Re: [Lightning] Re: Stack register control?

[Eli Barzilay]

On Sep 30, Paolo Bonzini wrote:
> 
> > We had an experience with both.  The two systems are not really
> > comparable.  LLVM is *huge*, and much more difficult to get into your
> > project.  But since it's a generic framework for compilers, you get
> > lots of advantages -- the most obvious is that you don't need to worry
> > as much about optimizations.  OTOH, Lightning is a simple native-code
> > generator, which is much easier to use, but it doesn't do more than
> > just that.  (BTW, IIRC GCing code was not a problem in LLVM.)
> > 
> > (Since it's off topic, feel free to mail me off list for more
> > details, if you're interested.)
> 
> Not really, feel free to keep this on list.

Well, there wasn't really anything that would surprise...  Comparing
lightning and llvm is like comparing a tool that does a basic job like
ed to something that throws the kitchen sink in like Emacs.  Sandro
asked if we tried to compare speeds -- we didn't, but I suspect that
this comparison depends heavily on the level of optimization that we
do, if any.  Our hope with the llvm experiment was to benefit from
some of its advanced optimization -- like good inlining and register
allocation, but we never got to actually compile code to the llvm
backend (we were working first on a high-level assembly language, and
that's were the project failed (for irrelevant reasons)).

In any case, the big advantage of LLVM is, of course, also its big
disadvantage -- incorporating LLVM to an existing project requires
much more work (probably on the same scale as embedding gcc) than
Lightning.  File sizes were also a problem -- the size is usually not
an issue when you have a 1mb library, but with an executable size of
1.8mb it does become a point of consideration.

-- 
          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
                  http://www.barzilay.org/                 Maze is Life!