Chinaunix首页 | 论坛 | 博客
  • 博客访问: 4735816
  • 博文数量: 930
  • 博客积分: 12070
  • 博客等级: 上将
  • 技术积分: 11448
  • 用 户 组: 普通用户
  • 注册时间: 2008-08-15 16:57
文章分类

全部博文(930)

文章存档

2011年(60)

2010年(220)

2009年(371)

2008年(279)

分类: LINUX

2009-05-20 08:55:12

  仰慕kernel的list.h良久了,还记得以前o2的李大哥号称list.h飘逸写手...自己也曾经牛刀小试过一把.今日得闲,暂探索性的分析一把.

1. 简约而又不简单的链表定义

于双向链表而言,内核中定义了如下简单结构:

  struct list_head {                     
    struct list_head *next, *prev;  
 };                                               
   
 这个不含任何数据项的结构,注定了它的通用性和未来使用的灵活性,例如前面的例子就可以按如下方式定义:
 
struct my_list{ 
void *mydata;
struct list_head list;
};  
  在此,进一步说明几点:
  1)list字段,隐藏了链表的指针特性,但正是它,把我们要链接的数据组织成了链表。
  2)struct list_head可以位于结构的任何位置
  3)可以给struct list_head起任何名字。
  4)在一个结构中可以有多个list

例如,我们对要完成的任务进行描述,而任务中又包含子任务,于是有如下结构:

-------------------------------------------------------------------------------------------------------------------------
struct todo_tasks{
char *task_name;
unsigned int name_len;
short int status;

int sub_tasks;

int subtasks_completed;
struct list_head completed_subtasks;/* 已完成的子任务形成链表 */

int subtasks_waiting;
struct list_head waiting_subtasks; /* 待完成的子任务形成链表 */

struct list_head todo_list; /* 要完成的任务形成链表 */
};
-----------------------------------------------------------------------
  简约而又不简单struct list_head,以此为基本对象,就衍生了对链表的插入、删除、合并以及遍历等各种操作:
2. 链表的声明和初始化宏
   

   实际上, struct list_head只定义了链表节点,并没有专门定义链表头,那么一个链表结构是如何建立起来的?让我们来看看下面两个宏:

#define LIST_HEAD_INIT(name) { &(name), &(name) }
#define LIST_HEAD(name) struct list_head name = LIST_HEAD_INIT(name)

   如果我们要申明并定义自己的链表头mylist,直接调用LIST_HEAD:

    LIST_HEAD(mylist)

    则mylist的next、prev指针都初始化为指向自己,这样,我们就有了一个空链表,如何判断链表是否为空,自己写一下这个简单的函数 ,也就是让头指针的next指向自己而已。


3. staitic inline函数-隐藏并展开
   在list.h中定义的函数大都是 staitic inline f()形式?为什么这样定义?
   关键字“static”加在函数前,表示这个函数是静态函数,所谓静态函数,实际上是对函数作用域的限制,指该函数的作用域仅
 局限于本文件。所以说,static具有信息隐藏作用。
     而关键字"inline“加在函数前,说明这个函数对编译程序是可见的,也就是说,编译程序在调用这个函数时就立即展开该函数。所以,关键字inline 必须与函数定义体放在一起才能使函数成为内联。inline函数一般放在头文件中。

4.  无处不在的隐藏特性
     我们分析一下在链表中增加一个节点的函数实现:
     有三个函数:
     static inline void __list_add();
     static inline void list_add();
     static inline void list_add_tail();


-------------------------------------------------------------------------------------------------
  /*
 * Insert a new entry between two known consecutive entries. 
 *
 * This is only for internal list manipulation where we know
 * the prev/next entries already!
 */
static inline void __list_add(struct list_head *new,
                  struct list_head *prev,
                  struct list_head *next)
{
    next->prev = new;
    new->next = next;
    new->prev = prev;
    prev->next = new;
}
--------------------------------------------------------------------------------------------------
/**
 * list_add - add a new entry
 * @new: new entry to be added
 * @head: list head to add it after
 *
 * Insert a new entry after the specified head.
 * This is good for implementing stacks.
 */
static inline void list_add(struct list_head *new, struct list_head *head)
{
    __list_add(new, head, head->next);
}
--------------------------------------------------------------------------------------------------
/**
 * list_add_tail - add a new entry
 * @new: new entry to be added
 * @head: list head to add it before
 *
 * Insert a new entry before the specified head.                                          

