diff options
Diffstat (limited to 'arch/x86/kvm/x86.c')
-rw-r--r-- | arch/x86/kvm/x86.c | 178 |
1 files changed, 105 insertions, 73 deletions
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 45608a7da9b3..19f9f9e05c2a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -36,7 +36,8 @@ #include <linux/kvm.h> #include <linux/fs.h> #include <linux/vmalloc.h> -#include <linux/module.h> +#include <linux/export.h> +#include <linux/moduleparam.h> #include <linux/mman.h> #include <linux/highmem.h> #include <linux/iommu.h> @@ -70,7 +71,8 @@ #define MAX_IO_MSRS 256 #define KVM_MAX_MCE_BANKS 32 -#define KVM_MCE_CAP_SUPPORTED (MCG_CTL_P | MCG_SER_P) +u64 __read_mostly kvm_mce_cap_supported = MCG_CTL_P | MCG_SER_P; +EXPORT_SYMBOL_GPL(kvm_mce_cap_supported); #define emul_to_vcpu(ctxt) \ container_of(ctxt, struct kvm_vcpu, arch.emulate_ctxt) @@ -89,8 +91,12 @@ static u64 __read_mostly efer_reserved_bits = ~((u64)EFER_SCE); #define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU +#define KVM_X2APIC_API_VALID_FLAGS (KVM_X2APIC_API_USE_32BIT_IDS | \ + KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK) + static void update_cr8_intercept(struct kvm_vcpu *vcpu); static void process_nmi(struct kvm_vcpu *vcpu); +static void enter_smm(struct kvm_vcpu *vcpu); static void __kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags); struct kvm_x86_ops *kvm_x86_ops __read_mostly; @@ -113,7 +119,8 @@ u8 __read_mostly kvm_tsc_scaling_ratio_frac_bits; EXPORT_SYMBOL_GPL(kvm_tsc_scaling_ratio_frac_bits); u64 __read_mostly kvm_max_tsc_scaling_ratio; EXPORT_SYMBOL_GPL(kvm_max_tsc_scaling_ratio); -static u64 __read_mostly kvm_default_tsc_scaling_ratio; +u64 __read_mostly kvm_default_tsc_scaling_ratio; +EXPORT_SYMBOL_GPL(kvm_default_tsc_scaling_ratio); /* tsc tolerance in parts per million - default to 1/2 of the NTP threshold */ static u32 __read_mostly tsc_tolerance_ppm = 250; @@ -537,7 +544,7 @@ int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3) goto out; } for (i = 0; i < ARRAY_SIZE(pdpte); ++i) { - if (is_present_gpte(pdpte[i]) && + if ((pdpte[i] & PT_PRESENT_MASK) && (pdpte[i] & vcpu->arch.mmu.guest_rsvd_check.rsvd_bits_mask[0][2])) { ret = 0; @@ -982,6 +989,7 @@ static u32 emulated_msrs[] = { MSR_IA32_MISC_ENABLE, MSR_IA32_MCG_STATUS, MSR_IA32_MCG_CTL, + MSR_IA32_MCG_EXT_CTL, MSR_IA32_SMBASE, }; @@ -1161,7 +1169,7 @@ static void kvm_write_wall_clock(struct kvm *kvm, gpa_t wall_clock) int version; int r; struct pvclock_wall_clock wc; - struct timespec boot; + struct timespec64 boot; if (!wall_clock) return; @@ -1184,13 +1192,13 @@ static void kvm_write_wall_clock(struct kvm *kvm, gpa_t wall_clock) * wall clock specified here. guest system time equals host * system time for us, thus we must fill in host boot time here. */ - getboottime(&boot); + getboottime64(&boot); if (kvm->arch.kvmclock_offset) { - struct timespec ts = ns_to_timespec(kvm->arch.kvmclock_offset); - boot = timespec_sub(boot, ts); + struct timespec64 ts = ns_to_timespec64(kvm->arch.kvmclock_offset); + boot = timespec64_sub(boot, ts); } - wc.sec = boot.tv_sec; + wc.sec = (u32)boot.tv_sec; /* overflow in 2106 guest time */ wc.nsec = boot.tv_nsec; wc.version = version; @@ -2615,6 +2623,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_TSC_CONTROL: r = kvm_has_tsc_control; break; + case