glibc/sysdeps/ieee754/dbl-64/e_exp.c
Joseph Myers d96164c330 Refactor code forcing underflow exceptions.
Various floating-point functions have code to force underflow
exceptions if a tiny result was computed in a way that might not have
resulted in such exceptions even though the result is inexact.  This
typically uses math_force_eval to ensure that the underflowing
expression is evaluated, but sometimes uses volatile.

This patch refactors such code to use three new macros
math_check_force_underflow, math_check_force_underflow_nonneg and
math_check_force_underflow_complex (which in turn use
math_force_eval).  In the limited number of cases not suited to a
simple conversion to these macros, existing uses of volatile are
changed to use math_force_eval instead.  The converted code does not
always execute exactly the same sequence of operations as the original
code, but the overall effects should be the same.

Tested for x86_64, x86, mips64 and powerpc.

	* sysdeps/generic/math_private.h (fabs_tg): New macro.
	(min_of_type): Likewise.
	(math_check_force_underflow): Likewise.
	(math_check_force_underflow_nonneg): Likewise.
	(math_check_force_underflow_complex): Likewise.
	* math/e_exp2l.c (__ieee754_exp2l): Use
	math_check_force_underflow_nonneg.
	* math/k_casinh.c (__kernel_casinh): Likewise.
	* math/k_casinhf.c (__kernel_casinhf): Likewise.
	* math/k_casinhl.c (__kernel_casinhl): Likewise.
	* math/s_catan.c (__catan): Use
	math_check_force_underflow_complex.
	* math/s_catanf.c (__catanf): Likewise.
	* math/s_catanh.c (__catanh): Likewise.
	* math/s_catanhf.c (__catanhf): Likewise.
	* math/s_catanhl.c (__catanhl): Likewise.
	* math/s_catanl.c (__catanl): Likewise.
	* math/s_ccosh.c (__ccosh): Likewise.
	* math/s_ccoshf.c (__ccoshf): Likewise.
	* math/s_ccoshl.c (__ccoshl): Likewise.
	* math/s_cexp.c (__cexp): Likewise.
	* math/s_cexpf.c (__cexpf): Likewise.
	* math/s_cexpl.c (__cexpl): Likewise.
	* math/s_clog.c (__clog): Use math_check_force_underflow_nonneg.
	* math/s_clog10.c (__clog10): Likewise.
	* math/s_clog10f.c (__clog10f): Likewise.
	* math/s_clog10l.c (__clog10l): Likewise.
	* math/s_clogf.c (__clogf): Likewise.
	* math/s_clogl.c (__clogl): Likewise.
	* math/s_csin.c (__csin): Use math_check_force_underflow_complex.
	* math/s_csinf.c (__csinf): Likewise.
	* math/s_csinh.c (__csinh): Likewise.
	* math/s_csinhf.c (__csinhf): Likewise.
	* math/s_csinhl.c (__csinhl): Likewise.
	* math/s_csinl.c (__csinl): Likewise.
	* math/s_csqrt.c (__csqrt): Use math_check_force_underflow.
	* math/s_csqrtf.c (__csqrtf): Likewise.
	* math/s_csqrtl.c (__csqrtl): Likewise.
	* math/s_ctan.c (__ctan): Use math_check_force_underflow_complex.
	* math/s_ctanf.c (__ctanf): Likewise.
	* math/s_ctanh.c (__ctanh): Likewise.
	* math/s_ctanhf.c (__ctanhf): Likewise.
	* math/s_ctanhl.c (__ctanhl): Likewise.
	* math/s_ctanl.c (__ctanl): Likewise.
	* stdlib/strtod_l.c (round_and_return): Use math_force_eval
	instead of volatile.
	* sysdeps/ieee754/dbl-64/e_asin.c (__ieee754_asin): Use
	math_check_force_underflow.
	* sysdeps/ieee754/dbl-64/e_atanh.c (__ieee754_atanh): Likewise.
	* sysdeps/ieee754/dbl-64/e_exp.c (__ieee754_exp): Do not use
	volatile when forcing underflow.
	* sysdeps/ieee754/dbl-64/e_exp2.c (__ieee754_exp2): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/dbl-64/e_gamma_r.c (__ieee754_gamma_r):
	Likewise.
	* sysdeps/ieee754/dbl-64/e_j1.c (__ieee754_j1): Use
	math_check_force_underflow.
