[BACK]Return to s_exp2f.c CVS log [TXT][DIR] Up to [cvs.NetBSD.org] / src / lib / libm / src

File: [cvs.NetBSD.org] / src / lib / libm / src / s_exp2f.c (download)

Revision 1.2, Sun Mar 16 22:30:43 2014 UTC (5 months, 4 weeks ago) by dsl
Branch: MAIN
CVS Tags: yamt-pagecache-base9, tls-maxphys-base, tls-earlyentropy-base, tls-earlyentropy, riastradh-xf86-video-intel-2-7-1-pre-2-21-15, riastradh-drm2-base3, netbsd-7-base, netbsd-7, HEAD
Changes since 1.1: +26 -5 lines

Fix overflow and underflow on i386.
The return value of a 'float' function is in the x87 %st(0) register.
This is an 80bit 'long double' register.
If you multiply 0x1p100f by 0x1p100f the caller sees 0x1p200 - not the
  expected infinity.
So use a 'double' value which goes through a store-load sequence to generate
  the required exception and value.

/*-
 * Copyright (c) 2005 David Schultz <das@FreeBSD.ORG>
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 */

#include <sys/cdefs.h>
__RCSID("$NetBSD: s_exp2f.c,v 1.2 2014/03/16 22:30:43 dsl Exp $");
#ifdef __FBSDID
__FBSDID("$FreeBSD: src/lib/msun/src/s_exp2f.c,v 1.9 2008/02/22 02:27:34 das Exp $");
#endif

#include <float.h>

#include "math.h"
#include "math_private.h"

#define	TBLBITS	4
#define	TBLSIZE	(1 << TBLBITS)

static const float
    redux   = 0x1.8p23f / TBLSIZE,
    P1	    = 0x1.62e430p-1f,
    P2	    = 0x1.ebfbe0p-3f,
    P3	    = 0x1.c6b348p-5f,
    P4	    = 0x1.3b2c9cp-7f;

/*
 * For out of range values we need to generate the appropriate
 * underflow or overflow trap as well as generating infinity or zero.
 * This means we have to get the fpu to execute an instruction that
 * will generate the trap (and not have the compiler optimise it away).
 * This is normally done by calculating 'huge * huge' or 'tiny * tiny'.
 *
 * i386 is particularly problematic.
 * The 'float' result is returned on the x87 stack, so is 'long double'.
 * If we just multiply two 'float' values the caller will see 0x1p+/-200
 * (not 0 or infinity). 
 * If we use 'double' the compiler does a store-load which will convert the
 * value and generate the required exception.
 */
#ifdef __i386__
static volatile double overflow = 0x1p+1000;
static volatile double underflow = 0x1p-1000;
#else
static volatile float huge = 0x1p+100;
static volatile float tiny = 0x1p-100;
#define overflow (huge * huge)
#define underflow (tiny * tiny)
#endif

static const double exp2ft[TBLSIZE] = {
	0x1.6a09e667f3bcdp-1,
	0x1.7a11473eb0187p-1,
	0x1.8ace5422aa0dbp-1,
	0x1.9c49182a3f090p-1,
	0x1.ae89f995ad3adp-1,
	0x1.c199bdd85529cp-1,
	0x1.d5818dcfba487p-1,
	0x1.ea4afa2a490dap-1,
	0x1.0000000000000p+0,
	0x1.0b5586cf9890fp+0,
	0x1.172b83c7d517bp+0,
	0x1.2387a6e756238p+0,
	0x1.306fe0a31b715p+0,
	0x1.3dea64c123422p+0,
	0x1.4bfdad5362a27p+0,
	0x1.5ab07dd485429p+0,
};
	
/*
 * exp2f(x): compute the base 2 exponential of x
 *
 * Accuracy: Peak error < 0.501 ulp; location of peak: -0.030110927.
 *
 * Method: (equally-spaced tables)
 *
 *   Reduce x:
 *     x = 2**k + y, for integer k and |y| <= 1/2.
 *     Thus we have exp2f(x) = 2**k * exp2(y).
 *
 *   Reduce y:
 *     y = i/TBLSIZE + z for integer i near y * TBLSIZE.
 *     Thus we have exp2(y) = exp2(i/TBLSIZE) * exp2(z),
 *     with |z| <= 2**-(TBLSIZE+1).
 *
 *   We compute exp2(i/TBLSIZE) via table lookup and exp2(z) via a
 *   degree-4 minimax polynomial with maximum error under 1.4 * 2**-33.
 *   Using double precision for everything except the reduction makes
 *   roundoff error insignificant and simplifies the scaling step.
 *
 *   This method is due to Tang, but I do not use his suggested parameters:
 *
 *	Tang, P.  Table-driven Implementation of the Exponential Function
 *	in IEEE Floating-Point Arithmetic.  TOMS 15(2), 144-157 (1989).
 */
float
exp2f(float x)
{
	double tv, twopk, u, z;
	float t;
	uint32_t hx, ix, i0;
	int32_t k;

	/* Filter out exceptional cases. */
	GET_FLOAT_WORD(hx, x);
	ix = hx & 0x7fffffff;		/* high word of |x| */
	if(ix >= 0x43000000) {			/* |x| >= 128 */
		if(ix >= 0x7f800000) {
			if ((ix & 0x7fffff) != 0 || (hx & 0x80000000) == 0)
				return (x + x);	/* x is NaN or +Inf */
			else 
				return (0.0);	/* x is -Inf */
		}
		if(x >= 0x1.0p7f)
			return overflow;	/* +infinity with overflow */
		if(x <= -0x1.2cp7f)
			return underflow;	/* zero with underflow */
	} else if (ix <= 0x33000000) {		/* |x| <= 0x1p-25 */
		return (1.0f + x);
	}

	/* Reduce x, computing z, i0, and k. */
	STRICT_ASSIGN(float, t, x + redux);
	GET_FLOAT_WORD(i0, t);
	i0 += TBLSIZE / 2;
	k = (i0 >> TBLBITS) << 20;
	i0 &= TBLSIZE - 1;
	t -= redux;
	z = x - t;
	INSERT_WORDS(twopk, 0x3ff00000 + k, 0);

	/* Compute r = exp2(y) = exp2ft[i0] * p(z). */
	tv = exp2ft[i0];
	u = tv * z;
	tv = tv + u * (P1 + z * P2) + u * (z * z) * (P3 + z * P4);

	/* Scale by 2**(k>>20). */
	return (tv * twopk);
}