Part A: Bitwise Logical and Shift Operations
Create a SPARC assembly language program that extracts a bit-field from the contents of register %l0. The position of the rightmost bit of the field is indicated in register %l1, and the number of bits in the field is indicated in register %l2. The extracted field should be put into register %l3, right-shifted so that field starts at bit 0; any bits outside of the extracted field should be set to 0. Structure your program so that it operates on 3 separate inputs, each with different input data, field positions, and field widths. Your TA will provide the inputs to work with. Do not hard code any bit masks; your program should create them using the appropriate bitwise operations. Optimize your program, eliminating nop instructions where possible. Do not use m4.
Use printf() to display in hexadecimal the contents of registers %l0, %l1, %l2 before each extraction, and %l3 afterwards. Also run your program in gdb, displaying the contents of registers %l0, %l1, %l2 before each extraction, and %l3 afterwards. Capture the gdb session using script. On a separate piece of paper, show the bit pattern for each hexadecimal number for the registers %l0 and %l3, circling the extracted field.
Part B: Integer Multiplication using Add and Shift Operations
Write a SPARC assembly language program that implements the following integer multiplication algorithm:
negative = multiplier >= 0 ? 0 : 1;
product = 0;
for (i = 0; i < 32; i++) {
if (multiplier & 1)
product += multiplicand;
(product and multiplier registers combined as a unit) >> 1;
}
if (negative)
product -= multiplicand;
Structure your program so that it shows 3 different multiplications: the first should multiply two positive numbers together, the second should multiply a positive number by a negative number, and the third should multiply two negative numbers together. Your TA will provide the input data to work with. Since we don't yet know how to create subroutines, you can simply cut and paste the multiplication code to do each multiplication.
Use printf() to display in hexadecimal the contents of the product, multiplier and multiplicand registers before and after each multiplication. Optimize your program, eliminating nop instructions where possible. Do not use m4. Also run the program in gdb, displaying the contents of key registers as the program executes; you should show that the algorithm is working as expected. Capture the gdb session using script. On a separate piece of paper, show the bit pattern (binary number) for each hexadecimal number, and its decimal equivalent (in other words, show the binary and decimal values of the multiplier, multiplicand, and product).
Other Requirements
Make sure your code is properly formatted into columns, is readable and fully documented, and includes identifying information at the top of each file. You must comment each line of assembly code. Your code should also be well designed: make sure it is well organized, clear, and concise. Your TA will specify the inputs to use for the above two programs.
New Skills Needed for this Assignment:
Use of bitwise logical and shift operations
Use of branching and condition code tests
Understanding of hexadecimal and binary numbers