	* sysdeps/ieee754/dbl-64/e_jn.c (__ieee754_jn): Likewise.
	* sysdeps/ieee754/dbl-64/e_sinh.c (__ieee754_sinh): Likewise.
	* sysdeps/ieee754/dbl-64/s_asinh.c (__asinh): Likewise.
	* sysdeps/ieee754/dbl-64/s_atan.c (atan): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/dbl-64/s_erf.c (__erf): Use
	math_check_force_underflow.
	* sysdeps/ieee754/dbl-64/s_expm1.c (__expm1): Likewise.
	* sysdeps/ieee754/dbl-64/s_fma.c (__fma): Use math_force_eval
	instead of volatile.
	* sysdeps/ieee754/dbl-64/s_log1p.c (__log1p): Use
	math_check_force_underflow.
	* sysdeps/ieee754/dbl-64/s_sin.c (__sin): Likewise.
	* sysdeps/ieee754/dbl-64/s_tan.c (tan): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/dbl-64/s_tanh.c (__tanh): Use
	math_check_force_underflow.
	* sysdeps/ieee754/flt-32/e_asinf.c (__ieee754_asinf): Likewise.
	* sysdeps/ieee754/flt-32/e_atanhf.c (__ieee754_atanhf): Likewise.
	* sysdeps/ieee754/flt-32/e_exp2f.c (__ieee754_exp2f): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/flt-32/e_gammaf_r.c (__ieee754_gammaf_r):
	Likewise.
	* sysdeps/ieee754/flt-32/e_j1f.c (__ieee754_j1f): Use
	math_check_force_underflow.
	* sysdeps/ieee754/flt-32/e_jnf.c (__ieee754_jnf): Likewise.
	* sysdeps/ieee754/flt-32/e_sinhf.c (__ieee754_sinhf): Likewise.
	* sysdeps/ieee754/flt-32/k_sinf.c (__kernel_sinf): Likewise.
	* sysdeps/ieee754/flt-32/k_tanf.c (__kernel_tanf): Likewise.
	* sysdeps/ieee754/flt-32/s_asinhf.c (__asinhf): Likewise.
	* sysdeps/ieee754/flt-32/s_atanf.c (__atanf): Likewise.
	* sysdeps/ieee754/flt-32/s_erff.c (__erff): Likewise.
	* sysdeps/ieee754/flt-32/s_expm1f.c (__expm1f): Likewise.
	* sysdeps/ieee754/flt-32/s_log1pf.c (__log1pf): Likewise.
	* sysdeps/ieee754/flt-32/s_tanhf.c (__tanhf): Likewise.
	* sysdeps/ieee754/ldbl-128/e_asinl.c (__ieee754_asinl): Likewise.
	* sysdeps/ieee754/ldbl-128/e_atanhl.c (__ieee754_atanhl):
	Likewise.
	* sysdeps/ieee754/ldbl-128/e_expl.c (__ieee754_expl): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/ldbl-128/e_gammal_r.c (__ieee754_gammal_r):
	Likewise.
	* sysdeps/ieee754/ldbl-128/e_j1l.c (__ieee754_j1l): Use
	math_check_force_underflow.
	* sysdeps/ieee754/ldbl-128/e_jnl.c (__ieee754_jnl): Likewise.
	* sysdeps/ieee754/ldbl-128/e_sinhl.c (__ieee754_sinhl): Likewise.
	* sysdeps/ieee754/ldbl-128/k_sincosl.c (__kernel_sincosl):
	Likewise.
	* sysdeps/ieee754/ldbl-128/k_sinl.c (__kernel_sinl): Likewise.
	* sysdeps/ieee754/ldbl-128/k_tanl.c (__kernel_tanl): Likewise.
	* sysdeps/ieee754/ldbl-128/s_asinhl.c (__asinhl): Likewise.
	* sysdeps/ieee754/ldbl-128/s_atanl.c (__atanl): Likewise.
	* sysdeps/ieee754/ldbl-128/s_erfl.c (__erfl): Likewise.
	* sysdeps/ieee754/ldbl-128/s_expm1l.c (__expm1l): Likewise.
	* sysdeps/ieee754/ldbl-128/s_fmal.c (__fmal): Use math_force_eval
	instead of volatile.
	* sysdeps/ieee754/ldbl-128/s_log1pl.c (__log1pl): Use
	math_check_force_underflow.
