_vcpu_is_legal_cr3(vcpu, save->cr3))) return false; } - if (CC(!kvm_is_valid_cr4(vcpu, save->cr4))) + /* Note, SVM doesn't have any additional restrictions on CR4. */ + if (CC(!__kvm_is_valid_cr4(vcpu, save->cr4))) + return false; + + if (CC(!kvm_valid_efer(vcpu, save->efer))) return false; return true; } -/* Common checks that apply to both L1 and L2 state. */ -static bool nested_vmcb_valid_sregs(struct kvm_vcpu *vcpu, - struct vmcb_save_area *save) +static bool nested_vmcb_check_save(struct kvm_vcpu *vcpu) { - /* - * FIXME: these should be done after copying the fields, - * to avoid TOC/TOU races. For these save area checks - * the possible damage is limited since kvm_set_cr0 and - * kvm_set_cr4 handle failure; EFER_SVME is an exception - * so it is force-set later in nested_prepare_vmcb_save. - */ - if (CC(!(save->efer & EFER_SVME))) - return false; + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb_save_area_cached *save = &svm->nested.save; - if (CC((save->cr0 & X86_CR0_CD) == 0 && (save->cr0 & X86_CR0_NW)) || - CC(save->cr0 & ~0xffffffffULL)) - return false; + return __nested_vmcb_check_save(vcpu, save); +} - if (CC(!kvm_dr6_valid(save->dr6)) || CC(!kvm_dr7_valid(save->dr7))) - return false; +static bool nested_vmcb_check_controls(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb_ctrl_area_cached *ctl = &svm->nested.ctl; - if (!nested_vmcb_check_cr3_cr4(vcpu, save)) - return false; + return __nested_vmcb_check_controls(vcpu, ctl); +} - if (CC(!kvm_valid_efer(vcpu, save->efer))) - return false; +static +void __nested_copy_vmcb_control_to_cache(struct kvm_vcpu *vcpu, + struct vmcb_ctrl_area_cached *to, + struct vmcb_control_area *from) +{ + unsigned int i; - return true; + for (i = 0; i < MAX_INTERCEPT; i++) + to->intercepts[i] = from->intercepts[i]; + + to->iopm_base_pa = from->iopm_base_pa; + to->msrpm_base_pa = from->msrpm_base_pa; + to->tsc_offset = from->tsc_offset; + to->tlb_ctl = from->tlb_ctl; + to->int_ctl = from->int_ctl; + to->int_vector = from->int_vector; + to->int_state = from->int_state; + to->exit_code = from->exit_code; + to->exit_code_hi = from->exit_code_hi; + to->exit_info_1 = from->exit_info_1; + to->exit_info_2 = from->exit_info_2; + to->exit_int_info = from->exit_int_info; + to->exit_int_info_err = from->exit_int_info_err; + to->nested_ctl = from->nested_ctl; + to->event_inj = from->event_inj; + to->event_inj_err = from->event_inj_err; + to->next_rip = from->next_rip; + to->nested_cr3 = from->nested_cr3; + to->virt_ext = from->virt_ext; + to->pause_filter_count = from->pause_filter_count; + to->pause_filter_thresh = from->pause_filter_thresh; + + /* Copy asid here because nested_vmcb_check_controls will check it. */ + to->asid = from->asid; + to->msrpm_base_pa &= ~0x0fffULL; + to->iopm_base_pa &= ~0x0fffULL; + +#ifdef CONFIG_KVM_HYPERV + /* Hyper-V extensions (Enlightened VMCB) */ + if (kvm_hv_hypercall_enabled(vcpu)) { + to->clean = from->clean; + memcpy(&to->hv_enlightenments, &from->hv_enlightenments, + sizeof(to->hv_enlightenments)); + } +#endif } -static void nested_load_control_from_vmcb12(struct vcpu_svm *svm, - struct vmcb_control_area *control) +void nested_copy_vmcb_control_to_cache(struct vcpu_svm *svm, + struct vmcb_control_area *control) { - copy_vmcb_control_area(&svm->nested.ctl, control); + __nested_copy_vmcb_control_to_cache(&svm->vcpu, &svm->nested.ctl, control); +} - /* Copy it here because nested_svm_check_controls will check it. */ - svm->nested.ctl.asid = control->asid; - svm->nested.ctl.msrpm_base_pa &= ~0x0fffULL; - svm->nested.ctl.iopm_base_pa &= ~0x0fffULL; +static void __nested_copy_vmcb_save_to_cache(struct vmcb_save_area_cached *to, + struct vmcb_save_area *from) +{ + /* + * Copy only fields that are validated, as we need them + * to avoid TOC/TOU races. + */ + to->efer = from->efer; + to->cr0 = from->cr0; + to->cr3 = from->cr3; + to->cr4 = from->cr4; + + to->dr6 = from->dr6; + to->dr7 = from->dr7; +} + +void nested_copy_vmcb_save_to_cache(struct vcpu_svm *svm, + struct vmcb_save_area *save) +{ + __nested_copy_vmcb_save_to_cache(&svm->nested.save, save); } /* @@ -327,18 +490,24 @@ void nested_sync_control_from_vmcb02(struct vcpu_svm *svm) /* Only a few fields of int_ctl are written by the processor. */ mask = V_IRQ_MASK | V_TPR_MASK; - if (!(svm->nested.ctl.int_ctl & V_INTR_MASKING_MASK) && - svm_is_intercept(svm, INTERCEPT_VINTR)) { - /* - * In order to request an interrupt window, L0 is usurping - * svm->vmcb->control.int_ctl and possibly setting V_IRQ - * even if it was clear in L1's VMCB. Restoring it would be - * wrong. However, in this case V_IRQ will remain true until - * interrupt_window_interception calls svm_clear_vintr and - * restores int_ctl. We can just leave it aside. - */ + /* + * Don't sync vmcb02 V_IRQ back to vmcb12 if KVM (L0) is intercepting + * virtual interrupts in order to request an interrupt window, as KVM + * has usurped vmcb02's int_ctl. If an interrupt window opens before + * the next VM-Exit, svm_clear_vintr() will restore vmcb12's int_ctl. + * If no window opens, V_IRQ will be correctly preserved in vmcb12's + * int_ctl (because it was never recognized while L2 was running). + */ + if (svm_is_intercept(svm, INTERCEPT_VINTR) && + !test_bit(INTERCEPT_VINTR, (unsigned long *)svm->nested.ctl.intercepts)) mask &= ~V_IRQ_MASK; - } + + if (nested_vgif_enabled(svm)) + mask |= V_GIF_MASK; + + if (nested_vnmi_enabled(svm)) + mask |= V_NMI_BLOCKING_MASK | V_NMI_PENDING_MASK; + svm->nested.ctl.int_ctl &= ~mask; svm->nested.ctl.int_ctl |= svm->vmcb->control.int_ctl & mask; } @@ -355,7 +524,7 @@ static void nested_save_pending_event_to_vmcb12(struct vcpu_svm *svm, unsigned int nr; if (vcpu->arch.exception.injected) { - nr = vcpu->arch.exception.nr; + nr = vcpu->arch.exception.vector; exit_int_info = nr | SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_EXEPT; if (vcpu->arch.exception.has_error_code) { @@ -380,13 +549,11 @@ static void nested_save_pending_event_to_vmcb12(struct vcpu_svm *svm, vmcb12->control.exit_int_info = exit_int_info; } -static inline bool nested_npt_enabled(struct vcpu_svm *svm) -{ - return svm->nested.ctl.nested_ctl & SVM_NESTED_CTL_NP_ENABLE; -} - static void nested_svm_transition_tlb_flush(struct kvm_vcpu *vcpu) { + /* Handle pending Hyper-V TLB flush requests */ + kvm_hv_nested_transtion_tlb_flush(vcpu, npt_enabled); + /* * TODO: optimize unconditional TLB flush/MMU sync. A partial list of * things to fix before this can be conditional: @@ -411,22 +578,21 @@ static void nested_svm_transition_tlb_flush(struct kvm_vcpu *vcpu) static int nested_svm_load_cr3(struct kvm_vcpu *vcpu, unsigned long cr3, bool nested_npt, bool reload_pdptrs) { - if (CC(kvm_vcpu_is_illegal_gpa(vcpu, cr3))) + if (CC(!kvm_vcpu_is_legal_cr3(vcpu, cr3))) return -EINVAL; if (reload_pdptrs && !nested_npt && is_pae_paging(vcpu) && - CC(!load_pdptrs(vcpu, vcpu->arch.walk_mmu, cr3))) + CC(!load_pdptrs(vcpu, cr3))) return -EINVAL; - if (!nested_npt) - kvm_mmu_new_pgd(vcpu, cr3); - vcpu->arch.cr3 = cr3; - kvm_register_mark_available(vcpu, VCPU_EXREG_CR3); /* Re-initialize the MMU, e.g. to pick up CR4 MMU role changes. */ kvm_init_mmu(vcpu); + if (!nested_npt) + kvm_mmu_new_pgd(vcpu, cr3); + return 0; } @@ -442,111 +608,263 @@ void nested_vmcb02_compute_g_pat(struct vcpu_svm *svm) static void nested_vmcb02_prepare_save(struct vcpu_svm *svm, struct vmcb *vmcb12) { bool new_vmcb12 = false; + struct vmcb *vmcb01 = svm->vmcb01.ptr; + struct vmcb *vmcb02 = svm->nested.vmcb02.ptr; + struct kvm_vcpu *vcpu = &svm->vcpu; nested_vmcb02_compute_g_pat(svm); + vmcb_mark_dirty(vmcb02, VMCB_NPT); /* Load the nested guest state */ if (svm->nested.vmcb12_gpa != svm->nested.last_vmcb12_gpa) { new_vmcb12 = true; svm->nested.last_vmcb12_gpa = svm->nested.vmcb12_gpa; + svm->nested.force_msr_bitmap_recalc = true; } if (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_SEG))) { - svm->vmcb->save.es = vmcb12->save.es; - svm->vmcb->save.cs = vmcb12->save.cs; - svm->vmcb->save.ss = vmcb12->save.ss; - svm->vmcb->save.ds = vmcb12->save.ds; - svm->vmcb->save.cpl = vmcb12->save.cpl; - vmcb_mark_dirty(svm->vmcb, VMCB_SEG); + vmcb02->save.es = vmcb12->save.es; + vmcb02->save.cs = vmcb12->save.cs; + vmcb02->save.ss = vmcb12->save.ss; + vmcb02->save.ds = vmcb12->save.ds; + vmcb02->save.cpl = vmcb12->save.cpl; + vmcb_mark_dirty(vmcb02, VMCB_SEG); } if (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_DT))) { - svm->vmcb->save.gdtr = vmcb12->save.gdtr; - svm->vmcb->save.idtr = vmcb12->save.idtr; - vmcb_mark_dirty(svm->vmcb, VMCB_DT); + vmcb02->save.gdtr = vmcb12->save.gdtr; + vmcb02->save.idtr = vmcb12->save.idtr; + vmcb_mark_dirty(vmcb02, VMCB_DT); } - kvm_set_rflags(&svm->vcpu, vmcb12->save.rflags | X86_EFLAGS_FIXED); + if (guest_cpu_cap_has(vcpu, X86_FEATURE_SHSTK) && + (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_CET)))) { + vmcb02->save.s_cet = vmcb12->save.s_cet; + vmcb02->save.isst_addr = vmcb12->save.isst_addr; + vmcb02->save.ssp = vmcb12->save.ssp; + vmcb_mark_dirty(vmcb02, VMCB_CET); + } - /* - * Force-set EFER_SVME even though it is checked earlier on the - * VMCB12, because the guest can flip the bit between the check - * and now. Clearing EFER_SVME would call svm_free_nested. - */ - svm_set_efer(&svm->vcpu, vmcb12->save.efer | EFER_SVME); + kvm_set_rflags(vcpu, vmcb12->save.rflags | X86_EFLAGS_FIXED); - svm_set_cr0(&svm->vcpu, vmcb12->save.cr0); - svm_set_cr4(&svm->vcpu, vmcb12->save.cr4); + svm_set_efer(vcpu, svm->nested.save.efer); + + svm_set_cr0(vcpu, svm->nested.save.cr0); + svm_set_cr4(vcpu, svm->nested.save.cr4); svm->vcpu.arch.cr2 = vmcb12->save.cr2; - kvm_rax_write(&svm->vcpu, vmcb12->save.rax); - kvm_rsp_write(&svm->vcpu, vmcb12->save.rsp); - kvm_rip_write(&svm->vcpu, vmcb12->save.rip); + kvm_rax_write(vcpu, vmcb12->save.rax); + kvm_rsp_write(vcpu, vmcb12->save.rsp); + kvm_rip_write(vcpu, vmcb12->save.rip); /* In case we don't even reach vcpu_run, the fields are not updated */ - svm->vmcb->save.rax = vmcb12->save.rax; - svm->vmcb->save.rsp = vmcb12->save.rsp; - svm->vmcb->save.rip = vmcb12->save.rip; + vmcb02->save.rax = vmcb12->save.rax; + vmcb02->save.rsp = vmcb12->save.rsp; + vmcb02->save.rip = vmcb12->save.rip; /* These bits will be set properly on the first execution when new_vmc12 is true */ if (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_DR))) { - svm->vmcb->save.dr7 = vmcb12->save.dr7 | DR7_FIXED_1; - svm->vcpu.arch.dr6 = vmcb12->save.dr6 | DR6_ACTIVE_LOW; - vmcb_mark_dirty(svm->vmcb, VMCB_DR); + vmcb02->save.dr7 = svm->nested.save.dr7 | DR7_FIXED_1; + svm->vcpu.arch.dr6 = svm->nested.save.dr6 | DR6_ACTIVE_LOW; + vmcb_mark_dirty(vmcb02, VMCB_DR); + } + + if (unlikely(guest_cpu_cap_has(vcpu, X86_FEATURE_LBRV) && + (svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK))) { + /* + * Reserved bits of DEBUGCTL are ignored. Be consistent with + * svm_set_msr's definition of reserved bits. + */ + svm_copy_lbrs(vmcb02, vmcb12); + vmcb02->save.dbgctl &= ~DEBUGCTL_RESERVED_BITS; + } else { + svm_copy_lbrs(vmcb02, vmcb01); } + svm_update_lbrv(&svm->vcpu); } -static void nested_vmcb02_prepare_control(struct vcpu_svm *svm) +static inline bool is_evtinj_soft(u32 evtinj) { - const u32 mask = V_INTR_MASKING_MASK | V_GIF_ENABLE_MASK | V_GIF_MASK; + u32 type = evtinj & SVM_EVTINJ_TYPE_MASK; + u8 vector = evtinj & SVM_EVTINJ_VEC_MASK; + + if (!(evtinj & SVM_EVTINJ_VALID)) + return false; + + if (type == SVM_EVTINJ_TYPE_SOFT) + return true; + + return type == SVM_EVTINJ_TYPE_EXEPT && kvm_exception_is_soft(vector); +} + +static bool is_evtinj_nmi(u32 evtinj) +{ + u32 type = evtinj & SVM_EVTINJ_TYPE_MASK; + + if (!(evtinj & SVM_EVTINJ_VALID)) + return false; + + return type == SVM_EVTINJ_TYPE_NMI; +} + +static void nested_vmcb02_prepare_control(struct vcpu_svm *svm, + unsigned long vmcb12_rip, + unsigned long vmcb12_csbase) +{ + u32 int_ctl_vmcb01_bits = V_INTR_MASKING_MASK; + u32 int_ctl_vmcb12_bits = V_TPR_MASK | V_IRQ_INJECTION_BITS_MASK; + struct kvm_vcpu *vcpu = &svm->vcpu; + struct vmcb *vmcb01 = svm->vmcb01.ptr; + struct vmcb *vmcb02 = svm->nested.vmcb02.ptr; + u32 pause_count12; + u32 pause_thresh12; + + nested_svm_transition_tlb_flush(vcpu); + + /* Enter Guest-Mode */ + enter_guest_mode(vcpu); /* * Filled at exit: exit_code, exit_code_hi, exit_info_1, exit_info_2, * exit_int_info, exit_int_info_err, next_rip, insn_len, insn_bytes. */ - /* - * Also covers avic_vapic_bar, avic_backing_page, avic_logical_id, - * avic_physical_id. - */ - WARN_ON(svm->vmcb01.ptr->control.int_ctl & AVIC_ENABLE_MASK); + if (guest_cpu_cap_has(vcpu, X86_FEATURE_VGIF) && + (svm->nested.ctl.int_ctl & V_GIF_ENABLE_MASK)) + int_ctl_vmcb12_bits |= (V_GIF_MASK | V_GIF_ENABLE_MASK); + else + int_ctl_vmcb01_bits |= (V_GIF_MASK | V_GIF_ENABLE_MASK); + + if (vnmi) { + if (vmcb01->control.int_ctl & V_NMI_PENDING_MASK) { + svm->vcpu.arch.nmi_pending++; + kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); + } + if (nested_vnmi_enabled(svm)) + int_ctl_vmcb12_bits |= (V_NMI_PENDING_MASK | + V_NMI_ENABLE_MASK | + V_NMI_BLOCKING_MASK); + } /* Copied from vmcb01. msrpm_base can be overwritten later. */ - svm->vmcb->control.nested_ctl = svm->vmcb01.ptr->control.nested_ctl; - svm->vmcb->control.iopm_base_pa = svm->vmcb01.ptr->control.iopm_base_pa; - svm->vmcb->control.msrpm_base_pa = svm->vmcb01.ptr->control.msrpm_base_pa; + vmcb02->control.nested_ctl = vmcb01->control.nested_ctl; + vmcb02->control.iopm_base_pa = vmcb01->control.iopm_base_pa; + vmcb02->control.msrpm_base_pa = vmcb01->control.msrpm_base_pa; + vmcb_mark_dirty(vmcb02, VMCB_PERM_MAP); + + /* + * Stash vmcb02's counter if the guest hasn't moved past the guilty + * instruction; otherwise, reset the counter to '0'. + * + * In order to detect if L2 has made forward progress or not, track the + * RIP at which a bus lock has occurred on a per-vmcb12 basis. If RIP + * is changed, guest has clearly made forward progress, bus_lock_counter + * still remained '1', so reset bus_lock_counter to '0'. Eg. In the + * scenario, where a buslock happened in L1 before VMRUN, the bus lock + * firmly happened on an instruction in the past. Even if vmcb01's + * counter is still '1', (because the guilty instruction got patched), + * the vCPU has clearly made forward progress and so KVM should reset + * vmcb02's counter to '0'. + * + * If the RIP hasn't changed, stash the bus lock counter at nested VMRUN + * to prevent the same guilty instruction from triggering a VM-Exit. Eg. + * if userspace rate-limits the vCPU, then it's entirely possible that + * L1's tick interrupt is pending by the time userspace re-runs the + * vCPU. If KVM unconditionally clears the counter on VMRUN, then when + * L1 re-enters L2, the same instruction will trigger a VM-Exit and the + * entire cycle start over. + */ + if (vmcb02->save.rip && (svm->nested.ctl.bus_lock_rip == vmcb02->save.rip)) + vmcb02->control.bus_lock_counter = 1; + else + vmcb02->control.bus_lock_counter = 0; /* Done at vmrun: asid. */ /* Also overwritten later if necessary. */ - svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING; + vmcb02->control.tlb_ctl = TLB_CONTROL_DO_NOTHING; /* nested_cr3. */ if (nested_npt_enabled(svm)) nested_svm_init_mmu_context(vcpu); - svm->vmcb->control.tsc_offset = vcpu->arch.tsc_offset = - vcpu->arch.l1_tsc_offset + svm->nested.ctl.tsc_offset; + vcpu->arch.tsc_offset = kvm_calc_nested_tsc_offset( + vcpu->arch.l1_tsc_offset, + svm->nested.ctl.tsc_offset, + svm->tsc_ratio_msr); - svm->vmcb->control.int_ctl = - (svm->nested.ctl.int_ctl & ~mask) | - (svm->vmcb01.ptr->control.int_ctl & mask); + vmcb02->control.tsc_offset = vcpu->arch.tsc_offset; - svm->vmcb->control.virt_ext = svm->nested.ctl.virt_ext; - svm->vmcb->control.int_vector = svm->nested.ctl.int_vector; - svm->vmcb->control.int_state = svm->nested.ctl.int_state; - svm->vmcb->control.event_inj = svm->nested.ctl.event_inj; - svm->vmcb->control.event_inj_err = svm->nested.ctl.event_inj_err; + if (guest_cpu_cap_has(vcpu, X86_FEATURE_TSCRATEMSR) && + svm->tsc_ratio_msr != kvm_caps.default_tsc_scaling_ratio) + nested_svm_update_tsc_ratio_msr(vcpu); - svm->vmcb->control.pause_filter_count = svm->nested.ctl.pause_filter_count; - svm->vmcb->control.pause_filter_thresh = svm->nested.ctl.pause_filter_thresh; + vmcb02->control.int_ctl = + (svm->nested.ctl.int_ctl & int_ctl_vmcb12_bits) | + (vmcb01->control.int_ctl & int_ctl_vmcb01_bits); - nested_svm_transition_tlb_flush(vcpu); + vmcb02->control.int_vector = svm->nested.ctl.int_vector; + vmcb02->control.int_state = svm->nested.ctl.int_state; + vmcb02->control.event_inj = svm->nested.ctl.event_inj; + vmcb02->control.event_inj_err = svm->nested.ctl.event_inj_err; - /* Enter Guest-Mode */ - enter_guest_mode(vcpu); + /* + * next_rip is consumed on VMRUN as the return address pushed on the + * stack for injected soft exceptions/interrupts. If nrips is exposed + * to L1, take it verbatim from vmcb12. If nrips is supported in + * hardware but not exposed to L1, stuff the actual L2 RIP to emulate + * what a nrips=0 CPU would do (L1 is responsible for advancing RIP + * prior to injecting the event). + */ + if (guest_cpu_cap_has(vcpu, X86_FEATURE_NRIPS)) + vmcb02->control.next_rip = svm->nested.ctl.next_rip; + else if (boot_cpu_has(X86_FEATURE_NRIPS)) + vmcb02->control.next_rip = vmcb12_rip; + + svm->nmi_l1_to_l2 = is_evtinj_nmi(vmcb02->control.event_inj); + if (is_evtinj_soft(vmcb02->control.event_inj)) { + svm->soft_int_injected = true; + svm->soft_int_csbase = vmcb12_csbase; + svm->soft_int_old_rip = vmcb12_rip; + if (guest_cpu_cap_has(vcpu, X86_FEATURE_NRIPS)) + svm->soft_int_next_rip = svm->nested.ctl.next_rip; + else + svm->soft_int_next_rip = vmcb12_rip; + } + + /* LBR_CTL_ENABLE_MASK is controlled by svm_update_lbrv() */ + + if (!nested_vmcb_needs_vls_intercept(svm)) + vmcb02->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; + + if (guest_cpu_cap_has(vcpu, X86_FEATURE_PAUSEFILTER)) + pause_count12 = svm->nested.ctl.pause_filter_count; + else + pause_count12 = 0; + if (guest_cpu_cap_has(vcpu, X86_FEATURE_PFTHRESHOLD)) + pause_thresh12 = svm->nested.ctl.pause_filter_thresh; + else + pause_thresh12 = 0; + if (kvm_pause_in_guest(svm->vcpu.kvm)) { + /* use guest values since host doesn't intercept PAUSE */ + vmcb02->control.pause_filter_count = pause_count12; + vmcb02->control.pause_filter_thresh = pause_thresh12; + + } else { + /* start from host values otherwise */ + vmcb02->control.pause_filter_count = vmcb01->control.pause_filter_count; + vmcb02->control.pause_filter_thresh = vmcb01->control.pause_filter_thresh; + + /* ... but ensure filtering is disabled if so requested. */ + if (vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_PAUSE)) { + if (!pause_count12) + vmcb02->control.pause_filter_count = 0; + if (!pause_thresh12) + vmcb02->control.pause_filter_thresh = 0; + } + } /* * Merge guest and host intercepts - must be called with vcpu in @@ -568,16 +886,20 @@ static void nested_svm_copy_common_state(struct vmcb *from_vmcb, struct vmcb *to } int enter_svm_guest_mode(struct kvm_vcpu *vcpu, u64 vmcb12_gpa, - struct vmcb *vmcb12) + struct vmcb *vmcb12, bool from_vmrun) { struct vcpu_svm *svm = to_svm(vcpu); int ret; - trace_kvm_nested_vmrun(svm->vmcb->save.rip, vmcb12_gpa, - vmcb12->save.rip, - vmcb12->control.int_ctl, - vmcb12->control.event_inj, - vmcb12->control.nested_ctl); + trace_kvm_nested_vmenter(svm->vmcb->save.rip, + vmcb12_gpa, + vmcb12->save.rip, + vmcb12->control.int_ctl, + vmcb12->control.event_inj, + vmcb12->control.nested_ctl, + vmcb12->control.nested_cr3, + vmcb12->save.cr3, + KVM_ISA_SVM); trace_kvm_nested_intercepts(vmcb12->control.intercepts[INTERCEPT_CR] & 0xffff, vmcb12->control.intercepts[INTERCEPT_CR] >> 16, @@ -594,19 +916,24 @@ int enter_svm_guest_mode(struct kvm_vcpu *vcpu, u64 vmcb12_gpa, nested_svm_copy_common_state(svm->vmcb01.ptr, svm->nested.vmcb02.ptr); svm_switch_vmcb(svm, &svm->nested.vmcb02); - nested_vmcb02_prepare_control(svm); + nested_vmcb02_prepare_control(svm, vmcb12->save.rip, vmcb12->save.cs.base); nested_vmcb02_prepare_save(svm, vmcb12); - ret = nested_svm_load_cr3(&svm->vcpu, vmcb12->save.cr3, - nested_npt_enabled(svm), true); + ret = nested_svm_load_cr3(&svm->vcpu, svm->nested.save.cr3, + nested_npt_enabled(svm), from_vmrun); if (ret) return ret; - if (!npt_enabled) - vcpu->arch.mmu->inject_page_fault = svm_inject_page_fault_nested; + if (!from_vmrun) + kvm_make_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu); svm_set_gif(svm, true); + if (kvm_vcpu_apicv_active(vcpu)) + kvm_make_request(KVM_REQ_APICV_UPDATE, vcpu); + + nested_svm_hv_update_vm_vp_ids(vcpu); + return 0; } @@ -617,12 +944,25 @@ int nested_svm_vmrun(struct kvm_vcpu *vcpu) struct vmcb *vmcb12; struct kvm_host_map map; u64 vmcb12_gpa; + struct vmcb *vmcb01 = svm->vmcb01.ptr; + + if (!svm->nested.hsave_msr) { + kvm_inject_gp(vcpu, 0); + return 1; + } if (is_smm(vcpu)) { kvm_queue_exception(vcpu, UD_VECTOR); return 1; } + /* This fails when VP assist page is enabled but the supplied GPA is bogus */ + ret = kvm_hv_verify_vp_assist(vcpu); + if (ret) { + kvm_inject_gp(vcpu, 0); + return ret; + } + vmcb12_gpa = svm->vmcb->save.rax; ret = kvm_vcpu_map(vcpu, gpa_to_gfn(vmcb12_gpa), &map); if (ret == -EINVAL) { @@ -639,10 +979,11 @@ int nested_svm_vmrun(struct kvm_vcpu *vcpu) if (WARN_ON_ONCE(!svm->nested.initialized)) return -EINVAL; - nested_load_control_from_vmcb12(svm, &vmcb12->control); + nested_copy_vmcb_control_to_cache(svm, &vmcb12->control); + nested_copy_vmcb_save_to_cache(svm, &vmcb12->save); - if (!nested_vmcb_valid_sregs(vcpu, &vmcb12->save) || - !nested_vmcb_check_controls(vcpu, &svm->nested.ctl)) { + if (!nested_vmcb_check_save(vcpu) || + !nested_vmcb_check_controls(vcpu)) { vmcb12->control.exit_code = SVM_EXIT_ERR; vmcb12->control.exit_code_hi = 0; vmcb12->control.exit_info_1 = 0; @@ -650,34 +991,31 @@ int nested_svm_vmrun(struct kvm_vcpu *vcpu) goto out; } - - /* Clear internal status */ - kvm_clear_exception_queue(vcpu); - kvm_clear_interrupt_queue(vcpu); - /* * Since vmcb01 is not in use, we can use it to store some of the L1 * state. */ - svm->vmcb01.ptr->save.efer = vcpu->arch.efer; - svm->vmcb01.ptr->save.cr0 = kvm_read_cr0(vcpu); - svm->vmcb01.ptr->save.cr4 = vcpu->arch.cr4; - svm->vmcb01.ptr->save.rflags = kvm_get_rflags(vcpu); - svm->vmcb01.ptr->save.rip = kvm_rip_read(vcpu); + vmcb01->save.efer = vcpu->arch.efer; + vmcb01->save.cr0 = kvm_read_cr0(vcpu); + vmcb01->save.cr4 = vcpu->arch.cr4; + vmcb01->save.rflags = kvm_get_rflags(vcpu); + vmcb01->save.rip = kvm_rip_read(vcpu); if (!npt_enabled) - svm->vmcb01.ptr->save.cr3 = kvm_read_cr3(vcpu); + vmcb01->save.cr3 = kvm_read_cr3(vcpu); svm->nested.nested_run_pending = 1; - if (enter_svm_guest_mode(vcpu, vmcb12_gpa, vmcb12)) + if (enter_svm_guest_mode(vcpu, vmcb12_gpa, vmcb12, true)) goto out_exit_err; - if (nested_svm_vmrun_msrpm(svm)) + if (nested_svm_merge_msrpm(vcpu)) goto out; out_exit_err: svm->nested.nested_run_pending = 0; + svm->nmi_l1_to_l2 = false; + svm->soft_int_injected = false; svm->vmcb->control.exit_code = SVM_EXIT_ERR; svm->vmcb->control.exit_code_hi = 0; @@ -687,12 +1025,39 @@ out_exit_err: nested_svm_vmexit(svm); out: - kvm_vcpu_unmap(vcpu, &map, true); + kvm_vcpu_unmap(vcpu, &map); return ret; } -void nested_svm_vmloadsave(struct vmcb *from_vmcb, struct vmcb *to_vmcb) +/* Copy state save area fields which are handled by VMRUN */ +void svm_copy_vmrun_state(struct vmcb_save_area *to_save, + struct vmcb_save_area *from_save) +{ + to_save->es = from_save->es; + to_save->cs = from_save->cs; + to_save->ss = from_save->ss; + to_save->ds = from_save->ds; + to_save->gdtr = from_save->gdtr; + to_save->idtr = from_save->idtr; + to_save->rflags = from_save->rflags | X86_EFLAGS_FIXED; + to_save->efer = from_save->efer; + to_save->cr0 = from_save->cr0; + to_save->cr3 = from_save->cr3; + to_save->cr4 = from_save->cr4; + to_save->rax = from_save->rax; + to_save->rsp = from_save->rsp; + to_save->rip = from_save->rip; + to_save->cpl = 0; + + if (kvm_cpu_cap_has(X86_FEATURE_SHSTK)) { + to_save->s_cet = from_save->s_cet; + to_save->isst_addr = from_save->isst_addr; + to_save->ssp = from_save->ssp; + } +} + +void svm_copy_vmloadsave_state(struct vmcb *to_vmcb, struct vmcb *from_vmcb) { to_vmcb->save.fs = from_vmcb->save.fs; to_vmcb->save.gs = from_vmcb->save.gs; @@ -711,14 +1076,12 @@ void nested_svm_vmloadsave(struct vmcb *from_vmcb, struct vmcb *to_vmcb) int nested_svm_vmexit(struct vcpu_svm *svm) { struct kvm_vcpu *vcpu = &svm->vcpu; + struct vmcb *vmcb01 = svm->vmcb01.ptr; + struct vmcb *vmcb02 = svm->nested.vmcb02.ptr; struct vmcb *vmcb12; - struct vmcb *vmcb = svm->vmcb; struct kvm_host_map map; int rc; - /* Triple faults in L2 should never escape. */ - WARN_ON_ONCE(kvm_check_request(KVM_REQ_TRIPLE_FAULT, vcpu)); - rc = kvm_vcpu_map(vcpu, gpa_to_gfn(svm->nested.vmcb12_gpa), &map); if (rc) { if (rc == -EINVAL) @@ -736,66 +1099,131 @@ int nested_svm_vmexit(struct vcpu_svm *svm) kvm_clear_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu); /* in case we halted in L2 */ - svm->vcpu.arch.mp_state = KVM_MP_STATE_RUNNABLE; + kvm_set_mp_state(vcpu, KVM_MP_STATE_RUNNABLE); /* Give the current vmcb to the guest */ - vmcb12->save.es = vmcb->save.es; - vmcb12->save.cs = vmcb->save.cs; - vmcb12->save.ss = vmcb->save.ss; - vmcb12->save.ds = vmcb->save.ds; - vmcb12->save.gdtr = vmcb->save.gdtr; - vmcb12->save.idtr = vmcb->save.idtr; + vmcb12->save.es = vmcb02->save.es; + vmcb12->save.cs = vmcb02->save.cs; + vmcb12->save.ss = vmcb02->save.ss; + vmcb12->save.ds = vmcb02->save.ds; + vmcb12->save.gdtr = vmcb02->save.gdtr; + vmcb12->save.idtr = vmcb02->save.idtr; vmcb12->save.efer = svm->vcpu.arch.efer; vmcb12->save.cr0 = kvm_read_cr0(vcpu); vmcb12->save.cr3 = kvm_read_cr3(vcpu); - vmcb12->save.cr2 = vmcb->save.cr2; + vmcb12->save.cr2 = vmcb02->save.cr2; vmcb12->save.cr4 = svm->vcpu.arch.cr4; vmcb12->save.rflags = kvm_get_rflags(vcpu); vmcb12->save.rip = kvm_rip_read(vcpu); vmcb12->save.rsp = kvm_rsp_read(vcpu); vmcb12->save.rax = kvm_rax_read(vcpu); - vmcb12->save.dr7 = vmcb->save.dr7; + vmcb12->save.dr7 = vmcb02->save.dr7; vmcb12->save.dr6 = svm->vcpu.arch.dr6; - vmcb12->save.cpl = vmcb->save.cpl; + vmcb12->save.cpl = vmcb02->save.cpl; - vmcb12->control.int_state = vmcb->control.int_state; - vmcb12->control.exit_code = vmcb->control.exit_code; - vmcb12->control.exit_code_hi = vmcb->control.exit_code_hi; - vmcb12->control.exit_info_1 = vmcb->control.exit_info_1; - vmcb12->control.exit_info_2 = vmcb->control.exit_info_2; + if (guest_cpu_cap_has(vcpu, X86_FEATURE_SHSTK)) { + vmcb12->save.s_cet = vmcb02->save.s_cet; + vmcb12->save.isst_addr = vmcb02->save.isst_addr; + vmcb12->save.ssp = vmcb02->save.ssp; + } + + vmcb12->control.int_state = vmcb02->control.int_state; + vmcb12->control.exit_code = vmcb02->control.exit_code; + vmcb12->control.exit_code_hi = vmcb02->control.exit_code_hi; + vmcb12->control.exit_info_1 = vmcb02->control.exit_info_1; + vmcb12->control.exit_info_2 = vmcb02->control.exit_info_2; if (vmcb12->control.exit_code != SVM_EXIT_ERR) nested_save_pending_event_to_vmcb12(svm, vmcb12); - if (svm->nrips_enabled) - vmcb12->control.next_rip = vmcb->control.next_rip; + if (guest_cpu_cap_has(vcpu, X86_FEATURE_NRIPS)) + vmcb12->control.next_rip = vmcb02->control.next_rip; vmcb12->control.int_ctl = svm->nested.ctl.int_ctl; - vmcb12->control.tlb_ctl = svm->nested.ctl.tlb_ctl; vmcb12->control.event_inj = svm->nested.ctl.event_inj; vmcb12->control.event_inj_err = svm->nested.ctl.event_inj_err; - vmcb12->control.pause_filter_count = - svm->vmcb->control.pause_filter_count; - vmcb12->control.pause_filter_thresh = - svm->vmcb->control.pause_filter_thresh; + if (!kvm_pause_in_guest(vcpu->kvm)) { + vmcb01->control.pause_filter_count = vmcb02->control.pause_filter_count; + vmcb_mark_dirty(vmcb01, VMCB_INTERCEPTS); + + } + + /* + * Invalidate bus_lock_rip unless KVM is still waiting for the guest + * to make forward progress before re-enabling bus lock detection. + */ + if (!vmcb02->control.bus_lock_counter) + svm->nested.ctl.bus_lock_rip = INVALID_GPA; nested_svm_copy_common_state(svm->nested.vmcb02.ptr, svm->vmcb01.ptr); + kvm_nested_vmexit_handle_ibrs(vcpu); + svm_switch_vmcb(svm, &svm->vmcb01); /* + * Rules for synchronizing int_ctl bits from vmcb02 to vmcb01: + * + * V_IRQ, V_IRQ_VECTOR, V_INTR_PRIO_MASK, V_IGN_TPR: If L1 doesn't + * intercept interrupts, then KVM will use vmcb02's V_IRQ (and related + * flags) to detect interrupt windows for L1 IRQs (even if L1 uses + * virtual interrupt masking). Raise KVM_REQ_EVENT to ensure that + * KVM re-requests an interrupt window if necessary, which implicitly + * copies this bits from vmcb02 to vmcb01. + * + * V_TPR: If L1 doesn't use virtual interrupt masking, then L1's vTPR + * is stored in vmcb02, but its value doesn't need to be copied from/to + * vmcb01 because it is copied from/to the virtual APIC's TPR register + * on each VM entry/exit. + * + * V_GIF: If nested vGIF is not used, KVM uses vmcb02's V_GIF for L1's + * V_GIF. However, GIF is architecturally clear on each VM exit, thus + * there is no need to copy V_GIF from vmcb02 to vmcb01. + */ + if (!nested_exit_on_intr(svm)) + kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); + + if (unlikely(guest_cpu_cap_has(vcpu, X86_FEATURE_LBRV) && + (svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK))) + svm_copy_lbrs(vmcb12, vmcb02); + else + svm_copy_lbrs(vmcb01, vmcb02); + + svm_update_lbrv(vcpu); + + if (vnmi) { + if (vmcb02->control.int_ctl & V_NMI_BLOCKING_MASK) + vmcb01->control.int_ctl |= V_NMI_BLOCKING_MASK; + else + vmcb01->control.int_ctl &= ~V_NMI_BLOCKING_MASK; + + if (vcpu->arch.nmi_pending) { + vcpu->arch.nmi_pending--; + vmcb01->control.int_ctl |= V_NMI_PENDING_MASK; + } else { + vmcb01->control.int_ctl &= ~V_NMI_PENDING_MASK; + } + } + + /* * On vmexit the GIF is set to false and * no event can be injected in L1. */ svm_set_gif(svm, false); - svm->vmcb->control.exit_int_info = 0; + vmcb01->control.exit_int_info = 0; svm->vcpu.arch.tsc_offset = svm->vcpu.arch.l1_tsc_offset; - if (svm->vmcb->control.tsc_offset != svm->vcpu.arch.tsc_offset) { - svm->vmcb->control.tsc_offset = svm->vcpu.arch.tsc_offset; - vmcb_mark_dirty(svm->vmcb, VMCB_INTERCEPTS); + if (vmcb01->control.tsc_offset != svm->vcpu.arch.tsc_offset) { + vmcb01->control.tsc_offset = svm->vcpu.arch.tsc_offset; + vmcb_mark_dirty(vmcb01, VMCB_INTERCEPTS); + } + + if (kvm_caps.has_tsc_control && + vcpu->arch.tsc_scaling_ratio != vcpu->arch.l1_tsc_scaling_ratio) { + vcpu->arch.tsc_scaling_ratio = vcpu->arch.l1_tsc_scaling_ratio; + svm_write_tsc_multiplier(vcpu); } svm->nested.ctl.nested_cr3 = 0; @@ -803,13 +1231,13 @@ int nested_svm_vmexit(struct vcpu_svm *svm) /* * Restore processor state that had been saved in vmcb01 */ - kvm_set_rflags(vcpu, svm->vmcb->save.rflags); - svm_set_efer(vcpu, svm->vmcb->save.efer); - svm_set_cr0(vcpu, svm->vmcb->save.cr0 | X86_CR0_PE); - svm_set_cr4(vcpu, svm->vmcb->save.cr4); - kvm_rax_write(vcpu, svm->vmcb->save.rax); - kvm_rsp_write(vcpu, svm->vmcb->save.rsp); - kvm_rip_write(vcpu, svm->vmcb->save.rip); + kvm_set_rflags(vcpu, vmcb01->save.rflags); + svm_set_efer(vcpu, vmcb01->save.efer); + svm_set_cr0(vcpu, vmcb01->save.cr0 | X86_CR0_PE); + svm_set_cr4(vcpu, vmcb01->save.cr4); + kvm_rax_write(vcpu, vmcb01->save.rax); + kvm_rsp_write(vcpu, vmcb01->save.rsp); + kvm_rip_write(vcpu, vmcb01->save.rip); svm->vcpu.arch.dr7 = DR7_FIXED_1; kvm_update_dr7(&svm->vcpu); @@ -821,13 +1249,13 @@ int nested_svm_vmexit(struct vcpu_svm *svm) vmcb12->control.exit_int_info_err, KVM_ISA_SVM); - kvm_vcpu_unmap(vcpu, &map, true); + kvm_vcpu_unmap(vcpu, &map); nested_svm_transition_tlb_flush(vcpu); nested_svm_uninit_mmu_context(vcpu); - rc = nested_svm_load_cr3(vcpu, svm->vmcb->save.cr3, false, true); + rc = nested_svm_load_cr3(vcpu, vmcb01->save.cr3, false, true); if (rc) return 1; @@ -845,14 +1273,27 @@ int nested_svm_vmexit(struct vcpu_svm *svm) * right now so that it an be accounted for before we execute * L1's next instruction. */ - if (unlikely(svm->vmcb->save.rflags & X86_EFLAGS_TF)) + if (unlikely(vmcb01->save.rflags & X86_EFLAGS_TF)) kvm_queue_exception(&(svm->vcpu), DB_VECTOR); + /* + * Un-inhibit the AVIC right away, so that other vCPUs can start + * to benefit from it right away. + */ + if (kvm_apicv_activated(vcpu->kvm)) + __kvm_vcpu_update_apicv(vcpu); + return 0; } static void nested_svm_triple_fault(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); + + if (!vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_SHUTDOWN)) + return; + + kvm_clear_request(KVM_REQ_TRIPLE_FAULT, vcpu); nested_svm_simple_vmexit(to_svm(vcpu), SVM_EXIT_SHUTDOWN); } @@ -863,7 +1304,7 @@ int svm_allocate_nested(struct vcpu_svm *svm) if (svm->nested.initialized) return 0; - vmcb02_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); + vmcb02_page = snp_safe_alloc_page(); if (!vmcb02_page) return -ENOMEM; svm->nested.vmcb02.ptr = page_address(vmcb02_page); @@ -872,7 +1313,6 @@ int svm_allocate_nested(struct vcpu_svm *svm) svm->nested.msrpm = svm_vcpu_alloc_msrpm(); if (!svm->nested.msrpm) goto err_free_vmcb02; - svm_vcpu_init_msrpm(&svm->vcpu, svm->nested.msrpm); svm->nested.initialized = true; return 0; @@ -887,6 +1327,9 @@ void svm_free_nested(struct vcpu_svm *svm) if (!svm->nested.initialized) return; + if (WARN_ON_ONCE(svm->vmcb != svm->vmcb01.ptr)) + svm_switch_vmcb(svm, &svm->vmcb01); + svm_vcpu_free_msrpm(svm->nested.msrpm); svm->nested.msrpm = NULL; @@ -905,12 +1348,9 @@ void svm_free_nested(struct vcpu_svm *svm) svm->nested.initialized = false; } -/* - * Forcibly leave nested mode in order to be able to reset the VCPU later on. - */ -void svm_leave_nested(struct vcpu_svm *svm) +void svm_leave_nested(struct kvm_vcpu *vcpu) { - struct kvm_vcpu *vcpu = &svm->vcpu; + struct vcpu_svm *svm = to_svm(vcpu); if (is_guest_mode(vcpu)) { svm->nested.nested_run_pending = 0; @@ -922,6 +1362,9 @@ void svm_leave_nested(struct vcpu_svm *svm) nested_svm_uninit_mmu_context(vcpu); vmcb_mark_all_dirty(svm->vmcb); + + if (kvm_apicv_activated(vcpu->kvm)) + kvm_make_request(KVM_REQ_APICV_UPDATE, vcpu); } kvm_clear_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu); @@ -929,26 +1372,26 @@ void svm_leave_nested(struct vcpu_svm *svm) static int nested_svm_exit_handled_msr(struct vcpu_svm *svm) { - u32 offset, msr, value; - int write, mask; + gpa_t base = svm->nested.ctl.msrpm_base_pa; + int write, bit_nr; + u8 value, mask; + u32 msr; - if (!(vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_MSR_PROT))) + if (!(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_MSR_PROT))) return NESTED_EXIT_HOST; msr = svm->vcpu.arch.regs[VCPU_REGS_RCX]; - offset = svm_msrpm_offset(msr); + bit_nr = svm_msrpm_bit_nr(msr); write = svm->vmcb->control.exit_info_1 & 1; - mask = 1 << ((2 * (msr & 0xf)) + write); - if (offset == MSR_INVALID) + if (bit_nr < 0) return NESTED_EXIT_DONE; - /* Offset is in 32 bit units but need in 8 bit units */ - offset *= 4; - - if (kvm_vcpu_read_guest(&svm->vcpu, svm->nested.ctl.msrpm_base_pa + offset, &value, 4)) + if (kvm_vcpu_read_guest(&svm->vcpu, base + bit_nr / BITS_PER_BYTE, + &value, sizeof(value))) return NESTED_EXIT_DONE; + mask = BIT(write) << (bit_nr & (BITS_PER_BYTE - 1)); return (value & mask) ? NESTED_EXIT_DONE : NESTED_EXIT_HOST; } @@ -959,7 +1402,7 @@ static int nested_svm_intercept_ioio(struct vcpu_svm *svm) u8 start_bit; u64 gpa; - if (!(vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_IOIO_PROT))) + if (!(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_IOIO_PROT))) return NESTED_EXIT_HOST; port = svm->vmcb->control.exit_info_1 >> 16; @@ -989,16 +1432,6 @@ static int nested_svm_intercept(struct vcpu_svm *svm) case SVM_EXIT_IOIO: vmexit = nested_svm_intercept_ioio(svm); break; - case SVM_EXIT_READ_CR0 ... SVM_EXIT_WRITE_CR8: { - if (vmcb_is_intercept(&svm->nested.ctl, exit_code)) - vmexit = NESTED_EXIT_DONE; - break; - } - case SVM_EXIT_READ_DR0 ... SVM_EXIT_WRITE_DR7: { - if (vmcb_is_intercept(&svm->nested.ctl, exit_code)) - vmexit = NESTED_EXIT_DONE; - break; - } case SVM_EXIT_EXCP_BASE ... SVM_EXIT_EXCP_BASE + 0x1f: { /* * Host-intercepted exceptions have been checked already in @@ -1013,7 +1446,7 @@ static int nested_svm_intercept(struct vcpu_svm *svm) break; } default: { - if (vmcb_is_intercept(&svm->nested.ctl, exit_code)) + if (vmcb12_is_intercept(&svm->nested.ctl, exit_code)) vmexit = NESTED_EXIT_DONE; } } @@ -1048,58 +1481,73 @@ int nested_svm_check_permissions(struct kvm_vcpu *vcpu) return 0; } -static bool nested_exit_on_exception(struct vcpu_svm *svm) +static bool nested_svm_is_exception_vmexit(struct kvm_vcpu *vcpu, u8 vector, + u32 error_code) { - unsigned int nr = svm->vcpu.arch.exception.nr; + struct vcpu_svm *svm = to_svm(vcpu); - return (svm->nested.ctl.intercepts[INTERCEPT_EXCEPTION] & BIT(nr)); + return (svm->nested.ctl.intercepts[INTERCEPT_EXCEPTION] & BIT(vector)); } -static void nested_svm_inject_exception_vmexit(struct vcpu_svm *svm) +static void nested_svm_inject_exception_vmexit(struct kvm_vcpu *vcpu) { - unsigned int nr = svm->vcpu.arch.exception.nr; + struct kvm_queued_exception *ex = &vcpu->arch.exception_vmexit; + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb *vmcb = svm->vmcb; - svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + nr; - svm->vmcb->control.exit_code_hi = 0; + vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + ex->vector; + vmcb->control.exit_code_hi = 0; - if (svm->vcpu.arch.exception.has_error_code) - svm->vmcb->control.exit_info_1 = svm->vcpu.arch.exception.error_code; + if (ex->has_error_code) + vmcb->control.exit_info_1 = ex->error_code; /* * EXITINFO2 is undefined for all exception intercepts other * than #PF. */ - if (nr == PF_VECTOR) { - if (svm->vcpu.arch.exception.nested_apf) - svm->vmcb->control.exit_info_2 = svm->vcpu.arch.apf.nested_apf_token; - else if (svm->vcpu.arch.exception.has_payload) - svm->vmcb->control.exit_info_2 = svm->vcpu.arch.exception.payload; + if (ex->vector == PF_VECTOR) { + if (ex->has_payload) + vmcb->control.exit_info_2 = ex->payload; else - svm->vmcb->control.exit_info_2 = svm->vcpu.arch.cr2; - } else if (nr == DB_VECTOR) { - /* See inject_pending_event. */ - kvm_deliver_exception_payload(&svm->vcpu); - if (svm->vcpu.arch.dr7 & DR7_GD) { - svm->vcpu.arch.dr7 &= ~DR7_GD; - kvm_update_dr7(&svm->vcpu); + vmcb->control.exit_info_2 = vcpu->arch.cr2; + } else if (ex->vector == DB_VECTOR) { + /* See kvm_check_and_inject_events(). */ + kvm_deliver_exception_payload(vcpu, ex); + + if (vcpu->arch.dr7 & DR7_GD) { + vcpu->arch.dr7 &= ~DR7_GD; + kvm_update_dr7(vcpu); } - } else - WARN_ON(svm->vcpu.arch.exception.has_payload); + } else { + WARN_ON(ex->has_payload); + } nested_svm_vmexit(svm); } static inline bool nested_exit_on_init(struct vcpu_svm *svm) { - return vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_INIT); + return vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_INIT); } static int svm_check_nested_events(struct kvm_vcpu *vcpu) { - struct vcpu_svm *svm = to_svm(vcpu); - bool block_nested_events = - kvm_event_needs_reinjection(vcpu) || svm->nested.nested_run_pending; struct kvm_lapic *apic = vcpu->arch.apic; + struct vcpu_svm *svm = to_svm(vcpu); + /* + * Only a pending nested run blocks a pending exception. If there is a + * previously injected event, the pending exception occurred while said + * event was being delivered and thus needs to be handled. + */ + bool block_nested_exceptions = svm->nested.nested_run_pending; + /* + * New events (not exceptions) are only recognized at instruction + * boundaries. If an event needs reinjection, then KVM is handling a + * VM-Exit that occurred _during_ instruction execution; new events are + * blocked until the instruction completes. + */ + bool block_nested_events = block_nested_exceptions || + kvm_event_needs_reinjection(vcpu); if (lapic_in_kernel(vcpu) && test_bit(KVM_APIC_INIT, &apic->pending_events)) { @@ -1111,21 +1559,20 @@ static int svm_check_nested_events(struct kvm_vcpu *vcpu) return 0; } - if (vcpu->arch.exception.pending) { - /* - * Only a pending nested run can block a pending exception. - * Otherwise an injected NMI/interrupt should either be - * lost or delivered to the nested hypervisor in the EXITINTINFO - * vmcb field, while delivering the pending exception. - */ - if (svm->nested.nested_run_pending) + if (vcpu->arch.exception_vmexit.pending) { + if (block_nested_exceptions) return -EBUSY; - if (!nested_exit_on_exception(svm)) - return 0; - nested_svm_inject_exception_vmexit(svm); + nested_svm_inject_exception_vmexit(vcpu); return 0; } + if (vcpu->arch.exception.pending) { + if (block_nested_exceptions) + return -EBUSY; + return 0; + } + +#ifdef CONFIG_KVM_SMM if (vcpu->arch.smi_pending && !svm_smi_blocked(vcpu)) { if (block_nested_events) return -EBUSY; @@ -1134,6 +1581,7 @@ static int svm_check_nested_events(struct kvm_vcpu *vcpu) nested_svm_simple_vmexit(svm, SVM_EXIT_SMI); return 0; } +#endif if (vcpu->arch.nmi_pending && !svm_nmi_blocked(vcpu)) { if (block_nested_events) @@ -1160,6 +1608,7 @@ static int svm_check_nested_events(struct kvm_vcpu *vcpu) int nested_svm_exit_special(struct vcpu_svm *svm) { u32 exit_code = svm->vmcb->control.exit_code; + struct kvm_vcpu *vcpu = &svm->vcpu; switch (exit_code) { case SVM_EXIT_INTR: @@ -1178,6 +1627,13 @@ int nested_svm_exit_special(struct vcpu_svm *svm) return NESTED_EXIT_HOST; break; } + case SVM_EXIT_VMMCALL: + /* Hyper-V L2 TLB flush hypercall is handled by L0 */ + if (guest_hv_cpuid_has_l2_tlb_flush(vcpu) && + nested_svm_l2_tlb_flush_enabled(vcpu) && + kvm_hv_is_tlb_flush_hcall(vcpu)) + return NESTED_EXIT_HOST; + break; default: break; } @@ -1185,11 +1641,59 @@ int nested_svm_exit_special(struct vcpu_svm *svm) return NESTED_EXIT_CONTINUE; } +void nested_svm_update_tsc_ratio_msr(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + vcpu->arch.tsc_scaling_ratio = + kvm_calc_nested_tsc_multiplier(vcpu->arch.l1_tsc_scaling_ratio, + svm->tsc_ratio_msr); + svm_write_tsc_multiplier(vcpu); +} + +/* Inverse operation of nested_copy_vmcb_control_to_cache(). asid is copied too. */ +static void nested_copy_vmcb_cache_to_control(struct vmcb_control_area *dst, + struct vmcb_ctrl_area_cached *from) +{ + unsigned int i; + + memset(dst, 0, sizeof(struct vmcb_control_area)); + + for (i = 0; i < MAX_INTERCEPT; i++) + dst->intercepts[i] = from->intercepts[i]; + + dst->iopm_base_pa = from->iopm_base_pa; + dst->msrpm_base_pa = from->msrpm_base_pa; + dst->tsc_offset = from->tsc_offset; + dst->asid = from->asid; + dst->tlb_ctl = from->tlb_ctl; + dst->int_ctl = from->int_ctl; + dst->int_vector = from->int_vector; + dst->int_state = from->int_state; + dst->exit_code = from->exit_code; + dst->exit_code_hi = from->exit_code_hi; + dst->exit_info_1 = from->exit_info_1; + dst->exit_info_2 = from->exit_info_2; + dst->exit_int_info = from->exit_int_info; + dst->exit_int_info_err = from->exit_int_info_err; + dst->nested_ctl = from->nested_ctl; + dst->event_inj = from->event_inj; + dst->event_inj_err = from->event_inj_err; + dst->next_rip = from->next_rip; + dst->nested_cr3 = from->nested_cr3; + dst->virt_ext = from->virt_ext; + dst->pause_filter_count = from->pause_filter_count; + dst->pause_filter_thresh = from->pause_filter_thresh; + /* 'clean' and 'hv_enlightenments' are not changed by KVM */ +} + static int svm_get_nested_state(struct kvm_vcpu *vcpu, struct kvm_nested_state __user *user_kvm_nested_state, u32 user_data_size) { struct vcpu_svm *svm; + struct vmcb_control_area *ctl; + unsigned long r; struct kvm_nested_state kvm_state = { .flags = 0, .format = KVM_STATE_NESTED_FORMAT_SVM, @@ -1231,9 +1735,18 @@ static int svm_get_nested_state(struct kvm_vcpu *vcpu, */ if (clear_user(user_vmcb, KVM_STATE_NESTED_SVM_VMCB_SIZE)) return -EFAULT; - if (copy_to_user(&user_vmcb->control, &svm->nested.ctl, - sizeof(user_vmcb->control))) + + ctl = kzalloc(sizeof(*ctl), GFP_KERNEL); + if (!ctl) + return -ENOMEM; + + nested_copy_vmcb_cache_to_control(ctl, &svm->nested.ctl); + r = copy_to_user(&user_vmcb->control, ctl, + sizeof(user_vmcb->control)); + kfree(ctl); + if (r) return -EFAULT; + if (copy_to_user(&user_vmcb->save, &svm->vmcb01.ptr->save, sizeof(user_vmcb->save))) return -EFAULT; @@ -1250,6 +1763,8 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, &user_kvm_nested_state->data.svm[0]; struct vmcb_control_area *ctl; struct vmcb_save_area *save; + struct vmcb_save_area_cached save_cached; + struct vmcb_ctrl_area_cached ctl_cached; unsigned long cr0; int ret; @@ -1279,7 +1794,7 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, return -EINVAL; if (!(kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE)) { - svm_leave_nested(svm); + svm_leave_nested(vcpu); svm_set_gif(svm, !!(kvm_state->flags & KVM_STATE_NESTED_GIF_SET)); return 0; } @@ -1289,20 +1804,19 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, if (kvm_state->size < sizeof(*kvm_state) + KVM_STATE_NESTED_SVM_VMCB_SIZE) return -EINVAL; - ret = -ENOMEM; - ctl = kzalloc(sizeof(*ctl), GFP_KERNEL_ACCOUNT); - save = kzalloc(sizeof(*save), GFP_KERNEL_ACCOUNT); - if (!ctl || !save) - goto out_free; + ctl = memdup_user(&user_vmcb->control, sizeof(*ctl)); + if (IS_ERR(ctl)) + return PTR_ERR(ctl); - ret = -EFAULT; - if (copy_from_user(ctl, &user_vmcb->control, sizeof(*ctl))) - goto out_free; - if (copy_from_user(save, &user_vmcb->save, sizeof(*save))) - goto out_free; + save = memdup_user(&user_vmcb->save, sizeof(*save)); + if (IS_ERR(save)) { + kfree(ctl); + return PTR_ERR(save); + } ret = -EINVAL; - if (!nested_vmcb_check_controls(vcpu, ctl)) + __nested_copy_vmcb_control_to_cache(vcpu, &ctl_cached, ctl); + if (!__nested_vmcb_check_controls(vcpu, &ctl_cached)) goto out_free; /* @@ -1317,22 +1831,11 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, * Validate host state saved from before VMRUN (see * nested_svm_check_permissions). */ + __nested_copy_vmcb_save_to_cache(&save_cached, save); if (!(save->cr0 & X86_CR0_PG) || !(save->cr0 & X86_CR0_PE) || (save->rflags & X86_EFLAGS_VM) || - !nested_vmcb_valid_sregs(vcpu, save)) - goto out_free; - - /* - * While the nested guest CR3 is already checked and set by - * KVM_SET_SREGS, it was set when nested state was yet loaded, - * thus MMU might not be initialized correctly. - * Set it again to fix this. - */ - - ret = nested_svm_load_cr3(&svm->vcpu, vcpu->arch.cr3, - nested_npt_enabled(svm), false); - if (WARN_ON_ONCE(ret)) + !__nested_vmcb_check_save(vcpu, &save_cached)) goto out_free; @@ -1344,7 +1847,7 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, */ if (is_guest_mode(vcpu)) - svm_leave_nested(svm); + svm_leave_nested(vcpu); else svm->nested.vmcb02.ptr->save = svm->vmcb01.ptr->save; @@ -1355,27 +1858,25 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, svm->nested.vmcb12_gpa = kvm_state->hdr.svm.vmcb_pa; - svm->vmcb01.ptr->save.es = save->es; - svm->vmcb01.ptr->save.cs = save->cs; - svm->vmcb01.ptr->save.ss = save->ss; - svm->vmcb01.ptr->save.ds = save->ds; - svm->vmcb01.ptr->save.gdtr = save->gdtr; - svm->vmcb01.ptr->save.idtr = save->idtr; - svm->vmcb01.ptr->save.rflags = save->rflags | X86_EFLAGS_FIXED; - svm->vmcb01.ptr->save.efer = save->efer; - svm->vmcb01.ptr->save.cr0 = save->cr0; - svm->vmcb01.ptr->save.cr3 = save->cr3; - svm->vmcb01.ptr->save.cr4 = save->cr4; - svm->vmcb01.ptr->save.rax = save->rax; - svm->vmcb01.ptr->save.rsp = save->rsp; - svm->vmcb01.ptr->save.rip = save->rip; - svm->vmcb01.ptr->save.cpl = 0; - - nested_load_control_from_vmcb12(svm, ctl); + svm_copy_vmrun_state(&svm->vmcb01.ptr->save, save); + nested_copy_vmcb_control_to_cache(svm, ctl); svm_switch_vmcb(svm, &svm->nested.vmcb02); + nested_vmcb02_prepare_control(svm, svm->vmcb->save.rip, svm->vmcb->save.cs.base); + + /* + * While the nested guest CR3 is already checked and set by + * KVM_SET_SREGS, it was set when nested state was yet loaded, + * thus MMU might not be initialized correctly. + * Set it again to fix this. + */ - nested_vmcb02_prepare_control(svm); + ret = nested_svm_load_cr3(&svm->vcpu, vcpu->arch.cr3, + nested_npt_enabled(svm), false); + if (WARN_ON_ONCE(ret)) + goto out_free; + + svm->nested.force_msr_bitmap_recalc = true; kvm_make_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu); ret = 0; @@ -1388,22 +1889,20 @@ out_free: static bool svm_get_nested_state_pages(struct kvm_vcpu *vcpu) { - struct vcpu_svm *svm = to_svm(vcpu); - if (WARN_ON(!is_guest_mode(vcpu))) return true; if (!vcpu->arch.pdptrs_from_userspace && - !nested_npt_enabled(svm) && is_pae_paging(vcpu)) + !nested_npt_enabled(to_svm(vcpu)) && is_pae_paging(vcpu)) /* * Reload the guest's PDPTRs since after a migration * the guest CR3 might be restored prior to setting the nested * state which can lead to a load of wrong PDPTRs. */ - if (CC(!load_pdptrs(vcpu, vcpu->arch.walk_mmu, vcpu->arch.cr3))) + if (CC(!load_pdptrs(vcpu, vcpu->arch.cr3))) return false; - if (!nested_svm_vmrun_msrpm(svm)) { + if (!nested_svm_merge_msrpm(vcpu)) { vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION; @@ -1411,13 +1910,19 @@ static bool svm_get_nested_state_pages(struct kvm_vcpu *vcpu) return false; } + if (kvm_hv_verify_vp_assist(vcpu)) + return false; + return true; } struct kvm_x86_nested_ops svm_nested_ops = { + .leave_nested = svm_leave_nested, + .is_exception_vmexit = nested_svm_is_exception_vmexit, .check_events = svm_check_nested_events, .triple_fault = nested_svm_triple_fault, .get_nested_state_pages = svm_get_nested_state_pages, .get_state = svm_get_nested_state, .set_state = svm_set_nested_state, + .hv_inject_synthetic_vmexit_post_tlb_flush = svm_hv_inject_synthetic_vmexit_post_tlb_flush, }; diff --git a/arch/x86/kvm/svm/pmu.c b/arch/x86/kvm/svm/pmu.c index fdf587f19c5f..bc062285fbf5 100644 --- a/arch/x86/kvm/svm/pmu.c +++ b/arch/x86/kvm/svm/pmu.c @@ -9,6 +9,8 @@ * * Implementation is based on pmu_intel.c file */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/types.h> #include <linux/kvm_host.h> #include <linux/perf_event.h> @@ -16,200 +18,76 @@ #include "cpuid.h" #include "lapic.h" #include "pmu.h" +#include "svm.h" enum pmu_type { PMU_TYPE_COUNTER = 0, PMU_TYPE_EVNTSEL, }; -enum index { - INDEX_ZERO = 0, - INDEX_ONE, - INDEX_TWO, - INDEX_THREE, - INDEX_FOUR, - INDEX_FIVE, - INDEX_ERROR, -}; - -/* duplicated from amd_perfmon_event_map, K7 and above should work. */ -static struct kvm_event_hw_type_mapping amd_event_mapping[] = { - [0] = { 0x76, 0x00, PERF_COUNT_HW_CPU_CYCLES }, - [1] = { 0xc0, 0x00, PERF_COUNT_HW_INSTRUCTIONS }, - [2] = { 0x7d, 0x07, PERF_COUNT_HW_CACHE_REFERENCES }, - [3] = { 0x7e, 0x07, PERF_COUNT_HW_CACHE_MISSES }, - [4] = { 0xc2, 0x00, PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, - [5] = { 0xc3, 0x00, PERF_COUNT_HW_BRANCH_MISSES }, - [6] = { 0xd0, 0x00, PERF_COUNT_HW_STALLED_CYCLES_FRONTEND }, - [7] = { 0xd1, 0x00, PERF_COUNT_HW_STALLED_CYCLES_BACKEND }, -}; - -static unsigned int get_msr_base(struct kvm_pmu *pmu, enum pmu_type type) +static struct kvm_pmc *amd_pmu_get_pmc(struct kvm_pmu *pmu, int pmc_idx) { - struct kvm_vcpu *vcpu = pmu_to_vcpu(pmu); + unsigned int num_counters = pmu->nr_arch_gp_counters; - if (guest_cpuid_has(vcpu, X86_FEATURE_PERFCTR_CORE)) { - if (type == PMU_TYPE_COUNTER) - return MSR_F15H_PERF_CTR; - else - return MSR_F15H_PERF_CTL; - } else { - if (type == PMU_TYPE_COUNTER) - return MSR_K7_PERFCTR0; - else - return MSR_K7_EVNTSEL0; - } -} + if (pmc_idx >= num_counters) + return NULL; -static enum index msr_to_index(u32 msr) -{ - switch (msr) { - case MSR_F15H_PERF_CTL0: - case MSR_F15H_PERF_CTR0: - case MSR_K7_EVNTSEL0: - case MSR_K7_PERFCTR0: - return INDEX_ZERO; - case MSR_F15H_PERF_CTL1: - case MSR_F15H_PERF_CTR1: - case MSR_K7_EVNTSEL1: - case MSR_K7_PERFCTR1: - return INDEX_ONE; - case MSR_F15H_PERF_CTL2: - case MSR_F15H_PERF_CTR2: - case MSR_K7_EVNTSEL2: - case MSR_K7_PERFCTR2: - return INDEX_TWO; - case MSR_F15H_PERF_CTL3: - case MSR_F15H_PERF_CTR3: - case MSR_K7_EVNTSEL3: - case MSR_K7_PERFCTR3: - return INDEX_THREE; - case MSR_F15H_PERF_CTL4: - case MSR_F15H_PERF_CTR4: - return INDEX_FOUR; - case MSR_F15H_PERF_CTL5: - case MSR_F15H_PERF_CTR5: - return INDEX_FIVE; - default: - return INDEX_ERROR; - } + return &pmu->gp_counters[array_index_nospec(pmc_idx, num_counters)]; } static inline struct kvm_pmc *get_gp_pmc_amd(struct kvm_pmu *pmu, u32 msr, enum pmu_type type) { struct kvm_vcpu *vcpu = pmu_to_vcpu(pmu); + unsigned int idx; + + if (!pmu->version) + return NULL; switch (msr) { - case MSR_F15H_PERF_CTL0: - case MSR_F15H_PERF_CTL1: - case MSR_F15H_PERF_CTL2: - case MSR_F15H_PERF_CTL3: - case MSR_F15H_PERF_CTL4: - case MSR_F15H_PERF_CTL5: - if (!guest_cpuid_has(vcpu, X86_FEATURE_PERFCTR_CORE)) + case MSR_F15H_PERF_CTL0 ... MSR_F15H_PERF_CTR5: + if (!guest_cpu_cap_has(vcpu, X86_FEATURE_PERFCTR_CORE)) return NULL; - fallthrough; + /* + * Each PMU counter has a pair of CTL and CTR MSRs. CTLn + * MSRs (accessed via EVNTSEL) are even, CTRn MSRs are odd. + */ + idx = (unsigned int)((msr - MSR_F15H_PERF_CTL0) / 2); + if (!(msr & 0x1) != (type == PMU_TYPE_EVNTSEL)) + return NULL; + break; case MSR_K7_EVNTSEL0 ... MSR_K7_EVNTSEL3: if (type != PMU_TYPE_EVNTSEL) return NULL; + idx = msr - MSR_K7_EVNTSEL0; break; - case MSR_F15H_PERF_CTR0: - case MSR_F15H_PERF_CTR1: - case MSR_F15H_PERF_CTR2: - case MSR_F15H_PERF_CTR3: - case MSR_F15H_PERF_CTR4: - case MSR_F15H_PERF_CTR5: - if (!guest_cpuid_has(vcpu, X86_FEATURE_PERFCTR_CORE)) - return NULL; - fallthrough; case MSR_K7_PERFCTR0 ... MSR_K7_PERFCTR3: if (type != PMU_TYPE_COUNTER) return NULL; + idx = msr - MSR_K7_PERFCTR0; break; default: return NULL; } - return &pmu->gp_counters[msr_to_index(msr)]; + return amd_pmu_get_pmc(pmu, idx); } -static unsigned amd_find_arch_event(struct kvm_pmu *pmu, - u8 event_select, - u8 unit_mask) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(amd_event_mapping); i++) - if (amd_event_mapping[i].eventsel == event_select - && amd_event_mapping[i].unit_mask == unit_mask) - break; - - if (i == ARRAY_SIZE(amd_event_mapping)) - return PERF_COUNT_HW_MAX; - - return amd_event_mapping[i].event_type; -} - -/* return PERF_COUNT_HW_MAX as AMD doesn't have fixed events */ -static unsigned amd_find_fixed_event(int idx) -{ - return PERF_COUNT_HW_MAX; -} - -/* check if a PMC is enabled by comparing it against global_ctrl bits. Because - * AMD CPU doesn't have global_ctrl MSR, all PMCs are enabled (return TRUE). - */ -static bool amd_pmc_is_enabled(struct kvm_pmc *pmc) -{ - return true; -} - -static struct kvm_pmc *amd_pmc_idx_to_pmc(struct kvm_pmu *pmu, int pmc_idx) -{ - unsigned int base = get_msr_base(pmu, PMU_TYPE_COUNTER); - struct kvm_vcpu *vcpu = pmu_to_vcpu(pmu); - - if (guest_cpuid_has(vcpu, X86_FEATURE_PERFCTR_CORE)) { - /* - * The idx is contiguous. The MSRs are not. The counter MSRs - * are interleaved with the event select MSRs. - */ - pmc_idx *= 2; - } - - return get_gp_pmc_amd(pmu, base + pmc_idx, PMU_TYPE_COUNTER); -} - -/* returns 0 if idx's corresponding MSR exists; otherwise returns 1. */ -static int amd_is_valid_rdpmc_ecx(struct kvm_vcpu *vcpu, unsigned int idx) +static int amd_check_rdpmc_early(struct kvm_vcpu *vcpu, unsigned int idx) { struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); - idx &= ~(3u << 30); + if (idx >= pmu->nr_arch_gp_counters) + return -EINVAL; - return (idx >= pmu->nr_arch_gp_counters); + return 0; } /* idx is the ECX register of RDPMC instruction */ static struct kvm_pmc *amd_rdpmc_ecx_to_pmc(struct kvm_vcpu *vcpu, unsigned int idx, u64 *mask) { - struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); - struct kvm_pmc *counters; - - idx &= ~(3u << 30); - if (idx >= pmu->nr_arch_gp_counters) - return NULL; - counters = pmu->gp_counters; - - return &counters[idx]; -} - -static bool amd_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr) -{ - /* All MSRs refer to exactly one PMC, so msr_idx_to_pmc is enough. */ - return false; + return amd_pmu_get_pmc(vcpu_to_pmu(vcpu), idx); } static struct kvm_pmc *amd_msr_idx_to_pmc(struct kvm_vcpu *vcpu, u32 msr) @@ -223,6 +101,30 @@ static struct kvm_pmc *amd_msr_idx_to_pmc(struct kvm_vcpu *vcpu, u32 msr) return pmc; } +static bool amd_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr) +{ + struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); + + switch (msr) { + case MSR_K7_EVNTSEL0 ... MSR_K7_PERFCTR3: + return pmu->version > 0; + case MSR_F15H_PERF_CTL0 ... MSR_F15H_PERF_CTR5: + return guest_cpu_cap_has(vcpu, X86_FEATURE_PERFCTR_CORE); + case MSR_AMD64_PERF_CNTR_GLOBAL_STATUS: + case MSR_AMD64_PERF_CNTR_GLOBAL_CTL: + case MSR_AMD64_PERF_CNTR_GLOBAL_STATUS_CLR: + case MSR_AMD64_PERF_CNTR_GLOBAL_STATUS_SET: + return pmu->version > 1; + default: + if (msr > MSR_F15H_PERF_CTR5 && + msr < MSR_F15H_PERF_CTL0 + 2 * pmu->nr_arch_gp_counters) + return pmu->version > 1; + break; + } + + return amd_msr_idx_to_pmc(vcpu, msr); +} + static int amd_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) { struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); @@ -255,18 +157,18 @@ static int amd_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) /* MSR_PERFCTRn */ pmc = get_gp_pmc_amd(pmu, msr, PMU_TYPE_COUNTER); if (pmc) { - pmc->counter += data - pmc_read_counter(pmc); + pmc_write_counter(pmc, data); return 0; } /* MSR_EVNTSELn */ pmc = get_gp_pmc_amd(pmu, msr, PMU_TYPE_EVNTSEL); if (pmc) { - if (data == pmc->eventsel) - return 0; - if (!(data & pmu->reserved_bits)) { - reprogram_gp_counter(pmc, data); - return 0; + data &= ~pmu->reserved_bits; + if (data != pmc->eventsel) { + pmc->eventsel = data; + kvm_pmu_request_counter_reprogram(pmc); } + return 0; } return 1; @@ -275,20 +177,39 @@ static int amd_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) static void amd_pmu_refresh(struct kvm_vcpu *vcpu) { struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); + union cpuid_0x80000022_ebx ebx; - if (guest_cpuid_has(vcpu, X86_FEATURE_PERFCTR_CORE)) + pmu->version = 1; + if (guest_cpu_cap_has(vcpu, X86_FEATURE_PERFMON_V2)) { + pmu->version = 2; + /* + * Note, PERFMON_V2 is also in 0x80000022.0x0, i.e. the guest + * CPUID entry is guaranteed to be non-NULL. + */ + BUILD_BUG_ON(x86_feature_cpuid(X86_FEATURE_PERFMON_V2).function != 0x80000022 || + x86_feature_cpuid(X86_FEATURE_PERFMON_V2).index); + ebx.full = kvm_find_cpuid_entry_index(vcpu, 0x80000022, 0)->ebx; + pmu->nr_arch_gp_counters = ebx.split.num_core_pmc; + } else if (guest_cpu_cap_has(vcpu, X86_FEATURE_PERFCTR_CORE)) { pmu->nr_arch_gp_counters = AMD64_NUM_COUNTERS_CORE; - else + } else { pmu->nr_arch_gp_counters = AMD64_NUM_COUNTERS; + } - pmu->counter_bitmask[KVM_PMC_GP] = ((u64)1 << 48) - 1; - pmu->reserved_bits = 0xffffffff00200000ull; - pmu->version = 1; + pmu->nr_arch_gp_counters = min_t(unsigned int, pmu->nr_arch_gp_counters, + kvm_pmu_cap.num_counters_gp); + + if (pmu->version > 1) { + pmu->global_ctrl_rsvd = ~(BIT_ULL(pmu->nr_arch_gp_counters) - 1); + pmu->global_status_rsvd = pmu->global_ctrl_rsvd; + } + + pmu->counter_bitmask[KVM_PMC_GP] = BIT_ULL(48) - 1; + pmu->reserved_bits = 0xfffffff000280000ull; + pmu->raw_event_mask = AMD64_RAW_EVENT_MASK; /* not applicable to AMD; but clean them to prevent any fall out */ pmu->counter_bitmask[KVM_PMC_FIXED] = 0; pmu->nr_arch_fixed_counters = 0; - pmu->global_status = 0; - bitmap_set(pmu->all_valid_pmc_idx, 0, pmu->nr_arch_gp_counters); } static void amd_pmu_init(struct kvm_vcpu *vcpu) @@ -296,9 +217,9 @@ static void amd_pmu_init(struct kvm_vcpu *vcpu) struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); int i; - BUILD_BUG_ON(AMD64_NUM_COUNTERS_CORE > INTEL_PMC_MAX_GENERIC); + BUILD_BUG_ON(KVM_MAX_NR_AMD_GP_COUNTERS > AMD64_NUM_COUNTERS_CORE); - for (i = 0; i < AMD64_NUM_COUNTERS_CORE ; i++) { + for (i = 0; i < KVM_MAX_NR_AMD_GP_COUNTERS; i++) { pmu->gp_counters[i].type = KVM_PMC_GP; pmu->gp_counters[i].vcpu = vcpu; pmu->gp_counters[i].idx = i; @@ -306,31 +227,16 @@ static void amd_pmu_init(struct kvm_vcpu *vcpu) } } -static void amd_pmu_reset(struct kvm_vcpu *vcpu) -{ - struct kvm_pmu *pmu = vcpu_to_pmu(vcpu); - int i; - - for (i = 0; i < AMD64_NUM_COUNTERS_CORE; i++) { - struct kvm_pmc *pmc = &pmu->gp_counters[i]; - - pmc_stop_counter(pmc); - pmc->counter = pmc->eventsel = 0; - } -} - -struct kvm_pmu_ops amd_pmu_ops = { - .find_arch_event = amd_find_arch_event, - .find_fixed_event = amd_find_fixed_event, - .pmc_is_enabled = amd_pmc_is_enabled, - .pmc_idx_to_pmc = amd_pmc_idx_to_pmc, +struct kvm_pmu_ops amd_pmu_ops __initdata = { .rdpmc_ecx_to_pmc = amd_rdpmc_ecx_to_pmc, .msr_idx_to_pmc = amd_msr_idx_to_pmc, - .is_valid_rdpmc_ecx = amd_is_valid_rdpmc_ecx, + .check_rdpmc_early = amd_check_rdpmc_early, .is_valid_msr = amd_is_valid_msr, .get_msr = amd_pmu_get_msr, .set_msr = amd_pmu_set_msr, .refresh = amd_pmu_refresh, .init = amd_pmu_init, - .reset = amd_pmu_reset, + .EVENTSEL_EVENT = AMD64_EVENTSEL_EVENT, + .MAX_NR_GP_COUNTERS = KVM_MAX_NR_AMD_GP_COUNTERS, + .MIN_NR_GP_COUNTERS = AMD64_NUM_COUNTERS, }; diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 62926f1a5f7b..f59c65abe3cf 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -6,46 +6,41 @@ * * Copyright 2010 Red Hat, Inc. and/or its affiliates. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kvm_types.h> #include <linux/kvm_host.h> #include <linux/kernel.h> #include <linux/highmem.h> +#include <linux/psp.h> #include <linux/psp-sev.h> #include <linux/pagemap.h> #include <linux/swap.h> #include <linux/misc_cgroup.h> #include <linux/processor.h> #include <linux/trace_events.h> -#include <asm/fpu/internal.h> +#include <uapi/linux/sev-guest.h> #include <asm/pkru.h> #include <asm/trapnr.h> +#include <asm/fpu/xcr.h> +#include <asm/fpu/xstate.h> +#include <asm/debugreg.h> +#include <asm/msr.h> +#include <asm/sev.h> +#include "mmu.h" #include "x86.h" #include "svm.h" #include "svm_ops.h" #include "cpuid.h" #include "trace.h" -#define __ex(x) __kvm_handle_fault_on_reboot(x) +#define GHCB_VERSION_MAX 2ULL +#define GHCB_VERSION_MIN 1ULL -#ifndef CONFIG_KVM_AMD_SEV -/* - * When this config is not defined, SEV feature is not supported and APIs in - * this file are not used but this file still gets compiled into the KVM AMD - * module. - * - * We will not have MISC_CG_RES_SEV and MISC_CG_RES_SEV_ES entries in the enum - * misc_res_type {} defined in linux/misc_cgroup.h. - * - * Below macros allow compilation to succeed. - */ -#define MISC_CG_RES_SEV MISC_CG_RES_TYPES -#define MISC_CG_RES_SEV_ES MISC_CG_RES_TYPES -#endif +#define GHCB_HV_FT_SUPPORTED (GHCB_HV_FT_SNP | GHCB_HV_FT_SNP_AP_CREATION) -#ifdef CONFIG_KVM_AMD_SEV /* enable/disable SEV support */ static bool sev_enabled = true; module_param_named(sev, sev_enabled, bool, 0444); @@ -53,20 +48,60 @@ module_param_named(sev, sev_enabled, bool, 0444); /* enable/disable SEV-ES support */ static bool sev_es_enabled = true; module_param_named(sev_es, sev_es_enabled, bool, 0444); -#else -#define sev_enabled false -#define sev_es_enabled false -#endif /* CONFIG_KVM_AMD_SEV */ + +/* enable/disable SEV-SNP support */ +static bool sev_snp_enabled = true; +module_param_named(sev_snp, sev_snp_enabled, bool, 0444); + +/* enable/disable SEV-ES DebugSwap support */ +static bool sev_es_debug_swap_enabled = true; +module_param_named(debug_swap, sev_es_debug_swap_enabled, bool, 0444); +static u64 sev_supported_vmsa_features; + +static unsigned int nr_ciphertext_hiding_asids; +module_param_named(ciphertext_hiding_asids, nr_ciphertext_hiding_asids, uint, 0444); + +#define AP_RESET_HOLD_NONE 0 +#define AP_RESET_HOLD_NAE_EVENT 1 +#define AP_RESET_HOLD_MSR_PROTO 2 + +/* + * SEV-SNP policy bits that can be supported by KVM. These include policy bits + * that have implementation support within KVM or policy bits that do not + * require implementation support within KVM to enforce the policy. + */ +#define KVM_SNP_POLICY_MASK_VALID (SNP_POLICY_MASK_API_MINOR | \ + SNP_POLICY_MASK_API_MAJOR | \ + SNP_POLICY_MASK_SMT | \ + SNP_POLICY_MASK_RSVD_MBO | \ + SNP_POLICY_MASK_DEBUG | \ + SNP_POLICY_MASK_SINGLE_SOCKET | \ + SNP_POLICY_MASK_CXL_ALLOW | \ + SNP_POLICY_MASK_MEM_AES_256_XTS | \ + SNP_POLICY_MASK_RAPL_DIS | \ + SNP_POLICY_MASK_CIPHERTEXT_HIDING_DRAM | \ + SNP_POLICY_MASK_PAGE_SWAP_DISABLE) + +static u64 snp_supported_policy_bits __ro_after_init; + +#define INITIAL_VMSA_GPA 0xFFFFFFFFF000 static u8 sev_enc_bit; static DECLARE_RWSEM(sev_deactivate_lock); static DEFINE_MUTEX(sev_bitmap_lock); unsigned int max_sev_asid; static unsigned int min_sev_asid; +static unsigned int max_sev_es_asid; +static unsigned int min_sev_es_asid; +static unsigned int max_snp_asid; +static unsigned int min_snp_asid; static unsigned long sev_me_mask; +static unsigned int nr_asids; static unsigned long *sev_asid_bitmap; static unsigned long *sev_reclaim_asid_bitmap; +static int snp_decommission_context(struct kvm *kvm); + struct enc_region { struct list_head list; unsigned long npages; @@ -76,13 +111,14 @@ struct enc_region { }; /* Called with the sev_bitmap_lock held, or on shutdown */ -static int sev_flush_asids(int min_asid, int max_asid) +static int sev_flush_asids(unsigned int min_asid, unsigned int max_asid) { - int ret, pos, error = 0; + int ret, error = 0; + unsigned int asid; /* Check if there are any ASIDs to reclaim before performing a flush */ - pos = find_next_bit(sev_reclaim_asid_bitmap, max_asid, min_asid); - if (pos >= max_asid) + asid = find_next_bit(sev_reclaim_asid_bitmap, nr_asids, min_asid); + if (asid > max_asid) return -EBUSY; /* @@ -91,46 +127,104 @@ static int sev_flush_asids(int min_asid, int max_asid) */ down_write(&sev_deactivate_lock); + /* SNP firmware requires use of WBINVD for ASID recycling. */ wbinvd_on_all_cpus(); - ret = sev_guest_df_flush(&error); + + if (sev_snp_enabled) + ret = sev_do_cmd(SEV_CMD_SNP_DF_FLUSH, NULL, &error); + else + ret = sev_guest_df_flush(&error); up_write(&sev_deactivate_lock); if (ret) - pr_err("SEV: DF_FLUSH failed, ret=%d, error=%#x\n", ret, error); + pr_err("SEV%s: DF_FLUSH failed, ret=%d, error=%#x\n", + sev_snp_enabled ? "-SNP" : "", ret, error); return ret; } static inline bool is_mirroring_enc_context(struct kvm *kvm) { - return !!to_kvm_svm(kvm)->sev_info.enc_context_owner; + return !!to_kvm_sev_info(kvm)->enc_context_owner; +} + +static bool sev_vcpu_has_debug_swap(struct vcpu_svm *svm) +{ + struct kvm_vcpu *vcpu = &svm->vcpu; + struct kvm_sev_info *sev = to_kvm_sev_info(vcpu->kvm); + + return sev->vmsa_features & SVM_SEV_FEAT_DEBUG_SWAP; +} + +static bool snp_is_secure_tsc_enabled(struct kvm *kvm) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + + return (sev->vmsa_features & SVM_SEV_FEAT_SECURE_TSC) && + !WARN_ON_ONCE(!sev_snp_guest(kvm)); } /* Must be called with the sev_bitmap_lock held */ -static bool __sev_recycle_asids(int min_asid, int max_asid) +static bool __sev_recycle_asids(unsigned int min_asid, unsigned int max_asid) { if (sev_flush_asids(min_asid, max_asid)) return false; /* The flush process will flush all reclaimable SEV and SEV-ES ASIDs */ bitmap_xor(sev_asid_bitmap, sev_asid_bitmap, sev_reclaim_asid_bitmap, - max_sev_asid); - bitmap_zero(sev_reclaim_asid_bitmap, max_sev_asid); + nr_asids); + bitmap_zero(sev_reclaim_asid_bitmap, nr_asids); return true; } -static int sev_asid_new(struct kvm_sev_info *sev) +static int sev_misc_cg_try_charge(struct kvm_sev_info *sev) { - int pos, min_asid, max_asid, ret; + enum misc_res_type type = sev->es_active ? MISC_CG_RES_SEV_ES : MISC_CG_RES_SEV; + return misc_cg_try_charge(type, sev->misc_cg, 1); +} + +static void sev_misc_cg_uncharge(struct kvm_sev_info *sev) +{ + enum misc_res_type type = sev->es_active ? MISC_CG_RES_SEV_ES : MISC_CG_RES_SEV; + misc_cg_uncharge(type, sev->misc_cg, 1); +} + +static int sev_asid_new(struct kvm_sev_info *sev, unsigned long vm_type) +{ + /* + * SEV-enabled guests must use asid from min_sev_asid to max_sev_asid. + * SEV-ES-enabled guest can use from 1 to min_sev_asid - 1. + */ + unsigned int min_asid, max_asid, asid; bool retry = true; - enum misc_res_type type; + int ret; + + if (vm_type == KVM_X86_SNP_VM) { + min_asid = min_snp_asid; + max_asid = max_snp_asid; + } else if (sev->es_active) { + min_asid = min_sev_es_asid; + max_asid = max_sev_es_asid; + } else { + min_asid = min_sev_asid; + max_asid = max_sev_asid; + } + + /* + * The min ASID can end up larger than the max if basic SEV support is + * effectively disabled by disallowing use of ASIDs for SEV guests. + * Similarly for SEV-ES guests the min ASID can end up larger than the + * max when ciphertext hiding is enabled, effectively disabling SEV-ES + * support. + */ + if (min_asid > max_asid) + return -ENOTTY; - type = sev->es_active ? MISC_CG_RES_SEV_ES : MISC_CG_RES_SEV; WARN_ON(sev->misc_cg); sev->misc_cg = get_current_misc_cg(); - ret = misc_cg_try_charge(type, sev->misc_cg, 1); + ret = sev_misc_cg_try_charge(sev); if (ret) { put_misc_cg(sev->misc_cg); sev->misc_cg = NULL; @@ -139,15 +233,9 @@ static int sev_asid_new(struct kvm_sev_info *sev) mutex_lock(&sev_bitmap_lock); - /* - * SEV-enabled guests must use asid from min_sev_asid to max_sev_asid. - * SEV-ES-enabled guest can use from 1 to min_sev_asid - 1. - */ - min_asid = sev->es_active ? 0 : min_sev_asid - 1; - max_asid = sev->es_active ? min_sev_asid - 1 : max_sev_asid; again: - pos = find_next_zero_bit(sev_asid_bitmap, max_sev_asid, min_asid); - if (pos >= max_asid) { + asid = find_next_zero_bit(sev_asid_bitmap, max_asid + 1, min_asid); + if (asid > max_asid) { if (retry && __sev_recycle_asids(min_asid, max_asid)) { retry = false; goto again; @@ -157,45 +245,41 @@ again: goto e_uncharge; } - __set_bit(pos, sev_asid_bitmap); + __set_bit(asid, sev_asid_bitmap); mutex_unlock(&sev_bitmap_lock); - return pos + 1; + sev->asid = asid; + return 0; e_uncharge: - misc_cg_uncharge(type, sev->misc_cg, 1); + sev_misc_cg_uncharge(sev); put_misc_cg(sev->misc_cg); sev->misc_cg = NULL; return ret; } -static int sev_get_asid(struct kvm *kvm) +static unsigned int sev_get_asid(struct kvm *kvm) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; - - return sev->asid; + return to_kvm_sev_info(kvm)->asid; } static void sev_asid_free(struct kvm_sev_info *sev) { struct svm_cpu_data *sd; - int cpu, pos; - enum misc_res_type type; + int cpu; mutex_lock(&sev_bitmap_lock); - pos = sev->asid - 1; - __set_bit(pos, sev_reclaim_asid_bitmap); + __set_bit(sev->asid, sev_reclaim_asid_bitmap); for_each_possible_cpu(cpu) { - sd = per_cpu(svm_data, cpu); - sd->sev_vmcbs[pos] = NULL; + sd = per_cpu_ptr(&svm_data, cpu); + sd->sev_vmcbs[sev->asid] = NULL; } mutex_unlock(&sev_bitmap_lock); - type = sev->es_active ? MISC_CG_RES_SEV_ES : MISC_CG_RES_SEV; - misc_cg_uncharge(type, sev->misc_cg, 1); + sev_misc_cg_uncharge(sev); put_misc_cg(sev->misc_cg); sev->misc_cg = NULL; } @@ -211,6 +295,53 @@ static void sev_decommission(unsigned int handle) sev_guest_decommission(&decommission, NULL); } +/* + * Transition a page to hypervisor-owned/shared state in the RMP table. This + * should not fail under normal conditions, but leak the page should that + * happen since it will no longer be usable by the host due to RMP protections. + */ +static int kvm_rmp_make_shared(struct kvm *kvm, u64 pfn, enum pg_level level) +{ + if (KVM_BUG_ON(rmp_make_shared(pfn, level), kvm)) { + snp_leak_pages(pfn, page_level_size(level) >> PAGE_SHIFT); + return -EIO; + } + + return 0; +} + +/* + * Certain page-states, such as Pre-Guest and Firmware pages (as documented + * in Chapter 5 of the SEV-SNP Firmware ABI under "Page States") cannot be + * directly transitioned back to normal/hypervisor-owned state via RMPUPDATE + * unless they are reclaimed first. + * + * Until they are reclaimed and subsequently transitioned via RMPUPDATE, they + * might not be usable by the host due to being set as immutable or still + * being associated with a guest ASID. + * + * Bug the VM and leak the page if reclaim fails, or if the RMP entry can't be + * converted back to shared, as the page is no longer usable due to RMP + * protections, and it's infeasible for the guest to continue on. + */ +static int snp_page_reclaim(struct kvm *kvm, u64 pfn) +{ + struct sev_data_snp_page_reclaim data = {0}; + int fw_err, rc; + + data.paddr = __sme_set(pfn << PAGE_SHIFT); + rc = sev_do_cmd(SEV_CMD_SNP_PAGE_RECLAIM, &data, &fw_err); + if (KVM_BUG(rc, kvm, "Failed to reclaim PFN %llx, rc %d fw_err %d", pfn, rc, fw_err)) { + snp_leak_pages(pfn, 1); + return -EIO; + } + + if (kvm_rmp_make_shared(kvm, pfn, PG_LEVEL_4K)) + return -EIO; + + return rc; +} + static void sev_unbind_asid(struct kvm *kvm, unsigned int handle) { struct sev_data_deactivate deactivate; @@ -228,47 +359,213 @@ static void sev_unbind_asid(struct kvm *kvm, unsigned int handle) sev_decommission(handle); } -static int sev_guest_init(struct kvm *kvm, struct kvm_sev_cmd *argp) +/* + * This sets up bounce buffers/firmware pages to handle SNP Guest Request + * messages (e.g. attestation requests). See "SNP Guest Request" in the GHCB + * 2.0 specification for more details. + * + * Technically, when an SNP Guest Request is issued, the guest will provide its + * own request/response pages, which could in theory be passed along directly + * to firmware rather than using bounce pages. However, these pages would need + * special care: + * + * - Both pages are from shared guest memory, so they need to be protected + * from migration/etc. occurring while firmware reads/writes to them. At a + * minimum, this requires elevating the ref counts and potentially needing + * an explicit pinning of the memory. This places additional restrictions + * on what type of memory backends userspace can use for shared guest + * memory since there is some reliance on using refcounted pages. + * + * - The response page needs to be switched to Firmware-owned[1] state + * before the firmware can write to it, which can lead to potential + * host RMP #PFs if the guest is misbehaved and hands the host a + * guest page that KVM might write to for other reasons (e.g. virtio + * buffers/etc.). + * + * Both of these issues can be avoided completely by using separately-allocated + * bounce pages for both the request/response pages and passing those to + * firmware instead. So that's what is being set up here. + * + * Guest requests rely on message sequence numbers to ensure requests are + * issued to firmware in the order the guest issues them, so concurrent guest + * requests generally shouldn't happen. But a misbehaved guest could issue + * concurrent guest requests in theory, so a mutex is used to serialize + * access to the bounce buffers. + * + * [1] See the "Page States" section of the SEV-SNP Firmware ABI for more + * details on Firmware-owned pages, along with "RMP and VMPL Access Checks" + * in the APM for details on the related RMP restrictions. + */ +static int snp_guest_req_init(struct kvm *kvm) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; - bool es_active = argp->id == KVM_SEV_ES_INIT; - int asid, ret; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct page *req_page; + + req_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); + if (!req_page) + return -ENOMEM; + + sev->guest_resp_buf = snp_alloc_firmware_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); + if (!sev->guest_resp_buf) { + __free_page(req_page); + return -EIO; + } + + sev->guest_req_buf = page_address(req_page); + mutex_init(&sev->guest_req_mutex); + + return 0; +} + +static void snp_guest_req_cleanup(struct kvm *kvm) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + + if (sev->guest_resp_buf) + snp_free_firmware_page(sev->guest_resp_buf); + + if (sev->guest_req_buf) + __free_page(virt_to_page(sev->guest_req_buf)); + + sev->guest_req_buf = NULL; + sev->guest_resp_buf = NULL; +} + +static int __sev_guest_init(struct kvm *kvm, struct kvm_sev_cmd *argp, + struct kvm_sev_init *data, + unsigned long vm_type) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct sev_platform_init_args init_args = {0}; + bool es_active = vm_type != KVM_X86_SEV_VM; + bool snp_active = vm_type == KVM_X86_SNP_VM; + u64 valid_vmsa_features = es_active ? sev_supported_vmsa_features : 0; + int ret; if (kvm->created_vcpus) return -EINVAL; - ret = -EBUSY; + if (data->flags) + return -EINVAL; + + if (!snp_active) + valid_vmsa_features &= ~SVM_SEV_FEAT_SECURE_TSC; + + if (data->vmsa_features & ~valid_vmsa_features) + return -EINVAL; + + if (data->ghcb_version > GHCB_VERSION_MAX || (!es_active && data->ghcb_version)) + return -EINVAL; + + /* + * KVM supports the full range of mandatory features defined by version + * 2 of the GHCB protocol, so default to that for SEV-ES guests created + * via KVM_SEV_INIT2 (KVM_SEV_INIT forces version 1). + */ + if (es_active && !data->ghcb_version) + data->ghcb_version = 2; + + if (snp_active && data->ghcb_version < 2) + return -EINVAL; + if (unlikely(sev->active)) - return ret; + return -EINVAL; + sev->active = true; sev->es_active = es_active; - asid = sev_asid_new(sev); - if (asid < 0) + sev->vmsa_features = data->vmsa_features; + sev->ghcb_version = data->ghcb_version; + + if (snp_active) + sev->vmsa_features |= SVM_SEV_FEAT_SNP_ACTIVE; + + ret = sev_asid_new(sev, vm_type); + if (ret) goto e_no_asid; - sev->asid = asid; - ret = sev_platform_init(&argp->error); + init_args.probe = false; + ret = sev_platform_init(&init_args); if (ret) - goto e_free; + goto e_free_asid; + + if (!zalloc_cpumask_var(&sev->have_run_cpus, GFP_KERNEL_ACCOUNT)) { + ret = -ENOMEM; + goto e_free_asid; + } + + /* This needs to happen after SEV/SNP firmware initialization. */ + if (snp_active) { + ret = snp_guest_req_init(kvm); + if (ret) + goto e_free; + } - sev->active = true; - sev->asid = asid; INIT_LIST_HEAD(&sev->regions_list); + INIT_LIST_HEAD(&sev->mirror_vms); + sev->need_init = false; + + kvm_set_apicv_inhibit(kvm, APICV_INHIBIT_REASON_SEV); return 0; e_free: + free_cpumask_var(sev->have_run_cpus); +e_free_asid: + argp->error = init_args.error; sev_asid_free(sev); sev->asid = 0; e_no_asid: + sev->vmsa_features = 0; sev->es_active = false; + sev->active = false; return ret; } +static int sev_guest_init(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct kvm_sev_init data = { + .vmsa_features = 0, + .ghcb_version = 0, + }; + unsigned long vm_type; + + if (kvm->arch.vm_type != KVM_X86_DEFAULT_VM) + return -EINVAL; + + vm_type = (argp->id == KVM_SEV_INIT ? KVM_X86_SEV_VM : KVM_X86_SEV_ES_VM); + + /* + * KVM_SEV_ES_INIT has been deprecated by KVM_SEV_INIT2, so it will + * continue to only ever support the minimal GHCB protocol version. + */ + if (vm_type == KVM_X86_SEV_ES_VM) + data.ghcb_version = GHCB_VERSION_MIN; + + return __sev_guest_init(kvm, argp, &data, vm_type); +} + +static int sev_guest_init2(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct kvm_sev_init data; + + if (!to_kvm_sev_info(kvm)->need_init) + return -EINVAL; + + if (kvm->arch.vm_type != KVM_X86_SEV_VM && + kvm->arch.vm_type != KVM_X86_SEV_ES_VM && + kvm->arch.vm_type != KVM_X86_SNP_VM) + return -EINVAL; + + if (copy_from_user(&data, u64_to_user_ptr(argp->data), sizeof(data))) + return -EFAULT; + + return __sev_guest_init(kvm, argp, &data, kvm->arch.vm_type); +} + static int sev_bind_asid(struct kvm *kvm, unsigned int handle, int *error) { + unsigned int asid = sev_get_asid(kvm); struct sev_data_activate activate; - int asid = sev_get_asid(kvm); int ret; /* activate ASID on the given handle */ @@ -281,29 +578,24 @@ static int sev_bind_asid(struct kvm *kvm, unsigned int handle, int *error) static int __sev_issue_cmd(int fd, int id, void *data, int *error) { - struct fd f; - int ret; + CLASS(fd, f)(fd); - f = fdget(fd); - if (!f.file) + if (fd_empty(f)) return -EBADF; - ret = sev_issue_cmd_external_user(f.file, id, data, error); - - fdput(f); - return ret; + return sev_issue_cmd_external_user(fd_file(f), id, data, error); } static int sev_issue_cmd(struct kvm *kvm, int id, void *data, int *error) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); return __sev_issue_cmd(sev->fd, id, data, error); } static int sev_launch_start(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); struct sev_data_launch_start start; struct kvm_sev_launch_start params; void *dh_blob, *session_blob; @@ -313,7 +605,7 @@ static int sev_launch_start(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, sizeof(params))) + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) return -EFAULT; memset(&start, 0, sizeof(start)); @@ -357,12 +649,13 @@ static int sev_launch_start(struct kvm *kvm, struct kvm_sev_cmd *argp) /* return handle to userspace */ params.handle = start.handle; - if (copy_to_user((void __user *)(uintptr_t)argp->data, ¶ms, sizeof(params))) { + if (copy_to_user(u64_to_user_ptr(argp->data), ¶ms, sizeof(params))) { sev_unbind_asid(kvm, start.handle); ret = -EFAULT; goto e_free_session; } + sev->policy = params.policy; sev->handle = start.handle; sev->fd = argp->sev_fd; @@ -375,9 +668,9 @@ e_free_dh: static struct page **sev_pin_memory(struct kvm *kvm, unsigned long uaddr, unsigned long ulen, unsigned long *n, - int write) + unsigned int flags) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); unsigned long npages, size; int npinned; unsigned long locked, lock_limit; @@ -408,7 +701,7 @@ static struct page **sev_pin_memory(struct kvm *kvm, unsigned long uaddr, /* Avoid using vmalloc for smaller buffers. */ size = npages * sizeof(struct page *); if (size > PAGE_SIZE) - pages = __vmalloc(size, GFP_KERNEL_ACCOUNT | __GFP_ZERO); + pages = __vmalloc(size, GFP_KERNEL_ACCOUNT); else pages = kmalloc(size, GFP_KERNEL_ACCOUNT); @@ -416,7 +709,7 @@ static struct page **sev_pin_memory(struct kvm *kvm, unsigned long uaddr, return ERR_PTR(-ENOMEM); /* Pin the user virtual address. */ - npinned = pin_user_pages_fast(uaddr, npages, write ? FOLL_WRITE : 0, pages); + npinned = pin_user_pages_fast(uaddr, npages, flags, pages); if (npinned != npages) { pr_err("SEV: Failure locking %lu pages.\n", npages); ret = -ENOMEM; @@ -439,11 +732,9 @@ err: static void sev_unpin_memory(struct kvm *kvm, struct page **pages, unsigned long npages) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; - unpin_user_pages(pages, npages); kvfree(pages); - sev->pages_locked -= npages; + to_kvm_sev_info(kvm)->pages_locked -= npages; } static void sev_clflush_pages(struct page *pages[], unsigned long npages) @@ -456,12 +747,36 @@ static void sev_clflush_pages(struct page *pages[], unsigned long npages) return; for (i = 0; i < npages; i++) { - page_virtual = kmap_atomic(pages[i]); + page_virtual = kmap_local_page(pages[i]); clflush_cache_range(page_virtual, PAGE_SIZE); - kunmap_atomic(page_virtual); + kunmap_local(page_virtual); + cond_resched(); } } +static void sev_writeback_caches(struct kvm *kvm) +{ + /* + * Ensure that all dirty guest tagged cache entries are written back + * before releasing the pages back to the system for use. CLFLUSH will + * not do this without SME_COHERENT, and flushing many cache lines + * individually is slower than blasting WBINVD for large VMs, so issue + * WBNOINVD (or WBINVD if the "no invalidate" variant is unsupported) + * on CPUs that have done VMRUN, i.e. may have dirtied data using the + * VM's ASID. + * + * For simplicity, never remove CPUs from the bitmap. Ideally, KVM + * would clear the mask when flushing caches, but doing so requires + * serializing multiple calls and having responding CPUs (to the IPI) + * mark themselves as still running if they are running (or about to + * run) a vCPU for the VM. + * + * Note, the caller is responsible for ensuring correctness if the mask + * can be modified, e.g. if a CPU could be doing VMRUN. + */ + wbnoinvd_on_cpus_mask(to_kvm_sev_info(kvm)->have_run_cpus); +} + static unsigned long get_num_contig_pages(unsigned long idx, struct page **inpages, unsigned long npages) { @@ -486,7 +801,6 @@ static unsigned long get_num_contig_pages(unsigned long idx, static int sev_launch_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) { unsigned long vaddr, vaddr_end, next_vaddr, npages, pages, size, i; - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct kvm_sev_launch_update_data params; struct sev_data_launch_update_data data; struct page **inpages; @@ -495,7 +809,7 @@ static int sev_launch_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, sizeof(params))) + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) return -EFAULT; vaddr = params.uaddr; @@ -503,7 +817,7 @@ static int sev_launch_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) vaddr_end = vaddr + size; /* Lock the user memory. */ - inpages = sev_pin_memory(kvm, vaddr, size, &npages, 1); + inpages = sev_pin_memory(kvm, vaddr, size, &npages, FOLL_WRITE); if (IS_ERR(inpages)) return PTR_ERR(inpages); @@ -514,7 +828,7 @@ static int sev_launch_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) sev_clflush_pages(inpages, npages); data.reserved = 0; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; for (i = 0; vaddr < vaddr_end; vaddr = next_vaddr, i += pages) { int offset, len; @@ -553,12 +867,26 @@ e_unpin: static int sev_es_sync_vmsa(struct vcpu_svm *svm) { - struct vmcb_save_area *save = &svm->vmcb->save; + struct kvm_vcpu *vcpu = &svm->vcpu; + struct kvm_sev_info *sev = to_kvm_sev_info(vcpu->kvm); + struct sev_es_save_area *save = svm->sev_es.vmsa; + struct xregs_state *xsave; + const u8 *s; + u8 *d; + int i; /* Check some debug related fields before encrypting the VMSA */ - if (svm->vcpu.guest_debug || (save->dr7 & ~DR7_FIXED_1)) + if (svm->vcpu.guest_debug || (svm->vmcb->save.dr7 & ~DR7_FIXED_1)) return -EINVAL; + /* + * SEV-ES will use a VMSA that is pointed to by the VMCB, not + * the traditional VMSA that is part of the VMCB. Copy the + * traditional VMSA as it has been built so far (in prep + * for LAUNCH_UPDATE_VMSA) to be the initial SEV-ES state. + */ + memcpy(save, &svm->vmcb->save, sizeof(svm->vmcb->save)); + /* Sync registgers */ save->rax = svm->vcpu.arch.regs[VCPU_REGS_RAX]; save->rbx = svm->vcpu.arch.regs[VCPU_REGS_RBX]; @@ -584,55 +912,124 @@ static int sev_es_sync_vmsa(struct vcpu_svm *svm) save->xcr0 = svm->vcpu.arch.xcr0; save->pkru = svm->vcpu.arch.pkru; save->xss = svm->vcpu.arch.ia32_xss; + save->dr6 = svm->vcpu.arch.dr6; + + save->sev_features = sev->vmsa_features; /* - * SEV-ES will use a VMSA that is pointed to by the VMCB, not - * the traditional VMSA that is part of the VMCB. Copy the - * traditional VMSA as it has been built so far (in prep - * for LAUNCH_UPDATE_VMSA) to be the initial SEV-ES state. + * Skip FPU and AVX setup with KVM_SEV_ES_INIT to avoid + * breaking older measurements. */ - memcpy(svm->vmsa, save, sizeof(*save)); + if (vcpu->kvm->arch.vm_type != KVM_X86_DEFAULT_VM) { + xsave = &vcpu->arch.guest_fpu.fpstate->regs.xsave; + save->x87_dp = xsave->i387.rdp; + save->mxcsr = xsave->i387.mxcsr; + save->x87_ftw = xsave->i387.twd; + save->x87_fsw = xsave->i387.swd; + save->x87_fcw = xsave->i387.cwd; + save->x87_fop = xsave->i387.fop; + save->x87_ds = 0; + save->x87_cs = 0; + save->x87_rip = xsave->i387.rip; + + for (i = 0; i < 8; i++) { + /* + * The format of the x87 save area is undocumented and + * definitely not what you would expect. It consists of + * an 8*8 bytes area with bytes 0-7, and an 8*2 bytes + * area with bytes 8-9 of each register. + */ + d = save->fpreg_x87 + i * 8; + s = ((u8 *)xsave->i387.st_space) + i * 16; + memcpy(d, s, 8); + save->fpreg_x87[64 + i * 2] = s[8]; + save->fpreg_x87[64 + i * 2 + 1] = s[9]; + } + memcpy(save->fpreg_xmm, xsave->i387.xmm_space, 256); + + s = get_xsave_addr(xsave, XFEATURE_YMM); + if (s) + memcpy(save->fpreg_ymm, s, 256); + else + memset(save->fpreg_ymm, 0, 256); + } + + pr_debug("Virtual Machine Save Area (VMSA):\n"); + print_hex_dump_debug("", DUMP_PREFIX_NONE, 16, 1, save, sizeof(*save), false); return 0; } -static int sev_launch_update_vmsa(struct kvm *kvm, struct kvm_sev_cmd *argp) +static int __sev_launch_update_vmsa(struct kvm *kvm, struct kvm_vcpu *vcpu, + int *error) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_launch_update_vmsa vmsa; + struct vcpu_svm *svm = to_svm(vcpu); + int ret; + + if (vcpu->guest_debug) { + pr_warn_once("KVM_SET_GUEST_DEBUG for SEV-ES guest is not supported"); + return -EINVAL; + } + + /* Perform some pre-encryption checks against the VMSA */ + ret = sev_es_sync_vmsa(svm); + if (ret) + return ret; + + /* + * The LAUNCH_UPDATE_VMSA command will perform in-place encryption of + * the VMSA memory content (i.e it will write the same memory region + * with the guest's key), so invalidate it first. + */ + clflush_cache_range(svm->sev_es.vmsa, PAGE_SIZE); + + vmsa.reserved = 0; + vmsa.handle = to_kvm_sev_info(kvm)->handle; + vmsa.address = __sme_pa(svm->sev_es.vmsa); + vmsa.len = PAGE_SIZE; + ret = sev_issue_cmd(kvm, SEV_CMD_LAUNCH_UPDATE_VMSA, &vmsa, error); + if (ret) + return ret; + + /* + * SEV-ES guests maintain an encrypted version of their FPU + * state which is restored and saved on VMRUN and VMEXIT. + * Mark vcpu->arch.guest_fpu->fpstate as scratch so it won't + * do xsave/xrstor on it. + */ + fpstate_set_confidential(&vcpu->arch.guest_fpu); + vcpu->arch.guest_state_protected = true; + + /* + * SEV-ES guest mandates LBR Virtualization to be _always_ ON. Enable it + * only after setting guest_state_protected because KVM_SET_MSRS allows + * dynamic toggling of LBRV (for performance reason) on write access to + * MSR_IA32_DEBUGCTLMSR when guest_state_protected is not set. + */ + svm_enable_lbrv(vcpu); + return 0; +} + +static int sev_launch_update_vmsa(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ struct kvm_vcpu *vcpu; - int i, ret; + unsigned long i; + int ret; if (!sev_es_guest(kvm)) return -ENOTTY; - vmsa.reserved = 0; - kvm_for_each_vcpu(i, vcpu, kvm) { - struct vcpu_svm *svm = to_svm(vcpu); - - /* Perform some pre-encryption checks against the VMSA */ - ret = sev_es_sync_vmsa(svm); + ret = mutex_lock_killable(&vcpu->mutex); if (ret) return ret; - /* - * The LAUNCH_UPDATE_VMSA command will perform in-place - * encryption of the VMSA memory content (i.e it will write - * the same memory region with the guest's key), so invalidate - * it first. - */ - clflush_cache_range(svm->vmsa, PAGE_SIZE); + ret = __sev_launch_update_vmsa(kvm, vcpu, &argp->error); - vmsa.handle = sev->handle; - vmsa.address = __sme_pa(svm->vmsa); - vmsa.len = PAGE_SIZE; - ret = sev_issue_cmd(kvm, SEV_CMD_LAUNCH_UPDATE_VMSA, &vmsa, - &argp->error); + mutex_unlock(&vcpu->mutex); if (ret) return ret; - - svm->vcpu.arch.guest_state_protected = true; } return 0; @@ -640,8 +1037,7 @@ static int sev_launch_update_vmsa(struct kvm *kvm, struct kvm_sev_cmd *argp) static int sev_launch_measure(struct kvm *kvm, struct kvm_sev_cmd *argp) { - void __user *measure = (void __user *)(uintptr_t)argp->data; - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + void __user *measure = u64_to_user_ptr(argp->data); struct sev_data_launch_measure data; struct kvm_sev_launch_measure params; void __user *p = NULL; @@ -660,12 +1056,12 @@ static int sev_launch_measure(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!params.len) goto cmd; - p = (void __user *)(uintptr_t)params.uaddr; + p = u64_to_user_ptr(params.uaddr); if (p) { if (params.len > SEV_FW_BLOB_MAX_SIZE) return -EINVAL; - blob = kmalloc(params.len, GFP_KERNEL_ACCOUNT); + blob = kzalloc(params.len, GFP_KERNEL_ACCOUNT); if (!blob) return -ENOMEM; @@ -674,7 +1070,7 @@ static int sev_launch_measure(struct kvm *kvm, struct kvm_sev_cmd *argp) } cmd: - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_LAUNCH_MEASURE, &data, &argp->error); /* @@ -702,19 +1098,17 @@ e_free_blob: static int sev_launch_finish(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_launch_finish data; if (!sev_guest(kvm)) return -ENOTTY; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; return sev_issue_cmd(kvm, SEV_CMD_LAUNCH_FINISH, &data, &argp->error); } static int sev_guest_status(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct kvm_sev_guest_status params; struct sev_data_guest_status data; int ret; @@ -724,7 +1118,7 @@ static int sev_guest_status(struct kvm *kvm, struct kvm_sev_cmd *argp) memset(&data, 0, sizeof(data)); - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_GUEST_STATUS, &data, &argp->error); if (ret) return ret; @@ -733,7 +1127,7 @@ static int sev_guest_status(struct kvm *kvm, struct kvm_sev_cmd *argp) params.state = data.state; params.handle = data.handle; - if (copy_to_user((void __user *)(uintptr_t)argp->data, ¶ms, sizeof(params))) + if (copy_to_user(u64_to_user_ptr(argp->data), ¶ms, sizeof(params))) ret = -EFAULT; return ret; @@ -743,11 +1137,10 @@ static int __sev_issue_dbg_cmd(struct kvm *kvm, unsigned long src, unsigned long dst, int size, int *error, bool enc) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_dbg data; data.reserved = 0; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; data.dst_addr = dst; data.src_addr = src; data.len = size; @@ -785,7 +1178,7 @@ static int __sev_dbg_decrypt_user(struct kvm *kvm, unsigned long paddr, if (!IS_ALIGNED(dst_paddr, 16) || !IS_ALIGNED(paddr, 16) || !IS_ALIGNED(size, 16)) { - tpage = (void *)alloc_page(GFP_KERNEL); + tpage = (void *)alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); if (!tpage) return -ENOMEM; @@ -821,7 +1214,7 @@ static int __sev_dbg_encrypt_user(struct kvm *kvm, unsigned long paddr, /* If source buffer is not aligned then use an intermediate buffer */ if (!IS_ALIGNED((unsigned long)vaddr, 16)) { - src_tpage = alloc_page(GFP_KERNEL); + src_tpage = alloc_page(GFP_KERNEL_ACCOUNT); if (!src_tpage) return -ENOMEM; @@ -842,7 +1235,7 @@ static int __sev_dbg_encrypt_user(struct kvm *kvm, unsigned long paddr, if (!IS_ALIGNED((unsigned long)dst_vaddr, 16) || !IS_ALIGNED(size, 16)) { int dst_offset; - dst_tpage = alloc_page(GFP_KERNEL); + dst_tpage = alloc_page(GFP_KERNEL_ACCOUNT); if (!dst_tpage) { ret = -ENOMEM; goto e_free; @@ -898,7 +1291,7 @@ static int sev_dbg_crypt(struct kvm *kvm, struct kvm_sev_cmd *argp, bool dec) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(&debug, (void __user *)(uintptr_t)argp->data, sizeof(debug))) + if (copy_from_user(&debug, u64_to_user_ptr(argp->data), sizeof(debug))) return -EFAULT; if (!debug.len || debug.src_uaddr + debug.len < debug.src_uaddr) @@ -919,7 +1312,7 @@ static int sev_dbg_crypt(struct kvm *kvm, struct kvm_sev_cmd *argp, bool dec) if (IS_ERR(src_p)) return PTR_ERR(src_p); - dst_p = sev_pin_memory(kvm, dst_vaddr & PAGE_MASK, PAGE_SIZE, &n, 1); + dst_p = sev_pin_memory(kvm, dst_vaddr & PAGE_MASK, PAGE_SIZE, &n, FOLL_WRITE); if (IS_ERR(dst_p)) { sev_unpin_memory(kvm, src_p, n); return PTR_ERR(dst_p); @@ -971,7 +1364,6 @@ err: static int sev_launch_secret(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_launch_secret data; struct kvm_sev_launch_secret params; struct page **pages; @@ -982,10 +1374,10 @@ static int sev_launch_secret(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, sizeof(params))) + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) return -EFAULT; - pages = sev_pin_memory(kvm, params.guest_uaddr, params.guest_len, &n, 1); + pages = sev_pin_memory(kvm, params.guest_uaddr, params.guest_len, &n, FOLL_WRITE); if (IS_ERR(pages)) return PTR_ERR(pages); @@ -1027,7 +1419,7 @@ static int sev_launch_secret(struct kvm *kvm, struct kvm_sev_cmd *argp) data.hdr_address = __psp_pa(hdr); data.hdr_len = params.hdr_len; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_LAUNCH_UPDATE_SECRET, &data, &argp->error); kfree(hdr); @@ -1046,8 +1438,7 @@ e_unpin_memory: static int sev_get_attestation_report(struct kvm *kvm, struct kvm_sev_cmd *argp) { - void __user *report = (void __user *)(uintptr_t)argp->data; - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + void __user *report = u64_to_user_ptr(argp->data); struct sev_data_attestation_report data; struct kvm_sev_attestation_report params; void __user *p; @@ -1057,7 +1448,7 @@ static int sev_get_attestation_report(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, sizeof(params))) + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) return -EFAULT; memset(&data, 0, sizeof(data)); @@ -1066,12 +1457,12 @@ static int sev_get_attestation_report(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!params.len) goto cmd; - p = (void __user *)(uintptr_t)params.uaddr; + p = u64_to_user_ptr(params.uaddr); if (p) { if (params.len > SEV_FW_BLOB_MAX_SIZE) return -EINVAL; - blob = kmalloc(params.len, GFP_KERNEL_ACCOUNT); + blob = kzalloc(params.len, GFP_KERNEL_ACCOUNT); if (!blob) return -ENOMEM; @@ -1080,7 +1471,7 @@ static int sev_get_attestation_report(struct kvm *kvm, struct kvm_sev_cmd *argp) memcpy(data.mnonce, params.mnonce, sizeof(params.mnonce)); } cmd: - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_ATTESTATION_REPORT, &data, &argp->error); /* * If we query the session length, FW responded with expected data. @@ -1110,16 +1501,15 @@ static int __sev_send_start_query_session_length(struct kvm *kvm, struct kvm_sev_cmd *argp, struct kvm_sev_send_start *params) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_send_start data; int ret; memset(&data, 0, sizeof(data)); - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_SEND_START, &data, &argp->error); params->session_len = data.session_len; - if (copy_to_user((void __user *)(uintptr_t)argp->data, params, + if (copy_to_user(u64_to_user_ptr(argp->data), params, sizeof(struct kvm_sev_send_start))) ret = -EFAULT; @@ -1128,7 +1518,6 @@ __sev_send_start_query_session_length(struct kvm *kvm, struct kvm_sev_cmd *argp, static int sev_send_start(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_send_start data; struct kvm_sev_send_start params; void *amd_certs, *session_data; @@ -1138,7 +1527,7 @@ static int sev_send_start(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(struct kvm_sev_send_start))) return -EFAULT; @@ -1153,7 +1542,7 @@ static int sev_send_start(struct kvm *kvm, struct kvm_sev_cmd *argp) return -EINVAL; /* allocate the memory to hold the session data blob */ - session_data = kmalloc(params.session_len, GFP_KERNEL_ACCOUNT); + session_data = kzalloc(params.session_len, GFP_KERNEL_ACCOUNT); if (!session_data) return -ENOMEM; @@ -1189,11 +1578,11 @@ static int sev_send_start(struct kvm *kvm, struct kvm_sev_cmd *argp) data.amd_certs_len = params.amd_certs_len; data.session_address = __psp_pa(session_data); data.session_len = params.session_len; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_SEND_START, &data, &argp->error); - if (!ret && copy_to_user((void __user *)(uintptr_t)params.session_uaddr, + if (!ret && copy_to_user(u64_to_user_ptr(params.session_uaddr), session_data, params.session_len)) { ret = -EFAULT; goto e_free_amd_cert; @@ -1201,7 +1590,7 @@ static int sev_send_start(struct kvm *kvm, struct kvm_sev_cmd *argp) params.policy = data.policy; params.session_len = data.session_len; - if (copy_to_user((void __user *)(uintptr_t)argp->data, ¶ms, + if (copy_to_user(u64_to_user_ptr(argp->data), ¶ms, sizeof(struct kvm_sev_send_start))) ret = -EFAULT; @@ -1221,18 +1610,17 @@ static int __sev_send_update_data_query_lengths(struct kvm *kvm, struct kvm_sev_cmd *argp, struct kvm_sev_send_update_data *params) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_send_update_data data; int ret; memset(&data, 0, sizeof(data)); - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_SEND_UPDATE_DATA, &data, &argp->error); params->hdr_len = data.hdr_len; params->trans_len = data.trans_len; - if (copy_to_user((void __user *)(uintptr_t)argp->data, params, + if (copy_to_user(u64_to_user_ptr(argp->data), params, sizeof(struct kvm_sev_send_update_data))) ret = -EFAULT; @@ -1241,7 +1629,6 @@ __sev_send_update_data_query_lengths(struct kvm *kvm, struct kvm_sev_cmd *argp, static int sev_send_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_send_update_data data; struct kvm_sev_send_update_data params; void *hdr, *trans_data; @@ -1252,7 +1639,7 @@ static int sev_send_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -ENOTTY; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(struct kvm_sev_send_update_data))) return -EFAULT; @@ -1266,22 +1653,22 @@ static int sev_send_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) /* Check if we are crossing the page boundary */ offset = params.guest_uaddr & (PAGE_SIZE - 1); - if ((params.guest_len + offset > PAGE_SIZE)) + if (params.guest_len > PAGE_SIZE || (params.guest_len + offset) > PAGE_SIZE) return -EINVAL; /* Pin guest memory */ guest_page = sev_pin_memory(kvm, params.guest_uaddr & PAGE_MASK, PAGE_SIZE, &n, 0); - if (!guest_page) - return -EFAULT; + if (IS_ERR(guest_page)) + return PTR_ERR(guest_page); /* allocate memory for header and transport buffer */ ret = -ENOMEM; - hdr = kmalloc(params.hdr_len, GFP_KERNEL_ACCOUNT); + hdr = kzalloc(params.hdr_len, GFP_KERNEL); if (!hdr) goto e_unpin; - trans_data = kmalloc(params.trans_len, GFP_KERNEL_ACCOUNT); + trans_data = kzalloc(params.trans_len, GFP_KERNEL); if (!trans_data) goto e_free_hdr; @@ -1295,7 +1682,7 @@ static int sev_send_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) data.guest_address = (page_to_pfn(guest_page[0]) << PAGE_SHIFT) + offset; data.guest_address |= sev_me_mask; data.guest_len = params.guest_len; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_SEND_UPDATE_DATA, &data, &argp->error); @@ -1303,15 +1690,16 @@ static int sev_send_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) goto e_free_trans_data; /* copy transport buffer to user space */ - if (copy_to_user((void __user *)(uintptr_t)params.trans_uaddr, + if (copy_to_user(u64_to_user_ptr(params.trans_uaddr), trans_data, params.trans_len)) { ret = -EFAULT; goto e_free_trans_data; } /* Copy packet header to userspace. */ - ret = copy_to_user((void __user *)(uintptr_t)params.hdr_uaddr, hdr, - params.hdr_len); + if (copy_to_user(u64_to_user_ptr(params.hdr_uaddr), hdr, + params.hdr_len)) + ret = -EFAULT; e_free_trans_data: kfree(trans_data); @@ -1325,31 +1713,29 @@ e_unpin: static int sev_send_finish(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_send_finish data; if (!sev_guest(kvm)) return -ENOTTY; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; return sev_issue_cmd(kvm, SEV_CMD_SEND_FINISH, &data, &argp->error); } static int sev_send_cancel(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_send_cancel data; if (!sev_guest(kvm)) return -ENOTTY; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; return sev_issue_cmd(kvm, SEV_CMD_SEND_CANCEL, &data, &argp->error); } static int sev_receive_start(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); struct sev_data_receive_start start; struct kvm_sev_receive_start params; int *error = &argp->error; @@ -1361,7 +1747,7 @@ static int sev_receive_start(struct kvm *kvm, struct kvm_sev_cmd *argp) return -ENOTTY; /* Get parameter from the userspace */ - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(struct kvm_sev_receive_start))) return -EFAULT; @@ -1397,11 +1783,13 @@ static int sev_receive_start(struct kvm *kvm, struct kvm_sev_cmd *argp) /* Bind ASID to this guest */ ret = sev_bind_asid(kvm, start.handle, error); - if (ret) + if (ret) { + sev_decommission(start.handle); goto e_free_session; + } params.handle = start.handle; - if (copy_to_user((void __user *)(uintptr_t)argp->data, + if (copy_to_user(u64_to_user_ptr(argp->data), ¶ms, sizeof(struct kvm_sev_receive_start))) { ret = -EFAULT; sev_unbind_asid(kvm, start.handle); @@ -1421,7 +1809,6 @@ e_free_pdh: static int sev_receive_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct kvm_sev_receive_update_data params; struct sev_data_receive_update_data data; void *hdr = NULL, *trans = NULL; @@ -1432,7 +1819,7 @@ static int sev_receive_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) if (!sev_guest(kvm)) return -EINVAL; - if (copy_from_user(¶ms, (void __user *)(uintptr_t)argp->data, + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(struct kvm_sev_receive_update_data))) return -EFAULT; @@ -1443,7 +1830,7 @@ static int sev_receive_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) /* Check if we are crossing the page boundary */ offset = params.guest_uaddr & (PAGE_SIZE - 1); - if ((params.guest_len + offset > PAGE_SIZE)) + if (params.guest_len > PAGE_SIZE || (params.guest_len + offset) > PAGE_SIZE) return -EINVAL; hdr = psp_copy_user_blob(params.hdr_uaddr, params.hdr_len); @@ -1463,17 +1850,25 @@ static int sev_receive_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp) data.trans_len = params.trans_len; /* Pin guest memory */ - ret = -EFAULT; guest_page = sev_pin_memory(kvm, params.guest_uaddr & PAGE_MASK, - PAGE_SIZE, &n, 0); - if (!guest_page) + PAGE_SIZE, &n, FOLL_WRITE); + if (IS_ERR(guest_page)) { + ret = PTR_ERR(guest_page); goto e_free_trans; + } + + /* + * Flush (on non-coherent CPUs) before RECEIVE_UPDATE_DATA, the PSP + * encrypts the written data with the guest's key, and the cache may + * contain dirty, unencrypted data. + */ + sev_clflush_pages(guest_page, n); /* The RECEIVE_UPDATE_DATA command requires C-bit to be always set. */ data.guest_address = (page_to_pfn(guest_page[0]) << PAGE_SHIFT) + offset; data.guest_address |= sev_me_mask; data.guest_len = params.guest_len; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; ret = sev_issue_cmd(kvm, SEV_CMD_RECEIVE_UPDATE_DATA, &data, &argp->error); @@ -1490,17 +1885,689 @@ e_free_hdr: static int sev_receive_finish(struct kvm *kvm, struct kvm_sev_cmd *argp) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; struct sev_data_receive_finish data; if (!sev_guest(kvm)) return -ENOTTY; - data.handle = sev->handle; + data.handle = to_kvm_sev_info(kvm)->handle; return sev_issue_cmd(kvm, SEV_CMD_RECEIVE_FINISH, &data, &argp->error); } -int svm_mem_enc_op(struct kvm *kvm, void __user *argp) +static bool is_cmd_allowed_from_mirror(u32 cmd_id) +{ + /* + * Allow mirrors VM to call KVM_SEV_LAUNCH_UPDATE_VMSA to enable SEV-ES + * active mirror VMs. Also allow the debugging and status commands. + */ + if (cmd_id == KVM_SEV_LAUNCH_UPDATE_VMSA || + cmd_id == KVM_SEV_GUEST_STATUS || cmd_id == KVM_SEV_DBG_DECRYPT || + cmd_id == KVM_SEV_DBG_ENCRYPT) + return true; + + return false; +} + +static int sev_lock_two_vms(struct kvm *dst_kvm, struct kvm *src_kvm) +{ + struct kvm_sev_info *dst_sev = to_kvm_sev_info(dst_kvm); + struct kvm_sev_info *src_sev = to_kvm_sev_info(src_kvm); + int r = -EBUSY; + + if (dst_kvm == src_kvm) + return -EINVAL; + + /* + * Bail if these VMs are already involved in a migration to avoid + * deadlock between two VMs trying to migrate to/from each other. + */ + if (atomic_cmpxchg_acquire(&dst_sev->migration_in_progress, 0, 1)) + return -EBUSY; + + if (atomic_cmpxchg_acquire(&src_sev->migration_in_progress, 0, 1)) + goto release_dst; + + r = -EINTR; + if (mutex_lock_killable(&dst_kvm->lock)) + goto release_src; + if (mutex_lock_killable_nested(&src_kvm->lock, SINGLE_DEPTH_NESTING)) + goto unlock_dst; + return 0; + +unlock_dst: + mutex_unlock(&dst_kvm->lock); +release_src: + atomic_set_release(&src_sev->migration_in_progress, 0); +release_dst: + atomic_set_release(&dst_sev->migration_in_progress, 0); + return r; +} + +static void sev_unlock_two_vms(struct kvm *dst_kvm, struct kvm *src_kvm) +{ + struct kvm_sev_info *dst_sev = to_kvm_sev_info(dst_kvm); + struct kvm_sev_info *src_sev = to_kvm_sev_info(src_kvm); + + mutex_unlock(&dst_kvm->lock); + mutex_unlock(&src_kvm->lock); + atomic_set_release(&dst_sev->migration_in_progress, 0); + atomic_set_release(&src_sev->migration_in_progress, 0); +} + +static void sev_migrate_from(struct kvm *dst_kvm, struct kvm *src_kvm) +{ + struct kvm_sev_info *dst = to_kvm_sev_info(dst_kvm); + struct kvm_sev_info *src = to_kvm_sev_info(src_kvm); + struct kvm_vcpu *dst_vcpu, *src_vcpu; + struct vcpu_svm *dst_svm, *src_svm; + struct kvm_sev_info *mirror; + unsigned long i; + + dst->active = true; + dst->asid = src->asid; + dst->handle = src->handle; + dst->pages_locked = src->pages_locked; + dst->enc_context_owner = src->enc_context_owner; + dst->es_active = src->es_active; + dst->vmsa_features = src->vmsa_features; + + src->asid = 0; + src->active = false; + src->handle = 0; + src->pages_locked = 0; + src->enc_context_owner = NULL; + src->es_active = false; + + list_cut_before(&dst->regions_list, &src->regions_list, &src->regions_list); + + /* + * If this VM has mirrors, "transfer" each mirror's refcount of the + * source to the destination (this KVM). The caller holds a reference + * to the source, so there's no danger of use-after-free. + */ + list_cut_before(&dst->mirror_vms, &src->mirror_vms, &src->mirror_vms); + list_for_each_entry(mirror, &dst->mirror_vms, mirror_entry) { + kvm_get_kvm(dst_kvm); + kvm_put_kvm(src_kvm); + mirror->enc_context_owner = dst_kvm; + } + + /* + * If this VM is a mirror, remove the old mirror from the owners list + * and add the new mirror to the list. + */ + if (is_mirroring_enc_context(dst_kvm)) { + struct kvm_sev_info *owner_sev_info = to_kvm_sev_info(dst->enc_context_owner); + + list_del(&src->mirror_entry); + list_add_tail(&dst->mirror_entry, &owner_sev_info->mirror_vms); + } + + kvm_for_each_vcpu(i, dst_vcpu, dst_kvm) { + dst_svm = to_svm(dst_vcpu); + + sev_init_vmcb(dst_svm, false); + + if (!dst->es_active) + continue; + + /* + * Note, the source is not required to have the same number of + * vCPUs as the destination when migrating a vanilla SEV VM. + */ + src_vcpu = kvm_get_vcpu(src_kvm, i); + src_svm = to_svm(src_vcpu); + + /* + * Transfer VMSA and GHCB state to the destination. Nullify and + * clear source fields as appropriate, the state now belongs to + * the destination. + */ + memcpy(&dst_svm->sev_es, &src_svm->sev_es, sizeof(src_svm->sev_es)); + dst_svm->vmcb->control.ghcb_gpa = src_svm->vmcb->control.ghcb_gpa; + dst_svm->vmcb->control.vmsa_pa = src_svm->vmcb->control.vmsa_pa; + dst_vcpu->arch.guest_state_protected = true; + + memset(&src_svm->sev_es, 0, sizeof(src_svm->sev_es)); + src_svm->vmcb->control.ghcb_gpa = INVALID_PAGE; + src_svm->vmcb->control.vmsa_pa = INVALID_PAGE; + src_vcpu->arch.guest_state_protected = false; + } +} + +static int sev_check_source_vcpus(struct kvm *dst, struct kvm *src) +{ + struct kvm_vcpu *src_vcpu; + unsigned long i; + + if (src->created_vcpus != atomic_read(&src->online_vcpus) || + dst->created_vcpus != atomic_read(&dst->online_vcpus)) + return -EBUSY; + + if (!sev_es_guest(src)) + return 0; + + if (atomic_read(&src->online_vcpus) != atomic_read(&dst->online_vcpus)) + return -EINVAL; + + kvm_for_each_vcpu(i, src_vcpu, src) { + if (!src_vcpu->arch.guest_state_protected) + return -EINVAL; + } + + return 0; +} + +int sev_vm_move_enc_context_from(struct kvm *kvm, unsigned int source_fd) +{ + struct kvm_sev_info *dst_sev = to_kvm_sev_info(kvm); + struct kvm_sev_info *src_sev, *cg_cleanup_sev; + CLASS(fd, f)(source_fd); + struct kvm *source_kvm; + bool charged = false; + int ret; + + if (fd_empty(f)) + return -EBADF; + + if (!file_is_kvm(fd_file(f))) + return -EBADF; + + source_kvm = fd_file(f)->private_data; + ret = sev_lock_two_vms(kvm, source_kvm); + if (ret) + return ret; + + if (kvm->arch.vm_type != source_kvm->arch.vm_type || + sev_guest(kvm) || !sev_guest(source_kvm)) { + ret = -EINVAL; + goto out_unlock; + } + + src_sev = to_kvm_sev_info(source_kvm); + + dst_sev->misc_cg = get_current_misc_cg(); + cg_cleanup_sev = dst_sev; + if (dst_sev->misc_cg != src_sev->misc_cg) { + ret = sev_misc_cg_try_charge(dst_sev); + if (ret) + goto out_dst_cgroup; + charged = true; + } + + ret = kvm_lock_all_vcpus(kvm); + if (ret) + goto out_dst_cgroup; + ret = kvm_lock_all_vcpus(source_kvm); + if (ret) + goto out_dst_vcpu; + + ret = sev_check_source_vcpus(kvm, source_kvm); + if (ret) + goto out_source_vcpu; + + /* + * Allocate a new have_run_cpus for the destination, i.e. don't copy + * the set of CPUs from the source. If a CPU was used to run a vCPU in + * the source VM but is never used for the destination VM, then the CPU + * can only have cached memory that was accessible to the source VM. + */ + if (!zalloc_cpumask_var(&dst_sev->have_run_cpus, GFP_KERNEL_ACCOUNT)) { + ret = -ENOMEM; + goto out_source_vcpu; + } + + sev_migrate_from(kvm, source_kvm); + kvm_vm_dead(source_kvm); + cg_cleanup_sev = src_sev; + ret = 0; + +out_source_vcpu: + kvm_unlock_all_vcpus(source_kvm); +out_dst_vcpu: + kvm_unlock_all_vcpus(kvm); +out_dst_cgroup: + /* Operates on the source on success, on the destination on failure. */ + if (charged) + sev_misc_cg_uncharge(cg_cleanup_sev); + put_misc_cg(cg_cleanup_sev->misc_cg); + cg_cleanup_sev->misc_cg = NULL; +out_unlock: + sev_unlock_two_vms(kvm, source_kvm); + return ret; +} + +int sev_dev_get_attr(u32 group, u64 attr, u64 *val) +{ + if (group != KVM_X86_GRP_SEV) + return -ENXIO; + + switch (attr) { + case KVM_X86_SEV_VMSA_FEATURES: + *val = sev_supported_vmsa_features; + return 0; + + case KVM_X86_SNP_POLICY_BITS: + *val = snp_supported_policy_bits; + return 0; + + default: + return -ENXIO; + } +} + +/* + * The guest context contains all the information, keys and metadata + * associated with the guest that the firmware tracks to implement SEV + * and SNP features. The firmware stores the guest context in hypervisor + * provide page via the SNP_GCTX_CREATE command. + */ +static void *snp_context_create(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct sev_data_snp_addr data = {}; + void *context; + int rc; + + /* Allocate memory for context page */ + context = snp_alloc_firmware_page(GFP_KERNEL_ACCOUNT); + if (!context) + return NULL; + + data.address = __psp_pa(context); + rc = __sev_issue_cmd(argp->sev_fd, SEV_CMD_SNP_GCTX_CREATE, &data, &argp->error); + if (rc) { + pr_warn("Failed to create SEV-SNP context, rc %d fw_error %d", + rc, argp->error); + snp_free_firmware_page(context); + return NULL; + } + + return context; +} + +static int snp_bind_asid(struct kvm *kvm, int *error) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct sev_data_snp_activate data = {0}; + + data.gctx_paddr = __psp_pa(sev->snp_context); + data.asid = sev_get_asid(kvm); + return sev_issue_cmd(kvm, SEV_CMD_SNP_ACTIVATE, &data, error); +} + +static int snp_launch_start(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct sev_data_snp_launch_start start = {0}; + struct kvm_sev_snp_launch_start params; + int rc; + + if (!sev_snp_guest(kvm)) + return -ENOTTY; + + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) + return -EFAULT; + + /* Don't allow userspace to allocate memory for more than 1 SNP context. */ + if (sev->snp_context) + return -EINVAL; + + if (params.flags) + return -EINVAL; + + if (params.policy & ~snp_supported_policy_bits) + return -EINVAL; + + /* Check for policy bits that must be set */ + if (!(params.policy & SNP_POLICY_MASK_RSVD_MBO)) + return -EINVAL; + + if (snp_is_secure_tsc_enabled(kvm)) { + if (WARN_ON_ONCE(!kvm->arch.default_tsc_khz)) + return -EINVAL; + + start.desired_tsc_khz = kvm->arch.default_tsc_khz; + } + + sev->snp_context = snp_context_create(kvm, argp); + if (!sev->snp_context) + return -ENOTTY; + + start.gctx_paddr = __psp_pa(sev->snp_context); + start.policy = params.policy; + + memcpy(start.gosvw, params.gosvw, sizeof(params.gosvw)); + rc = __sev_issue_cmd(argp->sev_fd, SEV_CMD_SNP_LAUNCH_START, &start, &argp->error); + if (rc) { + pr_debug("%s: SEV_CMD_SNP_LAUNCH_START firmware command failed, rc %d\n", + __func__, rc); + goto e_free_context; + } + + sev->policy = params.policy; + sev->fd = argp->sev_fd; + rc = snp_bind_asid(kvm, &argp->error); + if (rc) { + pr_debug("%s: Failed to bind ASID to SEV-SNP context, rc %d\n", + __func__, rc); + goto e_free_context; + } + + return 0; + +e_free_context: + snp_decommission_context(kvm); + + return rc; +} + +struct sev_gmem_populate_args { + __u8 type; + int sev_fd; + int fw_error; +}; + +static int sev_gmem_post_populate(struct kvm *kvm, gfn_t gfn_start, kvm_pfn_t pfn, + void __user *src, int order, void *opaque) +{ + struct sev_gmem_populate_args *sev_populate_args = opaque; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + int n_private = 0, ret, i; + int npages = (1 << order); + gfn_t gfn; + + if (WARN_ON_ONCE(sev_populate_args->type != KVM_SEV_SNP_PAGE_TYPE_ZERO && !src)) + return -EINVAL; + + for (gfn = gfn_start, i = 0; gfn < gfn_start + npages; gfn++, i++) { + struct sev_data_snp_launch_update fw_args = {0}; + bool assigned = false; + int level; + + ret = snp_lookup_rmpentry((u64)pfn + i, &assigned, &level); + if (ret || assigned) { + pr_debug("%s: Failed to ensure GFN 0x%llx RMP entry is initial shared state, ret: %d assigned: %d\n", + __func__, gfn, ret, assigned); + ret = ret ? -EINVAL : -EEXIST; + goto err; + } + + if (src) { + void *vaddr = kmap_local_pfn(pfn + i); + + if (copy_from_user(vaddr, src + i * PAGE_SIZE, PAGE_SIZE)) { + ret = -EFAULT; + goto err; + } + kunmap_local(vaddr); + } + + ret = rmp_make_private(pfn + i, gfn << PAGE_SHIFT, PG_LEVEL_4K, + sev_get_asid(kvm), true); + if (ret) + goto err; + + n_private++; + + fw_args.gctx_paddr = __psp_pa(sev->snp_context); + fw_args.address = __sme_set(pfn_to_hpa(pfn + i)); + fw_args.page_size = PG_LEVEL_TO_RMP(PG_LEVEL_4K); + fw_args.page_type = sev_populate_args->type; + + ret = __sev_issue_cmd(sev_populate_args->sev_fd, SEV_CMD_SNP_LAUNCH_UPDATE, + &fw_args, &sev_populate_args->fw_error); + if (ret) + goto fw_err; + } + + return 0; + +fw_err: + /* + * If the firmware command failed handle the reclaim and cleanup of that + * PFN specially vs. prior pages which can be cleaned up below without + * needing to reclaim in advance. + * + * Additionally, when invalid CPUID function entries are detected, + * firmware writes the expected values into the page and leaves it + * unencrypted so it can be used for debugging and error-reporting. + * + * Copy this page back into the source buffer so userspace can use this + * information to provide information on which CPUID leaves/fields + * failed CPUID validation. + */ + if (!snp_page_reclaim(kvm, pfn + i) && + sev_populate_args->type == KVM_SEV_SNP_PAGE_TYPE_CPUID && + sev_populate_args->fw_error == SEV_RET_INVALID_PARAM) { + void *vaddr = kmap_local_pfn(pfn + i); + + if (copy_to_user(src + i * PAGE_SIZE, vaddr, PAGE_SIZE)) + pr_debug("Failed to write CPUID page back to userspace\n"); + + kunmap_local(vaddr); + } + + /* pfn + i is hypervisor-owned now, so skip below cleanup for it. */ + n_private--; + +err: + pr_debug("%s: exiting with error ret %d (fw_error %d), restoring %d gmem PFNs to shared.\n", + __func__, ret, sev_populate_args->fw_error, n_private); + for (i = 0; i < n_private; i++) + kvm_rmp_make_shared(kvm, pfn + i, PG_LEVEL_4K); + + return ret; +} + +static int snp_launch_update(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct sev_gmem_populate_args sev_populate_args = {0}; + struct kvm_sev_snp_launch_update params; + struct kvm_memory_slot *memslot; + long npages, count; + void __user *src; + int ret = 0; + + if (!sev_snp_guest(kvm) || !sev->snp_context) + return -EINVAL; + + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) + return -EFAULT; + + pr_debug("%s: GFN start 0x%llx length 0x%llx type %d flags %d\n", __func__, + params.gfn_start, params.len, params.type, params.flags); + + if (!params.len || !PAGE_ALIGNED(params.len) || params.flags || + (params.type != KVM_SEV_SNP_PAGE_TYPE_NORMAL && + params.type != KVM_SEV_SNP_PAGE_TYPE_ZERO && + params.type != KVM_SEV_SNP_PAGE_TYPE_UNMEASURED && + params.type != KVM_SEV_SNP_PAGE_TYPE_SECRETS && + params.type != KVM_SEV_SNP_PAGE_TYPE_CPUID)) + return -EINVAL; + + npages = params.len / PAGE_SIZE; + + /* + * For each GFN that's being prepared as part of the initial guest + * state, the following pre-conditions are verified: + * + * 1) The backing memslot is a valid private memslot. + * 2) The GFN has been set to private via KVM_SET_MEMORY_ATTRIBUTES + * beforehand. + * 3) The PFN of the guest_memfd has not already been set to private + * in the RMP table. + * + * The KVM MMU relies on kvm->mmu_invalidate_seq to retry nested page + * faults if there's a race between a fault and an attribute update via + * KVM_SET_MEMORY_ATTRIBUTES, and a similar approach could be utilized + * here. However, kvm->slots_lock guards against both this as well as + * concurrent memslot updates occurring while these checks are being + * performed, so use that here to make it easier to reason about the + * initial expected state and better guard against unexpected + * situations. + */ + mutex_lock(&kvm->slots_lock); + + memslot = gfn_to_memslot(kvm, params.gfn_start); + if (!kvm_slot_has_gmem(memslot)) { + ret = -EINVAL; + goto out; + } + + sev_populate_args.sev_fd = argp->sev_fd; + sev_populate_args.type = params.type; + src = params.type == KVM_SEV_SNP_PAGE_TYPE_ZERO ? NULL : u64_to_user_ptr(params.uaddr); + + count = kvm_gmem_populate(kvm, params.gfn_start, src, npages, + sev_gmem_post_populate, &sev_populate_args); + if (count < 0) { + argp->error = sev_populate_args.fw_error; + pr_debug("%s: kvm_gmem_populate failed, ret %ld (fw_error %d)\n", + __func__, count, argp->error); + ret = -EIO; + } else { + params.gfn_start += count; + params.len -= count * PAGE_SIZE; + if (params.type != KVM_SEV_SNP_PAGE_TYPE_ZERO) + params.uaddr += count * PAGE_SIZE; + + ret = 0; + if (copy_to_user(u64_to_user_ptr(argp->data), ¶ms, sizeof(params))) + ret = -EFAULT; + } + +out: + mutex_unlock(&kvm->slots_lock); + + return ret; +} + +static int snp_launch_update_vmsa(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct sev_data_snp_launch_update data = {}; + struct kvm_vcpu *vcpu; + unsigned long i; + int ret; + + data.gctx_paddr = __psp_pa(sev->snp_context); + data.page_type = SNP_PAGE_TYPE_VMSA; + + kvm_for_each_vcpu(i, vcpu, kvm) { + struct vcpu_svm *svm = to_svm(vcpu); + u64 pfn = __pa(svm->sev_es.vmsa) >> PAGE_SHIFT; + + ret = sev_es_sync_vmsa(svm); + if (ret) + return ret; + + /* Transition the VMSA page to a firmware state. */ + ret = rmp_make_private(pfn, INITIAL_VMSA_GPA, PG_LEVEL_4K, sev->asid, true); + if (ret) + return ret; + + /* Issue the SNP command to encrypt the VMSA */ + data.address = __sme_pa(svm->sev_es.vmsa); + ret = __sev_issue_cmd(argp->sev_fd, SEV_CMD_SNP_LAUNCH_UPDATE, + &data, &argp->error); + if (ret) { + snp_page_reclaim(kvm, pfn); + + return ret; + } + + svm->vcpu.arch.guest_state_protected = true; + /* + * SEV-ES (and thus SNP) guest mandates LBR Virtualization to + * be _always_ ON. Enable it only after setting + * guest_state_protected because KVM_SET_MSRS allows dynamic + * toggling of LBRV (for performance reason) on write access to + * MSR_IA32_DEBUGCTLMSR when guest_state_protected is not set. + */ + svm_enable_lbrv(vcpu); + } + + return 0; +} + +static int snp_launch_finish(struct kvm *kvm, struct kvm_sev_cmd *argp) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct kvm_sev_snp_launch_finish params; + struct sev_data_snp_launch_finish *data; + void *id_block = NULL, *id_auth = NULL; + int ret; + + if (!sev_snp_guest(kvm)) + return -ENOTTY; + + if (!sev->snp_context) + return -EINVAL; + + if (copy_from_user(¶ms, u64_to_user_ptr(argp->data), sizeof(params))) + return -EFAULT; + + if (params.flags) + return -EINVAL; + + /* Measure all vCPUs using LAUNCH_UPDATE before finalizing the launch flow. */ + ret = snp_launch_update_vmsa(kvm, argp); + if (ret) + return ret; + + data = kzalloc(sizeof(*data), GFP_KERNEL_ACCOUNT); + if (!data) + return -ENOMEM; + + if (params.id_block_en) { + id_block = psp_copy_user_blob(params.id_block_uaddr, KVM_SEV_SNP_ID_BLOCK_SIZE); + if (IS_ERR(id_block)) { + ret = PTR_ERR(id_block); + goto e_free; + } + + data->id_block_en = 1; + data->id_block_paddr = __sme_pa(id_block); + + id_auth = psp_copy_user_blob(params.id_auth_uaddr, KVM_SEV_SNP_ID_AUTH_SIZE); + if (IS_ERR(id_auth)) { + ret = PTR_ERR(id_auth); + goto e_free_id_block; + } + + data->id_auth_paddr = __sme_pa(id_auth); + + if (params.auth_key_en) + data->auth_key_en = 1; + } + + data->vcek_disabled = params.vcek_disabled; + + memcpy(data->host_data, params.host_data, KVM_SEV_SNP_FINISH_DATA_SIZE); + data->gctx_paddr = __psp_pa(sev->snp_context); + ret = sev_issue_cmd(kvm, SEV_CMD_SNP_LAUNCH_FINISH, data, &argp->error); + + /* + * Now that there will be no more SNP_LAUNCH_UPDATE ioctls, private pages + * can be given to the guest simply by marking the RMP entry as private. + * This can happen on first access and also with KVM_PRE_FAULT_MEMORY. + */ + if (!ret) + kvm->arch.pre_fault_allowed = true; + + kfree(id_auth); + +e_free_id_block: + kfree(id_block); + +e_free: + kfree(data); + + return ret; +} + +int sev_mem_enc_ioctl(struct kvm *kvm, void __user *argp) { struct kvm_sev_cmd sev_cmd; int r; @@ -1516,12 +2583,22 @@ int svm_mem_enc_op(struct kvm *kvm, void __user *argp) mutex_lock(&kvm->lock); - /* enc_context_owner handles all memory enc operations */ - if (is_mirroring_enc_context(kvm)) { + /* Only the enc_context_owner handles some memory enc operations. */ + if (is_mirroring_enc_context(kvm) && + !is_cmd_allowed_from_mirror(sev_cmd.id)) { r = -EINVAL; goto out; } + /* + * Once KVM_SEV_INIT2 initializes a KVM instance as an SNP guest, only + * allow the use of SNP-specific commands. + */ + if (sev_snp_guest(kvm) && sev_cmd.id < KVM_SEV_SNP_LAUNCH_START) { + r = -EPERM; + goto out; + } + switch (sev_cmd.id) { case KVM_SEV_ES_INIT: if (!sev_es_enabled) { @@ -1532,6 +2609,9 @@ int svm_mem_enc_op(struct kvm *kvm, void __user *argp) case KVM_SEV_INIT: r = sev_guest_init(kvm, &sev_cmd); break; + case KVM_SEV_INIT2: + r = sev_guest_init2(kvm, &sev_cmd); + break; case KVM_SEV_LAUNCH_START: r = sev_launch_start(kvm, &sev_cmd); break; @@ -1583,6 +2663,15 @@ int svm_mem_enc_op(struct kvm *kvm, void __user *argp) case KVM_SEV_RECEIVE_FINISH: r = sev_receive_finish(kvm, &sev_cmd); break; + case KVM_SEV_SNP_LAUNCH_START: + r = snp_launch_start(kvm, &sev_cmd); + break; + case KVM_SEV_SNP_LAUNCH_UPDATE: + r = snp_launch_update(kvm, &sev_cmd); + break; + case KVM_SEV_SNP_LAUNCH_FINISH: + r = snp_launch_finish(kvm, &sev_cmd); + break; default: r = -EINVAL; goto out; @@ -1596,10 +2685,10 @@ out: return r; } -int svm_register_enc_region(struct kvm *kvm, - struct kvm_enc_region *range) +int sev_mem_enc_register_region(struct kvm *kvm, + struct kvm_enc_region *range) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); struct enc_region *region; int ret = 0; @@ -1618,27 +2707,30 @@ int svm_register_enc_region(struct kvm *kvm, return -ENOMEM; mutex_lock(&kvm->lock); - region->pages = sev_pin_memory(kvm, range->addr, range->size, ®ion->npages, 1); + region->pages = sev_pin_memory(kvm, range->addr, range->size, ®ion->npages, + FOLL_WRITE | FOLL_LONGTERM); if (IS_ERR(region->pages)) { ret = PTR_ERR(region->pages); mutex_unlock(&kvm->lock); goto e_free; } - region->uaddr = range->addr; - region->size = range->size; - - list_add_tail(®ion->list, &sev->regions_list); - mutex_unlock(&kvm->lock); - /* * The guest may change the memory encryption attribute from C=0 -> C=1 * or vice versa for this memory range. Lets make sure caches are * flushed to ensure that guest data gets written into memory with - * correct C-bit. + * correct C-bit. Note, this must be done before dropping kvm->lock, + * as region and its array of pages can be freed by a different task + * once kvm->lock is released. */ sev_clflush_pages(region->pages, region->npages); + region->uaddr = range->addr; + region->size = range->size; + + list_add_tail(®ion->list, &sev->regions_list); + mutex_unlock(&kvm->lock); + return ret; e_free: @@ -1649,7 +2741,7 @@ e_free: static struct enc_region * find_enc_region(struct kvm *kvm, struct kvm_enc_region *range) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); struct list_head *head = &sev->regions_list; struct enc_region *i; @@ -1670,8 +2762,8 @@ static void __unregister_enc_region_locked(struct kvm *kvm, kfree(region); } -int svm_unregister_enc_region(struct kvm *kvm, - struct kvm_enc_region *range) +int sev_mem_enc_unregister_region(struct kvm *kvm, + struct kvm_enc_region *range) { struct enc_region *region; int ret; @@ -1693,12 +2785,7 @@ int svm_unregister_enc_region(struct kvm *kvm, goto failed; } - /* - * Ensure that all guest tagged cache entries are flushed before - * releasing the pages back to the system for use. CLFLUSH will - * not do this, so issue a WBINVD. - */ - wbinvd_on_all_cpus(); + sev_writeback_caches(kvm); __unregister_enc_region_locked(kvm, region); @@ -1710,95 +2797,126 @@ failed: return ret; } -int svm_vm_copy_asid_from(struct kvm *kvm, unsigned int source_fd) +int sev_vm_copy_enc_context_from(struct kvm *kvm, unsigned int source_fd) { - struct file *source_kvm_file; + CLASS(fd, f)(source_fd); struct kvm *source_kvm; - struct kvm_sev_info *mirror_sev; - unsigned int asid; + struct kvm_sev_info *source_sev, *mirror_sev; int ret; - source_kvm_file = fget(source_fd); - if (!file_is_kvm(source_kvm_file)) { - ret = -EBADF; - goto e_source_put; - } + if (fd_empty(f)) + return -EBADF; - source_kvm = source_kvm_file->private_data; - mutex_lock(&source_kvm->lock); + if (!file_is_kvm(fd_file(f))) + return -EBADF; - if (!sev_guest(source_kvm)) { - ret = -EINVAL; - goto e_source_unlock; - } + source_kvm = fd_file(f)->private_data; + ret = sev_lock_two_vms(kvm, source_kvm); + if (ret) + return ret; - /* Mirrors of mirrors should work, but let's not get silly */ - if (is_mirroring_enc_context(source_kvm) || source_kvm == kvm) { + /* + * Mirrors of mirrors should work, but let's not get silly. Also + * disallow out-of-band SEV/SEV-ES init if the target is already an + * SEV guest, or if vCPUs have been created. KVM relies on vCPUs being + * created after SEV/SEV-ES initialization, e.g. to init intercepts. + */ + if (sev_guest(kvm) || !sev_guest(source_kvm) || + is_mirroring_enc_context(source_kvm) || kvm->created_vcpus) { ret = -EINVAL; - goto e_source_unlock; + goto e_unlock; } - asid = to_kvm_svm(source_kvm)->sev_info.asid; + mirror_sev = to_kvm_sev_info(kvm); + if (!zalloc_cpumask_var(&mirror_sev->have_run_cpus, GFP_KERNEL_ACCOUNT)) { + ret = -ENOMEM; + goto e_unlock; + } /* * The mirror kvm holds an enc_context_owner ref so its asid can't * disappear until we're done with it */ + source_sev = to_kvm_sev_info(source_kvm); kvm_get_kvm(source_kvm); - - fput(source_kvm_file); - mutex_unlock(&source_kvm->lock); - mutex_lock(&kvm->lock); - - if (sev_guest(kvm)) { - ret = -EINVAL; - goto e_mirror_unlock; - } + list_add_tail(&mirror_sev->mirror_entry, &source_sev->mirror_vms); /* Set enc_context_owner and copy its encryption context over */ - mirror_sev = &to_kvm_svm(kvm)->sev_info; mirror_sev->enc_context_owner = source_kvm; - mirror_sev->asid = asid; mirror_sev->active = true; + mirror_sev->asid = source_sev->asid; + mirror_sev->fd = source_sev->fd; + mirror_sev->es_active = source_sev->es_active; + mirror_sev->need_init = false; + mirror_sev->handle = source_sev->handle; + INIT_LIST_HEAD(&mirror_sev->regions_list); + INIT_LIST_HEAD(&mirror_sev->mirror_vms); + ret = 0; - mutex_unlock(&kvm->lock); - return 0; + /* + * Do not copy ap_jump_table. Since the mirror does not share the same + * KVM contexts as the original, and they may have different + * memory-views. + */ -e_mirror_unlock: - mutex_unlock(&kvm->lock); - kvm_put_kvm(source_kvm); - return ret; -e_source_unlock: - mutex_unlock(&source_kvm->lock); -e_source_put: - if (source_kvm_file) - fput(source_kvm_file); +e_unlock: + sev_unlock_two_vms(kvm, source_kvm); return ret; } +static int snp_decommission_context(struct kvm *kvm) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + struct sev_data_snp_addr data = {}; + int ret; + + /* If context is not created then do nothing */ + if (!sev->snp_context) + return 0; + + /* Do the decommision, which will unbind the ASID from the SNP context */ + data.address = __sme_pa(sev->snp_context); + down_write(&sev_deactivate_lock); + ret = sev_do_cmd(SEV_CMD_SNP_DECOMMISSION, &data, NULL); + up_write(&sev_deactivate_lock); + + if (WARN_ONCE(ret, "Failed to release guest context, ret %d", ret)) + return ret; + + snp_free_firmware_page(sev->snp_context); + sev->snp_context = NULL; + + return 0; +} + void sev_vm_destroy(struct kvm *kvm) { - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); struct list_head *head = &sev->regions_list; struct list_head *pos, *q; if (!sev_guest(kvm)) return; - /* If this is a mirror_kvm release the enc_context_owner and skip sev cleanup */ - if (is_mirroring_enc_context(kvm)) { - kvm_put_kvm(sev->enc_context_owner); - return; - } + WARN_ON(!list_empty(&sev->mirror_vms)); - mutex_lock(&kvm->lock); + free_cpumask_var(sev->have_run_cpus); /* - * Ensure that all guest tagged cache entries are flushed before - * releasing the pages back to the system for use. CLFLUSH will - * not do this, so issue a WBINVD. + * If this is a mirror VM, remove it from the owner's list of a mirrors + * and skip ASID cleanup (the ASID is tied to the lifetime of the owner). + * Note, mirror VMs don't support registering encrypted regions. */ - wbinvd_on_all_cpus(); + if (is_mirroring_enc_context(kvm)) { + struct kvm *owner_kvm = sev->enc_context_owner; + + mutex_lock(&owner_kvm->lock); + list_del(&sev->mirror_entry); + mutex_unlock(&owner_kvm->lock); + kvm_put_kvm(owner_kvm); + return; + } + /* * if userspace was terminated before unregistering the memory regions @@ -1812,32 +2930,96 @@ void sev_vm_destroy(struct kvm *kvm) } } - mutex_unlock(&kvm->lock); + if (sev_snp_guest(kvm)) { + snp_guest_req_cleanup(kvm); + + /* + * Decomission handles unbinding of the ASID. If it fails for + * some unexpected reason, just leak the ASID. + */ + if (snp_decommission_context(kvm)) + return; + } else { + sev_unbind_asid(kvm, sev->handle); + } - sev_unbind_asid(kvm, sev->handle); sev_asid_free(sev); } void __init sev_set_cpu_caps(void) { - if (!sev_enabled) - kvm_cpu_cap_clear(X86_FEATURE_SEV); - if (!sev_es_enabled) - kvm_cpu_cap_clear(X86_FEATURE_SEV_ES); + if (sev_enabled) { + kvm_cpu_cap_set(X86_FEATURE_SEV); + kvm_caps.supported_vm_types |= BIT(KVM_X86_SEV_VM); + } + if (sev_es_enabled) { + kvm_cpu_cap_set(X86_FEATURE_SEV_ES); + kvm_caps.supported_vm_types |= BIT(KVM_X86_SEV_ES_VM); + } + if (sev_snp_enabled) { + kvm_cpu_cap_set(X86_FEATURE_SEV_SNP); + kvm_caps.supported_vm_types |= BIT(KVM_X86_SNP_VM); + } +} + +static bool is_sev_snp_initialized(void) +{ + struct sev_user_data_snp_status *status; + struct sev_data_snp_addr buf; + bool initialized = false; + int ret, error = 0; + + status = snp_alloc_firmware_page(GFP_KERNEL | __GFP_ZERO); + if (!status) + return false; + + buf.address = __psp_pa(status); + ret = sev_do_cmd(SEV_CMD_SNP_PLATFORM_STATUS, &buf, &error); + if (ret) { + pr_err("SEV: SNP_PLATFORM_STATUS failed ret=%d, fw_error=%d (%#x)\n", + ret, error, error); + goto out; + } + + initialized = !!status->state; + +out: + snp_free_firmware_page(status); + + return initialized; } void __init sev_hardware_setup(void) { -#ifdef CONFIG_KVM_AMD_SEV unsigned int eax, ebx, ecx, edx, sev_asid_count, sev_es_asid_count; + struct sev_platform_init_args init_args = {0}; + bool sev_snp_supported = false; bool sev_es_supported = false; bool sev_supported = false; - if (!sev_enabled || !npt_enabled) + if (!sev_enabled || !npt_enabled || !nrips) goto out; - /* Does the CPU support SEV? */ - if (!boot_cpu_has(X86_FEATURE_SEV)) + /* + * SEV must obviously be supported in hardware. Sanity check that the + * CPU supports decode assists, which is mandatory for SEV guests to + * support instruction emulation. Ditto for flushing by ASID, as SEV + * guests are bound to a single ASID, i.e. KVM can't rotate to a new + * ASID to effect a TLB flush. + */ + if (!boot_cpu_has(X86_FEATURE_SEV) || + WARN_ON_ONCE(!boot_cpu_has(X86_FEATURE_DECODEASSISTS)) || + WARN_ON_ONCE(!boot_cpu_has(X86_FEATURE_FLUSHBYASID))) + goto out; + + /* + * The kernel's initcall infrastructure lacks the ability to express + * dependencies between initcalls, whereas the modules infrastructure + * automatically handles dependencies via symbol loading. Ensure the + * PSP SEV driver is initialized before proceeding if KVM is built-in, + * as the dependency isn't handled by the initcall infrastructure. + */ + if (IS_BUILTIN(CONFIG_KVM_AMD) && sev_module_init()) goto out; /* Retrieve SEV CPUID information */ @@ -1855,63 +3037,146 @@ void __init sev_hardware_setup(void) min_sev_asid = edx; sev_me_mask = 1UL << (ebx & 0x3f); - /* Initialize SEV ASID bitmaps */ - sev_asid_bitmap = bitmap_zalloc(max_sev_asid, GFP_KERNEL); + /* + * Initialize SEV ASID bitmaps. Allocate space for ASID 0 in the bitmap, + * even though it's never used, so that the bitmap is indexed by the + * actual ASID. + */ + nr_asids = max_sev_asid + 1; + sev_asid_bitmap = bitmap_zalloc(nr_asids, GFP_KERNEL); if (!sev_asid_bitmap) goto out; - sev_reclaim_asid_bitmap = bitmap_zalloc(max_sev_asid, GFP_KERNEL); + sev_reclaim_asid_bitmap = bitmap_zalloc(nr_asids, GFP_KERNEL); if (!sev_reclaim_asid_bitmap) { bitmap_free(sev_asid_bitmap); sev_asid_bitmap = NULL; goto out; } - sev_asid_count = max_sev_asid - min_sev_asid + 1; - if (misc_cg_set_capacity(MISC_CG_RES_SEV, sev_asid_count)) - goto out; - - pr_info("SEV supported: %u ASIDs\n", sev_asid_count); + if (min_sev_asid <= max_sev_asid) { + sev_asid_count = max_sev_asid - min_sev_asid + 1; + WARN_ON_ONCE(misc_cg_set_capacity(MISC_CG_RES_SEV, sev_asid_count)); + } sev_supported = true; /* SEV-ES support requested? */ if (!sev_es_enabled) goto out; + /* + * SEV-ES requires MMIO caching as KVM doesn't have access to the guest + * instruction stream, i.e. can't emulate in response to a #NPF and + * instead relies on #NPF(RSVD) being reflected into the guest as #VC + * (the guest can then do a #VMGEXIT to request MMIO emulation). + */ + if (!enable_mmio_caching) + goto out; + /* Does the CPU support SEV-ES? */ if (!boot_cpu_has(X86_FEATURE_SEV_ES)) goto out; + if (!lbrv) { + WARN_ONCE(!boot_cpu_has(X86_FEATURE_LBRV), + "LBRV must be present for SEV-ES support"); + goto out; + } + /* Has the system been allocated ASIDs for SEV-ES? */ if (min_sev_asid == 1) goto out; - sev_es_asid_count = min_sev_asid - 1; - if (misc_cg_set_capacity(MISC_CG_RES_SEV_ES, sev_es_asid_count)) - goto out; + min_sev_es_asid = min_snp_asid = 1; + max_sev_es_asid = max_snp_asid = min_sev_asid - 1; - pr_info("SEV-ES supported: %u ASIDs\n", sev_es_asid_count); + sev_es_asid_count = min_sev_asid - 1; + WARN_ON_ONCE(misc_cg_set_capacity(MISC_CG_RES_SEV_ES, sev_es_asid_count)); sev_es_supported = true; + sev_snp_supported = sev_snp_enabled && cc_platform_has(CC_ATTR_HOST_SEV_SNP); out: + if (sev_enabled) { + init_args.probe = true; + + if (sev_is_snp_ciphertext_hiding_supported()) + init_args.max_snp_asid = min(nr_ciphertext_hiding_asids, + min_sev_asid - 1); + + if (sev_platform_init(&init_args)) + sev_supported = sev_es_supported = sev_snp_supported = false; + else if (sev_snp_supported) + sev_snp_supported = is_sev_snp_initialized(); + + if (sev_snp_supported) { + snp_supported_policy_bits = sev_get_snp_policy_bits() & + KVM_SNP_POLICY_MASK_VALID; + nr_ciphertext_hiding_asids = init_args.max_snp_asid; + } + + /* + * If ciphertext hiding is enabled, the joint SEV-ES/SEV-SNP + * ASID range is partitioned into separate SEV-ES and SEV-SNP + * ASID ranges, with the SEV-SNP range being [1..max_snp_asid] + * and the SEV-ES range being (max_snp_asid..max_sev_es_asid]. + * Note, SEV-ES may effectively be disabled if all ASIDs from + * the joint range are assigned to SEV-SNP. + */ + if (nr_ciphertext_hiding_asids) { + max_snp_asid = nr_ciphertext_hiding_asids; + min_sev_es_asid = max_snp_asid + 1; + pr_info("SEV-SNP ciphertext hiding enabled\n"); + } + } + + if (boot_cpu_has(X86_FEATURE_SEV)) + pr_info("SEV %s (ASIDs %u - %u)\n", + sev_supported ? min_sev_asid <= max_sev_asid ? "enabled" : + "unusable" : + "disabled", + min_sev_asid, max_sev_asid); + if (boot_cpu_has(X86_FEATURE_SEV_ES)) + pr_info("SEV-ES %s (ASIDs %u - %u)\n", + sev_es_supported ? min_sev_es_asid <= max_sev_es_asid ? "enabled" : + "unusable" : + "disabled", + min_sev_es_asid, max_sev_es_asid); + if (boot_cpu_has(X86_FEATURE_SEV_SNP)) + pr_info("SEV-SNP %s (ASIDs %u - %u)\n", + str_enabled_disabled(sev_snp_supported), + min_snp_asid, max_snp_asid); + sev_enabled = sev_supported; sev_es_enabled = sev_es_supported; -#endif + sev_snp_enabled = sev_snp_supported; + + if (!sev_es_enabled || !cpu_feature_enabled(X86_FEATURE_DEBUG_SWAP) || + !cpu_feature_enabled(X86_FEATURE_NO_NESTED_DATA_BP)) + sev_es_debug_swap_enabled = false; + + sev_supported_vmsa_features = 0; + if (sev_es_debug_swap_enabled) + sev_supported_vmsa_features |= SVM_SEV_FEAT_DEBUG_SWAP; + + if (sev_snp_enabled && tsc_khz && cpu_feature_enabled(X86_FEATURE_SNP_SECURE_TSC)) + sev_supported_vmsa_features |= SVM_SEV_FEAT_SECURE_TSC; } -void sev_hardware_teardown(void) +void sev_hardware_unsetup(void) { if (!sev_enabled) return; /* No need to take sev_bitmap_lock, all VMs have been destroyed. */ - sev_flush_asids(0, max_sev_asid); + sev_flush_asids(1, max_sev_asid); bitmap_free(sev_asid_bitmap); bitmap_free(sev_reclaim_asid_bitmap); misc_cg_set_capacity(MISC_CG_RES_SEV, 0); misc_cg_set_capacity(MISC_CG_RES_SEV_ES, 0); + + sev_platform_shutdown(); } int sev_cpu_init(struct svm_cpu_data *sd) @@ -1919,7 +3184,7 @@ int sev_cpu_init(struct svm_cpu_data *sd) if (!sev_enabled) return 0; - sd->sev_vmcbs = kcalloc(max_sev_asid + 1, sizeof(void *), GFP_KERNEL); + sd->sev_vmcbs = kcalloc(nr_asids, sizeof(void *), GFP_KERNEL); if (!sd->sev_vmcbs) return -ENOMEM; @@ -1930,52 +3195,53 @@ int sev_cpu_init(struct svm_cpu_data *sd) * Pages used by hardware to hold guest encrypted state must be flushed before * returning them to the system. */ -static void sev_flush_guest_memory(struct vcpu_svm *svm, void *va, - unsigned long len) +static void sev_flush_encrypted_page(struct kvm_vcpu *vcpu, void *va) { + unsigned int asid = sev_get_asid(vcpu->kvm); + /* - * If hardware enforced cache coherency for encrypted mappings of the - * same physical page is supported, nothing to do. + * Note! The address must be a kernel address, as regular page walk + * checks are performed by VM_PAGE_FLUSH, i.e. operating on a user + * address is non-deterministic and unsafe. This function deliberately + * takes a pointer to deter passing in a user address. */ - if (boot_cpu_has(X86_FEATURE_SME_COHERENT)) - return; + unsigned long addr = (unsigned long)va; /* - * If the VM Page Flush MSR is supported, use it to flush the page - * (using the page virtual address and the guest ASID). + * If CPU enforced cache coherency for encrypted mappings of the + * same physical page is supported, use CLFLUSHOPT instead. NOTE: cache + * flush is still needed in order to work properly with DMA devices. */ - if (boot_cpu_has(X86_FEATURE_VM_PAGE_FLUSH)) { - struct kvm_sev_info *sev; - unsigned long va_start; - u64 start, stop; - - /* Align start and stop to page boundaries. */ - va_start = (unsigned long)va; - start = (u64)va_start & PAGE_MASK; - stop = PAGE_ALIGN((u64)va_start + len); - - if (start < stop) { - sev = &to_kvm_svm(svm->vcpu.kvm)->sev_info; + if (boot_cpu_has(X86_FEATURE_SME_COHERENT)) { + clflush_cache_range(va, PAGE_SIZE); + return; + } - while (start < stop) { - wrmsrl(MSR_AMD64_VM_PAGE_FLUSH, - start | sev->asid); + /* + * VM Page Flush takes a host virtual address and a guest ASID. Fall + * back to full writeback of caches if this faults so as not to make + * any problems worse by leaving stale encrypted data in the cache. + */ + if (WARN_ON_ONCE(wrmsrq_safe(MSR_AMD64_VM_PAGE_FLUSH, addr | asid))) + goto do_sev_writeback_caches; - start += PAGE_SIZE; - } + return; - return; - } - - WARN(1, "Address overflow, using WBINVD\n"); - } +do_sev_writeback_caches: + sev_writeback_caches(vcpu->kvm); +} +void sev_guest_memory_reclaimed(struct kvm *kvm) +{ /* - * Hardware should always have one of the above features, - * but if not, use WBINVD and issue a warning. + * With SNP+gmem, private/encrypted memory is unreachable via the + * hva-based mmu notifiers, i.e. these events are explicitly scoped to + * shared pages, where there's no need to flush caches. */ - WARN_ONCE(1, "Using WBINVD to flush guest memory\n"); - wbinvd_on_all_cpus(); + if (!sev_guest(kvm) || sev_snp_guest(kvm)) + return; + + sev_writeback_caches(kvm); } void sev_free_vcpu(struct kvm_vcpu *vcpu) @@ -1987,17 +3253,36 @@ void sev_free_vcpu(struct kvm_vcpu *vcpu) svm = to_svm(vcpu); + /* + * If it's an SNP guest, then the VMSA was marked in the RMP table as + * a guest-owned page. Transition the page to hypervisor state before + * releasing it back to the system. + */ + if (sev_snp_guest(vcpu->kvm)) { + u64 pfn = __pa(svm->sev_es.vmsa) >> PAGE_SHIFT; + + if (kvm_rmp_make_shared(vcpu->kvm, pfn, PG_LEVEL_4K)) + goto skip_vmsa_free; + } + if (vcpu->arch.guest_state_protected) - sev_flush_guest_memory(svm, svm->vmsa, PAGE_SIZE); - __free_page(virt_to_page(svm->vmsa)); + sev_flush_encrypted_page(vcpu, svm->sev_es.vmsa); + + __free_page(virt_to_page(svm->sev_es.vmsa)); - if (svm->ghcb_sa_free) - kfree(svm->ghcb_sa); +skip_vmsa_free: + if (svm->sev_es.ghcb_sa_free) + kvfree(svm->sev_es.ghcb_sa); +} + +static u64 kvm_get_cached_sw_exit_code(struct vmcb_control_area *control) +{ + return (((u64)control->exit_code_hi) << 32) | control->exit_code; } static void dump_ghcb(struct vcpu_svm *svm) { - struct ghcb *ghcb = svm->ghcb; + struct vmcb_control_area *control = &svm->vmcb->control; unsigned int nbits; /* Re-use the dump_invalid_vmcb module parameter */ @@ -2006,24 +3291,30 @@ static void dump_ghcb(struct vcpu_svm *svm) return; } - nbits = sizeof(ghcb->save.valid_bitmap) * 8; + nbits = sizeof(svm->sev_es.valid_bitmap) * 8; - pr_err("GHCB (GPA=%016llx):\n", svm->vmcb->control.ghcb_gpa); + /* + * Print KVM's snapshot of the GHCB values that were (unsuccessfully) + * used to handle the exit. If the guest has since modified the GHCB + * itself, dumping the raw GHCB won't help debug why KVM was unable to + * handle the VMGEXIT that KVM observed. + */ + pr_err("GHCB (GPA=%016llx) snapshot:\n", svm->vmcb->control.ghcb_gpa); pr_err("%-20s%016llx is_valid: %u\n", "sw_exit_code", - ghcb->save.sw_exit_code, ghcb_sw_exit_code_is_valid(ghcb)); + kvm_get_cached_sw_exit_code(control), kvm_ghcb_sw_exit_code_is_valid(svm)); pr_err("%-20s%016llx is_valid: %u\n", "sw_exit_info_1", - ghcb->save.sw_exit_info_1, ghcb_sw_exit_info_1_is_valid(ghcb)); + control->exit_info_1, kvm_ghcb_sw_exit_info_1_is_valid(svm)); pr_err("%-20s%016llx is_valid: %u\n", "sw_exit_info_2", - ghcb->save.sw_exit_info_2, ghcb_sw_exit_info_2_is_valid(ghcb)); + control->exit_info_2, kvm_ghcb_sw_exit_info_2_is_valid(svm)); pr_err("%-20s%016llx is_valid: %u\n", "sw_scratch", - ghcb->save.sw_scratch, ghcb_sw_scratch_is_valid(ghcb)); - pr_err("%-20s%*pb\n", "valid_bitmap", nbits, ghcb->save.valid_bitmap); + svm->sev_es.sw_scratch, kvm_ghcb_sw_scratch_is_valid(svm)); + pr_err("%-20s%*pb\n", "valid_bitmap", nbits, svm->sev_es.valid_bitmap); } static void sev_es_sync_to_ghcb(struct vcpu_svm *svm) { struct kvm_vcpu *vcpu = &svm->vcpu; - struct ghcb *ghcb = svm->ghcb; + struct ghcb *ghcb = svm->sev_es.ghcb; /* * The GHCB protocol so far allows for the following data @@ -2043,7 +3334,7 @@ static void sev_es_sync_from_ghcb(struct vcpu_svm *svm) { struct vmcb_control_area *control = &svm->vmcb->control; struct kvm_vcpu *vcpu = &svm->vcpu; - struct ghcb *ghcb = svm->ghcb; + struct ghcb *ghcb = svm->sev_es.ghcb; u64 exit_code; /* @@ -2060,25 +3351,30 @@ static void sev_es_sync_from_ghcb(struct vcpu_svm *svm) */ memset(vcpu->arch.regs, 0, sizeof(vcpu->arch.regs)); - vcpu->arch.regs[VCPU_REGS_RAX] = ghcb_get_rax_if_valid(ghcb); - vcpu->arch.regs[VCPU_REGS_RBX] = ghcb_get_rbx_if_valid(ghcb); - vcpu->arch.regs[VCPU_REGS_RCX] = ghcb_get_rcx_if_valid(ghcb); - vcpu->arch.regs[VCPU_REGS_RDX] = ghcb_get_rdx_if_valid(ghcb); - vcpu->arch.regs[VCPU_REGS_RSI] = ghcb_get_rsi_if_valid(ghcb); + BUILD_BUG_ON(sizeof(svm->sev_es.valid_bitmap) != sizeof(ghcb->save.valid_bitmap)); + memcpy(&svm->sev_es.valid_bitmap, &ghcb->save.valid_bitmap, sizeof(ghcb->save.valid_bitmap)); - svm->vmcb->save.cpl = ghcb_get_cpl_if_valid(ghcb); + vcpu->arch.regs[VCPU_REGS_RAX] = kvm_ghcb_get_rax_if_valid(svm); + vcpu->arch.regs[VCPU_REGS_RBX] = kvm_ghcb_get_rbx_if_valid(svm); + vcpu->arch.regs[VCPU_REGS_RCX] = kvm_ghcb_get_rcx_if_valid(svm); + vcpu->arch.regs[VCPU_REGS_RDX] = kvm_ghcb_get_rdx_if_valid(svm); + vcpu->arch.regs[VCPU_REGS_RSI] = kvm_ghcb_get_rsi_if_valid(svm); - if (ghcb_xcr0_is_valid(ghcb)) { - vcpu->arch.xcr0 = ghcb_get_xcr0(ghcb); - kvm_update_cpuid_runtime(vcpu); - } + svm->vmcb->save.cpl = kvm_ghcb_get_cpl_if_valid(svm); + + if (kvm_ghcb_xcr0_is_valid(svm)) + __kvm_set_xcr(vcpu, 0, kvm_ghcb_get_xcr0(svm)); + + if (kvm_ghcb_xss_is_valid(svm)) + __kvm_emulate_msr_write(vcpu, MSR_IA32_XSS, kvm_ghcb_get_xss(svm)); /* Copy the GHCB exit information into the VMCB fields */ - exit_code = ghcb_get_sw_exit_code(ghcb); + exit_code = kvm_ghcb_get_sw_exit_code(svm); control->exit_code = lower_32_bits(exit_code); control->exit_code_hi = upper_32_bits(exit_code); - control->exit_info_1 = ghcb_get_sw_exit_info_1(ghcb); - control->exit_info_2 = ghcb_get_sw_exit_info_2(ghcb); + control->exit_info_1 = kvm_ghcb_get_sw_exit_info_1(svm); + control->exit_info_2 = kvm_ghcb_get_sw_exit_info_2(svm); + svm->sev_es.sw_scratch = kvm_ghcb_get_sw_scratch_if_valid(svm); /* Clear the valid entries fields */ memset(ghcb->save.valid_bitmap, 0, sizeof(ghcb->save.valid_bitmap)); @@ -2086,72 +3382,75 @@ static void sev_es_sync_from_ghcb(struct vcpu_svm *svm) static int sev_es_validate_vmgexit(struct vcpu_svm *svm) { - struct kvm_vcpu *vcpu; - struct ghcb *ghcb; - u64 exit_code = 0; + struct vmcb_control_area *control = &svm->vmcb->control; + struct kvm_vcpu *vcpu = &svm->vcpu; + u64 exit_code; + u64 reason; - ghcb = svm->ghcb; + /* + * Retrieve the exit code now even though it may not be marked valid + * as it could help with debugging. + */ + exit_code = kvm_get_cached_sw_exit_code(control); /* Only GHCB Usage code 0 is supported */ - if (ghcb->ghcb_usage) + if (svm->sev_es.ghcb->ghcb_usage) { + reason = GHCB_ERR_INVALID_USAGE; goto vmgexit_err; + } - /* - * Retrieve the exit code now even though is may not be marked valid - * as it could help with debugging. - */ - exit_code = ghcb_get_sw_exit_code(ghcb); + reason = GHCB_ERR_MISSING_INPUT; - if (!ghcb_sw_exit_code_is_valid(ghcb) || - !ghcb_sw_exit_info_1_is_valid(ghcb) || - !ghcb_sw_exit_info_2_is_valid(ghcb)) + if (!kvm_ghcb_sw_exit_code_is_valid(svm) || + !kvm_ghcb_sw_exit_info_1_is_valid(svm) || + !kvm_ghcb_sw_exit_info_2_is_valid(svm)) goto vmgexit_err; - switch (ghcb_get_sw_exit_code(ghcb)) { + switch (exit_code) { case SVM_EXIT_READ_DR7: break; case SVM_EXIT_WRITE_DR7: - if (!ghcb_rax_is_valid(ghcb)) + if (!kvm_ghcb_rax_is_valid(svm)) goto vmgexit_err; break; case SVM_EXIT_RDTSC: break; case SVM_EXIT_RDPMC: - if (!ghcb_rcx_is_valid(ghcb)) + if (!kvm_ghcb_rcx_is_valid(svm)) goto vmgexit_err; break; case SVM_EXIT_CPUID: - if (!ghcb_rax_is_valid(ghcb) || - !ghcb_rcx_is_valid(ghcb)) + if (!kvm_ghcb_rax_is_valid(svm) || + !kvm_ghcb_rcx_is_valid(svm)) goto vmgexit_err; - if (ghcb_get_rax(ghcb) == 0xd) - if (!ghcb_xcr0_is_valid(ghcb)) + if (vcpu->arch.regs[VCPU_REGS_RAX] == 0xd) + if (!kvm_ghcb_xcr0_is_valid(svm)) goto vmgexit_err; break; case SVM_EXIT_INVD: break; case SVM_EXIT_IOIO: - if (ghcb_get_sw_exit_info_1(ghcb) & SVM_IOIO_STR_MASK) { - if (!ghcb_sw_scratch_is_valid(ghcb)) + if (control->exit_info_1 & SVM_IOIO_STR_MASK) { + if (!kvm_ghcb_sw_scratch_is_valid(svm)) goto vmgexit_err; } else { - if (!(ghcb_get_sw_exit_info_1(ghcb) & SVM_IOIO_TYPE_MASK)) - if (!ghcb_rax_is_valid(ghcb)) + if (!