Re: Compiler Optimisation?

[Ray Dillinger <bear@sonic.net>]

Iain Bate wrote:
>
> I am looking for some specific information on compiler optimisation
> and I was wondering if anyone could help. Basically, I would like to
> get some idea how much faster code executes due to optimisation. I am
> looking for the information at two levels:
>
> 1. How much faster code executes due to the overall optimisation
> process?

Depends on the language semantics among other things.  in LISP and its
cousins for example, optimization consists in large part of proving
that it's safe to eliminate symbol table checks and type lookups that
simply aren't required by the semantics of most languages -- as a
result, optimization of a LISP yields a much higher marginal return
than, say, optimization of Pascal.

> 2. How much faster code executes due to specific optimisation stages,
> e.g. the speed-up due to the peephole stage?

The speed-up due to the peephole stage in my experience runs between
ten and twenty percent of overall speed -- Although, of course, it
depends on what you put into the "peephole" stage.  I've purposefully
*not* done optimization in other stages that could be put off for the
peephole optimizer, which tends to make it seem a lot more important.

				Ray