Linux0.11内核源码解析-exec.c

主要实现对二进制可执行文件和shell文件的加载和执行，其中主要的函数是do_execve(),它是系统中断调用int 0x80的功能号__NR_execve()调用，是exec()函数的主要实现以下几点功能：

1.执行对参数和环境参数空间页面的初始化操作，初始化空间页面指针数组，根据执行文件名取执行对象的i节点，计算参数个数和环境变量个数，检查文件类型、执行权限

2.根据执行文件开始部分的头数据结构，对其中信息进行处理，根据被执行文件i节点读取文件头部信息，若是shell以#！开始，则分析shell程序名和其参数；根据幻数和段长度等信息判断是否可执行

3.对当前调用进程进行运行新文件前初始化操作，指向新执行文件的i节点，复位信号处理句柄，根据头结构信息设置局部描述符地址和段长；设置参数和环境参数页面指针；修改进程各执行字段内容

4.替换堆栈上原调用execve()程序的返回地址为新执行程序运行地址，运行新加载的程序

 create_tables函数用于根据给定的当前堆栈指针值p以及参数个数argc和环境变量个数envc，在新的程序堆栈中创建环境和参数变量指针表，并返回此时的堆栈指针值sp

 create_tables

用戶内存中解析环境变量和参数字符串，创建指针表，并将地址放到堆栈上，返回栈指针

/*
 * MAX_ARG_PAGES defines the number of pages allocated for arguments
 * and envelope for the new program. 32 should suffice, this gives
 * a maximum env+arg of 128kB !
 */
#define MAX_ARG_PAGES 32

/*
 * create_tables() parses the env- and arg-strings in new user
 * memory and creates the pointer tables from them, and puts their
 * addresses on the "stack", returning the new stack pointer value.
 */
static unsigned long * create_tables(char * p,int argc,int envc)
{
	unsigned long *argv,*envp;
	unsigned long * sp;
    //堆栈指针是以4字节为边界寻址
	sp = (unsigned long *) (0xfffffffc & (unsigned long) p);
    //sp向下移动，空出环境参数占用空间个数，让环境参数指针envp指向该处
	sp -= envc+1;
	envp = sp;
    //sp向下移动，空出环境参数占用的空间个数，并让环境参数指针envp指向该处
	sp -= argc+1;
	argv = sp;

    //argc、argv、envp
	put_fs_long((unsigned long)envp,--sp);
	put_fs_long((unsigned long)argv,--sp);
	put_fs_long((unsigned long)argc,--sp);

    //argc
	while (argc-->0) {
		put_fs_long((unsigned long) p,argv++);
		while (get_fs_byte(p++)) /* nothing */ ;
	}
    //最后放置NULL
	put_fs_long(0,argv);

    //envc
	while (envc-->0) {
		put_fs_long((unsigned long) p,envp++);
		while (get_fs_byte(p++)) /* nothing */ ;
	}
    //最后放置NULL
	put_fs_long(0,envp);
	return sp;
}

count

计算参数个数

/*
 * count() counts the number of arguments/envelopes
 */
static int count(char ** argv)
{
	int i=0;
	char ** tmp;
    
	if ((tmp = argv))
        //判断为NULL就退出
		while (get_fs_long((unsigned long *) (tmp++)))
			i++;

	return i;
}

copy_string

函数从用户内存空间拷贝参数和环境字符串到内核空闲页面内存中，argc

/*
 * 'copy_string()' copies argument/envelope strings from user
 * memory to free pages in kernel mem. These are in a format ready
 * to be put directly into the top of new user memory.
 *
 * Modified by TYT, 11/24/91 to add the from_kmem argument, which specifies
 * whether the string and the string array are from user or kernel segments:
 * 
 * from_kmem     argv *        argv **
 *    0          user space    user space
 *    1          kernel space  user space
 *    2          kernel space  kernel space
 * 
 * We do this by playing games with the fs segment register.  Since it
 * it is expensive to load a segment register, we try to avoid calling
 * set_fs() unless we absolutely have to.
 */
static unsigned long copy_strings(int argc,char ** argv,unsigned long *page,
		unsigned long p, int from_kmem)
{
	char *tmp, *pag=NULL;
	int len, offset = 0;
	unsigned long old_fs, new_fs;