 * This is useful for implementing queues.
 */
static inline void list_add_tail(struct list_head *new, struct list_head *head)
{
    __list_add(new, head->prev, head);
}

--------------------------------------------------------------------------------------------------
   
   仔细体会其实现代码,看起来简单有效,但实际上也是一种抽象和封装的体现。首先__list_add()函数做基本的操作,该函数仅仅是增加一个节点,至 于这个节点加到何处,暂不考虑。list_add()调用__list_add()这个内部函数,在链表头增加一个节点,实际上实现了栈在头部增加节点的 操作,而list_add_tail()在尾部增加一个节点,实际上实现了队的操作。
 
至于链表的删除、搬移和合并,比较简单,不再此一一讨论

5. 链表遍历-似走过千山万水
  遍历链表本是简单的,list.h中就定义了如下的宏:
--------------------------------------------------------------------------------------------------
  **
 * list_for_each    -    iterate over a list
 * @pos:    the &struct list_head to use as a loop counter.
 * @head:    the head for your list.
 */
#define list_for_each(pos, head) \
    for (pos = (head)->next; pos != (head); \
            pos = pos->next)

--------------------------------------------------------------------------------------------------
  这种遍历仅仅是找到一个个节点在链表中的位置pos,难点在于,如何通过pos获得节点的地址,从而可以使用节点中的数据? 于是 list.h中定义了晦涩难懂的list_entry()宏:


--------------------------------------------------------------------------------------------------
/**
 * list_entry - get the struct for this entry
 * @ptr:    the &struct list_head pointer.
 * @type:    the type of the struct this is embedded in.
 * @member:    the name of the list_struct within the struct.
 */
#define list_entry(ptr, type, member) \
    ((type *)((char *)(ptr)-(unsigned long)(&((type *)0)->member)))

--------------------------------------------------------------------------------------------------
指针ptr指向结构体type中的成员member;通过指针ptr,返回结构体type的起始地址,如图2。
      
          type

          |----------|
          |              |
          |              |
          |----------|
ptr--> | member --|
          |----------|
          |              |
          |              |
          |----------| 

        图2 list_entry()宏的示意图

为了便于理解,在此给予进一步说明。
 例如my_list结构:
struct my_list{ 
void *mydata;
struct list_head list;
};
struct list_head *pos;  

则list_entry(pos, mylist, list)宏,就可以根据pos的值,获取mylist的地址,也就是指向mylist的指针,这样,我们就可以存取mylist->mydata字段了。

可为什么能够达到这样的效果?
list_entry(pos, mylist, list) 展开以后为:

  ((struct my_list *)((char *)(pos) - (unsigned long)(&((struct my_list *)0)->list)))

这看起来会使大多数人眩晕,但仔细分析一下,实际很简单。
  ((size_t) &(type *)0)->member)把0地址转化为type结构的指针,然后获取该结构中member成员的指针,并将其强制转换为size_t类型。于 是,由于结构从0地址开始定义,因此,这样求出member的成员地址,实际上就是它在结构中的偏移量。为了更好的理解这些,我们可以写一段程序来验证:
---------------------------------------------------------------------------------------

#include
#include 

struct foobar{
unsigned int foo;
char bar;
char boo;
};

int main(int argc, char** argv){

struct foobar tmp;

printf("address of &tmp is= %p\n\n", &tmp);
printf("address of tmp->foo= %p \t offset of tmp->foo= %lu\n", &tmp.foo, (unsigned long) &((struct foobar *)0)->foo);
printf("address of tmp->bar= %p \t offset of tmp->bar= %lu\n", &tmp.bar, (unsigned long) &((struct foobar *)0)->bar);
printf("address of tmp->boo= %p \t offset of tmp->boo= %lu\n\n", &tmp.boo, (unsigned long) &((struct foobar *)0)->boo);