	* sysdeps/ieee754/ldbl-128/s_tanhl.c (__tanhl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_asinl.c (__ieee754_asinl): Use
	math_check_force_underflow.
	* sysdeps/ieee754/ldbl-128ibm/e_atanhl.c (__ieee754_atanhl):
	Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (__ieee754_gammal_r):
	Use math_check_force_underflow_nonneg.
	* sysdeps/ieee754/ldbl-128ibm/e_jnl.c (__ieee754_jnl): Use
	math_check_force_underflow.
	* sysdeps/ieee754/ldbl-128ibm/e_sinhl.c (__ieee754_sinhl):
	Likewise.
	* sysdeps/ieee754/ldbl-128ibm/k_sincosl.c (__kernel_sincosl):
	Likewise.
	* sysdeps/ieee754/ldbl-128ibm/k_sinl.c (__kernel_sinl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/k_tanl.c (__kernel_tanl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_asinhl.c (__asinhl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_atanl.c (__atanl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_erfl.c (__erfl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_tanhl.c (__tanhl): Likewise.
	* sysdeps/ieee754/ldbl-96/e_asinl.c (__ieee754_asinl): Likewise.
	* sysdeps/ieee754/ldbl-96/e_atanhl.c (__ieee754_atanhl): Likewise.
	* sysdeps/ieee754/ldbl-96/e_gammal_r.c (__ieee754_gammal_r): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/ldbl-96/e_j1l.c (__ieee754_j1l): Use
	math_check_force_underflow.
	* sysdeps/ieee754/ldbl-96/e_jnl.c (__ieee754_jnl): Likewise.
	* sysdeps/ieee754/ldbl-96/e_sinhl.c (__ieee754_sinhl): Likewise.
	* sysdeps/ieee754/ldbl-96/k_sinl.c (__kernel_sinl): Likewise.
	* sysdeps/ieee754/ldbl-96/k_tanl.c (__kernel_tanl): Use
	math_check_force_underflow_nonneg.
	* sysdeps/ieee754/ldbl-96/s_asinhl.c (__asinhl): Use
	math_check_force_underflow.
	* sysdeps/ieee754/ldbl-96/s_erfl.c (__erfl): Likewise.
	* sysdeps/ieee754/ldbl-96/s_fmal.c (__fmal): Use math_force_eval
	instead of volatile.
	* sysdeps/ieee754/ldbl-96/s_tanhl.c (__tanhl): Use
	math_check_force_underflow.
2015-09-23 22:42:30 +00:00

362 lines
9.6 KiB
C

/*
* IBM Accurate Mathematical Library
* written by International Business Machines Corp.
* Copyright (C) 2001-2015 Free Software Foundation, Inc.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation; either version 2.1 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
/***************************************************************************/
/* MODULE_NAME:uexp.c */
/* */
/* FUNCTION:uexp */
/* exp1 */
/* */
/* FILES NEEDED:dla.h endian.h mpa.h mydefs.h uexp.h */
/* mpa.c mpexp.x slowexp.c */
/* */
/* An ultimate exp routine. Given an IEEE double machine number x */
/* it computes the correctly rounded (to nearest) value of e^x */
/* Assumption: Machine arithmetic operations are performed in */
/* round to nearest mode of IEEE 754 standard. */
/* */
/***************************************************************************/
#include <math.h>
#include "endian.h"
#include "uexp.h"
#include "mydefs.h"
#include "MathLib.h"
#include "uexp.tbl"
#include <math_private.h>
#include <fenv.h>
#include <float.h>
#ifndef SECTION
# define SECTION
#endif
double __slowexp (double);
/* An ultimate exp routine. Given an IEEE double machine number x it computes
the correctly rounded (to nearest) value of e^x. */
double
SECTION
