linux的目标文件格式是,深入分析linux目标文件格式

所用程序elf.cc(平台为32位Intel x86的linux)：

#include

using namespace std;

int global_ini_var = 84;

int global_uninit_var;

void func1( int i )

{

cout << i << endl;

}

int main( void )

{

static int static_var = 85;

static int static_var2;

int a = 1;

int b;

func1( static_var + static_var2 + a + b );

return a;

}

// 用g++来生成目标文件(参数-c表示只编译不链接)

g++ -c elf.cc

生成2040字节的elf.o文件，使用的binutils的工具objdump来查看elf.o。

// 输入-h得到ELF文件的各个段的基本信息(-x会得到更多信息)

objdump -h elf.o

结果：

elf.o： 文件格式 elf32-i386

节：

Idx Name Size VMA LMA File off Algn

0 .text 000000be 00000000 00000000 00000034 2**0

CONTENTS, ALLOC, LOAD, RELOC, READONLY, CODE

1 .data 00000008 00000000 00000000 000000f4 2**2

CONTENTS, ALLOC, LOAD, DATA

2 .bss 0000000c 00000000 00000000 000000fc 2**2

ALLOC

3 .init_array 00000004 00000000 00000000 000000fc 2**2

CONTENTS, ALLOC, LOAD, RELOC, DATA

4 .comment 0000002c 00000000 00000000 00000100 2**0

CONTENTS, READONLY

5 .note.GNU-stack 00000000 00000000 00000000 0000012c 2**0

CONTENTS, READONLY

6 .eh_frame 00000098 00000000 00000000 0000012c 2**2

CONTENTS, ALLOC, LOAD, RELOC, READONLY, DATA

共有6个段，段的属性中Size表示段的长度，File Offset表示段的起始位置，VMA(Virtual Memory Address)表示虚拟内存地址

，LMA(Load Memory Address)表示内存装载地址。Algn表示对齐标识。

各段第二行表示各段属性,CONTENTS表示该段在文件中存在，其中.bss没有CONTENTS，表示它实际在ELF文件中不存在，

.note.GUN-stack有CONTENTS，但长度为0，我们认为它也不在ELF文件中存在。ALLOC表示其在进程中需要分配空间。

READONLY表示只读段。

下面是elf.o在ELF中的结构图：

下面挖掘各段的内容。objdump的“-s”参数可以将所有段的内容以十六进制的方式打印出来，“-d”参数可以将所有包含指令的段反汇编。

mimiasd@mimiasd-All-Series:~$ objdump -s -d elf.o

elf.o： 文件格式 elf32-i386

Contents of section .text:

0000 5589e583 ec188b45 08894424 04c70424 U......E..D$...$

0010 00000000 e8fcffff ffc74424 04000000 ..........D$....

0020 00890424 e8fcffff ffc9c355 89e583e4 ...$.......U....

0030 f083ec20 c7442418 01000000 8b150400 ... .D$.........

0040 0000a108 00000001 c28b4424 1801c28b ..........D$....

0050 44241c01 d0890424 e8fcffff ff8b4424 D$.....$......D$

0060 18c9c355 89e583ec 18837d08 01753181 ...U......}..u1.

0070 7d0cffff 00007528 c7042404 000000e8 }.....u(..$.....

0080 fcffffff c7442408 00000000 c7442404 .....D$......D$.

0090 04000000 c7042400 000000e8 fcffffff ......$.........

00a0 c9c35589 e583ec18 c7442404 ffff0000 ..U......D$.....

00b0 c7042401 000000e8 a7ffffff c9c3 ..$...........

Contents of section .data:

0000 54000000 55000000 T...U...

Contents of section .init_array:

0000 a2000000 ....

Contents of section .comment:

0000 00474343 3a202855 62756e74 7520342e .GCC: (Ubuntu 4.

0010 382e342d 32756275 6e747531 7e31342e 8.4-2ubuntu1~14.

0020 30342e33 2920342e 382e3400 04.3) 4.8.4.

Contents of section .eh_frame:

0000 14000000 00000000 017a5200 017c0801 .........zR..|..