(control->exit_info_1 & SVM_IOIO_TYPE_MASK)) + if (!kvm_ghcb_rax_is_valid(svm)) goto vmgexit_err; } break; case SVM_EXIT_MSR: - if (!ghcb_rcx_is_valid(ghcb)) + if (!kvm_ghcb_rcx_is_valid(svm)) goto vmgexit_err; - if (ghcb_get_sw_exit_info_1(ghcb)) { - if (!ghcb_rax_is_valid(ghcb) || - !ghcb_rdx_is_valid(ghcb)) + if (control->exit_info_1) { + if (!kvm_ghcb_rax_is_valid(svm) || + !kvm_ghcb_rdx_is_valid(svm)) goto vmgexit_err; } break; case SVM_EXIT_VMMCALL: - if (!ghcb_rax_is_valid(ghcb) || - !ghcb_cpl_is_valid(ghcb)) + if (!kvm_ghcb_rax_is_valid(svm) || + !kvm_ghcb_cpl_is_valid(svm)) goto vmgexit_err; break; case SVM_EXIT_RDTSCP: @@ -2159,88 +3458,130 @@ static int sev_es_validate_vmgexit(struct vcpu_svm *svm) case SVM_EXIT_WBINVD: break; case SVM_EXIT_MONITOR: - if (!ghcb_rax_is_valid(ghcb) || - !ghcb_rcx_is_valid(ghcb) || - !ghcb_rdx_is_valid(ghcb)) + if (!kvm_ghcb_rax_is_valid(svm) || + !kvm_ghcb_rcx_is_valid(svm) || + !kvm_ghcb_rdx_is_valid(svm)) goto vmgexit_err; break; case SVM_EXIT_MWAIT: - if (!ghcb_rax_is_valid(ghcb) || - !ghcb_rcx_is_valid(ghcb)) + if (!kvm_ghcb_rax_is_valid(svm) || + !kvm_ghcb_rcx_is_valid(svm)) goto vmgexit_err; break; case SVM_VMGEXIT_MMIO_READ: case SVM_VMGEXIT_MMIO_WRITE: - if (!ghcb_sw_scratch_is_valid(ghcb)) + if (!kvm_ghcb_sw_scratch_is_valid(svm)) goto vmgexit_err; break; + case SVM_VMGEXIT_AP_CREATION: + if (!sev_snp_guest(vcpu->kvm)) + goto vmgexit_err; + if (lower_32_bits(control->exit_info_1) != SVM_VMGEXIT_AP_DESTROY) + if (!kvm_ghcb_rax_is_valid(svm)) + goto vmgexit_err; + break; case SVM_VMGEXIT_NMI_COMPLETE: case SVM_VMGEXIT_AP_HLT_LOOP: case SVM_VMGEXIT_AP_JUMP_TABLE: case SVM_VMGEXIT_UNSUPPORTED_EVENT: + case SVM_VMGEXIT_HV_FEATURES: + case SVM_VMGEXIT_TERM_REQUEST: + break; + case SVM_VMGEXIT_PSC: + if (!sev_snp_guest(vcpu->kvm) || !kvm_ghcb_sw_scratch_is_valid(svm)) + goto vmgexit_err; + break; + case SVM_VMGEXIT_GUEST_REQUEST: + case SVM_VMGEXIT_EXT_GUEST_REQUEST: + if (!sev_snp_guest(vcpu->kvm) || + !PAGE_ALIGNED(control->exit_info_1) || + !PAGE_ALIGNED(control->exit_info_2) || + control->exit_info_1 == control->exit_info_2) + goto vmgexit_err; break; default: + reason = GHCB_ERR_INVALID_EVENT; goto vmgexit_err; } return 0; vmgexit_err: - vcpu = &svm->vcpu; - - if (ghcb->ghcb_usage) { + if (reason == GHCB_ERR_INVALID_USAGE) { vcpu_unimpl(vcpu, "vmgexit: ghcb usage %#x is not valid\n", - ghcb->ghcb_usage); + svm->sev_es.ghcb->ghcb_usage); + } else if (reason == GHCB_ERR_INVALID_EVENT) { + vcpu_unimpl(vcpu, "vmgexit: exit code %#llx is not valid\n", + exit_code); } else { - vcpu_unimpl(vcpu, "vmgexit: exit reason %#llx is not valid\n", + vcpu_unimpl(vcpu, "vmgexit: exit code %#llx input is not valid\n", exit_code); dump_ghcb(svm); } - vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_UNEXPECTED_EXIT_REASON; - vcpu->run->internal.ndata = 2; - vcpu->run->internal.data[0] = exit_code; - vcpu->run->internal.data[1] = vcpu->arch.last_vmentry_cpu; + svm_vmgexit_bad_input(svm, reason); - return -EINVAL; + /* Resume the guest to "return" the error code. */ + return 1; } void sev_es_unmap_ghcb(struct vcpu_svm *svm) { - if (!svm->ghcb) + /* Clear any indication that the vCPU is in a type of AP Reset Hold */ + svm->sev_es.ap_reset_hold_type = AP_RESET_HOLD_NONE; + + if (!svm->sev_es.ghcb) return; - if (svm->ghcb_sa_free) { + if (svm->sev_es.ghcb_sa_free) { /* * The scratch area lives outside the GHCB, so there is a * buffer that, depending on the operation performed, may * need to be synced, then freed. */ - if (svm->ghcb_sa_sync) { + if (svm->sev_es.ghcb_sa_sync) { kvm_write_guest(svm->vcpu.kvm, - ghcb_get_sw_scratch(svm->ghcb), - svm->ghcb_sa, svm->ghcb_sa_len); - svm->ghcb_sa_sync = false; + svm->sev_es.sw_scratch, + svm->sev_es.ghcb_sa, + svm->sev_es.ghcb_sa_len); + svm->sev_es.ghcb_sa_sync = false; } - kfree(svm->ghcb_sa); - svm->ghcb_sa = NULL; - svm->ghcb_sa_free = false; + kvfree(svm->sev_es.ghcb_sa); + svm->sev_es.ghcb_sa = NULL; + svm->sev_es.ghcb_sa_free = false; } - trace_kvm_vmgexit_exit(svm->vcpu.vcpu_id, svm->ghcb); + trace_kvm_vmgexit_exit(svm->vcpu.vcpu_id, svm->sev_es.ghcb); sev_es_sync_to_ghcb(svm); - kvm_vcpu_unmap(&svm->vcpu, &svm->ghcb_map, true); - svm->ghcb = NULL; + kvm_vcpu_unmap(&svm->vcpu, &svm->sev_es.ghcb_map); + svm->sev_es.ghcb = NULL; } -void pre_sev_run(struct vcpu_svm *svm, int cpu) +int pre_sev_run(struct vcpu_svm *svm, int cpu) { - struct svm_cpu_data *sd = per_cpu(svm_data, cpu); - int asid = sev_get_asid(svm->vcpu.kvm); + struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu); + struct kvm *kvm = svm->vcpu.kvm; + unsigned int asid = sev_get_asid(kvm); + + /* + * Reject KVM_RUN if userspace attempts to run the vCPU with an invalid + * VMSA, e.g. if userspace forces the vCPU to be RUNNABLE after an SNP + * AP Destroy event. + */ + if (sev_es_guest(kvm) && !VALID_PAGE(svm->vmcb->control.vmsa_pa)) + return -EINVAL; + + /* + * To optimize cache flushes when memory is reclaimed from an SEV VM, + * track physical CPUs that enter the guest for SEV VMs and thus can + * have encrypted, dirty data in the cache, and flush caches only for + * CPUs that have entered the guest. + */ + if (!cpumask_test_cpu(cpu, to_kvm_sev_info(kvm)->have_run_cpus)) + cpumask_set_cpu(cpu, to_kvm_sev_info(kvm)->have_run_cpus); /* Assign the asid allocated with this SEV guest */ svm->asid = asid; @@ -2253,33 +3594,33 @@ void pre_sev_run(struct vcpu_svm *svm, int cpu) */ if (sd->sev_vmcbs[asid] == svm->vmcb && svm->vcpu.arch.last_vmentry_cpu == cpu) - return; + return 0; sd->sev_vmcbs[asid] = svm->vmcb; svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ASID; vmcb_mark_dirty(svm->vmcb, VMCB_ASID); + return 0; } #define GHCB_SCRATCH_AREA_LIMIT (16ULL * PAGE_SIZE) -static bool setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) +static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) { struct vmcb_control_area *control = &svm->vmcb->control; - struct ghcb *ghcb = svm->ghcb; u64 ghcb_scratch_beg, ghcb_scratch_end; u64 scratch_gpa_beg, scratch_gpa_end; void *scratch_va; - scratch_gpa_beg = ghcb_get_sw_scratch(ghcb); + scratch_gpa_beg = svm->sev_es.sw_scratch; if (!scratch_gpa_beg) { pr_err("vmgexit: scratch gpa not provided\n"); - return false; + goto e_scratch; } scratch_gpa_end = scratch_gpa_beg + len; if (scratch_gpa_end < scratch_gpa_beg) { pr_err("vmgexit: scratch length (%#llx) not valid for scratch address (%#llx)\n", len, scratch_gpa_beg); - return false; + goto e_scratch; } if ((scratch_gpa_beg & PAGE_MASK) == control->ghcb_gpa) { @@ -2287,7 +3628,7 @@ static bool setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) ghcb_scratch_beg = control->ghcb_gpa + offsetof(struct ghcb, shared_buffer); ghcb_scratch_end = control->ghcb_gpa + - offsetof(struct ghcb, reserved_1); + offsetof(struct ghcb, reserved_0xff0); /* * If the scratch area begins within the GHCB, it must be @@ -2297,10 +3638,10 @@ static bool setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) scratch_gpa_end > ghcb_scratch_end) { pr_err("vmgexit: scratch area is outside of GHCB shared buffer area (%#llx - %#llx)\n", scratch_gpa_beg, scratch_gpa_end); - return false; + goto e_scratch; } - scratch_va = (void *)svm->ghcb; + scratch_va = (void *)svm->sev_es.ghcb; scratch_va += (scratch_gpa_beg - control->ghcb_gpa); } else { /* @@ -2310,18 +3651,18 @@ static bool setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) if (len > GHCB_SCRATCH_AREA_LIMIT) { pr_err("vmgexit: scratch area exceeds KVM limits (%#llx requested, %#llx limit)\n", len, GHCB_SCRATCH_AREA_LIMIT); - return false; + goto e_scratch; } - scratch_va = kzalloc(len, GFP_KERNEL_ACCOUNT); + scratch_va = kvzalloc(len, GFP_KERNEL_ACCOUNT); if (!scratch_va) - return false; + return -ENOMEM; if (kvm_read_guest(svm->vcpu.kvm, scratch_gpa_beg, scratch_va, len)) { /* Unable to copy scratch area from guest */ pr_err("vmgexit: kvm_read_guest for scratch area failed\n"); - kfree(scratch_va); - return false; + kvfree(scratch_va); + return -EFAULT; } /* @@ -2330,14 +3671,19 @@ static bool setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) * the vCPU next time (i.e. a read was requested so the data * must be written back to the guest memory). */ - svm->ghcb_sa_sync = sync; - svm->ghcb_sa_free = true; + svm->sev_es.ghcb_sa_sync = sync; + svm->sev_es.ghcb_sa_free = true; } - svm->ghcb_sa = scratch_va; - svm->ghcb_sa_len = len; + svm->sev_es.ghcb_sa = scratch_va; + svm->sev_es.ghcb_sa_len = len; - return true; + return 0; + +e_scratch: + svm_vmgexit_bad_input(svm, GHCB_ERR_INVALID_SCRATCH_AREA); + + return 1; } static void set_ghcb_msr_bits(struct vcpu_svm *svm, u64 value, u64 mask, @@ -2357,10 +3703,514 @@ static void set_ghcb_msr(struct vcpu_svm *svm, u64 value) svm->vmcb->control.ghcb_gpa = value; } +static int snp_rmptable_psmash(kvm_pfn_t pfn) +{ + int ret; + + pfn = pfn & ~(KVM_PAGES_PER_HPAGE(PG_LEVEL_2M) - 1); + + /* + * PSMASH_FAIL_INUSE indicates another processor is modifying the + * entry, so retry until that's no longer the case. + */ + do { + ret = psmash(pfn); + } while (ret == PSMASH_FAIL_INUSE); + + return ret; +} + +static int snp_complete_psc_msr(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + if (vcpu->run->hypercall.ret) + set_ghcb_msr(svm, GHCB_MSR_PSC_RESP_ERROR); + else + set_ghcb_msr(svm, GHCB_MSR_PSC_RESP); + + return 1; /* resume guest */ +} + +static int snp_begin_psc_msr(struct vcpu_svm *svm, u64 ghcb_msr) +{ + u64 gpa = gfn_to_gpa(GHCB_MSR_PSC_REQ_TO_GFN(ghcb_msr)); + u8 op = GHCB_MSR_PSC_REQ_TO_OP(ghcb_msr); + struct kvm_vcpu *vcpu = &svm->vcpu; + + if (op != SNP_PAGE_STATE_PRIVATE && op != SNP_PAGE_STATE_SHARED) { + set_ghcb_msr(svm, GHCB_MSR_PSC_RESP_ERROR); + return 1; /* resume guest */ + } + + if (!user_exit_on_hypercall(vcpu->kvm, KVM_HC_MAP_GPA_RANGE)) { + set_ghcb_msr(svm, GHCB_MSR_PSC_RESP_ERROR); + return 1; /* resume guest */ + } + + vcpu->run->exit_reason = KVM_EXIT_HYPERCALL; + vcpu->run->hypercall.nr = KVM_HC_MAP_GPA_RANGE; + /* + * In principle this should have been -KVM_ENOSYS, but userspace (QEMU <=9.2) + * assumed that vcpu->run->hypercall.ret is never changed by KVM and thus that + * it was always zero on KVM_EXIT_HYPERCALL. Since KVM is now overwriting + * vcpu->run->hypercall.ret, ensuring that it is zero to not break QEMU. + */ + vcpu->run->hypercall.ret = 0; + vcpu->run->hypercall.args[0] = gpa; + vcpu->run->hypercall.args[1] = 1; + vcpu->run->hypercall.args[2] = (op == SNP_PAGE_STATE_PRIVATE) + ? KVM_MAP_GPA_RANGE_ENCRYPTED + : KVM_MAP_GPA_RANGE_DECRYPTED; + vcpu->run->hypercall.args[2] |= KVM_MAP_GPA_RANGE_PAGE_SZ_4K; + + vcpu->arch.complete_userspace_io = snp_complete_psc_msr; + + return 0; /* forward request to userspace */ +} + +struct psc_buffer { + struct psc_hdr hdr; + struct psc_entry entries[]; +} __packed; + +static int snp_begin_psc(struct vcpu_svm *svm, struct psc_buffer *psc); + +static void snp_complete_psc(struct vcpu_svm *svm, u64 psc_ret) +{ + svm->sev_es.psc_inflight = 0; + svm->sev_es.psc_idx = 0; + svm->sev_es.psc_2m = false; + + /* + * PSC requests always get a "no action" response in SW_EXITINFO1, with + * a PSC-specific return code in SW_EXITINFO2 that provides the "real" + * return code. E.g. if the PSC request was interrupted, the need to + * retry is communicated via SW_EXITINFO2, not SW_EXITINFO1. + */ + svm_vmgexit_no_action(svm, psc_ret); +} + +static void __snp_complete_one_psc(struct vcpu_svm *svm) +{ + struct psc_buffer *psc = svm->sev_es.ghcb_sa; + struct psc_entry *entries = psc->entries; + struct psc_hdr *hdr = &psc->hdr; + __u16 idx; + + /* + * Everything in-flight has been processed successfully. Update the + * corresponding entries in the guest's PSC buffer and zero out the + * count of in-flight PSC entries. + */ + for (idx = svm->sev_es.psc_idx; svm->sev_es.psc_inflight; + svm->sev_es.psc_inflight--, idx++) { + struct psc_entry *entry = &entries[idx]; + + entry->cur_page = entry->pagesize ? 512 : 1; + } + + hdr->cur_entry = idx; +} + +static int snp_complete_one_psc(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct psc_buffer *psc = svm->sev_es.ghcb_sa; + + if (vcpu->run->hypercall.ret) { + snp_complete_psc(svm, VMGEXIT_PSC_ERROR_GENERIC); + return 1; /* resume guest */ + } + + __snp_complete_one_psc(svm); + + /* Handle the next range (if any). */ + return snp_begin_psc(svm, psc); +} + +static int snp_begin_psc(struct vcpu_svm *svm, struct psc_buffer *psc) +{ + struct psc_entry *entries = psc->entries; + struct kvm_vcpu *vcpu = &svm->vcpu; + struct psc_hdr *hdr = &psc->hdr; + struct psc_entry entry_start; + u16 idx, idx_start, idx_end; + int npages; + bool huge; + u64 gfn; + + if (!user_exit_on_hypercall(vcpu->kvm, KVM_HC_MAP_GPA_RANGE)) { + snp_complete_psc(svm, VMGEXIT_PSC_ERROR_GENERIC); + return 1; + } + +next_range: + /* There should be no other PSCs in-flight at this point. */ + if (WARN_ON_ONCE(svm->sev_es.psc_inflight)) { + snp_complete_psc(svm, VMGEXIT_PSC_ERROR_GENERIC); + return 1; + } + + /* + * The PSC descriptor buffer can be modified by a misbehaved guest after + * validation, so take care to only use validated copies of values used + * for things like array indexing. + */ + idx_start = hdr->cur_entry; + idx_end = hdr->end_entry; + + if (idx_end >= VMGEXIT_PSC_MAX_COUNT) { + snp_complete_psc(svm, VMGEXIT_PSC_ERROR_INVALID_HDR); + return 1; + } + + /* Find the start of the next range which needs processing. */ + for (idx = idx_start; idx <= idx_end; idx++, hdr->cur_entry++) { + entry_start = entries[idx]; + + gfn = entry_start.gfn; + huge = entry_start.pagesize; + npages = huge ? 512 : 1; + + if (entry_start.cur_page > npages || !IS_ALIGNED(gfn, npages)) { + snp_complete_psc(svm, VMGEXIT_PSC_ERROR_INVALID_ENTRY); + return 1; + } + + if (entry_start.cur_page) { + /* + * If this is a partially-completed 2M range, force 4K handling + * for the remaining pages since they're effectively split at + * this point. Subsequent code should ensure this doesn't get + * combined with adjacent PSC entries where 2M handling is still + * possible. + */ + npages -= entry_start.cur_page; + gfn += entry_start.cur_page; + huge = false; + } + + if (npages) + break; + } + + if (idx > idx_end) { + /* Nothing more to process. */ + snp_complete_psc(svm, 0); + return 1; + } + + svm->sev_es.psc_2m = huge; + svm->sev_es.psc_idx = idx; + svm->sev_es.psc_inflight = 1; + + /* + * Find all subsequent PSC entries that contain adjacent GPA + * ranges/operations and can be combined into a single + * KVM_HC_MAP_GPA_RANGE exit. + */ + while (++idx <= idx_end) { + struct psc_entry entry = entries[idx]; + + if (entry.operation != entry_start.operation || + entry.gfn != entry_start.gfn + npages || + entry.cur_page || !!entry.pagesize != huge) + break; + + svm->sev_es.psc_inflight++; + npages += huge ? 512 : 1; + } + + switch (entry_start.operation) { + case VMGEXIT_PSC_OP_PRIVATE: + case VMGEXIT_PSC_OP_SHARED: + vcpu->run->exit_reason = KVM_EXIT_HYPERCALL; + vcpu->run->hypercall.nr = KVM_HC_MAP_GPA_RANGE; + /* + * In principle this should have been -KVM_ENOSYS, but userspace (QEMU <=9.2) + * assumed that vcpu->run->hypercall.ret is never changed by KVM and thus that + * it was always zero on KVM_EXIT_HYPERCALL. Since KVM is now overwriting + * vcpu->run->hypercall.ret, ensuring that it is zero to not break QEMU. + */ + vcpu->run->hypercall.ret = 0; + vcpu->run->hypercall.args[0] = gfn_to_gpa(gfn); + vcpu->run->hypercall.args[1] = npages; + vcpu->run->hypercall.args[2] = entry_start.operation == VMGEXIT_PSC_OP_PRIVATE + ? KVM_MAP_GPA_RANGE_ENCRYPTED + : KVM_MAP_GPA_RANGE_DECRYPTED; + vcpu->run->hypercall.args[2] |= entry_start.pagesize + ? KVM_MAP_GPA_RANGE_PAGE_SZ_2M + : KVM_MAP_GPA_RANGE_PAGE_SZ_4K; + vcpu->arch.complete_userspace_io = snp_complete_one_psc; + return 0; /* forward request to userspace */ + default: + /* + * Only shared/private PSC operations are currently supported, so if the + * entire range consists of unsupported operations (e.g. SMASH/UNSMASH), + * then consider the entire range completed and avoid exiting to + * userspace. In theory snp_complete_psc() can always be called directly + * at this point to complete the current range and start the next one, + * but that could lead to unexpected levels of recursion. + */ + __snp_complete_one_psc(svm); + goto next_range; + } + + BUG(); +} + +/* + * Invoked as part of svm_vcpu_reset() processing of an init event. + */ +static void sev_snp_init_protected_guest_state(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct kvm_memory_slot *slot; + struct page *page; + kvm_pfn_t pfn; + gfn_t gfn; + + guard(mutex)(&svm->sev_es.snp_vmsa_mutex); + + if (!svm->sev_es.snp_ap_waiting_for_reset) + return; + + svm->sev_es.snp_ap_waiting_for_reset = false; + + /* Mark the vCPU as offline and not runnable */ + vcpu->arch.pv.pv_unhalted = false; + kvm_set_mp_state(vcpu, KVM_MP_STATE_HALTED); + + /* Clear use of the VMSA */ + svm->vmcb->control.vmsa_pa = INVALID_PAGE; + + /* + * When replacing the VMSA during SEV-SNP AP creation, + * mark the VMCB dirty so that full state is always reloaded. + */ + vmcb_mark_all_dirty(svm->vmcb); + + if (!VALID_PAGE(svm->sev_es.snp_vmsa_gpa)) + return; + + gfn = gpa_to_gfn(svm->sev_es.snp_vmsa_gpa); + svm->sev_es.snp_vmsa_gpa = INVALID_PAGE; + + slot = gfn_to_memslot(vcpu->kvm, gfn); + if (!slot) + return; + + /* + * The new VMSA will be private memory guest memory, so retrieve the + * PFN from the gmem backend. + */ + if (kvm_gmem_get_pfn(vcpu->kvm, slot, gfn, &pfn, &page, NULL)) + return; + + /* + * From this point forward, the VMSA will always be a guest-mapped page + * rather than the initial one allocated by KVM in svm->sev_es.vmsa. In + * theory, svm->sev_es.vmsa could be free'd and cleaned up here, but + * that involves cleanups like flushing caches, which would ideally be + * handled during teardown rather than guest boot. Deferring that also + * allows the existing logic for SEV-ES VMSAs to be re-used with + * minimal SNP-specific changes. + */ + svm->sev_es.snp_has_guest_vmsa = true; + + /* Use the new VMSA */ + svm->vmcb->control.vmsa_pa = pfn_to_hpa(pfn); + + /* Mark the vCPU as runnable */ + kvm_set_mp_state(vcpu, KVM_MP_STATE_RUNNABLE); + + /* + * gmem pages aren't currently migratable, but if this ever changes + * then care should be taken to ensure svm->sev_es.vmsa is pinned + * through some other means. + */ + kvm_release_page_clean(page); +} + +static int sev_snp_ap_creation(struct vcpu_svm *svm) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(svm->vcpu.kvm); + struct kvm_vcpu *vcpu = &svm->vcpu; + struct kvm_vcpu *target_vcpu; + struct vcpu_svm *target_svm; + unsigned int request; + unsigned int apic_id; + + request = lower_32_bits(svm->vmcb->control.exit_info_1); + apic_id = upper_32_bits(svm->vmcb->control.exit_info_1); + + /* Validate the APIC ID */ + target_vcpu = kvm_get_vcpu_by_id(vcpu->kvm, apic_id); + if (!target_vcpu) { + vcpu_unimpl(vcpu, "vmgexit: invalid AP APIC ID [%#x] from guest\n", + apic_id); + return -EINVAL; + } + + target_svm = to_svm(target_vcpu); + + guard(mutex)(&target_svm->sev_es.snp_vmsa_mutex); + + switch (request) { + case SVM_VMGEXIT_AP_CREATE_ON_INIT: + case SVM_VMGEXIT_AP_CREATE: + if (vcpu->arch.regs[VCPU_REGS_RAX] != sev->vmsa_features) { + vcpu_unimpl(vcpu, "vmgexit: mismatched AP sev_features [%#lx] != [%#llx] from guest\n", + vcpu->arch.regs[VCPU_REGS_RAX], sev->vmsa_features); + return -EINVAL; + } + + if (!page_address_valid(vcpu, svm->vmcb->control.exit_info_2)) { + vcpu_unimpl(vcpu, "vmgexit: invalid AP VMSA address [%#llx] from guest\n", + svm->vmcb->control.exit_info_2); + return -EINVAL; + } + + /* + * Malicious guest can RMPADJUST a large page into VMSA which + * will hit the SNP erratum where the CPU will incorrectly signal + * an RMP violation #PF if a hugepage collides with the RMP entry + * of VMSA page, reject the AP CREATE request if VMSA address from + * guest is 2M aligned. + */ + if (IS_ALIGNED(svm->vmcb->control.exit_info_2, PMD_SIZE)) { + vcpu_unimpl(vcpu, + "vmgexit: AP VMSA address [%llx] from guest is unsafe as it is 2M aligned\n", + svm->vmcb->control.exit_info_2); + return -EINVAL; + } + + target_svm->sev_es.snp_vmsa_gpa = svm->vmcb->control.exit_info_2; + break; + case SVM_VMGEXIT_AP_DESTROY: + target_svm->sev_es.snp_vmsa_gpa = INVALID_PAGE; + break; + default: + vcpu_unimpl(vcpu, "vmgexit: invalid AP creation request [%#x] from guest\n", + request); + return -EINVAL; + } + + target_svm->sev_es.snp_ap_waiting_for_reset = true; + + /* + * Unless Creation is deferred until INIT, signal the vCPU to update + * its state. + */ + if (request != SVM_VMGEXIT_AP_CREATE_ON_INIT) + kvm_make_request_and_kick(KVM_REQ_UPDATE_PROTECTED_GUEST_STATE, target_vcpu); + + return 0; +} + +static int snp_handle_guest_req(struct vcpu_svm *svm, gpa_t req_gpa, gpa_t resp_gpa) +{ + struct sev_data_snp_guest_request data = {0}; + struct kvm *kvm = svm->vcpu.kvm; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + sev_ret_code fw_err = 0; + int ret; + + if (!sev_snp_guest(kvm)) + return -EINVAL; + + mutex_lock(&sev->guest_req_mutex); + + if (kvm_read_guest(kvm, req_gpa, sev->guest_req_buf, PAGE_SIZE)) { + ret = -EIO; + goto out_unlock; + } + + data.gctx_paddr = __psp_pa(sev->snp_context); + data.req_paddr = __psp_pa(sev->guest_req_buf); + data.res_paddr = __psp_pa(sev->guest_resp_buf); + + /* + * Firmware failures are propagated on to guest, but any other failure + * condition along the way should be reported to userspace. E.g. if + * the PSP is dead and commands are timing out. + */ + ret = sev_issue_cmd(kvm, SEV_CMD_SNP_GUEST_REQUEST, &data, &fw_err); + if (ret && !fw_err) + goto out_unlock; + + if (kvm_write_guest(kvm, resp_gpa, sev->guest_resp_buf, PAGE_SIZE)) { + ret = -EIO; + goto out_unlock; + } + + /* No action is requested *from KVM* if there was a firmware error. */ + svm_vmgexit_no_action(svm, SNP_GUEST_ERR(0, fw_err)); + + ret = 1; /* resume guest */ + +out_unlock: + mutex_unlock(&sev->guest_req_mutex); + return ret; +} + +static int snp_handle_ext_guest_req(struct vcpu_svm *svm, gpa_t req_gpa, gpa_t resp_gpa) +{ + struct kvm *kvm = svm->vcpu.kvm; + u8 msg_type; + + if (!sev_snp_guest(kvm)) + return -EINVAL; + + if (kvm_read_guest(kvm, req_gpa + offsetof(struct snp_guest_msg_hdr, msg_type), + &msg_type, 1)) + return -EIO; + + /* + * As per GHCB spec, requests of type MSG_REPORT_REQ also allow for + * additional certificate data to be provided alongside the attestation + * report via the guest-provided data pages indicated by RAX/RBX. The + * certificate data is optional and requires additional KVM enablement + * to provide an interface for userspace to provide it, but KVM still + * needs to be able to handle extended guest requests either way. So + * provide a stub implementation that will always return an empty + * certificate table in the guest-provided data pages. + */ + if (msg_type == SNP_MSG_REPORT_REQ) { + struct kvm_vcpu *vcpu = &svm->vcpu; + u64 data_npages; + gpa_t data_gpa; + + if (!kvm_ghcb_rax_is_valid(svm) || !kvm_ghcb_rbx_is_valid(svm)) + goto request_invalid; + + data_gpa = vcpu->arch.regs[VCPU_REGS_RAX]; + data_npages = vcpu->arch.regs[VCPU_REGS_RBX]; + + if (!PAGE_ALIGNED(data_gpa)) + goto request_invalid; + + /* + * As per GHCB spec (see "SNP Extended Guest Request"), the + * certificate table is terminated by 24-bytes of zeroes. + */ + if (data_npages && kvm_clear_guest(kvm, data_gpa, 24)) + return -EIO; + } + + return snp_handle_guest_req(svm, req_gpa, resp_gpa); + +request_invalid: + svm_vmgexit_bad_input(svm, GHCB_ERR_INVALID_INPUT); + return 1; /* resume guest */ +} + static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm) { struct vmcb_control_area *control = &svm->vmcb->control; struct kvm_vcpu *vcpu = &svm->vcpu; + struct kvm_sev_info *sev = to_kvm_sev_info(vcpu->kvm); u64 ghcb_info; int ret = 1; @@ -2371,7 +4221,7 @@ static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm) switch (ghcb_info) { case GHCB_MSR_SEV_INFO_REQ: - set_ghcb_msr(svm, GHCB_MSR_SEV_INFO(GHCB_VERSION_MAX, + set_ghcb_msr(svm, GHCB_MSR_SEV_INFO((__u64)sev->ghcb_version, GHCB_VERSION_MIN, sev_enc_bit)); break; @@ -2388,7 +4238,7 @@ static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm) ret = svm_invoke_exit_handler(vcpu, SVM_EXIT_CPUID); if (!ret) { - ret = -EINVAL; + /* Error, keep GHCB MSR value as-is */ break; } @@ -2413,6 +4263,60 @@ static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm) GHCB_MSR_INFO_POS); break; } + case GHCB_MSR_AP_RESET_HOLD_REQ: + svm->sev_es.ap_reset_hold_type = AP_RESET_HOLD_MSR_PROTO; + ret = kvm_emulate_ap_reset_hold(&svm->vcpu); + + /* + * Preset the result to a non-SIPI return and then only set + * the result to non-zero when delivering a SIPI. + */ + set_ghcb_msr_bits(svm, 0, + GHCB_MSR_AP_RESET_HOLD_RESULT_MASK, + GHCB_MSR_AP_RESET_HOLD_RESULT_POS); + + set_ghcb_msr_bits(svm, GHCB_MSR_AP_RESET_HOLD_RESP, + GHCB_MSR_INFO_MASK, + GHCB_MSR_INFO_POS); + break; + case GHCB_MSR_HV_FT_REQ: + set_ghcb_msr_bits(svm, GHCB_HV_FT_SUPPORTED, + GHCB_MSR_HV_FT_MASK, GHCB_MSR_HV_FT_POS); + set_ghcb_msr_bits(svm, GHCB_MSR_HV_FT_RESP, + GHCB_MSR_INFO_MASK, GHCB_MSR_INFO_POS); + break; + case GHCB_MSR_PREF_GPA_REQ: + if (!sev_snp_guest(vcpu->kvm)) + goto out_terminate; + + set_ghcb_msr_bits(svm, GHCB_MSR_PREF_GPA_NONE, GHCB_MSR_GPA_VALUE_MASK, + GHCB_MSR_GPA_VALUE_POS); + set_ghcb_msr_bits(svm, GHCB_MSR_PREF_GPA_RESP, GHCB_MSR_INFO_MASK, + GHCB_MSR_INFO_POS); + break; + case GHCB_MSR_REG_GPA_REQ: { + u64 gfn; + + if (!sev_snp_guest(vcpu->kvm)) + goto out_terminate; + + gfn = get_ghcb_msr_bits(svm, GHCB_MSR_GPA_VALUE_MASK, + GHCB_MSR_GPA_VALUE_POS); + + svm->sev_es.ghcb_registered_gpa = gfn_to_gpa(gfn); + + set_ghcb_msr_bits(svm, gfn, GHCB_MSR_GPA_VALUE_MASK, + GHCB_MSR_GPA_VALUE_POS); + set_ghcb_msr_bits(svm, GHCB_MSR_REG_GPA_RESP, GHCB_MSR_INFO_MASK, + GHCB_MSR_INFO_POS); + break; + } + case GHCB_MSR_PSC_REQ: + if (!sev_snp_guest(vcpu->kvm)) + goto out_terminate; + + ret = snp_begin_psc_msr(svm, control->ghcb_gpa); + break; case GHCB_MSR_TERM_REQ: { u64 reason_set, reason_code; @@ -2424,16 +4328,26 @@ static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm) GHCB_MSR_TERM_REASON_POS); pr_info("SEV-ES guest requested termination: %#llx:%#llx\n", reason_set, reason_code); - fallthrough; + + goto out_terminate; } default: - ret = -EINVAL; + /* Error, keep GHCB MSR value as-is */ + break; } trace_kvm_vmgexit_msr_protocol_exit(svm->vcpu.vcpu_id, control->ghcb_gpa, ret); return ret; + +out_terminate: + vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT; + vcpu->run->system_event.type = KVM_SYSTEM_EVENT_SEV_TERM; + vcpu->run->system_event.ndata = 1; + vcpu->run->system_event.data[0] = control->ghcb_gpa; + + return 0; } int sev_handle_vmgexit(struct kvm_vcpu *vcpu) @@ -2441,7 +4355,6 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); struct vmcb_control_area *control = &svm->vmcb->control; u64 ghcb_gpa, exit_code; - struct ghcb *ghcb; int ret; /* Validate the GHCB */ @@ -2451,59 +4364,72 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) if (!ghcb_gpa) { vcpu_unimpl(vcpu, "vmgexit: GHCB gpa is not set\n"); - return -EINVAL; + + /* Without a GHCB, just return right back to the guest */ + return 1; } - if (kvm_vcpu_map(vcpu, ghcb_gpa >> PAGE_SHIFT, &svm->ghcb_map)) { + if (kvm_vcpu_map(vcpu, ghcb_gpa >> PAGE_SHIFT, &svm->sev_es.ghcb_map)) { /* Unable to map GHCB from guest */ vcpu_unimpl(vcpu, "vmgexit: error mapping GHCB [%#llx] from guest\n", ghcb_gpa); - return -EINVAL; + + /* Without a GHCB, just return right back to the guest */ + return 1; } - svm->ghcb = svm->ghcb_map.hva; - ghcb = svm->ghcb_map.hva; + svm->sev_es.ghcb = svm->sev_es.ghcb_map.hva; - trace_kvm_vmgexit_enter(vcpu->vcpu_id, ghcb); + trace_kvm_vmgexit_enter(vcpu->vcpu_id, svm->sev_es.ghcb); - exit_code = ghcb_get_sw_exit_code(ghcb); + sev_es_sync_from_ghcb(svm); + + /* SEV-SNP guest requires that the GHCB GPA must be registered */ + if (sev_snp_guest(svm->vcpu.kvm) && !ghcb_gpa_is_registered(svm, ghcb_gpa)) { + vcpu_unimpl(&svm->vcpu, "vmgexit: GHCB GPA [%#llx] is not registered.\n", ghcb_gpa); + return -EINVAL; + } ret = sev_es_validate_vmgexit(svm); if (ret) return ret; - sev_es_sync_from_ghcb(svm); - ghcb_set_sw_exit_info_1(ghcb, 0); - ghcb_set_sw_exit_info_2(ghcb, 0); + svm_vmgexit_success(svm, 0); - ret = -EINVAL; + exit_code = kvm_get_cached_sw_exit_code(control); switch (exit_code) { case SVM_VMGEXIT_MMIO_READ: - if (!setup_vmgexit_scratch(svm, true, control->exit_info_2)) + ret = setup_vmgexit_scratch(svm, true, control->exit_info_2); + if (ret) break; ret = kvm_sev_es_mmio_read(vcpu, control->exit_info_1, control->exit_info_2, - svm->ghcb_sa); + svm->sev_es.ghcb_sa); break; case SVM_VMGEXIT_MMIO_WRITE: - if (!setup_vmgexit_scratch(svm, false, control->exit_info_2)) + ret = setup_vmgexit_scratch(svm, false, control->exit_info_2); + if (ret) break; ret = kvm_sev_es_mmio_write(vcpu, control->exit_info_1, control->exit_info_2, - svm->ghcb_sa); + svm->sev_es.ghcb_sa); break; case SVM_VMGEXIT_NMI_COMPLETE: - ret = svm_invoke_exit_handler(vcpu, SVM_EXIT_IRET); + ++vcpu->stat.nmi_window_exits; + svm->nmi_masked = false; + kvm_make_request(KVM_REQ_EVENT, vcpu); + ret = 1; break; case SVM_VMGEXIT_AP_HLT_LOOP: + svm->sev_es.ap_reset_hold_type = AP_RESET_HOLD_NAE_EVENT; ret = kvm_emulate_ap_reset_hold(vcpu); break; case SVM_VMGEXIT_AP_JUMP_TABLE: { - struct kvm_sev_info *sev = &to_kvm_svm(vcpu->kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(vcpu->kvm); switch (control->exit_info_1) { case 0: @@ -2512,25 +4438,55 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) break; case 1: /* Get AP jump table address */ - ghcb_set_sw_exit_info_2(ghcb, sev->ap_jump_table); + svm_vmgexit_success(svm, sev->ap_jump_table); break; default: pr_err("svm: vmgexit: unsupported AP jump table request - exit_info_1=%#llx\n", control->exit_info_1); - ghcb_set_sw_exit_info_1(ghcb, 1); - ghcb_set_sw_exit_info_2(ghcb, - X86_TRAP_UD | - SVM_EVTINJ_TYPE_EXEPT | - SVM_EVTINJ_VALID); + svm_vmgexit_bad_input(svm, GHCB_ERR_INVALID_INPUT); } ret = 1; break; } + case SVM_VMGEXIT_HV_FEATURES: + svm_vmgexit_success(svm, GHCB_HV_FT_SUPPORTED); + ret = 1; + break; + case SVM_VMGEXIT_TERM_REQUEST: + pr_info("SEV-ES guest requested termination: reason %#llx info %#llx\n", + control->exit_info_1, control->exit_info_2); + vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT; + vcpu->run->system_event.type = KVM_SYSTEM_EVENT_SEV_TERM; + vcpu->run->system_event.ndata = 1; + vcpu->run->system_event.data[0] = control->ghcb_gpa; + break; + case SVM_VMGEXIT_PSC: + ret = setup_vmgexit_scratch(svm, true, control->exit_info_2); + if (ret) + break; + + ret = snp_begin_psc(svm, svm->sev_es.ghcb_sa); + break; + case SVM_VMGEXIT_AP_CREATION: + ret = sev_snp_ap_creation(svm); + if (ret) { + svm_vmgexit_bad_input(svm, GHCB_ERR_INVALID_INPUT); + } + + ret = 1; + break; + case SVM_VMGEXIT_GUEST_REQUEST: + ret = snp_handle_guest_req(svm, control->exit_info_1, control->exit_info_2); + break; + case SVM_VMGEXIT_EXT_GUEST_REQUEST: + ret = snp_handle_ext_guest_req(svm, control->exit_info_1, control->exit_info_2); + break; case SVM_VMGEXIT_UNSUPPORTED_EVENT: vcpu_unimpl(vcpu, "vmgexit: unsupported event - exit_info_1=%#llx, exit_info_2=%#llx\n", control->exit_info_1, control->exit_info_2); + ret = -EINVAL; break; default: ret = svm_invoke_exit_handler(vcpu, exit_code); @@ -2541,26 +4497,92 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) int sev_es_string_io(struct vcpu_svm *svm, int size, unsigned int port, int in) { - if (!setup_vmgexit_scratch(svm, in, svm->vmcb->control.exit_info_2)) + int count; + int bytes; + int r; + + if (svm->vmcb->control.exit_info_2 > INT_MAX) return -EINVAL; - return kvm_sev_es_string_io(&svm->vcpu, size, port, - svm->ghcb_sa, svm->ghcb_sa_len, in); + count = svm->vmcb->control.exit_info_2; + if (unlikely(check_mul_overflow(count, size, &bytes))) + return -EINVAL; + + r = setup_vmgexit_scratch(svm, in, bytes); + if (r) + return r; + + return kvm_sev_es_string_io(&svm->vcpu, size, port, svm->sev_es.ghcb_sa, + count, in); } -void sev_es_init_vmcb(struct vcpu_svm *svm) +void sev_es_recalc_msr_intercepts(struct kvm_vcpu *vcpu) +{ + /* Clear intercepts on MSRs that are context switched by hardware. */ + svm_disable_intercept_for_msr(vcpu, MSR_AMD64_SEV_ES_GHCB, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_EFER, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_IA32_CR_PAT, MSR_TYPE_RW); + + if (boot_cpu_has(X86_FEATURE_V_TSC_AUX)) + svm_set_intercept_for_msr(vcpu, MSR_TSC_AUX, MSR_TYPE_RW, + !guest_cpu_cap_has(vcpu, X86_FEATURE_RDTSCP) && + !guest_cpu_cap_has(vcpu, X86_FEATURE_RDPID)); + + svm_set_intercept_for_msr(vcpu, MSR_AMD64_GUEST_TSC_FREQ, MSR_TYPE_R, + !snp_is_secure_tsc_enabled(vcpu->kvm)); + + /* + * For SEV-ES, accesses to MSR_IA32_XSS should not be intercepted if + * the host/guest supports its use. + * + * KVM treats the guest as being capable of using XSAVES even if XSAVES + * isn't enabled in guest CPUID as there is no intercept for XSAVES, + * i.e. the guest can use XSAVES/XRSTOR to read/write XSS if XSAVE is + * exposed to the guest and XSAVES is supported in hardware. Condition + * full XSS passthrough on the guest being able to use XSAVES *and* + * XSAVES being exposed to the guest so that KVM can at least honor + * guest CPUID for RDMSR and WRMSR. + */ + svm_set_intercept_for_msr(vcpu, MSR_IA32_XSS, MSR_TYPE_RW, + !guest_cpu_cap_has(vcpu, X86_FEATURE_XSAVES) || + !guest_cpuid_has(vcpu, X86_FEATURE_XSAVES)); +} + +void sev_vcpu_after_set_cpuid(struct vcpu_svm *svm) { struct kvm_vcpu *vcpu = &svm->vcpu; + struct kvm_cpuid_entry2 *best; + + /* For sev guests, the memory encryption bit is not reserved in CR3. */ + best = kvm_find_cpuid_entry(vcpu, 0x8000001F); + if (best) + vcpu->arch.reserved_gpa_bits &= ~(1UL << (best->ebx & 0x3f)); +} + +static void sev_es_init_vmcb(struct vcpu_svm *svm, bool init_event) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(svm->vcpu.kvm); + struct vmcb *vmcb = svm->vmcb01.ptr; svm->vmcb->control.nested_ctl |= SVM_NESTED_CTL_SEV_ES_ENABLE; - svm->vmcb->control.virt_ext |= LBR_CTL_ENABLE_MASK; /* * An SEV-ES guest requires a VMSA area that is a separate from the * VMCB page. Do not include the encryption mask on the VMSA physical - * address since hardware will access it using the guest key. + * address since hardware will access it using the guest key. Note, + * the VMSA will be NULL if this vCPU is the destination for intrahost + * migration, and will be copied later. */ - svm->vmcb->control.vmsa_pa = __pa(svm->vmsa); + if (!svm->sev_es.snp_has_guest_vmsa) { + if (svm->sev_es.vmsa) + svm->vmcb->control.vmsa_pa = __pa(svm->sev_es.vmsa); + else + svm->vmcb->control.vmsa_pa = INVALID_PAGE; + } + + if (cpu_feature_enabled(X86_FEATURE_ALLOWED_SEV_FEATURES)) + svm->vmcb->control.allowed_sev_features = sev->vmsa_features | + VMCB_ALLOWED_SEV_FEATURES_VALID; /* Can't intercept CR register access, HV can't modify CR registers */ svm_clr_intercept(svm, INTERCEPT_CR0_READ); @@ -2578,53 +4600,140 @@ void sev_es_init_vmcb(struct vcpu_svm *svm) svm_set_intercept(svm, TRAP_CR4_WRITE); svm_set_intercept(svm, TRAP_CR8_WRITE); - /* No support for enable_vmware_backdoor */ - clr_exception_intercept(svm, GP_VECTOR); + vmcb->control.intercepts[INTERCEPT_DR] = 0; + if (!sev_vcpu_has_debug_swap(svm)) { + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_WRITE); + recalc_intercepts(svm); + } else { + /* + * Disable #DB intercept iff DebugSwap is enabled. KVM doesn't + * allow debugging SEV-ES guests, and enables DebugSwap iff + * NO_NESTED_DATA_BP is supported, so there's no reason to + * intercept #DB when DebugSwap is enabled. For simplicity + * with respect to guest debug, intercept #DB for other VMs + * even if NO_NESTED_DATA_BP is supported, i.e. even if the + * guest can't DoS the CPU with infinite #DB vectoring. + */ + clr_exception_intercept(svm, DB_VECTOR); + } /* Can't intercept XSETBV, HV can't modify XCR0 directly */ svm_clr_intercept(svm, INTERCEPT_XSETBV); - /* Clear intercepts on selected MSRs */ - set_msr_interception(vcpu, svm->msrpm, MSR_EFER, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_CR_PAT, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHFROMIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 1, 1); + /* + * Set the GHCB MSR value as per the GHCB specification when emulating + * vCPU RESET for an SEV-ES guest. + */ + if (!init_event) + set_ghcb_msr(svm, GHCB_MSR_SEV_INFO((__u64)sev->ghcb_version, + GHCB_VERSION_MIN, + sev_enc_bit)); } -void sev_es_create_vcpu(struct vcpu_svm *svm) +void sev_init_vmcb(struct vcpu_svm *svm, bool init_event) { + struct kvm_vcpu *vcpu = &svm->vcpu; + + svm->vmcb->control.nested_ctl |= SVM_NESTED_CTL_SEV_ENABLE; + clr_exception_intercept(svm, UD_VECTOR); + /* - * Set the GHCB MSR value as per the GHCB specification when creating - * a vCPU for an SEV-ES guest. + * Don't intercept #GP for SEV guests, e.g. for the VMware backdoor, as + * KVM can't decrypt guest memory to decode the faulting instruction. */ - set_ghcb_msr(svm, GHCB_MSR_SEV_INFO(GHCB_VERSION_MAX, - GHCB_VERSION_MIN, - sev_enc_bit)); + clr_exception_intercept(svm, GP_VECTOR); + + if (init_event && sev_snp_guest(vcpu->kvm)) + sev_snp_init_protected_guest_state(vcpu); + + if (sev_es_guest(vcpu->kvm)) + sev_es_init_vmcb(svm, init_event); } -void sev_es_prepare_guest_switch(struct vcpu_svm *svm, unsigned int cpu) +int sev_vcpu_create(struct kvm_vcpu *vcpu) { - struct svm_cpu_data *sd = per_cpu(svm_data, cpu); - struct vmcb_save_area *hostsa; + struct vcpu_svm *svm = to_svm(vcpu); + struct page *vmsa_page; + + mutex_init(&svm->sev_es.snp_vmsa_mutex); + + if (!sev_es_guest(vcpu->kvm)) + return 0; /* - * As an SEV-ES guest, hardware will restore the host state on VMEXIT, - * of which one step is to perform a VMLOAD. Since hardware does not - * perform a VMSAVE on VMRUN, the host savearea must be updated. + * SEV-ES guests require a separate (from the VMCB) VMSA page used to + * contain the encrypted register state of the guest. */ - vmsave(__sme_page_pa(sd->save_area)); + vmsa_page = snp_safe_alloc_page(); + if (!vmsa_page) + return -ENOMEM; - /* XCR0 is restored on VMEXIT, save the current host value */ - hostsa = (struct vmcb_save_area *)(page_address(sd->save_area) + 0x400); - hostsa->xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); + svm->sev_es.vmsa = page_address(vmsa_page); - /* PKRU is restored on VMEXIT, save the current host value */ + vcpu->arch.guest_tsc_protected = snp_is_secure_tsc_enabled(vcpu->kvm); + + return 0; +} + +void sev_es_prepare_switch_to_guest(struct vcpu_svm *svm, struct sev_es_save_area *hostsa) +{ + struct kvm *kvm = svm->vcpu.kvm; + + /* + * All host state for SEV-ES guests is categorized into three swap types + * based on how it is handled by hardware during a world switch: + * + * A: VMRUN: Host state saved in host save area + * VMEXIT: Host state loaded from host save area + * + * B: VMRUN: Host state _NOT_ saved in host save area + * VMEXIT: Host state loaded from host save area + * + * C: VMRUN: Host state _NOT_ saved in host save area + * VMEXIT: Host state initialized to default(reset) values + * + * Manually save type-B state, i.e. state that is loaded by VMEXIT but + * isn't saved by VMRUN, that isn't already saved by VMSAVE (performed + * by common SVM code). + */ + hostsa->xcr0 = kvm_host.xcr0; hostsa->pkru = read_pkru(); + hostsa->xss = kvm_host.xss; - /* MSR_IA32_XSS is restored on VMEXIT, save the currnet host value */ - hostsa->xss = host_xss; + /* + * If DebugSwap is enabled, debug registers are loaded but NOT saved by + * the CPU (Type-B). If DebugSwap is disabled/unsupported, the CPU does + * not save or load debug registers. Sadly, KVM can't prevent SNP + * guests from lying about DebugSwap on secondary vCPUs, i.e. the + * SEV_FEATURES provided at "AP Create" isn't guaranteed to match what + * the guest has actually enabled (or not!) in the VMSA. + * + * If DebugSwap is *possible*, save the masks so that they're restored + * if the guest enables DebugSwap. But for the DRs themselves, do NOT + * rely on the CPU to restore the host values; KVM will restore them as + * needed in common code, via hw_breakpoint_restore(). Note, KVM does + * NOT support virtualizing Breakpoint Extensions, i.e. the mask MSRs + * don't need to be restored per se, KVM just needs to ensure they are + * loaded with the correct values *if* the CPU writes the MSRs. + */ + if (sev_vcpu_has_debug_swap(svm) || + (sev_snp_guest(kvm) && cpu_feature_enabled(X86_FEATURE_DEBUG_SWAP))) { + hostsa->dr0_addr_mask = amd_get_dr_addr_mask(0); + hostsa->dr1_addr_mask = amd_get_dr_addr_mask(1); + hostsa->dr2_addr_mask = amd_get_dr_addr_mask(2); + hostsa->dr3_addr_mask = amd_get_dr_addr_mask(3); + } + + /* + * TSC_AUX is always virtualized for SEV-ES guests when the feature is + * available, i.e. TSC_AUX is loaded on #VMEXIT from the host save area. + * Set the save area to the current hardware value, i.e. the current + * user return value, so that the correct value is restored on #VMEXIT. + */ + if (cpu_feature_enabled(X86_FEATURE_V_TSC_AUX) && + !WARN_ON_ONCE(tsc_aux_uret_slot < 0)) + hostsa->tsc_aux = kvm_get_user_return_msr(tsc_aux_uret_slot); } void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector) @@ -2632,18 +4741,429 @@ void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector) struct vcpu_svm *svm = to_svm(vcpu); /* First SIPI: Use the values as initially set by the VMM */ - if (!svm->received_first_sipi) { - svm->received_first_sipi = true; + if (!svm->sev_es.received_first_sipi) { + svm->sev_es.received_first_sipi = true; + return; + } + + /* Subsequent SIPI */ + switch (svm->sev_es.ap_reset_hold_type) { + case AP_RESET_HOLD_NAE_EVENT: + /* + * Return from an AP Reset Hold VMGEXIT, where the guest will + * set the CS and RIP. Set SW_EXIT_INFO_2 to a non-zero value. + */ + svm_vmgexit_success(svm, 1); + break; + case AP_RESET_HOLD_MSR_PROTO: + /* + * Return from an AP Reset Hold VMGEXIT, where the guest will + * set the CS and RIP. Set GHCB data field to a non-zero value. + */ + set_ghcb_msr_bits(svm, 1, + GHCB_MSR_AP_RESET_HOLD_RESULT_MASK, + GHCB_MSR_AP_RESET_HOLD_RESULT_POS); + + set_ghcb_msr_bits(svm, GHCB_MSR_AP_RESET_HOLD_RESP, + GHCB_MSR_INFO_MASK, + GHCB_MSR_INFO_POS); + break; + default: + break; + } +} + +struct page *snp_safe_alloc_page_node(int node, gfp_t gfp) +{ + unsigned long pfn; + struct page *p; + + if (!cc_platform_has(CC_ATTR_HOST_SEV_SNP)) + return alloc_pages_node(node, gfp | __GFP_ZERO, 0); + + /* + * Allocate an SNP-safe page to workaround the SNP erratum where + * the CPU will incorrectly signal an RMP violation #PF if a + * hugepage (2MB or 1GB) collides with the RMP entry of a + * 2MB-aligned VMCB, VMSA, or AVIC backing page. + * + * Allocate one extra page, choose a page which is not + * 2MB-aligned, and free the other. + */ + p = alloc_pages_node(node, gfp | __GFP_ZERO, 1); + if (!p) + return NULL; + + split_page(p, 1); + + pfn = page_to_pfn(p); + if (IS_ALIGNED(pfn, PTRS_PER_PMD)) + __free_page(p++); + else + __free_page(p + 1); + + return p; +} + +void sev_handle_rmp_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u64 error_code) +{ + struct kvm_memory_slot *slot; + struct kvm *kvm = vcpu->kvm; + int order, rmp_level, ret; + struct page *page; + bool assigned; + kvm_pfn_t pfn; + gfn_t gfn; + + gfn = gpa >> PAGE_SHIFT; + + /* + * The only time RMP faults occur for shared pages is when the guest is + * triggering an RMP fault for an implicit page-state change from + * shared->private. Implicit page-state changes are forwarded to + * userspace via KVM_EXIT_MEMORY_FAULT events, however, so RMP faults + * for shared pages should not end up here. + */ + if (!kvm_mem_is_private(kvm, gfn)) { + pr_warn_ratelimited("SEV: Unexpected RMP fault for non-private GPA 0x%llx\n", + gpa); + return; + } + + slot = gfn_to_memslot(kvm, gfn); + if (!kvm_slot_has_gmem(slot)) { + pr_warn_ratelimited("SEV: Unexpected RMP fault, non-private slot for GPA 0x%llx\n", + gpa); + return; + } + + ret = kvm_gmem_get_pfn(kvm, slot, gfn, &pfn, &page, &order); + if (ret) { + pr_warn_ratelimited("SEV: Unexpected RMP fault, no backing page for private GPA 0x%llx\n", + gpa); + return; + } + + ret = snp_lookup_rmpentry(pfn, &assigned, &rmp_level); + if (ret || !assigned) { + pr_warn_ratelimited("SEV: Unexpected RMP fault, no assigned RMP entry found for GPA 0x%llx PFN 0x%llx error %d\n", + gpa, pfn, ret); + goto out_no_trace; + } + + /* + * There are 2 cases where a PSMASH may be needed to resolve an #NPF + * with PFERR_GUEST_RMP_BIT set: + * + * 1) RMPADJUST/PVALIDATE can trigger an #NPF with PFERR_GUEST_SIZEM + * bit set if the guest issues them with a smaller granularity than + * what is indicated by the page-size bit in the 2MB RMP entry for + * the PFN that backs the GPA. + * + * 2) Guest access via NPT can trigger an #NPF if the NPT mapping is + * smaller than what is indicated by the 2MB RMP entry for the PFN + * that backs the GPA. + * + * In both these cases, the corresponding 2M RMP entry needs to + * be PSMASH'd to 512 4K RMP entries. If the RMP entry is already + * split into 4K RMP entries, then this is likely a spurious case which + * can occur when there are concurrent accesses by the guest to a 2MB + * GPA range that is backed by a 2MB-aligned PFN who's RMP entry is in + * the process of being PMASH'd into 4K entries. These cases should + * resolve automatically on subsequent accesses, so just ignore them + * here. + */ + if (rmp_level == PG_LEVEL_4K) + goto out; + + ret = snp_rmptable_psmash(pfn); + if (ret) { + /* + * Look it up again. If it's 4K now then the PSMASH may have + * raced with another process and the issue has already resolved + * itself. + */ + if (!snp_lookup_rmpentry(pfn, &assigned, &rmp_level) && + assigned && rmp_level == PG_LEVEL_4K) + goto out; + + pr_warn_ratelimited("SEV: Unable to split RMP entry for GPA 0x%llx PFN 0x%llx ret %d\n", + gpa, pfn, ret); + } + + kvm_zap_gfn_range(kvm, gfn, gfn + PTRS_PER_PMD); +out: + trace_kvm_rmp_fault(vcpu, gpa, pfn, error_code, rmp_level, ret); +out_no_trace: + kvm_release_page_unused(page); +} + +static bool is_pfn_range_shared(kvm_pfn_t start, kvm_pfn_t end) +{ + kvm_pfn_t pfn = start; + + while (pfn < end) { + int ret, rmp_level; + bool assigned; + + ret = snp_lookup_rmpentry(pfn, &assigned, &rmp_level); + if (ret) { + pr_warn_ratelimited("SEV: Failed to retrieve RMP entry: PFN 0x%llx GFN start 0x%llx GFN end 0x%llx RMP level %d error %d\n", + pfn, start, end, rmp_level, ret); + return false; + } + + if (assigned) { + pr_debug("%s: overlap detected, PFN 0x%llx start 0x%llx end 0x%llx RMP level %d\n", + __func__, pfn, start, end, rmp_level); + return false; + } + + pfn++; + } + + return true; +} + +static u8 max_level_for_order(int order) +{ + if (order >= KVM_HPAGE_GFN_SHIFT(PG_LEVEL_2M)) + return PG_LEVEL_2M; + + return PG_LEVEL_4K; +} + +static bool is_large_rmp_possible(struct kvm *kvm, kvm_pfn_t pfn, int order) +{ + kvm_pfn_t pfn_aligned = ALIGN_DOWN(pfn, PTRS_PER_PMD); + + /* + * If this is a large folio, and the entire 2M range containing the + * PFN is currently shared, then the entire 2M-aligned range can be + * set to private via a single 2M RMP entry. + */ + if (max_level_for_order(order) > PG_LEVEL_4K && + is_pfn_range_shared(pfn_aligned, pfn_aligned + PTRS_PER_PMD)) + return true; + + return false; +} + +int sev_gmem_prepare(struct kvm *kvm, kvm_pfn_t pfn, gfn_t gfn, int max_order) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); + kvm_pfn_t pfn_aligned; + gfn_t gfn_aligned; + int level, rc; + bool assigned; + + if (!sev_snp_guest(kvm)) + return 0; + + rc = snp_lookup_rmpentry(pfn, &assigned, &level); + if (rc) { + pr_err_ratelimited("SEV: Failed to look up RMP entry: GFN %llx PFN %llx error %d\n", + gfn, pfn, rc); + return -ENOENT; + } + + if (assigned) { + pr_debug("%s: already assigned: gfn %llx pfn %llx max_order %d level %d\n", + __func__, gfn, pfn, max_order, level); + return 0; + } + + if (is_large_rmp_possible(kvm, pfn, max_order)) { + level = PG_LEVEL_2M; + pfn_aligned = ALIGN_DOWN(pfn, PTRS_PER_PMD); + gfn_aligned = ALIGN_DOWN(gfn, PTRS_PER_PMD); + } else { + level = PG_LEVEL_4K; + pfn_aligned = pfn; + gfn_aligned = gfn; + } + + rc = rmp_make_private(pfn_aligned, gfn_to_gpa(gfn_aligned), level, sev->asid, false); + if (rc) { + pr_err_ratelimited("SEV: Failed to update RMP entry: GFN %llx PFN %llx level %d error %d\n", + gfn, pfn, level, rc); + return -EINVAL; + } + + pr_debug("%s: updated: gfn %llx pfn %llx pfn_aligned %llx max_order %d level %d\n", + __func__, gfn, pfn, pfn_aligned, max_order, level); + + return 0; +} + +void sev_gmem_invalidate(kvm_pfn_t start, kvm_pfn_t end) +{ + kvm_pfn_t pfn; + + if (!cc_platform_has(CC_ATTR_HOST_SEV_SNP)) return; + + pr_debug("%s: PFN start 0x%llx PFN end 0x%llx\n", __func__, start, end); + + for (pfn = start; pfn < end;) { + bool use_2m_update = false; + int rc, rmp_level; + bool assigned; + + rc = snp_lookup_rmpentry(pfn, &assigned, &rmp_level); + if (rc || !assigned) + goto next_pfn; + + use_2m_update = IS_ALIGNED(pfn, PTRS_PER_PMD) && + end >= (pfn + PTRS_PER_PMD) && + rmp_level > PG_LEVEL_4K; + + /* + * If an unaligned PFN corresponds to a 2M region assigned as a + * large page in the RMP table, PSMASH the region into individual + * 4K RMP entries before attempting to convert a 4K sub-page. + */ + if (!use_2m_update && rmp_level > PG_LEVEL_4K) { + /* + * This shouldn't fail, but if it does, report it, but + * still try to update RMP entry to shared and pray this + * was a spurious error that can be addressed later. + */ + rc = snp_rmptable_psmash(pfn); + WARN_ONCE(rc, "SEV: Failed to PSMASH RMP entry for PFN 0x%llx error %d\n", + pfn, rc); + } + + rc = rmp_make_shared(pfn, use_2m_update ? PG_LEVEL_2M : PG_LEVEL_4K); + if (WARN_ONCE(rc, "SEV: Failed to update RMP entry for PFN 0x%llx error %d\n", + pfn, rc)) + goto next_pfn; + + /* + * SEV-ES avoids host/guest cache coherency issues through + * WBNOINVD hooks issued via MMU notifiers during run-time, and + * KVM's VM destroy path at shutdown. Those MMU notifier events + * don't cover gmem since there is no requirement to map pages + * to a HVA in order to use them for a running guest. While the + * shutdown path would still likely cover things for SNP guests, + * userspace may also free gmem pages during run-time via + * hole-punching operations on the guest_memfd, so flush the + * cache entries for these pages before free'ing them back to + * the host. + */ + clflush_cache_range(__va(pfn_to_hpa(pfn)), + use_2m_update ? PMD_SIZE : PAGE_SIZE); +next_pfn: + pfn += use_2m_update ? PTRS_PER_PMD : 1; + cond_resched(); } +} + +int sev_gmem_max_mapping_level(struct kvm *kvm, kvm_pfn_t pfn, bool is_private) +{ + int level, rc; + bool assigned; + + if (!sev_snp_guest(kvm)) + return 0; + + rc = snp_lookup_rmpentry(pfn, &assigned, &level); + if (rc || !assigned) + return PG_LEVEL_4K; + + return level; +} + +struct vmcb_save_area *sev_decrypt_vmsa(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb_save_area *vmsa; + struct kvm_sev_info *sev; + int error = 0; + int ret; + + if (!sev_es_guest(vcpu->kvm)) + return NULL; /* - * Subsequent SIPI: Return from an AP Reset Hold VMGEXIT, where - * the guest will set the CS and RIP. Set SW_EXIT_INFO_2 to a - * non-zero value. + * If the VMSA has not yet been encrypted, return a pointer to the + * current un-encrypted VMSA. */ - if (!svm->ghcb) + if (!vcpu->arch.guest_state_protected) + return (struct vmcb_save_area *)svm->sev_es.vmsa; + + sev = to_kvm_sev_info(vcpu->kvm); + + /* Check if the SEV policy allows debugging */ + if (sev_snp_guest(vcpu->kvm)) { + if (!