ernel 3.10内核源码分析--KVM相关--虚拟机运行

1、基本原理
KVM虚拟机通过字符设备/dev/kvm的ioctl接口创建和运行，相关原理见之前的文章说明。
虚拟机的运行通过/dev/kvm设备ioctl VCPU接口的KVM_RUN指令实现， 在 VM 和 VCPU 创建好并完成初始化后，就可以调度该虚拟机 运行了，通常，一个VCPU对应于一个线程，虚拟机运行的本质为调度该虚拟机相关的VCPU所在线程运行。 虚拟机 (VCPU) 的运行主要任务是要进行上下文切换，上下文主要包括相关寄存器、 APIC 状态、 TLB 等，通常上下文切换的过程如下：
1、     保存当前的上下文。
2、     使用 kvm_vcpu 结构体中的上下文信息，加载到物理 CPU 中。
3、     执行 kvm_x86_ops 中的 run_vcpu 函数，调用硬件相关的指令 ( 如 VMLAUNCH) ，进入虚拟机运行环境中 。
虚拟机运行于qemu-kvm的进程上下文中，从硬件的角度看，虚拟机的运行过程，实质为相关指令的执行过程，虚拟机编译后的也就是相应的CPU指令序列，而虚拟机的指令跟Host机的指令执行过程并没有太多的差别，最关键的差别为“敏感指令”(通常为IO、内存等关键操作)的执行，这也是虚拟化实现的本质所在，当在虚拟机中(Guest模式)执行“敏感指令”时，会触发（由硬件触发）VM-exit，使当前CPU从Guest模式(non-root模式)切换到root模式，当前CPU的控制权随之转交给VMM(Hypervisor，KVM中即Host)，由VMM进行相应的处理，处理完成后再次通过应该硬件指令(如VMLAUNCH)，重新进入到Guest模式，从而进入虚拟机运行环境中继续运行。
本文简单解释及分析在3.10版本内核代码中的相关流程，用户态qemu-kvm部分暂不包括。

2、大致流程：
Qemu-kvm 可以通过 ioctl(KVM_RUN…) 使虚拟机运行，最终进入内核态，由KVM相关内核流程处理，在内核态执行的大致过程如下：
kvm_vcpu_ioctl  -->
    kvm_arch_vcpu_ioctl_run
具体由内核函数 kvm_arch_vcpu_ioctl_run 完成相关工作。主要流程如下：

1、    Sigprocmask()屏蔽信号，防止在此过程中受到信号的干扰。

2、    设置当前VCPU状态为KVM_MP_STATE_UNINITIALIZED

3、    配置APIC和mmio相关信息

4、    将VCPU中保存的上下文信息写入指定位置

5、    然后的工作交由__vcpu_run完成

6、    __vcpu_run最终调用vcpu_enter_guest，该函数实现了进入Guest，并执行Guest OS具体指令的操作。       

7、    vcpu_enter_guest最终调用kvm_x86_ops中的run函数运行。对应于Intel平台，该函数为vmx_vcpu_run(设置Guest CR3和其他寄存器、EPT/影子页表相关设置、汇编代码VMLAUNCH切换到非根模式，执行Guest目标代码)。

8、    Guest代码执行到敏感指令或因其他原因(比如中断/异常)，VM-Exit退出非根模式，返回到vcpu_enter_guest函数继续执行。

9、    vcpu_enter_guest函数中会判断VM-Exit原因，并进行相应处理。

10、处理完成后VM-Entry到Guest重新执行Guest代码，或重新等待下次调度。

3、代码分析
kvm_vcpu_ioctl():

点击(此处)折叠或打开

 /*
   * kvm ioctl VCPU指令的入口，传入的fd为KVM_CREATE_VCPU中返回的fd。
   * 主要针对具体的VCPU进行参数设置。如：相关寄存器的读
   * 写、中断控制等
   */

 static long kvm_vcpu_ioctl(struct file *filp,
              unsigned int ioctl, unsigned long arg)
 {
     struct kvm_vcpu *vcpu = filp->private_data;
     void __user *argp = (void __user *)arg;
     int r;
     struct kvm_fpu *fpu = NULL;
     struct kvm_sregs *kvm_sregs = NULL;

     if (vcpu->kvm->mm != current->mm)
         return -EIO;

