本文详细解析了一段使用C语言编写的代码片段，重点讲解了内存分配、指针操作以及数组与指针之间的关系。通过代码实例，展示了如何在内存中分配变量、指针和数组的空间，并阐述了编译器如何处理指针和数组名以实现内存的高效利用。文章还特别强调了指针与数组元素之间的转换，以及指针常量与普通指针的区别，帮助读者更好地理解C语言中内存管理的核心概念。
/********************************test,c***********************
song
8.13
*************************************************************/
#include<stdio.h>
int i=6;
int cc[9]={1,2,3,4,5,6,7,8,9};
int * const pa=cc;//注意const的位置：不是const int * pa，
int * ppa=cc;
int bb[2][3]={{0x50,0x51,0x52},{0x53,0x54,0x55}};
int main()
{
i=3;
i=3;
*(cc+1)=6;
*(pa+1)=8;
*(ppa+1)=9;
*(*(bb+1)+1)=0x99;
i=2;
i=2;
printf("%x\n",i);
printf("%x\n",*(cc+1));
printf("%x\n",*(pa+1));
printf("%x\n",*(ppa+1));
printf("%x\n",*(*(bb+1)+1));
printf("%x\n",cc);
printf("%x\n",pa);
printf("%x\n",ppa);
printf("%x\n",bb);
printf("%x\n",&i);
printf("%x\n",&cc);
printf("%x\n",&pa);
printf("%x\n",&ppa);
printf("%x\n",&bb);
}[root@localhost mmap]# ./test
2
9
9
9
99
8049800
8049800
8049800
8049828
80497e0
8049800
8048634
8049824
8049828
//print sections
[root@localhost mmap]# readelf -S test
[Nr] Name Type Addr Off Size ES Flg Lk Inf Al
[15] .rodata PROGBITS 08048628 000628 000014 00 A 0 0 4
[24] .data PROGBITS 080497c0 0007c0 000080 00 WA 0 0 32
//print symblols
[root@localhost mmap]# readelf -s test
Num: Value Size Type Bind Vis Ndx Name
46: 08048634 4 OBJECT GLOBAL DEFAULT 15 pa
53: 08049800 36 OBJECT GLOBAL DEFAULT 24 cc
57: 080497e0 4 OBJECT GLOBAL DEFAULT 24 i
62: 08049828 24 OBJECT GLOBAL DEFAULT 24 bb
65: 08049824 4 OBJECT GLOBAL DEFAULT 24 ppa
[root@localhost mmap]# objdump -s test
Contents of section .rodata:
8048628 03000000 01000200 00000000 00980408 ................
/*
the address of pa is starting from 8048634 ,length 4 bytes,it's content
is "00980408" displaying as little endian
actually it is 08049800---the address of cc
*/
Contents of section .data:
80497c0 00000000 00000000 00000000 00000000 ................
80497d0 00000000 00000000 00000000 00000000 ................
80497e0 06000000 00000000 00000000 00000000 ................
80497f0 00000000 00000000 00000000 00000000 ................
8049800 01000000 02000000 03000000 04000000 ................
8049810 05000000 06000000 07000000 08000000 ................
8049820 09000000 00980408 50000000 51000000 ........P...Q...
8049830 52000000 53000000 54000000 55000000 R...S...T...U...
