NASM Syntax

NASM, or the Netwide Assembler, is the assembler that will be used to convert your assembly language code to binary executables. Below is a description of a few NASM features you will find useful. For more detailed information on these or other NASM features see the NASM Manual (852 KB, requires Acrobat Reader; use "acroread" in UNIX and Linux).

Layout of a NASM Source Line

In general, each line of assembly source code has the following form:

label:    instruction   operands      ; comment

For example you may have the following lines in a piece of assembly:

MyFunction:
	add	ax, bx		; add bx to ax

As this example shows, labels do not have to be typed on the same line as an instruction. The address of the next instruction will be assumed for the label. Also, with NASM, only the labels are case sensitive.

Numbers

When entering numbers in an assembly program, the following forms are allowed:

decimal:      100
hexadecimal:  0a2h  (The 0 is needed before numbers beginning with letter)
hexadecimal:  0xa2
octal:        777q
binary:       10010011b

Declaring Data

To declare data in an assembly program the db, dw, and dd pseudo-instructions are used:

	db	value		; Allocate a byte sized value
	dw	value		; Allocate a word sized value
	dd	value		; Allocate a dword sized value

For example, to allocate a word sized value initialized to the value 3 that can be referred to by the name MyValue, one could use the following lines of assembly:

MyValue:
	dw	3

This value could then be accessed with instructions like:

	mov	ax, [MyValue]		; Load MyValue into ax
	mov	[MyValue], 2		; Store 2 into MyValue

The times pseudo-instruction causes the assembler to repeat an instruction or pseudo-instruction. Its syntax is:

	times	numRep instruction operands

For example, to declare 20 bytes of data, initialized to 0, you could use the following line of assembly:

	times	20	db	0

Strings can be declared in several ways. The following are valid NASM examples:

	db	"abcd",0				; Same as "abcd" in C (with NULL termination)
	db	'a',"b",'cd',0				; Same as previous
	db	'"quotes"',"and 'apostrophes'.",0	; The string: "quotes" and 'apostrophes'.

Other NASM Directives

Other useful NASM directives include the org and the align directives. Org allows you to specify the program origin (i.e., the address offset at which you intend the program to be loaded). If not given, NASM assumes that the origin is 0. This information is needed in order to convert label names to address offsets during assembly. For example, to set the origin to 0x100 one could place the following line at the beginning of an assembly file:

	org	0x100

The align directive allows programmers to align their code to word, dword, or larger boundaries in memory. To align to a word boundary, the following line of assembly could be used:

	align	2	; Align to nearest 2-byte boundary

This will cause an unused byte to be inserted if the address of the next instruction or data would have been odd. The parameter given to align must be a power of 2. Code and data alignment are important in ensuring memory performance.