Designing Embedded Hardware, 2nd Edition

Loops

It is often useful to repeat an instruction or series of instructions. Using branch instructions allows us to redirect the program counter and hence control the flow of the program. Here is a simple example of a loop that executes 15 times:

       LDAA #$0F  ; load the count into the A accumulator
again             ;
       DECA       ; count down (decrement the A accumulator)
       BNE again  ; repeat until we have counted down to zero

DECA decrements the A accumulator and sets the status flags as appropriate. If the result of the decrement was a zero in the A accumulator, the zero flag in the CCR (Condition Code Register) is set; otherwise, the flag is cleared. The BNE instruction is "branch not equal (to zero)." This checks the state of the zero flag. If it is clear (not equal to zero), then the branch is taken; otherwise, execution continues on with the next instruction after BNE. Hence, the above code fragment will start with the A accumulator equal to 15 and will decrement this value until is has reached zero. At this point, the loop will terminate. Now, to get the loop to actually perform work for you, place your code between the again label and the DECA instruction.

In machine code, this is:

Address Opcodes       Assembly
0100    86 0F         LDAA #$0F
0102    4A      again DECA
0103    26 FD         BNE  again
0105    ...           ; address of instruction after BNE

Note the instruction BNE. Where did the 0xFD come from? We want the processor to branch back to address 0x0102. As with all instructions, when executing the BNE the program counter points to ...