(sev->policy & SNP_POLICY_MASK_DEBUG)) + return NULL; + } else { + if (sev->policy & SEV_POLICY_MASK_NODBG) + return NULL; + } + + if (sev_snp_guest(vcpu->kvm)) { + struct sev_data_snp_dbg dbg = {0}; + + vmsa = snp_alloc_firmware_page(__GFP_ZERO); + if (!vmsa) + return NULL; + + dbg.gctx_paddr = __psp_pa(sev->snp_context); + dbg.src_addr = svm->vmcb->control.vmsa_pa; + dbg.dst_addr = __psp_pa(vmsa); + + ret = sev_do_cmd(SEV_CMD_SNP_DBG_DECRYPT, &dbg, &error); + + /* + * Return the target page to a hypervisor page no matter what. + * If this fails, the page can't be used, so leak it and don't + * try to use it. + */ + if (snp_page_reclaim(vcpu->kvm, PHYS_PFN(__pa(vmsa)))) + return NULL; + + if (ret) { + pr_err("SEV: SNP_DBG_DECRYPT failed ret=%d, fw_error=%d (%#x)\n", + ret, error, error); + free_page((unsigned long)vmsa); + + return NULL; + } + } else { + struct sev_data_dbg dbg = {0}; + struct page *vmsa_page; + + vmsa_page = alloc_page(GFP_KERNEL); + if (!vmsa_page) + return NULL; + + vmsa = page_address(vmsa_page); + + dbg.handle = sev->handle; + dbg.src_addr = svm->vmcb->control.vmsa_pa; + dbg.dst_addr = __psp_pa(vmsa); + dbg.len = PAGE_SIZE; + + ret = sev_do_cmd(SEV_CMD_DBG_DECRYPT, &dbg, &error); + if (ret) { + pr_err("SEV: SEV_CMD_DBG_DECRYPT failed ret=%d, fw_error=%d (0x%x)\n", + ret, error, error); + __free_page(vmsa_page); + + return NULL; + } + } + + return vmsa; +} + +void sev_free_decrypted_vmsa(struct kvm_vcpu *vcpu, struct vmcb_save_area *vmsa) +{ + /* If the VMSA has not yet been encrypted, nothing was allocated */ + if (!vcpu->arch.guest_state_protected || !vmsa) return; - ghcb_set_sw_exit_info_2(svm->ghcb, 1); + free_page((unsigned long)vmsa); } diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 8834822c00cd..f56c2d895011 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -1,4 +1,4 @@ -#define pr_fmt(fmt) "SVM: " fmt +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kvm_host.h> @@ -6,6 +6,7 @@ #include "mmu.h" #include "kvm_cache_regs.h" #include "x86.h" +#include "smm.h" #include "cpuid.h" #include "pmu.h" @@ -25,8 +26,13 @@ #include <linux/pagemap.h> #include <linux/swap.h> #include <linux/rwsem.h> +#include <linux/cc_platform.h> +#include <linux/smp.h> +#include <linux/string_choices.h> +#include <linux/mutex.h> #include <asm/apic.h> +#include <asm/msr.h> #include <asm/perf_event.h> #include <asm/tlbflush.h> #include <asm/desc.h> @@ -36,8 +42,11 @@ #include <asm/spec-ctrl.h> #include <asm/cpu_device_id.h> #include <asm/traps.h> +#include <asm/reboot.h> +#include <asm/fpu/api.h> + +#include <trace/events/ipi.h> -#include <asm/virtext.h> #include "trace.h" #include "svm.h" @@ -46,9 +55,8 @@ #include "kvm_onhyperv.h" #include "svm_onhyperv.h" -#define __ex(x) __kvm_handle_fault_on_reboot(x) - MODULE_AUTHOR("Qumranet"); +MODULE_DESCRIPTION("KVM support for SVM (AMD-V) extensions"); MODULE_LICENSE("GPL"); #ifdef MODULE @@ -62,24 +70,8 @@ MODULE_DEVICE_TABLE(x86cpu, svm_cpu_id); #define SEG_TYPE_LDT 2 #define SEG_TYPE_BUSY_TSS16 3 -#define SVM_FEATURE_LBRV (1 << 1) -#define SVM_FEATURE_SVML (1 << 2) -#define SVM_FEATURE_TSC_RATE (1 << 4) -#define SVM_FEATURE_VMCB_CLEAN (1 << 5) -#define SVM_FEATURE_FLUSH_ASID (1 << 6) -#define SVM_FEATURE_DECODE_ASSIST (1 << 7) -#define SVM_FEATURE_PAUSE_FILTER (1 << 10) - -#define DEBUGCTL_RESERVED_BITS (~(0x3fULL)) - -#define TSC_RATIO_RSVD 0xffffff0000000000ULL -#define TSC_RATIO_MIN 0x0000000000000001ULL -#define TSC_RATIO_MAX 0x000000ffffffffffULL - static bool erratum_383_found __read_mostly; -u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly; - /* * Set osvw_len to higher value when updated Revision Guides * are published and we know what the new status bits are @@ -87,35 +79,6 @@ u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly; static uint64_t osvw_len = 4, osvw_status; static DEFINE_PER_CPU(u64, current_tsc_ratio); -#define TSC_RATIO_DEFAULT 0x0100000000ULL - -static const struct svm_direct_access_msrs { - u32 index; /* Index of the MSR */ - bool always; /* True if intercept is initially cleared */ -} direct_access_msrs[MAX_DIRECT_ACCESS_MSRS] = { - { .index = MSR_STAR, .always = true }, - { .index = MSR_IA32_SYSENTER_CS, .always = true }, - { .index = MSR_IA32_SYSENTER_EIP, .always = false }, - { .index = MSR_IA32_SYSENTER_ESP, .always = false }, -#ifdef CONFIG_X86_64 - { .index = MSR_GS_BASE, .always = true }, - { .index = MSR_FS_BASE, .always = true }, - { .index = MSR_KERNEL_GS_BASE, .always = true }, - { .index = MSR_LSTAR, .always = true }, - { .index = MSR_CSTAR, .always = true }, - { .index = MSR_SYSCALL_MASK, .always = true }, -#endif - { .index = MSR_IA32_SPEC_CTRL, .always = false }, - { .index = MSR_IA32_PRED_CMD, .always = false }, - { .index = MSR_IA32_LASTBRANCHFROMIP, .always = false }, - { .index = MSR_IA32_LASTBRANCHTOIP, .always = false }, - { .index = MSR_IA32_LASTINTFROMIP, .always = false }, - { .index = MSR_IA32_LASTINTTOIP, .always = false }, - { .index = MSR_EFER, .always = false }, - { .index = MSR_IA32_CR_PAT, .always = false }, - { .index = MSR_AMD64_SEV_ES_GHCB, .always = true }, - { .index = MSR_INVALID, .always = false }, -}; /* * These 2 parameters are used to config the controls for Pause-Loop Exiting: @@ -174,10 +137,10 @@ module_param_named(npt, npt_enabled, bool, 0444); /* allow nested virtualization in KVM/SVM */ static int nested = true; -module_param(nested, int, S_IRUGO); +module_param(nested, int, 0444); /* enable/disable Next RIP Save */ -static int nrips = true; +int nrips = true; module_param(nrips, int, 0444); /* enable/disable Virtual VMLOAD VMSAVE */ @@ -185,35 +148,37 @@ static int vls = true; module_param(vls, int, 0444); /* enable/disable Virtual GIF */ -static int vgif = true; +int vgif = true; module_param(vgif, int, 0444); -/* - * enable / disable AVIC. Because the defaults differ for APICv - * support between VMX and SVM we cannot use module_param_named. - */ -static bool avic; -module_param(avic, bool, 0444); +/* enable/disable LBR virtualization */ +int lbrv = true; +module_param(lbrv, int, 0444); + +static int tsc_scaling = true; +module_param(tsc_scaling, int, 0444); + +module_param(enable_device_posted_irqs, bool, 0444); bool __read_mostly dump_invalid_vmcb; module_param(dump_invalid_vmcb, bool, 0644); + +bool intercept_smi = true; +module_param(intercept_smi, bool, 0444); + +bool vnmi = true; +module_param(vnmi, bool, 0444); + static bool svm_gp_erratum_intercept = true; static u8 rsm_ins_bytes[] = "\x0f\xaa"; static unsigned long iopm_base; -struct kvm_ldttss_desc { - u16 limit0; - u16 base0; - unsigned base1:8, type:5, dpl:2, p:1; - unsigned limit1:4, zero0:3, g:1, base2:8; - u32 base3; - u32 zero1; -} __attribute__((packed)); +DEFINE_PER_CPU(struct svm_cpu_data, svm_data); -DEFINE_PER_CPU(struct svm_cpu_data *, svm_data); +static DEFINE_MUTEX(vmcb_dump_mutex); /* * Only MSR_TSC_AUX is switched via the user return hook. EFER is switched via @@ -222,41 +187,12 @@ DEFINE_PER_CPU(struct svm_cpu_data *, svm_data); * RDTSCP and RDPID are not used in the kernel, specifically to allow KVM to * defer the restoration of TSC_AUX until the CPU returns to userspace. */ -static int tsc_aux_uret_slot __read_mostly = -1; - -static const u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000}; - -#define NUM_MSR_MAPS ARRAY_SIZE(msrpm_ranges) -#define MSRS_RANGE_SIZE 2048 -#define MSRS_IN_RANGE (MSRS_RANGE_SIZE * 8 / 2) - -u32 svm_msrpm_offset(u32 msr) -{ - u32 offset; - int i; - - for (i = 0; i < NUM_MSR_MAPS; i++) { - if (msr < msrpm_ranges[i] || - msr >= msrpm_ranges[i] + MSRS_IN_RANGE) - continue; +int tsc_aux_uret_slot __ro_after_init = -1; - offset = (msr - msrpm_ranges[i]) / 4; /* 4 msrs per u8 */ - offset += (i * MSRS_RANGE_SIZE); /* add range offset */ - - /* Now we have the u8 offset - but need the u32 offset */ - return offset / 4; - } - - /* MSR not in any range */ - return MSR_INVALID; -} - -#define MAX_INST_SIZE 15 - -static int get_max_npt_level(void) +static int get_npt_level(void) { #ifdef CONFIG_X86_64 - return PT64_ROOT_4LEVEL; + return pgtable_l5_enabled() ? PT64_ROOT_5LEVEL : PT64_ROOT_4LEVEL; #else return PT32E_ROOT_LEVEL; #endif @@ -278,7 +214,7 @@ int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) if ((old_efer & EFER_SVME) != (efer & EFER_SVME)) { if (!(efer & EFER_SVME)) { - svm_leave_nested(svm); + svm_leave_nested(vcpu); svm_set_gif(svm, true); /* #GP intercept is still needed for vmware backdoor */ if (!enable_vmware_backdoor) @@ -300,7 +236,11 @@ int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) return ret; } - if (svm_gp_erratum_intercept) + /* + * Never intercept #GP for SEV guests, KVM can't + * decrypt guest memory to workaround the erratum. + */ + if (svm_gp_erratum_intercept && !sev_guest(vcpu->kvm)) set_exception_intercept(svm, GP_VECTOR); } } @@ -310,12 +250,6 @@ int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) return 0; } -static int is_external_interrupt(u32 info) -{ - info &= SVM_EVTINJ_TYPE_MASK | SVM_EVTINJ_VALID; - return info == (SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR); -} - static u32 svm_get_interrupt_shadow(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -337,9 +271,12 @@ static void svm_set_interrupt_shadow(struct kvm_vcpu *vcpu, int mask) } -static int skip_emulated_instruction(struct kvm_vcpu *vcpu) +static int __svm_skip_emulated_instruction(struct kvm_vcpu *vcpu, + int emul_type, + bool commit_side_effects) { struct vcpu_svm *svm = to_svm(vcpu); + unsigned long old_rflags; /* * SEV-ES does not expose the next RIP. The RIP update is controlled by @@ -354,70 +291,109 @@ static int skip_emulated_instruction(struct kvm_vcpu *vcpu) } if (!svm->next_rip) { - if (!kvm_emulate_instruction(vcpu, EMULTYPE_SKIP)) + if (unlikely(!commit_side_effects)) + old_rflags = svm->vmcb->save.rflags; + + if (!kvm_emulate_instruction(vcpu, emul_type)) return 0; + + if (unlikely(!commit_side_effects)) + svm->vmcb->save.rflags = old_rflags; } else { kvm_rip_write(vcpu, svm->next_rip); } done: - svm_set_interrupt_shadow(vcpu, 0); + if (likely(commit_side_effects)) + svm_set_interrupt_shadow(vcpu, 0); return 1; } -static void svm_queue_exception(struct kvm_vcpu *vcpu) +static int svm_skip_emulated_instruction(struct kvm_vcpu *vcpu) +{ + return __svm_skip_emulated_instruction(vcpu, EMULTYPE_SKIP, true); +} + +static int svm_update_soft_interrupt_rip(struct kvm_vcpu *vcpu, u8 vector) { + const int emul_type = EMULTYPE_SKIP | EMULTYPE_SKIP_SOFT_INT | + EMULTYPE_SET_SOFT_INT_VECTOR(vector); + unsigned long rip, old_rip = kvm_rip_read(vcpu); struct vcpu_svm *svm = to_svm(vcpu); - unsigned nr = vcpu->arch.exception.nr; - bool has_error_code = vcpu->arch.exception.has_error_code; - u32 error_code = vcpu->arch.exception.error_code; - kvm_deliver_exception_payload(vcpu); + /* + * Due to architectural shortcomings, the CPU doesn't always provide + * NextRIP, e.g. if KVM intercepted an exception that occurred while + * the CPU was vectoring an INTO/INT3 in the guest. Temporarily skip + * the instruction even if NextRIP is supported to acquire the next + * RIP so that it can be shoved into the NextRIP field, otherwise + * hardware will fail to advance guest RIP during event injection. + * Drop the exception/interrupt if emulation fails and effectively + * retry the instruction, it's the least awful option. If NRIPS is + * in use, the skip must not commit any side effects such as clearing + * the interrupt shadow or RFLAGS.RF. + */ + if (!__svm_skip_emulated_instruction(vcpu, emul_type, !nrips)) + return -EIO; - if (nr == BP_VECTOR && !nrips) { - unsigned long rip, old_rip = kvm_rip_read(vcpu); + rip = kvm_rip_read(vcpu); - /* - * For guest debugging where we have to reinject #BP if some - * INT3 is guest-owned: - * Emulate nRIP by moving RIP forward. Will fail if injection - * raises a fault that is not intercepted. Still better than - * failing in all cases. - */ - (void)skip_emulated_instruction(vcpu); - rip = kvm_rip_read(vcpu); - svm->int3_rip = rip + svm->vmcb->save.cs.base; - svm->int3_injected = rip - old_rip; - } + /* + * Save the injection information, even when using next_rip, as the + * VMCB's next_rip will be lost (cleared on VM-Exit) if the injection + * doesn't complete due to a VM-Exit occurring while the CPU is + * vectoring the event. Decoding the instruction isn't guaranteed to + * work as there may be no backing instruction, e.g. if the event is + * being injected by L1 for L2, or if the guest is patching INT3 into + * a different instruction. + */ + svm->soft_int_injected = true; + svm->soft_int_csbase = svm->vmcb->save.cs.base; + svm->soft_int_old_rip = old_rip; + svm->soft_int_next_rip = rip; + + if (nrips) + kvm_rip_write(vcpu, old_rip); + + if (static_cpu_has(X86_FEATURE_NRIPS)) + svm->vmcb->control.next_rip = rip; + + return 0; +} + +static void svm_inject_exception(struct kvm_vcpu *vcpu) +{ + struct kvm_queued_exception *ex = &vcpu->arch.exception; + struct vcpu_svm *svm = to_svm(vcpu); + + kvm_deliver_exception_payload(vcpu, ex); - svm->vmcb->control.event_inj = nr + if (kvm_exception_is_soft(ex->vector) && + svm_update_soft_interrupt_rip(vcpu, ex->vector)) + return; + + svm->vmcb->control.event_inj = ex->vector | SVM_EVTINJ_VALID - | (has_error_code ? SVM_EVTINJ_VALID_ERR : 0) + | (ex->has_error_code ? SVM_EVTINJ_VALID_ERR : 0) | SVM_EVTINJ_TYPE_EXEPT; - svm->vmcb->control.event_inj_err = error_code; + svm->vmcb->control.event_inj_err = ex->error_code; } static void svm_init_erratum_383(void) { - u32 low, high; - int err; u64 val; if (!static_cpu_has_bug(X86_BUG_AMD_TLB_MMATCH)) return; /* Use _safe variants to not break nested virtualization */ - val = native_read_msr_safe(MSR_AMD64_DC_CFG, &err); - if (err) + if (native_read_msr_safe(MSR_AMD64_DC_CFG, &val)) return; val |= (1ULL << 47); - low = lower_32_bits(val); - high = upper_32_bits(val); - - native_write_msr_safe(MSR_AMD64_DC_CFG, low, high); + native_write_msr_safe(MSR_AMD64_DC_CFG, val); erratum_383_found = true; } @@ -443,76 +419,124 @@ static void svm_init_osvw(struct kvm_vcpu *vcpu) vcpu->arch.osvw.status |= 1; } -static int has_svm(void) +static bool __kvm_is_svm_supported(void) { - const char *msg; + int cpu = smp_processor_id(); + struct cpuinfo_x86 *c = &cpu_data(cpu); - if (!cpu_has_svm(&msg)) { - printk(KERN_INFO "has_svm: %s\n", msg); - return 0; + if (c->x86_vendor != X86_VENDOR_AMD && + c->x86_vendor != X86_VENDOR_HYGON) { + pr_err("CPU %d isn't AMD or Hygon\n", cpu); + return false; } - if (sev_active()) { + if (!cpu_has(c, X86_FEATURE_SVM)) { + pr_err("SVM not supported by CPU %d\n", cpu); + return false; + } + + if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) { pr_info("KVM is unsupported when running as an SEV guest\n"); - return 0; + return false; } - if (pgtable_l5_enabled()) { - pr_info("KVM doesn't yet support 5-level paging on AMD SVM\n"); - return 0; + return true; +} + +static bool kvm_is_svm_supported(void) +{ + bool supported; + + migrate_disable(); + supported = __kvm_is_svm_supported(); + migrate_enable(); + + return supported; +} + +static int svm_check_processor_compat(void) +{ + if (!__kvm_is_svm_supported()) + return -EIO; + + return 0; +} + +static void __svm_write_tsc_multiplier(u64 multiplier) +{ + if (multiplier == __this_cpu_read(current_tsc_ratio)) + return; + + wrmsrq(MSR_AMD64_TSC_RATIO, multiplier); + __this_cpu_write(current_tsc_ratio, multiplier); +} + +static __always_inline struct sev_es_save_area *sev_es_host_save_area(struct svm_cpu_data *sd) +{ + return &sd->save_area->host_sev_es_save; +} + +static inline void kvm_cpu_svm_disable(void) +{ + uint64_t efer; + + wrmsrq(MSR_VM_HSAVE_PA, 0); + rdmsrq(MSR_EFER, efer); + if (efer & EFER_SVME) { + /* + * Force GIF=1 prior to disabling SVM, e.g. to ensure INIT and + * NMI aren't blocked. + */ + stgi(); + wrmsrq(MSR_EFER, efer & ~EFER_SVME); } +} - return 1; +static void svm_emergency_disable_virtualization_cpu(void) +{ + kvm_rebooting = true; + + kvm_cpu_svm_disable(); } -static void svm_hardware_disable(void) +static void svm_disable_virtualization_cpu(void) { /* Make sure we clean up behind us */ - if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) - wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT); + if (tsc_scaling) + __svm_write_tsc_multiplier(SVM_TSC_RATIO_DEFAULT); - cpu_svm_disable(); + kvm_cpu_svm_disable(); amd_pmu_disable_virt(); } -static int svm_hardware_enable(void) +static int svm_enable_virtualization_cpu(void) { struct svm_cpu_data *sd; uint64_t efer; - struct desc_struct *gdt; int me = raw_smp_processor_id(); - rdmsrl(MSR_EFER, efer); + rdmsrq(MSR_EFER, efer); if (efer & EFER_SVME) return -EBUSY; - if (!has_svm()) { - pr_err("%s: err EOPNOTSUPP on %d\n", __func__, me); - return -EINVAL; - } - sd = per_cpu(svm_data, me); - if (!sd) { - pr_err("%s: svm_data is NULL on %d\n", __func__, me); - return -EINVAL; - } - + sd = per_cpu_ptr(&svm_data, me); sd->asid_generation = 1; sd->max_asid = cpuid_ebx(SVM_CPUID_FUNC) - 1; sd->next_asid = sd->max_asid + 1; sd->min_asid = max_sev_asid + 1; - gdt = get_current_gdt_rw(); - sd->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS); - - wrmsrl(MSR_EFER, efer | EFER_SVME); + wrmsrq(MSR_EFER, efer | EFER_SVME); - wrmsrl(MSR_VM_HSAVE_PA, __sme_page_pa(sd->save_area)); + wrmsrq(MSR_VM_HSAVE_PA, sd->save_area_pa); if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) { - wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT); - __this_cpu_write(current_tsc_ratio, TSC_RATIO_DEFAULT); + /* + * Set the default value, even if we don't use TSC scaling + * to avoid having stale value in the msr + */ + __svm_write_tsc_multiplier(SVM_TSC_RATIO_DEFAULT); } @@ -526,13 +550,12 @@ static int svm_hardware_enable(void) * erratum is present everywhere). */ if (cpu_has(&boot_cpu_data, X86_FEATURE_OSVW)) { - uint64_t len, status = 0; + u64 len, status = 0; int err; - len = native_read_msr_safe(MSR_AMD64_OSVW_ID_LENGTH, &err); + err = native_read_msr_safe(MSR_AMD64_OSVW_ID_LENGTH, &len); if (!err) - status = native_read_msr_safe(MSR_AMD64_OSVW_STATUS, - &err); + err = native_read_msr_safe(MSR_AMD64_OSVW_STATUS, &status); if (err) osvw_status = osvw_len = 0; @@ -554,261 +577,281 @@ static int svm_hardware_enable(void) static void svm_cpu_uninit(int cpu) { - struct svm_cpu_data *sd = per_cpu(svm_data, cpu); + struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu); - if (!sd) + if (!sd->save_area) return; - per_cpu(svm_data, cpu) = NULL; kfree(sd->sev_vmcbs); - __free_page(sd->save_area); - kfree(sd); + __free_page(__sme_pa_to_page(sd->save_area_pa)); + sd->save_area_pa = 0; + sd->save_area = NULL; } static int svm_cpu_init(int cpu) { - struct svm_cpu_data *sd; + struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu); + struct page *save_area_page; int ret = -ENOMEM; - sd = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL); - if (!sd) + memset(sd, 0, sizeof(struct svm_cpu_data)); + save_area_page = snp_safe_alloc_page_node(cpu_to_node(cpu), GFP_KERNEL); + if (!save_area_page) return ret; - sd->cpu = cpu; - sd->save_area = alloc_page(GFP_KERNEL); - if (!sd->save_area) - goto free_cpu_data; - - clear_page(page_address(sd->save_area)); ret = sev_cpu_init(sd); if (ret) goto free_save_area; - per_cpu(svm_data, cpu) = sd; - + sd->save_area = page_address(save_area_page); + sd->save_area_pa = __sme_page_pa(save_area_page); return 0; free_save_area: - __free_page(sd->save_area); -free_cpu_data: - kfree(sd); + __free_page(save_area_page); return ret; } -static int direct_access_msr_slot(u32 msr) +static void set_dr_intercepts(struct vcpu_svm *svm) { - u32 i; + struct vmcb *vmcb = svm->vmcb01.ptr; - for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++) - if (direct_access_msrs[i].index == msr) - return i; + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR0_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR1_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR2_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR3_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR4_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR5_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR6_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR0_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR1_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR2_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR3_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR4_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR5_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR6_WRITE); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_READ); + vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_WRITE); - return -ENOENT; + recalc_intercepts(svm); } -static void set_shadow_msr_intercept(struct kvm_vcpu *vcpu, u32 msr, int read, - int write) +static void clr_dr_intercepts(struct vcpu_svm *svm) { - struct vcpu_svm *svm = to_svm(vcpu); - int slot = direct_access_msr_slot(msr); + struct vmcb *vmcb = svm->vmcb01.ptr; - if (slot == -ENOENT) - return; - - /* Set the shadow bitmaps to the desired intercept states */ - if (read) - set_bit(slot, svm->shadow_msr_intercept.read); - else - clear_bit(slot, svm->shadow_msr_intercept.read); + vmcb->control.intercepts[INTERCEPT_DR] = 0; - if (write) - set_bit(slot, svm->shadow_msr_intercept.write); - else - clear_bit(slot, svm->shadow_msr_intercept.write); -} - -static bool valid_msr_intercept(u32 index) -{ - return direct_access_msr_slot(index) != -ENOENT; + recalc_intercepts(svm); } static bool msr_write_intercepted(struct kvm_vcpu *vcpu, u32 msr) { - u8 bit_write; - unsigned long tmp; - u32 offset; - u32 *msrpm; - - msrpm = is_guest_mode(vcpu) ? to_svm(vcpu)->nested.msrpm: - to_svm(vcpu)->msrpm; - - offset = svm_msrpm_offset(msr); - bit_write = 2 * (msr & 0x0f) + 1; - tmp = msrpm[offset]; - - BUG_ON(offset == MSR_INVALID); - - return !!test_bit(bit_write, &tmp); -} - -static void set_msr_interception_bitmap(struct kvm_vcpu *vcpu, u32 *msrpm, - u32 msr, int read, int write) -{ - u8 bit_read, bit_write; - unsigned long tmp; - u32 offset; - /* - * If this warning triggers extend the direct_access_msrs list at the - * beginning of the file + * For non-nested case: + * If the L01 MSR bitmap does not intercept the MSR, then we need to + * save it. + * + * For nested case: + * If the L02 MSR bitmap does not intercept the MSR, then we need to + * save it. */ - WARN_ON(!valid_msr_intercept(msr)); - - /* Enforce non allowed MSRs to trap */ - if (read && !kvm_msr_allowed(vcpu, msr, KVM_MSR_FILTER_READ)) - read = 0; - - if (write && !kvm_msr_allowed(vcpu, msr, KVM_MSR_FILTER_WRITE)) - write = 0; + void *msrpm = is_guest_mode(vcpu) ? to_svm(vcpu)->nested.msrpm : + to_svm(vcpu)->msrpm; - offset = svm_msrpm_offset(msr); - bit_read = 2 * (msr & 0x0f); - bit_write = 2 * (msr & 0x0f) + 1; - tmp = msrpm[offset]; + return svm_test_msr_bitmap_write(msrpm, msr); +} - BUG_ON(offset == MSR_INVALID); +void svm_set_intercept_for_msr(struct kvm_vcpu *vcpu, u32 msr, int type, bool set) +{ + struct vcpu_svm *svm = to_svm(vcpu); + void *msrpm = svm->msrpm; - read ? clear_bit(bit_read, &tmp) : set_bit(bit_read, &tmp); - write ? clear_bit(bit_write, &tmp) : set_bit(bit_write, &tmp); + /* Don't disable interception for MSRs userspace wants to handle. */ + if (type & MSR_TYPE_R) { + if (!set && kvm_msr_allowed(vcpu, msr, KVM_MSR_FILTER_READ)) + svm_clear_msr_bitmap_read(msrpm, msr); + else + svm_set_msr_bitmap_read(msrpm, msr); + } - msrpm[offset] = tmp; + if (type & MSR_TYPE_W) { + if (!set && kvm_msr_allowed(vcpu, msr, KVM_MSR_FILTER_WRITE)) + svm_clear_msr_bitmap_write(msrpm, msr); + else + svm_set_msr_bitmap_write(msrpm, msr); + } svm_hv_vmcb_dirty_nested_enlightenments(vcpu); - -} - -void set_msr_interception(struct kvm_vcpu *vcpu, u32 *msrpm, u32 msr, - int read, int write) -{ - set_shadow_msr_intercept(vcpu, msr, read, write); - set_msr_interception_bitmap(vcpu, msrpm, msr, read, write); + svm->nested.force_msr_bitmap_recalc = true; } -u32 *svm_vcpu_alloc_msrpm(void) +void *svm_alloc_permissions_map(unsigned long size, gfp_t gfp_mask) { - unsigned int order = get_order(MSRPM_SIZE); - struct page *pages = alloc_pages(GFP_KERNEL_ACCOUNT, order); - u32 *msrpm; + unsigned int order = get_order(size); + struct page *pages = alloc_pages(gfp_mask, order); + void *pm; if (!pages) return NULL; - msrpm = page_address(pages); - memset(msrpm, 0xff, PAGE_SIZE * (1 << order)); + /* + * Set all bits in the permissions map so that all MSR and I/O accesses + * are intercepted by default. + */ + pm = page_address(pages); + memset(pm, 0xff, PAGE_SIZE * (1 << order)); - return msrpm; + return pm; } -void svm_vcpu_init_msrpm(struct kvm_vcpu *vcpu, u32 *msrpm) +static void svm_recalc_lbr_msr_intercepts(struct kvm_vcpu *vcpu) { - int i; + struct vcpu_svm *svm = to_svm(vcpu); + bool intercept = !(svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK); - for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++) { - if (!direct_access_msrs[i].always) - continue; - set_msr_interception(vcpu, msrpm, direct_access_msrs[i].index, 1, 1); - } -} + if (intercept == svm->lbr_msrs_intercepted) + return; + + svm_set_intercept_for_msr(vcpu, MSR_IA32_LASTBRANCHFROMIP, MSR_TYPE_RW, intercept); + svm_set_intercept_for_msr(vcpu, MSR_IA32_LASTBRANCHTOIP, MSR_TYPE_RW, intercept); + svm_set_intercept_for_msr(vcpu, MSR_IA32_LASTINTFROMIP, MSR_TYPE_RW, intercept); + svm_set_intercept_for_msr(vcpu, MSR_IA32_LASTINTTOIP, MSR_TYPE_RW, intercept); + if (sev_es_guest(vcpu->kvm)) + svm_set_intercept_for_msr(vcpu, MSR_IA32_DEBUGCTLMSR, MSR_TYPE_RW, intercept); + + svm->lbr_msrs_intercepted = intercept; +} -void svm_vcpu_free_msrpm(u32 *msrpm) +void svm_vcpu_free_msrpm(void *msrpm) { __free_pages(virt_to_page(msrpm), get_order(MSRPM_SIZE)); } -static void svm_msr_filter_changed(struct kvm_vcpu *vcpu) +static void svm_recalc_msr_intercepts(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - u32 i; - /* - * Set intercept permissions for all direct access MSRs again. They - * will automatically get filtered through the MSR filter, so we are - * back in sync after this. - */ - for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++) { - u32 msr = direct_access_msrs[i].index; - u32 read = test_bit(i, svm->shadow_msr_intercept.read); - u32 write = test_bit(i, svm->shadow_msr_intercept.write); + svm_disable_intercept_for_msr(vcpu, MSR_STAR, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_IA32_SYSENTER_CS, MSR_TYPE_RW); - set_msr_interception_bitmap(vcpu, svm->msrpm, msr, read, write); - } -} +#ifdef CONFIG_X86_64 + svm_disable_intercept_for_msr(vcpu, MSR_GS_BASE, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_FS_BASE, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_KERNEL_GS_BASE, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_LSTAR, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_CSTAR, MSR_TYPE_RW); + svm_disable_intercept_for_msr(vcpu, MSR_SYSCALL_MASK, MSR_TYPE_RW); +#endif -static void add_msr_offset(u32 offset) -{ - int i; + if (lbrv) + svm_recalc_lbr_msr_intercepts(vcpu); - for (i = 0; i < MSRPM_OFFSETS; ++i) { + if (cpu_feature_enabled(X86_FEATURE_IBPB)) + svm_set_intercept_for_msr(vcpu, MSR_IA32_PRED_CMD, MSR_TYPE_W, + !guest_has_pred_cmd_msr(vcpu)); - /* Offset already in list? */ - if (msrpm_offsets[i] == offset) - return; + if (cpu_feature_enabled(X86_FEATURE_FLUSH_L1D)) + svm_set_intercept_for_msr(vcpu, MSR_IA32_FLUSH_CMD, MSR_TYPE_W, + !guest_cpu_cap_has(vcpu, X86_FEATURE_FLUSH_L1D)); + + /* + * Disable interception of SPEC_CTRL if KVM doesn't need to manually + * context switch the MSR (SPEC_CTRL is virtualized by the CPU), or if + * the guest has a non-zero SPEC_CTRL value, i.e. is likely actively + * using SPEC_CTRL. + */ + if (cpu_feature_enabled(X86_FEATURE_V_SPEC_CTRL)) + svm_set_intercept_for_msr(vcpu, MSR_IA32_SPEC_CTRL, MSR_TYPE_RW, + !guest_has_spec_ctrl_msr(vcpu)); + else + svm_set_intercept_for_msr(vcpu, MSR_IA32_SPEC_CTRL, MSR_TYPE_RW, + !svm->spec_ctrl); - /* Slot used by another offset? */ - if (msrpm_offsets[i] != MSR_INVALID) - continue; + /* + * Intercept SYSENTER_EIP and SYSENTER_ESP when emulating an Intel CPU, + * as AMD hardware only store 32 bits, whereas Intel CPUs track 64 bits. + */ + svm_set_intercept_for_msr(vcpu, MSR_IA32_SYSENTER_EIP, MSR_TYPE_RW, + guest_cpuid_is_intel_compatible(vcpu)); + svm_set_intercept_for_msr(vcpu, MSR_IA32_SYSENTER_ESP, MSR_TYPE_RW, + guest_cpuid_is_intel_compatible(vcpu)); + + if (kvm_aperfmperf_in_guest(vcpu->kvm)) { + svm_disable_intercept_for_msr(vcpu, MSR_IA32_APERF, MSR_TYPE_R); + svm_disable_intercept_for_msr(vcpu, MSR_IA32_MPERF, MSR_TYPE_R); + } - /* Add offset to list */ - msrpm_offsets[i] = offset; + if (kvm_cpu_cap_has(X86_FEATURE_SHSTK)) { + bool shstk_enabled = guest_cpu_cap_has(vcpu, X86_FEATURE_SHSTK); - return; + svm_set_intercept_for_msr(vcpu, MSR_IA32_U_CET, MSR_TYPE_RW, !shstk_enabled); + svm_set_intercept_for_msr(vcpu, MSR_IA32_S_CET, MSR_TYPE_RW, !shstk_enabled); + svm_set_intercept_for_msr(vcpu, MSR_IA32_PL0_SSP, MSR_TYPE_RW, !shstk_enabled); + svm_set_intercept_for_msr(vcpu, MSR_IA32_PL1_SSP, MSR_TYPE_RW, !shstk_enabled); + svm_set_intercept_for_msr(vcpu, MSR_IA32_PL2_SSP, MSR_TYPE_RW, !shstk_enabled); + svm_set_intercept_for_msr(vcpu, MSR_IA32_PL3_SSP, MSR_TYPE_RW, !shstk_enabled); } + if (sev_es_guest(vcpu->kvm)) + sev_es_recalc_msr_intercepts(vcpu); + /* - * If this BUG triggers the msrpm_offsets table has an overflow. Just - * increase MSRPM_OFFSETS in this case. + * x2APIC intercepts are modified on-demand and cannot be filtered by + * userspace. */ - BUG(); } -static void init_msrpm_offsets(void) +void svm_copy_lbrs(struct vmcb *to_vmcb, struct vmcb *from_vmcb) { - int i; + to_vmcb->save.dbgctl = from_vmcb->save.dbgctl; + to_vmcb->save.br_from = from_vmcb->save.br_from; + to_vmcb->save.br_to = from_vmcb->save.br_to; + to_vmcb->save.last_excp_from = from_vmcb->save.last_excp_from; + to_vmcb->save.last_excp_to = from_vmcb->save.last_excp_to; - memset(msrpm_offsets, 0xff, sizeof(msrpm_offsets)); - - for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++) { - u32 offset; - - offset = svm_msrpm_offset(direct_access_msrs[i].index); - BUG_ON(offset == MSR_INVALID); + vmcb_mark_dirty(to_vmcb, VMCB_LBR); +} - add_msr_offset(offset); - } +static void __svm_enable_lbrv(struct kvm_vcpu *vcpu) +{ + to_svm(vcpu)->vmcb->control.virt_ext |= LBR_CTL_ENABLE_MASK; } -static void svm_enable_lbrv(struct kvm_vcpu *vcpu) +void svm_enable_lbrv(struct kvm_vcpu *vcpu) { - struct vcpu_svm *svm = to_svm(vcpu); + __svm_enable_lbrv(vcpu); + svm_recalc_lbr_msr_intercepts(vcpu); +} - svm->vmcb->control.virt_ext |= LBR_CTL_ENABLE_MASK; - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHFROMIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 1, 1); +static void __svm_disable_lbrv(struct kvm_vcpu *vcpu) +{ + KVM_BUG_ON(sev_es_guest(vcpu->kvm), vcpu->kvm); + to_svm(vcpu)->vmcb->control.virt_ext &= ~LBR_CTL_ENABLE_MASK; } -static void svm_disable_lbrv(struct kvm_vcpu *vcpu) +void svm_update_lbrv(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); + bool current_enable_lbrv = svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK; + bool enable_lbrv = (svm->vmcb->save.dbgctl & DEBUGCTLMSR_LBR) || + (is_guest_mode(vcpu) && guest_cpu_cap_has(vcpu, X86_FEATURE_LBRV) && + (svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK)); - svm->vmcb->control.virt_ext &= ~LBR_CTL_ENABLE_MASK; - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHFROMIP, 0, 0); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 0, 0); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 0, 0); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 0, 0); + if (enable_lbrv && !current_enable_lbrv) + __svm_enable_lbrv(vcpu); + else if (!enable_lbrv && current_enable_lbrv) + __svm_disable_lbrv(vcpu); + + /* + * During nested transitions, it is possible that the current VMCB has + * LBR_CTL set, but the previous LBR_CTL had it cleared (or vice versa). + * In this case, even though LBR_CTL does not need an update, intercepts + * do, so always recalculate the intercepts here. + */ + svm_recalc_lbr_msr_intercepts(vcpu); } void disable_nmi_singlestep(struct vcpu_svm *svm) @@ -830,6 +873,9 @@ static void grow_ple_window(struct kvm_vcpu *vcpu) struct vmcb_control_area *control = &svm->vmcb->control; int old = control->pause_filter_count; + if (kvm_pause_in_guest(vcpu->kvm)) + return; + control->pause_filter_count = __grow_ple_window(old, pause_filter_count, pause_filter_count_grow, @@ -848,6 +894,9 @@ static void shrink_ple_window(struct kvm_vcpu *vcpu) struct vmcb_control_area *control = &svm->vmcb->control; int old = control->pause_filter_count; + if (kvm_pause_in_guest(vcpu->kvm)) + return; + control->pause_filter_count = __shrink_ple_window(old, pause_filter_count, @@ -860,229 +909,21 @@ static void shrink_ple_window(struct kvm_vcpu *vcpu) } } -/* - * The default MMIO mask is a single bit (excluding the present bit), - * which could conflict with the memory encryption bit. Check for - * memory encryption support and override the default MMIO mask if - * memory encryption is enabled. - */ -static __init void svm_adjust_mmio_mask(void) -{ - unsigned int enc_bit, mask_bit; - u64 msr, mask; - - /* If there is no memory encryption support, use existing mask */ - if (cpuid_eax(0x80000000) < 0x8000001f) - return; - - /* If memory encryption is not enabled, use existing mask */ - rdmsrl(MSR_AMD64_SYSCFG, msr); - if (!(msr & MSR_AMD64_SYSCFG_MEM_ENCRYPT)) - return; - - enc_bit = cpuid_ebx(0x8000001f) & 0x3f; - mask_bit = boot_cpu_data.x86_phys_bits; - - /* Increment the mask bit if it is the same as the encryption bit */ - if (enc_bit == mask_bit) - mask_bit++; - - /* - * If the mask bit location is below 52, then some bits above the - * physical addressing limit will always be reserved, so use the - * rsvd_bits() function to generate the mask. This mask, along with - * the present bit, will be used to generate a page fault with - * PFER.RSV = 1. - * - * If the mask bit location is 52 (or above), then clear the mask. - */ - mask = (mask_bit < 52) ? rsvd_bits(mask_bit, 51) | PT_PRESENT_MASK : 0; - - kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK); -} - -static void svm_hardware_teardown(void) +static void svm_hardware_unsetup(void) { int cpu; - sev_hardware_teardown(); + avic_hardware_unsetup(); + + sev_hardware_unsetup(); for_each_possible_cpu(cpu) svm_cpu_uninit(cpu); - __free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), - get_order(IOPM_SIZE)); + __free_pages(__sme_pa_to_page(iopm_base), get_order(IOPM_SIZE)); iopm_base = 0; } -static __init void svm_set_cpu_caps(void) -{ - kvm_set_cpu_caps(); - - supported_xss = 0; - - /* CPUID 0x80000001 and 0x8000000A (SVM features) */ - if (nested) { - kvm_cpu_cap_set(X86_FEATURE_SVM); - - if (nrips) - kvm_cpu_cap_set(X86_FEATURE_NRIPS); - - if (npt_enabled) - kvm_cpu_cap_set(X86_FEATURE_NPT); - - /* Nested VM can receive #VMEXIT instead of triggering #GP */ - kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK); - } - - /* CPUID 0x80000008 */ - if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD) || - boot_cpu_has(X86_FEATURE_AMD_SSBD)) - kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD); - - /* CPUID 0x8000001F (SME/SEV features) */ - sev_set_cpu_caps(); -} - -static __init int svm_hardware_setup(void) -{ - int cpu; - struct page *iopm_pages; - void *iopm_va; - int r; - unsigned int order = get_order(IOPM_SIZE); - - /* - * NX is required for shadow paging and for NPT if the NX huge pages - * mitigation is enabled. - */ - if (!boot_cpu_has(X86_FEATURE_NX)) { - pr_err_ratelimited("NX (Execute Disable) not supported\n"); - return -EOPNOTSUPP; - } - kvm_enable_efer_bits(EFER_NX); - - iopm_pages = alloc_pages(GFP_KERNEL, order); - - if (!iopm_pages) - return -ENOMEM; - - iopm_va = page_address(iopm_pages); - memset(iopm_va, 0xff, PAGE_SIZE * (1 << order)); - iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT; - - init_msrpm_offsets(); - - supported_xcr0 &= ~(XFEATURE_MASK_BNDREGS | XFEATURE_MASK_BNDCSR); - - if (boot_cpu_has(X86_FEATURE_FXSR_OPT)) - kvm_enable_efer_bits(EFER_FFXSR); - - if (boot_cpu_has(X86_FEATURE_TSCRATEMSR)) { - kvm_has_tsc_control = true; - kvm_max_tsc_scaling_ratio = TSC_RATIO_MAX; - kvm_tsc_scaling_ratio_frac_bits = 32; - } - - tsc_aux_uret_slot = kvm_add_user_return_msr(MSR_TSC_AUX); - - /* Check for pause filtering support */ - if (!boot_cpu_has(X86_FEATURE_PAUSEFILTER)) { - pause_filter_count = 0; - pause_filter_thresh = 0; - } else if (!boot_cpu_has(X86_FEATURE_PFTHRESHOLD)) { - pause_filter_thresh = 0; - } - - if (nested) { - printk(KERN_INFO "kvm: Nested Virtualization enabled\n"); - kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE); - } - - /* - * KVM's MMU doesn't support using 2-level paging for itself, and thus - * NPT isn't supported if the host is using 2-level paging since host - * CR4 is unchanged on VMRUN. - */ - if (!IS_ENABLED(CONFIG_X86_64) && !IS_ENABLED(CONFIG_X86_PAE)) - npt_enabled = false; - - if (!boot_cpu_has(X86_FEATURE_NPT)) - npt_enabled = false; - - kvm_configure_mmu(npt_enabled, get_max_npt_level(), PG_LEVEL_1G); - pr_info("kvm: Nested Paging %sabled\n", npt_enabled ? "en" : "dis"); - - /* Note, SEV setup consumes npt_enabled. */ - sev_hardware_setup(); - - svm_hv_hardware_setup(); - - svm_adjust_mmio_mask(); - - for_each_possible_cpu(cpu) { - r = svm_cpu_init(cpu); - if (r) - goto err; - } - - if (nrips) { - if (!boot_cpu_has(X86_FEATURE_NRIPS)) - nrips = false; - } - - enable_apicv = avic = avic && npt_enabled && boot_cpu_has(X86_FEATURE_AVIC); - - if (enable_apicv) { - pr_info("AVIC enabled\n"); - - amd_iommu_register_ga_log_notifier(&avic_ga_log_notifier); - } - - if (vls) { - if (!npt_enabled || - !boot_cpu_has(X86_FEATURE_V_VMSAVE_VMLOAD) || - !IS_ENABLED(CONFIG_X86_64)) { - vls = false; - } else { - pr_info("Virtual VMLOAD VMSAVE supported\n"); - } - } - - if (boot_cpu_has(X86_FEATURE_SVME_ADDR_CHK)) - svm_gp_erratum_intercept = false; - - if (vgif) { - if (!boot_cpu_has(X86_FEATURE_VGIF)) - vgif = false; - else - pr_info("Virtual GIF supported\n"); - } - - svm_set_cpu_caps(); - - /* - * It seems that on AMD processors PTE's accessed bit is - * being set by the CPU hardware before the NPF vmexit. - * This is not expected behaviour and our tests fail because - * of it. - * A workaround here is to disable support for - * GUEST_MAXPHYADDR < HOST_MAXPHYADDR if NPT is enabled. - * In this case userspace can know if there is support using - * KVM_CAP_SMALLER_MAXPHYADDR extension and decide how to handle - * it - * If future AMD CPU models change the behaviour described above, - * this variable can be changed accordingly - */ - allow_smaller_maxphyaddr = !npt_enabled; - - return 0; - -err: - svm_hardware_teardown(); - return r; -} - static void init_seg(struct vmcb_seg *seg) { seg->selector = 0; @@ -1109,54 +950,81 @@ static u64 svm_get_l2_tsc_offset(struct kvm_vcpu *vcpu) static u64 svm_get_l2_tsc_multiplier(struct kvm_vcpu *vcpu) { - return kvm_default_tsc_scaling_ratio; + struct vcpu_svm *svm = to_svm(vcpu); + + return svm->tsc_ratio_msr; } -static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset) +static void svm_write_tsc_offset(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); svm->vmcb01.ptr->control.tsc_offset = vcpu->arch.l1_tsc_offset; - svm->vmcb->control.tsc_offset = offset; + svm->vmcb->control.tsc_offset = vcpu->arch.tsc_offset; vmcb_mark_dirty(svm->vmcb, VMCB_INTERCEPTS); } -static void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu, u64 multiplier) +void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu) { - wrmsrl(MSR_AMD64_TSC_RATIO, multiplier); + preempt_disable(); + if (to_svm(vcpu)->guest_state_loaded) + __svm_write_tsc_multiplier(vcpu->arch.tsc_scaling_ratio); + preempt_enable(); } /* Evaluate instruction intercepts that depend on guest CPUID features. */ -static void svm_recalc_instruction_intercepts(struct kvm_vcpu *vcpu, - struct vcpu_svm *svm) +static void svm_recalc_instruction_intercepts(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); + /* * Intercept INVPCID if shadow paging is enabled to sync/free shadow * roots, or if INVPCID is disabled in the guest to inject #UD. */ if (kvm_cpu_cap_has(X86_FEATURE_INVPCID)) { if (!npt_enabled || - !guest_cpuid_has(&svm->vcpu, X86_FEATURE_INVPCID)) + !guest_cpu_cap_has(&svm->vcpu, X86_FEATURE_INVPCID)) svm_set_intercept(svm, INTERCEPT_INVPCID); else svm_clr_intercept(svm, INTERCEPT_INVPCID); } if (kvm_cpu_cap_has(X86_FEATURE_RDTSCP)) { - if (guest_cpuid_has(vcpu, X86_FEATURE_RDTSCP)) + if (guest_cpu_cap_has(vcpu, X86_FEATURE_RDTSCP)) svm_clr_intercept(svm, INTERCEPT_RDTSCP); else svm_set_intercept(svm, INTERCEPT_RDTSCP); } + + if (guest_cpuid_is_intel_compatible(vcpu)) { + svm_set_intercept(svm, INTERCEPT_VMLOAD); + svm_set_intercept(svm, INTERCEPT_VMSAVE); + svm->vmcb->control.virt_ext &= ~VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; + } else { + /* + * If hardware supports Virtual VMLOAD VMSAVE then enable it + * in VMCB and clear intercepts to avoid #VMEXIT. + */ + if (vls) { + svm_clr_intercept(svm, INTERCEPT_VMLOAD); + svm_clr_intercept(svm, INTERCEPT_VMSAVE); + svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; + } + } } -static void init_vmcb(struct kvm_vcpu *vcpu) +static void svm_recalc_intercepts(struct kvm_vcpu *vcpu) { - struct vcpu_svm *svm = to_svm(vcpu); - struct vmcb_control_area *control = &svm->vmcb->control; - struct vmcb_save_area *save = &svm->vmcb->save; + svm_recalc_instruction_intercepts(vcpu); + svm_recalc_msr_intercepts(vcpu); +} - vcpu->arch.hflags = 0; +static void init_vmcb(struct kvm_vcpu *vcpu, bool init_event) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb *vmcb = svm->vmcb01.ptr; + struct vmcb_control_area *control = &vmcb->control; + struct vmcb_save_area *save = &vmcb->save; svm_set_intercept(svm, INTERCEPT_CR0_READ); svm_set_intercept(svm, INTERCEPT_CR3_READ); @@ -1185,7 +1053,10 @@ static void init_vmcb(struct kvm_vcpu *vcpu) svm_set_intercept(svm, INTERCEPT_INTR); svm_set_intercept(svm, INTERCEPT_NMI); - svm_set_intercept(svm, INTERCEPT_SMI); + + if (intercept_smi) + svm_set_intercept(svm, INTERCEPT_SMI); + svm_set_intercept(svm, INTERCEPT_SELECTIVE_CR0); svm_set_intercept(svm, INTERCEPT_RDPMC); svm_set_intercept(svm, INTERCEPT_CPUID); @@ -1213,10 +1084,14 @@ static void init_vmcb(struct kvm_vcpu *vcpu) svm_set_intercept(svm, INTERCEPT_MWAIT); } - if (!kvm_hlt_in_guest(vcpu->kvm)) - svm_set_intercept(svm, INTERCEPT_HLT); + if (!kvm_hlt_in_guest(vcpu->kvm)) { + if (cpu_feature_enabled(X86_FEATURE_IDLE_HLT)) + svm_set_intercept(svm, INTERCEPT_IDLE_HLT); + else + svm_set_intercept(svm, INTERCEPT_HLT); + } - control->iopm_base_pa = __sme_set(iopm_base); + control->iopm_base_pa = iopm_base; control->msrpm_base_pa = __sme_set(__pa(svm->msrpm)); control->int_ctl = V_INTR_MASKING_MASK; @@ -1233,29 +1108,14 @@ static void init_vmcb(struct kvm_vcpu *vcpu) SVM_SELECTOR_S_MASK | SVM_SELECTOR_CODE_MASK; save->cs.limit = 0xffff; + save->gdtr.base = 0; save->gdtr.limit = 0xffff; + save->idtr.base = 0; save->idtr.limit = 0xffff; init_sys_seg(&save->ldtr, SEG_TYPE_LDT); init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16); - svm_set_cr4(vcpu, 0); - svm_set_efer(vcpu, 0); - save->dr6 = 0xffff0ff0; - kvm_set_rflags(vcpu, X86_EFLAGS_FIXED); - save->rip = 0x0000fff0; - vcpu->arch.regs[VCPU_REGS_RIP] = save->rip; - - /* - * svm_set_cr0() sets PG and WP and clears NW and CD on save->cr0. - * It also updates the guest-visible cr0 value. - */ - svm_set_cr0(vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); - kvm_mmu_reset_context(vcpu); - - save->cr4 = X86_CR4_PAE; - /* rdx = ?? */ - if (npt_enabled) { /* Setup VMCB for Nested Paging */ control->nested_ctl |= SVM_NESTED_CTL_NP_ENABLE; @@ -1265,14 +1125,12 @@ static void init_vmcb(struct kvm_vcpu *vcpu) svm_clr_intercept(svm, INTERCEPT_CR3_WRITE); save->g_pat = vcpu->arch.pat; save->cr3 = 0; - save->cr4 = 0; } svm->current_vmcb->asid_generation = 0; svm->asid = 0; svm->nested.vmcb12_gpa = INVALID_GPA; svm->nested.last_vmcb12_gpa = INVALID_GPA; - vcpu->arch.hflags = 0; if (!kvm_pause_in_guest(vcpu->kvm)) { control->pause_filter_count = pause_filter_count; @@ -1283,17 +1141,11 @@ static void init_vmcb(struct kvm_vcpu *vcpu) svm_clr_intercept(svm, INTERCEPT_PAUSE); } - svm_recalc_instruction_intercepts(vcpu, svm); - - /* - * If the host supports V_SPEC_CTRL then disable the interception - * of MSR_IA32_SPEC_CTRL. - */ - if (boot_cpu_has(X86_FEATURE_V_SPEC_CTRL)) - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SPEC_CTRL, 1, 1); - if (kvm_vcpu_apicv_active(vcpu)) - avic_init_vmcb(svm); + avic_init_vmcb(svm, vmcb); + + if (vnmi) + svm->vmcb->control.int_ctl |= V_NMI_ENABLE_MASK; if (vgif) { svm_clr_intercept(svm, INTERCEPT_STGI); @@ -1301,46 +1153,46 @@ static void init_vmcb(struct kvm_vcpu *vcpu) svm->vmcb->control.int_ctl |= V_GIF_ENABLE_MASK; } - if (sev_guest(vcpu->kvm)) { - svm->vmcb->control.nested_ctl |= SVM_NESTED_CTL_SEV_ENABLE; - clr_exception_intercept(svm, UD_VECTOR); + if (vcpu->kvm->arch.bus_lock_detection_enabled) + svm_set_intercept(svm, INTERCEPT_BUSLOCK); - if (sev_es_guest(vcpu->kvm)) { - /* Perform SEV-ES specific VMCB updates */ - sev_es_init_vmcb(svm); - } - } + if (sev_guest(vcpu->kvm)) + sev_init_vmcb(svm, init_event); + + svm_hv_init_vmcb(vmcb); - svm_hv_init_vmcb(svm->vmcb); + kvm_make_request(KVM_REQ_RECALC_INTERCEPTS, vcpu); - vmcb_mark_all_dirty(svm->vmcb); + vmcb_mark_all_dirty(vmcb); enable_gif(svm); +} +static void __svm_vcpu_reset(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + svm_init_osvw(vcpu); + + if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_STUFF_FEATURE_MSRS)) + vcpu->arch.microcode_version = 0x01000065; + svm->tsc_ratio_msr = kvm_caps.default_tsc_scaling_ratio; + + svm->nmi_masked = false; + svm->awaiting_iret_completion = false; } static void svm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) { struct vcpu_svm *svm = to_svm(vcpu); - u32 dummy; - u32 eax = 1; svm->spec_ctrl = 0; svm->virt_spec_ctrl = 0; - if (!init_event) { - vcpu->arch.apic_base = APIC_DEFAULT_PHYS_BASE | - MSR_IA32_APICBASE_ENABLE; - if (kvm_vcpu_is_reset_bsp(vcpu)) - vcpu->arch.apic_base |= MSR_IA32_APICBASE_BSP; - } - init_vmcb(vcpu); - - kvm_cpuid(vcpu, &eax, &dummy, &dummy, &dummy, false); - kvm_rdx_write(vcpu, eax); + init_vmcb(vcpu, init_event); - if (kvm_vcpu_apicv_active(vcpu) && !init_event) - avic_update_vapic_bar(svm, APIC_DEFAULT_PHYS_BASE); + if (!init_event) + __svm_vcpu_reset(vcpu); } void svm_switch_vmcb(struct vcpu_svm *svm, struct kvm_vmcb_info *target_vmcb) @@ -1349,117 +1201,134 @@ void svm_switch_vmcb(struct vcpu_svm *svm, struct kvm_vmcb_info *target_vmcb) svm->vmcb = target_vmcb->ptr; } -static int svm_create_vcpu(struct kvm_vcpu *vcpu) +static int svm_vcpu_precreate(struct kvm *kvm) +{ + return avic_alloc_physical_id_table(kvm); +} + +static int svm_vcpu_create(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm; struct page *vmcb01_page; - struct page *vmsa_page = NULL; int err; BUILD_BUG_ON(offsetof(struct vcpu_svm, vcpu) != 0); svm = to_svm(vcpu); err = -ENOMEM; - vmcb01_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); + vmcb01_page = snp_safe_alloc_page(); if (!vmcb01_page) goto out; - if (sev_es_guest(vcpu->kvm)) { - /* - * SEV-ES guests require a separate VMSA page used to contain - * the encrypted register state of the guest. - */ - vmsa_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); - if (!vmsa_page) - goto error_free_vmcb_page; - - /* - * SEV-ES guests maintain an encrypted version of their FPU - * state which is restored and saved on VMRUN and VMEXIT. - * Free the fpu structure to prevent KVM from attempting to - * access the FPU state. - */ - kvm_free_guest_fpu(vcpu); - } + err = sev_vcpu_create(vcpu); + if (err) + goto error_free_vmcb_page; err = avic_init_vcpu(svm); if (err) - goto error_free_vmsa_page; - - /* We initialize this flag to true to make sure that the is_running - * bit would be set the first time the vcpu is loaded. - */ - if (irqchip_in_kernel(vcpu->kvm) && kvm_apicv_activated(vcpu->kvm)) - svm->avic_is_running = true; + goto error_free_sev; svm->msrpm = svm_vcpu_alloc_msrpm(); if (!svm->msrpm) { err = -ENOMEM; - goto error_free_vmsa_page; + goto error_free_sev; } - svm_vcpu_init_msrpm(vcpu, svm->msrpm); + svm->x2avic_msrs_intercepted = true; + svm->lbr_msrs_intercepted = true; svm->vmcb01.ptr = page_address(vmcb01_page); svm->vmcb01.pa = __sme_set(page_to_pfn(vmcb01_page) << PAGE_SHIFT); - - if (vmsa_page) - svm->vmsa = page_address(vmsa_page); - - svm->guest_state_loaded = false; - svm_switch_vmcb(svm, &svm->vmcb01); - init_vmcb(vcpu); - - svm_init_osvw(vcpu); - vcpu->arch.microcode_version = 0x01000065; - if (sev_es_guest(vcpu->kvm)) - /* Perform SEV-ES specific VMCB creation updates */ - sev_es_create_vcpu(svm); + svm->guest_state_loaded = false; return 0; -error_free_vmsa_page: - if (vmsa_page) - __free_page(vmsa_page); +error_free_sev: + sev_free_vcpu(vcpu); error_free_vmcb_page: __free_page(vmcb01_page); out: return err; } -static void svm_clear_current_vmcb(struct vmcb *vmcb) +static void svm_vcpu_free(struct kvm_vcpu *vcpu) { - int i; + struct vcpu_svm *svm = to_svm(vcpu); - for_each_online_cpu(i) - cmpxchg(&per_cpu(svm_data, i)->current_vmcb, vmcb, NULL); + WARN_ON_ONCE(!list_empty(&svm->ir_list)); + + svm_leave_nested(vcpu); + svm_free_nested(svm); + + sev_free_vcpu(vcpu); + + __free_page(__sme_pa_to_page(svm->vmcb01.pa)); + svm_vcpu_free_msrpm(svm->msrpm); } -static void svm_free_vcpu(struct kvm_vcpu *vcpu) +#ifdef CONFIG_CPU_MITIGATIONS +static DEFINE_SPINLOCK(srso_lock); +static atomic_t srso_nr_vms; + +static void svm_srso_clear_bp_spec_reduce(void *ign) { - struct vcpu_svm *svm = to_svm(vcpu); + struct svm_cpu_data *sd = this_cpu_ptr(&svm_data); + + if (!sd->bp_spec_reduce_set) + return; + + msr_clear_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT); + sd->bp_spec_reduce_set = false; +} + +static void svm_srso_vm_destroy(void) +{ + if (!cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) + return; + + if (atomic_dec_return(&srso_nr_vms)) + return; + + guard(spinlock)(&srso_lock); /* - * The vmcb page can be recycled, causing a false negative in - * svm_vcpu_load(). So, ensure that no logical CPU has this - * vmcb page recorded as its current vmcb. + * Verify a new VM didn't come along, acquire the lock, and increment + * the count before this task acquired the lock. */ - svm_clear_current_vmcb(svm->vmcb); + if (atomic_read(&srso_nr_vms)) + return; - svm_free_nested(svm); + on_each_cpu(svm_srso_clear_bp_spec_reduce, NULL, 1); +} - sev_free_vcpu(vcpu); +static void svm_srso_vm_init(void) +{ + if (!cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) + return; - __free_page(pfn_to_page(__sme_clr(svm->vmcb01.pa) >> PAGE_SHIFT)); - __free_pages(virt_to_page(svm->msrpm), get_order(MSRPM_SIZE)); + /* + * Acquire the lock on 0 => 1 transitions to ensure a potential 1 => 0 + * transition, i.e. destroying the last VM, is fully complete, e.g. so + * that a delayed IPI doesn't clear BP_SPEC_REDUCE after a vCPU runs. + */ + if (atomic_inc_not_zero(&srso_nr_vms)) + return; + + guard(spinlock)(&srso_lock); + + atomic_inc(&srso_nr_vms); } +#else +static void svm_srso_vm_init(void) { } +static void svm_srso_vm_destroy(void) { } +#endif -static void svm_prepare_guest_switch(struct kvm_vcpu *vcpu) +static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); + struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu); if (sev_es_guest(vcpu->kvm)) sev_es_unmap_ghcb(svm); @@ -1471,23 +1340,28 @@ static void svm_prepare_guest_switch(struct kvm_vcpu *vcpu) * Save additional host state that will be restored on VMEXIT (sev-es) * or subsequent vmload of host save area. */ - if (sev_es_guest(vcpu->kvm)) { - sev_es_prepare_guest_switch(svm, vcpu->cpu); - } else { - vmsave(__sme_page_pa(sd->save_area)); - } + vmsave(sd->save_area_pa); + if (sev_es_guest(vcpu->kvm)) + sev_es_prepare_switch_to_guest(svm, sev_es_host_save_area(sd)); - if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) { - u64 tsc_ratio = vcpu->arch.tsc_scaling_ratio; - if (tsc_ratio != __this_cpu_read(current_tsc_ratio)) { - __this_cpu_write(current_tsc_ratio, tsc_ratio); - wrmsrl(MSR_AMD64_TSC_RATIO, tsc_ratio); - } - } + if (tsc_scaling) + __svm_write_tsc_multiplier(vcpu->arch.tsc_scaling_ratio); - if (likely(tsc_aux_uret_slot >= 0)) + /* + * TSC_AUX is always virtualized (context switched by hardware) for + * SEV-ES guests when the feature is available. For non-SEV-ES guests, + * context switch TSC_AUX via the user_return MSR infrastructure (not + * all CPUs support TSC_AUX virtualization). + */ + if (likely(tsc_aux_uret_slot >= 0) && + (!boot_cpu_has(X86_FEATURE_V_TSC_AUX) || !sev_es_guest(vcpu->kvm))) kvm_set_user_return_msr(tsc_aux_uret_slot, svm->tsc_aux, -1ull); + if (cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE) && + !sd->bp_spec_reduce_set) { + sd->bp_spec_reduce_set = true; + msr_set_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT); + } svm->guest_state_loaded = true; } @@ -1498,19 +1372,18 @@ static void svm_prepare_host_switch(struct kvm_vcpu *vcpu) static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { - struct vcpu_svm *svm = to_svm(vcpu); - struct svm_cpu_data *sd = per_cpu(svm_data, cpu); + if (vcpu->scheduled_out && !kvm_pause_in_guest(vcpu->kvm)) + shrink_ple_window(vcpu); - if (sd->current_vmcb != svm->vmcb) { - sd->current_vmcb = svm->vmcb; - indirect_branch_prediction_barrier(); - } - avic_vcpu_load(vcpu, cpu); + if (kvm_vcpu_apicv_active(vcpu)) + avic_vcpu_load(vcpu, cpu); } static void svm_vcpu_put(struct kvm_vcpu *vcpu) { - avic_vcpu_put(vcpu); + if (kvm_vcpu_apicv_active(vcpu)) + avic_vcpu_put(vcpu); + svm_prepare_host_switch(vcpu); ++vcpu->stat.host_state_reload; @@ -1544,15 +1417,30 @@ static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) to_svm(vcpu)->vmcb->save.rflags = rflags; } +static bool svm_get_if_flag(struct kvm_vcpu *vcpu) +{ + struct vmcb *vmcb = to_svm(vcpu)->vmcb; + + return sev_es_guest(vcpu->kvm) + ? vmcb->control.int_state & SVM_GUEST_INTERRUPT_MASK + : kvm_get_rflags(vcpu) & X86_EFLAGS_IF; +} + static void svm_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) { + kvm_register_mark_available(vcpu, reg); + switch (reg) { case VCPU_EXREG_PDPTR: - BUG_ON(!npt_enabled); - load_pdptrs(vcpu, vcpu->arch.walk_mmu, kvm_read_cr3(vcpu)); + /* + * When !npt_enabled, mmu->pdptrs[] is already available since + * it is always updated per SDM when moving to CRs. + */ + if (npt_enabled) + load_pdptrs(vcpu, kvm_read_cr3(vcpu)); break; default: - WARN_ON_ONCE(1); + KVM_BUG_ON(1, vcpu->kvm); } } @@ -1560,11 +1448,24 @@ static void svm_set_vintr(struct vcpu_svm *svm) { struct vmcb_control_area *control; - /* The following fields are ignored when AVIC is enabled */ - WARN_ON(kvm_vcpu_apicv_active(&svm->vcpu)); + /* + * The following fields are ignored when AVIC is enabled + */ + WARN_ON(kvm_vcpu_apicv_activated(&svm->vcpu)); + svm_set_intercept(svm, INTERCEPT_VINTR); /* + * Recalculating intercepts may have cleared the VINTR intercept. If + * V_INTR_MASKING is enabled in vmcb12, then the effective RFLAGS.IF + * for L1 physical interrupts is L1's RFLAGS.IF at the time of VMRUN. + * Requesting an interrupt window if save.RFLAGS.IF=0 is pointless as + * interrupts will never be unblocked while L2 is running. + */ + if (!svm_is_intercept(svm, INTERCEPT_VINTR)) + return; + + /* * This is just a dummy VINTR to actually cause a vmexit to happen. * Actual injection of virtual interrupts happens through EVENTINJ. */ @@ -1578,17 +1479,20 @@ static void svm_set_vintr(struct vcpu_svm *svm) static void svm_clear_vintr(struct vcpu_svm *svm) { - const u32 mask = V_TPR_MASK | V_GIF_ENABLE_MASK | V_GIF_MASK | V_INTR_MASKING_MASK; svm_clr_intercept(svm, INTERCEPT_VINTR); /* Drop int_ctl fields related to VINTR injection. */ - svm->vmcb->control.int_ctl &= mask; + svm->vmcb->control.int_ctl &= ~V_IRQ_INJECTION_BITS_MASK; if (is_guest_mode(&svm->vcpu)) { - svm->vmcb01.ptr->control.int_ctl &= mask; + svm->vmcb01.ptr->control.int_ctl &= ~V_IRQ_INJECTION_BITS_MASK; WARN_ON((svm->vmcb->control.int_ctl & V_TPR_MASK) != (svm->nested.ctl.int_ctl & V_TPR_MASK)); - svm->vmcb->control.int_ctl |= svm->nested.ctl.int_ctl & ~mask; + + svm->vmcb->control.int_ctl |= svm->nested.ctl.int_ctl & + V_IRQ_INJECTION_BITS_MASK; + + svm->vmcb->control.int_vector = svm->nested.ctl.int_vector; } vmcb_mark_dirty(svm->vmcb, VMCB_INTR); @@ -1696,6 +1600,15 @@ static int svm_get_cpl(struct kvm_vcpu *vcpu) return save->cpl; } +static void svm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) +{ + struct kvm_segment cs; + + svm_get_segment(vcpu, &cs, VCPU_SREG_CS); + *db = cs.db; + *l = cs.l; +} + static void svm_get_idt(struct kvm_vcpu *vcpu, struct desc_ptr *dt) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1730,28 +1643,57 @@ static void svm_set_gdt(struct kvm_vcpu *vcpu, struct desc_ptr *dt) vmcb_mark_dirty(svm->vmcb, VMCB_DT); } +static void sev_post_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + /* + * For guests that don't set guest_state_protected, the cr3 update is + * handled via kvm_mmu_load() while entering the guest. For guests + * that do (SEV-ES/SEV-SNP), the cr3 update needs to be written to + * VMCB save area now, since the save area will become the initial + * contents of the VMSA, and future VMCB save area updates won't be + * seen. + */ + if (sev_es_guest(vcpu->kvm)) { + svm->vmcb->save.cr3 = cr3; + vmcb_mark_dirty(svm->vmcb, VMCB_CR); + } +} + +static bool svm_is_valid_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +{ + return true; +} + void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) { struct vcpu_svm *svm = to_svm(vcpu); u64 hcr0 = cr0; + bool old_paging = is_paging(vcpu); #ifdef CONFIG_X86_64 - if (vcpu->arch.efer & EFER_LME && !vcpu->arch.guest_state_protected) { + if (vcpu->arch.efer & EFER_LME) { if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) { vcpu->arch.efer |= EFER_LMA; - svm->vmcb->save.efer |= EFER_LMA | EFER_LME; + if (!vcpu->arch.guest_state_protected) + svm->vmcb->save.efer |= EFER_LMA | EFER_LME; } if (is_paging(vcpu) && !