Re: A fast scheme implementation?

William D Clinger <cesura17@xxxxxxxxx> writes:

Think about it: a double integral. He's talking
about a general purpose library procedure, highly
tuned, that performs numerical integration. You'd
pass it a procedure in Scheme, or a function pointer
in C, and the two limits of integration.

Or, in C++, it could be templated on a class having the function as a
member; then the compiler will, pardon my French, defunctorize the
result and thus specialize the integral routine for that function.

C compilers don't know how to optimize across library
boundaries, however.

Not entirely true. If the compiler squirrels away a serialized
representation of its IR along with the object code, then when invoked
to do linking it can perform whole-program optimizations.

Such things have been known in commercial C compilers for some time;
more recently the LLVM project has been working in this area, and a
separate effort is known to exist for GCC.

--
(let ((C call-with-current-continuation)) (apply (lambda (x y) (x y)) (map
((lambda (r) ((C C) (lambda (s) (r (lambda l (apply (s s) l)))))) (lambda
(f) (lambda (l) (if (null? l) C (lambda (k) (display (car l)) ((f (cdr l))
(C k))))))) '((#\J #\d #\D #\v #\s) (#\e #\space #\a #\i #\newline)))))
.

Relevant Pages

  • Re: Where to start ? Too many options to choose from.
    ... compiler and is really old anyway. ... the latter has supposedly better integration but it also ... with evaluation board I have a Win CE 5.0 Pro license. ... Hence I'd need platform builder. ...
    (microsoft.public.windowsce.app.development)
  • Re: what werent structure/union/map made part of F77/90 ?
    ... > standard for how many places something gets said about derived types. ... > integration, intended only to standardize old codes, then that would not ... Visual Fortran compiler. ... Kinda hard to track them all down and eliminate them with limited staff. ...
    (comp.lang.fortran)
  • Re: explicit versus implicit method for the heat equation
    ... Many people are solving the one-dimensional heat (diffusion) equation using central finite differences in space and forward Euler in time. ... To be able to take larger time steps, some people therefore use the implicit backward Euler scheme. ... However, as both time integration methods are first order methods, and the error of the full numerical scheme is ... I am wondering if putting much effort in using the backward Euler scheme has much sense. ...
    (sci.math.num-analysis)
  • Re: The crux of peoples issues with PLT Scheme?
    ...  * COM integration on Windows ... I acknowledge there is probably a lot more in store for PLT library ... them and start using Bigloo (because a fast, mature Scheme is already ...
    (comp.lang.scheme)
  • Re: Intel nonVisual fortran
    ... I am not using Lahey anymore. ... buying VS separate and getting it from Intel with the Fortran compiler? ... like the Intel compiler and its integration in VS very much. ...
    (comp.lang.fortran)