mirror of
https://github.com/autc04/Retro68.git
synced 2024-12-11 03:52:59 +00:00
48 lines
1.1 KiB
C
48 lines
1.1 KiB
C
/*
|
|
* ====================================================
|
|
* Copyright (C) 1998,2002 by Red Hat Inc. All rights reserved.
|
|
*
|
|
* Permission to use, copy, modify, and distribute this
|
|
* software is freely granted, provided that this notice
|
|
* is preserved.
|
|
* ====================================================
|
|
*/
|
|
|
|
#if !defined(_SOFT_FLOAT)
|
|
|
|
/*
|
|
Fast version of exp using Intel float instructions.
|
|
|
|
double _f_exp (double x);
|
|
|
|
Function computes e ** x. The following special cases exist:
|
|
1. if x is 0.0 ==> return 1.0
|
|
2. if x is infinity ==> return infinity
|
|
3. if x is -infinity ==> return 0.0
|
|
4. if x is NaN ==> return x
|
|
There is no error checking or setting of errno.
|
|
*/
|
|
|
|
|
|
#include <math.h>
|
|
#include <ieeefp.h>
|
|
#include "f_math.h"
|
|
|
|
double _f_exp (double x)
|
|
{
|
|
if (check_finite(x))
|
|
{
|
|
double result;
|
|
asm ("fldl2e; fmulp; fld %%st; frndint; fsub %%st,%%st(1); fxch;" \
|
|
"fchs; f2xm1; fld1; faddp; fxch; fld1; fscale; fstp %%st(1); fmulp" :
|
|
"=t"(result) : "0"(x));
|
|
return result;
|
|
}
|
|
else if (x == -infinity())
|
|
return 0.0;
|
|
|
|
return x;
|
|
}
|
|
|
|
#endif
|