Saturday, February 1, 10pm
See Assignment 3 on Canvas for the Github Classroom link.
Submit the contents of your repository via Gradescope. See Deliverables below for what to submit.
This is an individual assignment.
This assignment will take you through 2 tasks implementing and using functions in assembly. You will be also asked to write some C (pseudo-)code as a blueprint for your assembly. The C code does not need to compile or run, but needs to reflect the functionality. Feel free to rely on your Java knowledge here. If in doubt, leave comments.
This assignment will also introduce separate compilation. For one of the tasks you will write a function called by an already implemented main
function. For the other, your task will be to write a main
function for an already implemented function. Read the instruction carefully to make sure you understand what is being asked in each of the tasks.
For each of the required functions, you need to follow the Assembly Design Recipe from class. You can optimize your code once you’ve written it – if, for example, there is some repetition that can be trivially removed.
You are provided with a “mystery library”. This pre-compiled library contains a single function, crunch
, which takes two signed long integers and returns a long integer.
Your task is to complete the mystery
program by implementing the main
function in mystery-main.s
, which needs to call the provided crunch
function. This is also called a “driver”. When compiled, the program should have the following features and behaviour:
Accept exactly two arguments. You can assume that, when provided, these arguments are valid signed long integers.
The provided crunch
function should be called with these two numbers, converting them from strings as necessary.
Based on the result from the function, the program should print one of the following strings (with a newline at the end) and exit with an exit status of 0
:
hat
, if crunch
returned a negative integertea
, if crunch
returned 0
beer
, if crunch
returned a positive integer.If fewer or more than 2 arguments are provided, the program should print “Two arguments required.
”, followed by a newline, and exit with a status of 1
.
First, write the main
function in C and save it in mystery-main.c
, where we provided a “stub” for you. The program does not have to compile, but it should be a fairly accurate high-level representation of the assembly program. Do not spend too much time on this, but give it your best shot. Try to use your knowledge of Java and provide comments if you are struggling with C.
Second, implement an assembly version of the main
function in mystery-main.s
. Your program must compile using the provided Makefile by running make mystery
. The command make mystery
will combine libmystery.a
(provided by us) and mystery-main.s
(written by you) into the executable mystery
.
Sample test cases with mystery
:
$ $ ./mystery 1 2
tea
$ ./mystery 2344 12
beer
$ ./mystery -1345 321
hat
$ ./mystery 1
Two arguments required.
The second task is to write the function:
unsigned long array_max(unsigned long n, unsigned long *items)
This function will return the maximum value of an array of long integers ≥ 0. The first argument provided is the number of elements, the second argument is the address of the first element.
You do not need to write a C version of this function, but we recommend doing so.
We have provided the driver program in array-max-main.c
, which processes the command line arguments and calls the array_max
function. The Makefile will compile both array-max-main.c
and your implementation of the array_max
function to produce the executable array-max
. Once compiled, the interactions with array-max
should look as follows:
$ ./array-max 1 2
2
$ ./array-max 42 1
42
$ ./array-max 3 1 5 8 2 4 8 20 1
20
Your program must compile without any modification to the provided array-max-main.c
file, using the provided Makefile. The command make array-max
will combine array-max-main.c
(provided by us) and array-max.s
(written by you) into the executable array-max
.
We have provided a Makefile for you. You can use it as follows on the command line:
make
: compile all programs (mystery
, array-max
)make clean
: basic cleanup, remove binariesmake mystery
: compile the mystery
program using mystery-main.s
and libmystery.a
make array-max
: compile the array-max
program using array-max.s
and array-max-main.c
Modify the files mystery-main.s
(implementation) and mystery-main.c
(blueprint) and commit them to your repository.
Modify the file array-max.s
and commit it to your repository.
Do not include any executables, object files, or any other binary, intermediate or hidden files.
Finally, go to our Gradescope and submit a ZIP archive of your repository (which can be downloaded from Github).
While inspecting C compiler output to learn about assembly is a good approach, you are not, under any circumstances, allowed to submit compiler (or any other machine-generated) output as your .s
files. Doing so will result in an automatic 0 for the assignment. The code must be clearly written by you, following the Assembly Design Recipe
Read the assignment carefully and make sure you understand what is being asked in each Task.
Start early. This doesn’t mean you start writing code right away, but you should at least read the description, clone the assignment repository, and look for “missing links”.
Make sure you understand the provided starter code. Ask questions early if not.
Each argument to a program is a string, that is, the value is actually a memory address pointing to the first character of the string in memory.
The above is reflected in the signature of main
, which is int main(int argc, char *argv[])
– argc
is the argument count, and argv
is an array of strings. Each string is an array of characters, ending with a null byte/character (i.e., a character whose numeric ASCII value is 0). An array is represented as the address of (= a pointer to) its first element.
The first element of argv
is always the path and name of the executable. This means the actual arguments start at argv[1]
.
The return value of main
is the exit code or exit status of the program. We return 0 by default to signal success.
You can use standard C functions like atol
where needed.
Pay close attention to assembly calling conventions and the use of registers when calling C functions, or when writing functions.
Use examples from the lectures and the labs to help you get unstuck and ask questions.
Learn more about Makefiles: https://makefiletutorial.com/