KVM_CAP_X2APIC_API: + r = KVM_X2APIC_API_VALID_FLAGS; + break; default: r = 0; break; @@ -2677,11 +2688,9 @@ long kvm_arch_dev_ioctl(struct file *filp, break; } case KVM_X86_GET_MCE_CAP_SUPPORTED: { - u64 mce_cap; - - mce_cap = KVM_MCE_CAP_SUPPORTED; r = -EFAULT; - if (copy_to_user(argp, &mce_cap, sizeof mce_cap)) + if (copy_to_user(argp, &kvm_mce_cap_supported, + sizeof(kvm_mce_cap_supported))) goto out; r = 0; break; @@ -2733,6 +2742,11 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) rdtsc() - vcpu->arch.last_host_tsc; if (tsc_delta < 0) mark_tsc_unstable("KVM discovered backwards TSC"); + + if (kvm_lapic_hv_timer_in_use(vcpu) && + kvm_x86_ops->set_hv_timer(vcpu, + kvm_get_lapic_tscdeadline_msr(vcpu))) + kvm_lapic_switch_to_sw_timer(vcpu); if (check_tsc_unstable()) { u64 offset = kvm_compute_tsc_offset(vcpu, vcpu->arch.last_guest_tsc); @@ -2766,15 +2780,17 @@ static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu, if (vcpu->arch.apicv_active) kvm_x86_ops->sync_pir_to_irr(vcpu); - memcpy(s->regs, vcpu->arch.apic->regs, sizeof *s); - - return 0; + return kvm_apic_get_state(vcpu, s); } static int kvm_vcpu_ioctl_set_lapic(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s) { - kvm_apic_post_state_restore(vcpu, s); + int r; + + r = kvm_apic_set_state(vcpu, s); + if (r) + return r; update_cr8_intercept(vcpu); return 0; @@ -2859,7 +2875,7 @@ static int kvm_vcpu_ioctl_x86_setup_mce(struct kvm_vcpu *vcpu, r = -EINVAL; if (!bank_num || bank_num >= KVM_MAX_MCE_BANKS) goto out; - if (mcg_cap & ~(KVM_MCE_CAP_SUPPORTED | 0xff | 0xff0000)) + if (mcg_cap & ~(kvm_mce_cap_supported | 0xff | 0xff0000)) goto out; r = 0; vcpu->arch.mcg_cap = mcg_cap; @@ -2869,6 +2885,9 @@ static int kvm_vcpu_ioctl_x86_setup_mce(struct kvm_vcpu *vcpu, /* Init IA32_MCi_CTL to all 1s */ for (bank = 0; bank < bank_num; bank++) vcpu->arch.mce_banks[bank*4] = ~(u64)0; + + if (kvm_x86_ops->setup_mce) + kvm_x86_ops->setup_mce(vcpu); out: return r; } @@ -3767,7 +3786,7 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, r = -EEXIST; if (irqchip_in_kernel(kvm)) goto split_irqchip_unlock; - if (atomic_read(&kvm->online_vcpus)) + if (kvm->created_vcpus) goto split_irqchip_unlock; r = kvm_setup_empty_irq_routing(kvm); if (r) @@ -3781,6 +3800,18 @@ split_irqchip_unlock: mutex_unlock(&kvm->lock); break; } + case KVM_CAP_X2APIC_API: + r = -EINVAL; + if (cap->args[0] & ~KVM_X2APIC_API_VALID_FLAGS) + break; + + if (cap->args[0] & KVM_X2APIC_API_USE_32BIT_IDS) + kvm->arch.x2apic_format = true; + if (cap->args[0] & KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK) + kvm->arch.x2apic_broadcast_quirk_disabled = true; + + r = 0; + break; default: r = -EINVAL; break; @@ -3832,7 +3863,7 @@ long kvm_arch_vm_ioctl(struct file *filp, if (kvm->arch.vpic) goto create_irqchip_unlock; r = -EINVAL; - if (atomic_read(&kvm->online_vcpus)) + if (kvm->created_vcpus) goto create_irqchip_unlock; r = -ENOMEM; vpic = kvm_create_pic(kvm); @@ -3872,7 +3903,7 @@ long kvm_arch_vm_ioctl(struct file *filp, sizeof(struct kvm_pit_config))) goto out; create_pit: - mutex_lock(&kvm->slots_lock); + mutex_lock(&kvm->lock); r = -EEXIST; if (kvm->arch.vpit) goto create_pit_unlock; @@ -3881,7 +3912,7 @@ long kvm_arch_vm_ioctl(struct file *filp, if (kvm->arch.vpit) r = 0; create_pit_unlock: - mutex_unlock(&kvm->slots_lock); + mutex_unlock(&kvm->lock); break; case KVM_GET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ @@ -3988,7 +4019,7 @@ long kvm_arch_vm_ioctl(struct file *filp, case KVM_SET_BOOT_CPU_ID: r = 0; mutex_lock(&kvm->lock); - if (atomic_read(&kvm->online_vcpus) != 0) + if (kvm->created_vcpus) r = -EBUSY; else kvm->arch.bsp_vcpu_id = arg; @@ -5296,13 +5327,8 @@ static void kvm_smm_changed(struct kvm_vcpu *vcpu) /* This is a good place to trace that we are exiting SMM. */ trace_kvm_enter_smm(vcpu->vcpu_id, vcpu->arch.smbase, false); - if (unlikely(vcpu->arch.smi_pending)) { - kvm_make_request(KVM_REQ_SMI, vcpu); - vcpu->arch.smi_pending = 0; - } else { - /* Process a latched INIT, if any. */ - kvm_make_request(KVM_REQ_EVENT, vcpu); - } + /* Process a latched INIT or SMI, if any. */ + kvm_make_request(KVM_REQ_EVENT, vcpu); } kvm_mmu_reset_context(vcpu); @@ -5848,8 +5874,8 @@ int kvm_arch_init(void *opaque) kvm_x86_ops = ops; kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK, - PT_DIRTY_MASK, PT64_NX_MASK, 0); - + PT_DIRTY_MASK, PT64_NX_MASK, 0, + PT_PRESENT_MASK); kvm_timer_init(); perf_register_guest_info_callbacks(&kvm_guest_cbs); @@ -6083,7 +6109,10 @@ static int inject_pending_event(struct kvm_vcpu *vcpu, bool req_int_win) } /* try to inject new event if pending */ - if (vcpu->arch.nmi_pending && kvm_x86_ops->nmi_allowed(vcpu)) { + if (vcpu->arch.smi_pending && !is_smm(vcpu)) { + vcpu->arch.smi_pending = false; + enter_smm(vcpu); + } else if (vcpu->arch.nmi_pending && kvm_x86_ops->nmi_allowed(vcpu)) { --vcpu->arch.nmi_pending; vcpu->arch.nmi_injected = true; kvm_x86_ops->set_nmi(vcpu); @@ -6106,6 +6135,7 @@ static int inject_pending_event(struct kvm_vcpu *vcpu, bool req_int_win) kvm_x86_ops->set_irq(vcpu); } } + return 0; } @@ -6129,7 +6159,7 @@ static void process_nmi(struct kvm_vcpu *vcpu) #define put_smstate(type, buf, offset, val) \ *(type *)((buf) + (offset) - 0x7e00) = val -static u32 process_smi_get_segment_flags(struct kvm_segment *seg) +static u32 enter_smm_get_segment_flags(struct kvm_segment *seg) { u32 flags = 0; flags |= seg->g << 23; @@ -6143,7 +6173,7 @@ static u32 process_smi_get_segment_flags(struct kvm_segment *seg) return flags; } -static void process_smi_save_seg_32(struct kvm_vcpu *vcpu, char *buf, int n) +static void enter_smm_save_seg_32(struct kvm_vcpu *vcpu, char *buf, int n) { struct kvm_segment seg; int offset; @@ -6158,11 +6188,11 @@ static void process_smi_save_seg_32(struct kvm_vcpu *vcpu, char *buf, int n) put_smstate(u32, buf, offset + 8, seg.base); put_smstate(u32, buf, offset + 4, seg.limit); - put_smstate(u32, buf, offset, process_smi_get_segment_flags(&seg)); + put_smstate(u32, buf, offset, enter_smm_get_segment_flags(&seg)); } #ifdef CONFIG_X86_64 -static void process_smi_save_seg_64(struct kvm_vcpu *vcpu, char *buf, int n) +static void enter_smm_save_seg_64(struct kvm_vcpu *vcpu, char *buf, int n) { struct kvm_segment seg; int offset; @@ -6171,7 +6201,7 @@ static void process_smi_save_seg_64(struct kvm_vcpu *vcpu, char *buf, int n) kvm_get_segment(vcpu, &seg, n); offset = 0x7e00 + n * 16; - flags = process_smi_get_segment_flags(&seg) >> 8; + flags = enter_smm_get_segment_flags(&seg) >> 8; put_smstate(u16, buf, offset, seg.selector); put_smstate(u16, buf, offset + 