__FILE__,__LINE__,FUNCTION__实现代码跟踪调试（下c语言编程 ）先看下简单的初始代码：注意其编译运行后的结果。
root@xuanfei-desktop:~/cpropram/2# cat global.h //头文件#ifndef CLOBAL_H #define GLOBAL_H #include <stdio.h> int funca(void); int funcb(void);#endifroot@xuanfei-desktop:~/cpropram/2# cat funca.c //函数a#include "global.h"int funca(void){ printf ("this is function\n");return 0;}root@xuanfei-desktop:~/cpropram/2# cat funcb.c //函数b#include "global.h"int funcb(void){ printf ("this is function\n");return 0;}root@xuanfei-desktop:~/cpropram/2# gcc -Wall funca.c funcb.c main.c //联合编译root@xuanfei-desktop:~/cpropram/2# ./a.out //运行this is mainthis is this is mainthis is functionthis is main相同结果很难让人看出那里出错，下面我们用用 __FILE__,__LINE__,__FUNCTION__加入代码，看看有什么区别吗.把 __FILE__,__LINE__,__FUNCTION__加入到mail.c中root@xuanfei-desktop:~/cpropram/2# cat main.c#include "global.h"int main(int argc, char **argv){ printf("%s(%d)-%s: this is main\n",__FILE__,__LINE__,__FUNCTION__); funca(); printf("%s(%d)-%s: this is main\n",__FILE__,__LINE__,__FUNCTION__); funcb(); printf("%s(%d)-%s: this is main\n",__FILE__,__LINE__,__FUNCTION__); return 0;}root@xuanfei-desktop:~/cpropram/2# gcc -Wall funca.c funcb.c main.croot@xuanfei-desktop:~/cpropram/2# ./a.outmain.c(4)-main: this is mainthis is functionmain.c(6)-main: this is mainthis is functionmain.c(8)-main: this is main上面的结果main.c(4)-main:this is main 表示在mian.c源代码的第四行main函数里边打印出来的 this is main那样的话就很方便的让程序员对自己的程序进行排错！为了更方便的使用它我们可以通过在global.h代码中进行宏定义root@xuanfei-desktop:~/cpropram/2# cat global.h#ifndef CLOBAL_H #define GLOBAL_H #include <stdio.h> int funca(void); int funcb(void); #define DEBUGFMT "%s(%d)-%s" #define DEBUGARGS __FILE__,__LINE__,__FUNCTION__#endifroot@xuanfei-desktop:~/cpropram/2# cat funca.c#include "global.h"int funca(void){ printf (DEBUGFMT " this is function\n",DEBUGARGS);return 0;}root@xuanfei-desktop:~/cpropram/2# cat funcb.c#include "global.h"int funcb(void){ printf (DEBUGFMT " this is function\n",DEBUGARGS);return 0;}root@xuanfei-desktop:~/cpropram/2# cat main.c#include "global.h"int main(int argc, char **argv){ printf(DEBUGFMT "this is main\n", DEBUGARGS); funca(); printf(DEBUGFMT "this is main\n", DEBUGARGS); funcb(); printf(DEBUGFMT "this is main\n", DEBUGARGS); return 0;}root@xuanfei-desktop:~/cpropram/2# gcc -Wall funca.c funcb.c main.croot@xuanfei-desktop:~/cpropram/2# ./a.outmain.c(4)-mainthis is mainfunca.c(4)-funca this is functionmain.c(6)-mainthis is mainfuncb.c(4)-funcb this is functionmain.c(8)-mainthis is mainroot@xuanfei-desktop:~/cpropram/2#这就是通过定义__FILE__,__LINE__,FUNCTION__的宏来简单实现代码的跟踪调试：）下面是一个可供调试用的头文件#ifndef _GOLD_DEBUG_H#define _GOLD_DEBUG_H