 #if defined(CONFIG_S390) || defined(CONFIG_PPC) || defined(CONFIG_MIPS)
     /*
      * Special cases: vcpu ioctls that are asynchronous to vcpu execution,
      * so vcpu_load() would break it.
      */
     if (ioctl == KVM_S390_INTERRUPT || ioctl == KVM_INTERRUPT)
         return kvm_arch_vcpu_ioctl(filp, ioctl, arg);
 #endif

     // KVM虚拟机VCPU数据结构载入物理CPU
     r = vcpu_load(vcpu);
     if (r)
         return r;
     switch (ioctl) {
     /*
      * 运行虚拟机，最终通过执行VMLAUNCH指令进入non root模式，
      * 进入虚拟机运行。当虚拟机内部执行敏感指令时，由硬
      * 件触发VM-exit，返回到root模式
      */
     case KVM_RUN:
         r = -EINVAL;
         // 不能带参数。
         if (arg)
             goto out;
         // 运行VCPU(即运行虚拟机)的入口函数
         r = kvm_arch_vcpu_ioctl_run(vcpu, vcpu->run);
         trace_kvm_userspace_exit(vcpu->run->exit_reason, r);
         break;
 ...

kvm_vcpu_ioctl()-->kvm_arch_vcpu_ioctl_run ()-->__vcpu_run():

点击(此处)折叠或打开

 static int __vcpu_run(struct kvm_vcpu *vcpu)
 {
     int r;
     struct kvm *kvm = vcpu->kvm;

     vcpu->srcu_idx = srcu_read_lock(&kvm->srcu);
     /*设置vcpu->arch.apic->vapic_page*/
     r = vapic_enter(vcpu);
     if (r) {
         srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx);
         return r;
     }

     r = 1;
     while (r > 0) {
         /*检查状态*/
         if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE &&
          !vcpu->arch.apf.halted)
          /* 进入Guest模式，最终通过VMLAUNCH指令实现*/
             r = vcpu_enter_guest(vcpu);
         else {/*什么情况下会走到这里?*/
             srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx);
             /*阻塞VCPU，其实就是schddule()调度出去，但在有特殊情况时(比如有挂起的定时器或信号时)，不进行调度而直接退出*/
             kvm_vcpu_block(vcpu);
             vcpu->srcu_idx = srcu_read_lock(&kvm->srcu);
             if (kvm_check_request(KVM_REQ_UNHALT, vcpu)) {
                 kvm_apic_accept_events(vcpu);
                 switch(vcpu->arch.mp_state) {
                 case KVM_MP_STATE_HALTED:
                     vcpu->arch.pv.pv_unhalted = false;
                     vcpu->arch.mp_state =
                         KVM_MP_STATE_RUNNABLE;
                 case KVM_MP_STATE_RUNNABLE:
                     vcpu->arch.apf.halted = false;
                     break;
                 case KVM_MP_STATE_INIT_RECEIVED:
                     break;
                 default:
                     r = -EINTR;
                     break;
                 }
             }
         }

         if (r <= 0)
             break;

         clear_bit(KVM_REQ_PENDING_TIMER, &vcpu->requests);
         if (kvm_cpu_has_pending_timer(vcpu))
             kvm_inject_pending_timer_irqs(vcpu);

         if (dm_request_for_irq_injection(vcpu)) {
             r = -EINTR;
             vcpu->run->exit_reason = KVM_EXIT_INTR;
             ++vcpu->stat.request_irq_exits;
         }

         kvm_check_async_pf_completion(vcpu);

         if (signal_pending(current)) {
             r = -EINTR;
             vcpu->run->exit_reason = KVM_EXIT_INTR;
             ++vcpu->stat.signal_exits;
         }
         /*这是kvm中的一个调度时机点，即选择新VCPU运行的时机点*/
         if (need_resched()) {
             srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx);
             kvm_resched(vcpu);
             vcpu->srcu_idx = srcu_read_lock(&kvm->srcu);
         }
     }

     srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx);

     vapic_exit(vcpu);

     return r;
 }

kvm_vcpu_ioctl()-->kvm_arch_vcpu_ioctl_run()-->__vcpu_run() -->vcpu_enter_guest():

