traps.c

xen虚拟机源代码安装包

    domain_crash(v->domain);
    return 0;
}

static int write_stack_trampoline(
    char *stack, char *stack_bottom, uint16_t cs_seg)
{
    /* movq %rsp, saversp(%rip) */
    stack[0] = 0x48;
    stack[1] = 0x89;
    stack[2] = 0x25;
    *(u32 *)&stack[3] = (stack_bottom - &stack[7]) - 16;

    /* leaq saversp(%rip), %rsp */
    stack[7] = 0x48;
    stack[8] = 0x8d;
    stack[9] = 0x25;
    *(u32 *)&stack[10] = (stack_bottom - &stack[14]) - 16;

    /* pushq %r11 */
    stack[14] = 0x41;
    stack[15] = 0x53;

    /* pushq $<cs_seg> */
    stack[16] = 0x68;
    *(u32 *)&stack[17] = cs_seg;

    /* movq $syscall_enter,%r11 */
    stack[21] = 0x49;
    stack[22] = 0xbb;
    *(void **)&stack[23] = (void *)syscall_enter;

    /* jmpq *%r11 */
    stack[31] = 0x41;
    stack[32] = 0xff;
    stack[33] = 0xe3;

    return 34;
}

void __devinit subarch_percpu_traps_init(void)
{
    char *stack_bottom, *stack;
    int   cpu = smp_processor_id();

    if ( cpu == 0 )
    {
        /* Specify dedicated interrupt stacks for NMI, #DF, and #MC. */
        set_intr_gate(TRAP_double_fault, &double_fault);
        idt_table[TRAP_double_fault].a  |= IST_DF << 32;
        idt_table[TRAP_nmi].a           |= IST_NMI << 32;
        idt_table[TRAP_machine_check].a |= IST_MCE << 32;

        /*
         * The 32-on-64 hypercall entry vector is only accessible from ring 1.
         * Also note that this is a trap gate, not an interrupt gate.
         */
        _set_gate(idt_table+HYPERCALL_VECTOR, 15, 1, &compat_hypercall);

        /* Fast trap for int80 (faster than taking the #GP-fixup path). */
        _set_gate(idt_table+0x80, 15, 3, &int80_direct_trap);
    }

    stack_bottom = (char *)get_stack_bottom();
    stack        = (char *)((unsigned long)stack_bottom & ~(STACK_SIZE - 1));

    /* IST_MAX IST pages + 1 syscall page + 1 guard page + primary stack. */
    BUILD_BUG_ON((IST_MAX + 2) * PAGE_SIZE + PRIMARY_STACK_SIZE > STACK_SIZE);

    /* Machine Check handler has its own per-CPU 4kB stack. */
    init_tss[cpu].ist[IST_MCE] = (unsigned long)&stack[IST_MCE * PAGE_SIZE];

    /* Double-fault handler has its own per-CPU 4kB stack. */
    init_tss[cpu].ist[IST_DF] = (unsigned long)&stack[IST_DF * PAGE_SIZE];

    /* NMI handler has its own per-CPU 4kB stack. */
    init_tss[cpu].ist[IST_NMI] = (unsigned long)&stack[IST_NMI * PAGE_SIZE];

    /* Trampoline for SYSCALL entry from long mode. */
    stack = &stack[IST_MAX * PAGE_SIZE]; /* Skip the IST stacks. */
    wrmsrl(MSR_LSTAR, (unsigned long)stack);
    stack += write_stack_trampoline(stack, stack_bottom, FLAT_KERNEL_CS64);

    if ( boot_cpu_data.x86_vendor == X86_VENDOR_INTEL )
    {
        /* SYSENTER entry. */
        wrmsrl(MSR_IA32_SYSENTER_ESP, (unsigned long)stack_bottom);
        wrmsrl(MSR_IA32_SYSENTER_EIP, (unsigned long)sysenter_entry);
        wrmsr(MSR_IA32_SYSENTER_CS, __HYPERVISOR_CS, 0);
    }

    /* Trampoline for SYSCALL entry from compatibility mode. */
    stack = (char *)L1_CACHE_ALIGN((unsigned long)stack);
    wrmsrl(MSR_CSTAR, (unsigned long)stack);
    stack += write_stack_trampoline(stack, stack_bottom, FLAT_USER_CS32);

    /* Common SYSCALL parameters. */
    wrmsr(MSR_STAR, 0, (FLAT_RING3_CS32<<16) | __HYPERVISOR_CS);
    wrmsr(MSR_SYSCALL_MASK, EF_VM|EF_RF|EF_NT|EF_DF|EF_IE|EF_TF, 0U);
}

void init_int80_direct_trap(struct vcpu *v)
{
    struct trap_info *ti = &v->arch.guest_context.trap_ctxt[0x80];
    struct trap_bounce *tb = &v->arch.int80_bounce;

    tb->flags = TBF_EXCEPTION;
    tb->cs    = ti->cs;
    tb->eip   = ti->address;

    if ( null_trap_bounce(v, tb) )
        tb->flags = 0;
}

static long register_guest_callback(struct callback_register *reg)
{
    long ret = 0;
    struct vcpu *v = current;

    if ( !is_canonical_address(reg->address) )
        return -EINVAL;

    switch ( reg->type )
    {
    case CALLBACKTYPE_event:
        v->arch.guest_context.event_callback_eip    = reg->address;
        break;