2, flags); put_smstate(u32, buf, offset + 4, seg.limit); @@ -6179,7 +6209,7 @@ static void process_smi_save_seg_64(struct kvm_vcpu *vcpu, char *buf, int n) } #endif -static void process_smi_save_state_32(struct kvm_vcpu *vcpu, char *buf) +static void enter_smm_save_state_32(struct kvm_vcpu *vcpu, char *buf) { struct desc_ptr dt; struct kvm_segment seg; @@ -6203,13 +6233,13 @@ static void process_smi_save_state_32(struct kvm_vcpu *vcpu, char *buf) put_smstate(u32, buf, 0x7fc4, seg.selector); put_smstate(u32, buf, 0x7f64, seg.base); put_smstate(u32, buf, 0x7f60, seg.limit); - put_smstate(u32, buf, 0x7f5c, process_smi_get_segment_flags(&seg)); + put_smstate(u32, buf, 0x7f5c, enter_smm_get_segment_flags(&seg)); kvm_get_segment(vcpu, &seg, VCPU_SREG_LDTR); put_smstate(u32, buf, 0x7fc0, seg.selector); put_smstate(u32, buf, 0x7f80, seg.base); put_smstate(u32, buf, 0x7f7c, seg.limit); - put_smstate(u32, buf, 0x7f78, process_smi_get_segment_flags(&seg)); + put_smstate(u32, buf, 0x7f78, enter_smm_get_segment_flags(&seg)); kvm_x86_ops->get_gdt(vcpu, &dt); put_smstate(u32, buf, 0x7f74, dt.address); @@ -6220,7 +6250,7 @@ static void process_smi_save_state_32(struct kvm_vcpu *vcpu, char *buf) put_smstate(u32, buf, 0x7f54, dt.size); for (i = 0; i < 6; i++) - process_smi_save_seg_32(vcpu, buf, i); + enter_smm_save_seg_32(vcpu, buf, i); put_smstate(u32, buf, 0x7f14, kvm_read_cr4(vcpu)); @@ -6229,7 +6259,7 @@ static void process_smi_save_state_32(struct kvm_vcpu *vcpu, char *buf) put_smstate(u32, buf, 0x7ef8, vcpu->arch.smbase); } -static void process_smi_save_state_64(struct kvm_vcpu *vcpu, char *buf) +static void enter_smm_save_state_64(struct kvm_vcpu *vcpu, char *buf) { #ifdef CONFIG_X86_64 struct desc_ptr dt; @@ -6261,7 +6291,7 @@ static void process_smi_save_state_64(struct kvm_vcpu *vcpu, char *buf) kvm_get_segment(vcpu, &seg, VCPU_SREG_TR); put_smstate(u16, buf, 0x7e90, seg.selector); - put_smstate(u16, buf, 0x7e92, process_smi_get_segment_flags(&seg) >> 8); + put_smstate(u16, buf, 0x7e92, enter_smm_get_segment_flags(&seg) >> 8); put_smstate(u32, buf, 0x7e94, seg.limit); put_smstate(u64, buf, 0x7e98, seg.base); @@ -6271,7 +6301,7 @@ static void process_smi_save_state_64(struct kvm_vcpu *vcpu, char *buf) kvm_get_segment(vcpu, &seg, VCPU_SREG_LDTR); put_smstate(u16, buf, 0x7e70, seg.selector); - put_smstate(u16, buf, 0x7e72, process_smi_get_segment_flags(&seg) >> 8); + put_smstate(u16, buf, 0x7e72, enter_smm_get_segment_flags(&seg) >> 8); put_smstate(u32, buf, 0x7e74, seg.limit); put_smstate(u64, buf, 0x7e78, seg.base); @@ -6280,31 +6310,26 @@ static void process_smi_save_state_64(struct kvm_vcpu *vcpu, char *buf) put_smstate(u64, buf, 0x7e68, dt.address); for (i = 0; i < 6; i++) - process_smi_save_seg_64(vcpu, buf, i); + enter_smm_save_seg_64(vcpu, buf, i); #else WARN_ON_ONCE(1); #endif } -static void process_smi(struct kvm_vcpu *vcpu) +static void enter_smm(struct kvm_vcpu *vcpu) { struct kvm_segment cs, ds; struct desc_ptr dt; char buf[512]; u32 cr0; - if (is_smm(vcpu)) { - vcpu->arch.smi_pending = true; - return; - } - trace_kvm_enter_smm(vcpu->vcpu_id, vcpu->arch.smbase, true); vcpu->arch.hflags |= HF_SMM_MASK; memset(buf, 0, 512); if (guest_cpuid_has_longmode(vcpu)) - process_smi_save_state_64(vcpu, buf); + enter_smm_save_state_64(vcpu, buf); else - process_smi_save_state_32(vcpu, buf); + enter_smm_save_state_32(vcpu, buf); kvm_vcpu_write_guest(vcpu, vcpu->arch.smbase + 0xfe00, buf, sizeof(buf)); @@ -6360,6 +6385,12 @@ static void process_smi(struct kvm_vcpu *vcpu) kvm_mmu_reset_context(vcpu); } +static void process_smi(struct kvm_vcpu *vcpu) +{ + vcpu->arch.smi_pending = true; + kvm_make_request(KVM_REQ_EVENT, vcpu); +} + void kvm_make_scan_ioapic_request(struct kvm *kvm) { kvm_make_all_cpus_request(kvm, KVM_REQ_SCAN_IOAPIC); @@ -6554,8 +6585,18 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) if (inject_pending_event(vcpu, req_int_win) != 0) req_immediate_exit = true; - /* enable NMI/IRQ window open exits if needed */ else { + /* Enable NMI/IRQ window open exits if needed. + * + * SMIs have two cases: 1) they can be nested, and + * then there is nothing to do here because RSM will + * cause a vmexit anyway; 2) or the SMI can be pending + * because inject_pending_event has completed the + * injection of an IRQ or NMI from the previous vmexit, + * and then we request an immediate exit to inject the SMI. + */ + if (vcpu->arch.smi_pending && !is_smm(vcpu)) + req_immediate_exit = true; if (vcpu->arch.nmi_pending) kvm_x86_ops->enable_nmi_window(vcpu); if (kvm_cpu_has_injectable_intr(vcpu) || req_int_win) @@ -6606,12 +6647,14 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_load_guest_xcr0(vcpu); - if (req_immediate_exit) + if (req_immediate_exit) { + kvm_make_request(KVM_REQ_EVENT, vcpu); smp_send_reschedule(vcpu->cpu); + } trace_kvm_entry(vcpu->vcpu_id); wait_lapic_expire(vcpu); - __kvm_guest_enter(); + guest_enter_irqoff(); if (unlikely(vcpu->arch.switch_db_regs)) { set_debugreg(0, 7); @@ -6662,16 +6705,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) ++vcpu->stat.exits; - /* - * We must have an instruction between local_irq_enable() and - * kvm_guest_exit(), so the timer interrupt isn't delayed by - * the interrupt shadow. The stat.exits increment will do nicely. - * But we need to prevent reordering, hence this barrier(): - */ - barrier(); - - kvm_guest_exit(); + guest_exit_irqoff(); + local_irq_enable(); preempt_enable(); vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); @@ -7408,6 +7444,7 @@ void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) { vcpu->arch.hflags = 0; + vcpu->arch.smi_pending = 0; atomic_set(&vcpu->arch.nmi_queued, 0); vcpu->arch.nmi_pending = 0; vcpu->arch.nmi_injected = false; @@ -7600,11 +7637,6 @@ bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu) return (vcpu->arch.apic_base & MSR_IA32_APICBASE_BSP) != 0; } -bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu) -{ - return irqchip_in_kernel(vcpu->kvm) == lapic_in_kernel(vcpu); -} - struct static_key kvm_no_apic_vcpu __read_mostly; EXPORT_SYMBOL_GPL(kvm_no_apic_vcpu); @@ -7871,7 +7903,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm) kfree(kvm->arch.vpic); kfree(kvm->arch.vioapic); kvm_free_vcpus(kvm); - kfree(rcu_dereference_check(kvm->arch.apic_map, 1)); + kvfree(rcu_dereference_check(kvm->arch.apic_map, 1)); kvm_mmu_uninit_vm(kvm); } @@ -8379,7 +8411,7 @@ void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons, /* * 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 + * when the irq is masked/disabled 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); |