Designing Embedded Hardware, 2nd Edition

Position-Independent Code

When we jump to an address or subroutine within our program, we could use an absolute address. For example, JSR $0200 jumps to a subroutine at the absolute address 0x0200. This means that for our program to work, that subroutine must always be at address 0x0200. A program (and hence its subroutines) could be located (or relocated) by the computer's operating system to anywhere in memory. If this were to happen, JSR $0200 would no longer jump to the location of our subroutine because that subroutine would no longer be at address 0x0200. Therefore the program would crash. For this reason, it is good practice to employ position-independent code . This method of programming avoids the use of absolute addresses (except where appropriate). It means that rather than jumping to the absolute address of our subroutine, we branch to the subroutine relative to the program counter's contents. It sounds complicated, but all it really means is that we should use BRA (or BSR) rather than JMP (or JSR). By using a branch rather than a jump, we are adding a number to the program counter rather than replacing its contents. This means that we are not jumping to an absolute position in memory, but branching relative to our current location.

Absolute addressing is used when we branch or jump to some code that we know will always be at a given address. For example, a subroutine that is located in ROM (permanent memory) is not going to move somewhere else.