(cr0 & X86_CR0_PG)) { vcpu->arch.efer &= ~EFER_LMA; - svm->vmcb->save.efer &= ~(EFER_LMA | EFER_LME); + if (!vcpu->arch.guest_state_protected) + svm->vmcb->save.efer &= ~(EFER_LMA | EFER_LME); } } #endif vcpu->arch.cr0 = cr0; - if (!npt_enabled) + if (!npt_enabled) { hcr0 |= X86_CR0_PG | X86_CR0_WP; + if (old_paging != is_paging(vcpu)) + svm_set_cr4(vcpu, kvm_read_cr4(vcpu)); + } /* * re-enable caching here because the QEMU bios @@ -1791,18 +1733,19 @@ void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) unsigned long host_cr4_mce = cr4_read_shadow() & X86_CR4_MCE; unsigned long old_cr4 = vcpu->arch.cr4; - if (npt_enabled && ((old_cr4 ^ cr4) & X86_CR4_PGE)) - svm_flush_tlb(vcpu); - vcpu->arch.cr4 = cr4; - if (!npt_enabled) + if (!npt_enabled) { cr4 |= X86_CR4_PAE; + + if (!is_paging(vcpu)) + cr4 &= ~(X86_CR4_SMEP | X86_CR4_SMAP | X86_CR4_PKE); + } cr4 |= host_cr4_mce; to_svm(vcpu)->vmcb->save.cr4 = cr4; vmcb_mark_dirty(to_svm(vcpu)->vmcb, VMCB_CR); if ((cr4 ^ old_cr4) & (X86_CR4_OSXSAVE | X86_CR4_PKE)) - kvm_update_cpuid_runtime(vcpu); + vcpu->arch.cpuid_dynamic_bits_dirty = true; } static void svm_set_segment(struct kvm_vcpu *vcpu, @@ -1861,11 +1804,11 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd) svm->asid = sd->next_asid++; } -static void svm_set_dr6(struct vcpu_svm *svm, unsigned long value) +static void svm_set_dr6(struct kvm_vcpu *vcpu, unsigned long value) { - struct vmcb *vmcb = svm->vmcb; + struct vmcb *vmcb = to_svm(vcpu)->vmcb; - if (svm->vcpu.arch.guest_state_protected) + if (vcpu->arch.guest_state_protected) return; if (unlikely(value != vmcb->save.dr6)) { @@ -1878,7 +1821,7 @@ static void svm_sync_dirty_debug_regs(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - if (vcpu->arch.guest_state_protected) + if (WARN_ON_ONCE(sev_es_guest(vcpu->kvm))) return; get_debugreg(vcpu->arch.db[0], 0); @@ -1922,15 +1865,33 @@ static int pf_interception(struct kvm_vcpu *vcpu) static int npf_interception(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); + int rc; - u64 fault_address = __sme_clr(svm->vmcb->control.exit_info_2); + u64 fault_address = svm->vmcb->control.exit_info_2; u64 error_code = svm->vmcb->control.exit_info_1; - trace_kvm_page_fault(fault_address, error_code); - return kvm_mmu_page_fault(vcpu, fault_address, error_code, - static_cpu_has(X86_FEATURE_DECODEASSISTS) ? - svm->vmcb->control.insn_bytes : NULL, - svm->vmcb->control.insn_len); + /* + * WARN if hardware generates a fault with an error code that collides + * with KVM-defined sythentic flags. Clear the flags and continue on, + * i.e. don't terminate the VM, as KVM can't possibly be relying on a + * flag that KVM doesn't know about. + */ + if (WARN_ON_ONCE(error_code & PFERR_SYNTHETIC_MASK)) + error_code &= ~PFERR_SYNTHETIC_MASK; + + if (sev_snp_guest(vcpu->kvm) && (error_code & PFERR_GUEST_ENC_MASK)) + error_code |= PFERR_PRIVATE_ACCESS; + + trace_kvm_page_fault(vcpu, fault_address, error_code); + rc = kvm_mmu_page_fault(vcpu, fault_address, error_code, + static_cpu_has(X86_FEATURE_DECODEASSISTS) ? + svm->vmcb->control.insn_bytes : NULL, + svm->vmcb->control.insn_len); + + if (rc > 0 && error_code & PFERR_GUEST_RMP_MASK) + sev_handle_rmp_fault(vcpu, fault_address, error_code); + + return rc; } static int db_interception(struct kvm_vcpu *vcpu) @@ -1990,14 +1951,13 @@ static int ac_interception(struct kvm_vcpu *vcpu) static bool is_erratum_383(void) { - int err, i; + int i; u64 value; if (!erratum_383_found) return false; - value = native_read_msr_safe(MSR_IA32_MC0_STATUS, &err); - if (err) + if (native_read_msr_safe(MSR_IA32_MC0_STATUS, &value)) return false; /* Bit 62 may or may not be set for this mce */ @@ -2008,17 +1968,11 @@ static bool is_erratum_383(void) /* Clear MCi_STATUS registers */ for (i = 0; i < 6; ++i) - native_write_msr_safe(MSR_IA32_MCx_STATUS(i), 0, 0); - - value = native_read_msr_safe(MSR_IA32_MCG_STATUS, &err); - if (!err) { - u32 low, high; + native_write_msr_safe(MSR_IA32_MCx_STATUS(i), 0); + if (!native_read_msr_safe(MSR_IA32_MCG_STATUS, &value)) { value &= ~(1ULL << 2); - low = lower_32_bits(value); - high = upper_32_bits(value); - - native_write_msr_safe(MSR_IA32_MCG_STATUS, low, high); + native_write_msr_safe(MSR_IA32_MCG_STATUS, value); } /* Flush tlb to evict multi-match entries */ @@ -2034,7 +1988,7 @@ static void svm_handle_mce(struct kvm_vcpu *vcpu) * Erratum 383 triggered. Guest state is corrupt so kill the * guest. */ - pr_err("KVM: Guest triggered AMD Erratum 383\n"); + pr_err("Guest triggered AMD Erratum 383\n"); kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); @@ -2058,19 +2012,26 @@ static int shutdown_interception(struct kvm_vcpu *vcpu) struct kvm_run *kvm_run = vcpu->run; struct vcpu_svm *svm = to_svm(vcpu); - /* - * The VM save area has already been encrypted so it - * cannot be reinitialized - just terminate. - */ - if (sev_es_guest(vcpu->kvm)) - return -EINVAL; /* - * VMCB is undefined after a SHUTDOWN intercept - * so reinitialize it. + * VMCB is undefined after a SHUTDOWN intercept. INIT the vCPU to put + * the VMCB in a known good state. Unfortuately, KVM doesn't have + * KVM_MP_STATE_SHUTDOWN and can't add it without potentially breaking + * userspace. At a platform view, INIT is acceptable behavior as + * there exist bare metal platforms that automatically INIT the CPU + * in response to shutdown. + * + * The VM save area for SEV-ES guests has already been encrypted so it + * cannot be reinitialized, i.e. synthesizing INIT is futile. */ - clear_page(svm->vmcb); - init_vmcb(vcpu); + if (!sev_es_guest(vcpu->kvm)) { + clear_page(svm->vmcb); +#ifdef CONFIG_KVM_SMM + if (is_smm(vcpu)) + kvm_smm_changed(vcpu, false); +#endif + kvm_vcpu_reset(vcpu, true); + } kvm_run->exit_reason = KVM_EXIT_SHUTDOWN; return 0; @@ -2106,6 +2067,11 @@ static int nmi_interception(struct kvm_vcpu *vcpu) return 1; } +static int smi_interception(struct kvm_vcpu *vcpu) +{ + return 1; +} + static int intr_interception(struct kvm_vcpu *vcpu) { ++vcpu->stat.irq_exits; @@ -2134,13 +2100,14 @@ static int vmload_vmsave_interception(struct kvm_vcpu *vcpu, bool vmload) ret = kvm_skip_emulated_instruction(vcpu); if (vmload) { - nested_svm_vmloadsave(vmcb12, svm->vmcb); + svm_copy_vmloadsave_state(svm->vmcb, vmcb12); svm->sysenter_eip_hi = 0; svm->sysenter_esp_hi = 0; - } else - nested_svm_vmloadsave(svm->vmcb, vmcb12); + } else { + svm_copy_vmloadsave_state(vmcb12, svm->vmcb); + } - kvm_vcpu_unmap(vcpu, &map, true); + kvm_vcpu_unmap(vcpu, &map); return ret; } @@ -2252,8 +2219,13 @@ static int gp_interception(struct kvm_vcpu *vcpu) if (!is_guest_mode(vcpu)) return kvm_emulate_instruction(vcpu, EMULTYPE_VMWARE_GP | EMULTYPE_NO_DECODE); - } else + } else { + /* All SVM instructions expect page aligned RAX */ + if (svm->vmcb->save.rax & ~PAGE_MASK) + goto reinject; + return emulate_svm_instr(vcpu, opcode); + } reinject: kvm_queue_exception_e(vcpu, GP_VECTOR, error_code); @@ -2269,7 +2241,7 @@ void svm_set_gif(struct vcpu_svm *svm, bool value) * Likewise, clear the VINTR intercept, we will set it * again while processing KVM_REQ_EVENT if needed. */ - if (vgif_enabled(svm)) + if (vgif) svm_clr_intercept(svm, INTERCEPT_STGI); if (svm_is_intercept(svm, INTERCEPT_VINTR)) svm_clear_vintr(svm); @@ -2277,7 +2249,8 @@ void svm_set_gif(struct vcpu_svm *svm, bool value) enable_gif(svm); if (svm->vcpu.arch.smi_pending || svm->vcpu.arch.nmi_pending || - kvm_cpu_has_injectable_intr(&svm->vcpu)) + kvm_cpu_has_injectable_intr(&svm->vcpu) || + kvm_apic_has_pending_init_or_sipi(&svm->vcpu)) kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); } else { disable_gif(svm); @@ -2287,7 +2260,7 @@ void svm_set_gif(struct vcpu_svm *svm, bool value) * in use, we still rely on the VINTR intercept (rather than * STGI) to detect an open interrupt window. */ - if (!vgif_enabled(svm)) + if (!vgif) svm_clear_vintr(svm); } } @@ -2384,6 +2357,7 @@ static int task_switch_interception(struct kvm_vcpu *vcpu) kvm_clear_exception_queue(vcpu); break; case SVM_EXITINTINFO_TYPE_INTR: + case SVM_EXITINTINFO_TYPE_SOFT: kvm_clear_interrupt_queue(vcpu); break; default: @@ -2395,7 +2369,7 @@ static int task_switch_interception(struct kvm_vcpu *vcpu) int_type == SVM_EXITINTINFO_TYPE_SOFT || (int_type == SVM_EXITINTINFO_TYPE_EXEPT && (int_vec == OF_VECTOR || int_vec == BP_VECTOR))) { - if (!skip_emulated_instruction(vcpu)) + if (!svm_skip_emulated_instruction(vcpu)) return 0; } @@ -2406,16 +2380,30 @@ static int task_switch_interception(struct kvm_vcpu *vcpu) has_error_code, error_code); } +static void svm_clr_iret_intercept(struct vcpu_svm *svm) +{ + if (!sev_es_guest(svm->vcpu.kvm)) + svm_clr_intercept(svm, INTERCEPT_IRET); +} + +static void svm_set_iret_intercept(struct vcpu_svm *svm) +{ + if (!sev_es_guest(svm->vcpu.kvm)) + svm_set_intercept(svm, INTERCEPT_IRET); +} + static int iret_interception(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); + WARN_ON_ONCE(sev_es_guest(vcpu->kvm)); + ++vcpu->stat.nmi_window_exits; - vcpu->arch.hflags |= HF_IRET_MASK; - if (!sev_es_guest(vcpu->kvm)) { - svm_clr_intercept(svm, INTERCEPT_IRET); - svm->nmi_iret_rip = kvm_rip_read(vcpu); - } + svm->awaiting_iret_completion = true; + + svm_clr_iret_intercept(svm); + svm->nmi_iret_rip = kvm_rip_read(vcpu); + kvm_make_request(KVM_REQ_EVENT, vcpu); return 1; } @@ -2447,7 +2435,7 @@ static bool check_selective_cr0_intercepted(struct kvm_vcpu *vcpu, bool ret = false; if (!is_guest_mode(vcpu) || - (!(vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0)))) + (!(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0)))) return false; cr0 &= ~SVM_CR0_SELECTIVE_MASK; @@ -2576,9 +2564,15 @@ static int dr_interception(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); int reg, dr; - unsigned long val; int err = 0; + /* + * SEV-ES intercepts DR7 only to disable guest debugging and the guest issues a VMGEXIT + * for DR7 write only. KVM cannot change DR7 (always swapped as type 'A') so return early. + */ + if (sev_es_guest(vcpu->kvm)) + return 1; + if (vcpu->guest_debug == 0) { /* * No more DR vmexits; force a reload of the debug registers @@ -2597,11 +2591,9 @@ static int dr_interception(struct kvm_vcpu *vcpu) dr = svm->vmcb->control.exit_code - SVM_EXIT_READ_DR0; if (dr >= 16) { /* mov to DRn */ dr -= 16; - val = kvm_register_read(vcpu, reg); - err = kvm_set_dr(vcpu, dr, val); + err = kvm_set_dr(vcpu, dr, kvm_register_read(vcpu, reg)); } else { - kvm_get_dr(vcpu, dr, &val); - kvm_register_write(vcpu, reg, val); + kvm_register_write(vcpu, reg, kvm_get_dr(vcpu, dr)); } return kvm_complete_insn_gp(vcpu, err); @@ -2641,29 +2633,46 @@ static int efer_trap(struct kvm_vcpu *vcpu) return kvm_complete_insn_gp(vcpu, ret); } -static int svm_get_msr_feature(struct kvm_msr_entry *msr) +static int svm_get_feature_msr(u32 msr, u64 *data) { - msr->data = 0; + *data = 0; - switch (msr->index) { - case MSR_F10H_DECFG: - if (boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) - msr->data |= MSR_F10H_DECFG_LFENCE_SERIALIZE; + switch (msr) { + case MSR_AMD64_DE_CFG: + if (cpu_feature_enabled(X86_FEATURE_LFENCE_RDTSC)) + *data |= MSR_AMD64_DE_CFG_LFENCE_SERIALIZE; break; - case MSR_IA32_PERF_CAPABILITIES: - return 0; default: - return KVM_MSR_RET_INVALID; + return KVM_MSR_RET_UNSUPPORTED; } return 0; } +static bool sev_es_prevent_msr_access(struct kvm_vcpu *vcpu, + struct msr_data *msr_info) +{ + return sev_es_guest(vcpu->kvm) && vcpu->arch.guest_state_protected && + msr_info->index != MSR_IA32_XSS && + !msr_write_intercepted(vcpu, msr_info->index); +} + static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) { struct vcpu_svm *svm = to_svm(vcpu); + if (sev_es_prevent_msr_access(vcpu, msr_info)) { + msr_info->data = 0; + return vcpu->kvm->arch.has_protected_state ? -EINVAL : 0; + } + switch (msr_info->index) { + case MSR_AMD64_TSC_RATIO: + if (!msr_info->host_initiated && + !guest_cpu_cap_has(vcpu, X86_FEATURE_TSCRATEMSR)) + return 1; + msr_info->data = svm->tsc_ratio_msr; + break; case MSR_STAR: msr_info->data = svm->vmcb01.ptr->save.star; break; @@ -2674,6 +2683,12 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) case MSR_CSTAR: msr_info->data = svm->vmcb01.ptr->save.cstar; break; + case MSR_GS_BASE: + msr_info->data = svm->vmcb01.ptr->save.gs.base; + break; + case MSR_FS_BASE: + msr_info->data = svm->vmcb01.ptr->save.fs.base; + break; case MSR_KERNEL_GS_BASE: msr_info->data = svm->vmcb01.ptr->save.kernel_gs_base; break; @@ -2686,22 +2701,26 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) break; case MSR_IA32_SYSENTER_EIP: msr_info->data = (u32)svm->vmcb01.ptr->save.sysenter_eip; - if (guest_cpuid_is_intel(vcpu)) + if (guest_cpuid_is_intel_compatible(vcpu)) msr_info->data |= (u64)svm->sysenter_eip_hi << 32; break; case MSR_IA32_SYSENTER_ESP: msr_info->data = svm->vmcb01.ptr->save.sysenter_esp; - if (guest_cpuid_is_intel(vcpu)) + if (guest_cpuid_is_intel_compatible(vcpu)) msr_info->data |= (u64)svm->sysenter_esp_hi << 32; break; + case MSR_IA32_S_CET: + msr_info->data = svm->vmcb->save.s_cet; + break; + case MSR_IA32_INT_SSP_TAB: + msr_info->data = svm->vmcb->save.isst_addr; + break; + case MSR_KVM_INTERNAL_GUEST_SSP: + msr_info->data = svm->vmcb->save.ssp; + break; case MSR_TSC_AUX: msr_info->data = svm->tsc_aux; break; - /* - * Nobody will change the following 5 values in the VMCB so we can - * safely return them on rdmsr. They will always be 0 until LBRV is - * implemented. - */ case MSR_IA32_DEBUGCTLMSR: msr_info->data = svm->vmcb->save.dbgctl; break; @@ -2735,7 +2754,7 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) break; case MSR_AMD64_VIRT_SPEC_CTRL: if (!msr_info->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_VIRT_SSBD)) + !guest_cpu_cap_has(vcpu, X86_FEATURE_VIRT_SSBD)) return 1; msr_info->data = svm->virt_spec_ctrl; @@ -2757,7 +2776,7 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = 0x1E; } break; - case MSR_F10H_DECFG: + case MSR_AMD64_DE_CFG: msr_info->data = svm->msr_decfg; break; default: @@ -2769,14 +2788,10 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) static int svm_complete_emulated_msr(struct kvm_vcpu *vcpu, int err) { struct vcpu_svm *svm = to_svm(vcpu); - if (!err || !sev_es_guest(vcpu->kvm) || WARN_ON_ONCE(!svm->ghcb)) + if (!err || !sev_es_guest(vcpu->kvm) || WARN_ON_ONCE(!svm->sev_es.ghcb)) return kvm_complete_insn_gp(vcpu, err); - ghcb_set_sw_exit_info_1(svm->ghcb, 1); - ghcb_set_sw_exit_info_2(svm->ghcb, - X86_TRAP_GP | - SVM_EVTINJ_TYPE_EXEPT | - SVM_EVTINJ_VALID); + svm_vmgexit_inject_exception(svm, X86_TRAP_GP); return 1; } @@ -2808,15 +2823,49 @@ static int svm_set_vm_cr(struct kvm_vcpu *vcpu, u64 data) static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) { struct vcpu_svm *svm = to_svm(vcpu); - int r; + int ret = 0; u32 ecx = msr->index; u64 data = msr->data; + + if (sev_es_prevent_msr_access(vcpu, msr)) + return vcpu->kvm->arch.has_protected_state ? -EINVAL : 0; + switch (ecx) { - case MSR_IA32_CR_PAT: - if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data)) + case MSR_AMD64_TSC_RATIO: + + if (!guest_cpu_cap_has(vcpu, X86_FEATURE_TSCRATEMSR)) { + + if (!msr->host_initiated) + return 1; + /* + * In case TSC scaling is not enabled, always + * leave this MSR at the default value. + * + * Due to bug in qemu 6.2.0, it would try to set + * this msr to 0 if tsc scaling is not enabled. + * Ignore this value as well. + */ + if (data != 0 && data != svm->tsc_ratio_msr) + return 1; + break; + } + + if (data & SVM_TSC_RATIO_RSVD) return 1; - vcpu->arch.pat = data; + + svm->tsc_ratio_msr = data; + + if (guest_cpu_cap_has(vcpu, X86_FEATURE_TSCRATEMSR) && + is_guest_mode(vcpu)) + nested_svm_update_tsc_ratio_msr(vcpu); + + break; + case MSR_IA32_CR_PAT: + ret = kvm_set_msr_common(vcpu, msr); + if (ret) + break; + svm->vmcb01.ptr->save.g_pat = data; if (is_guest_mode(vcpu)) nested_vmcb02_compute_g_pat(svm); @@ -2844,30 +2893,15 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) * * For nested: * The handling of the MSR bitmap for L2 guests is done in - * nested_svm_vmrun_msrpm. + * nested_svm_merge_msrpm(). * We update the L1 MSR bit as well since it will end up * touching the MSR anyway now. */ - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SPEC_CTRL, 1, 1); - break; - case MSR_IA32_PRED_CMD: - if (!msr->host_initiated && - !guest_has_pred_cmd_msr(vcpu)) - return 1; - - if (data & ~PRED_CMD_IBPB) - return 1; - if (!boot_cpu_has(X86_FEATURE_IBPB)) - return 1; - if (!data) - break; - - wrmsrl(MSR_IA32_PRED_CMD, PRED_CMD_IBPB); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_PRED_CMD, 0, 1); + svm_disable_intercept_for_msr(vcpu, MSR_IA32_SPEC_CTRL, MSR_TYPE_RW); break; case MSR_AMD64_VIRT_SPEC_CTRL: if (!msr->host_initiated && - !guest_cpuid_has(vcpu, X86_FEATURE_VIRT_SSBD)) + !guest_cpu_cap_has(vcpu, X86_FEATURE_VIRT_SSBD)) return 1; if (data & ~SPEC_CTRL_SSBD) @@ -2885,6 +2919,12 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) case MSR_CSTAR: svm->vmcb01.ptr->save.cstar = data; break; + case MSR_GS_BASE: + svm->vmcb01.ptr->save.gs.base = data; + break; + case MSR_FS_BASE: + svm->vmcb01.ptr->save.fs.base = data; + break; case MSR_KERNEL_GS_BASE: svm->vmcb01.ptr->save.kernel_gs_base = data; break; @@ -2904,76 +2944,104 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) * 32 bit part of these msrs to support Intel's * implementation of SYSENTER/SYSEXIT. */ - svm->sysenter_eip_hi = guest_cpuid_is_intel(vcpu) ? (data >> 32) : 0; + svm->sysenter_eip_hi = guest_cpuid_is_intel_compatible(vcpu) ? (data >> 32) : 0; break; case MSR_IA32_SYSENTER_ESP: svm->vmcb01.ptr->save.sysenter_esp = (u32)data; - svm->sysenter_esp_hi = guest_cpuid_is_intel(vcpu) ? (data >> 32) : 0; + svm->sysenter_esp_hi = guest_cpuid_is_intel_compatible(vcpu) ? (data >> 32) : 0; + break; + case MSR_IA32_S_CET: + svm->vmcb->save.s_cet = data; + vmcb_mark_dirty(svm->vmcb01.ptr, VMCB_CET); + break; + case MSR_IA32_INT_SSP_TAB: + svm->vmcb->save.isst_addr = data; + vmcb_mark_dirty(svm->vmcb01.ptr, VMCB_CET); + break; + case MSR_KVM_INTERNAL_GUEST_SSP: + svm->vmcb->save.ssp = data; + vmcb_mark_dirty(svm->vmcb01.ptr, VMCB_CET); break; case MSR_TSC_AUX: /* + * TSC_AUX is always virtualized for SEV-ES guests when the + * feature is available. The user return MSR support is not + * required in this case because TSC_AUX is restored on #VMEXIT + * from the host save area. + */ + if (boot_cpu_has(X86_FEATURE_V_TSC_AUX) && sev_es_guest(vcpu->kvm)) + break; + + /* * TSC_AUX is usually changed only during boot and never read - * directly. Intercept TSC_AUX instead of exposing it to the - * guest via direct_access_msrs, and switch it via user return. + * directly. Intercept TSC_AUX and switch it via user return. */ preempt_disable(); - r = kvm_set_user_return_msr(tsc_aux_uret_slot, data, -1ull); + ret = kvm_set_user_return_msr(tsc_aux_uret_slot, data, -1ull); preempt_enable(); - if (r) - return 1; + if (ret) + break; svm->tsc_aux = data; break; case MSR_IA32_DEBUGCTLMSR: - if (!boot_cpu_has(X86_FEATURE_LBRV)) { - vcpu_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTL 0x%llx, nop\n", - __func__, data); + if (!lbrv) { + kvm_pr_unimpl_wrmsr(vcpu, ecx, data); break; } + + /* + * Suppress BTF as KVM doesn't virtualize BTF, but there's no + * way to communicate lack of support to the guest. + */ + if (data & DEBUGCTLMSR_BTF) { + kvm_pr_unimpl_wrmsr(vcpu, MSR_IA32_DEBUGCTLMSR, data); + data &= ~DEBUGCTLMSR_BTF; + } + if (data & DEBUGCTL_RESERVED_BITS) return 1; + if (svm->vmcb->save.dbgctl == data) + break; + svm->vmcb->save.dbgctl = data; vmcb_mark_dirty(svm->vmcb, VMCB_LBR); - if (data & (1ULL<<0)) - svm_enable_lbrv(vcpu); - else - svm_disable_lbrv(vcpu); + svm_update_lbrv(vcpu); break; case MSR_VM_HSAVE_PA: - svm->nested.hsave_msr = data; + /* + * Old kernels did not validate the value written to + * MSR_VM_HSAVE_PA. Allow KVM_SET_MSR to set an invalid + * value to allow live migrating buggy or malicious guests + * originating from those kernels. + */ + if (!msr->host_initiated && !page_address_valid(vcpu, data)) + return 1; + + svm->nested.hsave_msr = data & PAGE_MASK; break; case MSR_VM_CR: return svm_set_vm_cr(vcpu, data); case MSR_VM_IGNNE: - vcpu_unimpl(vcpu, "unimplemented wrmsr: 0x%x data 0x%llx\n", ecx, data); + kvm_pr_unimpl_wrmsr(vcpu, ecx, data); break; - case MSR_F10H_DECFG: { - struct kvm_msr_entry msr_entry; + case MSR_AMD64_DE_CFG: { + u64 supported_de_cfg; - msr_entry.index = msr->index; - if (svm_get_msr_feature(&msr_entry)) + if (svm_get_feature_msr(ecx, &supported_de_cfg)) return 1; - /* Check the supported bits */ - if (data & ~msr_entry.data) - return 1; - - /* Don't allow the guest to change a bit, #GP */ - if (!msr->host_initiated && (data ^ msr_entry.data)) + if (data & ~supported_de_cfg) return 1; svm->msr_decfg = data; break; } - case MSR_IA32_APICBASE: - if (kvm_vcpu_apicv_active(vcpu)) - avic_update_vapic_bar(to_svm(vcpu), data); - fallthrough; default: return kvm_set_msr_common(vcpu, msr); } - return 0; + return ret; } static int msr_interception(struct kvm_vcpu *vcpu) @@ -2990,11 +3058,18 @@ static int interrupt_window_interception(struct kvm_vcpu *vcpu) svm_clear_vintr(to_svm(vcpu)); /* - * For AVIC, the only reason to end up here is ExtINTs. + * If not running nested, for AVIC, the only reason to end up here is ExtINTs. * In this case AVIC was temporarily disabled for * requesting the IRQ window and we have to re-enable it. + * + * If running nested, still remove the VM wide AVIC inhibit to + * support case in which the interrupt window was requested when the + * vCPU was not running nested. + + * All vCPUs which run still run nested, will remain to have their + * AVIC still inhibited due to per-cpu AVIC inhibition. */ - svm_toggle_avic_for_irq_window(vcpu, true); + kvm_clear_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN); ++vcpu->stat.irq_window_exits; return 1; @@ -3003,7 +3078,6 @@ static int interrupt_window_interception(struct kvm_vcpu *vcpu) static int pause_interception(struct kvm_vcpu *vcpu) { bool in_kernel; - /* * CPL is not made available for an SEV-ES guest, therefore * vcpu->arch.preempted_in_kernel can never be true. Just @@ -3011,8 +3085,7 @@ static int pause_interception(struct kvm_vcpu *vcpu) */ in_kernel = !sev_es_guest(vcpu->kvm) && svm_get_cpl(vcpu) == 0; - if (!kvm_pause_in_guest(vcpu->kvm)) - grow_ple_window(vcpu); + grow_ple_window(vcpu); kvm_vcpu_on_spin(vcpu, in_kernel); return kvm_skip_emulated_instruction(vcpu); @@ -3024,7 +3097,7 @@ static int invpcid_interception(struct kvm_vcpu *vcpu) unsigned long type; gva_t gva; - if (!guest_cpuid_has(vcpu, X86_FEATURE_INVPCID)) { + if (!guest_cpu_cap_has(vcpu, X86_FEATURE_INVPCID)) { kvm_queue_exception(vcpu, UD_VECTOR); return 1; } @@ -3037,14 +3110,51 @@ static int invpcid_interception(struct kvm_vcpu *vcpu) type = svm->vmcb->control.exit_info_2; gva = svm->vmcb->control.exit_info_1; - if (type > 3) { - kvm_inject_gp(vcpu, 0); + /* + * FIXME: Perform segment checks for 32-bit mode, and inject #SS if the + * stack segment is used. The intercept takes priority over all + * #GP checks except CPL>0, but somehow still generates a linear + * address? The APM is sorely lacking. + */ + if (is_noncanonical_address(gva, vcpu, 0)) { + kvm_queue_exception_e(vcpu, GP_VECTOR, 0); return 1; } return kvm_handle_invpcid(vcpu, type, gva); } +static inline int complete_userspace_buslock(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + /* + * If userspace has NOT changed RIP, then KVM's ABI is to let the guest + * execute the bus-locking instruction. Set the bus lock counter to '1' + * to effectively step past the bus lock. + */ + if (kvm_is_linear_rip(vcpu, vcpu->arch.cui_linear_rip)) + svm->vmcb->control.bus_lock_counter = 1; + + return 1; +} + +static int bus_lock_exit(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + vcpu->run->exit_reason = KVM_EXIT_X86_BUS_LOCK; + vcpu->run->flags |= KVM_RUN_X86_BUS_LOCK; + + vcpu->arch.cui_linear_rip = kvm_get_linear_rip(vcpu); + vcpu->arch.complete_userspace_io = complete_userspace_buslock; + + if (is_guest_mode(vcpu)) + svm->nested.ctl.bus_lock_rip = vcpu->arch.cui_linear_rip; + + return 0; +} + static int (*const svm_exit_handlers[])(struct kvm_vcpu *vcpu) = { [SVM_EXIT_READ_CR0] = cr_interception, [SVM_EXIT_READ_CR3] = cr_interception, @@ -3080,8 +3190,7 @@ static int (*const svm_exit_handlers[])(struct kvm_vcpu *vcpu) = { [SVM_EXIT_EXCP_BASE + GP_VECTOR] = gp_interception, [SVM_EXIT_INTR] = intr_interception, [SVM_EXIT_NMI] = nmi_interception, - [SVM_EXIT_SMI] = kvm_emulate_as_nop, - [SVM_EXIT_INIT] = kvm_emulate_as_nop, + [SVM_EXIT_SMI] = smi_interception, [SVM_EXIT_VINTR] = interrupt_window_interception, [SVM_EXIT_RDPMC] = kvm_emulate_rdpmc, [SVM_EXIT_CPUID] = kvm_emulate_cpuid, @@ -3113,11 +3222,15 @@ static int (*const svm_exit_handlers[])(struct kvm_vcpu *vcpu) = { [SVM_EXIT_CR4_WRITE_TRAP] = cr_trap, [SVM_EXIT_CR8_WRITE_TRAP] = cr_trap, [SVM_EXIT_INVPCID] = invpcid_interception, + [SVM_EXIT_IDLE_HLT] = kvm_emulate_halt, [SVM_EXIT_NPF] = npf_interception, + [SVM_EXIT_BUS_LOCK] = bus_lock_exit, [SVM_EXIT_RSM] = rsm_interception, [SVM_EXIT_AVIC_INCOMPLETE_IPI] = avic_incomplete_ipi_interception, [SVM_EXIT_AVIC_UNACCELERATED_ACCESS] = avic_unaccelerated_access_interception, +#ifdef CONFIG_KVM_AMD_SEV [SVM_EXIT_VMGEXIT] = sev_handle_vmgexit, +#endif }; static void dump_vmcb(struct kvm_vcpu *vcpu) @@ -3126,14 +3239,21 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) struct vmcb_control_area *control = &svm->vmcb->control; struct vmcb_save_area *save = &svm->vmcb->save; struct vmcb_save_area *save01 = &svm->vmcb01.ptr->save; + char *vm_type; if (!dump_invalid_vmcb) { pr_warn_ratelimited("set kvm_amd.dump_invalid_vmcb=1 to dump internal KVM state.\n"); return; } - pr_err("VMCB %p, last attempted VMRUN on CPU %d\n", - svm->current_vmcb->ptr, vcpu->arch.last_vmentry_cpu); + guard(mutex)(&vmcb_dump_mutex); + + vm_type = sev_snp_guest(vcpu->kvm) ? "SEV-SNP" : + sev_es_guest(vcpu->kvm) ? "SEV-ES" : + sev_guest(vcpu->kvm) ? "SEV" : "SVM"; + + pr_err("%s vCPU%u VMCB %p, last attempted VMRUN on CPU %d\n", + vm_type, vcpu->vcpu_id, svm->current_vmcb->ptr, vcpu->arch.last_vmentry_cpu); pr_err("VMCB Control Area:\n"); pr_err("%-20s%04x\n", "cr_read:", control->intercepts[INTERCEPT_CR] & 0xffff); pr_err("%-20s%04x\n", "cr_write:", control->intercepts[INTERCEPT_CR] >> 16); @@ -3171,6 +3291,17 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) pr_err("%-20s%016llx\n", "avic_logical_id:", control->avic_logical_id); pr_err("%-20s%016llx\n", "avic_physical_id:", control->avic_physical_id); pr_err("%-20s%016llx\n", "vmsa_pa:", control->vmsa_pa); + pr_err("%-20s%016llx\n", "allowed_sev_features:", control->allowed_sev_features); + pr_err("%-20s%016llx\n", "guest_sev_features:", control->guest_sev_features); + + if (sev_es_guest(vcpu->kvm)) { + save = sev_decrypt_vmsa(vcpu); + if (!save) + goto no_vmsa; + + save01 = save; + } + pr_err("VMCB State Save Area:\n"); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "es:", @@ -3212,8 +3343,8 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) "tr:", save01->tr.selector, save01->tr.attrib, save01->tr.limit, save01->tr.base); - pr_err("cpl: %d efer: %016llx\n", - save->cpl, save->efer); + pr_err("vmpl: %d cpl: %d efer: %016llx\n", + save->vmpl, save->cpl, save->efer); pr_err("%-15s %016llx %-13s %016llx\n", "cr0:", save->cr0, "cr2:", save->cr2); pr_err("%-15s %016llx %-13s %016llx\n", @@ -3225,6 +3356,10 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) pr_err("%-15s %016llx %-13s %016llx\n", "rsp:", save->rsp, "rax:", save->rax); pr_err("%-15s %016llx %-13s %016llx\n", + "s_cet:", save->s_cet, "ssp:", save->ssp); + pr_err("%-15s %016llx\n", + "isst_addr:", save->isst_addr); + pr_err("%-15s %016llx %-13s %016llx\n", "star:", save01->star, "lstar:", save01->lstar); pr_err("%-15s %016llx %-13s %016llx\n", "cstar:", save01->cstar, "sfmask:", save01->sfmask); @@ -3241,50 +3376,116 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) pr_err("%-15s %016llx %-13s %016llx\n", "excp_from:", save->last_excp_from, "excp_to:", save->last_excp_to); + + if (sev_es_guest(vcpu->kvm)) { + struct sev_es_save_area *vmsa = (struct sev_es_save_area *)save; + + pr_err("%-15s %016llx\n", + "sev_features", vmsa->sev_features); + + pr_err("%-15s %016llx %-13s %016llx\n", + "pl0_ssp:", vmsa->pl0_ssp, "pl1_ssp:", vmsa->pl1_ssp); + pr_err("%-15s %016llx %-13s %016llx\n", + "pl2_ssp:", vmsa->pl2_ssp, "pl3_ssp:", vmsa->pl3_ssp); + pr_err("%-15s %016llx\n", + "u_cet:", vmsa->u_cet); + + pr_err("%-15s %016llx %-13s %016llx\n", + "rax:", vmsa->rax, "rbx:", vmsa->rbx); + pr_err("%-15s %016llx %-13s %016llx\n", + "rcx:", vmsa->rcx, "rdx:", vmsa->rdx); + pr_err("%-15s %016llx %-13s %016llx\n", + "rsi:", vmsa->rsi, "rdi:", vmsa->rdi); + pr_err("%-15s %016llx %-13s %016llx\n", + "rbp:", vmsa->rbp, "rsp:", vmsa->rsp); + pr_err("%-15s %016llx %-13s %016llx\n", + "r8:", vmsa->r8, "r9:", vmsa->r9); + pr_err("%-15s %016llx %-13s %016llx\n", + "r10:", vmsa->r10, "r11:", vmsa->r11); + pr_err("%-15s %016llx %-13s %016llx\n", + "r12:", vmsa->r12, "r13:", vmsa->r13); + pr_err("%-15s %016llx %-13s %016llx\n", + "r14:", vmsa->r14, "r15:", vmsa->r15); + pr_err("%-15s %016llx %-13s %016llx\n", + "xcr0:", vmsa->xcr0, "xss:", vmsa->xss); + } else { + pr_err("%-15s %016llx %-13s %016lx\n", + "rax:", save->rax, "rbx:", + vcpu->arch.regs[VCPU_REGS_RBX]); + pr_err("%-15s %016lx %-13s %016lx\n", + "rcx:", vcpu->arch.regs[VCPU_REGS_RCX], + "rdx:", vcpu->arch.regs[VCPU_REGS_RDX]); + pr_err("%-15s %016lx %-13s %016lx\n", + "rsi:", vcpu->arch.regs[VCPU_REGS_RSI], + "rdi:", vcpu->arch.regs[VCPU_REGS_RDI]); + pr_err("%-15s %016lx %-13s %016llx\n", + "rbp:", vcpu->arch.regs[VCPU_REGS_RBP], + "rsp:", save->rsp); +#ifdef CONFIG_X86_64 + pr_err("%-15s %016lx %-13s %016lx\n", + "r8:", vcpu->arch.regs[VCPU_REGS_R8], + "r9:", vcpu->arch.regs[VCPU_REGS_R9]); + pr_err("%-15s %016lx %-13s %016lx\n", + "r10:", vcpu->arch.regs[VCPU_REGS_R10], + "r11:", vcpu->arch.regs[VCPU_REGS_R11]); + pr_err("%-15s %016lx %-13s %016lx\n", + "r12:", vcpu->arch.regs[VCPU_REGS_R12], + "r13:", vcpu->arch.regs[VCPU_REGS_R13]); + pr_err("%-15s %016lx %-13s %016lx\n", + "r14:", vcpu->arch.regs[VCPU_REGS_R14], + "r15:", vcpu->arch.regs[VCPU_REGS_R15]); +#endif + } + +no_vmsa: + if (sev_es_guest(vcpu->kvm)) + sev_free_decrypted_vmsa(vcpu, save); } -static int svm_handle_invalid_exit(struct kvm_vcpu *vcpu, u64 exit_code) +static bool svm_check_exit_valid(u64 exit_code) { - if (exit_code < ARRAY_SIZE(svm_exit_handlers) && - svm_exit_handlers[exit_code]) - return 0; + return (exit_code < ARRAY_SIZE(svm_exit_handlers) && + svm_exit_handlers[exit_code]); +} - vcpu_unimpl(vcpu, "svm: unexpected exit reason 0x%llx\n", exit_code); +static int svm_handle_invalid_exit(struct kvm_vcpu *vcpu, u64 exit_code) +{ dump_vmcb(vcpu); - vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_UNEXPECTED_EXIT_REASON; - vcpu->run->internal.ndata = 2; - vcpu->run->internal.data[0] = exit_code; - vcpu->run->internal.data[1] = vcpu->arch.last_vmentry_cpu; - - return -EINVAL; + kvm_prepare_unexpected_reason_exit(vcpu, exit_code); + return 0; } int svm_invoke_exit_handler(struct kvm_vcpu *vcpu, u64 exit_code) { - if (svm_handle_invalid_exit(vcpu, exit_code)) - return 0; + if (!svm_check_exit_valid(exit_code)) + return svm_handle_invalid_exit(vcpu, exit_code); -#ifdef CONFIG_RETPOLINE +#ifdef CONFIG_MITIGATION_RETPOLINE if (exit_code == SVM_EXIT_MSR) return msr_interception(vcpu); else if (exit_code == SVM_EXIT_VINTR) return interrupt_window_interception(vcpu); else if (exit_code == SVM_EXIT_INTR) return intr_interception(vcpu); - else if (exit_code == SVM_EXIT_HLT) + else if (exit_code == SVM_EXIT_HLT || exit_code == SVM_EXIT_IDLE_HLT) return kvm_emulate_halt(vcpu); else if (exit_code == SVM_EXIT_NPF) return npf_interception(vcpu); +#ifdef CONFIG_KVM_AMD_SEV + else if (exit_code == SVM_EXIT_VMGEXIT) + return sev_handle_vmgexit(vcpu); +#endif #endif return svm_exit_handlers[exit_code](vcpu); } -static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2, +static void svm_get_exit_info(struct kvm_vcpu *vcpu, u32 *reason, + u64 *info1, u64 *info2, u32 *intr_info, u32 *error_code) { struct vmcb_control_area *control = &to_svm(vcpu)->vmcb->control; + *reason = control->exit_code; *info1 = control->exit_info_1; *info2 = control->exit_info_2; *intr_info = control->exit_int_info; @@ -3295,14 +3496,27 @@ static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2, *error_code = 0; } -static int handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) +static void svm_get_entry_info(struct kvm_vcpu *vcpu, u32 *intr_info, + u32 *error_code) +{ + struct vmcb_control_area *control = &to_svm(vcpu)->vmcb->control; + + *intr_info = control->event_inj; + + if ((*intr_info & SVM_EXITINTINFO_VALID) && + (*intr_info & SVM_EXITINTINFO_VALID_ERR)) + *error_code = control->event_inj_err; + else + *error_code = 0; + +} + +static int svm_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) { struct vcpu_svm *svm = to_svm(vcpu); struct kvm_run *kvm_run = vcpu->run; u32 exit_code = svm->vmcb->control.exit_code; - trace_kvm_exit(exit_code, vcpu, KVM_ISA_SVM); - /* SEV-ES guests must use the CR write traps to track CR registers. */ if (!sev_es_guest(vcpu->kvm)) { if (!svm_is_intercept(svm, INTERCEPT_CR0_WRITE)) @@ -3314,7 +3528,7 @@ static int handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) if (is_guest_mode(vcpu)) { int vmexit; - trace_kvm_nested_vmexit(exit_code, vcpu, KVM_ISA_SVM); + trace_kvm_nested_vmexit(vcpu, KVM_ISA_SVM); vmexit = nested_svm_exit_special(svm); @@ -3334,32 +3548,15 @@ static int handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) return 0; } - if (is_external_interrupt(svm->vmcb->control.exit_int_info) && - exit_code != SVM_EXIT_EXCP_BASE + PF_VECTOR && - exit_code != SVM_EXIT_NPF && exit_code != SVM_EXIT_TASK_SWITCH && - exit_code != SVM_EXIT_INTR && exit_code != SVM_EXIT_NMI) - printk(KERN_ERR "%s: unexpected exit_int_info 0x%x " - "exit_code 0x%x\n", - __func__, svm->vmcb->control.exit_int_info, - exit_code); - if (exit_fastpath != EXIT_FASTPATH_NONE) return 1; return svm_invoke_exit_handler(vcpu, exit_code); } -static void reload_tss(struct kvm_vcpu *vcpu) -{ - struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); - - sd->tss_desc->type = 9; /* available 32/64-bit TSS */ - load_TR_desc(); -} - -static void pre_svm_run(struct kvm_vcpu *vcpu) +static int pre_svm_run(struct kvm_vcpu *vcpu) { - struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); + struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu); struct vcpu_svm *svm = to_svm(vcpu); /* @@ -3379,6 +3576,8 @@ static void pre_svm_run(struct kvm_vcpu *vcpu) /* FIXME: handle wraparound of asid_generation */ if (svm->current_vmcb->asid_generation != sd->asid_generation) new_asid(svm, sd); + + return 0; } static void svm_inject_nmi(struct kvm_vcpu *vcpu) @@ -3386,23 +3585,124 @@ static void svm_inject_nmi(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_NMI; - vcpu->arch.hflags |= HF_NMI_MASK; - if (!sev_es_guest(vcpu->kvm)) - svm_set_intercept(svm, INTERCEPT_IRET); + + if (svm->nmi_l1_to_l2) + return; + + /* + * No need to manually track NMI masking when vNMI is enabled, hardware + * automatically sets V_NMI_BLOCKING_MASK as appropriate, including the + * case where software directly injects an NMI. + */ + if (!is_vnmi_enabled(svm)) { + svm->nmi_masked = true; + svm_set_iret_intercept(svm); + } ++vcpu->stat.nmi_injections; } -static void svm_set_irq(struct kvm_vcpu *vcpu) +static bool svm_is_vnmi_pending(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - BUG_ON(!(gif_set(svm))); + if (!is_vnmi_enabled(svm)) + return false; + + return !!(svm->vmcb->control.int_ctl & V_NMI_PENDING_MASK); +} + +static bool svm_set_vnmi_pending(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + if (!is_vnmi_enabled(svm)) + return false; + + if (svm->vmcb->control.int_ctl & V_NMI_PENDING_MASK) + return false; + + svm->vmcb->control.int_ctl |= V_NMI_PENDING_MASK; + vmcb_mark_dirty(svm->vmcb, VMCB_INTR); + + /* + * Because the pending NMI is serviced by hardware, KVM can't know when + * the NMI is "injected", but for all intents and purposes, passing the + * NMI off to hardware counts as injection. + */ + ++vcpu->stat.nmi_injections; + + return true; +} + +static void svm_inject_irq(struct kvm_vcpu *vcpu, bool reinjected) +{ + struct kvm_queued_interrupt *intr = &vcpu->arch.interrupt; + struct vcpu_svm *svm = to_svm(vcpu); + u32 type; + + if (intr->soft) { + if (svm_update_soft_interrupt_rip(vcpu, intr->nr)) + return; + + type = SVM_EVTINJ_TYPE_SOFT; + } else { + type = SVM_EVTINJ_TYPE_INTR; + } - trace_kvm_inj_virq(vcpu->arch.interrupt.nr); + trace_kvm_inj_virq(intr->nr, intr->soft, reinjected); ++vcpu->stat.irq_injections; - svm->vmcb->control.event_inj = vcpu->arch.interrupt.nr | - SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR; + svm->vmcb->control.event_inj = intr->nr | SVM_EVTINJ_VALID | type; +} + +void svm_complete_interrupt_delivery(struct kvm_vcpu *vcpu, int delivery_mode, + int trig_mode, int vector) +{ + /* + * apic->apicv_active must be read after vcpu->mode. + * Pairs with smp_store_release in vcpu_enter_guest. + */ + bool in_guest_mode = (smp_load_acquire(&vcpu->mode) == IN_GUEST_MODE); + + /* Note, this is called iff the local APIC is in-kernel. */ + if (!READ_ONCE(vcpu->arch.apic->apicv_active)) { + /* Process the interrupt via kvm_check_and_inject_events(). */ + kvm_make_request(KVM_REQ_EVENT, vcpu); + kvm_vcpu_kick(vcpu); + return; + } + + trace_kvm_apicv_accept_irq(vcpu->vcpu_id, delivery_mode, trig_mode, vector); + if (in_guest_mode) { + /* + * Signal the doorbell to tell hardware to inject the IRQ. If + * the vCPU exits the guest before the doorbell chimes, hardware + * will automatically process AVIC interrupts at the next VMRUN. + */ + avic_ring_doorbell(vcpu); + } else { + /* + * Wake the vCPU if it was blocking. KVM will then detect the + * pending IRQ when checking if the vCPU has a wake event. + */ + kvm_vcpu_wake_up(vcpu); + } +} + +static void svm_deliver_interrupt(struct kvm_lapic *apic, int delivery_mode, + int trig_mode, int vector) +{ + kvm_lapic_set_irr(vector, apic); + + /* + * Pairs with the smp_mb_*() after setting vcpu->guest_mode in + * vcpu_enter_guest() to ensure the write to the vIRR is ordered before + * the read of guest_mode. This guarantees that either VMRUN will see + * and process the new vIRR entry, or that svm_complete_interrupt_delivery + * will signal the doorbell if the CPU has already entered the guest. + */ + smp_mb__after_atomic(); + svm_complete_interrupt_delivery(apic->vcpu, delivery_mode, trig_mode, vector); } static void svm_update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr) @@ -3428,11 +3728,39 @@ static void svm_update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr) svm_set_intercept(svm, INTERCEPT_CR8_WRITE); } +static bool svm_get_nmi_mask(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + if (is_vnmi_enabled(svm)) + return svm->vmcb->control.int_ctl & V_NMI_BLOCKING_MASK; + else + return svm->nmi_masked; +} + +static void svm_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + if (is_vnmi_enabled(svm)) { + if (masked) + svm->vmcb->control.int_ctl |= V_NMI_BLOCKING_MASK; + else + svm->vmcb->control.int_ctl &= ~V_NMI_BLOCKING_MASK; + + } else { + svm->nmi_masked = masked; + if (masked) + svm_set_iret_intercept(svm); + else + svm_clr_iret_intercept(svm); + } +} + bool svm_nmi_blocked(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); struct vmcb *vmcb = svm->vmcb; - bool ret; if (!gif_set(svm)) return true; @@ -3440,10 +3768,10 @@ bool svm_nmi_blocked(struct kvm_vcpu *vcpu) if (is_guest_mode(vcpu) && nested_exit_on_nmi(svm)) return false; - ret = (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) || - (vcpu->arch.hflags & HF_NMI_MASK); + if (svm_get_nmi_mask(vcpu)) + return true; - return ret; + return vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK; } static int svm_nmi_allowed(struct kvm_vcpu *vcpu, bool for_injection) @@ -3452,31 +3780,13 @@ static int svm_nmi_allowed(struct kvm_vcpu *vcpu, bool for_injection) if (svm->nested.nested_run_pending) return -EBUSY; + if (svm_nmi_blocked(vcpu)) + return 0; + /* An NMI must not be injected into L2 if it's supposed to VM-Exit. */ if (for_injection && is_guest_mode(vcpu) && nested_exit_on_nmi(svm)) return -EBUSY; - - return !svm_nmi_blocked(vcpu); -} - -static bool svm_get_nmi_mask(struct kvm_vcpu *vcpu) -{ - return !!(vcpu->arch.hflags & HF_NMI_MASK); -} - -static void svm_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked) -{ - struct vcpu_svm *svm = to_svm(vcpu); - - if (masked) { - vcpu->arch.hflags |= HF_NMI_MASK; - if (!sev_es_guest(vcpu->kvm)) - svm_set_intercept(svm, INTERCEPT_IRET); - } else { - vcpu->arch.hflags &= ~HF_NMI_MASK; - if (!sev_es_guest(vcpu->kvm)) - svm_clr_intercept(svm, INTERCEPT_IRET); - } + return 1; } bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) @@ -3487,14 +3797,7 @@ bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) if (!gif_set(svm)) return true; - if (sev_es_guest(vcpu->kvm)) { - /* - * SEV-ES guests to not expose RFLAGS. Use the VMCB interrupt mask - * bit to determine the state of the IF flag. - */ - if (!(vmcb->control.int_state & SVM_GUEST_INTERRUPT_MASK)) - return true; - } else if (is_guest_mode(vcpu)) { + if (is_guest_mode(vcpu)) { /* As long as interrupts are being delivered... */ if ((svm->nested.ctl.int_ctl & V_INTR_MASKING_MASK) ? !(svm->vmcb01.ptr->save.rflags & X86_EFLAGS_IF) @@ -3505,7 +3808,7 @@ bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) if (nested_exit_on_intr(svm)) return false; } else { - if (!