Describe the different stages of the assembly process for a two-pass assembler

4.2 Assembly Language – The Two‑Pass Assembler

Think of the assembler as a translator that turns your human‑readable assembly code into the machine’s binary language. A two‑pass assembler does this in two stages, just like a detective first gathers clues and then solves the mystery.

Pass 1 – Building the Symbol Table

1. Read the source line by line. For each line, record the address where the instruction will live. This is like noting the page number for each clue.
2. When you see a label (e.g., LOOP:), store it in the symbol table with its current address.
3. Ignore forward references. If a label is used before it’s defined, just remember that it’s a placeholder – the real address will come later.
4. Calculate instruction sizes. Each instruction may be 1, 2, or 3 bytes. Knowing the size lets you keep the address counter correct.

At the end of Pass 1 you have a complete symbol table:

Pass 2 – Generating Machine Code

1. Reset the address counter. Start again at the beginning of the source.
2. Translate each instruction. Replace mnemonics with opcodes and operands with binary values.
3. Resolve labels. Look up each label in the symbol table to get its real address.
4. Output the machine code. Write the binary (or hex) representation to the object file.

Now the assembler has produced a binary file that the CPU can execute. 🎉

Analogy: Building a LEGO City

Imagine you’re building a LEGO city:

• Pass 1 is like planning the city map. You decide where each building (label) will go and note its coordinates (address).
• Pass 2 is the actual construction. You follow the map, placing each building at the right spot.

If you forget a building’s location in Pass 1, you’ll have to rebuild it in Pass 2 – just like a missing label causes a problem.

Exam Tips – What the Mark‑scheme Looks For

Quick Quiz – Test Your Knowledge

1️⃣ If an instruction at address $0004$ is 2 bytes long, what will be the address of the next instruction?

Answer: $0006$ (because $0004 + 2 = 0006$).

2️⃣ Why does a two‑pass assembler need to read the source twice?

Answer: The first pass collects all labels and their addresses; the second pass can then replace label references with actual addresses.

Keep practicing, and soon you’ll be assembling code faster than a cheetah on a treadmill! 🏃‍♂️💨