Do I need to explicitly handle negative numbers or zero when summing squared digits?

I recently had a test in my class. One of the problems was the following:

Given a number n, write a function in C/C++ that returns the sum of the digits of the number squared. (The following is important). The range of n is [ -(10^7), 10^7 ]. Example: If n = 123, your function should return 14 (1^2 + 2^2 + 3^2 = 14).

This is the function that I wrote:

int sum_of_digits_squared(int n) 
{
    int s = 0, c;

    while (n) {
        c = n % 10;
        s += (c * c);
        n /= 10;
    }

    return s;
}

Looked right to me. So now the test came back and I found that the teacher didn’t give me all the points for a reason that I do not understand. According to him, for my function to be complete, I should’ve have added the following detail:

int sum_of_digits_squared(int n) 
 {
    int s = 0, c;

    if (n == 0) {      //
        return 0;      //
    }                  //
                       // THIS APPARENTLY SHOULD'VE 
    if (n < 0) {       // BEEN IN THE FUNCTION FOR IT
        n = n * (-1);  // TO BE CORRECT
    }                  //

    while (n) {
        c = n % 10;
        s += (c * c);
        n /= 10;
    }

    return s;
}

The argument for this is that the number n is in the range [-(10^7), 10^7], so it can be a negative number. But I don’t see where my own version of the function fails. If I understand correctly, the meaning of while(n) is while(n != 0), not while (n > 0), so in my version of the function the number n wouldn’t fail to enter the loop. It would work just the same.

Then, I tried both versions of the function on my computer at home and I got exactly the same answers for all the examples that I tried. So, sum_of_digits_squared(-123) is equal to sum_of_digits_squared(123) (which again, is equal to 14) (even without the detail that I apparently should’ve added). Indeed, if I try to print on the screen the digits of the number (from least to greatest in importance), in the 123 case I get 3 2 1 and in the -123 case I get -3 -2 -1 (which is actually kind of interesting). But in this problem it wouldn’t matter since we square the digits.

So, who’s wrong?

EDIT: My bad, I forgot to specify and didn’t know it was important. The version of C used in our class and tests has to be C99 or newer. So I guess (by reading the comments) that my version would get the correct answer in any way.