[gambit-list] getting float* out of a f32vector

Adrien Piérard pierarda at iro.umontreal.ca
Tue Apr 21 11:30:55 EDT 2009


Since Scheme requires exact arithmetic, are there any chances that
there is no such macro/function, and that inexact arithmetic is
handled with, say, {U,S}64 ?
(In my guess, then, exact decimal arithmetic would be handled like
bignums are, through pointers to a structure. But I've never
implemented a RnRS compliant compiler in C, and don't know how this is
usually handled)

P!

2009/4/21 lowly coder <lowlycoder at huoyanjinjing.com>:
> So for something like a u8vector, I can get the unsigned char* out of it as
> follows:
>
> &___FETCH_U8(___BODY,(___arg1),___INT(0))
>
> how do I do this for something like a floating point? Grepping through
> gambit.h, I see a bunch of ___FETCH_{U/S}{8/16/32/64}, but not any
> ___FETCH_F32
>
> Thanks!

-- 
Français, English, 日本語, 한국어



More information about the Gambit-list mailing list