C for loop almost right

Question

I am working on a homework problem involving comparing a recursive solution to a problem with its iterative counterpart. Currently I have the recursive function working for the given set of values, however the iterative version seems to work for every value other than the last one which seems odd to me.

The iterative function is as follows

uint64_t
normalPower(uint64_t k, uint64_t n, uint32_t m)
{
    uint64_t answer = 1;
    for (int i = 1; i <= n; i++)
    {
    answer = answer * k % m;
    }
    return answer;
    return EXIT_SUCCESS;

}

with the other relevant information being

uint64_t n;
for (int i = 0; i <= 10; i++)
{
    n = n * 10;
}

and

uint64_t k = 1835;
uint32_t m = 590623782;

When I run the recursive function I get the answer 424171147 for the final value which I have confirmed with other classmates. I am just confused as to why the algorithm works for all of the previous values but not the final one. Any and all help is greatly appreciated.

On a side note: I am aware the iterative version is wildly inefficient, that is the point of the assignment.

As requested here is the recursive function

uint64_t
fastPower(uint64_t k, uint64_t n, uint32_t m)
{
    uint64_t answer = 0;
    if (n % 2 == 0 && n > 1)
    {
    uint64_t kn = fastPower(k, n / 2, m);
    answer = (kn * kn) % m;
    return answer;
    }
    else if (n > 1)
    {
    answer = fastPower(k, 1, m) * fastPower(k, n - 1, m) % m;
    return answer; 
    }
    else
    answer = k % m;
        return answer;
    return EXIT_SUCCESS;
}

Also n is initialized as 1 when it is declared.

`n = n * 10;`. Is that really the actual code? `n` is used uninitialised. — kaylum, Oct 22 '15 at 01:31
Check that you're using the same last number in both methods and please show the code of the recursive method. — SlySherZ, Oct 22 '15 at 02:10
I just ran it and k, n and m are the same for each iteration of both functions. I can't seem to figure out why the iterative one would only differ when n = 10^10 — tacoofdoomk, Oct 22 '15 at 13:04

score 1 · Accepted Answer · answered Oct 22 '15 at 13:43

Your code is 'correct' mathematically. What you're running into is an integer overflow on your loop counter in the iterative function. Since i is a signed int, it is wrapping over into negative numbers (which when compared to your unsigned n, is interpreted as unsigned, which will lead to strange results)

If you change

for (int i = 1; i <= n; i++)
{
    answer = answer * k % m;
}

to

for (uint64_t i = 1; i <= n; i++)
{
    answer = answer * k % m;
}

Then i will match n in size and sign, and the answers will match

C for loop almost right

1 Answers1