Labels and Entry Point ( start)

Labels and Entry Point (_start)

This page explains how labels work in assembly programming and how the program’s entry point is defined.

What are labels

A label is a symbolic name for a memory address. It marks a location in code or data that can be referenced later.

Syntax:

LabelName:

Example:

SECTION .data
EatMsg: db "Eat at Joe’s"

SECTION .text
mov rcx, EatMsg   ; rcx ← address of the string
mov rdx, [EatMsg] ; rdx ← first 8 bytes of the string

Here:

• EatMsg refers to the address of the string.
• [EatMsg] refers to the contents at that address.

Why labels are needed

Labels replace hardcoded addresses and make programs relocatable:

• You don’t know where in memory your program will be loaded.
• The linker can adjust label references automatically.
• The code becomes modular and position-independent.

Code labels

Labels can also be attached to code lines. They mark destinations for jumps and branches, forming the basis for control flow.

Example:

mov eax, 5

loop_start:

dec eax
jnz loop_start   ; jump back if eax ≠ 0

Here, loop_start marks the instruction to jump to. Such code labels are essential for loops, branches, and function calls.

The _start label

_start: defines the entry point of an assembly program on Linux. It is the first instruction executed when the program runs.

Example:

global _start

_start:

mov rax, 60      ; system call for exit
xor rdi, rdi     ; return code 0
syscall

• The keyword global makes _start visible to the linker.
• The operating system looks for this symbol to know where to begin execution.

Every NASM program for Linux must define exactly one entry point, usually called _start.

Summary

• Labels are symbolic names for addresses.
• Data labels point to memory locations.
• Code labels mark positions for jumps.
• The _start label defines where execution begins.