printf("computed address of &tmp using:\n");
printf("\taddress and offset of tmp->foo= %p\n",
(struct foobar *) (((char *) &tmp.foo) - ((unsigned long) &((struct foobar *)0)->foo)));
printf("\taddress and offset of tmp->bar= %p\n",
(struct foobar *) (((char *) &tmp.bar) - ((unsigned long) &((struct foobar *)0)->bar)));
printf("\taddress and offset of tmp->boo= %p\n",
(struct foobar *) (((char *) &tmp.boo) - ((unsigned long) &((struct foobar *)0)->boo)));

return 0;
}
Output from this code is:
address of &tmp is= 0xbfffed00

address of tmp->foo= 0xbfffed00 offset of tmp->foo= 0
address of tmp->bar= 0xbfffed04 offset of tmp->bar= 4
address of tmp->boo= 0xbfffed05 offset of tmp->boo= 5

computed address of &tmp using:
address and offset of tmp->foo= 0xbfffed00
address and offset of tmp->bar= 0xbfffed00
address and offset of tmp->boo= 0xbfffed00

----------------------------------------------------------------------------------------

  到此,我们对链表的实现机制有所了解,但在此止步的话,我们依然无法领略这风景背后的韵味。
  尽管list.h是内核代码中的头文件,但我们可以把它移植到用户空间使用。且看下一讲,链表接口之应用。


[root@mip-123456 list]# cat list.h
#ifndef _LINUX_LIST_H
#define _LINUX_LIST_H

#define LIST_POISON1 ((void *) 0x00100100)
#define LIST_POISON2 ((void *) 0x00200200)

#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
#define container_of(ptr, type, member) ({\
           const typeof( ((type *)0)->member ) *__mptr = (ptr);\
           (type *)( (char *)__mptr - offsetof(type,member) );})

struct list_head {
        struct list_head *next, *prev;
};

#define LIST_HEAD_INIT(name) { &(name), &(name) }

#define LIST_HEAD(name) \
        struct list_head name = LIST_HEAD_INIT(name)

static inline void INIT_LIST_HEAD(struct list_head *list)
{
        list->next = list;
        list->prev = list;
}


static inline void __list_add(struct list_head *new,
                              struct list_head *prev,
                              struct list_head *next)
{
        next->prev = new;
        new->next = next;
        new->prev = prev;
        prev->next = new;
}


static inline void list_add(struct list_head *new, struct list_head *head)
{
        __list_add(new, head, head->next);
}

static inline void list_add_tail(struct list_head *new, struct list_head *head)
{
        __list_add(new, head->prev, head);
}

static inline void __list_del(struct list_head * prev, struct list_head * next)
{
        next->prev = prev;
        prev->next = next;
}

static inline void list_del(struct list_head *entry)
{
        __list_del(entry->prev, entry->next);
        entry->next = LIST_POISON1;
        entry->prev = LIST_POISON2;
}

static inline void list_replace(struct list_head *old,
                                struct list_head *new)
{
        new->next = old->next;
        new->next->prev = new;
        new->prev = old->prev;
        new->prev->next = new;
}

static inline void list_replace_init(struct list_head *old,
                                        struct list_head *new)
{
        list_replace(old, new);
        INIT_LIST_HEAD(old);
}

static inline void list_del_init(struct list_head *entry)
{
        __list_del(entry->prev, entry->next);
        INIT_LIST_HEAD(entry);
}

static inline void list_move(struct list_head *list, struct list_head *head)
{
        __list_del(list->prev, list->next);
        list_add(list, head);
}

static inline void list_move_tail(struct list_head *list,
                                  struct list_head *head)
{
        __list_del(list->prev, list->next);
        list_add_tail(list, head);
}

static inline int list_is_last(const struct list_head *list,
                                const struct list_head *head)
{
        return list->next == head;
}

static inline int list_empty(const struct list_head *head)
{
        return head->next == head;
}

static inline int list_empty_careful(const struct list_head *head)
{
        struct list_head *next = head->next;
        return (next == head) && (next == head->prev);
}

static inline void __list_splice(struct list_head *list,
                                 struct list_head *head)
{
        struct list_head *first = list->next;
        struct list_head *last = list->prev;
        struct list_head *at = head->next;

        first->prev = head;
        head->next = first;

        last->next = at;
        at->prev = last;
}

static inline void list_splice(struct list_head *list, struct list_head *head)
{
        if (!list_empty(list))
                __list_splice(list, head);
}

static inline void list_splice_init(struct list_head *list,
                                    struct list_head *head)
{
        if (!list_empty(list)) {
                __list_splice(list, head);
                INIT_LIST_HEAD(list);
        }
}

