C语言全局数组位于其反汇编代码中的哪个位置
[英]where is the C language global array located in its disassembly code
问题描述
我写了一个非常简单的C代码如下：
int data_items[] = {3,67,32,4,89,6,34,2,9,0};
int max(int* pt)
{
int val = *pt;
while(*pt != 0)
{
if (*pt > val)
{
val = *pt;
}
++pt;
}
return val;
}
int main()
{
max(data_items);
return 0;
}
然后我用gcc编译它：
gcc main.c
然后将其拆解为：
objdump -d a.out
最后我得到了汇编代码：
a.out: file format elf64-x86-64
Disassembly of section .init:
00000000004003a8 <_init>:
4003a8: 48 83 ec 08 sub $0x8,%rsp
4003ac: 48 8b 05 45 0c 20 00 mov 0x200c45(%rip),%rax # 600ff8 <_DYNAMIC+0x1d0>
4003b3: 48 85 c0 test %rax,%rax
4003b6: 74 05 je 4003bd <_init+0x15>
4003b8: e8 33 00 00 00 callq 4003f0 <__gmon_start__@plt>
4003bd: 48 83 c4 08 add $0x8,%rsp
4003c1: c3 retq
Disassembly of section .plt:
00000000004003d0 <__libc_start_main@plt-0x10>:
4003d0: ff 35 32 0c 20 00 pushq 0x200c32(%rip) # 601008 <_GLOBAL_OFFSET_TABLE_+0x8>
4003d6: ff 25 34 0c 20 00 jmpq *0x200c34(%rip) # 601010 <_GLOBAL_OFFSET_TABLE_+0x10>
4003dc: 0f 1f 40 00 nopl 0x0(%rax)
00000000004003e0 <__libc_start_main@plt>:
4003e0: ff 25 32 0c 20 00 jmpq *0x200c32(%rip) # 601018 <_GLOBAL_OFFSET_TABLE_+0x18>
4003e6: 68 00 00 00 00 pushq $0x0
4003eb: e9 e0 ff ff ff jmpq 4003d0 <_init+0x28>
00000000004003f0 <__gmon_start__@plt>:
4003f0: ff 25 2a 0c 20 00 jmpq *0x200c2a(%rip) # 601020 <_GLOBAL_OFFSET_TABLE_+0x20>
4003f6: 68 01 00 00 00 pushq $0x1
4003fb: e9 d0 ff ff ff jmpq 4003d0 <_init+0x28>
Disassembly of section .text:
0000000000400400 <_start>:
400400: 31 ed xor %ebp,%ebp
400402: 49 89 d1 mov %rdx,%r9
400405: 5e pop %rsi
400406: 48 89 e2 mov %rsp,%rdx
400409: 48 83 e4 f0 and $0xfffffffffffffff0,%rsp
40040d: 50 push %rax
40040e: 54 push %rsp
40040f: 49 c7 c0 b0 05 40 00 mov $0x4005b0,%r8
400416: 48 c7 c1 40 05 40 00 mov $0x400540,%rcx
40041d: 48 c7 c7 28 05 40 00 mov $0x400528,%rdi
400424: e8 b7 ff ff ff callq 4003e0 <__libc_start_main@plt>
400429: f4 hlt
40042a: 66 0f 1f 44 00 00 nopw 0x0(%rax,%rax,1)
0000000000400430 <deregister_tm_clones>:
400430: b8 8f 10 60 00 mov $0x60108f,%eax
400435: 55 push %rbp
400436: 48 2d 88 10 60 00 sub $0x601088,%rax
40043c: 48 83 f8 0e cmp $0xe,%rax
400440: 48 89 e5 mov %rsp,%rbp
400443: 77 02 ja 400447 <deregister_tm_clones+0x17>
400445: 5d pop %rbp
400446: c3 retq
400447: b8 00 00 00 00 mov $0x0,%eax
40044c: 48 85 c0 test %rax,%rax
40044f: 74 f4 je 400445 <deregister_tm_clones+0x15>
400451: 5d pop %rbp
400452: bf 88 10 60 00 mov $0x601088,%edi
400457: ff e0 jmpq *%rax
400459: 0f 1f 80 00 00 00 00 nopl 0x0(%rax)
0000000000400460 <register_tm_clones>:
400460: b8 88 10 60 00 mov $0x601088,%eax
400465: 55 push %rbp
400466: 48 2d 88 10 60 00 sub $0x601088,%rax
40046c: 48 c1 f8 03 sar $0x3,%rax
400470: 48 89 e5 mov %rsp,%rbp
400473: 48 89 c2 mov %rax,%rdx
400476: 48 c1 ea 3f shr $0x3f,%rdx