	if (!p)
		return 0;	/* bullet-proofing */
	new_fs = get_ds();
	old_fs = get_fs();
	if (from_kmem==2)
		set_fs(new_fs);
	while (argc-- > 0) {
		if (from_kmem == 1)
			set_fs(new_fs);
		if (!(tmp = (char *)get_fs_long(((unsigned long *)argv)+argc)))
			panic("argc is wrong");
		if (from_kmem == 1)
			set_fs(old_fs);
		len=0;		/* remember zero-padding */
        //计算字符串的长度
		do {
			len++;
		} while (get_fs_byte(tmp++));
        //字符串长度判断
		if (p-len < 0) {	/* this shouldn't happen - 128kB */
			set_fs(old_fs);
			return 0;
		}
		while (len) {
			--p; --tmp; --len;
            //首次复制字符串，offset复制为0
			if (--offset < 0) {
                //p指针在页内的偏移
				offset = p % PAGE_SIZE;
                //恢复fs段寄存器
				if (from_kmem==2)
					set_fs(old_fs);
                //如果pag不为NULL，说明页面在之前的拷贝过程得到分配
                //pag指针为NULL，说明该页面还没有分配内存空间
				if (!(pag = (char *) page[p/PAGE_SIZE]) &&
				    !(pag = (char *) (page[p/PAGE_SIZE] =
				      (unsigned long *) get_free_page()))) 
					return 0;
				if (from_kmem==2)
					set_fs(new_fs);

			}
			*(pag + offset) = get_fs_byte(tmp);
		}
	}
    //还原fs
	if (from_kmem==2)
		set_fs(old_fs);
	return p;
}

修改局部描述符表中的描述符基址和段限长，并将参数和环境页面放置在数据段末端

参数:text_size 执行文件头部中a_text字段给出的代码段长度值，page参数和环境空间页面指针数组

static unsigned long change_ldt(unsigned long text_size,unsigned long * page)
{
	unsigned long code_limit,data_limit,code_base,data_base;
	int i;
//根据执行文件头部a_text的值，计算以页面长度为边界代码段限长，并设置数据段长度为64MB
	code_limit = text_size+PAGE_SIZE -1;
	code_limit &= 0xFFFFF000;
	data_limit = 0x4000000;
//取当前进程中局部描述符代码段描述符中代码段基地址，代码段基址和数据段基址相同
	code_base = get_base(current->ldt[1]);
	data_base = code_base;
//重新设置局部表中代码段和数据描述符的基址和段限长
	set_base(current->ldt[1],code_base);
	set_limit(current->ldt[1],code_limit);
	set_base(current->ldt[2],data_base);
	set_limit(current->ldt[2],data_limit);
/* make sure fs points to the NEW data segment */
	__asm__("pushl $0x17\n\tpop %%fs"::);
	data_base += data_limit;
	for (i=MAX_ARG_PAGES-1 ; i>=0 ; i--) {
		data_base -= PAGE_SIZE;
		if (page[i])
			put_page(page[i],data_base);
	}
	return data_limit;
}

execve系统中断调用函数，加载并执行子进程

函数系统中断调用int 0x80功能号__NR_execve，

参数：eip指向堆栈中调用系统中断的程序代码指针eip，

tmp：_sys_execve时返回地址

filename 被执行程序文件名

argv命令行参数指针数组

envp环境变量指针数组

/*
 * 'do_execve()' executes a new program.
 */
int do_execve(unsigned long * eip,long tmp,char * filename,
	char ** argv, char ** envp)
{
	struct m_inode * inode;
	struct buffer_head * bh;
	struct exec ex;
	unsigned long page[MAX_ARG_PAGES];
	int i,argc,envc;
	int e_uid, e_gid;
	int retval;
	int sh_bang = 0;
	unsigned long p=PAGE_SIZE*MAX_ARG_PAGES-4;

	if ((0xffff & eip[1]) != 0x000f)
		panic("execve called from supervisor mode");
	for (i=0 ; ii_mode)) {	/* must be regular file */
		retval = -EACCES;
		goto exec_error2;
	}
	i = inode->i_mode;
	e_uid = (i & S_ISUID) ? inode->i_uid : current->euid;
	e_gid = (i & S_ISGID) ? inode->i_gid : current->egid;
	if (current->euid == inode->i_uid)
		i >>= 6;
	else if (current->egid == inode->i_gid)
		i >>= 3;
	if (!(i & 1) &&
	    !((inode->i_mode & 0111) && suser())) {
		retval = -ENOEXEC;
		goto exec_error2;
	}
	if (!(bh = bread(inode->i_dev,inode->i_zone[0]))) {
		retval = -EACCES;
		goto exec_error2;
	}
	ex = *((struct exec *) bh->b_data);	/* read exec-header */
	if ((bh->b_data[0] == '#') && (bh->b_data[1] == '!') && (!sh_bang)) {
		/*
		 * This section does the #! interpretation.
		 * Sorta complicated, but hopefully it will work.  -TYT
		 */

