在ubuntu下用nasm和gcc的ld链接程…

我们先写一个.asm文件

[SECTION .data]

StrHello db "Hello, kaito!", 0Ah

HelloLen equ $ - StrHello

[SECTION .text]

global _start ;must be decleared for linker(ld)

_start: ;tell linker entry point

mov edx, HelloLen ;message length

mov ecx, StrHello ;message to write

mov ebx, 1 ;file descriptor(stdout, standard output)

mov eax, 4 ;system call number(sys_write)

int 0x80 ;call kernel

mov ebx, 0

mov eax, 1 ;system call number(sys_exit)

int 0x80 ;call kernel

在nasm中可以指定输出文件格式，其中有如下

valid output formats for -f are (`*' denotes default): 
  * bin       flat-form binary files (e.g. DOS .COM, .SYS) 
    ith       Intel hex 
    srec      Motorola S-records 
    aout      Linux a.out object files 
    aoutb     NetBSD/FreeBSD a.out object files 
    coff      COFF (i386) object files (e.g. DJGPP for DOS) 
    elf32     ELF32 (i386) object files (e.g. Linux) 
    elf64     ELF64 (x86_64) object files (e.g. Linux) 
    as86      Linux as86 (bin86 version 0.3) object files 
    obj       MS-DOS 16-bit/32-bit OMF object files 
    win32     Microsoft Win32 (i386) object files 
    win64     Microsoft Win64 (x86-64) object files 
    rdf       Relocatable Dynamic Object File Format v2.0 
    ieee      IEEE-695 (LADsoft variant) object file format 
    macho32   NeXTstep/OpenStep/Rhapsody/Darwin/MacOS X (i386) object files 
    macho64   NeXTstep/OpenStep/Rhapsody/Darwin/MacOS X (x86_64) object files 
    dbg       Trace of all info passed to output stage 
    elf       ELF (short name for ELF32) 
    macho     MACHO (short name for MACHO32) 
    win       WIN (short name for WIN32) 

由于我的虚拟机是i386:x86-64的，所以在编译的时候要选择elf64

nasm -f elf64 hello.asm -o hello.o

ld -s hello.o -o hello

运行hello文件

./hello

如果想要在64位Linux上编译出32位.o文件，则在编译时加上编译选项   -melf_i386