[gambit-list] Need new flonum functions---are they available on all systems of interest?

Marc Feeley feeley at iro.umontreal.ca
Fri Dec 6 09:23:32 EST 2013


On Nov 25, 2013, at 3:30 PM, Bradley Lucier <lucier at math.purdue.edu> wrote:

> Marc:
> 
> I have the following definitions in a file I'm using when updating Gambit's trigonometric (and adding hyperbolic) functions:
> 
> ;;; #include <math.h> is in gambit.h
> 
> (define ##scalbn (c-lambda (double int) double "scalbn"))
> (define ##ilogb  (c-lambda (double) int "ilogb"))
> 
> (define ##flsinh (c-lambda (double) double "sinh"))
> (define ##flcosh (c-lambda (double) double "cosh"))
> (define ##fltanh (c-lambda (double) double "tanh"))
> 
> (define ##flasinh (c-lambda (double) double "asinh"))
> (define ##flacosh (c-lambda (double) double "acosh"))
> (define ##flatanh (c-lambda (double) double "atanh"))
> 
> (define ##fllog1p (c-lambda (double) double "log1p"))
> (define ##flexpm1 (c-lambda (double) double "expm1"))
> 
> It would be really helpful for me if these could be added to gambit.h, the compiler, and _num.scm; then I could just propose patches to _num.scm for my changes, which will come in stages.
> 
> Also, it would be good to know whether these are available on all systems that Gambit runs on.
> 
> Brad

I have put this on my TODO.

Marc




More information about the Gambit-list mailing list