一、位域简介
接触过Linux内核网络协议栈的人，大概都见过位域的表达方式。 如下是摘自Linux内核代码(include/linux/tcp.h)中关于tcp头部的定义:
1 struct tcphdr {
2 __be16 source;
3 __be16 dest;
4 __be32 seq;
5 __be32 ack_seq;
6 #if defined(__LITTLE_ENDIAN_BITFIELD)
7 __u16 res1:4,
8 doff:4,
9 fin:1,
10 syn:1,
11 rst:1,
12 psh:1,
13 ack:1,
14 urg:1,
15 ece:1,
16 cwr:1;
17 #elif defined(__BIG_ENDIAN_BITFIELD)
18 __u16 doff:4,
19 res1:4,
20 cwr:1,
21 ece:1,
22 urg:1,
23 ack:1,
24 psh:1,
25 rst:1,
26 syn:1,
27 fin:1;
28 #else
29 #error "Adjust your <asm/byteorder.h> defines"
30 #endif
31 __be16 window;
32 __sum16 check;
33 __be16 urg_ptr;
34 };
位域的表达方式就是变量名:位数。 从上面tcphdr的定义可以看出，位域是跟实现有关的。 下面是C1999标准中关于位域的一个样例：
EXAMPLE 3 The following obscure constructions
typedef signed int t;
typedef int plain;
struct tag {
unsigned t:4;
const t:5;
plain r:5;
};
declare a typedef name t with type signed int, a typedef name plain with type int, and a structure
with three bit-field members, one named t that contains values in the range [0, 15], an unnamed constqualified
bit-field which (if it could be accessed) would contain values in either the range [−15, +15] or
[−16, +15], and one named r that contains values in one of the ranges [0, 31], [−15, +15], or [−16, +15].
(The choice of range is implementation-defined.) The first two bit-field declarations differ in that
unsigned is a type specifier (which forces t to be the name of a structure member), while const is a
type qualifier (which modifies t which is still visible as a typedef name).
样例中给出了几个匿名的结构体成员， 如文中解释的，位域成员的取值范围是跟实现相关的。 我对由位域构成的结构体所占内存的大小比较感兴趣，就用sizeof()测试了一下，如下：
1 #include<stdio.h>
2 #include<stdlib.h>
3 #include<string.h>
4 /*
5 **Sample code by virHappy
6 */
7
8 typedef signed int t;
9 typedef int plain;
10
11 //anoymous member
12 struct tag {
13 unsigned t:4;
14 const t:5;
15 plain r:5;
16 };
17 // member is char
18 struct rec {
19 unsigned char a:1;
20 unsigned char b:1;
21 unsigned char c:1;
22 unsigned char d:1;
23 };
24
25 // member is unsigned int
26 struct rec_int {
27 unsigned int a:1;
28 unsigned int b:1;
29 unsigned int c:1;
30 unsigned int d:1;
31 };
32
33
34 #define TEST_AND_SET_BIT(x) \
35 do{ \
36 if ((x)) { \
37 printf("bit alread set.\n"); \
38 } else { \
39 (x) = 1; \
40 } \
41 }while(0)
42
43 int main()
44 {
45 struct tag st;
46 struct rec sr;
47
48 printf("size of tag is: %d\n", sizeof(st));
49 printf("size of rec is: %d\n", sizeof(sr));
50 printf("size of rec is: %d\n", sizeof(struct rec_int));
51 printf("size of rec is: %d\n", sizeof(int));
52
53 memset(&sr, 0, sizeof(struct rec));
54 TEST_AND_SET_BIT(sr.a);
55 TEST_AND_SET_BIT(sr.b);
56 TEST_AND_SET_BIT(sr.c);
57 TEST_AND_SET_BIT(sr.d);
58
59 TEST_AND_SET_BIT(sr.a);
60 TEST_AND_SET_BIT(sr.b);
61 TEST_AND_SET_BIT(sr.c);
62 TEST_AND_SET_BIT(sr.d);
63 return 0;
64 }
输出为：
root@host]# gcc -Wall bitfield.c -o bf
[root@host]# ./bf
size of tag is: 4
size of rec is: 1
size of rec is: 4
size of rec is: 4
bit alread set.
bit alread set.
bit alread set.
bit alread set.
结果显示由int类型说明符修饰的位域成员构成的结构体为4byte, 由char类型说明符修改的位域成员构成的结构体为1byte，即使实际上只声明了4个位长度大小的成员。
二、 位域的作用
1. 在看一些rfc文档时，关于包结构部分的描述， 常常看到具体的某一位具有特定的功能。而内核的网络协议栈中对应的实现就是通过位域来实现的。
2. 配置文件解析时， 有时候需要比较新的配置和已有的配置的区别， 这时需要做一些标记。 位域在这个时候就派上了用场。 优点的是占内存少。
3. 其它？