diff options
Diffstat (limited to 'arch/x86/kvm/x86.c')
-rw-r--r-- | arch/x86/kvm/x86.c | 256 |
1 files changed, 197 insertions, 59 deletions
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index bda65690788e..4a6eff166fc6 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -51,6 +51,8 @@ #include <linux/pci.h> #include <linux/timekeeper_internal.h> #include <linux/pvclock_gtod.h> +#include <linux/kvm_irqfd.h> +#include <linux/irqbypass.h> #include <trace/events/kvm.h> #define CREATE_TRACE_POINTS @@ -64,6 +66,7 @@ #include <asm/fpu/internal.h> /* Ugh! */ #include <asm/pvclock.h> #include <asm/div64.h> +#include <asm/irq_remapping.h> #define MAX_IO_MSRS 256 #define KVM_MAX_MCE_BANKS 32 @@ -622,7 +625,9 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) if ((cr0 ^ old_cr0) & update_bits) kvm_mmu_reset_context(vcpu); - if ((cr0 ^ old_cr0) & X86_CR0_CD) + if (((cr0 ^ old_cr0) & X86_CR0_CD) && + kvm_arch_has_noncoherent_dma(vcpu->kvm) && + !kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED)) kvm_zap_gfn_range(vcpu->kvm, 0, ~0ULL); return 0; @@ -789,7 +794,7 @@ int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) { if (cr8 & CR8_RESERVED_BITS) return 1; - if (irqchip_in_kernel(vcpu->kvm)) + if (lapic_in_kernel(vcpu)) kvm_lapic_set_tpr(vcpu, cr8); else vcpu->arch.cr8 = cr8; @@ -799,7 +804,7 @@ EXPORT_SYMBOL_GPL(kvm_set_cr8); unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu) { - if (irqchip_in_kernel(vcpu->kvm)) + if (lapic_in_kernel(vcpu)) return kvm_lapic_get_cr8(vcpu); else return vcpu->arch.cr8; @@ -953,6 +958,9 @@ static u32 emulated_msrs[] = { HV_X64_MSR_TIME_REF_COUNT, HV_X64_MSR_REFERENCE_TSC, HV_X64_MSR_CRASH_P0, HV_X64_MSR_CRASH_P1, HV_X64_MSR_CRASH_P2, HV_X64_MSR_CRASH_P3, HV_X64_MSR_CRASH_P4, HV_X64_MSR_CRASH_CTL, + HV_X64_MSR_RESET, + HV_X64_MSR_VP_INDEX, + HV_X64_MSR_VP_RUNTIME, HV_X64_MSR_APIC_ASSIST_PAGE, MSR_KVM_ASYNC_PF_EN, MSR_KVM_STEAL_TIME, MSR_KVM_PV_EOI_EN, @@ -1898,6 +1906,8 @@ static void accumulate_steal_time(struct kvm_vcpu *vcpu) static void record_steal_time(struct kvm_vcpu *vcpu) { + accumulate_steal_time(vcpu); + if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED)) return; @@ -2048,12 +2058,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) if (!(data & KVM_MSR_ENABLED)) break; - vcpu->arch.st.last_steal = current->sched_info.run_delay; - - preempt_disable(); - accumulate_steal_time(vcpu); - preempt_enable(); - kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); break; @@ -2449,6 +2453,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_ENABLE_CAP_VM: case KVM_CAP_DISABLE_QUIRKS: case KVM_CAP_SET_BOOT_CPU_ID: + case KVM_CAP_SPLIT_IRQCHIP: #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT case KVM_CAP_ASSIGN_DEV_IRQ: case KVM_CAP_PCI_2_3: @@ -2628,7 +2633,6 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) vcpu->cpu = cpu; } - accumulate_steal_time(vcpu); kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); } @@ -2662,12 +2666,24 @@ static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, { if (irq->irq >= KVM_NR_INTERRUPTS) return -EINVAL; - if (irqchip_in_kernel(vcpu->kvm)) + + if (!irqchip_in_kernel(vcpu->kvm)) { + kvm_queue_interrupt(vcpu, irq->irq, false); + kvm_make_request(KVM_REQ_EVENT, vcpu); + return 0; + } + + /* + * With in-kernel LAPIC, we only use this to inject