40047a: 48 01 d0 add %rdx,%rax
40047d: 48 d1 f8 sar %rax
400480: 75 02 jne 400484 <register_tm_clones+0x24>
400482: 5d pop %rbp
400483: c3 retq
400484: ba 00 00 00 00 mov $0x0,%edx
400489: 48 85 d2 test %rdx,%rdx
40048c: 74 f4 je 400482 <register_tm_clones+0x22>
40048e: 5d pop %rbp
40048f: 48 89 c6 mov %rax,%rsi
400492: bf 88 10 60 00 mov $0x601088,%edi
400497: ff e2 jmpq *%rdx
400499: 0f 1f 80 00 00 00 00 nopl 0x0(%rax)
00000000004004a0 <__do_global_dtors_aux>:
4004a0: 80 3d e1 0b 20 00 00 cmpb $0x0,0x200be1(%rip) # 601088 <__TMC_END__>
4004a7: 75 11 jne 4004ba <__do_global_dtors_aux+0x1a>
4004a9: 55 push %rbp
4004aa: 48 89 e5 mov %rsp,%rbp
4004ad: e8 7e ff ff ff callq 400430 <deregister_tm_clones>
4004b2: 5d pop %rbp
4004b3: c6 05 ce 0b 20 00 01 movb $0x1,0x200bce(%rip) # 601088 <__TMC_END__>
4004ba: f3 c3 repz retq
4004bc: 0f 1f 40 00 nopl 0x0(%rax)
00000000004004c0 <frame_dummy>:
4004c0: 48 83 3d 58 09 20 00 cmpq $0x0,0x200958(%rip) # 600e20 <__JCR_END__>
4004c7: 00
4004c8: 74 1e je 4004e8 <frame_dummy+0x28>
4004ca: b8 00 00 00 00 mov $0x0,%eax
4004cf: 48 85 c0 test %rax,%rax
4004d2: 74 14 je 4004e8 <frame_dummy+0x28>
4004d4: 55 push %rbp
4004d5: bf 20 0e 60 00 mov $0x600e20,%edi
4004da: 48 89 e5 mov %rsp,%rbp
4004dd: ff d0 callq *%rax
4004df: 5d pop %rbp
4004e0: e9 7b ff ff ff jmpq 400460 <register_tm_clones>
4004e5: 0f 1f 00 nopl (%rax)
4004e8: e9 73 ff ff ff jmpq 400460 <register_tm_clones>
00000000004004ed <max>:
4004ed: 55 push %rbp
4004ee: 48 89 e5 mov %rsp,%rbp
4004f1: 48 89 7d e8 mov %rdi,-0x18(%rbp)
4004f5: 48 8b 45 e8 mov -0x18(%rbp),%rax
4004f9: 8b 00 mov (%rax),%eax
4004fb: 89 45 fc mov %eax,-0x4(%rbp)
4004fe: eb 19 jmp 400519 <max+0x2c>
400500: 48 8b 45 e8 mov -0x18(%rbp),%rax
400504: 8b 00 mov (%rax),%eax
400506: 3b 45 fc cmp -0x4(%rbp),%eax
400509: 7e 09 jle 400514 <max+0x27>
40050b: 48 8b 45 e8 mov -0x18(%rbp),%rax
40050f: 8b 00 mov (%rax),%eax
400511: 89 45 fc mov %eax,-0x4(%rbp)
400514: 48 83 45 e8 04 addq $0x4,-0x18(%rbp)
400519: 48 8b 45 e8 mov -0x18(%rbp),%rax
40051d: 8b 00 mov (%rax),%eax
40051f: 85 c0 test %eax,%eax
400521: 75 dd jne 400500 <max+0x13>
400523: 8b 45 fc mov -0x4(%rbp),%eax
400526: 5d pop %rbp
400527: c3 retq
0000000000400528 <main>:
400528: 55 push %rbp
400529: 48 89 e5 mov %rsp,%rbp
40052c: bf 60 10 60 00 mov $0x601060,%edi
400531: e8 b7 ff ff ff callq 4004ed <max>
400536: b8 00 00 00 00 mov $0x0,%eax
40053b: 5d pop %rbp
40053c: c3 retq
40053d: 0f 1f 00 nopl (%rax)
0000000000400540 <__libc_csu_init>:
400540: 41 57 push %r15
400542: 41 89 ff mov %edi,%r15d
400545: 41 56 push %r14
400547: 49 89 f6 mov %rsi,%r14
40054a: 41 55 push %r13
40054c: 49 89 d5 mov %rdx,%r13
40054f: 41 54 push %r12
400551: 4c 8d 25 b8 08 20 00 lea 0x2008b8(%rip),%r12 # 600e10 <__frame_dummy_init_array_entry>
