On Tue, 14 May 2002, Oliver Bandel wrote:

> On Tue, 14 May 2002, Francois Pottier wrote:
> 
> [...]
> > O'Caml's type inference algorithm is sub-optimal for at least one
> > reason: it performs the occurs check at every unification step,
> > instead of delaying it until the current `let' binding is exited.
> 
> Would it help to use the Lazy-module and rewrite some parts
> od the application that uses so much time, if written "unlazy"?
> 
> BTW: When (in which kind of application) does the Lazy-module
>      really helps increasing performance?

The compiler itself uses Lazy in environment manager (typing/env.ml)
Xavier et al can probably shed more light on benefits.

-- 
mailto:malc@pulsesoft.com

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners