Debugging Documentation

Introduction

The objective was to debug a C++ program performing matrix multiplication. An issue was identified where the program was not providing the correct matrix multiplication output.

Steps Taken for Debugging

0. Initial Run

Ran the program to observe the problem firsthand and identified that the issue was in the resultant multiplication output, as it provided an incorrect answer.

1. Preliminary Checks

The Matrix was printed successfully with the right data, validating the setData and display methods.
The problem was narrowed down to the multiply method.

2. Inspecting the Multiply Method

The message "Matrix dimensions are incompatible for multiplication." (line 34) wasn't printed, so the code did not enter this condition.
A breakpoint was set at line 38 and rows and other.cols were printed to verify their values, which were correct.

g++ -g -O0 -o Matrix Matrix.cpp
gdb Matrix

break 38
print rows
print other.cols

3. Loop Logic Verification

The 3 nested loops at line 40 were investigated, focusing on the assignment operation within them.
A breakpoint was set at the beginning of the loops and cols was printed to verify its value.

break 40
print cols

PS: We don't need to check rows and other.cols again as they were previously checked and found to be correct.

4. Value Tracking

Tracked the changes to result.data to observe how its value was being altered.
Observed the first cell result.data[0][0] which was initially 0. After the first iteration, it became 7 (1*7), which was expected.
The second iteration posed an issue where the value was 18 (2*9), but it should have been 25 (1*7 + 2*9).

print result.data 
continue
print result.data
continue
print result.data

5. Issue Identification

Identified that line 43 was performing an assignment operation instead of an addition to the previous value:
```
result.data[i][j] = data[i][k] * other.data[k][j];
```

6. Issue Fixing

Corrected the line to be as follows:

result.data[i][j] += data[i][k] * other.data[k][j];

7. Re-testing the Application

Re-ran the application with various test cases to ensure the corrected code now provides the expected and accurate matrix multiplication output.

Conclusion

The matrix multiplication logic was corrected by ensuring that the products of the respective elements were accumulated instead of being overwritten. This was achieved by altering the assignment operation to accumulate the values.

helmy162/Debugging-and-Logging-Assignment

README