(kvm_get_rflags(vcpu) & X86_EFLAGS_IF)) + if (!svm_get_if_flag(vcpu)) return true; } @@ -3515,9 +3818,13 @@ bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) static int svm_interrupt_allowed(struct kvm_vcpu *vcpu, bool for_injection) { struct vcpu_svm *svm = to_svm(vcpu); + if (svm->nested.nested_run_pending) return -EBUSY; + if (svm_interrupt_blocked(vcpu)) + return 0; + /* * An IRQ must not be injected into L2 if it's supposed to VM-Exit, * e.g. if the IRQ arrived asynchronously after checking nested events. @@ -3525,7 +3832,7 @@ static int svm_interrupt_allowed(struct kvm_vcpu *vcpu, bool for_injection) if (for_injection && is_guest_mode(vcpu) && nested_exit_on_intr(svm)) return -EBUSY; - return !svm_interrupt_blocked(vcpu); + return 1; } static void svm_enable_irq_window(struct kvm_vcpu *vcpu) @@ -3540,14 +3847,20 @@ static void svm_enable_irq_window(struct kvm_vcpu *vcpu) * enabled, the STGI interception will not occur. Enable the irq * window under the assumption that the hardware will set the GIF. */ - if (vgif_enabled(svm) || gif_set(svm)) { + if (vgif || gif_set(svm)) { /* * IRQ window is not needed when AVIC is enabled, * unless we have pending ExtINT since it cannot be injected - * via AVIC. In such case, we need to temporarily disable AVIC, + * via AVIC. In such case, KVM needs to temporarily disable AVIC, * and fallback to injecting IRQ via V_IRQ. + * + * If running nested, AVIC is already locally inhibited + * on this vCPU, therefore there is no need to request + * the VM wide AVIC inhibition. */ - svm_toggle_avic_for_irq_window(vcpu, false); + if (!is_guest_mode(vcpu)) + kvm_set_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN); + svm_set_vintr(svm); } } @@ -3556,11 +3869,44 @@ static void svm_enable_nmi_window(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - if ((vcpu->arch.hflags & (HF_NMI_MASK | HF_IRET_MASK)) == HF_NMI_MASK) - return; /* IRET will cause a vm exit */ + /* + * If NMIs are outright masked, i.e. the vCPU is already handling an + * NMI, and KVM has not yet intercepted an IRET, then there is nothing + * more to do at this time as KVM has already enabled IRET intercepts. + * If KVM has already intercepted IRET, then single-step over the IRET, + * as NMIs aren't architecturally unmasked until the IRET completes. + * + * If vNMI is enabled, KVM should never request an NMI window if NMIs + * are masked, as KVM allows at most one to-be-injected NMI and one + * pending NMI. If two NMIs arrive simultaneously, KVM will inject one + * NMI and set V_NMI_PENDING for the other, but if and only if NMIs are + * unmasked. KVM _will_ request an NMI window in some situations, e.g. + * if the vCPU is in an STI shadow or if GIF=0, KVM can't immediately + * inject the NMI. In those situations, KVM needs to single-step over + * the STI shadow or intercept STGI. + */ + if (svm_get_nmi_mask(vcpu)) { + WARN_ON_ONCE(is_vnmi_enabled(svm)); + + if (!svm->awaiting_iret_completion) + return; /* IRET will cause a vm exit */ + } + + /* + * SEV-ES guests are responsible for signaling when a vCPU is ready to + * receive a new NMI, as SEV-ES guests can't be single-stepped, i.e. + * KVM can't intercept and single-step IRET to detect when NMIs are + * unblocked (architecturally speaking). See SVM_VMGEXIT_NMI_COMPLETE. + * + * Note, GIF is guaranteed to be '1' for SEV-ES guests as hardware + * ignores SEV-ES guest writes to EFER.SVME *and* CLGI/STGI are not + * supported NAEs in the GHCB protocol. + */ + if (sev_es_guest(vcpu->kvm)) + return; if (!gif_set(svm)) { - if (vgif_enabled(svm)) + if (vgif) svm_set_intercept(svm, INTERCEPT_STGI); return; /* STGI will cause a vm exit */ } @@ -3574,21 +3920,18 @@ static void svm_enable_nmi_window(struct kvm_vcpu *vcpu) svm->vmcb->save.rflags |= (X86_EFLAGS_TF | X86_EFLAGS_RF); } -static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr) -{ - return 0; -} - -static int svm_set_identity_map_addr(struct kvm *kvm, u64 ident_addr) -{ - return 0; -} - -void svm_flush_tlb(struct kvm_vcpu *vcpu) +static void svm_flush_tlb_asid(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); /* + * Unlike VMX, SVM doesn't provide a way to flush only NPT TLB entries. + * A TLB flush for the current ASID flushes both "host" and "guest" TLB + * entries, and thus is a superset of Hyper-V's fine grained flushing. + */ + kvm_hv_vcpu_purge_flush_tlb(vcpu); + + /* * Flush only the current ASID even if the TLB flush was invoked via * kvm_flush_remote_tlbs(). Although flushing remote TLBs requires all * ASIDs to be flushed, KVM uses a single ASID for L1 and L2, and @@ -3601,6 +3944,37 @@ void svm_flush_tlb(struct kvm_vcpu *vcpu) svm->current_vmcb->asid_generation--; } +static void svm_flush_tlb_current(struct kvm_vcpu *vcpu) +{ + hpa_t root_tdp = vcpu->arch.mmu->root.hpa; + + /* + * When running on Hyper-V with EnlightenedNptTlb enabled, explicitly + * flush the NPT mappings via hypercall as flushing the ASID only + * affects virtual to physical mappings, it does not invalidate guest + * physical to host physical mappings. + */ + if (svm_hv_is_enlightened_tlb_enabled(vcpu) && VALID_PAGE(root_tdp)) + hyperv_flush_guest_mapping(root_tdp); + + svm_flush_tlb_asid(vcpu); +} + +static void svm_flush_tlb_all(struct kvm_vcpu *vcpu) +{ + /* + * When running on Hyper-V with EnlightenedNptTlb enabled, remote TLB + * flushes should be routed to hv_flush_remote_tlbs() without requesting + * a "regular" remote flush. Reaching this point means either there's + * a KVM bug or a prior hv_flush_remote_tlbs() call failed, both of + * which might be fatal to the guest. Yell, but try to recover. + */ + if (WARN_ON_ONCE(svm_hv_is_enlightened_tlb_enabled(vcpu))) + hv_flush_remote_tlbs(vcpu->kvm); + + svm_flush_tlb_asid(vcpu); +} + static void svm_flush_tlb_gva(struct kvm_vcpu *vcpu, gva_t gva) { struct vcpu_svm *svm = to_svm(vcpu); @@ -3626,8 +4000,7 @@ static inline void sync_lapic_to_cr8(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); u64 cr8; - if (nested_svm_virtualize_tpr(vcpu) || - kvm_vcpu_apicv_active(vcpu)) + if (nested_svm_virtualize_tpr(vcpu)) return; cr8 = kvm_get_cr8(vcpu); @@ -3635,24 +4008,58 @@ static inline void sync_lapic_to_cr8(struct kvm_vcpu *vcpu) svm->vmcb->control.int_ctl |= cr8 & V_TPR_MASK; } +static void svm_complete_soft_interrupt(struct kvm_vcpu *vcpu, u8 vector, + int type) +{ + bool is_exception = (type == SVM_EXITINTINFO_TYPE_EXEPT); + bool is_soft = (type == SVM_EXITINTINFO_TYPE_SOFT); + struct vcpu_svm *svm = to_svm(vcpu); + + /* + * If NRIPS is enabled, KVM must snapshot the pre-VMRUN next_rip that's + * associated with the original soft exception/interrupt. next_rip is + * cleared on all exits that can occur while vectoring an event, so KVM + * needs to manually set next_rip for re-injection. Unlike the !nrips + * case below, this needs to be done if and only if KVM is re-injecting + * the same event, i.e. if the event is a soft exception/interrupt, + * otherwise next_rip is unused on VMRUN. + */ + if (nrips && (is_soft || (is_exception && kvm_exception_is_soft(vector))) && + kvm_is_linear_rip(vcpu, svm->soft_int_old_rip + svm->soft_int_csbase)) + svm->vmcb->control.next_rip = svm->soft_int_next_rip; + /* + * If NRIPS isn't enabled, KVM must manually advance RIP prior to + * injecting the soft exception/interrupt. That advancement needs to + * be unwound if vectoring didn't complete. Note, the new event may + * not be the injected event, e.g. if KVM injected an INTn, the INTn + * hit a #NP in the guest, and the #NP encountered a #PF, the #NP will + * be the reported vectored event, but RIP still needs to be unwound. + */ + else if (!nrips && (is_soft || is_exception) && + kvm_is_linear_rip(vcpu, svm->soft_int_next_rip + svm->soft_int_csbase)) + kvm_rip_write(vcpu, svm->soft_int_old_rip); +} + static void svm_complete_interrupts(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); u8 vector; int type; u32 exitintinfo = svm->vmcb->control.exit_int_info; - unsigned int3_injected = svm->int3_injected; + bool nmi_l1_to_l2 = svm->nmi_l1_to_l2; + bool soft_int_injected = svm->soft_int_injected; - svm->int3_injected = 0; + svm->nmi_l1_to_l2 = false; + svm->soft_int_injected = false; /* - * If we've made progress since setting HF_IRET_MASK, we've + * If we've made progress since setting awaiting_iret_completion, we've * executed an IRET and can allow NMI injection. */ - if ((vcpu->arch.hflags & HF_IRET_MASK) && - (sev_es_guest(vcpu->kvm) || - kvm_rip_read(vcpu) != svm->nmi_iret_rip)) { - vcpu->arch.hflags &= ~(HF_NMI_MASK | HF_IRET_MASK); + if (svm->awaiting_iret_completion && + kvm_rip_read(vcpu) != svm->nmi_iret_rip) { + svm->awaiting_iret_completion = false; + svm->nmi_masked = false; kvm_make_request(KVM_REQ_EVENT, vcpu); } @@ -3668,42 +4075,41 @@ static void svm_complete_interrupts(struct kvm_vcpu *vcpu) vector = exitintinfo & SVM_EXITINTINFO_VEC_MASK; type = exitintinfo & SVM_EXITINTINFO_TYPE_MASK; + if (soft_int_injected) + svm_complete_soft_interrupt(vcpu, vector, type); + switch (type) { case SVM_EXITINTINFO_TYPE_NMI: vcpu->arch.nmi_injected = true; + svm->nmi_l1_to_l2 = nmi_l1_to_l2; break; - case SVM_EXITINTINFO_TYPE_EXEPT: + case SVM_EXITINTINFO_TYPE_EXEPT: { + u32 error_code = 0; + /* * Never re-inject a #VC exception. */ if (vector == X86_TRAP_VC) break; - /* - * In case of software exceptions, do not reinject the vector, - * but re-execute the instruction instead. Rewind RIP first - * if we emulated INT3 before. - */ - if (kvm_exception_is_soft(vector)) { - if (vector == BP_VECTOR && int3_injected && - kvm_is_linear_rip(vcpu, svm->int3_rip)) - kvm_rip_write(vcpu, - kvm_rip_read(vcpu) - int3_injected); - break; - } - if (exitintinfo & SVM_EXITINTINFO_VALID_ERR) { - u32 err = svm->vmcb->control.exit_int_info_err; - kvm_requeue_exception_e(vcpu, vector, err); + if (exitintinfo & SVM_EXITINTINFO_VALID_ERR) + error_code = svm->vmcb->control.exit_int_info_err; - } else - kvm_requeue_exception(vcpu, vector); + kvm_requeue_exception(vcpu, vector, + exitintinfo & SVM_EXITINTINFO_VALID_ERR, + error_code); break; + } case SVM_EXITINTINFO_TYPE_INTR: kvm_queue_interrupt(vcpu, vector, false); break; + case SVM_EXITINTINFO_TYPE_SOFT: + kvm_queue_interrupt(vcpu, vector, true); + break; default: break; } + } static void svm_cancel_injection(struct kvm_vcpu *vcpu) @@ -3717,48 +4123,84 @@ static void svm_cancel_injection(struct kvm_vcpu *vcpu) svm_complete_interrupts(vcpu); } +static int svm_vcpu_pre_run(struct kvm_vcpu *vcpu) +{ + if (to_kvm_sev_info(vcpu->kvm)->need_init) + return -EINVAL; + + return 1; +} + static fastpath_t svm_exit_handlers_fastpath(struct kvm_vcpu *vcpu) { - if (to_svm(vcpu)->vmcb->control.exit_code == SVM_EXIT_MSR && - to_svm(vcpu)->vmcb->control.exit_info_1) - return handle_fastpath_set_msr_irqoff(vcpu); + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb_control_area *control = &svm->vmcb->control; + + /* + * Next RIP must be provided as IRQs are disabled, and accessing guest + * memory to decode the instruction might fault, i.e. might sleep. + */ + if (!nrips || !control->next_rip) + return EXIT_FASTPATH_NONE; + + if (is_guest_mode(vcpu)) + return EXIT_FASTPATH_NONE; + + switch (control->exit_code) { + case SVM_EXIT_MSR: + if (!control->exit_info_1) + break; + return handle_fastpath_wrmsr(vcpu); + case SVM_EXIT_HLT: + return handle_fastpath_hlt(vcpu); + case SVM_EXIT_INVD: + return handle_fastpath_invd(vcpu); + default: + break; + } return EXIT_FASTPATH_NONE; } -static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu) +static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu, bool spec_ctrl_intercepted) { + struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu); struct vcpu_svm *svm = to_svm(vcpu); - unsigned long vmcb_pa = svm->current_vmcb->pa; - kvm_guest_enter_irqoff(); + guest_state_enter_irqoff(); - if (sev_es_guest(vcpu->kvm)) { - __svm_sev_es_vcpu_run(vmcb_pa); - } else { - struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); + /* + * Set RFLAGS.IF prior to VMRUN, as the host's RFLAGS.IF at the time of + * VMRUN controls whether or not physical IRQs are masked (KVM always + * runs with V_INTR_MASKING_MASK). Toggle RFLAGS.IF here to avoid the + * temptation to do STI+VMRUN+CLI, as AMD CPUs bleed the STI shadow + * into guest state if delivery of an event during VMRUN triggers a + * #VMEXIT, and the guest_state transitions already tell lockdep that + * IRQs are being enabled/disabled. Note! GIF=0 for the entirety of + * this path, so IRQs aren't actually unmasked while running host code. + */ + raw_local_irq_enable(); - /* - * Use a single vmcb (vmcb01 because it's always valid) for - * context switching guest state via VMLOAD/VMSAVE, that way - * the state doesn't need to be copied between vmcb01 and - * vmcb02 when switching vmcbs for nested virtualization. - */ - vmload(svm->vmcb01.pa); - __svm_vcpu_run(vmcb_pa, (unsigned long *)&vcpu->arch.regs); - vmsave(svm->vmcb01.pa); + amd_clear_divider(); - vmload(__sme_page_pa(sd->save_area)); - } + if (sev_es_guest(vcpu->kvm)) + __svm_sev_es_vcpu_run(svm, spec_ctrl_intercepted, + sev_es_host_save_area(sd)); + else + __svm_vcpu_run(svm, spec_ctrl_intercepted); + + raw_local_irq_disable(); - kvm_guest_exit_irqoff(); + guest_state_exit_irqoff(); } -static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) +static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu, u64 run_flags) { + bool force_immediate_exit = run_flags & KVM_RUN_FORCE_IMMEDIATE_EXIT; struct vcpu_svm *svm = to_svm(vcpu); + bool spec_ctrl_intercepted = msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL); - trace_kvm_entry(vcpu); + trace_kvm_entry(vcpu, force_immediate_exit); svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; @@ -3777,10 +4219,18 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) * is enough to force an immediate vmexit. */ disable_nmi_singlestep(svm); - smp_send_reschedule(vcpu->cpu); + force_immediate_exit = true; } - pre_svm_run(vcpu); + if (force_immediate_exit) + smp_send_reschedule(vcpu->cpu); + + if (pre_svm_run(vcpu)) { + vcpu->run->exit_reason = KVM_EXIT_FAIL_ENTRY; + vcpu->run->fail_entry.hardware_entry_failure_reason = SVM_EXIT_ERR; + vcpu->run->fail_entry.cpu = vcpu->cpu; + return EXIT_FASTPATH_EXIT_USERSPACE; + } sync_lapic_to_cr8(vcpu); @@ -3793,16 +4243,26 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) svm_hv_update_vp_id(svm->vmcb, vcpu); /* - * Run with all-zero DR6 unless needed, so that we can get the exact cause - * of a #DB. + * Run with all-zero DR6 unless the guest can write DR6 freely, so that + * KVM can get the exact cause of a #DB. Note, loading guest DR6 from + * KVM's snapshot is only necessary when DR accesses won't exit. */ - if (unlikely(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) - svm_set_dr6(svm, vcpu->arch.dr6); - else - svm_set_dr6(svm, DR6_ACTIVE_LOW); + if (unlikely(run_flags & KVM_RUN_LOAD_GUEST_DR6)) + svm_set_dr6(vcpu, vcpu->arch.dr6); + else if (likely(!(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT))) + svm_set_dr6(vcpu, DR6_ACTIVE_LOW); clgi(); - kvm_load_guest_xsave_state(vcpu); + + /* + * Hardware only context switches DEBUGCTL if LBR virtualization is + * enabled. Manually load DEBUGCTL if necessary (and restore it after + * VM-Exit), as running with the host's DEBUGCTL can negatively affect + * guest state and can even be fatal, e.g. due to Bus Lock Detect. + */ + if (!(svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK) && + vcpu->arch.host_debugctl != svm->vmcb->save.dbgctl) + update_debugctlmsr(svm->vmcb->save.dbgctl); kvm_wait_lapic_expire(vcpu); @@ -3813,34 +4273,12 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) * being speculatively taken. */ if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL)) - x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl); + x86_spec_ctrl_set_guest(svm->virt_spec_ctrl); - svm_vcpu_enter_exit(vcpu); - - /* - * We do not use IBRS in the kernel. If this vCPU has used the - * SPEC_CTRL MSR it may have left it on; save the value and - * turn it off. This is much more efficient than blindly adding - * it to the atomic save/restore list. Especially as the former - * (Saving guest MSRs on vmexit) doesn't even exist in KVM. - * - * For non-nested case: - * If the L01 MSR bitmap does not intercept the MSR, then we need to - * save it. - * - * For nested case: - * If the L02 MSR bitmap does not intercept the MSR, then we need to - * save it. - */ - if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL) && - unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) - svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL); - - if (!sev_es_guest(vcpu->kvm)) - reload_tss(vcpu); + svm_vcpu_enter_exit(vcpu, spec_ctrl_intercepted); if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL)) - x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl); + x86_spec_ctrl_restore_host(svm->virt_spec_ctrl); if (!sev_es_guest(vcpu->kvm)) { vcpu->arch.cr2 = svm->vmcb->save.cr2; @@ -3848,11 +4286,15 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; } + vcpu->arch.regs_dirty = 0; if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_NMI)) - kvm_before_interrupt(vcpu); + kvm_before_interrupt(vcpu, KVM_HANDLING_NMI); + + if (!(svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK) && + vcpu->arch.host_debugctl != svm->vmcb->save.dbgctl) + update_debugctlmsr(vcpu->arch.host_debugctl); - kvm_load_host_xsave_state(vcpu); stgi(); /* Any pending NMI will happen here */ @@ -3882,22 +4324,12 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) vcpu->arch.apf.host_apf_flags = kvm_read_and_reset_apf_flags(); - if (npt_enabled) - kvm_register_clear_available(vcpu, VCPU_EXREG_PDPTR); + vcpu->arch.regs_avail &= ~SVM_REGS_LAZY_LOAD_SET; - /* - * We need to handle MC intercepts here before the vcpu has a chance to - * change the physical cpu - */ - if (unlikely(svm->vmcb->control.exit_code == - SVM_EXIT_EXCP_BASE + MC_VECTOR)) - svm_handle_mce(vcpu); + trace_kvm_exit(vcpu, KVM_ISA_SVM); svm_complete_interrupts(vcpu); - if (is_guest_mode(vcpu)) - return EXIT_FASTPATH_NONE; - return svm_exit_handlers_fastpath(vcpu); } @@ -3913,11 +4345,8 @@ static void svm_load_mmu_pgd(struct kvm_vcpu *vcpu, hpa_t root_hpa, hv_track_root_tdp(vcpu, root_hpa); - /* Loading L2's CR3 is handled by enter_svm_guest_mode. */ - if (!test_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail)) - return; cr3 = vcpu->arch.cr3; - } else if (vcpu->arch.mmu->shadow_root_level >= PT64_ROOT_4LEVEL) { + } else if (root_level >= PT64_ROOT_4LEVEL) { cr3 = __sme_set(root_hpa) | kvm_get_active_pcid(vcpu); } else { /* PCID in the guest should be impossible with a 32-bit MMU. */ @@ -3929,17 +4358,6 @@ static void svm_load_mmu_pgd(struct kvm_vcpu *vcpu, hpa_t root_hpa, vmcb_mark_dirty(svm->vmcb, VMCB_CR); } -static int is_disabled(void) -{ - u64 vm_cr; - - rdmsrl(MSR_VM_CR, vm_cr); - if (vm_cr & (1 << SVM_VM_CR_SVM_DISABLE)) - return 1; - - return 0; -} - static void svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall) { @@ -3951,16 +4369,6 @@ svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall) hypercall[2] = 0xd9; } -static int __init svm_check_processor_compat(void) -{ - return 0; -} - -static bool svm_cpu_has_accelerated_tpr(void) -{ - return false; -} - /* * The kvm parameter can be NULL (module initialization, or invocation before * VM creation). Be sure to check the kvm parameter before using it. @@ -3969,9 +4377,11 @@ static bool svm_has_emulated_msr(struct kvm *kvm, u32 index) { switch (index) { case MSR_IA32_MCG_EXT_CTL: - case MSR_IA32_VMX_BASIC ... MSR_IA32_VMX_VMFUNC: + case KVM_FIRST_EMULATED_VMX_MSR ... KVM_LAST_EMULATED_VMX_MSR: return false; case MSR_IA32_SMBASE: + if (!IS_ENABLED(CONFIG_KVM_SMM)) + return false; /* SEV-ES guests do not support SMM, so report false */ if (kvm && sev_es_guest(kvm)) return false; @@ -3983,77 +4393,34 @@ static bool svm_has_emulated_msr(struct kvm *kvm, u32 index) return true; } -static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) -{ - return 0; -} - static void svm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - struct kvm_cpuid_entry2 *best; - - vcpu->arch.xsaves_enabled = guest_cpuid_has(vcpu, X86_FEATURE_XSAVE) && - boot_cpu_has(X86_FEATURE_XSAVE) && - boot_cpu_has(X86_FEATURE_XSAVES); - - /* Update nrips enabled cache */ - svm->nrips_enabled = kvm_cpu_cap_has(X86_FEATURE_NRIPS) && - guest_cpuid_has(vcpu, X86_FEATURE_NRIPS); - - svm_recalc_instruction_intercepts(vcpu, svm); - /* For sev guests, the memory encryption bit is not reserved in CR3. */ - if (sev_guest(vcpu->kvm)) { - best = kvm_find_cpuid_entry(vcpu, 0x8000001F, 0); - if (best) - vcpu->arch.reserved_gpa_bits &= ~(1UL << (best->ebx & 0x3f)); - } - - if (kvm_vcpu_apicv_active(vcpu)) { - /* - * AVIC does not work with an x2APIC mode guest. If the X2APIC feature - * is exposed to the guest, disable AVIC. - */ - if (guest_cpuid_has(vcpu, X86_FEATURE_X2APIC)) - kvm_request_apicv_update(vcpu->kvm, false, - APICV_INHIBIT_REASON_X2APIC); - - /* - * Currently, AVIC does not work with nested virtualization. - * So, we disable AVIC when cpuid for SVM is set in the L1 guest. - */ - if (nested && guest_cpuid_has(vcpu, X86_FEATURE_SVM)) - kvm_request_apicv_update(vcpu->kvm, false, - APICV_INHIBIT_REASON_NESTED); - } + /* + * SVM doesn't provide a way to disable just XSAVES in the guest, KVM + * can only disable all variants of by disallowing CR4.OSXSAVE from + * being set. As a result, if the host has XSAVE and XSAVES, and the + * guest has XSAVE enabled, the guest can execute XSAVES without + * faulting. Treat XSAVES as enabled in this case regardless of + * whether it's advertised to the guest so that KVM context switches + * XSS on VM-Enter/VM-Exit. Failure to do so would effectively give + * the guest read/write access to the host's XSS. + */ + guest_cpu_cap_change(vcpu, X86_FEATURE_XSAVES, + boot_cpu_has(X86_FEATURE_XSAVES) && + guest_cpu_cap_has(vcpu, X86_FEATURE_XSAVE)); - if (guest_cpuid_is_intel(vcpu)) { - /* - * We must intercept SYSENTER_EIP and SYSENTER_ESP - * accesses because the processor only stores 32 bits. - * For the same reason we cannot use virtual VMLOAD/VMSAVE. - */ - svm_set_intercept(svm, INTERCEPT_VMLOAD); - svm_set_intercept(svm, INTERCEPT_VMSAVE); - svm->vmcb->control.virt_ext &= ~VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; + /* + * Intercept VMLOAD if the vCPU model is Intel in order to emulate that + * VMLOAD drops bits 63:32 of SYSENTER (ignoring the fact that exposing + * SVM on Intel is bonkers and extremely unlikely to work). + */ + if (guest_cpuid_is_intel_compatible(vcpu)) + guest_cpu_cap_clear(vcpu, X86_FEATURE_V_VMSAVE_VMLOAD); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 0, 0); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 0, 0); - } else { - /* - * If hardware supports Virtual VMLOAD VMSAVE then enable it - * in VMCB and clear intercepts to avoid #VMEXIT. - */ - if (vls) { - svm_clr_intercept(svm, INTERCEPT_VMLOAD); - svm_clr_intercept(svm, INTERCEPT_VMSAVE); - svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; - } - /* No need to intercept these MSRs */ - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 1, 1); - set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 1, 1); - } + if (sev_guest(vcpu->kvm)) + sev_vcpu_after_set_cpuid(svm); } static bool svm_has_wbinvd_exit(void) @@ -4151,31 +4518,45 @@ static int svm_check_intercept(struct kvm_vcpu *vcpu, case SVM_EXIT_WRITE_CR0: { unsigned long cr0, val; - if (info->intercept == x86_intercept_cr_write) + /* + * Adjust the exit code accordingly if a CR other than CR0 is + * being written, and skip straight to the common handling as + * only CR0 has an additional selective intercept. + */ + if (info->intercept == x86_intercept_cr_write && info->modrm_reg) { icpt_info.exit_code += info->modrm_reg; - - if (icpt_info.exit_code != SVM_EXIT_WRITE_CR0 || - info->intercept == x86_intercept_clts) break; + } - if (!(vmcb_is_intercept(&svm->nested.ctl, - INTERCEPT_SELECTIVE_CR0))) + /* + * Convert the exit_code to SVM_EXIT_CR0_SEL_WRITE if a + * selective CR0 intercept is triggered (the common logic will + * treat the selective intercept as being enabled). Note, the + * unconditional intercept has higher priority, i.e. this is + * only relevant if *only* the selective intercept is enabled. + */ + if (vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_CR0_WRITE) || + !(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0))) break; - cr0 = vcpu->arch.cr0 & ~SVM_CR0_SELECTIVE_MASK; - val = info->src_val & ~SVM_CR0_SELECTIVE_MASK; + /* CLTS never triggers INTERCEPT_SELECTIVE_CR0 */ + if (info->intercept == x86_intercept_clts) + break; + /* LMSW always triggers INTERCEPT_SELECTIVE_CR0 */ if (info->intercept == x86_intercept_lmsw) { - cr0 &= 0xfUL; - val &= 0xfUL; - /* lmsw can't clear PE - catch this here */ - if (cr0 & X86_CR0_PE) - val |= X86_CR0_PE; + icpt_info.exit_code = SVM_EXIT_CR0_SEL_WRITE; + break; } + /* + * MOV-to-CR0 only triggers INTERCEPT_SELECTIVE_CR0 if any bit + * other than SVM_CR0_SELECTIVE_MASK is changed. + */ + cr0 = vcpu->arch.cr0 & ~SVM_CR0_SELECTIVE_MASK; + val = info->src_val & ~SVM_CR0_SELECTIVE_MASK; if (cr0 ^ val) icpt_info.exit_code = SVM_EXIT_CR0_SEL_WRITE; - break; } case SVM_EXIT_READ_DR0: @@ -4247,12 +4628,16 @@ out: static void svm_handle_exit_irqoff(struct kvm_vcpu *vcpu) { -} - -static void svm_sched_in(struct kvm_vcpu *vcpu, int cpu) -{ - if (!kvm_pause_in_guest(vcpu->kvm)) - shrink_ple_window(vcpu); + switch (to_svm(vcpu)->vmcb->control.exit_code) { + case SVM_EXIT_EXCP_BASE + MC_VECTOR: + svm_handle_mce(vcpu); + break; + case SVM_EXIT_INTR: + vcpu->arch.at_instruction_boundary = true; + break; + default: + break; + } } static void svm_setup_mce(struct kvm_vcpu *vcpu) @@ -4261,6 +4646,7 @@ static void svm_setup_mce(struct kvm_vcpu *vcpu) vcpu->arch.mcg_cap &= 0x1ff; } +#ifdef CONFIG_KVM_SMM bool svm_smi_blocked(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -4278,65 +4664,127 @@ static int svm_smi_allowed(struct kvm_vcpu *vcpu, bool for_injection) if (svm->nested.nested_run_pending) return -EBUSY; + if (svm_smi_blocked(vcpu)) + return 0; + /* An SMI must not be injected into L2 if it's supposed to VM-Exit. */ if (for_injection && is_guest_mode(vcpu) && nested_exit_on_smi(svm)) return -EBUSY; - return !svm_smi_blocked(vcpu); + return 1; } -static int svm_enter_smm(struct kvm_vcpu *vcpu, char *smstate) +static int svm_enter_smm(struct kvm_vcpu *vcpu, union kvm_smram *smram) { struct vcpu_svm *svm = to_svm(vcpu); + struct kvm_host_map map_save; int ret; - if (is_guest_mode(vcpu)) { - /* FED8h - SVM Guest */ - put_smstate(u64, smstate, 0x7ed8, 1); - /* FEE0h - SVM Guest VMCB Physical Address */ - put_smstate(u64, smstate, 0x7ee0, svm->nested.vmcb12_gpa); + if (!is_guest_mode(vcpu)) + return 0; - svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; - svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; - svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP]; + /* + * 32-bit SMRAM format doesn't preserve EFER and SVM state. Userspace is + * responsible for ensuring nested SVM and SMIs are mutually exclusive. + */ - ret = nested_svm_vmexit(svm); - if (ret) - return ret; - } + if (!guest_cpu_cap_has(vcpu, X86_FEATURE_LM)) + return 1; + + smram->smram64.svm_guest_flag = 1; + smram->smram64.svm_guest_vmcb_gpa = svm->nested.vmcb12_gpa; + + svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; + svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; + svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP]; + + ret = nested_svm_simple_vmexit(svm, SVM_EXIT_SW); + if (ret) + return ret; + + /* + * KVM uses VMCB01 to store L1 host state while L2 runs but + * VMCB01 is going to be used during SMM and thus the state will + * be lost. Temporary save non-VMLOAD/VMSAVE state to the host save + * area pointed to by MSR_VM_HSAVE_PA. APM guarantees that the + * format of the area is identical to guest save area offsetted + * by 0x400 (matches the offset of 'struct vmcb_save_area' + * within 'struct vmcb'). Note: HSAVE area may also be used by + * L1 hypervisor to save additional host context (e.g. KVM does + * that, see svm_prepare_switch_to_guest()) which must be + * preserved. + */ + if (kvm_vcpu_map(vcpu, gpa_to_gfn(svm->nested.hsave_msr), &map_save)) + return 1; + + BUILD_BUG_ON(offsetof(struct vmcb, save) != 0x400); + + svm_copy_vmrun_state(map_save.hva + 0x400, + &svm->vmcb01.ptr->save); + + kvm_vcpu_unmap(vcpu, &map_save); return 0; } -static int svm_leave_smm(struct kvm_vcpu *vcpu, const char *smstate) +static int svm_leave_smm(struct kvm_vcpu *vcpu, const union kvm_smram *smram) { struct vcpu_svm *svm = to_svm(vcpu); - struct kvm_host_map map; - int ret = 0; + struct kvm_host_map map, map_save; + struct vmcb *vmcb12; + int ret; - if (guest_cpuid_has(vcpu, X86_FEATURE_LM)) { - u64 saved_efer = GET_SMSTATE(u64, smstate, 0x7ed0); - u64 guest = GET_SMSTATE(u64, smstate, 0x7ed8); - u64 vmcb12_gpa = GET_SMSTATE(u64, smstate, 0x7ee0); + const struct kvm_smram_state_64 *smram64 = &smram->smram64; - if (guest) { - if (!guest_cpuid_has(vcpu, X86_FEATURE_SVM)) - return 1; + if (!guest_cpu_cap_has(vcpu, X86_FEATURE_LM)) + return 0; - if (!(saved_efer & EFER_SVME)) - return 1; + /* Non-zero if SMI arrived while vCPU was in guest mode. */ + if (!smram64->svm_guest_flag) + return 0; - if (kvm_vcpu_map(vcpu, - gpa_to_gfn(vmcb12_gpa), &map) == -EINVAL) - return 1; + if (!guest_cpu_cap_has(vcpu, X86_FEATURE_SVM)) + return 1; - if (svm_allocate_nested(svm)) - return 1; + if (!(smram64->efer & EFER_SVME)) + return 1; - ret = enter_svm_guest_mode(vcpu, vmcb12_gpa, map.hva); - kvm_vcpu_unmap(vcpu, &map, true); - } - } + if (kvm_vcpu_map(vcpu, gpa_to_gfn(smram64->svm_guest_vmcb_gpa), &map)) + return 1; + + ret = 1; + if (kvm_vcpu_map(vcpu, gpa_to_gfn(svm->nested.hsave_msr), &map_save)) + goto unmap_map; + if (svm_allocate_nested(svm)) + goto unmap_save; + + /* + * Restore L1 host state from L1 HSAVE area as VMCB01 was + * used during SMM (see svm_enter_smm()) + */ + + svm_copy_vmrun_state(&svm->vmcb01.ptr->save, map_save.hva + 0x400); + + /* + * Enter the nested guest now + */ + + vmcb_mark_all_dirty(svm->vmcb01.ptr); + + vmcb12 = map.hva; + nested_copy_vmcb_control_to_cache(svm, &vmcb12->control); + nested_copy_vmcb_save_to_cache(svm, &vmcb12->save); + ret = enter_svm_guest_mode(vcpu, smram64->svm_guest_vmcb_gpa, vmcb12, false); + + if (ret) + goto unmap_save; + + svm->nested.nested_run_pending = 1; + +unmap_save: + kvm_vcpu_unmap(vcpu, &map_save); +unmap_map: + kvm_vcpu_unmap(vcpu, &map); return ret; } @@ -4345,103 +4793,173 @@ static void svm_enable_smi_window(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); if (!gif_set(svm)) { - if (vgif_enabled(svm)) + if (vgif) svm_set_intercept(svm, INTERCEPT_STGI); /* STGI will cause a vm exit */ } else { /* We must be in SMM; RSM will cause a vmexit anyway. */ } } +#endif -static bool svm_can_emulate_instruction(struct kvm_vcpu *vcpu, void *insn, int insn_len) +static int svm_check_emulate_instruction(struct kvm_vcpu *vcpu, int emul_type, + void *insn, int insn_len) { + struct vcpu_svm *svm = to_svm(vcpu); bool smep, smap, is_user; - unsigned long cr4; + u64 error_code; + + /* Check that emulation is possible during event vectoring */ + if ((svm->vmcb->control.exit_int_info & SVM_EXITINTINFO_TYPE_MASK) && + !kvm_can_emulate_event_vectoring(emul_type)) + return X86EMUL_UNHANDLEABLE_VECTORING; + + /* Emulation is always possible when KVM has access to all guest state. */ + if (!sev_guest(vcpu->kvm)) + return X86EMUL_CONTINUE; + + /* #UD and #GP should never be intercepted for SEV guests. */ + WARN_ON_ONCE(emul_type & (EMULTYPE_TRAP_UD | + EMULTYPE_TRAP_UD_FORCED | + EMULTYPE_VMWARE_GP)); /* - * When the guest is an SEV-ES guest, emulation is not possible. + * Emulation is impossible for SEV-ES guests as KVM doesn't have access + * to guest register state. */ if (sev_es_guest(vcpu->kvm)) - return false; + return X86EMUL_RETRY_INSTR; + + /* + * Emulation is possible if the instruction is already decoded, e.g. + * when completing I/O after returning from userspace. + */ + if (emul_type & EMULTYPE_NO_DECODE) + return X86EMUL_CONTINUE; + + /* + * Emulation is possible for SEV guests if and only if a prefilled + * buffer containing the bytes of the intercepted instruction is + * available. SEV guest memory is encrypted with a guest specific key + * and cannot be decrypted by KVM, i.e. KVM would read ciphertext and + * decode garbage. + * + * If KVM is NOT trying to simply skip an instruction, inject #UD if + * KVM reached this point without an instruction buffer. In practice, + * this path should never be hit by a well-behaved guest, e.g. KVM + * doesn't intercept #UD or #GP for SEV guests, but this path is still + * theoretically reachable, e.g. via unaccelerated fault-like AVIC + * access, and needs to be handled by KVM to avoid putting the guest + * into an infinite loop. Injecting #UD is somewhat arbitrary, but + * its the least awful option given lack of insight into the guest. + * + * If KVM is trying to skip an instruction, simply resume the guest. + * If a #NPF occurs while the guest is vectoring an INT3/INTO, then KVM + * will attempt to re-inject the INT3/INTO and skip the instruction. + * In that scenario, retrying the INT3/INTO and hoping the guest will + * make forward progress is the only option that has a chance of + * success (and in practice it will work the vast majority of the time). + */ + if (unlikely(!insn)) { + if (emul_type & EMULTYPE_SKIP) + return X86EMUL_UNHANDLEABLE; + + kvm_queue_exception(vcpu, UD_VECTOR); + return X86EMUL_PROPAGATE_FAULT; + } + + /* + * Emulate for SEV guests if the insn buffer is not empty. The buffer + * will be empty if the DecodeAssist microcode cannot fetch bytes for + * the faulting instruction because the code fetch itself faulted, e.g. + * the guest attempted to fetch from emulated MMIO or a guest page + * table used to translate CS:RIP resides in emulated MMIO. + */ + if (likely(insn_len)) + return X86EMUL_CONTINUE; /* * Detect and workaround Errata 1096 Fam_17h_00_0Fh. * * Errata: - * When CPU raise #NPF on guest data access and vCPU CR4.SMAP=1, it is - * possible that CPU microcode implementing DecodeAssist will fail - * to read bytes of instruction which caused #NPF. In this case, - * GuestIntrBytes field of the VMCB on a VMEXIT will incorrectly - * return 0 instead of the correct guest instruction bytes. - * - * This happens because CPU microcode reading instruction bytes - * uses a special opcode which attempts to read data using CPL=0 - * privileges. The microcode reads CS:RIP and if it hits a SMAP - * fault, it gives up and returns no instruction bytes. + * When CPU raises #NPF on guest data access and vCPU CR4.SMAP=1, it is + * possible that CPU microcode implementing DecodeAssist will fail to + * read guest memory at CS:RIP and vmcb.GuestIntrBytes will incorrectly + * be '0'. This happens because microcode reads CS:RIP using a _data_ + * loap uop with CPL=0 privileges. If the load hits a SMAP #PF, ucode + * gives up and does not fill the instruction bytes buffer. * - * Detection: - * We reach here in case CPU supports DecodeAssist, raised #NPF and - * returned 0 in GuestIntrBytes field of the VMCB. - * First, errata can only be triggered in case vCPU CR4.SMAP=1. - * Second, if vCPU CR4.SMEP=1, errata could only be triggered - * in case vCPU CPL==3 (Because otherwise guest would have triggered - * a SMEP fault instead of #NPF). - * Otherwise, vCPU CR4.SMEP=0, errata could be triggered by any vCPU CPL. - * As most guests enable SMAP if they have also enabled SMEP, use above - * logic in order to attempt minimize false-positive of detecting errata - * while still preserving all cases semantic correctness. + * As above, KVM reaches this point iff the VM is an SEV guest, the CPU + * supports DecodeAssist, a #NPF was raised, KVM's page fault handler + * triggered emulation (e.g. for MMIO), and the CPU returned 0 in the + * GuestIntrBytes field of the VMCB. * - * Workaround: - * To determine what instruction the guest was executing, the hypervisor - * will have to decode the instruction at the instruction pointer. + * This does _not_ mean that the erratum has been encountered, as the + * DecodeAssist will also fail if the load for CS:RIP hits a legitimate + * #PF, e.g. if the guest attempt to execute from emulated MMIO and + * encountered a reserved/not-present #PF. * - * In non SEV guest, hypervisor will be able to read the guest - * memory to decode the instruction pointer when insn_len is zero - * so we return true to indicate that decoding is possible. + * To hit the erratum, the following conditions must be true: + * 1. CR4.SMAP=1 (obviously). + * 2. CR4.SMEP=0 || CPL=3. If SMEP=1 and CPL<3, the erratum cannot + * have been hit as the guest would have encountered a SMEP + * violation #PF, not a #NPF. + * 3. The #NPF is not due to a code fetch, in which case failure to + * retrieve the instruction bytes is legitimate (see abvoe). * - * But in the SEV guest, the guest memory is encrypted with the - * guest specific key and hypervisor will not be able to decode the - * instruction pointer so we will not able to workaround it. Lets - * print the error and request to kill the guest. - */ - if (likely(!insn || insn_len)) - return true; - - /* - * If RIP is invalid, go ahead with emulation which will cause an - * internal error exit. + * In addition, don't apply the erratum workaround if the #NPF occurred + * while translating guest page tables (see below). */ - if (!kvm_vcpu_gfn_to_memslot(vcpu, kvm_rip_read(vcpu) >> PAGE_SHIFT)) - return true; + error_code = svm->vmcb->control.exit_info_1; + if (error_code & (PFERR_GUEST_PAGE_MASK | PFERR_FETCH_MASK)) + goto resume_guest; - cr4 = kvm_read_cr4(vcpu); - smep = cr4 & X86_CR4_SMEP; - smap = cr4 & X86_CR4_SMAP; + smep = kvm_is_cr4_bit_set(vcpu, X86_CR4_SMEP); + smap = kvm_is_cr4_bit_set(vcpu, X86_CR4_SMAP); is_user = svm_get_cpl(vcpu) == 3; if (smap && (!smep || is_user)) { - if (!sev_guest(vcpu->kvm)) - return true; + pr_err_ratelimited("SEV Guest triggered AMD Erratum 1096\n"); - pr_err_ratelimited("KVM: SEV Guest triggered AMD Erratum 1096\n"); - kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); + /* + * If the fault occurred in userspace, arbitrarily inject #GP + * to avoid killing the guest and to hopefully avoid confusing + * the guest kernel too much, e.g. injecting #PF would not be + * coherent with respect to the guest's page tables. Request + * triple fault if the fault occurred in the kernel as there's + * no fault that KVM can inject without confusing the guest. + * In practice, the triple fault is moot as no sane SEV kernel + * will execute from user memory while also running with SMAP=1. + */ + if (is_user) + kvm_inject_gp(vcpu, 0); + else + kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); + return X86EMUL_PROPAGATE_FAULT; } - return false; +resume_guest: + /* + * If the erratum was not hit, simply resume the guest and let it fault + * again. While awful, e.g. the vCPU may get stuck in an infinite loop + * if the fault is at CPL=0, it's the lesser of all evils. Exiting to + * userspace will kill the guest, and letting the emulator read garbage + * will yield random behavior and potentially corrupt the guest. + * + * Simply resuming the guest is technically not a violation of the SEV + * architecture. AMD's APM states that all code fetches and page table + * accesses for SEV guest are encrypted, regardless of the C-Bit. The + * APM also states that encrypted accesses to MMIO are "ignored", but + * doesn't explicitly define "ignored", i.e. doing nothing and letting + * the guest spin is technically "ignoring" the access. + */ + return X86EMUL_RETRY_INSTR; } static bool svm_apic_init_signal_blocked(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - /* - * TODO: Last condition latch INIT signals on vCPU when - * vCPU is in guest-mode and vmcb12 defines intercept on INIT. - * To properly emulate the INIT intercept, - * svm_check_nested_events() should call nested_svm_vmexit() - * if an INIT signal is pending. - */ - return !gif_set(svm) || - (vmcb_is_intercept(&svm->vmcb->control, INTERCEPT_INIT)); + return !gif_set(svm); } static void svm_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector) @@ -4456,12 +4974,26 @@ static void svm_vm_destroy(struct kvm *kvm) { avic_vm_destroy(kvm); sev_vm_destroy(kvm); + + svm_srso_vm_destroy(); } static int svm_vm_init(struct kvm *kvm) { + int type = kvm->arch.vm_type; + + if (type != KVM_X86_DEFAULT_VM && + type != KVM_X86_SW_PROTECTED_VM) { + kvm->arch.has_protected_state = + (type == KVM_X86_SEV_ES_VM || type == KVM_X86_SNP_VM); + to_kvm_sev_info(kvm)->need_init = true; + + kvm->arch.has_private_mem = (type == KVM_X86_SNP_VM); + kvm->arch.pre_fault_allowed = !kvm->arch.has_private_mem; + } + if (!pause_filter_count || !pause_filter_thresh) - kvm->arch.pause_in_guest = true; + kvm_disable_exits(kvm, KVM_X86_DISABLE_EXITS_PAUSE); if (enable_apicv) { int ret = avic_vm_init(kvm); @@ -4469,40 +5001,59 @@ static int svm_vm_init(struct kvm *kvm) return ret; } + svm_srso_vm_init(); return 0; } -static struct kvm_x86_ops svm_x86_ops __initdata = { - .hardware_unsetup = svm_hardware_teardown, - .hardware_enable = svm_hardware_enable, - .hardware_disable = svm_hardware_disable, - .cpu_has_accelerated_tpr = svm_cpu_has_accelerated_tpr, +static void *svm_alloc_apic_backing_page(struct kvm_vcpu *vcpu) +{ + struct page *page = snp_safe_alloc_page(); + + if (!page) + return NULL; + + return page_address(page); +} + +struct kvm_x86_ops svm_x86_ops __initdata = { + .name = KBUILD_MODNAME, + + .check_processor_compatibility = svm_check_processor_compat, + + .hardware_unsetup = svm_hardware_unsetup, + .enable_virtualization_cpu = svm_enable_virtualization_cpu, + .disable_virtualization_cpu = svm_disable_virtualization_cpu, + .emergency_disable_virtualization_cpu = svm_emergency_disable_virtualization_cpu, .has_emulated_msr = svm_has_emulated_msr, - .vcpu_create = svm_create_vcpu, - .vcpu_free = svm_free_vcpu, + .vcpu_precreate = svm_vcpu_precreate, + .vcpu_create = svm_vcpu_create, + .vcpu_free = svm_vcpu_free, .vcpu_reset = svm_vcpu_reset, .vm_size = sizeof(struct kvm_svm), .vm_init = svm_vm_init, .vm_destroy = svm_vm_destroy, - .prepare_guest_switch = svm_prepare_guest_switch, + .prepare_switch_to_guest = svm_prepare_switch_to_guest, .vcpu_load = svm_vcpu_load, .vcpu_put = svm_vcpu_put, - .vcpu_blocking = svm_vcpu_blocking, - .vcpu_unblocking = svm_vcpu_unblocking, + .vcpu_blocking = avic_vcpu_blocking, + .vcpu_unblocking = avic_vcpu_unblocking, .update_exception_bitmap = svm_update_exception_bitmap, - .get_msr_feature = svm_get_msr_feature, + .get_feature_msr = svm_get_feature_msr, .get_msr = svm_get_msr, .set_msr = svm_set_msr, .get_segment_base = svm_get_segment_base, .get_segment = svm_get_segment, .set_segment = svm_set_segment, .get_cpl = svm_get_cpl, - .get_cs_db_l_bits = kvm_get_cs_db_l_bits, + .get_cpl_no_cache = svm_get_cpl, + .get_cs_db_l_bits = svm_get_cs_db_l_bits, + .is_valid_cr0 = svm_is_valid_cr0, .set_cr0 = svm_set_cr0, + .post_set_cr3 = sev_post_set_cr3, .is_valid_cr4 = svm_is_valid_cr4, .set_cr4 = svm_set_cr4, .set_efer = svm_set_efer, @@ -4515,22 +5066,26 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .cache_reg = svm_cache_reg, .get_rflags = svm_get_rflags, .set_rflags = svm_set_rflags, + .get_if_flag = svm_get_if_flag, - .tlb_flush_all = svm_flush_tlb, - .tlb_flush_current = svm_flush_tlb, - .tlb_flush_gva = svm_flush_tlb_gva, - .tlb_flush_guest = svm_flush_tlb, + .flush_tlb_all = svm_flush_tlb_all, + .flush_tlb_current = svm_flush_tlb_current, + .flush_tlb_gva = svm_flush_tlb_gva, + .flush_tlb_guest = svm_flush_tlb_asid, - .run = svm_vcpu_run, - .handle_exit = handle_exit, - .skip_emulated_instruction = skip_emulated_instruction, + .vcpu_pre_run = svm_vcpu_pre_run, + .vcpu_run = svm_vcpu_run, + .handle_exit = svm_handle_exit, + .skip_emulated_instruction = svm_skip_emulated_instruction, .update_emulated_instruction = NULL, .set_interrupt_shadow = svm_set_interrupt_shadow, .get_interrupt_shadow = svm_get_interrupt_shadow, .patch_hypercall = svm_patch_hypercall, - .set_irq = svm_set_irq, - .set_nmi = svm_inject_nmi, - .queue_exception = svm_queue_exception, + .inject_irq = svm_inject_irq, + .inject_nmi = svm_inject_nmi, + .is_vnmi_pending = svm_is_vnmi_pending, + .set_vnmi_pending = svm_set_vnmi_pending, + .inject_exception = svm_inject_exception, .cancel_injection = svm_cancel_injection, .interrupt_allowed = svm_interrupt_allowed, .nmi_allowed = svm_nmi_allowed, @@ -4539,21 +5094,15 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .enable_nmi_window = svm_enable_nmi_window, .enable_irq_window = svm_enable_irq_window, .update_cr8_intercept = svm_update_cr8_intercept, - .set_virtual_apic_mode = svm_set_virtual_apic_mode, - .refresh_apicv_exec_ctrl = svm_refresh_apicv_exec_ctrl, - .check_apicv_inhibit_reasons = svm_check_apicv_inhibit_reasons, - .pre_update_apicv_exec_ctrl = svm_pre_update_apicv_exec_ctrl, - .load_eoi_exitmap = svm_load_eoi_exitmap, - .hwapic_irr_update = svm_hwapic_irr_update, - .hwapic_isr_update = svm_hwapic_isr_update, - .sync_pir_to_irr = kvm_lapic_find_highest_irr, - .apicv_post_state_restore = avic_post_state_restore, - - .set_tss_addr = svm_set_tss_addr, - .set_identity_map_addr = svm_set_identity_map_addr, - .get_mt_mask = svm_get_mt_mask, + + .x2apic_icr_is_split = true, + .set_virtual_apic_mode = avic_refresh_virtual_apic_mode, + .refresh_apicv_exec_ctrl = avic_refresh_apicv_exec_ctrl, + .apicv_post_state_restore = avic_apicv_post_state_restore, + .required_apicv_inhibits = AVIC_REQUIRED_APICV_INHIBITS, .get_exit_info = svm_get_exit_info, + .get_entry_info = svm_get_entry_info, .vcpu_after_set_cpuid = svm_vcpu_after_set_cpuid, @@ -4569,59 +5118,392 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .check_intercept = svm_check_intercept, .handle_exit_irqoff = svm_handle_exit_irqoff, - .request_immediate_exit = __kvm_request_immediate_exit, - - .sched_in = svm_sched_in, - - .pmu_ops = &amd_pmu_ops, .nested_ops = &svm_nested_ops, - .deliver_posted_interrupt = svm_deliver_avic_intr, - .dy_apicv_has_pending_interrupt = svm_dy_apicv_has_pending_interrupt, - .update_pi_irte = svm_update_pi_irte, + .deliver_interrupt = svm_deliver_interrupt, + .pi_update_irte = avic_pi_update_irte, .setup_mce = svm_setup_mce, +#ifdef CONFIG_KVM_SMM .smi_allowed = svm_smi_allowed, .enter_smm = svm_enter_smm, .leave_smm = svm_leave_smm, .enable_smi_window = svm_enable_smi_window, +#endif - .mem_enc_op = svm_mem_enc_op, - .mem_enc_reg_region = svm_register_enc_region, - .mem_enc_unreg_region = svm_unregister_enc_region, - - .vm_copy_enc_context_from = svm_vm_copy_asid_from, +#ifdef CONFIG_KVM_AMD_SEV + .dev_get_attr = sev_dev_get_attr, + .mem_enc_ioctl = sev_mem_enc_ioctl, + .mem_enc_register_region = sev_mem_enc_register_region, + .mem_enc_unregister_region = sev_mem_enc_unregister_region, + .guest_memory_reclaimed = sev_guest_memory_reclaimed, - .can_emulate_instruction = svm_can_emulate_instruction, + .vm_copy_enc_context_from = sev_vm_copy_enc_context_from, + .vm_move_enc_context_from = sev_vm_move_enc_context_from, +#endif + .check_emulate_instruction = svm_check_emulate_instruction, .apic_init_signal_blocked = svm_apic_init_signal_blocked, - .msr_filter_changed = svm_msr_filter_changed, + .recalc_intercepts = svm_recalc_intercepts, .complete_emulated_msr = svm_complete_emulated_msr, .vcpu_deliver_sipi_vector = svm_vcpu_deliver_sipi_vector, + .vcpu_get_apicv_inhibit_reasons = avic_vcpu_get_apicv_inhibit_reasons, + .alloc_apic_backing_page = svm_alloc_apic_backing_page, + + .gmem_prepare = sev_gmem_prepare, + .gmem_invalidate = sev_gmem_invalidate, + .gmem_max_mapping_level = sev_gmem_max_mapping_level, }; +/* + * The default MMIO mask is a single bit (excluding the present bit), + * which could conflict with the memory encryption bit. Check for + * memory encryption support and override the default MMIO mask if + * memory encryption is enabled. + */ +static __init void svm_adjust_mmio_mask(void) +{ + unsigned int enc_bit, mask_bit; + u64 msr, mask; + + /* If there is no memory encryption support, use existing mask */ + if (cpuid_eax(0x80000000) < 0x8000001f) + return; + + /* If memory encryption is not enabled, use existing mask */ + rdmsrq(MSR_AMD64_SYSCFG, msr); + if (!(msr & MSR_AMD64_SYSCFG_MEM_ENCRYPT)) + return; + + enc_bit = cpuid_ebx(0x8000001f) & 0x3f; + mask_bit = boot_cpu_data.x86_phys_bits; + + /* Increment the mask bit if it is the same as the encryption bit */ + if (enc_bit == mask_bit) + mask_bit++; + + /* + * If the mask bit location is below 52, then some bits above the + * physical addressing limit will always be reserved, so use the + * rsvd_bits() function to generate the mask. This mask, along with + * the present bit, will be used to generate a page fault with + * PFER.RSV = 1. + * + * If the mask bit location is 52 (or above), then clear the mask. + */ + mask = (mask_bit < 52) ? rsvd_bits(mask_bit, 51) | PT_PRESENT_MASK : 0; + + kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK); +} + +static __init void svm_set_cpu_caps(void) +{ + kvm_set_cpu_caps(); + + kvm_caps.supported_perf_cap = 0; + + kvm_cpu_cap_clear(X86_FEATURE_IBT); + + /* CPUID 0x80000001 and 0x8000000A (SVM features) */ + if (nested) { + kvm_cpu_cap_set(X86_FEATURE_SVM); + kvm_cpu_cap_set(X86_FEATURE_VMCBCLEAN); + + /* + * KVM currently flushes TLBs on *every* nested SVM transition, + * and so for all intents and purposes KVM supports flushing by + * ASID, i.e. KVM is guaranteed to honor every L1 ASID flush. + */ + kvm_cpu_cap_set(X86_FEATURE_FLUSHBYASID); + + if (nrips) + kvm_cpu_cap_set(X86_FEATURE_NRIPS); + + if (npt_enabled) + kvm_cpu_cap_set(X86_FEATURE_NPT); + + if (tsc_scaling) + kvm_cpu_cap_set(X86_FEATURE_TSCRATEMSR); + + if (vls) + kvm_cpu_cap_set(X86_FEATURE_V_VMSAVE_VMLOAD); + if (lbrv) + kvm_cpu_cap_set(X86_FEATURE_LBRV); + + if (boot_cpu_has(X86_FEATURE_PAUSEFILTER)) + kvm_cpu_cap_set(X86_FEATURE_PAUSEFILTER); + + if (boot_cpu_has(X86_FEATURE_PFTHRESHOLD)) + kvm_cpu_cap_set(X86_FEATURE_PFTHRESHOLD); + + if (vgif) + kvm_cpu_cap_set(X86_FEATURE_VGIF); + + if (vnmi) + kvm_cpu_cap_set(X86_FEATURE_VNMI); + + /* Nested VM can receive #VMEXIT instead of triggering #GP */ + kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK); + } + + if (cpu_feature_enabled(X86_FEATURE_BUS_LOCK_THRESHOLD)) + kvm_caps.has_bus_lock_exit = true; + + /* CPUID 0x80000008 */ + if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD) || + boot_cpu_has(X86_FEATURE_AMD_SSBD)) + kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD); + + if (enable_pmu) { + /* + * Enumerate support for PERFCTR_CORE if and only if KVM has + * access to enough counters to virtualize "core" support, + * otherwise limit vPMU support to the legacy number of counters. + */ + if (kvm_pmu_cap.num_counters_gp < AMD64_NUM_COUNTERS_CORE) + kvm_pmu_cap.num_counters_gp = min(AMD64_NUM_COUNTERS, + kvm_pmu_cap.num_counters_gp); + else + kvm_cpu_cap_check_and_set(X86_FEATURE_PERFCTR_CORE); + + if (kvm_pmu_cap.version != 2 || + !kvm_cpu_cap_has(X86_FEATURE_PERFCTR_CORE)) + kvm_cpu_cap_clear(X86_FEATURE_PERFMON_V2); + } + + /* CPUID 0x8000001F (SME/SEV features) */ + sev_set_cpu_caps(); + + /* + * Clear capabilities that are automatically configured by common code, + * but that require explicit SVM support (that isn't yet implemented). + */ + kvm_cpu_cap_clear(X86_FEATURE_BUS_LOCK_DETECT); + kvm_cpu_cap_clear(X86_FEATURE_MSR_IMM); +} + +static __init int svm_hardware_setup(void) +{ + void *iopm_va; + int cpu, r; + + /* + * NX is required for shadow paging and for NPT if the NX huge pages + * mitigation is enabled. + */ + if (!boot_cpu_has(X86_FEATURE_NX)) { + pr_err_ratelimited("NX (Execute Disable) not supported\n"); + return -EOPNOTSUPP; + } + kvm_enable_efer_bits(EFER_NX); + + kvm_caps.supported_xcr0 &= ~(XFEATURE_MASK_BNDREGS | + XFEATURE_MASK_BNDCSR); + + if (boot_cpu_has(X86_FEATURE_FXSR_OPT)) + kvm_enable_efer_bits(EFER_FFXSR); + + if (tsc_scaling) { + if (!boot_cpu_has(X86_FEATURE_TSCRATEMSR)) { + tsc_scaling = false; + } else { + pr_info("TSC scaling supported\n"); + kvm_caps.has_tsc_control = true; + } + } + kvm_caps.max_tsc_scaling_ratio = SVM_TSC_RATIO_MAX; + kvm_caps.tsc_scaling_ratio_frac_bits = 32; + + tsc_aux_uret_slot = kvm_add_user_return_msr(MSR_TSC_AUX); + + if (boot_cpu_has(X86_FEATURE_AUTOIBRS)) + kvm_enable_efer_bits(EFER_AUTOIBRS); + + /* Check for pause filtering support */ + if (!boot_cpu_has(X86_FEATURE_PAUSEFILTER)) { + pause_filter_count = 0; + pause_filter_thresh = 0; + } else if (!boot_cpu_has(X86_FEATURE_PFTHRESHOLD)) { + pause_filter_thresh = 0; + } + + if (nested) { + pr_info("Nested Virtualization enabled\n"); + kvm_enable_efer_bits(EFER_SVME); + if (!boot_cpu_has(X86_FEATURE_EFER_LMSLE_MBZ)) + kvm_enable_efer_bits(EFER_LMSLE); + + r = nested_svm_init_msrpm_merge_offsets(); + if (r) + return r; + } + + /* + * KVM's MMU doesn't support using 2-level paging for itself, and thus + * NPT isn't supported if the host is using 2-level paging since host + * CR4 is unchanged on VMRUN. + */ + if (!IS_ENABLED(CONFIG_X86_64) && !IS_ENABLED(CONFIG_X86_PAE)) + npt_enabled = false; + + if (!boot_cpu_has(X86_FEATURE_NPT)) + npt_enabled = false; + + /* Force VM NPT level equal to the host's paging level */ + kvm_configure_mmu(npt_enabled, get_npt_level(), + get_npt_level(), PG_LEVEL_1G); + pr_info("Nested Paging %s\n", str_enabled_disabled(npt_enabled)); + + /* + * It seems that on AMD processors PTE's accessed bit is + * being set by the CPU hardware before the NPF vmexit. + * This is not expected behaviour and our tests fail because + * of it. + * A workaround here is to disable support for + * GUEST_MAXPHYADDR < HOST_MAXPHYADDR if NPT is enabled. + * In this case userspace can know if there is support using + * KVM_CAP_SMALLER_MAXPHYADDR extension and decide how to handle + * it + * If future AMD CPU models change the behaviour described above, + * this variable can be changed accordingly + */ + allow_smaller_maxphyaddr = !npt_enabled; + + /* Setup shadow_me_value and shadow_me_mask */ + kvm_mmu_set_me_spte_mask(sme_me_mask, sme_me_mask); + + svm_adjust_mmio_mask(); + + nrips = nrips && boot_cpu_has(X86_FEATURE_NRIPS); + + if (lbrv) { + if (!boot_cpu_has(X86_FEATURE_LBRV)) + lbrv = false; + else + pr_info("LBR virtualization supported\n"); + } + + iopm_va = svm_alloc_permissions_map(IOPM_SIZE, GFP_KERNEL); + if (!iopm_va) + return -ENOMEM; + + iopm_base = __sme_set(__pa(iopm_va)); + + /* + * Note, SEV setup consumes npt_enabled and enable_mmio_caching (which + * may be modified by svm_adjust_mmio_mask()), as well as nrips. + */ + sev_hardware_setup(); + + svm_hv_hardware_setup(); + + enable_apicv = avic_hardware_setup(); + if (!enable_apicv) { + enable_ipiv = false; + svm_x86_ops.vcpu_blocking = NULL; + svm_x86_ops.vcpu_unblocking = NULL; + svm_x86_ops.vcpu_get_apicv_inhibit_reasons = NULL; + } + + if (vls) { + if (!npt_enabled || + !boot_cpu_has(X86_FEATURE_V_VMSAVE_VMLOAD) || + !IS_ENABLED(CONFIG_X86_64)) { + vls = false; + } else { + pr_info("Virtual VMLOAD VMSAVE supported\n"); + } + } + + if (boot_cpu_has(X86_FEATURE_SVME_ADDR_CHK)) + svm_gp_erratum_intercept = false; + + if (vgif) { + if (!boot_cpu_has(X86_FEATURE_VGIF)) + vgif = false; + else + pr_info("Virtual GIF supported\n"); + } + + vnmi = vgif && vnmi && boot_cpu_has(X86_FEATURE_VNMI); + if (vnmi) + pr_info("Virtual NMI enabled\n"); + + if (!vnmi) { + svm_x86_ops.is_vnmi_pending = NULL; + svm_x86_ops.set_vnmi_pending = NULL; + } + + if (!enable_pmu) + pr_info("PMU virtualization is disabled\n"); + + svm_set_cpu_caps(); + + kvm_caps.inapplicable_quirks &= ~KVM_X86_QUIRK_CD_NW_CLEARED; + + for_each_possible_cpu(cpu) { + r = svm_cpu_init(cpu); + if (r) + goto err; + } + + return 0; + +err: + svm_hardware_unsetup(); + return r; +} + + static struct kvm_x86_init_ops svm_init_ops __initdata = { - .cpu_has_kvm_support = has_svm, - .disabled_by_bios = is_disabled, .hardware_setup = svm_hardware_setup, - .check_processor_compatibility = svm_check_processor_compat, .runtime_ops = &svm_x86_ops, + .pmu_ops = &amd_pmu_ops, }; +static void __svm_exit(void) +{ + kvm_x86_vendor_exit(); +} + static int __init svm_init(void) { + int r; + + KVM_SANITY_CHECK_VM_STRUCT_SIZE(kvm_svm); + __unused_size_checks(); - return kvm_init(&svm_init_ops, sizeof(struct vcpu_svm), - __alignof__(struct vcpu_svm), THIS_MODULE); + if (!kvm_is_svm_supported()) + return -EOPNOTSUPP; + + r = kvm_x86_vendor_init(&svm_init_ops); + if (r) + return r; + + /* + * Common KVM initialization _must_ come last, after this, /dev/kvm is + * exposed to userspace! + */ + r = kvm_init(sizeof(struct vcpu_svm), __alignof__(struct vcpu_svm), + THIS_MODULE); + if (r) + goto err_kvm_init; + + return 0; + +err_kvm_init: + __svm_exit(); + return r; } static void __exit svm_exit(void) { kvm_exit(); + __svm_exit(); } module_init(svm_init) diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h index f89b623bb591..9e151dbdef25 100644 --- a/arch/x86/kvm/svm/svm.h +++ b/arch/x86/kvm/svm/svm.h @@ -22,15 +22,38 @@ #include <asm/svm.h> #include <asm/sev-common.h> -#define __sme_page_pa(x) __sme_set(page_to_pfn(x) << PAGE_SHIFT) +#include "cpuid.h" +#include "kvm_cache_regs.h" + +/* + * Helpers to convert to/from physical addresses for pages whose address is + * consumed directly by hardware. Even though it's a physical address, SVM + * often restricts the address to the natural width, hence 'unsigned long' + * instead of 'hpa_t'. + */ +static inline unsigned long __sme_page_pa(struct page *page) +{ + return __sme_set(page_to_pfn(page) << PAGE_SHIFT); +} + +static inline struct page *__sme_pa_to_page(unsigned long pa) +{ + return pfn_to_page(__sme_clr(pa) >> PAGE_SHIFT); +} #define IOPM_SIZE PAGE_SIZE * 3 #define MSRPM_SIZE PAGE_SIZE * 2 -#define MAX_DIRECT_ACCESS_MSRS 20 -#define MSRPM_OFFSETS 16 -extern u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly; extern bool npt_enabled; +extern int nrips; +extern int vgif; +extern bool intercept_smi; +extern bool vnmi; +extern int lbrv; + +extern int tsc_aux_uret_slot __ro_after_init; + +extern struct kvm_x86_ops svm_x86_ops __initdata; /* * Clean bits in VMCB. @@ -54,6 +77,7 @@ enum { * AVIC PHYSICAL_TABLE pointer, * AVIC LOGICAL_TABLE pointer */ + VMCB_CET, /* S_CET, SSP, ISST_ADDR */ VMCB_SW = 31, /* Reserved for hypervisor/software use */ }; @@ -62,7 +86,7 @@ enum { (1U << VMCB_ASID) | (1U << VMCB_INTR) | \ (1U << VMCB_NPT) | (1U << VMCB_CR) | (1U << VMCB_DR) | \ (1U << VMCB_DT) | (1U << VMCB_SEG) | (1U << VMCB_CR2) | \ - (1U << VMCB_LBR) | (1U << VMCB_AVIC) | \ + (1U << VMCB_LBR) | (1U << VMCB_AVIC) | (1U << VMCB_CET) | \ (1U << VMCB_SW)) /* TPR and CR2 are always written before VMRUN */ @@ -71,14 +95,26 @@ enum { struct kvm_sev_info { bool active; /* SEV enabled guest */ bool