/*
the address of cc is starting from 08049800 ,4*9=36 bytes length ,we can see it's
content is 1 2 3 4 5 6 7 8 9 from data section
the address of bb is starting from 08049828 , 4*2*3=24 bytes length
cotent is 50 51 52 53 54 55
the address of i is starting from 080497e0 , 4 bytes length,cotent is 6
the address of ppa is starting from 08049824 , the length and cotent are the same as pa
located in rodata section
*/
可用下图表示变量，指针，数组的内存分分配
[root@localhost mmap]# objdump -S test
080483c4 <main>:
80483c4: 8d 4c 24 04 lea 0x4(%esp),%ecx
80483c8: 83 e4 f0 and $0xfffffff0,%esp
80483cb: ff 71 fc pushl -0x4(%ecx)
80483ce: 55 push %ebp
80483cf: 89 e5 mov %esp,%ebp
80483d1: 51 push %ecx
80483d2: 83 ec 14 sub $0x14,%esp
80483d5: c7 05 e0 97 04 08 03 movl $0x3,0x80497e0//i=3;
80483dc: 00 00 00
80483df: c7 05 e0 97 04 08 03 movl $0x3,0x80497e0//i=3;
80483e6: 00 00 00
/*$表示一个立即数，%表示一个寄存器,( )表示寄存器间接寻址
编译器将i用其地址0x80497e0替换掉
翻译为将立即数3移到内存单元0x80497e0*/
80483e9: b8 04 98 04 08 mov $0x8049804,%eax//*(cc+1)=6;
80483ee: c7 00 06 00 00 00 movl $0x6,(%eax)
/*
编译器将cc+1用cc+1所表示的地址值替换掉
cc+1-->8049800+4=8049804
编译器将*(cc+1)翻译为将计算出来的地址塞进累加器eax，
利用寄存器间接寻址将立即数6塞进内存单元8049804
*/
80483f4: a1 34 86 04 08 mov 0x8048634,%eax//*(pa+1)=8;
80483f9: 83 c0 04 add $0x4,%eax
80483fc: c7 00 08 00 00 00 movl $0x8,(%eax)
/*
编译器将pa用pa之地址替换下来
编译器将*(pa+1)翻译为将pa地址8048634塞进eax
执行+1操作，即eax=8048634+4=8048638
将立即数塞进8048638内存单元
*/
8048402: a1 24 98 04 08 mov 0x8049824,%eax//*(ppa+1)=9;
8048407: 83 c0 04 add $0x4,%eax
804840a: c7 00 09 00 00 00 movl $0x9,(%eax)
/*
同*(pa+1)=8;
可以看出，指针常量和指针都被编译器翻译成一样的代码，
数组名cc+1,cc[1]等都是编译器编译时执行的的运算
pa+1是程序运行时执行的运算
所以，数组名只是一个c语言的规定，用于编译器计算数组元素的地址值，
c语言在数组范畴规定了这么多的东西，要求我们按照他说的规范去写，然后他在编译的时候，直接套用
规范计算出实际地址，而不是在程序运行的时候去计算
指针常量和指针都是有一个32b的内存单元来装其值得
数组名没有
数组名在内存中更像是普通变量(的集合)
但是在c语言中,大部分场合，要将数组名"假想"作一个指针去用,
这是c语言规范，因为 编译时，是将数组名当做指针对待
比如将cc[1]先在内部译成*(cc+1)
这也就解释了指针也可以写成数组的形式，因为即使你写成了数组的形式（不管实际是数组还是指针）
，编译器还是会将其换成指针
一下来自 thq c程序设计
int *p;
int a[10];
p=&a[0](或p=a);
则
p+i，a+i都是a[i]的地址.a[i]也可写作p[i]
*(p+i)，*(a+i)都是p+i或a+i指向的数组元素即a[i]的值
比如p+2,代表a[2]的地址，或说p[2]的地址
*/
8048410: b8 28 98 04 08 mov $0x8049828,%eax//*(*(bb+1)+1)=0x99;
8048415: 83 c0 10 add $0x10,%eax
8048418: c7 00 99 00 00 00 movl $0x99,(%eax)
/*
8049828+12+4
*/
804841e: c7 05 e0 97 04 08 02 movl $0x2,0x80497e0//i=2;
8048425: 00 00 00
8048428: c7 05 e0 97 04 08 02 movl $0x2,0x80497e0//i=2;
804842f: 00 00 00
8048432: a1 e0 97 04 08 mov 0x80497e0,%eax
8048437: 89 44 24 04 mov %eax,0x4(%esp)
804843b: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
8048442: e8 ad fe ff ff call 80482f4 <printf@plt>
8048447: b8 04 98 04 08 mov $0x8049804,%eax
804844c: 8b 00 mov (%eax),%eax
804844e: 89 44 24 04 mov %eax,0x4(%esp)
8048452: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
