inet_ntoa的实现

从一段代码开始:

[cpp] view plain copy print ?

1. int test_ntoa() 
2. { 
3.     struct sockaddr_in recv = {0}; 
4.     struct sockaddr_in sa   = {0}; 
5.     recv.sin_addr.s_addr = -217732928; //192.168.5.243   
6.     sa.sin_addr.s_addr   = -939415360; //192.168.5.200 
7.     printf("ip of recv:%s ip of sa:%s\n", inet_ntoa(recv.sin_addr), inet_ntoa(sa.sin_addr)); 
8.     printf("ip of recv:%s\n", inet_ntoa(recv.sin_addr)); 
9.     printf("ip of sa  :%s\n", inet_ntoa(sa.sin_addr));  
10.     return 0;    
11. } 

int test_ntoa() { struct sockaddr_in recv = {0}; struct sockaddr_in sa = {0}; recv.sin_addr.s_addr = -217732928; //192.168.5.243 sa.sin_addr.s_addr = -939415360; //192.168.5.200 printf("ip of recv:%s ip of sa:%s\n", inet_ntoa(recv.sin_addr), inet_ntoa(sa.sin_addr)); printf("ip of recv:%s\n", inet_ntoa(recv.sin_addr)); printf("ip of sa :%s\n", inet_ntoa(sa.sin_addr)); return 0; }

结果有点诧异:

[cpp] view plain copy print ?

1. ip of recv:192.168.5.243 ip of sa:192.168.5.243 
2. ip of recv:192.168.5.243 
3. ip of sa   :192.168.1.200 

ip of recv:192.168.5.243 ip of sa:192.168.5.243 ip of recv:192.168.5.243 ip of sa :192.168.1.200

C函数出现这类问题,多半是由于其实现使用了静态内存所知.

google其源码:

[cpp] view plain copy print ?

1. char * 
2. inet_ntoa (struct in_addr in) 
3. { 
4.   __libc_once_define (static, once); 
5.   char *buffer; 
6.   unsigned char *bytes; 
7.  
8.   /* If we have not yet initialized the buffer do it now.  */ 
9.   __libc_once (once, init); 
10.  
11.   if (static_buf != NULL) 
12.     buffer = static_buf; 
13.   else 
14.     { 
15.       /* We don't use the static buffer and so we have a key.  Use it
16.      to get the thread-specific buffer.  */ 
17.       buffer = __libc_getspecific (key); 
18.       if (buffer == NULL) 
19.     { 
20.       /* No buffer allocated so far.  */ 
21.       buffer = malloc (18); 
22.       if (buffer == NULL) 
23.         /* No more memory available.  We use the static buffer.  */ 
24.         buffer = local_buf; 
25.       else 
26.         __libc_setspecific (key, buffer); 
27.     } 
28.     } 
29.  
30.   bytes = (unsigned char *) ∈ 
31.   __snprintf (buffer, 18, "%d.%d.%d.%d", 
32.           bytes[0], bytes[1], bytes[2], bytes[3]); 
33.  
34.   return buffer; 
35. } 

char * inet_ntoa (struct in_addr in) { __libc_once_define (static, once); char *buffer; unsigned char *bytes; /* If we have not yet initialized the buffer do it now. */ __libc_once (once, init); if (static_buf != NULL) buffer = static_buf; else { /* We don't use the static buffer and so we have a key. Use it to get the thread-specific buffer. */ buffer = __libc_getspecific (key); if (buffer == NULL) { /* No buffer allocated so far. */ buffer = malloc (18); if (buffer == NULL) /* No more memory available. We use the static buffer. */ buffer = local_buf; else __libc_setspecific (key, buffer); } } bytes = (unsigned char *) ∈ __snprintf (buffer, 18, "%d.%d.%d.%d", bytes[0], bytes[1], bytes[2], bytes[3]); return buffer; }
这里使用了静态的缓存buffer,函数把转换后的ip写入缓存buffer,然后返回buffer的首地址值.

看似无奈的选择.glibc里的某些底层函数都使用这类处理方式.

其实也可以选择外部传参的做法解决:

[cpp] view plain copy print ?

1. <pre name="code"class="cpp">char * 
2. inet_ntoa (struct in_addr in, char* buf, int buflen) 
3. { 
4.       ... ... 
5. } 

<pre name="code" class="cpp">char * inet_ntoa (struct in_addr in, char* buf, int buflen) { ... ... }

这样的处理方式,效率略低于第一种.或许这也是glibc选择第一种方式的原因.