mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-22 04:50:07 +00:00
51 lines
1.2 KiB
C
51 lines
1.2 KiB
C
/* w_gammaf.c -- float version of w_gamma.c.
|
|
* Conversion to float by Ian Lance Taylor, Cygnus Support, ian@cygnus.com.
|
|
*/
|
|
|
|
/*
|
|
* ====================================================
|
|
* Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.
|
|
*
|
|
* Developed at SunPro, a Sun Microsystems, Inc. business.
|
|
* Permission to use, copy, modify, and distribute this
|
|
* software is freely granted, provided that this notice
|
|
* is preserved.
|
|
* ====================================================
|
|
*/
|
|
|
|
#if defined(LIBM_SCCS) && !defined(lint)
|
|
static char rcsid[] = "$NetBSD: w_gammaf.c,v 1.4 1995/11/20 22:06:48 jtc Exp $";
|
|
#endif
|
|
|
|
#include "math.h"
|
|
#include "math_private.h"
|
|
|
|
#ifdef __STDC__
|
|
float __tgammaf(float x)
|
|
#else
|
|
float __tgammaf(x)
|
|
float x;
|
|
#endif
|
|
{
|
|
float y;
|
|
int local_signgam;
|
|
y = __ieee754_gammaf_r(x,&local_signgam);
|
|
if (local_signgam < 0) y = -y;
|
|
#ifdef _IEEE_LIBM
|
|
return y;
|
|
#else
|
|
if(_LIB_VERSION == _IEEE_) return y;
|
|
|
|
if(!__finitef(y)&&__finitef(x)) {
|
|
if(__floorf(x)==x&&x<=(float)0.0)
|
|
/* tgammaf pole */
|
|
return (float)__kernel_standard((double)x,(double)x,141);
|
|
else
|
|
/* tgammaf overflow */
|
|
return (float)__kernel_standard((double)x,(double)x,140);
|
|
}
|
|
return y;
|
|
#endif
|
|
}
|
|
weak_alias (__tgammaf, tgammaf)
|