__ieee754_exp (double x)
{
double bexp, t, eps, del, base, y, al, bet, res, rem, cor;
mynumber junk1, junk2, binexp = {{0, 0}};
int4 i, j, m, n, ex;
double retval;
{
SET_RESTORE_ROUND (FE_TONEAREST);
junk1.x = x;
m = junk1.i[HIGH_HALF];
n = m & hugeint;
if (n > smallint && n < bigint)
{
y = x * log2e.x + three51.x;
bexp = y - three51.x; /* multiply the result by 2**bexp */
junk1.x = y;
eps = bexp * ln_two2.x; /* x = bexp*ln(2) + t - eps */
t = x - bexp * ln_two1.x;
y = t + three33.x;
base = y - three33.x; /* t rounded to a multiple of 2**-18 */
junk2.x = y;
del = (t - base) - eps; /* x = bexp*ln(2) + base + del */
eps = del + del * del * (p3.x * del + p2.x);
binexp.i[HIGH_HALF] = (junk1.i[LOW_HALF] + 1023) << 20;
i = ((junk2.i[LOW_HALF] >> 8) & 0xfffffffe) + 356;
j = (junk2.i[LOW_HALF] & 511) << 1;
al = coar.x[i] * fine.x[j];
bet = ((coar.x[i] * fine.x[j + 1] + coar.x[i + 1] * fine.x[j])
+ coar.x[i + 1] * fine.x[j + 1]);
rem = (bet + bet * eps) + al * eps;
res = al + rem;
cor = (al - res) + rem;
if (res == (res + cor * err_0))
{
retval = res * binexp.x;
goto ret;
}
else
{
retval = __slowexp (x);
goto ret;
} /*if error is over bound */
}
if (n <= smallint)
{
retval = 1.0;
goto ret;
}
if (n >= badint)
{
if (n > infint)
{
retval = x + x;
goto ret;
} /* x is NaN */
if (n < infint)
{
if (x > 0)
goto ret_huge;
else
goto ret_tiny;
}
/* x is finite, cause either overflow or underflow */
if (junk1.i[LOW_HALF] != 0)
{
retval = x + x;
goto ret;
} /* x is NaN */
retval = (x > 0) ? inf.x : zero; /* |x| = inf; return either inf or 0 */
goto ret;
}
y = x * log2e.x + three51.x;
bexp = y - three51.x;
junk1.x = y;
eps = bexp * ln_two2.x;
t = x - bexp * ln_two1.x;
y = t + three33.x;
base = y - three33.x;
junk2.x = y;
del = (t - base) - eps;
eps = del + del * del * (p3.x * del + p2.x);
i = ((junk2.i[LOW_HALF] >> 8) & 0xfffffffe) + 356;
j = (junk2.i[LOW_HALF] & 511) << 1;
al = coar.x[i] * fine.x[j];
bet = ((coar.x[i] * fine.x[j + 1] + coar.x[i + 1] * fine.x[j])
+ coar.x[i + 1] * fine.x[j + 1]);
rem = (bet + bet * eps) + al * eps;
res = al + rem;
cor = (al - res) + rem;
if (m >> 31)
{
ex = junk1.i[LOW_HALF];
if (res < 1.0)
{
res += res;
cor += cor;
ex -= 1;
}
if (ex >= -1022)
{
binexp.i[HIGH_HALF] = (1023 + ex) << 20;
if (res == (res + cor * err_0))
{
retval = res * binexp.x;
goto ret;
}
else
{
retval = __slowexp (x);
goto check_uflow_ret;
} /*if error is over bound */
}
ex = -(1022 + ex);
binexp.i[HIGH_HALF] = (1023 - ex) << 20;
res *= binexp.x;
cor *= binexp.x;
eps = 1.0000000001 + err_0 * binexp.x;
t = 1.0 + res;
y = ((1.0 - t) + res) + cor;
res = t + y;
cor = (t - res) + y;
if (res == (res + eps * cor))
{
binexp.i[HIGH_HALF] = 0x00100000;
retval = (res - 1.0) * binexp.x;
goto check_uflow_ret;
}
else
{
retval = __slowexp (x);
goto check_uflow_ret;
} /* if error is over bound */
check_uflow_ret:
if (retval < DBL_MIN)
{
double force_underflow = tiny * tiny;
math_force_eval (force_underflow);
}
if (retval == 0)
goto ret_tiny;
goto ret;
}
else
{
binexp.i[HIGH_HALF] = (junk1.i[LOW_HALF] + 767) << 20;
if (res == (res + cor * err_0))
retval = res * binexp.x * t256.x;
else
retval = __slowexp (x);
if (isinf (retval))
goto ret_huge;
else
goto ret;
}
}
ret:
return retval;
ret_huge:
return hhuge * hhuge;
ret_tiny:
return tiny * tiny;
}
#ifndef __ieee754_exp
strong_alias (__ieee754_exp, __exp_finite)
#endif
/* Compute e^(x+xx). The routine also receives bound of error of previous
calculation. If after computing exp the error exceeds the allowed bounds,
the routine returns a non-positive number. Otherwise it returns the
computed result, which is always positive. */
double
SECTION
__exp1 (double x, double xx, double error)
{
double bexp, t, eps, del, base, y, al, bet, res, rem, cor;
mynumber junk1, junk2, binexp = {{0, 0}};
int4 i, j, m, n, ex;
junk1.x = x;
m = junk1.i[HIGH_HALF];
n = m & hugeint; /* no sign */
if (n > smallint && n < bigint)
{
y = x * log2e.x + three51.x;
bexp = y - three51.x; /* multiply the result by 2**bexp */
junk1.x = y;
eps = bexp * ln_two2.x; /* x = bexp*ln(2) + t - eps */
t = x - bexp * ln_two1.x;
y = t + three33.x;
base = y - three33.x; /* t rounded to a multiple of 2**-18 */
junk2.x = y;
del = (t - base) + (xx - eps); /* x = bexp*ln(2) + base + del */
eps = del + del * del * (p3.x * del + p2.x);
binexp.i[HIGH_HALF] = (junk1.i[LOW_HALF] + 1023) << 20;
i = ((junk2.i[LOW_HALF] >> 8) & 0xfffffffe) + 356;
j = (junk2.i[LOW_HALF] & 511) << 1;
al = coar.x[i] * fine.x[j];
bet = ((coar.x[i] * fine.x[j + 1] + coar.x[i + 1] * fine.x[j])
+ coar.x[i + 1] * fine.x[j + 1]);
rem = (bet + bet * eps) + al * eps;
res = al + rem;
cor = (al - res) + rem;
if (res == (res + cor * (1.0 + error + err_1)))
return res * binexp.x;
else
return -10.0;
}
if (n <= smallint)
return 1.0; /* if x->0 e^x=1 */
if (n >= badint)
{
if (n > infint)
return (zero / zero); /* x is NaN, return invalid */
if (n < infint)
return ((x > 0) ? (hhuge * hhuge) : (tiny * tiny));
/* x is finite, cause either overflow or underflow */
if (junk1.i[LOW_HALF] != 0)
return (zero / zero); /* x is NaN */
return ((x > 0) ? inf.x : zero); /* |x| = inf; return either inf or 0 */
}
y = x * log2e.x + three51.x;
bexp = y - three51.x;
junk1.x = y;
eps = bexp * ln_two2.x;
t = x - bexp * ln_two1.x;
y = t + three33.x;
base = y - three33.x;
junk2.x = y;
del = (t - base) + (xx - eps);
eps = del + del * del * (p3.x * del + p2.x);
i = ((junk2.i[LOW_HALF] >> 8) & 0xfffffffe) + 356;
j = (junk2.i[LOW_HALF] & 511) << 1;
al = coar.x[i] * fine.x[j];
bet = ((coar.x[i] * fine.x[j + 1] + coar.x[i + 1] * fine.x[j])
+ coar.x[i + 1] * fine.x[j + 1]);
rem = (bet + bet * eps) + al * eps;
res = al + rem;
cor = (al - res) + rem;
if (m >> 31)
{
ex = junk1.i[LOW_HALF];
if (res < 1.0)
{
res += res;
cor += cor;
ex -= 1;
}
if (ex >= -1022)
{
binexp.i[HIGH_HALF] = (1023 + ex) << 20;
if (res == (res + cor * (1.0 + error + err_1)))
return res * binexp.x;
else
return -10.0;
}
ex = -(1022 + ex);
binexp.i[HIGH_HALF] = (1023 - ex) << 20;
res *= binexp.x;
cor *= binexp.x;
eps = 1.00000000001 + (error + err_1) * binexp.x;
t = 1.0 + res;
y = ((1.0 - t) + res) + cor;
res = t + y;
cor = (t - res) + y;
if (res == (res + eps * cor))
{
binexp.i[HIGH_HALF] = 0x00100000;
return (res - 1.0) * binexp.x;
}
else
return -10.0;
}
else
{
binexp.i[HIGH_HALF] = (junk1.i[LOW_HALF] + 767) << 20;
if (res == (res + cor * (1.0 + error + err_1)))
return res * binexp.x * t256.x;
else
return -10.0;
}
}