Re: [clisp-list] Stack size problem

[Sam S. <sd...@gn...>]

> * In message <5.1...@mi...>
> * On the subject of "Re: [clisp-list] Stack size problem"
> * Sent on Sat, 05 Jan 2002 23:06:52 +0000
> * Honorable Miguel Gon=E7alves <m...@mi...> writes:
>
> BTW, why running (fact 250000) (that's a huge number :) I am just
> trying out LISP's mathematical capabilities as I intend to use it very
> soon for a project that involves lots of mathematics) in non-compiled
> mode CLISP aborts with:
>=20
> C:\CLISP>clisp factorial.lsp
>=20
> C:\CLISP>c:\clisp\lisp.exe -m 128MB -M c:\clisp\lispinit.mem -B c:\clisp
> factori
> al.lsp
>=20
> *** - Lisp stack overflow. RESET
>=20
> and running with the -C flag (id est, compiling before running) it all
> goes well (it completes the calculation in about 60 seconds -- real
> time)?

compiled code is faster and uses less stack - are you surprised?

note that your implementation of factorial is very inefficient.
see CLOCC/CLLIB/math.lisp (product-from-to, fibonacci &c)
<http://www.podval.org/~sds/data/cllib.html>

note also that CLISP has some mathematical functions built-in.
<http://clisp.cons.org/impnotes.html#d39e4421>

--=20
Sam Steingold (http://www.podval.org/~sds)
Keep Jerusalem united! <http://www.onejerusalem.org/Petition.asp>
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
Your mouse has moved - WinNT has to be restarted for this to take effect.