I've received several submissions just like this but have always avoided posting them because they seemed too short. Well, that, and I couldn't think of anything to say about this sort of (il)logic. I still got nothing, but fortunately Telek added a little more background:

I found this in source code that one of our customers sent us. They were complaining that when processing large incoming buffers our device ran too slow. I took a quick look through their code and it was no wonder why it was slow. However this one function, run on every character of the incoming/outgoing buffer, deserved special mention.

ushort GetAsciiValue (uchar ch)
{
    ushort i;
    for (i=0; i<=255; i++)
    {
        if(ch == i)
            return i;
    }
    return 0;
}