编译阶段翻译成平台汇编代码，是在预编译阶段上加码，将C/C++代码翻译成平台相关的汇编代码。
示例：
1）用户程序
/*@brief test demo-for-compile-to-asm? show you here@author wen`xuanpei@email 15873152445@163.com(query for any question here)
*/
#include <stdio.h>int main(int argc, char **argv){if(argc == 1){puts("hello world-0!\n");}else {--argc;switch(argc){default:break;case 1:/*dead-loop-1*/puts("hello world-1!\n");for(;;){;}break;case 2:/*dead-loop-2*/puts("hello world-2!\n");while(1){;}break;case 3:/*dead-loop-3*/puts("hello world-3!\n");do{;}while(1);break;case 4:/*dead-loop-4*/puts("hello world-4!\n");__dead_loop:{;}goto __dead_loop;break;}}return 0;
}2）编译
gcc test.c -S
g++ test.c -S3）翻译成平台相关的汇编代码
.file "test.c".section .rodata
.LC0:.string "hello world-0!\n"
.LC1:.string "hello world-1!\n"
.LC2:.string "hello world-2!\n"
.LC3:.string "hello world-3!\n"
.LC4:.string "hello world-4!\n".text.globl main.type main, @function
main:
.LFB0:.cfi_startprocpushq %rbp.cfi_def_cfa_offset 16.cfi_offset 6, -16movq %rsp, %rbp.cfi_def_cfa_register 6subq $16, %rspmovl %edi, -4(%rbp)movq %rsi, -16(%rbp)cmpl $1, -4(%rbp)jne .L2movl $.LC0, %edicall putsjmp .L3
.L2:subl $1, -4(%rbp)movl -4(%rbp), %eaxcmpl $2, %eaxje .L5cmpl $2, %eaxjg .L6cmpl $1, %eaxje .L7jmp .L3
.L6:cmpl $3, %eaxje .L8cmpl $4, %eaxje .L9jmp .L3
.L7:movl $.LC1, %edicall puts
.L10:jmp .L10
.L5:movl $.LC2, %edicall puts
.L11:jmp .L11
.L8:movl $.LC3, %edicall puts
.L12:jmp .L12
.L9:movl $.LC4, %edicall puts
.L13:jmp .L13
.L3:movl $0, %eaxleave.cfi_def_cfa 7, 8ret.cfi_endproc
.LFE0:.size main, .-main.ident "GCC: (GNU) 4.8.5 20150623 (Red Hat 4.8.5-44)".section .note.GNU-stack,"",@progbits
尾声：如果不知道预编译，可以参考下列文章
1）gcc/g++：预编译阶段宏定义-CSDN博客
2）gcc/g++：预编译阶段取消宏定义-CSDN博客
3）gcc/g++：预编译阶段查看模块生成目标的层级依赖-CSDN博客
4）gcc/g++：预编译阶段查看模块生成目标的直接依赖-CSDN博客
5）gcc/g++：预编译阶段查看层级依赖可用宏列表-CSDN博客
6）gcc/g++：预编译阶段嵌入头文件并完成替换-CSDN博客
7）gcc/g++：预编译阶段寻找头文件-CSDN博客
