|
From: | felix winkelmann |
Subject: | Re: [Chicken-users] Cleaning up floating-point-specific operators |
Date: | Fri, 1 Sep 2006 08:48:01 +0200 |
On 8/31/06, Will M Farr <address@hidden> wrote:
Hello all, Due to the discussion of floating-point performance on this list initiated by Carlos Pita, I've been writing some toy benchmarks which use the fp+, fp-, etc operators. The situation with these operators is a *mess*:
I have added your patch. Thanks for clearing this up. I modified your changes to throw a type-error instead of an arithmetic error if non-flonum arguments are passed to fpXXX operations. Also, "flonum?" is detected by the compiler and rewritten to a ##core#inline form (with extended bindings enabled). I would be grateful if you could document these changes in the wiki. cheers, felix -- http://galinha.ucpel.tche.br:8081/blog/blog.ssp
[Prev in Thread] | Current Thread | [Next in Thread] |