点击(此处)折叠或打开

 /* 进入Guest模式，最终通过VMLAUNCH指令实现*/
 static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 {
     int r;
     bool req_int_win = !irqchip_in_kernel(vcpu->kvm) &&
         vcpu->run->request_interrupt_window;
     bool req_immediate_exit = false;
     /*进入Guest模式前先处理相关挂起的请求*/
     if (vcpu->requests) {
         /*卸载MMU*/
         if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu))
             kvm_mmu_unload(vcpu);
         /*定时器迁移*/
         if (kvm_check_request(KVM_REQ_MIGRATE_TIMER, vcpu))
             __kvm_migrate_timers(vcpu);
         /*主时钟更新*/
         if (kvm_check_request(KVM_REQ_MASTERCLOCK_UPDATE, vcpu))
             kvm_gen_update_masterclock(vcpu->kvm);
         /*全局时钟更新*/
         if (kvm_check_request(KVM_REQ_GLOBAL_CLOCK_UPDATE, vcpu))
             kvm_gen_kvmclock_update(vcpu);
         /*虚拟机时钟更新*/
         if (kvm_check_request(KVM_REQ_CLOCK_UPDATE, vcpu)) {
             r = kvm_guest_time_update(vcpu);
             if (unlikely(r))
                 goto out;
         }
         /*更新mmu*/
         if (kvm_check_request(KVM_REQ_MMU_SYNC, vcpu))
             kvm_mmu_sync_roots(vcpu);
         /*刷新TLB*/
         if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu))
             kvm_x86_ops->tlb_flush(vcpu);
         if (kvm_check_request(KVM_REQ_REPORT_TPR_ACCESS, vcpu)) {
             vcpu->run->exit_reason = KVM_EXIT_TPR_ACCESS;
             r = 0;
             goto out;
         }
         if (kvm_check_request(KVM_REQ_TRIPLE_FAULT, vcpu)) {
             vcpu->run->exit_reason = KVM_EXIT_SHUTDOWN;
             r = 0;
             goto out;
         }
         if (kvm_check_request(KVM_REQ_DEACTIVATE_FPU, vcpu)) {
             vcpu->fpu_active = 0;
             kvm_x86_ops->fpu_deactivate(vcpu);
         }
         if (kvm_check_request(KVM_REQ_APF_HALT, vcpu)) {
             /* Page is swapped out. Do synthetic halt */
             vcpu->arch.apf.halted = true;
             r = 1;
             goto out;
         }
         if (kvm_check_request(KVM_REQ_STEAL_UPDATE, vcpu))
             record_steal_time(vcpu);
         if (kvm_check_request(KVM_REQ_NMI, vcpu))
             process_nmi(vcpu);
         if (kvm_check_request(KVM_REQ_PMU, vcpu))
             kvm_handle_pmu_event(vcpu);
         if (kvm_check_request(KVM_REQ_PMI, vcpu))
             kvm_deliver_pmi(vcpu);
         if (kvm_check_request(KVM_REQ_SCAN_IOAPIC, vcpu))
             vcpu_scan_ioapic(vcpu);
     }
     // 检查是否有事件请求
     if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) {
         kvm_apic_accept_events(vcpu);
         if (vcpu->arch.mp_state == KVM_MP_STATE_INIT_RECEIVED) {
             r = 1;
             goto out;
         }
         // 注入阻塞的事件，中断，异常和nmi等
         inject_pending_event(vcpu);