#ifdef __cplusplus
#if __cplusplusextern "C"{ #endif#endif /* __cplusplus *///#define GI_DEBUG
#ifdef GI_DEBUG
#define GI_DEBUG_POINT() printf("\n\n[File:%s Line:%d] Fun:%s\n\n", __FILE__, __LINE__, __FUNCTION__)
#define dbg_printf(arg...) printf(arg);#define GI_ASSERT(expr) \
do{ \ if (!(expr)) { \ printf("\nASSERT failed at:\n >File name: %s\n >Function : %s\n >Line No. : %d\n >Condition: %s\n", \ __FILE__,__FUNCTION__, __LINE__, #expr);\ } \ }while(0);/*调试宏, 用于暂停*/
#define GI_DEBUG_PAUSE() \ do \ { \ GI_DEBUG_POINT(); \ printf("pause for debug, press 'q' to exit!\n"); \ char c; \ while( ( c = getchar() ) ) \ { \ if('q' == c) \ { \ getchar(); \ break; \ } \ } \ }while(0);#define GI_DEBUG_PAUSE_ARG(arg...) \ do \ { \ printf(arg); \ GI_DEBUG_PAUSE() \ }while(0); #define GI_DEBUG_ASSERT(expression) \if(!(expression)) \{ \ printf("[ASSERT],%s,%s:%d\n", __FILE__, __FUNCTION__, __LINE__);\ exit(-1); \}#else#define GI_ASSERT(expr)#define GI_DEBUG_PAUSE()#define GI_DEBUG_PAUSE_ARG(arg...) #define GI_DEBUG_POINT()#define dbg_printf(arg...)#define GI_DEBUG_ASSERT(expression)#endif
#ifdef __cplusplus
#if __cplusplus}#endif#endif /* __cplusplus */ #endifC语言常用宏定义
01: 防止一个头文件被重复包含
#ifndef COMDEF_H#define COMDEF_H//头文件内容#endif02: 重新定义一些类型,防止由于各种平台和编译器的不同,而产生的类型字节数差异,方便移植。typedef unsigned char boolean; /* Boolean value type. */typedef unsigned long int uint32; /* Unsigned 32 bit value */typedef unsigned short uint16; /* Unsigned 16 bit value */typedef unsigned char uint8; /* Unsigned 8 bit value */typedef signed long int int32; /* Signed 32 bit value */typedef signed short int16; /* Signed 16 bit value */typedef signed char int8; /* Signed 8 bit value *///下面的不建议使用typedef unsigned char byte; /* Unsigned 8 bit value type. */typedef unsigned short word; /* Unsinged 16 bit value type. */typedef unsigned long dword; /* Unsigned 32 bit value type. */typedef unsigned char uint1; /* Unsigned 8 bit value type. */typedef unsigned short uint2; /* Unsigned 16 bit value type. */typedef unsigned long uint4; /* Unsigned 32 bit value type. */typedef signed char int1; /* Signed 8 bit value type. */typedef signed short int2; /* Signed 16 bit value type. */typedef long int int4; /* Signed 32 bit value type. */typedef signed long sint31; /* Signed 32 bit value */typedef signed short sint15; /* Signed 16 bit value */typedef signed char sint7; /* Signed 8 bit value */03: 得到指定地址上的一个字节或字#define MEM_B(x) (*((byte *)(x)))#define MEM_W(x) (*((word *)(x)))04: 求最大值和最小值#define MAX(x,y) (((x)>(y)) ? (x) : (y))#define MIN(x,y) (((x) < (y)) ? (x) : (y))05: 