Exercise 2: Tracing Condition Codes Through a Sequence — Possible Solution
====================================================================

THE SEQUENCE
------------------------------
1. AND R2, R2, #0        ; R2 becomes 0
2. ADD R2, R2, #7         ; R2 becomes 7
3. LEA R3, LABEL          ; R3 becomes LABEL's address (given as negative)
4. ADD R2, R2, #-7        ; R2 becomes 0

TRACE, LINE BY LINE
------------------------------
Line 1 -- AND R2, R2, #0
  Result written to R2 is 0 -> condition code set: Z

Line 2 -- ADD R2, R2, #7
  Result written to R2 is 7 (positive) -> condition code set: P

Line 3 -- LEA R3, LABEL
  LEA writes a register (R3), so it DOES set condition codes, per the
  chapter's own list of register-writing instructions. The problem
  states LABEL's computed address, interpreted as a signed value, is
  negative -> condition code set: N

Line 4 -- ADD R2, R2, #-7
  R2 (currently 7) + (-7) = 0 -> condition code set: Z

FINAL CONDITION CODES
------------------------------
After all four instructions, Z is set (N and P cleared) -- reflecting
line 4's result, NOT line 2's ADD R2, R2, #7, even though that was
also an addition into R2.

WHICH INSTRUCTION'S RESULT DO THEY REFLECT
------------------------------
The final condition codes reflect ONLY the most recent register-
writing instruction -- line 4 (ADD R2, R2, #-7) -- exactly per the
chapter's own warn-box: "condition codes reflect only the LAST
register write." Notably, line 3's LEA (writing to R3, a completely
different register than R2) still overwrote the condition codes set
by line 2, even though R2 itself wasn't touched by that LEA. This is
exactly the trap the warn-box calls out: ANY register write updates
N/Z/P, regardless of which register it targets.

WHY THIS WORKS AS AN ANSWER
------------------------------
It correctly identifies that LEA sets condition codes (since it
writes a register), tracks N/Z/P through all four lines using the
chapter's own set-on-every-register-write rule, and explicitly calls
out that the LEA in line 3 overwrote line 2's flags despite operating
on a different register -- the specific gotcha the chapter's warn-box
warns about.