         /* enable NMI/IRQ window open exits if needed */
         /*
          * 使能NMI/IRQ window，参见Intel64 System Programming Guide 25.3节
          * 当使能了interrupt-window exiting或NMI-window exiting(由VMCS中相关字段控制)，
          * 表示在刚进入虚拟机后，就会立刻因为有pending或注入的中断导致VM-exit
          */
         if (vcpu->arch.nmi_pending)
             req_immediate_exit =
                 kvm_x86_ops->enable_nmi_window(vcpu) != 0;
         else if (kvm_cpu_has_injectable_intr(vcpu) || req_int_win)
             req_immediate_exit =
                 kvm_x86_ops->enable_irq_window(vcpu) != 0;

         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);
         }
     }
     // 装载MMU，待深入分析
     r = kvm_mmu_reload(vcpu);
     if (unlikely(r)) {
         goto cancel_injection;
     }

     preempt_disable();
     // 进入Guest前期准备，架构相关
     kvm_x86_ops->prepare_guest_switch(vcpu);
     if (vcpu->fpu_active)
         kvm_load_guest_fpu(vcpu);
     kvm_load_guest_xcr0(vcpu);

     vcpu->mode = IN_GUEST_MODE;

     /* We should set ->mode before check ->requests,
      * see the comment in make_all_cpus_request.
      */
     smp_mb();

     local_irq_disable();
     /*
      * 如果VCPU处于EXITING_GUEST_MODE或者vcpu->requests(?)或者需要调度或者
      * 有挂起的信号，则放弃
      */
     if (vcpu->mode == EXITING_GUEST_MODE || vcpu->requests
      || need_resched() || signal_pending(current)) {
         vcpu->mode = OUTSIDE_GUEST_MODE;
         smp_wmb();
         local_irq_enable();
         preempt_enable();
         r = 1;
         goto cancel_injection;
     }

     srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
     // req_immediate_exit在前面使能NMI/IRQ window失败时设置，此时需要立即退出，触发重新调度
     if (req_immediate_exit)
         smp_send_reschedule(vcpu->cpu);
     // 计算虚拟机的enter时间
     kvm_guest_enter();
     // 调试相关
     if (unlikely(vcpu->arch.switch_db_regs)) {
         set_debugreg(0, 7);
         set_debugreg(vcpu->arch.eff_db[0], 0);
         set_debugreg(vcpu->arch.eff_db[1], 1);
         set_debugreg(vcpu->arch.eff_db[2], 2);
         set_debugreg(vcpu->arch.eff_db[3], 3);
     }

     trace_kvm_entry(vcpu->vcpu_id);
     // 调用架构相关的run接口(vmx_vcpu_run)，进入Guest模式
     kvm_x86_ops->run(vcpu);


     // 此处开始，说明已经发生了VM-exit，退出了Guest模式
     /*
      * If the guest has used debug registers, at least dr7
      * will be disabled while returning to the host.
      * If we don't have active breakpoints in the host, we don't
      * care about the messed up debug address registers. But if
      * we have some of them active, restore the old state.
      */
     if (hw_breakpoint_active())
         hw_breakpoint_restore();
     /*记录Guest退出前的TSC时钟*/
     vcpu->arch.last_guest_tsc = kvm_x86_ops->read_l1_tsc(vcpu,
                              native_read_tsc());
     // 设置模式
     vcpu->mode = OUTSIDE_GUEST_MODE;
     smp_wmb();

     /* Interrupt is enabled by handle_external_intr() */
     kvm_x86_ops->handle_external_intr(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();
     // 计算虚拟机的退出时间，其中还开中断了?
     kvm_guest_exit();

     preempt_enable();

     vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);