EXTINT, so + * fail for in-kernel 8259. + */ + if (pic_in_kernel(vcpu->kvm)) return -ENXIO; - kvm_queue_interrupt(vcpu, irq->irq, false); - kvm_make_request(KVM_REQ_EVENT, vcpu); + if (vcpu->arch.pending_external_vector != -1) + return -EEXIST; + vcpu->arch.pending_external_vector = irq->irq; return 0; } @@ -3176,7 +3192,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp, struct kvm_vapic_addr va; r = -EINVAL; - if (!irqchip_in_kernel(vcpu->kvm)) + if (!lapic_in_kernel(vcpu)) goto out; r = -EFAULT; if (copy_from_user(&va, argp, sizeof va)) @@ -3425,41 +3441,35 @@ static int kvm_vm_ioctl_set_irqchip(struct kvm *kvm, struct kvm_irqchip *chip) static int kvm_vm_ioctl_get_pit(struct kvm *kvm, struct kvm_pit_state *ps) { - int r = 0; - mutex_lock(&kvm->arch.vpit->pit_state.lock); memcpy(ps, &kvm->arch.vpit->pit_state, sizeof(struct kvm_pit_state)); mutex_unlock(&kvm->arch.vpit->pit_state.lock); - return r; + return 0; } static int kvm_vm_ioctl_set_pit(struct kvm *kvm, struct kvm_pit_state *ps) { - int r = 0; - mutex_lock(&kvm->arch.vpit->pit_state.lock); memcpy(&kvm->arch.vpit->pit_state, ps, sizeof(struct kvm_pit_state)); kvm_pit_load_count(kvm, 0, ps->channels[0].count, 0); mutex_unlock(&kvm->arch.vpit->pit_state.lock); - return r; + return 0; } static int kvm_vm_ioctl_get_pit2(struct kvm *kvm, struct kvm_pit_state2 *ps) { - int r = 0; - mutex_lock(&kvm->arch.vpit->pit_state.lock); memcpy(ps->channels, &kvm->arch.vpit->pit_state.channels, sizeof(ps->channels)); ps->flags = kvm->arch.vpit->pit_state.flags; mutex_unlock(&kvm->arch.vpit->pit_state.lock); memset(&ps->reserved, 0, sizeof(ps->reserved)); - return r; + return 0; } static int kvm_vm_ioctl_set_pit2(struct kvm *kvm, struct kvm_pit_state2 *ps) { - int r = 0, start = 0; + int start = 0; u32 prev_legacy, cur_legacy; mutex_lock(&kvm->arch.vpit->pit_state.lock); prev_legacy = kvm->arch.vpit->pit_state.flags & KVM_PIT_FLAGS_HPET_LEGACY; @@ -3471,7 +3481,7 @@ static int kvm_vm_ioctl_set_pit2(struct kvm *kvm, struct kvm_pit_state2 *ps) kvm->arch.vpit->pit_state.flags = ps->flags; kvm_pit_load_count(kvm, 0, kvm->arch.vpit->pit_state.channels[0].count, start); mutex_unlock(&kvm->arch.vpit->pit_state.lock); - return r; + return 0; } static int kvm_vm_ioctl_reinject(struct kvm *kvm, @@ -3556,6 +3566,28 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, kvm->arch.disabled_quirks = cap->args[0]; r = 0; break; + case KVM_CAP_SPLIT_IRQCHIP: { + mutex_lock(&kvm->lock); + r = -EINVAL; + if (cap->args[0] > MAX_NR_RESERVED_IOAPIC_PINS) + goto split_irqchip_unlock; + r = -EEXIST; + if (irqchip_in_kernel(kvm)) + goto split_irqchip_unlock; + if (atomic_read(&kvm->online_vcpus)) + goto split_irqchip_unlock; + r = kvm_setup_empty_irq_routing(kvm); + if (r) + goto split_irqchip_unlock; + /* Pairs with irqchip_in_kernel. */ + smp_wmb(); + kvm->arch.irqchip_split = true; + kvm->arch.nr_reserved_ioapic_pins = cap->args[0]; + r = 0; +split_irqchip_unlock: + mutex_unlock(&kvm->lock); + break; + } default: r = -EINVAL; break; @@ -3669,7 +3701,7 @@ long kvm_arch_vm_ioctl(struct file *filp, } r = -ENXIO; - if (!irqchip_in_kernel(kvm)) + if (!irqchip_in_kernel(kvm) || irqchip_split(kvm)) goto get_irqchip_out; r = kvm_vm_ioctl_get_irqchip(kvm, chip); if (r) @@ -3693,7 +3725,7 @@ long kvm_arch_vm_ioctl(struct file *filp, } r = -ENXIO; - if (!irqchip_in_kernel(kvm)) + if (!irqchip_in_kernel(kvm) || irqchip_split(kvm)) goto set_irqchip_out; r = kvm_vm_ioctl_set_irqchip(kvm, chip); if (r) @@ -4060,6 +4092,15 @@ static int kvm_read_guest_virt_system(struct x86_emulate_ctxt *ctxt, return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, 0, exception); } +static int kvm_read_guest_phys_system(struct x86_emulate_ctxt *ctxt, + unsigned long addr, void *val, unsigned int bytes) +{ + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + int r = kvm_vcpu_read_guest(vcpu, addr, val, bytes); + + return r < 0 ? X86EMUL_IO_NEEDED : X86EMUL_CONTINUE; +} + int kvm_write_guest_virt_system(struct x86_emulate_ctxt *ctxt, gva_t addr, void *val, unsigned int bytes, @@ -4795,6 +4836,7 @@ static const struct x86_emulate_ops emulate_ops = { .write_gpr = emulator_write_gpr, .read_std = kvm_read_guest_virt_system, .write_std = kvm_write_guest_virt_system, + .read_phys = kvm_read_guest_phys_system, .fetch = kvm_fetch_guest_virt, .read_emulated = emulator_read_emulated, .write_emulated = emulator_write_emulated, @@ -5667,7 +5709,7 @@ void kvm_arch_exit(void) int kvm_vcpu_halt(struct kvm_vcpu *vcpu) { ++vcpu->stat.halt_exits; - if (irqchip_in_kernel(vcpu->kvm)) { + if (lapic_in_kernel(vcpu)) { vcpu->arch.mp_state = KVM_MP_STATE_HALTED; return 1; } else { @@ -5774,9 +5816,15 @@ static int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt) */ static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu) { - return (!irqchip_in_kernel(vcpu->kvm) && !kvm_cpu_has_interrupt(vcpu) && - vcpu->run->request_interrupt_window && - kvm_arch_interrupt_allowed(vcpu)); + if (!vcpu->run->request_interrupt_window || pic_in_kernel(vcpu->kvm)) + return false; + + if (kvm_cpu_has_interrupt(vcpu)) + return false; + + return (irqchip_split(vcpu->kvm) + ? kvm_apic_accept_pic_intr(vcpu) + : kvm_arch_interrupt_allowed(vcpu)); } static void post_kvm_run_save(struct kvm_vcpu *vcpu) @@ -5787,13 +5835,17 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu) kvm_run->flags = is_smm(vcpu) ? KVM_RUN_X86_SMM : 0; kvm_run->cr8 = kvm_get_cr8(vcpu); kvm_run->apic_base = kvm_get_apic_base(vcpu); - if (irqchip_in_kernel(vcpu->kvm)) - kvm_run->ready_for_interrupt_injection = 1; - else + if (!irqchip_in_kernel(vcpu->kvm)) kvm_run->ready_for_interrupt_injection = kvm_arch_interrupt_allowed(vcpu) && !kvm_cpu_has_interrupt(vcpu) && !kvm_event_needs_reinjection(vcpu); + else if (!pic_in_kernel(vcpu->kvm)) + kvm_run->ready_for_interrupt_injection = + kvm_apic_accept_pic_intr(vcpu) && + !kvm_cpu_has_interrupt(vcpu); + else + kvm_run->ready_for_interrupt_injection = 1; } static void update_cr8_intercept(struct kvm_vcpu *vcpu) @@ -6144,18 +6196,18 @@ static void process_smi(struct kvm_vcpu *vcpu) static void vcpu_scan_ioapic(struct kvm_vcpu *vcpu) { - u64 eoi_exit_bitmap[4]; - u32 tmr[8]; - if (!kvm_apic_hw_enabled(vcpu->arch.apic)) return; - memset(eoi_exit_bitmap, 0, 32); - memset(tmr, 0, 32); + memset(vcpu->arch.eoi_exit_bitmap, 0, 256 / 8); - kvm_ioapic_scan_entry(vcpu, eoi_exit_bitmap, tmr); - kvm_x86_ops->load_eoi_exitmap(vcpu, eoi_exit_bitmap); - kvm_apic_update_tmr(vcpu, tmr); + if (irqchip_split(vcpu->kvm)) + kvm_scan_ioapic_routes(vcpu, vcpu->arch.eoi_exit_bitmap); + else { + kvm_x86_ops->sync_pir_to_irr(vcpu); + kvm_ioapic_scan_entry(vcpu, vcpu->arch.eoi_exit_bitmap); + } + kvm_x86_ops->load_eoi_exitmap(vcpu); } static