A simple script containing a bootable header.
This is (almost) the minimum example of a (useless) x86 NASM boot loader. It does nothing more than define a header, which makes it bootable. For BIOS to identify the script as being bootable it has two requirements.
- The script must be a total of 512 bytes
- The last two bytes must be the (magic) hexadecimal number
0xaa00
The above requirements are met by lines 6 and 7 in the script. Lines 1 and 2 are optional, they are so called "assembler directives", which gives hints to the assembler, that it can adhere to if it wants to. (see comments in the file).
x86 uses something called "byte addressing", which means that the least (meaningful) common denominator as far as memory allocation is concerned is a byte, which is 8 bits large. A chunk of memory of that size is referred to as just that, a byte. Everything above that is referred to as a variation of a word. read more in volume 1, section 4.1 in the intel x86 manual
NASM, much like other assembly languages, provides syntax to define these memory chunks. This is done via the pseudo-instructions that is generalized as dx. The x varies depending on size of the "chunk", see section 3.2.1 in the manual. For example, a dw (define word) allocates two bytes (16 bits) of memory, and dd (define doubleword) allocates four bytes (32 bits) of memory. In relation to a higher level language, such as C, a char is a byte, a short is two bytes and an int is four bytes.
The syntax for declaring data is as follows: dx {data}
TIMES is another pseudo-instruction that repeats an instruction n times, just like TIMES {n} {instruction}. see section 3.2.5 in the manual.
The above instructions are called pseudo instructions because they are not real x86 machine instructions, but are used in the instruction field anyway. see section 3.2 in the manual