     /*
      * Profile KVM exit RIPs:
      */
     // Profile(采样计数，用于性能分析和调优)相关
     if (unlikely(prof_on == KVM_PROFILING)) {
         unsigned long rip = kvm_rip_read(vcpu);
         profile_hit(KVM_PROFILING, (void *)rip);
     }

     if (unlikely(vcpu->arch.tsc_always_catchup))
         kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);

     if (vcpu->arch.apic_attention)
         kvm_lapic_sync_from_vapic(vcpu);
     /*
      * 调用vmx_handle_exit()处理虚拟机异常，异常原因及其它关键信息
      * 已经在之前获取。
      */
     r = kvm_x86_ops->handle_exit(vcpu);
     return r;

 cancel_injection:
     kvm_x86_ops->cancel_injection(vcpu);
     if (unlikely(vcpu->arch.apic_attention))
         kvm_lapic_sync_from_vapic(vcpu);
 out:
     return r;
 }

kvm_vcpu_ioctl()-->kvm_arch_vcpu_ioctl_run()-->__vcpu_run()-->vcpu_enter_guest() -->vmx_vcpu_run():

点击(此处)折叠或打开

 /*
   * 运行虚拟机，进入Guest模式，即non root模式
   */
 static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 {
     struct vcpu_vmx *vmx = to_vmx(vcpu);
     unsigned long debugctlmsr;

     /* Record the guest's net vcpu time for enforced NMI injections. */
     // nmi注入?跟nmi_watchdog相关?
     if (unlikely(!cpu_has_virtual_nmis() && vmx->soft_vnmi_blocked))
         vmx->entry_time = ktime_get();

     /* Don't enter VMX if guest state is invalid, let the exit handler
      start emulation until we arrive back to a valid state */
     if (vmx->emulation_required)
         return;

     if (vmx->nested.sync_shadow_vmcs) {
         copy_vmcs12_to_shadow(vmx);
         vmx->nested.sync_shadow_vmcs = false;
     }
     // 写入Guest的RSP寄存器信息至VMCS相关位置中
     if (test_bit(VCPU_REGS_RSP, (unsigned long *)&vcpu->arch.regs_dirty))
         vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]);
     // 写入Guest的RIP寄存器信息至VMCS相关位置中
     if (test_bit(VCPU_REGS_RIP, (unsigned long *)&vcpu->arch.regs_dirty))
         vmcs_writel(GUEST_RIP, vcpu->arch.regs[VCPU_REGS_RIP]);

     /* When single-stepping over STI and MOV SS, we must clear the
      * corresponding interruptibility bits in the guest state. Otherwise
      * vmentry fails as it then expects bit 14 (BS) in pending debug
      * exceptions being set, but that's not correct for the guest debugging
      * case. */
     // 单步调试时，需要禁用Guest中断
     if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
         vmx_set_interrupt_shadow(vcpu, 0);

