手撕编译原理：汇编语言不会编

NASM

Installation of NASM

wget -c http://www.nasm.us/pub/nasm/releasebuilds/2.14.02/nasm-2.14.02.tar.xz
tar -xf nasm-2.14.02.tar.xz
cd nasm-2.14.02
ls
./configure --prefix=/usr && make
sudo make install

version

$ nasm -v
NASM version 2.14.02 compiled on Jun  1 2020

demo

global main

main:
    mov eax, 5
    ret

Makefile

SRCS := first.asm
OBJS := $(SRCS:.asm=.o)
TARGET := $(OBJS:.o=)

all: $(TARGET)

$(TARGET): $(OBJS)
        gcc -m32 $< -o $@

%.o: %.asm
        nasm -f elf $< -o $@

.PHONY: all clean

clean:
        -rm -rf $(TARGET) $(OBJS)

make

$ make
gcc -m32 first.o -o first
/usr/bin/ld: cannot find crt1.o: No such file or directory
/usr/bin/ld: cannot find crti.o: No such file or directory
/usr/bin/ld: cannot find /lib/libgcc_s.so.1
collect2: error: ld returned 1 exit status
make: *** [Makefile:7: first] Error 1

glibc-devel

sudo yum install glibc-devel.i686 -y
sudo yum install libgcc.i686 -y

make

nasm -f elf first.asm -o first.o
gcc -m32 first.o -o first

查看返回值

# ./first ; echo $?
./first
echo $?

运算

add.asm

global main

main:
        mov eax, 15
        mov ebx, 19
        add eax, ebx
        ret

gdb

$ gdb add
(gdb) set disassembly-flavor intel
(gdb) disassemble main
Dump of assembler code for function main:
   0x08048480 <+0>:     mov    eax,0xf
   0x08048485 <+5>:     mov    ebx,0x13
   0x0804848a <+10>:    add    eax,ebx
   0x0804848c <+12>:    ret
   0x0804848d <+13>:    xchg   ax,ax
   0x0804848f <+15>:    nop
End of assembler dump.
(gdb) b *0x08048485# break *0x08048485
(gdb) r # run
(gdb) i r eax # info registers eax
eax            0xf                 15
(gdb) si # stepi Step one instruction exactly.
(gdb) info registers ebx
ebx            0x13                19
...
(gdb) i r # info registers
eax            0x22                34
ecx            0xbddbc2e           199081006
edx            0xffffd514          -10988
ebx            0x13                19
esp            0xffffd4ec          0xffffd4ec
ebp            0x0                 0x0
esi            0xf7fc0000          -134479872
edi            0x0                 0
eip            0x804848c           0x804848c <main+12>
eflags         0x216               [ PF AF IF ]
cs             0x23                35
ss             0x2b                43
ds             0x2b                43
es             0x2b                43
fs             0x0                 0
gs             0x63                99

...
(gdb) c # continue
Continuing.
[Inferior 1 (process 21718) exited with code 042]
(bdb) q

跳转

cmp & jmp

后缀	意义
a	above
e	equal
b	below
n	not
g	greater
l	lower
s	signed
z	zero

标志寄存器 eflags

eip : CPU 接下来要执行的代码

循环

for, while, do-while ==> goto + if

• goto: jmp
• if: cmp & jmp

函数

• call + ret

• 保存现场

• 传递参数

• 返回

x86 的环境下，栈是朝着低地址的方向伸长的。

• 栈: call 指令保存 eip

• esp(stack pointer): 寄存器

• 汇编中发生函数调用相关的指令call和ret

• call指令会产生跳转动作，与jmp不同的是，call之后可以通过ret指令跳回来
  call和ret的配合是依靠保存eip的值到栈里，返回时恢复eip实现的
  esp记录着当前栈顶所在的位置，每次call和ret执行都会伴随着入栈和出栈，也就是esp会发生变化

递归

• push + pop

函数调用相关指令

1. 通过寄存器传递参数和返回值
2. 函数调用后的返回地址会保存到堆栈中
3. 函数的局部状态也可以保存到堆栈中

明天手撕《汇编语言》第3版。

参考：

1. https://www.nasm.us/
2. http://www.linuxfromscratch.org/blfs/view/svn/general/nasm.html
3. https://zhuanlan.zhihu.com/c_144694924