I was running some mutation tests on this C-testsuite. One of my mutations caused the following test 00143 to either seg fault, or to run to completion with a stack smashing detected error. My mutation changed n = (count+7) / 8; to n = (count+7) * 8;
I have posted the mutated code below:
#include <stdio.h>
int main()
{
    int  count, n;
    short *from, *to;
    short a[39], b[39];
    for(n = 0; n < 39; n++) {
        a[n] = n;
        b[n] = 0;
    }
    from = a;
    to = b;
    count = 39;
    n = (count + 7) * 8;
    switch (count % 8) {
    case 0: do { *to++ = *from++;
    case 7:      *to++ = *from++;
    case 6:      *to++ = *from++;
    case 5:      *to++ = *from++;
    case 4:      *to++ = *from++;
    case 3:      *to++ = *from++;
    case 2:      *to++ = *from++;
    case 1:      *to++ = *from++;
            } while (--n > 0);
    }
    for(n = 0; n < 39; n++)
        if(a[n] != b[n])
            return 1;
    return 0;
}
You can see that n should not go out of bounds as the last for loop is between 0 <= n < 39.
My question is why does a segmentation fault, or stack smashing occur if I am not accessing out of bound arrays? Additionally why do I get flaky behavior?