     atomic_switch_perf_msrs(vmx);
     debugctlmsr = get_debugctlmsr();
     // vmx->__launched用于判断当前VCPU是否已经VMLAUNCH了
     vmx->__launched = vmx->loaded_vmcs->launched;
     // 执行VMLAUNCH指令进入Guest模式，虚拟机开始运行
     asm(
         /* Store host registers */
         /*将相关寄存器压栈*/
         "push %%" _ASM_DX "; push %%" _ASM_BP ";"/*BP压栈*/
         /*为guest的rcx寄存器保留个位置，所以这里压两次栈*/
         "push %%" _ASM_CX " \n\t" /* placeholder for guest rcx */
         "push %%" _ASM_CX " \n\t"
         /*
          * %c表示用来表示使用立即数替换，但不使用立即数的语法，at&t汇编中表示立即数的语法前面有一个$，而用了%c后，就去掉了这个$。
          * 主要是用在间接寻址的情况，这种情况下如果直接使用$立即数的方式的话，会报语法错误。
          * [host_rsp]是后面输入部分定义的tag，使用%tag方式可以直接引用，%0是后面输入输出部分中的第一个操作数，即vmx，这里是间接寻址
          * %c[host_rsp](%0)整体来看就是vmx(以寄存器ecx传入)中的host_rsp成员。
          * 所以，如下语句的整体含义就是比较当前SP寄存器和vmx->host_rsp的值。
          */
         /*如果当前RSP和vmx->rsp相等，那就不用mov了，否则将当前RSP保存到vmx中*/
         "cmp %%" _ASM_SP ", %c[host_rsp](%0) \n\t"
         "je 1f \n\t"
         "mov %%" _ASM_SP ", %c[host_rsp](%0) \n\t"
        /*         * 执行ASM_VMX_VMWRITE_RSP_RDX指令(Writes the contents of a primary source operand (register or memory) to a specified field in a VMCS，即将RSP的值写入vmcs中，field由RDX寄存器指定，         * 而此时的RDX寄存器的内容由后面的约束条件:"d"((unsigned long)HOST_RSP指定为HOST_RSP，所以这句命令的作用为:将rsp的值写vmcs，field是HOST_RSP。)，         * 当出现异常时直接重启，由__ex()实现         */
         __ex(ASM_VMX_VMWRITE_RSP_RDX) "\n\t"
         "1: \n\t"
         /* Reload cr2 if changed */
         /*比较当前CR2寄存器和vmx中保存的CR2寄存器内容，如果不相等，就从vmx中重新CR2内容到当前CR2寄存器中*/
         "mov %c[cr2](%0), %%" _ASM_AX " \n\t"
         "mov %%cr2, %%" _ASM_DX " \n\t"
         "cmp %%" _ASM_AX ", %%" _ASM_DX " \n\t"
         "je 2f \n\t"
         "mov %%" _ASM_AX", %%cr2 \n\t"
         "2: \n\t"
         /* Check if vmlaunch of vmresume is needed */
         /*判断vcpu_vmx->__launched，确认是否需要执行VMLAUNCH*/
         "cmpl $0, %c[launched](%0) \n\t"
         /* Load guest registers. Don't clobber flags. */
         /*加载guest寄存器，其实就是从vmx中加载*/
         "mov %c[rax](%0), %%" _ASM_AX " \n\t"
         "mov %c[rbx](%0), %%" _ASM_BX " \n\t"
         "mov %c[rdx](%0), %%" _ASM_DX " \n\t"
         "mov %c[rsi](%0), %%" _ASM_SI " \n\t"
         "mov %c[rdi](%0), %%" _ASM_DI " \n\t"
         "mov %c[rbp](%0), %%" _ASM_BP " \n\t"
 #ifdef CONFIG_X86_64
         "mov %c[r8](%0), %%r8 \n\t"
         "mov %c[r9](%0), %%r9 \n\t"
         "mov %c[r10](%0), %%r10 \n\t"
         "mov %c[r11](%0), %%r11 \n\t"
         "mov %c[r12](%0), %%r12 \n\t"
         "mov %c[r13](%0), %%r13 \n\t"
         "mov %c[r14](%0), %%r14 \n\t"
         "mov %c[r15](%0), %%r15 \n\t"
 #endif
         "mov %c[rcx](%0), %%" _ASM_CX " \n\t" /* kills %0 (ecx) */

         /* Enter guest mode */
         "jne 1f \n\t"
         /* 执行VMLAUNCH指令，进入Guest模式*/
         __ex(ASM_VMX_VMLAUNCH) "\n\t"
         "jmp 2f \n\t"
         /* 执行VMRESUME指令，从Guest模式恢复到root模式*/
         "1: " __ex(ASM_VMX_VMRESUME) "\n\t"
         "2: "
         /* Save guest registers, load host registers, keep flags */
         "mov %0, %c[wordsize](%%" _ASM_SP ") \n\t"
         "pop %0 \n\t"
         "mov %%" _ASM_AX ", %c[rax](%0) \n\t"
         "mov %%" _ASM_BX ", %c[rbx](%0) \n\t"
         __ASM_SIZE(pop) " %c[rcx](%0) \n\t"
         "mov %%" _ASM_DX ", %c[rdx](%0) \n\t"
         "mov %%" _ASM_SI ", %c[rsi](%0) \n\t"
         "mov %%" _ASM_DI ", %c[rdi](%0) \n\t"
         "mov %%" _ASM_BP ", %c[rbp](%0) \n\t"
 #ifdef CONFIG_X86_64
         "mov %%r8, %c[r8](%0) \n\t"
         "mov %%r9, %c[r9](%0) \n\t"
         "mov %%r10, %c[r10](%0) \n\t"
         "mov %%r11, %c[r11](%0) \n\t"
         "mov %%r12, %c[r12](%0) \n\t"
         "mov %%r13, %c[r13](%0) \n\t"
         "mov %%r14, %c[r14](%0) \n\t"
         "mov %%r15, %c[r15](%0) \n\t"
 #endif
         "mov %%cr2, %%" _ASM_AX " \n\t"
         "mov %%" _ASM_AX ", %c[cr2](%0) \n\t"