#define list_entry(ptr, type, member) \
        container_of(ptr, type, member)

#define list_for_each(pos, head) \
        for (pos = (head)->next;pos != (head); \
                pos = pos->next)

#define __list_for_each(pos, head) \
        for (pos = (head)->next; pos != (head); pos = pos->next)

#define list_for_each_prev(pos, head) \
        for (pos = (head)->prev; pos != (head); \
                pos = pos->prev)

#define list_for_each_safe(pos, n, head) \
        for (pos = (head)->next, n = pos->next; pos != (head); \
                pos = n, n = pos->next)

#define list_for_each_entry(pos, head, member) \
        for (pos = list_entry((head)->next, typeof(*pos), member); \
             &pos->member != (head); \
             pos = list_entry(pos->member.next, typeof(*pos), member))

#define list_for_each_entry_reverse(pos, head, member) \
        for (pos = list_entry((head)->prev, typeof(*pos), member); \
             &pos->member != (head); \
             pos = list_entry(pos->member.prev, typeof(*pos), member))

#define list_prepare_entry(pos, head, member) \
        ((pos) ? : list_entry(head, typeof(*pos), member))


#define list_for_each_entry_continue(pos, head, member) \
        for (pos = list_entry(pos->member.next, typeof(*pos), member); \
             prefetch(pos->member.next), &pos->member != (head); \
             pos = list_entry(pos->member.next, typeof(*pos), member))

#define list_for_each_entry_from(pos, head, member) \
        for (; prefetch(pos->member.next), &pos->member != (head); \
             pos = list_entry(pos->member.next, typeof(*pos), member))


#define list_for_each_entry_safe(pos, n, head, member) \
        for (pos = list_entry((head)->next, typeof(*pos), member), \
                n = list_entry(pos->member.next, typeof(*pos), member); \
             &pos->member != (head); \
             pos = n, n = list_entry(n->member.next, typeof(*n), member))


#define list_for_each_entry_safe_continue(pos, n, head, member) \
        for (pos = list_entry(pos->member.next, typeof(*pos), member), \
                n = list_entry(pos->member.next, typeof(*pos), member); \
             &pos->member != (head); \
             pos = n, n = list_entry(n->member.next, typeof(*n), member))

#define list_for_each_entry_safe_from(pos, n, head, member) \
        for (n = list_entry(pos->member.next, typeof(*pos), member); \
             &pos->member != (head); \
             pos = n, n = list_entry(n->member.next, typeof(*n), member))

#define list_for_each_entry_safe_reverse(pos, n, head, member) \
        for (pos = list_entry((head)->prev, typeof(*pos), member), \
                n = list_entry(pos->member.prev, typeof(*pos), member); \
             &pos->member != (head); \
             pos = n, n = list_entry(n->member.prev, typeof(*n), member))

#endif

[root@mip-123456 list]# cat list.c
#include <stdio.h>
#include <stdlib.h>

#include "list.h"

struct jimmy_list{
        int id;
        struct list_head list;
        char name[10];
        };

int main(int argc, char **argv){

        struct jimmy_list *tmp;
        struct list_head *pos, *q;
        unsigned int i;

        struct jimmy_list mylist;
        INIT_LIST_HEAD(&mylist.list); /*初始化链表头*/

        for(i=0; i<5; i++){
        tmp= (struct jimmy_list *)malloc(sizeof(struct jimmy_list));
        
        tmp->id = i+1;
        sprintf(tmp->name, "jimmy %d", i+1);

        /*list_add每次添加都是添加在list的第一个节点处*/
        list_add(&(tmp->list), &(mylist.list));
        }
        /*list 5->4->3->2->1*/