得到一个field在结构体(struct)中的偏移量#define FPOS(type,field) ((dword)&((type *)0)->field)06: 得到一个结构体中field所占用的字节数#define FSIZ(type,field) sizeof(((type *)0)->field)07: 按照LSB格式把两个字节转化为一个Word#define FLIPW(ray) ((((word)(ray)[0]) * 256) + (ray)[1])08: 按照LSB格式把一个Word转化为两个字节#define FLOPW(ray,val) (ray)[0] = ((val)/256); (ray)[1] = ((val) & 0xFF)09: 得到一个变量的地址（word宽度）#define B_PTR(var) ((byte *) (void *) &(var))#define W_PTR(var) ((word *) (void *) &(var))10: 得到一个字的高位和低位字节#define WORD_LO(xxx) ((byte) ((word)(xxx) & 255))#define WORD_HI(xxx) ((byte) ((word)(xxx) >> 8))11: 返回一个比X大的最接近的8的倍数#define RND8(x) ((((x) + 7)/8) * 8)12: 将一个字母转换为大写#define UPCASE(c) (((c)>='a' && (c) <= 'z') ? ((c) - 0x20) : (c))13: 判断字符是不是10进值的数字#define DECCHK(c) ((c)>='0' && (c)<='9')
14: 判断字符是不是16进值的数字#define HEXCHK(c) (((c) >= '0' && (c)<='9') ((c)>='A' && (c)<= 'F') \((c)>='a' && (c)<='f'))
15: 防止溢出的一个方法#define INC_SAT(val) (val=((val)+1>(val)) ? (val)+1 : (val))
16: 返回数组元素的个数#define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))17: 返回一个无符号数n尾的值MOD_BY_POWER_OF_TWO(X,n)=X%(2^n)#define MOD_BY_POWER_OF_TWO( val, mod_by ) ((dword)(val) & (dword)((mod_by)-1))18: 对于IO空间映射在存储空间的结构,输入输出处理#define inp(port) (*((volatile byte *)(port)))#define inpw(port) (*((volatile word *)(port)))#define inpdw(port) (*((volatile dword *)(port)))#define outp(port,val) (*((volatile byte *)(port))=((byte)(val)))#define outpw(port, val) (*((volatile word *)(port))=((word)(val)))#define outpdw(port, val) (*((volatile dword *)(port))=((dword)(val)))19: 使用一些宏跟踪调试ANSI标准说明了五个预定义的宏名。它们是：__LINE____FILE____DATE____TIME____STDC__C++中还定义了 __cplusplus如果编译器不是标准的,则可能仅支持以上宏名中的几个,或根本不支持。记住编译程序也许还提供其它预定义的宏名。__LINE__ 及 __FILE__ 宏指示，#line指令可以改变它的值，简单的讲，编译时，它们包含程序的当前行数和文件名。__DATE__ 宏指令含有形式为月/日/年的串,表示源文件被翻译到代码时的日期。__TIME__ 宏指令包含程序编译的时间。时间用字符串表示，其形式为： 分：秒__STDC__ 宏指令的意义是编译时定义的。一般来讲，如果__STDC__已经定义，编译器将仅接受不包含任何非标准扩展的标准C/C++代码。如果实现是标准的,则宏__STDC__含有十进制常量1。如果它含有任何其它数,则实现是非标准的。__cplusplus 与标准c++一致的编译器把它定义为一个包含至少6为的数值。与标准c++不一致的编译器将使用具有5位或更少的数值。可以定义宏,例如:当定义了_DEBUG,输出数据信息和所在文件所在行#ifdef _DEBUG#define DEBUGMSG(msg,date) printf(msg);printf(“%d%d%d”,date,_LINE_,_FILE_)#else#define DEBUGMSG(msg,date) #endif 20： 宏定义防止错误使用小括号包含。例如：有问题的定义：#define DUMP_WRITE(addr,nr) {memcpy(bufp,addr,nr); bufp += nr;}应该使用的定义： #difne DO(a,b) do{a+b;a++;}while(0)例如：if(addr) DUMP_WRITE(addr,nr);else do_somethong_else();宏展开以后变成这样:if(addr) {memcpy(bufp,addr,nr); bufp += nr;};else do_something_else();gcc 在碰到else前面的“；”时就认为if语句已经结束，因而后面的else不在if语句中。而采用do{} while(0)的定义，在任何情况下都没有问题。而改为 #difne DO(a,b) do{a+b;a++;}while(0) 的定义则在任何情况下都不会出错。