         "pop %%" _ASM_BP "; pop %%" _ASM_DX " \n\t"
         "setbe %c[fail](%0) \n\t"
         ".pushsection .rodata \n\t"
         ".global vmx_return \n\t"
         "vmx_return: " _ASM_PTR " 2b \n\t"
         ".popsection"
      : : "c"(vmx), "d"((unsigned long)HOST_RSP),
         [launched]"i"(offsetof(struct vcpu_vmx, __launched)),
         [fail]"i"(offsetof(struct vcpu_vmx, fail)),
         /*[host_rsp]是tag，可以在前面以%[host_rsp]方式引用*/
         [host_rsp]"i"(offsetof(struct vcpu_vmx, host_rsp)),
         [rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])),
         [rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])),
         [rcx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RCX])),
         [rdx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDX])),
         [rsi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RSI])),
         [rdi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDI])),
         [rbp]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBP])),
 #ifdef CONFIG_X86_64
         [r8]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R8])),
         [r9]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R9])),
         [r10]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R10])),
         [r11]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R11])),
         [r12]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R12])),
         [r13]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R13])),
         [r14]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R14])),
         [r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])),
 #endif
         [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2)),
         [wordsize]"i"(sizeof(ulong))
      : "cc", "memory"/*clobber list，cc表示寄存器，memory表示内存*/
 #ifdef CONFIG_X86_64
         , "rax", "rbx", "rdi", "rsi"
         , "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15"
 #else
         , "eax", "ebx", "edi", "esi"
 #endif
      );
     // 运行到这里，说明已经发生了VM-exit，返回到了root模式
     /* MSR_IA32_DEBUGCTLMSR is zeroed on vmexit. Restore it if needed */
     if (debugctlmsr)
         update_debugctlmsr(debugctlmsr);

 #ifndef CONFIG_X86_64
     /*
      * The sysexit path does not restore ds/es, so we must set them to
      * a reasonable value ourselves.
      *
      * We can't defer this to vmx_load_host_state() since that function
      * may be executed in interrupt context, which saves and restore segments
      * around it, nullifying its effect.
      */
     /*重新加载ds/es段寄存器，因为VM-exit不会自动加载他们*/
     loadsegment(ds, __USER_DS);
     loadsegment(es, __USER_DS);
 #endif

     vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP)
                  | (1 << VCPU_EXREG_RFLAGS)
                  | (1 << VCPU_EXREG_CPL)
                  | (1 << VCPU_EXREG_PDPTR)
                  | (1 << VCPU_EXREG_SEGMENTS)
                  | (1 << VCPU_EXREG_CR3));
     vcpu->arch.regs_dirty = 0;
     // 从硬件VMCS中读取中断向量表信息
     vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD);

     vmx->loaded_vmcs->launched = 1;
     // 从硬件VMCS中读取VM-exit原因信息，这些信息是VM-exit过程中由硬件自动写入的
     vmx->exit_reason = vmcs_read32(VM_EXIT_REASON);
     trace_kvm_exit(vmx->exit_reason, vcpu, KVM_ISA_VMX);
     /*处理MCE异常和NMI中断*/
     vmx_complete_atomic_exit(vmx);
     vmx_recover_nmi_blocking(vmx);
     vmx_complete_interrupts(vmx);
 }