---

Returning from a leaf subroutine

---

• Another form of the mov instruction
  • Previously, we have learned how to store a (small) constant number into a register:

    mov r0, #smallNumber // Store smallNumber into register r0

    ---

  • The mov instruction can also be used to store (= copy) the value in a register into another register:

    mov r0, r1 // Copy r1 to r0 // I.e.: assign r0 = r1

    You can use any register for r₀ and r₁.

  ---

  ---

  ---

  ---

• Returning from a subroutine call in ARM
  • The Return from Subroutine in the ARM processor can be achieved using a mov instruction

    mov pc, lr Effect: Copy the return address saved by the bl instruction into the Program Counter (PC)

    Explanation:

    When you update the PC (register) with the retun address in the lr register, the ARM processor will fetch the next instruction at the return address !!! The effect of the mov pc, lr instruction is: Branch to the return address given in the lr register

    ---

    ---

  • Example:

    main: mov r0, #4 mov r1, #4 bl myMethod mov r2, #4 mov r3, #4 bl myMethod mov r4, #4 mov r5, #4 myMethod: mov r10, #4 mov r9, #4 mov pc, lr // Return to caller !!!

    The main program calls the method myMethod twice.

    In each call, the return address is saved in the lr register

    So the mov pc, lr instruction in the myMethod function is able to jump back to the location where the function call took place !!!

    ---

  • Example Program: (Demo above code)                                                
    • Prog file: /home/cs255001/demo/asm/8-sub/bl+rts.s

    How to run the program:

    To compile:   as255 bl+rts          To run: use EGTAPI

  ---

  ---

  ---

  ---

• Marked differences between methods in high level language and assembler

  • Methods (or subroutines) in high level languages have very nice syntax structures to highlight where the method begins and ends...

    Example:

    public int myMethod ( int p1, int p2 ) { ... }

    You can see the method name and the curly braces { .. } that denote the start and the end of the method clearly

    ---

    ---

    Methods (or subroutines) in assembler are nothing more than a series of instructions marked by a label

    Example:

    myMethod: ... ... (assembler instructions) ... mov pc, lr

    The labels used to mark the beginning of functions/methods in assembler code are no different from the labels you used in if and while statements !!!

    It's very hard to tell in assembler program which label to mark a method start location !!!

    So it's a good practice to use comments, e.g.:

    /* ************************************************** myMethod: this method wil do the following .... bla bla bla ... ************************************************** */ myMethod: ... ... (assembler instructions) ... mov pc, lr