induction variable optimization

[guy at gorodish.UUCP]

>So long as an optimization is guaranteed never to alter program semantics,
>it really should be done by the compiler rather than by the programmer, who
>generally has more important things to worry about.

And who is, more often than not these days, writing C code rather
than writing code that happens to be in C for some particular
machine (even when writing kernel code!), so that they *can't* know
what code is better or worse.