Assembly programming 15 points consider the x86 32

Problem 4: Assembly Programming (15 points)
Consider the X86 32-bit assembly code.
puzzle:
pushl %ebp
movl %esp, %ebp
movl 12(%ebp), %eax
movl (%eax), %eax
movl 8(%ebp), %edx
imull (%edx), %eax
popl %ebp
ret
• (2 points) How many memory accesses are performed by the body of the puzzle function excluding
stack pushes and pops?
• 13 points Write the C code for the above assembly? Clearly show the function body, function signa-
tures and function return values?
7

Problem 5: Assembly Programming (25 points)
1. (9 points) Consider the below assembly code.
movl (%eax), %edx
movl (%edx), %ebx
If the register %eax maps to a C variable p and the register %edx maps to a C variable q .
• What is the data type of p in C?
• What is the data type of q in C?
• Write the equivalent C code for the above operation?
2. (6 points) Are these operations legal? If not, why?
3. (6 points) What is the effective address for each of the following:
%eax: 15
%edx: 0x40
%ebp: 0x6000
(a) 0 x 4 ( % ebp ) :
(b) ( , % ebp , 4 ) :
(c) ( % ebp , % edx , 4 ) :
8

4. (4 points) When you generate assembly code, you often see the following code.
test %eax, %eax
Why does the compiler generate this code? What is another alternative way of performing the same
operation?
9