400558: 55 push %rbp
400559: 48 8d 2d b8 08 20 00 lea 0x2008b8(%rip),%rbp # 600e18 <__init_array_end>
400560: 53 push %rbx
400561: 4c 29 e5 sub %r12,%rbp
400564: 31 db xor %ebx,%ebx
400566: 48 c1 fd 03 sar $0x3,%rbp
40056a: 48 83 ec 08 sub $0x8,%rsp
40056e: e8 35 fe ff ff callq 4003a8 <_init>
400573: 48 85 ed test %rbp,%rbp
400576: 74 1e je 400596 <__libc_csu_init+0x56>
400578: 0f 1f 84 00 00 00 00 nopl 0x0(%rax,%rax,1)
40057f: 00
400580: 4c 89 ea mov %r13,%rdx
400583: 4c 89 f6 mov %r14,%rsi
400586: 44 89 ff mov %r15d,%edi
400589: 41 ff 14 dc callq *(%r12,%rbx,8)
40058d: 48 83 c3 01 add $0x1,%rbx
400591: 48 39 eb cmp %rbp,%rbx
400594: 75 ea jne 400580 <__libc_csu_init+0x40>
400596: 48 83 c4 08 add $0x8,%rsp
40059a: 5b pop %rbx
40059b: 5d pop %rbp
40059c: 41 5c pop %r12
40059e: 41 5d pop %r13
4005a0: 41 5e pop %r14
4005a2: 41 5f pop %r15
4005a4: c3 retq
4005a5: 66 66 2e 0f 1f 84 00 data32 nopw %cs:0x0(%rax,%rax,1)
4005ac: 00 00 00 00
00000000004005b0 <__libc_csu_fini>:
4005b0: f3 c3 repz retq
Disassembly of section .fini:
00000000004005b4 <_fini>:
4005b4: 48 83 ec 08 sub $0x8,%rsp
4005b8: 48 83 c4 08 add $0x8,%rsp
4005bc: c3 retq
我的问题是，在主要功能中：
40052c: bf 60 10 60 00 mov $0x601060,%edi
地址$ 0x601060应该是C代码中数组data_items的地址。 但我在汇编代码中找不到data_items数组值，它在哪里？
顺便说一下，在max函数代码中：
4004f1: 48 89 7d e8 mov %rdi,-0x18(%rbp)
我认为堆栈应该是：
mov %rdi, -0x8(%rbp)
为什么编译器会在堆栈中产生一些漏洞？
我的系统是Ubuntu 14.04 LTS。
4 个解决方案
解决方案1
7 已采纳 2014-10-08 08:11:56
objdump -d只拆解可执行文件中应该有代码的部分。
使用objdump -D来反汇编所有部分，你会发现：
Disassembly of section .data:
08049760 <__data_start>:
...
08049780 <data_items>:
8049780: 03 00 add (%eax),%eax
8049782: 00 00 add %al,(%eax)
8049784: 43 inc %ebx
8049785: 00 00 add %al,(%eax)
8049787: 00 20 add %ah,(%eax)
8049789: 00 00 add %al,(%eax)
804978b: 00 04 00 add %al,(%eax,%eax,1)
804978e: 00 00 add %al,(%eax)
那是你的阵容。 objdump将尝试将其拆解为好像是机器码，因此组装不会有意义。
解决方案2
2 2014-10-08 08:17:27
如果运行nm a.out ，您将看到0x601060是data_items的地址。
解决方案3
1 2014-10-08 09:14:33
说到运行时dissasembly gdb本身可以非常方便：
(gdb) break main
Breakpoint 1 at 0x4004f3
(gdb) break max
Breakpoint 2 at 0x4004b8
(gdb) run
Starting program: /home/dtarcatu/workspace/ctest/a.out
Breakpoint 1, 0x00000000004004f3 in main ()
(gdb) disas
Dump of assembler code for function main:
0x00000000004004ef <+0>: push %rbp
0x00000000004004f0 <+1>: mov %rsp,%rbp
=> 0x00000000004004f3 <+4>: mov $0x601040,%edi
0x00000000004004f8 <+9>: callq 0x4004b4 <max>
0x00000000004004fd <+14>: mov $0x0,%eax
0x0000000000400502 <+19>: pop %rbp
0x0000000000400503 <+20>: retq
End of assembler dump.
(gdb) x 0x601040
0x601040 <data_items>: 0x00000003
(gdb) x /10d 0x601040
0x601040 <data_items>: 3 67 32 4
0x601050 <data_items+16>: 89 6 34 2
0x601060 <data_items+32>: 9 0
现在关于你的堆栈规则dillema我不确定我能得到一个非常好的答案，但我认为这与x86-64 红区优化有关 。
(gdb) c
Continuing.