8048459: e8 96 fe ff ff call 80482f4 <printf@plt>
804845e: a1 34 86 04 08 mov 0x8048634,%eax
8048463: 83 c0 04 add $0x4,%eax
8048466: 8b 00 mov (%eax),%eax
8048468: 89 44 24 04 mov %eax,0x4(%esp)
804846c: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
8048473: e8 7c fe ff ff call 80482f4 <printf@plt>
8048478: a1 24 98 04 08 mov 0x8049824,%eax
804847d: 83 c0 04 add $0x4,%eax
8048480: 8b 00 mov (%eax),%eax
8048482: 89 44 24 04 mov %eax,0x4(%esp)
8048486: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
804848d: e8 62 fe ff ff call 80482f4 <printf@plt>
8048492: b8 28 98 04 08 mov $0x8049828,%eax
8048497: 83 c0 10 add $0x10,%eax
804849a: 8b 00 mov (%eax),%eax
804849c: 89 44 24 04 mov %eax,0x4(%esp)
80484a0: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
80484a7: e8 48 fe ff ff call 80482f4 <printf@plt>
80484ac: c7 44 24 04 00 98 04 movl $0x8049800,0x4(%esp)
80484b3: 08
80484b4: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
80484bb: e8 34 fe ff ff call 80482f4 <printf@plt>
80484c0: a1 34 86 04 08 mov 0x8048634,%eax
80484c5: 89 44 24 04 mov %eax,0x4(%esp)
80484c9: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
80484d0: e8 1f fe ff ff call 80482f4 <printf@plt>
80484d5: a1 24 98 04 08 mov 0x8049824,%eax
80484da: 89 44 24 04 mov %eax,0x4(%esp)
80484de: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
80484e5: e8 0a fe ff ff call 80482f4 <printf@plt>
80484ea: c7 44 24 04 28 98 04 movl $0x8049828,0x4(%esp)
80484f1: 08
80484f2: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
80484f9: e8 f6 fd ff ff call 80482f4 <printf@plt>
80484fe: c7 44 24 04 e0 97 04 movl $0x80497e0,0x4(%esp)
8048505: 08
8048506: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
804850d: e8 e2 fd ff ff call 80482f4 <printf@plt>
8048512: c7 44 24 04 00 98 04 movl $0x8049800,0x4(%esp)
8048519: 08
804851a: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
8048521: e8 ce fd ff ff call 80482f4 <printf@plt>
8048526: c7 44 24 04 34 86 04 movl $0x8048634,0x4(%esp)
804852d: 08
804852e: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
8048535: e8 ba fd ff ff call 80482f4 <printf@plt>
804853a: c7 44 24 04 24 98 04 movl $0x8049824,0x4(%esp)
8048541: 08
8048542: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
8048549: e8 a6 fd ff ff call 80482f4 <printf@plt>
804854e: c7 44 24 04 28 98 04 movl $0x8049828,0x4(%esp)
8048555: 08
8048556: c7 04 24 38 86 04 08 movl $0x8048638,(%esp)
804855d: e8 92 fd ff ff call 80482f4 <printf@plt>
8048562: 83 c4 14 add $0x14,%esp
8048565: 59 pop %ecx
8048566: 5d pop %ebp
8048567: 8d 61 fc lea -0x4(%ecx),%esp
804856a: c3 ret
804856b: 90 nop
804856c: 90 nop
804856d: 90 nop
804856e: 90 nop
804856f: 90 nop
扫一扫
1100
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