        /*这里添加在尾部list_add_tail*/
        for(i=5; i<7; i++){
        tmp= (struct jimmy_list *)malloc(sizeof(struct jimmy_list));
        
        tmp->id = i+1;
        sprintf(tmp->name, "kenthy %d", i+1);
        /*list_add每次添加都是添加在list的第一个节点处*/
        list_add_tail(&(tmp->list), &(mylist.list));
        }
        /*list 5->4->3->2->1->6->7*/


        /*两种遍历list的方法--正向遍历*/
        printf("traversing the list using list_for_each()\n");
        list_for_each(pos, &mylist.list){
              tmp= list_entry(pos, struct jimmy_list, list);
              printf("id= %d name= %s\n", tmp->id, tmp->name);
        }
        printf("\n");

        /*这里添加一个替换玩玩*/
        #if 1
        tmp= (struct jimmy_list *)malloc(sizeof(struct jimmy_list));
        tmp->id = 55;
        sprintf(tmp->name, "kenthy %d", 55);
        INIT_LIST_HEAD(&(tmp->list));
        list_replace((&mylist.list)->next,&(tmp->list));
        #endif
        /*list 55->4->3->2->1->6->7*/

        printf("traversing the list using list_for_each_entry()\n");
        list_for_each_entry(tmp, &mylist.list, list)
          printf("id= %d name= %s\n", tmp->id, tmp->name);
        printf("\n");

        /*两种遍历list的方法--反向遍历*/
        printf("reverse the list using list_for_each_prev()\n");
        list_for_each_prev(pos, &mylist.list){
              tmp= list_entry(pos, struct jimmy_list, list);
              printf("id= %d name= %s\n", tmp->id, tmp->name);
        }
        printf("\n");
        
        printf("reverse the list using list_for_each_entry()\n");
        list_for_each_entry_reverse(tmp, &mylist.list, list)
          printf("id= %d name= %s\n", tmp->id, tmp->name);
        printf("\n");


 

        printf("deleting the list using list_for_each_safe()\n");
        list_for_each_safe(pos, q, &mylist.list){
        tmp= list_entry(pos, struct jimmy_list, list);
        printf("freeing item id= %d name= %s\n", tmp->id, tmp->name);
        list_del(pos);
        free(tmp);
        }

        if(list_empty(&mylist.list))
          printf("now the list if empty\n");
        return 0;
}

[root@mip-123456 list]# ./list
traversing the list using list_for_each()
id= 5 name= jimmy 5
id= 4 name= jimmy 4
id= 3 name= jimmy 3
id= 2 name= jimmy 2
id= 1 name= jimmy 1
id= 6 name= kenthy 6
id= 7 name= kenthy 7

traversing the list using list_for_each_entry()
id= 55 name= kenthy 55
id= 4 name= jimmy 4
id= 3 name= jimmy 3
id= 2 name= jimmy 2
id= 1 name= jimmy 1
id= 6 name= kenthy 6
id= 7 name= kenthy 7

reverse the list using list_for_each_prev()
id= 7 name= kenthy 7
id= 6 name= kenthy 6
id= 1 name= jimmy 1
id= 2 name= jimmy 2
id= 3 name= jimmy 3
id= 4 name= jimmy 4
id= 55 name= kenthy 55

reverse the list using list_for_each_entry()
id= 7 name= kenthy 7
id= 6 name= kenthy 6
id= 1 name= jimmy 1
id= 2 name= jimmy 2
id= 3 name= jimmy 3
id= 4 name= jimmy 4
id= 55 name= kenthy 55

deleting the list using list_for_each_safe()
freeing item id= 55 name= kenthy 55
freeing item id= 4 name= jimmy 4
freeing item id= 3 name= jimmy 3
freeing item id= 2 name= jimmy 2
freeing item id= 1 name= jimmy 1
freeing item id= 6 name= kenthy 6
freeing item id= 7 name= kenthy 7
now the list if empty

阅读(1981) | 评论(2) | 转发(0) |
给主人留下些什么吧!~~

caojiangfeng2010-07-31 15:41:23

作者对list.h分析的很好,而且用简单的例子实现运用了一下,很好!学习了!

chinaunix网友2009-05-22 14:12:38

好服月租型IT服务台登录中国,诚邀阁下莅临体验! 月租型ITSM软件,注册即可使用! 详情请登录官方网站:http://www.servicezon.com