void kvm_vcpu_flush_tlb(struct kvm_vcpu *vcpu) @@ -6168,7 +6220,7 @@ void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu) { struct page *page = NULL; - if (!irqchip_in_kernel(vcpu->kvm)) + if (!lapic_in_kernel(vcpu)) return; if (!kvm_x86_ops->set_apic_access_page_addr) @@ -6206,7 +6258,7 @@ void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, static int vcpu_enter_guest(struct kvm_vcpu *vcpu) { int r; - bool req_int_win = !irqchip_in_kernel(vcpu->kvm) && + bool req_int_win = !lapic_in_kernel(vcpu) && vcpu->run->request_interrupt_window; bool req_immediate_exit = false; @@ -6258,6 +6310,17 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_pmu_handle_event(vcpu); if (kvm_check_request(KVM_REQ_PMI, vcpu)) kvm_pmu_deliver_pmi(vcpu); + if (kvm_check_request(KVM_REQ_IOAPIC_EOI_EXIT, vcpu)) { + BUG_ON(vcpu->arch.pending_ioapic_eoi > 255); + if (test_bit(vcpu->arch.pending_ioapic_eoi, + (void *) vcpu->arch.eoi_exit_bitmap)) { + vcpu->run->exit_reason = KVM_EXIT_IOAPIC_EOI; + vcpu->run->eoi.vector = + vcpu->arch.pending_ioapic_eoi; + r = 0; + goto out; + } + } if (kvm_check_request(KVM_REQ_SCAN_IOAPIC, vcpu)) vcpu_scan_ioapic(vcpu); if (kvm_check_request(KVM_REQ_APIC_PAGE_RELOAD, vcpu)) @@ -6268,6 +6331,26 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) r = 0; goto out; } + if (kvm_check_request(KVM_REQ_HV_RESET, vcpu)) { + vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT; + vcpu->run->system_event.type = KVM_SYSTEM_EVENT_RESET; + r = 0; + goto out; + } + } + + /* + * KVM_REQ_EVENT is not set when posted interrupts are set by + * VT-d hardware, so we have to update RVI unconditionally. + */ + if (kvm_lapic_enabled(vcpu)) { + /* + * Update architecture specific hints for APIC + * virtual interrupt delivery. + */ + if (kvm_x86_ops->hwapic_irr_update) + kvm_x86_ops->hwapic_irr_update(vcpu, + kvm_lapic_find_highest_irr(vcpu)); } if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) { @@ -6286,13 +6369,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_x86_ops->enable_irq_window(vcpu); if (kvm_lapic_enabled(vcpu)) { - /* - * Update architecture specific hints for APIC - * virtual interrupt delivery. - */ - if (kvm_x86_ops->hwapic_irr_update) - kvm_x86_ops->hwapic_irr_update(vcpu, - kvm_lapic_find_highest_irr(vcpu)); update_cr8_intercept(vcpu); kvm_lapic_sync_to_vapic(vcpu); } @@ -6428,10 +6504,15 @@ out: static inline int vcpu_block(struct kvm *kvm, struct kvm_vcpu *vcpu) { - if (!kvm_arch_vcpu_runnable(vcpu)) { + if (!kvm_arch_vcpu_runnable(vcpu) && + (!kvm_x86_ops->pre_block || kvm_x86_ops->pre_block(vcpu) == 0)) { srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx); kvm_vcpu_block(vcpu); vcpu->srcu_idx = srcu_read_lock(&kvm->srcu); + + if (kvm_x86_ops->post_block) + kvm_x86_ops->post_block(vcpu); + if (!kvm_check_request(KVM_REQ_UNHALT, vcpu)) return 1; } @@ -6468,10 +6549,12 @@ static int vcpu_run(struct kvm_vcpu *vcpu) vcpu->srcu_idx = srcu_read_lock(&kvm->srcu); for (;;) { - if (kvm_vcpu_running(vcpu)) + if (kvm_vcpu_running(vcpu)) { r = vcpu_enter_guest(vcpu); - else + } else { r = vcpu_block(kvm, vcpu); + } + if (r <= 0) break; @@ -6480,8 +6563,8 @@ static int vcpu_run(struct kvm_vcpu *vcpu) kvm_inject_pending_timer_irqs(vcpu); if (dm_request_for_irq_injection(vcpu)) { - r = -EINTR; - vcpu->run->exit_reason = KVM_EXIT_INTR; + r = 0; + vcpu->run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN; ++vcpu->stat.request_irq_exits; break; } @@ -6608,7 +6691,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } /* re-sync apic's tpr */ - if (!irqchip_in_kernel(vcpu->kvm)) { + if (!lapic_in_kernel(vcpu)) { if (kvm_set_cr8(vcpu, kvm_run->cr8) != 0) { r = -EINVAL; goto out; @@ -7308,7 +7391,7 @@ bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu) bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu) { - return irqchip_in_kernel(vcpu->kvm) == (vcpu->arch.apic != NULL); + return irqchip_in_kernel(vcpu->kvm) == lapic_in_kernel(vcpu); } struct static_key kvm_no_apic_vcpu __read_mostly; @@ -7377,6 +7460,8 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) kvm_async_pf_hash_reset(vcpu); kvm_pmu_init(vcpu); + vcpu->arch.pending_external_vector = -1; + return 0; fail_free_mce_banks: @@ -7402,7 +7487,7 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) kvm_mmu_destroy(vcpu); srcu_read_unlock(&vcpu->kvm->srcu, idx); free_page((unsigned long)vcpu->arch.pio_data); - if (!irqchip_in_kernel(vcpu->kvm)) + if (!lapic_in_kernel(vcpu)) static_key_slow_dec(&kvm_no_apic_vcpu); } @@ -8029,7 +8114,59 @@ bool kvm_arch_has_noncoherent_dma(struct kvm *kvm) } EXPORT_SYMBOL_GPL(kvm_arch_has_noncoherent_dma); +int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons, + struct irq_bypass_producer *prod) +{ + struct kvm_kernel_irqfd *irqfd = + container_of(cons, struct kvm_kernel_irqfd, consumer); + + if (kvm_x86_ops->update_pi_irte) { + irqfd->producer = prod; + return kvm_x86_ops->update_pi_irte(irqfd->kvm, + prod->irq, irqfd->gsi, 1); + } + + return -EINVAL; +} + +void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons, + struct irq_bypass_producer *prod) +{ + int ret; + struct kvm_kernel_irqfd *irqfd = + container_of(cons, struct kvm_kernel_irqfd, consumer); + + if (!kvm_x86_ops->update_pi_irte) { + WARN_ON(irqfd->producer != NULL); + return; + } + + WARN_ON(irqfd->producer != prod); + irqfd->producer = NULL; + + /* + * When producer of consumer is unregistered, we change back to + * remapped mode, so we can re-use the current implementation + * when the irq is masked/disabed or the consumer side (KVM + * int this case doesn't want to receive the interrupts. + */ + ret = kvm_x86_ops->update_pi_irte(irqfd->kvm, prod->irq, irqfd->gsi, 0); + if (ret) + printk(KERN_INFO "irq bypass consumer (token %p) unregistration" + " fails: %d\n", irqfd->consumer.token, ret); +} + +int kvm_arch_update_irqfd_routing(struct kvm *kvm, unsigned int host_irq, + uint32_t guest_irq, bool set) +{ + if (!kvm_x86_ops->update_pi_irte) + return -EINVAL; + + return kvm_x86_ops->update_pi_irte(kvm, host_irq, guest_irq, set); +} + EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_fast_mmio); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_inj_virq); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_page_fault); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_msr); @@ -8044,3 +8181,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_write_tsc_offset); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ple_window); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_pml_full); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_pi_irte_update); |