es_active; /* SEV-ES enabled guest */ + bool need_init; /* waiting for SEV_INIT2 */ unsigned int asid; /* ASID used for this guest */ unsigned int handle; /* SEV firmware handle */ int fd; /* SEV device fd */ + unsigned long policy; unsigned long pages_locked; /* Number of pages locked */ struct list_head regions_list; /* List of registered regions */ u64 ap_jump_table; /* SEV-ES AP Jump Table address */ + u64 vmsa_features; + u16 ghcb_version; /* Highest guest GHCB protocol version allowed */ struct kvm *enc_context_owner; /* Owner of copied encryption context */ + struct list_head mirror_vms; /* List of VMs mirroring */ + struct list_head mirror_entry; /* Use as a list entry of mirrors */ struct misc_cg *misc_cg; /* For misc cgroup accounting */ + atomic_t migration_in_progress; + void *snp_context; /* SNP guest context page */ + void *guest_req_buf; /* Bounce buffer for SNP Guest Request input */ + void *guest_resp_buf; /* Bounce buffer for SNP Guest Request output */ + struct mutex guest_req_mutex; /* Must acquire before using bounce buffers */ + cpumask_var_t have_run_cpus; /* CPUs that have done VMRUN for this VM. */ }; struct kvm_svm { @@ -86,8 +122,8 @@ struct kvm_svm { /* Struct members for AVIC */ u32 avic_vm_id; - struct page *avic_logical_id_table_page; - struct page *avic_physical_id_table_page; + u32 *avic_logical_id_table; + u64 *avic_physical_id_table; struct hlist_node hnode; struct kvm_sev_info sev_info; @@ -102,6 +138,49 @@ struct kvm_vmcb_info { uint64_t asid_generation; }; +struct vmcb_save_area_cached { + u64 efer; + u64 cr4; + u64 cr3; + u64 cr0; + u64 dr7; + u64 dr6; +}; + +struct vmcb_ctrl_area_cached { + u32 intercepts[MAX_INTERCEPT]; + u16 pause_filter_thresh; + u16 pause_filter_count; + u64 iopm_base_pa; + u64 msrpm_base_pa; + u64 tsc_offset; + u32 asid; + u8 tlb_ctl; + u32 int_ctl; + u32 int_vector; + u32 int_state; + u32 exit_code; + u32 exit_code_hi; + u64 exit_info_1; + u64 exit_info_2; + u32 exit_int_info; + u32 exit_int_info_err; + u64 nested_ctl; + u32 event_inj; + u32 event_inj_err; + u64 next_rip; + u64 nested_cr3; + u64 virt_ext; + u32 clean; + u64 bus_lock_rip; + union { +#if IS_ENABLED(CONFIG_HYPERV) || IS_ENABLED(CONFIG_KVM_HYPERV) + struct hv_vmcb_enlightenments hv_enlightenments; +#endif + u8 reserved_sw[32]; + }; +}; + struct svm_nested_state { struct kvm_vmcb_info vmcb02; u64 hsave_msr; @@ -109,17 +188,64 @@ struct svm_nested_state { u64 vmcb12_gpa; u64 last_vmcb12_gpa; - /* These are the merged vectors */ - u32 *msrpm; + /* + * The MSR permissions map used for vmcb02, which is the merge result + * of vmcb01 and vmcb12 + */ + void *msrpm; /* A VMRUN has started but has not yet been performed, so * we cannot inject a nested vmexit yet. */ bool nested_run_pending; /* cache for control fields of the guest */ - struct vmcb_control_area ctl; + struct vmcb_ctrl_area_cached ctl; + + /* + * Note: this struct is not kept up-to-date while L2 runs; it is only + * valid within nested_svm_vmrun. + */ + struct vmcb_save_area_cached save; bool initialized; + + /* + * Indicates whether MSR bitmap for L2 needs to be rebuilt due to + * changes in MSR bitmap for L1 or switching to a different L2. Note, + * this flag can only be used reliably in conjunction with a paravirt L1 + * which informs L0 whether any changes to MSR bitmap for L2 were done + * on its side. + */ + bool force_msr_bitmap_recalc; +}; + +struct vcpu_sev_es_state { + /* SEV-ES support */ + struct sev_es_save_area *vmsa; + struct ghcb *ghcb; + u8 valid_bitmap[16]; + struct kvm_host_map ghcb_map; + bool received_first_sipi; + unsigned int ap_reset_hold_type; + + /* SEV-ES scratch area support */ + u64 sw_scratch; + void *ghcb_sa; + u32 ghcb_sa_len; + bool ghcb_sa_sync; + bool ghcb_sa_free; + + /* SNP Page-State-Change buffer entries currently being processed */ + u16 psc_idx; + u16 psc_inflight; + bool psc_2m; + + u64 ghcb_registered_gpa; + + struct mutex snp_vmsa_mutex; /* Used to handle concurrent updates of VMSA. */ + gpa_t snp_vmsa_gpa; + bool snp_ap_waiting_for_reset; + bool snp_has_guest_vmsa; }; struct vcpu_svm { @@ -128,7 +254,6 @@ struct vcpu_svm { struct vmcb *vmcb; struct kvm_vmcb_info vmcb01; struct kvm_vmcb_info *current_vmcb; - struct svm_cpu_data *svm_data; u32 asid; u32 sysenter_esp_hi; u32 sysenter_eip_hi; @@ -139,6 +264,8 @@ struct vcpu_svm { u64 next_rip; u64 spec_ctrl; + + u64 tsc_ratio_msr; /* * Contains guest-controlled bits of VIRT_SPEC_CTRL, which will be * translated into the appropriate L2_CFG bits on the host to @@ -146,102 +273,129 @@ struct vcpu_svm { */ u64 virt_spec_ctrl; - u32 *msrpm; + void *msrpm; ulong nmi_iret_rip; struct svm_nested_state nested; + /* NMI mask value, used when vNMI is not enabled */ + bool nmi_masked; + + /* + * True when NMIs are still masked but guest IRET was just intercepted + * and KVM is waiting for RIP to change, which will signal that the + * intercepted IRET was retired and thus NMI can be unmasked. + */ + bool awaiting_iret_completion; + + /* + * Set when KVM is awaiting IRET completion and needs to inject NMIs as + * soon as the IRET completes (e.g. NMI is pending injection). KVM + * temporarily steals RFLAGS.TF to single-step the guest in this case + * in order to regain control as soon as the NMI-blocking condition + * goes away. + */ bool nmi_singlestep; u64 nmi_singlestep_guest_rflags; - unsigned int3_injected; - unsigned long int3_rip; + bool nmi_l1_to_l2; - /* cached guest cpuid flags for faster access */ - bool nrips_enabled : 1; + unsigned long soft_int_csbase; + unsigned long soft_int_old_rip; + unsigned long soft_int_next_rip; + bool soft_int_injected; u32 ldr_reg; u32 dfr_reg; - struct page *avic_backing_page; - u64 *avic_physical_id_cache; - bool avic_is_running; + + /* This is essentially a shadow of the vCPU's actual entry in the + * Physical ID table that is programmed into the VMCB, i.e. that is + * seen by the CPU. If IPI virtualization is disabled, IsRunning is + * only ever set in the shadow, i.e. is never propagated to the "real" + * table, so that hardware never sees IsRunning=1. + */ + u64 avic_physical_id_entry; /* - * Per-vcpu list of struct amd_svm_iommu_ir: - * This is used mainly to store interrupt remapping information used - * when update the vcpu affinity. This avoids the need to scan for - * IRTE and try to match ga_tag in the IOMMU driver. + * Per-vCPU list of irqfds that are eligible to post IRQs directly to + * the vCPU (a.k.a. device posted IRQs, a.k.a. IRQ bypass). The list + * is used to reconfigure IRTEs when the vCPU is loaded/put (to set the + * target pCPU), when AVIC is toggled on/off (to (de)activate bypass), + * and if the irqfd becomes ineligible for posting (to put the IRTE + * back into remapped mode). */ struct list_head ir_list; - spinlock_t ir_list_lock; + raw_spinlock_t ir_list_lock; - /* Save desired MSR intercept (read: pass-through) state */ - struct { - DECLARE_BITMAP(read, MAX_DIRECT_ACCESS_MSRS); - DECLARE_BITMAP(write, MAX_DIRECT_ACCESS_MSRS); - } shadow_msr_intercept; + struct vcpu_sev_es_state sev_es; - /* SEV-ES support */ - struct vmcb_save_area *vmsa; - struct ghcb *ghcb; - struct kvm_host_map ghcb_map; - bool received_first_sipi; + bool guest_state_loaded; - /* SEV-ES scratch area support */ - void *ghcb_sa; - u64 ghcb_sa_len; - bool ghcb_sa_sync; - bool ghcb_sa_free; + bool x2avic_msrs_intercepted; + bool lbr_msrs_intercepted; - bool guest_state_loaded; + /* Guest GIF value, used when vGIF is not enabled */ + bool guest_gif; }; struct svm_cpu_data { - int cpu; - u64 asid_generation; u32 max_asid; u32 next_asid; u32 min_asid; - struct kvm_ldttss_desc *tss_desc; - struct page *save_area; - struct vmcb *current_vmcb; + bool bp_spec_reduce_set; + + struct vmcb *save_area; + unsigned long save_area_pa; /* index = sev_asid, value = vmcb pointer */ struct vmcb **sev_vmcbs; }; -DECLARE_PER_CPU(struct svm_cpu_data *, svm_data); +DECLARE_PER_CPU(struct svm_cpu_data, svm_data); void recalc_intercepts(struct vcpu_svm *svm); -static inline struct kvm_svm *to_kvm_svm(struct kvm *kvm) +static __always_inline struct kvm_svm *to_kvm_svm(struct kvm *kvm) { return container_of(kvm, struct kvm_svm, kvm); } -static inline bool sev_guest(struct kvm *kvm) +static __always_inline struct kvm_sev_info *to_kvm_sev_info(struct kvm *kvm) { + return &to_kvm_svm(kvm)->sev_info; +} + #ifdef CONFIG_KVM_AMD_SEV - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; +static __always_inline bool sev_guest(struct kvm *kvm) +{ + return to_kvm_sev_info(kvm)->active; +} +static __always_inline bool sev_es_guest(struct kvm *kvm) +{ + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); - return sev->active; -#else - return false; -#endif + return sev->es_active && !WARN_ON_ONCE(!sev->active); } -static inline bool sev_es_guest(struct kvm *kvm) +static __always_inline bool sev_snp_guest(struct kvm *kvm) { -#ifdef CONFIG_KVM_AMD_SEV - struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info; + struct kvm_sev_info *sev = to_kvm_sev_info(kvm); - return sev_guest(kvm) && sev->es_active; + return (sev->vmsa_features & SVM_SEV_FEAT_SNP_ACTIVE) && + !WARN_ON_ONCE(!sev_es_guest(kvm)); +} #else - return false; +#define sev_guest(kvm) false +#define sev_es_guest(kvm) false +#define sev_snp_guest(kvm) false #endif + +static inline bool ghcb_gpa_is_registered(struct vcpu_svm *svm, u64 val) +{ + return svm->sev_es.ghcb_registered_gpa == val; } static inline void vmcb_mark_all_dirty(struct vmcb *vmcb) @@ -255,11 +409,6 @@ static inline void vmcb_mark_all_clean(struct vmcb *vmcb) & ~VMCB_ALWAYS_DIRTY_MASK; } -static inline bool vmcb_is_clean(struct vmcb *vmcb, int bit) -{ - return (vmcb->control.clean & (1 << bit)); -} - static inline void vmcb_mark_dirty(struct vmcb *vmcb, int bit) { vmcb->control.clean &= ~(1 << bit); @@ -270,11 +419,21 @@ static inline bool vmcb_is_dirty(struct vmcb *vmcb, int bit) return !test_bit(bit, (unsigned long *)&vmcb->control.clean); } -static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu) +static __always_inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu) { return container_of(vcpu, struct vcpu_svm, vcpu); } +/* + * Only the PDPTRs are loaded on demand into the shadow MMU. All other + * fields are synchronized on VM-Exit, because accessing the VMCB is cheap. + * + * CR3 might be out of date in the VMCB but it is not marked dirty; instead, + * KVM_REQ_LOAD_MMU_PGD is always requested when the cached vcpu->arch.cr3 + * is changed. svm_load_mmu_pgd() then syncs the new CR3 value into the VMCB. + */ +#define SVM_REGS_LAZY_LOAD_SET (1 << VCPU_EXREG_PDPTR) + static inline void vmcb_set_intercept(struct vmcb_control_area *control, u32 bit) { WARN_ON_ONCE(bit >= 32 * MAX_INTERCEPT); @@ -293,46 +452,10 @@ static inline bool vmcb_is_intercept(struct vmcb_control_area *control, u32 bit) return test_bit(bit, (unsigned long *)&control->intercepts); } -static inline void set_dr_intercepts(struct vcpu_svm *svm) +static inline bool vmcb12_is_intercept(struct vmcb_ctrl_area_cached *control, u32 bit) { - struct vmcb *vmcb = svm->vmcb01.ptr; - - if (!sev_es_guest(svm->vcpu.kvm)) { - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR0_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR1_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR2_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR3_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR4_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR5_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR6_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR0_WRITE); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR1_WRITE); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR2_WRITE); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR3_WRITE); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR4_WRITE); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR5_WRITE); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR6_WRITE); - } - - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_WRITE); - - recalc_intercepts(svm); -} - -static inline void clr_dr_intercepts(struct vcpu_svm *svm) -{ - struct vmcb *vmcb = svm->vmcb01.ptr; - - vmcb->control.intercepts[INTERCEPT_DR] = 0; - - /* DR7 access must remain intercepted for an SEV-ES guest */ - if (sev_es_guest(svm->vcpu.kvm)) { - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_READ); - vmcb_set_intercept(&vmcb->control, INTERCEPT_DR7_WRITE); - } - - recalc_intercepts(svm); + WARN_ON_ONCE(bit >= 32 * MAX_INTERCEPT); + return test_bit(bit, (unsigned long *)&control->intercepts); } static inline void set_exception_intercept(struct vcpu_svm *svm, u32 bit) @@ -378,49 +501,198 @@ static inline bool svm_is_intercept(struct vcpu_svm *svm, int bit) return vmcb_is_intercept(&svm->vmcb->control, bit); } -static inline bool vgif_enabled(struct vcpu_svm *svm) +static inline bool nested_vgif_enabled(struct vcpu_svm *svm) { - return !!(svm->vmcb->control.int_ctl & V_GIF_ENABLE_MASK); + return guest_cpu_cap_has(&svm->vcpu, X86_FEATURE_VGIF) && + (svm->nested.ctl.int_ctl & V_GIF_ENABLE_MASK); +} + +static inline struct vmcb *get_vgif_vmcb(struct vcpu_svm *svm) +{ + if (!vgif) + return NULL; + + if (is_guest_mode(&svm->vcpu) && !nested_vgif_enabled(svm)) + return svm->nested.vmcb02.ptr; + else + return svm->vmcb01.ptr; } static inline void enable_gif(struct vcpu_svm *svm) { - if (vgif_enabled(svm)) - svm->vmcb->control.int_ctl |= V_GIF_MASK; + struct vmcb *vmcb = get_vgif_vmcb(svm); + + if (vmcb) + vmcb->control.int_ctl |= V_GIF_MASK; else - svm->vcpu.arch.hflags |= HF_GIF_MASK; + svm->guest_gif = true; } static inline void disable_gif(struct vcpu_svm *svm) { - if (vgif_enabled(svm)) - svm->vmcb->control.int_ctl &= ~V_GIF_MASK; + struct vmcb *vmcb = get_vgif_vmcb(svm); + + if (vmcb) + vmcb->control.int_ctl &= ~V_GIF_MASK; else - svm->vcpu.arch.hflags &= ~HF_GIF_MASK; + svm->guest_gif = false; } static inline bool gif_set(struct vcpu_svm *svm) { - if (vgif_enabled(svm)) - return !!(svm->vmcb->control.int_ctl & V_GIF_MASK); + struct vmcb *vmcb = get_vgif_vmcb(svm); + + if (vmcb) + return !!(vmcb->control.int_ctl & V_GIF_MASK); else - return !!(svm->vcpu.arch.hflags & HF_GIF_MASK); + return svm->guest_gif; } -/* svm.c */ -#define MSR_INVALID 0xffffffffU +static inline bool nested_npt_enabled(struct vcpu_svm *svm) +{ + return svm->nested.ctl.nested_ctl & SVM_NESTED_CTL_NP_ENABLE; +} +static inline bool nested_vnmi_enabled(struct vcpu_svm *svm) +{ + return guest_cpu_cap_has(&svm->vcpu, X86_FEATURE_VNMI) && + (svm->nested.ctl.int_ctl & V_NMI_ENABLE_MASK); +} + +static inline bool is_x2apic_msrpm_offset(u32 offset) +{ + /* 4 msrs per u8, and 4 u8 in u32 */ + u32 msr = offset * 16; + + return (msr >= APIC_BASE_MSR) && + (msr < (APIC_BASE_MSR + 0x100)); +} + +static inline struct vmcb *get_vnmi_vmcb_l1(struct vcpu_svm *svm) +{ + if (!vnmi) + return NULL; + + if (is_guest_mode(&svm->vcpu)) + return NULL; + else + return svm->vmcb01.ptr; +} + +static inline bool is_vnmi_enabled(struct vcpu_svm *svm) +{ + struct vmcb *vmcb = get_vnmi_vmcb_l1(svm); + + if (vmcb) + return !!(vmcb->control.int_ctl & V_NMI_ENABLE_MASK); + else + return false; +} + +static inline void svm_vmgexit_set_return_code(struct vcpu_svm *svm, + u64 response, u64 data) +{ + ghcb_set_sw_exit_info_1(svm->sev_es.ghcb, response); + ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, data); +} + +static inline void svm_vmgexit_inject_exception(struct vcpu_svm *svm, u8 vector) +{ + u64 data = SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_EXEPT | vector; + + svm_vmgexit_set_return_code(svm, GHCB_HV_RESP_ISSUE_EXCEPTION, data); +} + +static inline void svm_vmgexit_bad_input(struct vcpu_svm *svm, u64 suberror) +{ + svm_vmgexit_set_return_code(svm, GHCB_HV_RESP_MALFORMED_INPUT, suberror); +} + +static inline void svm_vmgexit_success(struct vcpu_svm *svm, u64 data) +{ + svm_vmgexit_set_return_code(svm, GHCB_HV_RESP_NO_ACTION, data); +} + +static inline void svm_vmgexit_no_action(struct vcpu_svm *svm, u64 data) +{ + svm_vmgexit_set_return_code(svm, GHCB_HV_RESP_NO_ACTION, data); +} + +/* + * The MSRPM is 8KiB in size, divided into four 2KiB ranges (the fourth range + * is reserved). Each MSR within a range is covered by two bits, one each for + * read (bit 0) and write (bit 1), where a bit value of '1' means intercepted. + */ +#define SVM_MSRPM_BYTES_PER_RANGE 2048 +#define SVM_BITS_PER_MSR 2 +#define SVM_MSRS_PER_BYTE (BITS_PER_BYTE / SVM_BITS_PER_MSR) +#define SVM_MSRS_PER_RANGE (SVM_MSRPM_BYTES_PER_RANGE * SVM_MSRS_PER_BYTE) +static_assert(SVM_MSRS_PER_RANGE == 8192); +#define SVM_MSRPM_OFFSET_MASK (SVM_MSRS_PER_RANGE - 1) + +static __always_inline int svm_msrpm_bit_nr(u32 msr) +{ + int range_nr; + + switch (msr & ~SVM_MSRPM_OFFSET_MASK) { + case 0: + range_nr = 0; + break; + case 0xc0000000: + range_nr = 1; + break; + case 0xc0010000: + range_nr = 2; + break; + default: + return -EINVAL; + } + + return range_nr * SVM_MSRPM_BYTES_PER_RANGE * BITS_PER_BYTE + + (msr & SVM_MSRPM_OFFSET_MASK) * SVM_BITS_PER_MSR; +} + +#define __BUILD_SVM_MSR_BITMAP_HELPER(rtype, action, bitop, access, bit_rw) \ +static inline rtype svm_##action##_msr_bitmap_##access(unsigned long *bitmap, \ + u32 msr) \ +{ \ + int bit_nr; \ + \ + bit_nr = svm_msrpm_bit_nr(msr); \ + if (bit_nr < 0) \ + return (rtype)true; \ + \ + return bitop##_bit(bit_nr + bit_rw, bitmap); \ +} + +#define BUILD_SVM_MSR_BITMAP_HELPERS(ret_type, action, bitop) \ + __BUILD_SVM_MSR_BITMAP_HELPER(ret_type, action, bitop, read, 0) \ + __BUILD_SVM_MSR_BITMAP_HELPER(ret_type, action, bitop, write, 1) + +BUILD_SVM_MSR_BITMAP_HELPERS(bool, test, test) +BUILD_SVM_MSR_BITMAP_HELPERS(void, clear, __clear) +BUILD_SVM_MSR_BITMAP_HELPERS(void, set, __set) + +#define DEBUGCTL_RESERVED_BITS (~DEBUGCTLMSR_LBR) + +/* svm.c */ extern bool dump_invalid_vmcb; -u32 svm_msrpm_offset(u32 msr); -u32 *svm_vcpu_alloc_msrpm(void); -void svm_vcpu_init_msrpm(struct kvm_vcpu *vcpu, u32 *msrpm); -void svm_vcpu_free_msrpm(u32 *msrpm); +void *svm_alloc_permissions_map(unsigned long size, gfp_t gfp_mask); + +static inline void *svm_vcpu_alloc_msrpm(void) +{ + return svm_alloc_permissions_map(MSRPM_SIZE, GFP_KERNEL_ACCOUNT); +} + +void svm_vcpu_free_msrpm(void *msrpm); +void svm_copy_lbrs(struct vmcb *to_vmcb, struct vmcb *from_vmcb); +void svm_enable_lbrv(struct kvm_vcpu *vcpu); +void svm_update_lbrv(struct kvm_vcpu *vcpu); int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer); void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4); -void svm_flush_tlb(struct kvm_vcpu *vcpu); void disable_nmi_singlestep(struct vcpu_svm *svm); bool svm_smi_blocked(struct kvm_vcpu *vcpu); bool svm_nmi_blocked(struct kvm_vcpu *vcpu); @@ -429,6 +701,22 @@ void svm_set_gif(struct vcpu_svm *svm, bool value); int svm_invoke_exit_handler(struct kvm_vcpu *vcpu, u64 exit_code); void set_msr_interception(struct kvm_vcpu *vcpu, u32 *msrpm, u32 msr, int read, int write); +void svm_complete_interrupt_delivery(struct kvm_vcpu *vcpu, int delivery_mode, + int trig_mode, int vec); + +void svm_set_intercept_for_msr(struct kvm_vcpu *vcpu, u32 msr, int type, bool set); + +static inline void svm_disable_intercept_for_msr(struct kvm_vcpu *vcpu, + u32 msr, int type) +{ + svm_set_intercept_for_msr(vcpu, msr, type, false); +} + +static inline void svm_enable_intercept_for_msr(struct kvm_vcpu *vcpu, + u32 msr, int type) +{ + svm_set_intercept_for_msr(vcpu, msr, type, true); +} /* nested.c */ @@ -445,25 +733,30 @@ static inline bool nested_svm_virtualize_tpr(struct kvm_vcpu *vcpu) static inline bool nested_exit_on_smi(struct vcpu_svm *svm) { - return vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_SMI); + return vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_SMI); } static inline bool nested_exit_on_intr(struct vcpu_svm *svm) { - return vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_INTR); + return vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_INTR); } static inline bool nested_exit_on_nmi(struct vcpu_svm *svm) { - return vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_NMI); + return vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_NMI); } -int enter_svm_guest_mode(struct kvm_vcpu *vcpu, u64 vmcb_gpa, struct vmcb *vmcb12); -void svm_leave_nested(struct vcpu_svm *svm); +int __init nested_svm_init_msrpm_merge_offsets(void); + +int enter_svm_guest_mode(struct kvm_vcpu *vcpu, + u64 vmcb_gpa, struct vmcb *vmcb12, bool from_vmrun); +void svm_leave_nested(struct kvm_vcpu *vcpu); void svm_free_nested(struct vcpu_svm *svm); int svm_allocate_nested(struct vcpu_svm *svm); int nested_svm_vmrun(struct kvm_vcpu *vcpu); -void nested_svm_vmloadsave(struct vmcb *from_vmcb, struct vmcb *to_vmcb); +void svm_copy_vmrun_state(struct vmcb_save_area *to_save, + struct vmcb_save_area *from_save); +void svm_copy_vmloadsave_state(struct vmcb *to_vmcb, struct vmcb *from_vmcb); int nested_svm_vmexit(struct vcpu_svm *svm); static inline int nested_svm_simple_vmexit(struct vcpu_svm *svm, u32 exit_code) @@ -479,6 +772,12 @@ int nested_svm_check_permissions(struct kvm_vcpu *vcpu); int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, bool has_error_code, u32 error_code); int nested_svm_exit_special(struct vcpu_svm *svm); +void nested_svm_update_tsc_ratio_msr(struct kvm_vcpu *vcpu); +void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu); +void nested_copy_vmcb_control_to_cache(struct vcpu_svm *svm, + struct vmcb_control_area *control); +void nested_copy_vmcb_save_to_cache(struct vcpu_svm *svm, + struct vmcb_save_area *save); void nested_sync_control_from_vmcb02(struct vcpu_svm *svm); void nested_vmcb02_compute_g_pat(struct vcpu_svm *svm); void svm_switch_vmcb(struct vcpu_svm *svm, struct kvm_vmcb_info *target_vmcb); @@ -486,92 +785,163 @@ void svm_switch_vmcb(struct vcpu_svm *svm, struct kvm_vmcb_info *target_vmcb); extern struct kvm_x86_nested_ops svm_nested_ops; /* avic.c */ - -#define AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK (0xFF) -#define AVIC_LOGICAL_ID_ENTRY_VALID_BIT 31 -#define AVIC_LOGICAL_ID_ENTRY_VALID_MASK (1 << 31) - -#define AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK (0xFFULL) -#define AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK (0xFFFFFFFFFFULL << 12) -#define AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK (1ULL << 62) -#define AVIC_PHYSICAL_ID_ENTRY_VALID_MASK (1ULL << 63) - -#define VMCB_AVIC_APIC_BAR_MASK 0xFFFFFFFFFF000ULL - -static inline void avic_update_vapic_bar(struct vcpu_svm *svm, u64 data) -{ - svm->vmcb->control.avic_vapic_bar = data & VMCB_AVIC_APIC_BAR_MASK; - vmcb_mark_dirty(svm->vmcb, VMCB_AVIC); -} - -static inline bool avic_vcpu_is_running(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - u64 *entry = svm->avic_physical_id_cache; - - if (!entry) - return false; - - return (READ_ONCE(*entry) & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK); -} - -int avic_ga_log_notifier(u32 ga_tag); +#define AVIC_REQUIRED_APICV_INHIBITS \ +( \ + BIT(APICV_INHIBIT_REASON_DISABLED) | \ + BIT(APICV_INHIBIT_REASON_ABSENT) | \ + BIT(APICV_INHIBIT_REASON_HYPERV) | \ + BIT(APICV_INHIBIT_REASON_NESTED) | \ + BIT(APICV_INHIBIT_REASON_IRQWIN) | \ + BIT(APICV_INHIBIT_REASON_PIT_REINJ) | \ + BIT(APICV_INHIBIT_REASON_BLOCKIRQ) | \ + BIT(APICV_INHIBIT_REASON_SEV) | \ + BIT(APICV_INHIBIT_REASON_PHYSICAL_ID_ALIASED) | \ + BIT(APICV_INHIBIT_REASON_APIC_ID_MODIFIED) | \ + BIT(APICV_INHIBIT_REASON_APIC_BASE_MODIFIED) | \ + BIT(APICV_INHIBIT_REASON_LOGICAL_ID_ALIASED) | \ + BIT(APICV_INHIBIT_REASON_PHYSICAL_ID_TOO_BIG) \ +) + +bool __init avic_hardware_setup(void); +void avic_hardware_unsetup(void); +int avic_alloc_physical_id_table(struct kvm *kvm); void avic_vm_destroy(struct kvm *kvm); int avic_vm_init(struct kvm *kvm); -void avic_init_vmcb(struct vcpu_svm *svm); -void svm_toggle_avic_for_irq_window(struct kvm_vcpu *vcpu, bool activate); +void avic_init_vmcb(struct vcpu_svm *svm, struct vmcb *vmcb); int avic_incomplete_ipi_interception(struct kvm_vcpu *vcpu); int avic_unaccelerated_access_interception(struct kvm_vcpu *vcpu); int avic_init_vcpu(struct vcpu_svm *svm); void avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu); void avic_vcpu_put(struct kvm_vcpu *vcpu); -void avic_post_state_restore(struct kvm_vcpu *vcpu); -void svm_set_virtual_apic_mode(struct kvm_vcpu *vcpu); -void svm_refresh_apicv_exec_ctrl(struct kvm_vcpu *vcpu); -bool svm_check_apicv_inhibit_reasons(ulong bit); -void svm_pre_update_apicv_exec_ctrl(struct kvm *kvm, bool activate); -void svm_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap); -void svm_hwapic_irr_update(struct kvm_vcpu *vcpu, int max_irr); -void svm_hwapic_isr_update(struct kvm_vcpu *vcpu, int max_isr); -int svm_deliver_avic_intr(struct kvm_vcpu *vcpu, int vec); -bool svm_dy_apicv_has_pending_interrupt(struct kvm_vcpu *vcpu); -int svm_update_pi_irte(struct kvm *kvm, unsigned int host_irq, - uint32_t guest_irq, bool set); -void svm_vcpu_blocking(struct kvm_vcpu *vcpu); -void svm_vcpu_unblocking(struct kvm_vcpu *vcpu); +void avic_apicv_post_state_restore(struct kvm_vcpu *vcpu); +void avic_refresh_apicv_exec_ctrl(struct kvm_vcpu *vcpu); +int avic_pi_update_irte(struct kvm_kernel_irqfd *irqfd, struct kvm *kvm, + unsigned int host_irq, uint32_t guest_irq, + struct kvm_vcpu *vcpu, u32 vector); +void avic_vcpu_blocking(struct kvm_vcpu *vcpu); +void avic_vcpu_unblocking(struct kvm_vcpu *vcpu); +void avic_ring_doorbell(struct kvm_vcpu *vcpu); +unsigned long avic_vcpu_get_apicv_inhibit_reasons(struct kvm_vcpu *vcpu); +void avic_refresh_virtual_apic_mode(struct kvm_vcpu *vcpu); + /* sev.c */ -#define GHCB_VERSION_MAX 1ULL -#define GHCB_VERSION_MIN 1ULL +int pre_sev_run(struct vcpu_svm *svm, int cpu); +void sev_init_vmcb(struct vcpu_svm *svm, bool init_event); +void sev_vcpu_after_set_cpuid(struct vcpu_svm *svm); +int sev_es_string_io(struct vcpu_svm *svm, int size, unsigned int port, int in); +void sev_es_recalc_msr_intercepts(struct kvm_vcpu *vcpu); +void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector); +void sev_es_prepare_switch_to_guest(struct vcpu_svm *svm, struct sev_es_save_area *hostsa); +void sev_es_unmap_ghcb(struct vcpu_svm *svm); +#ifdef CONFIG_KVM_AMD_SEV +int sev_mem_enc_ioctl(struct kvm *kvm, void __user *argp); +int sev_mem_enc_register_region(struct kvm *kvm, + struct kvm_enc_region *range); +int sev_mem_enc_unregister_region(struct kvm *kvm, + struct kvm_enc_region *range); +int sev_vm_copy_enc_context_from(struct kvm *kvm, unsigned int source_fd); +int sev_vm_move_enc_context_from(struct kvm *kvm, unsigned int source_fd); +void sev_guest_memory_reclaimed(struct kvm *kvm); +int sev_handle_vmgexit(struct kvm_vcpu *vcpu); -extern unsigned int max_sev_asid; +/* These symbols are used in common code and are stubbed below. */ +struct page *snp_safe_alloc_page_node(int node, gfp_t gfp); +static inline struct page *snp_safe_alloc_page(void) +{ + return snp_safe_alloc_page_node(numa_node_id(), GFP_KERNEL_ACCOUNT); +} + +int sev_vcpu_create(struct kvm_vcpu *vcpu); +void sev_free_vcpu(struct kvm_vcpu *vcpu); void sev_vm_destroy(struct kvm *kvm); -int svm_mem_enc_op(struct kvm *kvm, void __user *argp); -int svm_register_enc_region(struct kvm *kvm, - struct kvm_enc_region *range); -int svm_unregister_enc_region(struct kvm *kvm, - struct kvm_enc_region *range); -int svm_vm_copy_asid_from(struct kvm *kvm, unsigned int source_fd); -void pre_sev_run(struct vcpu_svm *svm, int cpu); void __init sev_set_cpu_caps(void); void __init sev_hardware_setup(void); -void sev_hardware_teardown(void); +void sev_hardware_unsetup(void); int sev_cpu_init(struct svm_cpu_data *sd); -void sev_free_vcpu(struct kvm_vcpu *vcpu); -int sev_handle_vmgexit(struct kvm_vcpu *vcpu); -int sev_es_string_io(struct vcpu_svm *svm, int size, unsigned int port, int in); -void sev_es_init_vmcb(struct vcpu_svm *svm); -void sev_es_create_vcpu(struct vcpu_svm *svm); -void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector); -void sev_es_prepare_guest_switch(struct vcpu_svm *svm, unsigned int cpu); -void sev_es_unmap_ghcb(struct vcpu_svm *svm); +int sev_dev_get_attr(u32 group, u64 attr, u64 *val); +extern unsigned int max_sev_asid; +void sev_handle_rmp_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u64 error_code); +int sev_gmem_prepare(struct kvm *kvm, kvm_pfn_t pfn, gfn_t gfn, int max_order); +void sev_gmem_invalidate(kvm_pfn_t start, kvm_pfn_t end); +int sev_gmem_max_mapping_level(struct kvm *kvm, kvm_pfn_t pfn, bool is_private); +struct vmcb_save_area *sev_decrypt_vmsa(struct kvm_vcpu *vcpu); +void sev_free_decrypted_vmsa(struct kvm_vcpu *vcpu, struct vmcb_save_area *vmsa); +#else +static inline struct page *snp_safe_alloc_page_node(int node, gfp_t gfp) +{ + return alloc_pages_node(node, gfp | __GFP_ZERO, 0); +} + +static inline struct page *snp_safe_alloc_page(void) +{ + return snp_safe_alloc_page_node(numa_node_id(), GFP_KERNEL_ACCOUNT); +} + +static inline int sev_vcpu_create(struct kvm_vcpu *vcpu) { return 0; } +static inline void sev_free_vcpu(struct kvm_vcpu *vcpu) {} +static inline void sev_vm_destroy(struct kvm *kvm) {} +static inline void __init sev_set_cpu_caps(void) {} +static inline void __init sev_hardware_setup(void) {} +static inline void sev_hardware_unsetup(void) {} +static inline int sev_cpu_init(struct svm_cpu_data *sd) { return 0; } +static inline int sev_dev_get_attr(u32 group, u64 attr, u64 *val) { return -ENXIO; } +#define max_sev_asid 0 +static inline void sev_handle_rmp_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u64 error_code) {} +static inline int sev_gmem_prepare(struct kvm *kvm, kvm_pfn_t pfn, gfn_t gfn, int max_order) +{ + return 0; +} +static inline void sev_gmem_invalidate(kvm_pfn_t start, kvm_pfn_t end) {} +static inline int sev_gmem_max_mapping_level(struct kvm *kvm, kvm_pfn_t pfn, bool is_private) +{ + return 0; +} + +static inline struct vmcb_save_area *sev_decrypt_vmsa(struct kvm_vcpu *vcpu) +{ + return NULL; +} +static inline void sev_free_decrypted_vmsa(struct kvm_vcpu *vcpu, struct vmcb_save_area *vmsa) {} +#endif /* vmenter.S */ -void __svm_sev_es_vcpu_run(unsigned long vmcb_pa); -void __svm_vcpu_run(unsigned long vmcb_pa, unsigned long *regs); +void __svm_sev_es_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted, + struct sev_es_save_area *hostsa); +void __svm_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted); + +#define DEFINE_KVM_GHCB_ACCESSORS(field) \ +static __always_inline u64 kvm_ghcb_get_##field(struct vcpu_svm *svm) \ +{ \ + return READ_ONCE(svm->sev_es.ghcb->save.field); \ +} \ + \ +static __always_inline bool kvm_ghcb_##field##_is_valid(const struct vcpu_svm *svm) \ +{ \ + return test_bit(GHCB_BITMAP_IDX(field), \ + (unsigned long *)&svm->sev_es.valid_bitmap); \ +} \ + \ +static __always_inline u64 kvm_ghcb_get_##field##_if_valid(struct vcpu_svm *svm) \ +{ \ + return kvm_ghcb_##field##_is_valid(svm) ? kvm_ghcb_get_##field(svm) : 0; \ +} + +DEFINE_KVM_GHCB_ACCESSORS(cpl) +DEFINE_KVM_GHCB_ACCESSORS(rax) +DEFINE_KVM_GHCB_ACCESSORS(rcx) +DEFINE_KVM_GHCB_ACCESSORS(rdx) +DEFINE_KVM_GHCB_ACCESSORS(rbx) +DEFINE_KVM_GHCB_ACCESSORS(rsi) +DEFINE_KVM_GHCB_ACCESSORS(sw_exit_code) +DEFINE_KVM_GHCB_ACCESSORS(sw_exit_info_1) +DEFINE_KVM_GHCB_ACCESSORS(sw_exit_info_2) +DEFINE_KVM_GHCB_ACCESSORS(sw_scratch) +DEFINE_KVM_GHCB_ACCESSORS(xcr0) +DEFINE_KVM_GHCB_ACCESSORS(xss) #endif diff --git a/arch/x86/kvm/svm/svm_onhyperv.c b/arch/x86/kvm/svm/svm_onhyperv.c index 98aa981c04ec..a8e78c0e5956 100644 --- a/arch/x86/kvm/svm/svm_onhyperv.c +++ b/arch/x86/kvm/svm/svm_onhyperv.c @@ -2,9 +2,9 @@ /* * KVM L1 hypervisor optimizations on Hyper-V for SVM. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kvm_host.h> -#include "kvm_cache_regs.h" #include <asm/mshyperv.h> @@ -15,27 +15,49 @@ #include "kvm_onhyperv.h" #include "svm_onhyperv.h" -int svm_hv_enable_direct_tlbflush(struct kvm_vcpu *vcpu) +static int svm_hv_enable_l2_tlb_flush(struct kvm_vcpu *vcpu) { - struct hv_enlightenments *hve; - struct hv_partition_assist_pg **p_hv_pa_pg = - &to_kvm_hv(vcpu->kvm)->hv_pa_pg; + struct hv_vmcb_enlightenments *hve; + hpa_t partition_assist_page = hv_get_partition_assist_page(vcpu); - if (!*p_hv_pa_pg) - *p_hv_pa_pg = kzalloc(PAGE_SIZE, GFP_KERNEL); - - if (!*p_hv_pa_pg) + if (partition_assist_page == INVALID_PAGE) return -ENOMEM; - hve = (struct hv_enlightenments *)to_svm(vcpu)->vmcb->control.reserved_sw; + hve = &to_svm(vcpu)->vmcb->control.hv_enlightenments; - hve->partition_assist_page = __pa(*p_hv_pa_pg); + hve->partition_assist_page = partition_assist_page; hve->hv_vm_id = (unsigned long)vcpu->kvm; if (!hve->hv_enlightenments_control.nested_flush_hypercall) { hve->hv_enlightenments_control.nested_flush_hypercall = 1; - vmcb_mark_dirty(to_svm(vcpu)->vmcb, VMCB_HV_NESTED_ENLIGHTENMENTS); + vmcb_mark_dirty(to_svm(vcpu)->vmcb, HV_VMCB_NESTED_ENLIGHTENMENTS); } return 0; } +__init void svm_hv_hardware_setup(void) +{ + if (npt_enabled && + ms_hyperv.nested_features & HV_X64_NESTED_ENLIGHTENED_TLB) { + pr_info(KBUILD_MODNAME ": Hyper-V enlightened NPT TLB flush enabled\n"); + svm_x86_ops.flush_remote_tlbs = hv_flush_remote_tlbs; + svm_x86_ops.flush_remote_tlbs_range = hv_flush_remote_tlbs_range; + } + + if (ms_hyperv.nested_features & HV_X64_NESTED_DIRECT_FLUSH) { + int cpu; + + pr_info(KBUILD_MODNAME ": Hyper-V Direct TLB Flush enabled\n"); + for_each_online_cpu(cpu) { + struct hv_vp_assist_page *vp_ap = + hv_get_vp_assist_page(cpu); + + if (!vp_ap) + continue; + + vp_ap->nested_control.features.directhypercall = 1; + } + svm_x86_ops.enable_l2_tlb_flush = + svm_hv_enable_l2_tlb_flush; + } +} diff --git a/arch/x86/kvm/svm/svm_onhyperv.h b/arch/x86/kvm/svm/svm_onhyperv.h index 9b9a55abc29f..08f14e6f195c 100644 --- a/arch/x86/kvm/svm/svm_onhyperv.h +++ b/arch/x86/kvm/svm/svm_onhyperv.h @@ -6,113 +6,70 @@ #ifndef __ARCH_X86_KVM_SVM_ONHYPERV_H__ #define __ARCH_X86_KVM_SVM_ONHYPERV_H__ -#if IS_ENABLED(CONFIG_HYPERV) #include <asm/mshyperv.h> -#include "hyperv.h" +#if IS_ENABLED(CONFIG_HYPERV) + #include "kvm_onhyperv.h" +#include "svm/hyperv.h" -static struct kvm_x86_ops svm_x86_ops; +__init void svm_hv_hardware_setup(void); -/* - * Hyper-V uses the software reserved 32 bytes in VMCB - * control area to expose SVM enlightenments to guests. - */ -struct hv_enlightenments { - struct __packed hv_enlightenments_control { - u32 nested_flush_hypercall:1; - u32 msr_bitmap:1; - u32 enlightened_npt_tlb: 1; - u32 reserved:29; - } __packed hv_enlightenments_control; - u32 hv_vp_id; - u64 hv_vm_id; - u64 partition_assist_page; - u64 reserved; -} __packed; - -/* - * Hyper-V uses the software reserved clean bit in VMCB - */ -#define VMCB_HV_NESTED_ENLIGHTENMENTS VMCB_SW +static inline bool svm_hv_is_enlightened_tlb_enabled(struct kvm_vcpu *vcpu) +{ + struct hv_vmcb_enlightenments *hve = &to_svm(vcpu)->vmcb->control.hv_enlightenments; -int svm_hv_enable_direct_tlbflush(struct kvm_vcpu *vcpu); + return ms_hyperv.nested_features & HV_X64_NESTED_ENLIGHTENED_TLB && + !!hve->hv_enlightenments_control.enlightened_npt_tlb; +} static inline void svm_hv_init_vmcb(struct vmcb *vmcb) { - struct hv_enlightenments *hve = - (struct hv_enlightenments *)vmcb->control.reserved_sw; + struct hv_vmcb_enlightenments *hve = &vmcb->control.hv_enlightenments; + + BUILD_BUG_ON(sizeof(vmcb->control.hv_enlightenments) != + sizeof(vmcb->control.reserved_sw)); if (npt_enabled && ms_hyperv.nested_features & HV_X64_NESTED_ENLIGHTENED_TLB) hve->hv_enlightenments_control.enlightened_npt_tlb = 1; -} - -static inline void svm_hv_hardware_setup(void) -{ - if (npt_enabled && - ms_hyperv.nested_features & HV_X64_NESTED_ENLIGHTENED_TLB) { - pr_info("kvm: Hyper-V enlightened NPT TLB flush enabled\n"); - svm_x86_ops.tlb_remote_flush = hv_remote_flush_tlb; - svm_x86_ops.tlb_remote_flush_with_range = - hv_remote_flush_tlb_with_range; - } - - if (ms_hyperv.nested_features & HV_X64_NESTED_DIRECT_FLUSH) { - int cpu; - pr_info("kvm: Hyper-V Direct TLB Flush enabled\n"); - for_each_online_cpu(cpu) { - struct hv_vp_assist_page *vp_ap = - hv_get_vp_assist_page(cpu); - - if (!vp_ap) - continue; - - vp_ap->nested_control.features.directhypercall = 1; - } - svm_x86_ops.enable_direct_tlbflush = - svm_hv_enable_direct_tlbflush; - } + if (ms_hyperv.nested_features & HV_X64_NESTED_MSR_BITMAP) + hve->hv_enlightenments_control.msr_bitmap = 1; } static inline void svm_hv_vmcb_dirty_nested_enlightenments( struct kvm_vcpu *vcpu) { struct vmcb *vmcb = to_svm(vcpu)->vmcb; - struct hv_enlightenments *hve = - (struct hv_enlightenments *)vmcb->control.reserved_sw; - - /* - * vmcb can be NULL if called during early vcpu init. - * And its okay not to mark vmcb dirty during vcpu init - * as we mark it dirty unconditionally towards end of vcpu - * init phase. - */ - if (vmcb && vmcb_is_clean(vmcb, VMCB_HV_NESTED_ENLIGHTENMENTS) && - hve->hv_enlightenments_control.msr_bitmap) - vmcb_mark_dirty(vmcb, VMCB_HV_NESTED_ENLIGHTENMENTS); + struct hv_vmcb_enlightenments *hve = &vmcb->control.hv_enlightenments; + + if (hve->hv_enlightenments_control.msr_bitmap) + vmcb_mark_dirty(vmcb, HV_VMCB_NESTED_ENLIGHTENMENTS); } -static inline void svm_hv_update_vp_id(struct vmcb *vmcb, - struct kvm_vcpu *vcpu) +static inline void svm_hv_update_vp_id(struct vmcb *vmcb, struct kvm_vcpu *vcpu) { - struct hv_enlightenments *hve = - (struct hv_enlightenments *)vmcb->control.reserved_sw; + struct hv_vmcb_enlightenments *hve = &vmcb->control.hv_enlightenments; u32 vp_index = kvm_hv_get_vpindex(vcpu); if (hve->hv_vp_id != vp_index) { hve->hv_vp_id = vp_index; - vmcb_mark_dirty(vmcb, VMCB_HV_NESTED_ENLIGHTENMENTS); + vmcb_mark_dirty(vmcb, HV_VMCB_NESTED_ENLIGHTENMENTS); } } #else +static inline bool svm_hv_is_enlightened_tlb_enabled(struct kvm_vcpu *vcpu) +{ + return false; +} + static inline void svm_hv_init_vmcb(struct vmcb *vmcb) { } -static inline void svm_hv_hardware_setup(void) +static inline __init void svm_hv_hardware_setup(void) { } diff --git a/arch/x86/kvm/svm/svm_ops.h b/arch/x86/kvm/svm/svm_ops.h index 8170f2a5a16f..4e725854c63a 100644 --- a/arch/x86/kvm/svm/svm_ops.h +++ b/arch/x86/kvm/svm/svm_ops.h @@ -4,11 +4,11 @@ #include <linux/compiler_types.h> -#include <asm/kvm_host.h> +#include "x86.h" #define svm_asm(insn, clobber...) \ do { \ - asm_volatile_goto("1: " __stringify(insn) "\n\t" \ + asm goto("1: " __stringify(insn) "\n\t" \ _ASM_EXTABLE(1b, %l[fault]) \ ::: clobber : fault); \ return; \ @@ -18,7 +18,7 @@ fault: \ #define svm_asm1(insn, op1, clobber...) \ do { \ - asm_volatile_goto("1: " __stringify(insn) " %0\n\t" \ + asm goto("1: " __stringify(insn) " %0\n\t" \ _ASM_EXTABLE(1b, %l[fault]) \ :: op1 : clobber : fault); \ return; \ @@ -28,7 +28,7 @@ fault: \ #define svm_asm2(insn, op1, op2, clobber...) \ do { \ - asm_volatile_goto("1: " __stringify(insn) " %1, %0\n\t" \ + asm goto("1: " __stringify(insn) " %1, %0\n\t" \ _ASM_EXTABLE(1b, %l[fault]) \ :: op1, op2 : clobber : fault); \ return; \ @@ -56,14 +56,9 @@ static inline void invlpga(unsigned long addr, u32 asid) * VMSAVE, VMLOAD, etc... is still controlled by the effective address size, * hence 'unsigned long' instead of 'hpa_t'. */ -static inline void vmsave(unsigned long pa) +static __always_inline void vmsave(unsigned long pa) { svm_asm1(vmsave, "a" (pa), "memory"); } -static inline void vmload(unsigned long pa) -{ - svm_asm1(vmload, "a" (pa), "memory"); -} - #endif /* __KVM_X86_SVM_OPS_H */ diff --git a/arch/x86/kvm/svm/vmenter.S b/arch/x86/kvm/svm/vmenter.S index 4fa17df123cd..3392bcadfb89 100644 --- a/arch/x86/kvm/svm/vmenter.S +++ b/arch/x86/kvm/svm/vmenter.S @@ -1,41 +1,132 @@ /* SPDX-License-Identifier: GPL-2.0 */ #include <linux/linkage.h> #include <asm/asm.h> +#include <asm/asm-offsets.h> #include <asm/bitsperlong.h> +#include <asm/frame.h> #include <asm/kvm_vcpu_regs.h> #include <asm/nospec-branch.h> +#include "kvm-asm-offsets.h" #define WORD_SIZE (BITS_PER_LONG / 8) /* Intentionally omit RAX as it's context switched by hardware */ -#define VCPU_RCX __VCPU_REGS_RCX * WORD_SIZE -#define VCPU_RDX __VCPU_REGS_RDX * WORD_SIZE -#define VCPU_RBX __VCPU_REGS_RBX * WORD_SIZE +#define VCPU_RCX (SVM_vcpu_arch_regs + __VCPU_REGS_RCX * WORD_SIZE) +#define VCPU_RDX (SVM_vcpu_arch_regs + __VCPU_REGS_RDX * WORD_SIZE) +#define VCPU_RBX (SVM_vcpu_arch_regs + __VCPU_REGS_RBX * WORD_SIZE) /* Intentionally omit RSP as it's context switched by hardware */ -#define VCPU_RBP __VCPU_REGS_RBP * WORD_SIZE -#define VCPU_RSI __VCPU_REGS_RSI * WORD_SIZE -#define VCPU_RDI __VCPU_REGS_RDI * WORD_SIZE +#define VCPU_RBP (SVM_vcpu_arch_regs + __VCPU_REGS_RBP * WORD_SIZE) +#define VCPU_RSI (SVM_vcpu_arch_regs + __VCPU_REGS_RSI * WORD_SIZE) +#define VCPU_RDI (SVM_vcpu_arch_regs + __VCPU_REGS_RDI * WORD_SIZE) #ifdef CONFIG_X86_64 -#define VCPU_R8 __VCPU_REGS_R8 * WORD_SIZE -#define VCPU_R9 __VCPU_REGS_R9 * WORD_SIZE -#define VCPU_R10 __VCPU_REGS_R10 * WORD_SIZE -#define VCPU_R11 __VCPU_REGS_R11 * WORD_SIZE -#define VCPU_R12 __VCPU_REGS_R12 * WORD_SIZE -#define VCPU_R13 __VCPU_REGS_R13 * WORD_SIZE -#define VCPU_R14 __VCPU_REGS_R14 * WORD_SIZE -#define VCPU_R15 __VCPU_REGS_R15 * WORD_SIZE +#define VCPU_R8 (SVM_vcpu_arch_regs + __VCPU_REGS_R8 * WORD_SIZE) +#define VCPU_R9 (SVM_vcpu_arch_regs + __VCPU_REGS_R9 * WORD_SIZE) +#define VCPU_R10 (SVM_vcpu_arch_regs + __VCPU_REGS_R10 * WORD_SIZE) +#define VCPU_R11 (SVM_vcpu_arch_regs + __VCPU_REGS_R11 * WORD_SIZE) +#define VCPU_R12 (SVM_vcpu_arch_regs + __VCPU_REGS_R12 * WORD_SIZE) +#define VCPU_R13 (SVM_vcpu_arch_regs + __VCPU_REGS_R13 * WORD_SIZE) +#define VCPU_R14 (SVM_vcpu_arch_regs + __VCPU_REGS_R14 * WORD_SIZE) +#define VCPU_R15 (SVM_vcpu_arch_regs + __VCPU_REGS_R15 * WORD_SIZE) #endif +#define SVM_vmcb01_pa (SVM_vmcb01 + KVM_VMCB_pa) + .section .noinstr.text, "ax" +.macro RESTORE_GUEST_SPEC_CTRL + /* No need to do anything if SPEC_CTRL is unset or V_SPEC_CTRL is set */ + ALTERNATIVE_2 "", \ + "jmp 800f", X86_FEATURE_MSR_SPEC_CTRL, \ + "", X86_FEATURE_V_SPEC_CTRL +801: +.endm +.macro RESTORE_GUEST_SPEC_CTRL_BODY +800: + /* + * SPEC_CTRL handling: if the guest's SPEC_CTRL value differs from the + * host's, write the MSR. This is kept out-of-line so that the common + * case does not have to jump. + * + * IMPORTANT: To avoid RSB underflow attacks and any other nastiness, + * there must not be any returns or indirect branches between this code + * and vmentry. + */ +#ifdef CONFIG_X86_64 + mov SVM_spec_ctrl(%rdi), %rdx + cmp PER_CPU_VAR(x86_spec_ctrl_current), %rdx + je 801b + movl %edx, %eax + shr $32, %rdx +#else + mov SVM_spec_ctrl(%edi), %eax + mov PER_CPU_VAR(x86_spec_ctrl_current), %ecx + xor %eax, %ecx + mov SVM_spec_ctrl + 4(%edi), %edx + mov PER_CPU_VAR(x86_spec_ctrl_current + 4), %esi + xor %edx, %esi + or %esi, %ecx + je 801b +#endif + mov $MSR_IA32_SPEC_CTRL, %ecx + wrmsr + jmp 801b +.endm + +.macro RESTORE_HOST_SPEC_CTRL + /* No need to do anything if SPEC_CTRL is unset or V_SPEC_CTRL is set */ + ALTERNATIVE_2 "", \ + "jmp 900f", X86_FEATURE_MSR_SPEC_CTRL, \ + "", X86_FEATURE_V_SPEC_CTRL +901: +.endm +.macro RESTORE_HOST_SPEC_CTRL_BODY spec_ctrl_intercepted:req +900: + /* Same for after vmexit. */ + mov $MSR_IA32_SPEC_CTRL, %ecx + + /* + * Load the value that the guest had written into MSR_IA32_SPEC_CTRL, + * if it was not intercepted during guest execution. + */ + cmpb $0, \spec_ctrl_intercepted + jnz 998f + rdmsr + movl %eax, SVM_spec_ctrl(%_ASM_DI) + movl %edx, SVM_spec_ctrl + 4(%_ASM_DI) +998: + /* Now restore the host value of the MSR if different from the guest's. */ +#ifdef CONFIG_X86_64 + mov PER_CPU_VAR(x86_spec_ctrl_current), %rdx + cmp SVM_spec_ctrl(%rdi), %rdx + je 901b + movl %edx, %eax + shr $32, %rdx +#else + mov PER_CPU_VAR(x86_spec_ctrl_current), %eax + mov SVM_spec_ctrl(%edi), %esi + xor %eax, %esi + mov PER_CPU_VAR(x86_spec_ctrl_current + 4), %edx + mov SVM_spec_ctrl + 4(%edi), %edi + xor %edx, %edi + or %edi, %esi + je 901b +#endif + wrmsr + jmp 901b +.endm + +#define SVM_CLEAR_CPU_BUFFERS \ + ALTERNATIVE "", __CLEAR_CPU_BUFFERS, X86_FEATURE_CLEAR_CPU_BUF_VM + /** * __svm_vcpu_run - Run a vCPU via a transition to SVM guest mode - * @vmcb_pa: unsigned long - * @regs: unsigned long * (to guest registers) + * @svm: struct vcpu_svm * + * @spec_ctrl_intercepted: bool */ SYM_FUNC_START(__svm_vcpu_run) push %_ASM_BP + mov %_ASM_SP, %_ASM_BP #ifdef CONFIG_X86_64 push %r15 push %r14 @@ -47,49 +138,70 @@ SYM_FUNC_START(__svm_vcpu_run) #endif push %_ASM_BX - /* Save @regs. */ + /* + * Save variables needed after vmexit on the stack, in inverse + * order compared to when they are needed. + */ + + /* Accessed directly from the stack in RESTORE_HOST_SPEC_CTRL. */ push %_ASM_ARG2 - /* Save @vmcb. */ + /* Needed to restore access to percpu variables. */ + __ASM_SIZE(push) PER_CPU_VAR(svm_data + SD_save_area_pa) + + /* Finally save @svm. */ push %_ASM_ARG1 - /* Move @regs to RAX. */ - mov %_ASM_ARG2, %_ASM_AX +.ifnc _ASM_ARG1, _ASM_DI + /* + * Stash @svm in RDI early. On 32-bit, arguments are in RAX, RCX + * and RDX which are clobbered by RESTORE_GUEST_SPEC_CTRL. + */ + mov %_ASM_ARG1, %_ASM_DI +.endif + + /* Clobbers RAX, RCX, RDX (and ESI on 32-bit), consumes RDI (@svm). */ + RESTORE_GUEST_SPEC_CTRL + + /* + * Use a single vmcb (vmcb01 because it's always valid) for + * context switching guest state via VMLOAD/VMSAVE, that way + * the state doesn't need to be copied between vmcb01 and + * vmcb02 when switching vmcbs for nested virtualization. + */ + mov SVM_vmcb01_pa(%_ASM_DI), %_ASM_AX +1: vmload %_ASM_AX +2: + + /* Get svm->current_vmcb->pa into RAX. */ + mov SVM_current_vmcb(%_ASM_DI), %_ASM_AX + mov KVM_VMCB_pa(%_ASM_AX), %_ASM_AX /* Load guest registers. */ - mov VCPU_RCX(%_ASM_AX), %_ASM_CX - mov VCPU_RDX(%_ASM_AX), %_ASM_DX - mov VCPU_RBX(%_ASM_AX), %_ASM_BX - mov VCPU_RBP(%_ASM_AX), %_ASM_BP - mov VCPU_RSI(%_ASM_AX), %_ASM_SI - mov VCPU_RDI(%_ASM_AX), %_ASM_DI + mov VCPU_RCX(%_ASM_DI), %_ASM_CX + mov VCPU_RDX(%_ASM_DI), %_ASM_DX + mov VCPU_RBX(%_ASM_DI), %_ASM_BX + mov VCPU_RBP(%_ASM_DI), %_ASM_BP + mov VCPU_RSI(%_ASM_DI), %_ASM_SI #ifdef CONFIG_X86_64 - mov VCPU_R8 (%_ASM_AX), %r8 - mov VCPU_R9 (%_ASM_AX), %r9 - mov VCPU_R10(%_ASM_AX), %r10 - mov VCPU_R11(%_ASM_AX), %r11 - mov VCPU_R12(%_ASM_AX), %r12 - mov VCPU_R13(%_ASM_AX), %r13 - mov VCPU_R14(%_ASM_AX), %r14 - mov VCPU_R15(%_ASM_AX), %r15 + mov VCPU_R8 (%_ASM_DI), %r8 + mov VCPU_R9 (%_ASM_DI), %r9 + mov VCPU_R10(%_ASM_DI), %r10 + mov VCPU_R11(%_ASM_DI), %r11 + mov VCPU_R12(%_ASM_DI), %r12 + mov VCPU_R13(%_ASM_DI), %r13 + mov VCPU_R14(%_ASM_DI), %r14 + mov VCPU_R15(%_ASM_DI), %r15 #endif + mov VCPU_RDI(%_ASM_DI), %_ASM_DI - /* "POP" @vmcb to RAX. */ - pop %_ASM_AX + /* Clobbers EFLAGS.ZF */ + SVM_CLEAR_CPU_BUFFERS /* Enter guest mode */ - sti - -1: vmrun %_ASM_AX - -2: cli - -#ifdef CONFIG_RETPOLINE - /* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */ - FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE -#endif - - /* "POP" @regs to RAX. */ +3: vmrun %_ASM_AX +4: + /* Pop @svm to RAX while it's the only available register. */ pop %_ASM_AX /* Save all guest registers. */ @@ -110,6 +222,36 @@ SYM_FUNC_START(__svm_vcpu_run) mov %r15, VCPU_R15(%_ASM_AX) #endif + /* @svm can stay in RDI from now on. */ + mov %_ASM_AX, %_ASM_DI + + mov SVM_vmcb01_pa(%_ASM_DI), %_ASM_AX +5: vmsave %_ASM_AX +6: + + /* Restores GSBASE among other things, allowing access to percpu data. */ + pop %_ASM_AX +7: vmload %_ASM_AX +8: + + /* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */ + FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RSB_VMEXIT + + /* + * Clobbers RAX, RCX, RDX (and ESI, EDI on 32-bit), consumes RDI (@svm) + * and RSP (pointer to @spec_ctrl_intercepted). + */ + RESTORE_HOST_SPEC_CTRL + + /* + * Mitigate RETBleed for AMD/Hygon Zen uarch. RET should be + * untrained as soon as we exit the VM and are back to the + * kernel. This should be done before re-enabling interrupts + * because interrupt handlers won't sanitize 'ret' if the return is + * from the kernel. + */ + UNTRAIN_RET_VM + /* * Clear all general purpose registers except RSP and RAX to prevent * speculative use of the guest's values, even those that are reloaded @@ -136,6 +278,9 @@ SYM_FUNC_START(__svm_vcpu_run) xor %r15d, %r15d #endif + /* "Pop" @spec_ctrl_intercepted. */ + pop %_ASM_BX + pop %_ASM_BX #ifdef CONFIG_X86_64 @@ -148,66 +293,112 @@ SYM_FUNC_START(__svm_vcpu_run) pop %edi #endif pop %_ASM_BP - ret + RET + + RESTORE_GUEST_SPEC_CTRL_BODY + RESTORE_HOST_SPEC_CTRL_BODY (%_ASM_SP) -3: cmpb $0, kvm_rebooting +10: cmpb $0, _ASM_RIP(kvm_rebooting) jne 2b ud2 +30: cmpb $0, _ASM_RIP(kvm_rebooting) + jne 4b + ud2 +50: cmpb $0, _ASM_RIP(kvm_rebooting) + jne 6b + ud2 +70: cmpb $0, _ASM_RIP(kvm_rebooting) + jne 8b + ud2 - _ASM_EXTABLE(1b, 3b) + _ASM_EXTABLE(1b, 10b) + _ASM_EXTABLE(3b, 30b) + _ASM_EXTABLE(5b, 50b) + _ASM_EXTABLE(7b, 70b) SYM_FUNC_END(__svm_vcpu_run) +#ifdef CONFIG_KVM_AMD_SEV + + +#ifdef CONFIG_X86_64 +#define SEV_ES_GPRS_BASE 0x300 +#define SEV_ES_RBX (SEV_ES_GPRS_BASE + __VCPU_REGS_RBX * WORD_SIZE) +#define SEV_ES_RBP (SEV_ES_GPRS_BASE + __VCPU_REGS_RBP * WORD_SIZE) +#define SEV_ES_RSI (SEV_ES_GPRS_BASE + __VCPU_REGS_RSI * WORD_SIZE) +#define SEV_ES_RDI (SEV_ES_GPRS_BASE + __VCPU_REGS_RDI * WORD_SIZE) +#define SEV_ES_R12 (SEV_ES_GPRS_BASE + __VCPU_REGS_R12 * WORD_SIZE) +#define SEV_ES_R13 (SEV_ES_GPRS_BASE + __VCPU_REGS_R13 * WORD_SIZE) +#define SEV_ES_R14 (SEV_ES_GPRS_BASE + __VCPU_REGS_R14 * WORD_SIZE) +#define SEV_ES_R15 (SEV_ES_GPRS_BASE + __VCPU_REGS_R15 * WORD_SIZE) +#endif + /** * __svm_sev_es_vcpu_run - Run a SEV-ES vCPU via a transition to SVM guest mode - * @vmcb_pa: unsigned long + * @svm: struct vcpu_svm * + * @spec_ctrl_intercepted: bool */ SYM_FUNC_START(__svm_sev_es_vcpu_run) - push %_ASM_BP -#ifdef CONFIG_X86_64 - push %r15 - push %r14 - push %r13 - push %r12 -#else - push %edi - push %esi -#endif - push %_ASM_BX + FRAME_BEGIN - /* Move @vmcb to RAX. */ - mov %_ASM_ARG1, %_ASM_AX + /* + * Save non-volatile (callee-saved) registers to the host save area. + * Except for RAX and RSP, all GPRs are restored on #VMEXIT, but not + * saved on VMRUN. + */ + mov %rbp, SEV_ES_RBP (%rdx) + mov %r15, SEV_ES_R15 (%rdx) + mov %r14, SEV_ES_R14 (%rdx) + mov %r13, SEV_ES_R13 (%rdx) + mov %r12, SEV_ES_R12 (%rdx) + mov %rbx, SEV_ES_RBX (%rdx) - /* Enter guest mode */ - sti + /* + * Save volatile registers that hold arguments that are needed after + * #VMEXIT (RDI=@svm and RSI=@spec_ctrl_intercepted). + */ + mov %rdi, SEV_ES_RDI (%rdx) + mov %rsi, SEV_ES_RSI (%rdx) + + /* Clobbers RAX, RCX, and RDX (@hostsa), consumes RDI (@svm). */ + RESTORE_GUEST_SPEC_CTRL -1: vmrun %_ASM_AX + /* Get svm->current_vmcb->pa into RAX. */ + mov SVM_current_vmcb(%rdi), %rax + mov KVM_VMCB_pa(%rax), %rax -2: cli + /* Clobbers EFLAGS.ZF */ + SVM_CLEAR_CPU_BUFFERS -#ifdef CONFIG_RETPOLINE + /* Enter guest mode */ +1: vmrun %rax +2: /* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */ - FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE -#endif + FILL_RETURN_BUFFER %rax, RSB_CLEAR_LOOPS, X86_FEATURE_RSB_VMEXIT - pop %_ASM_BX + /* Clobbers RAX, RCX, RDX, consumes RDI (@svm) and RSI (@spec_ctrl_intercepted). */ + RESTORE_HOST_SPEC_CTRL -#ifdef CONFIG_X86_64 - pop %r12 - pop %r13 - pop %r14 - pop %r15 -#else - pop %esi - pop %edi -#endif - pop %_ASM_BP - ret + /* + * Mitigate RETBleed for AMD/Hygon Zen uarch. RET should be + * untrained as soon as we exit the VM and are back to the + * kernel. This should be done before re-enabling interrupts + * because interrupt handlers won't sanitize RET if the return is + * from the kernel. + */ + UNTRAIN_RET_VM + + FRAME_END + RET + + RESTORE_GUEST_SPEC_CTRL_BODY + RESTORE_HOST_SPEC_CTRL_BODY %sil -3: cmpb $0, kvm_rebooting +3: cmpb $0, kvm_rebooting(%rip) jne 2b ud2 _ASM_EXTABLE(1b, 3b) SYM_FUNC_END(__svm_sev_es_vcpu_run) +#endif /* CONFIG_KVM_AMD_SEV */ |