Disassembly Windows htonl () function

Because I write a network program in the system kernel sometimes need to call htons htonl such functions to convert, but because the kernel can only call the C runtime Library, the other API can not be called. I have also been exposed to a little assembly, never to learn. Read the old code to learn the way the first few chapters, as their own anti-compilation test, the results of their own really anti-out, for people who know the assembly is very easy indeed.

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

ULONG myHtonl(ULONG i) {     ULONG eax,edx,ecx;     eax = i;     edx = i;     edx = edx << 16;     eax = eax & 0x0ff00;     eax = eax | edx;     edx = i;     edx = edx & 0x0ff0000;     ecx = i >> 16;     edx = edx | ecx;     eax = eax << 8;     edx = edx >> 8;     eax = eax | edx;     return eax; }

Variables are used to register the name, so as to avoid confusion. I feel this code is not a bit of a problem, how he did not judge his system is not small end on the conversion, if their own computer is high-end how to do?

I looked at the back of a person on the Internet to write analog code.

http://wxxweb.blog.163.com/blog/static/13512690020103145256909/

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 8 5 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111

ypedef unsigned short int uint16;typedef unsigned long int uint32;  // 短整型大小端互换#define BigLittleSwap16(A)  ((((uint16)(A) & 0xff00) >> 8) | \                                                 (((uint16)(A) & 0x00ff) << 8)) // 长整型大小端互换#define BigLittleSwap32(A)  ((((uint32)(A) & 0xff000000) >> 24) | \                                                  (((uint32)(A) & 0x00ff0000) >> 8) | \                                                 (((uint32)(A) & 0x0000ff00) << 8) | \                                                  (((uint32)(A) & 0x000000ff) << 24)) // 本机大端返回1，小端返回0int checkCPUendian(){       union{              unsigned long int i;              unsigned char s[4];       }c;         c.i = 0x12345678;       return (0x12 == c.s[0]);} // 模拟htonl函数，本机字节序转网络字节序unsigned long int HtoNl(unsigned long int h){       // 若本机为大端，与网络字节序同，直接返回       // 若本机为小端，转换成大端再返回       return checkCPUendian() ? h : BigLittleSwap32(h);} // 模拟ntohl函数，网络字节序转本机字节序 unsigned long int NtoHl(unsigned long int n){       // 若本机为大端，与网络字节序同，直接返回       // 若本机为小端，网络数据转换成小端再返回       return checkCPUendian() ? n : BigLittleSwap32(n);} // 模拟htons函数，本机字节序转网络字节序unsigned short int HtoNs(unsigned short int h){       // 若本机为大端，与网络字节序同，直接返回        // 若本机为小端，转换成大端再返回       return checkCPUendian() ? h : BigLittleSwap16(h);} // 模拟ntohs函数，网络字节序转本机字节序unsigned short int NtoHs(unsigned short int n){       // 若本机为大端，与网络字节序同，直接返回       // 若本机为小端，网络数据转换成小端再返回       return checkCPUendian() ? n : BigLittleSwap16(n);}

Looking at his own added judgment, I estimate that the Windows system is estimated to be replaced by a macro honts these functions, if the small-end definition of the conversion content, if the high-end definition of the direct return can be. That's the way Windows likes to write code.