Breakpoint 2, 0x00000000004004b8 in max ()
(gdb) disas
Dump of assembler code for function max:
0x00000000004004b4 <+0>: push %rbp
0x00000000004004b5 <+1>: mov %rsp,%rbp
=> 0x00000000004004b8 <+4>: mov %rdi,-0x18(%rbp)
0x00000000004004bc <+8>: mov -0x18(%rbp),%rax
0x00000000004004c0 <+12>: mov (%rax),%eax
0x00000000004004c2 <+14>: mov %eax,-0x4(%rbp)
0x00000000004004c5 <+17>: jmp 0x4004e0 <max+44>
0x00000000004004c7 <+19>: mov -0x18(%rbp),%rax
0x00000000004004cb <+23>: mov (%rax),%eax
0x00000000004004cd <+25>: cmp -0x4(%rbp),%eax
0x00000000004004d0 <+28>: jle 0x4004db <max+39>
0x00000000004004d2 <+30>: mov -0x18(%rbp),%rax
0x00000000004004d6 <+34>: mov (%rax),%eax
0x00000000004004d8 <+36>: mov %eax,-0x4(%rbp)
0x00000000004004db <+39>: addq $0x4,-0x18(%rbp)
0x00000000004004e0 <+44>: mov -0x18(%rbp),%rax
0x00000000004004e4 <+48>: mov (%rax),%eax
0x00000000004004e6 <+50>: test %eax,%eax
0x00000000004004e8 <+52>: jne 0x4004c7 <max+19>
0x00000000004004ea <+54>: mov -0x4(%rbp),%eax
0x00000000004004ed <+57>: pop %rbp
0x00000000004004ee <+58>: retq
End of assembler dump.
(gdb) ni
0x00000000004004bc in max ()
(gdb) print $rbp
$1 = (void *) 0x7fffffffdf10
(gdb) print $rsp
$2 = (void *) 0x7fffffffdf10
-0x18(%rbp)位置绝对位于红色区域内。 我不确定编译器是如何使用它的，但它可以做任何它想要的黑客。 您的本地int似乎稍后存储在-0x4(%rbp) ，所以我猜-0x18(%rbp)只是某种临时缓冲区。
解决方案4
0 2014-10-09 07:10:18
您还可以要求编译器输出其发出的汇编代码。 请注意，程序中没有可观察到的副作用，因此编译器可以将其优化为nop - 或者在编译时计算最大值等。
但是，使用（GCC 4.9.1 Debian / x86-64）编译代码
gcc -fverbose-asm -O -S main.c
我得到一个文件main.s包含（前后几行）：
.globl max
.type max, @function
max:
.LFB0:
.file 1 "main.c"
.loc 1 4 0
.cfi_startproc
.LVL0:
.loc 1 5 0
movl (%rdi), %eax # *pt_5(D), val
.LVL1:
.loc 1 6 0
testl %eax, %eax # val
je .L2 #,
movl %eax, %edx # val, val
.L3:
cmpl %edx, %eax # val, val
cmovl %edx, %eax # val,, val, val
.LVL2:
.loc 1 12 0
addq $4, %rdi #, pt
.LVL3:
.loc 1 6 0
movl (%rdi), %edx # MEM[base: pt_8, offset: 0B], val
testl %edx, %edx # val
jne .L3 #,
.L2:
.loc 1 15 0
rep ret
.cfi_endproc
.LFE0:
.size max, .-max
.globl main
.type main, @function
main:
.LFB1:
.loc 1 18 0
.cfi_startproc
.loc 1 20 0
movl $data_items, %edi #,
call max #
.LVL4:
.loc 1 23 0
movl $0, %eax #,
ret
.cfi_endproc
.LFE1:
.size main, .-main
.globl data_items
.data
.align 32
.type data_items, @object
.size data_items, 40
data_items:
.long 3
.long 67
.long 32
.long 4
.long 89
.long 6
.long 34
.long 2
.long 9
.long 0
.text
所以你看到data_items进入数据部分（因为.data指令）
顺便说一句，通常GCC发射由所述转化的一个汇编文件as汇编成ELF文件main.o包含目标代码 。 该目标文件包含重定位指令和几个稍后由ld 链接器处理的部分。
如何创建一个包含 100 个整数的数组，其中数组的每个元素都是它在 c 语言中的索引
[英]How to create an array of 100 integers where every element of array is its index in c language
C 语言中 memory 分配的本地和全局数组有什么区别？
[英]What is difference for local and global array for memory allocation in C language?
声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.