```c
int main(int argc, char **argv)
{
        static const size_t len = 2ull*1024*1024*1024;
        char *src = malloc(len);
        char dst[10];
        int loops;

        memset(src, 'A', len - 1);
        src[len - 1] = 0;

        loops = atoi(argv[1]);
        for (int i = 0; i < loops; i++) {
                strlcpy(dst, src, sizeof(dst));
        }
        return (dst[0]+dst[1]);
}
```

(Substituting strlcpy implementation.  I verified the optimizer was not cheating and inlining the loop.)

Invoked:

```
for i in $(jot 5) ; do /usr/bin/time ./a.exe 20 ; done
```

```
x dataa
+ datab
+----------------------------------------------------------------------------------------+
|+                                                                                       |
|+                                                                                   x   |
|+                                                                                   x   |
|+                                                                                   x   |
|+                                                                                   x  x|
|A                                                                                   MA| |
+----------------------------------------------------------------------------------------+
    N           Min           Max        Median           Avg        Stddev
x   5         16.93          17.5         16.99        17.084    0.23479779
+   5          0.81          0.85          0.85         0.834   0.021908902
Difference at 95.0% confidence
        -16.25 +/- 0.243193
        -95.1182% +/- 0.149259%
        (Student's t, pooled s = 0.166748)
```