0010 1b0c0404 88010000 1c000000 1c000000 ................

0020 00000000 2b000000 00410e08 8502420d ....+....A....B.

0030 0567c50c 04040000 1c000000 3c000000 .g..........<...>

0040 2b000000 38000000 00410e08 8502420d +...8....A....B.

0050 0574c50c 04040000 1c000000 5c000000 .t..........\...

0060 63000000 3f000000 00410e08 8502420d c...?....A....B.

0070 057bc50c 04040000 1c000000 7c000000 .{..........|...

0080 a2000000 1c000000 00410e08 8502420d .........A....B.

0090 0558c50c 04040000 .X......

Disassembly of section .text:

00000000 <_z5func1i>:

0:55 push %ebp

1:89 e5 mov %esp,%ebp

3:83 ec 18 sub $0x18,%esp

6:8b 45 08 mov 0x8(%ebp),%eax

9:89 44 24 04 mov %eax,0x4(%esp)

d:c7 04 24 00 00 00 00 movl $0x0,(%esp)

14:e8 fc ff ff ff call 15 <_z5func1i>

19:c7 44 24 04 00 00 00 movl $0x0,0x4(%esp)

20:00

21:89 04 24 mov %eax,(%esp)

24:e8 fc ff ff ff call 25 <_z5func1i>

29:c9 leave

2a:c3 ret

0000002b :

2b:55 push %ebp

2c:89 e5 mov %esp,%ebp

2e:83 e4 f0 and $0xfffffff0,%esp

31:83 ec 20 sub $0x20,%esp

34:c7 44 24 18 01 00 00 movl $0x1,0x18(%esp)

3b:00

3c:8b 15 04 00 00 00 mov 0x4,%edx

42:a1 08 00 00 00 mov 0x8,%eax

47:01 c2 add %eax,%edx

49:8b 44 24 18 mov 0x18(%esp),%eax

4d:01 c2 add %eax,%edx

4f:8b 44 24 1c mov 0x1c(%esp),%eax

53:01 d0 add %edx,%eax

55:89 04 24 mov %eax,(%esp)

58:e8 fc ff ff ff call 59

5d:8b 44 24 18 mov 0x18(%esp),%eax

61:c9 leave

62:c3 ret

00000063 <_z41__static_initialization_and_destruction_0ii>:

63:55 push %ebp

64:89 e5 mov %esp,%ebp

66:83 ec 18 sub $0x18,%esp

69:83 7d 08 01 cmpl $0x1,0x8(%ebp)

6d:75 31 jne a0 <_z41__static_initialization_and_destruction_0ii>

6f:81 7d 0c ff ff 00 00 cmpl $0xffff,0xc(%ebp)

76:75 28 jne a0 <_z41__static_initialization_and_destruction_0ii>

78:c7 04 24 04 00 00 00 movl $0x4,(%esp)

7f:e8 fc ff ff ff call 80 <_z41__static_initialization_and_destruction_0ii>

84:c7 44 24 08 00 00 00 movl $0x0,0x8(%esp)

8b:00

8c:c7 44 24 04 04 00 00 movl $0x4,0x4(%esp)

93:00

94:c7 04 24 00 00 00 00 movl $0x0,(%esp)

9b:e8 fc ff ff ff call 9c <_z41__static_initialization_and_destruction_0ii>

a0:c9 leave

a1:c3 ret

000000a2 <_global__sub_i_global_ini_var>:

a2:55 push %ebp

a3:89 e5 mov %esp,%ebp

a5:83 ec 18 sub $0x18,%esp

a8:c7 44 24 04 ff ff 00 movl $0xffff,0x4(%esp)

af:00

b0:c7 04 24 01 00 00 00 movl $0x1,(%esp)

b7:e8 a7 ff ff ff call 63 <_z41__static_initialization_and_destruction_0ii>

bc:c9 leave

bd:c3 ret

自定义段：

__attribute__( ( section( "FOO" ) ) ) int global = 42;

__attribute__( ( section( "BAR" ) ) ) void foo()

{

}