		char buf[1023], *cp, *interp, *i_name, *i_arg;
		unsigned long old_fs;

		strncpy(buf, bh->b_data+2, 1022);
		brelse(bh);
		iput(inode);
		buf[1022] = '\0';
		if ((cp = strchr(buf, '\n'))) {
			*cp = '\0';
			for (cp = buf; (*cp == ' ') || (*cp == '\t'); cp++);
		}
		if (!cp || *cp == '\0') {
			retval = -ENOEXEC; /* No interpreter name found */
			goto exec_error1;
		}
		interp = i_name = cp;
		i_arg = 0;
		for ( ; *cp && (*cp != ' ') && (*cp != '\t'); cp++) {
 			if (*cp == '/')
				i_name = cp+1;
		}
		if (*cp) {
			*cp++ = '\0';
			i_arg = cp;
		}
		/*
		 * OK, we've parsed out the interpreter name and
		 * (optional) argument.
		 */
		if (sh_bang++ == 0) {
			p = copy_strings(envc, envp, page, p, 0);
			p = copy_strings(--argc, argv+1, page, p, 0);
		}
		/*
		 * Splice in (1) the interpreter's name for argv[0]
		 *           (2) (optional) argument to interpreter
		 *           (3) filename of shell script
		 *
		 * This is done in reverse order, because of how the
		 * user environment and arguments are stored.
		 */
		p = copy_strings(1, &filename, page, p, 1);
		argc++;
		if (i_arg) {
			p = copy_strings(1, &i_arg, page, p, 2);
			argc++;
		}
		p = copy_strings(1, &i_name, page, p, 2);
		argc++;
		if (!p) {
			retval = -ENOMEM;
			goto exec_error1;
		}
		/*
		 * OK, now restart the process with the interpreter's inode.
		 */
		old_fs = get_fs();
		set_fs(get_ds());
		if (!(inode=namei(interp))) { /* get executables inode */
			set_fs(old_fs);
			retval = -ENOENT;
			goto exec_error1;
		}
		set_fs(old_fs);
		goto restart_interp;
	}
	brelse(bh);
	if (N_MAGIC(ex) != ZMAGIC || ex.a_trsize || ex.a_drsize ||
		ex.a_text+ex.a_data+ex.a_bss>0x3000000 ||
		inode->i_size < ex.a_text+ex.a_data+ex.a_syms+N_TXTOFF(ex)) {
		retval = -ENOEXEC;
		goto exec_error2;
	}
	if (N_TXTOFF(ex) != BLOCK_SIZE) {
		printk("%s: N_TXTOFF != BLOCK_SIZE. See a.out.h.", filename);
		retval = -ENOEXEC;
		goto exec_error2;
	}
	if (!sh_bang) {
		p = copy_strings(envc,envp,page,p,0);
		p = copy_strings(argc,argv,page,p,0);
		if (!p) {
			retval = -ENOMEM;
			goto exec_error2;
		}
	}
/* OK, This is the point of no return */
	if (current->executable)
		iput(current->executable);
	current->executable = inode;
	for (i=0 ; i<32 ; i++)
		current->sigaction[i].sa_handler = NULL;
	for (i=0 ; iclose_on_exec>>i)&1)
			sys_close(i);
	current->close_on_exec = 0;
	free_page_tables(get_base(current->ldt[1]),get_limit(0x0f));
	free_page_tables(get_base(current->ldt[2]),get_limit(0x17));
	if (last_task_used_math == current)
		last_task_used_math = NULL;
	current->used_math = 0;
	p += change_ldt(ex.a_text,page)-MAX_ARG_PAGES*PAGE_SIZE;
	p = (unsigned long) create_tables((char *)p,argc,envc);
	current->brk = ex.a_bss +
		(current->end_data = ex.a_data +
		(current->end_code = ex.a_text));
	current->start_stack = p & 0xfffff000;
	current->euid = e_uid;
	current->egid = e_gid;
	i = ex.a_text+ex.a_data;
	while (i&0xfff)
		put_fs_byte(0,(char *) (i++));
	eip[0] = ex.a_entry;		/* eip, magic happens :-) */
	eip[3] = p;			/* stack pointer */
	return 0;
exec_error2:
	iput(inode);
exec_error1:
	for (i=0 ; i