    case CALLBACKTYPE_failsafe:
        v->arch.guest_context.failsafe_callback_eip = reg->address;
        if ( reg->flags & CALLBACKF_mask_events )
            set_bit(_VGCF_failsafe_disables_events,
                    &v->arch.guest_context.flags);
        else
            clear_bit(_VGCF_failsafe_disables_events,
                      &v->arch.guest_context.flags);
        break;

    case CALLBACKTYPE_syscall:
        v->arch.guest_context.syscall_callback_eip  = reg->address;
        if ( reg->flags & CALLBACKF_mask_events )
            set_bit(_VGCF_syscall_disables_events,
                    &v->arch.guest_context.flags);
        else
            clear_bit(_VGCF_syscall_disables_events,
                      &v->arch.guest_context.flags);
        break;

    case CALLBACKTYPE_syscall32:
        v->arch.syscall32_callback_eip = reg->address;
        v->arch.syscall32_disables_events =
            !!(reg->flags & CALLBACKF_mask_events);
        break;

    case CALLBACKTYPE_sysenter:
        v->arch.sysenter_callback_eip = reg->address;
        v->arch.sysenter_disables_events =
            !!(reg->flags & CALLBACKF_mask_events);
        break;

    case CALLBACKTYPE_nmi:
        ret = register_guest_nmi_callback(reg->address);
        break;

    default:
        ret = -ENOSYS;
        break;
    }

    return ret;
}

static long unregister_guest_callback(struct callback_unregister *unreg)
{
    long ret;

    switch ( unreg->type )
    {
    case CALLBACKTYPE_event:
    case CALLBACKTYPE_failsafe:
    case CALLBACKTYPE_syscall:
    case CALLBACKTYPE_syscall32:
    case CALLBACKTYPE_sysenter:
        ret = -EINVAL;
        break;

    case CALLBACKTYPE_nmi:
        ret = unregister_guest_nmi_callback();
        break;

    default:
        ret = -ENOSYS;
        break;
    }

    return ret;
}


long do_callback_op(int cmd, XEN_GUEST_HANDLE(const_void) arg)
{
    long ret;

    switch ( cmd )
    {
    case CALLBACKOP_register:
    {
        struct callback_register reg;

        ret = -EFAULT;
        if ( copy_from_guest(&reg, arg, 1) )
            break;

        ret = register_guest_callback(&reg);
    }
    break;

    case CALLBACKOP_unregister:
    {
        struct callback_unregister unreg;

        ret = -EFAULT;
        if ( copy_from_guest(&unreg, arg, 1) )
            break;

        ret = unregister_guest_callback(&unreg);
    }
    break;

    default:
        ret = -ENOSYS;
        break;
    }

    return ret;
}

long do_set_callbacks(unsigned long event_address,
                      unsigned long failsafe_address,
                      unsigned long syscall_address)
{
    struct callback_register event = {
        .type = CALLBACKTYPE_event,
        .address = event_address,
    };
    struct callback_register failsafe = {
        .type = CALLBACKTYPE_failsafe,
        .address = failsafe_address,
    };
    struct callback_register syscall = {
        .type = CALLBACKTYPE_syscall,
        .address = syscall_address,
    };

    register_guest_callback(&event);
    register_guest_callback(&failsafe);
    register_guest_callback(&syscall);

    return 0;
}

static void hypercall_page_initialise_ring3_kernel(void *hypercall_page)
{
    char *p;
    int i;

    /* Fill in all the transfer points with template machine code. */
    for ( i = 0; i < (PAGE_SIZE / 32); i++ )
    {
        p = (char *)(hypercall_page + (i * 32));
        *(u8  *)(p+ 0) = 0x51;    /* push %rcx */
        *(u16 *)(p+ 1) = 0x5341;  /* push %r11 */
        *(u8  *)(p+ 3) = 0xb8;    /* mov  $<i>,%eax */
        *(u32 *)(p+ 4) = i;
        *(u16 *)(p+ 8) = 0x050f;  /* syscall */
        *(u16 *)(p+10) = 0x5b41;  /* pop  %r11 */
        *(u8  *)(p+12) = 0x59;    /* pop  %rcx */
        *(u8  *)(p+13) = 0xc3;    /* ret */
    }

    /*
     * HYPERVISOR_iret is special because it doesn't return and expects a 
     * special stack frame. Guests jump at this transfer point instead of 
     * calling it.
     */
    p = (char *)(hypercall_page + (__HYPERVISOR_iret * 32));
    *(u8  *)(p+ 0) = 0x51;    /* push %rcx */
    *(u16 *)(p+ 1) = 0x5341;  /* push %r11 */
    *(u8  *)(p+ 3) = 0x50;    /* push %rax */
    *(u8  *)(p+ 4) = 0xb8;    /* mov  $__HYPERVISOR_iret,%eax */
    *(u32 *)(p+ 5) = __HYPERVISOR_iret;
    *(u16 *)(p+ 9) = 0x050f;  /* syscall */
}

#include "compat/traps.c"

void hypercall_page_initialise(struct domain *d, void *hypercall_page)
{
    memset(hypercall_page, 0xCC, PAGE_SIZE);
    if ( is_hvm_domain(d) )
        hvm_hypercall_page_initialise(d, hypercall_page);
    else if ( !is_pv_32bit_domain(d) )
        hypercall_page_initialise_ring3_kernel(hypercall_page);
    else
        hypercall_page_initialise_ring1_kernel(hypercall_page);
}

/*
 * Local variables:
 * mode: C
 * c-set-style: "BSD"
 * c-basic-offset: 4
 * tab-width: 4
 * indent-tabs-mode: nil
 * End:
 */