2006-01-28 00:15:15 +00:00
|
|
|
/* expm1l.c
|
|
|
|
*
|
|
|
|
* Exponential function, minus 1
|
|
|
|
* 128-bit long double precision
|
|
|
|
*
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* SYNOPSIS:
|
|
|
|
*
|
|
|
|
* long double x, y, expm1l();
|
|
|
|
*
|
|
|
|
* y = expm1l( x );
|
|
|
|
*
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* DESCRIPTION:
|
|
|
|
*
|
|
|
|
* Returns e (2.71828...) raised to the x power, minus one.
|
|
|
|
*
|
|
|
|
* Range reduction is accomplished by separating the argument
|
|
|
|
* into an integer k and fraction f such that
|
|
|
|
*
|
|
|
|
* x k f
|
|
|
|
* e = 2 e.
|
|
|
|
*
|
|
|
|
* An expansion x + .5 x^2 + x^3 R(x) approximates exp(f) - 1
|
|
|
|
* in the basic range [-0.5 ln 2, 0.5 ln 2].
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* ACCURACY:
|
|
|
|
*
|
|
|
|
* Relative error:
|
|
|
|
* arithmetic domain # trials peak rms
|
|
|
|
* IEEE -79,+MAXLOG 100,000 1.7e-34 4.5e-35
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* Copyright 2001 by Stephen L. Moshier
|
|
|
|
|
|
|
|
This library 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 library 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
|
2012-02-09 23:18:22 +00:00
|
|
|
License along with this library; if not, see
|
|
|
|
<http://www.gnu.org/licenses/>. */
|
2006-01-28 00:15:15 +00:00
|
|
|
|
2009-05-22 15:35:15 +00:00
|
|
|
#include <errno.h>
|
2012-03-09 19:29:16 +00:00
|
|
|
#include <math.h>
|
|
|
|
#include <math_private.h>
|
2006-01-28 00:15:15 +00:00
|
|
|
#include <math_ldbl_opt.h>
|
|
|
|
|
|
|
|
/* exp(x) - 1 = x + 0.5 x^2 + x^3 P(x)/Q(x)
|
|
|
|
-.5 ln 2 < x < .5 ln 2
|
|
|
|
Theoretical peak relative error = 8.1e-36 */
|
|
|
|
|
|
|
|
static const long double
|
|
|
|
P0 = 2.943520915569954073888921213330863757240E8L,
|
|
|
|
P1 = -5.722847283900608941516165725053359168840E7L,
|
|
|
|
P2 = 8.944630806357575461578107295909719817253E6L,
|
|
|
|
P3 = -7.212432713558031519943281748462837065308E5L,
|
|
|
|
P4 = 4.578962475841642634225390068461943438441E4L,
|
|
|
|
P5 = -1.716772506388927649032068540558788106762E3L,
|
|
|
|
P6 = 4.401308817383362136048032038528753151144E1L,
|
|
|
|
P7 = -4.888737542888633647784737721812546636240E-1L,
|
|
|
|
Q0 = 1.766112549341972444333352727998584753865E9L,
|
|
|
|
Q1 = -7.848989743695296475743081255027098295771E8L,
|
|
|
|
Q2 = 1.615869009634292424463780387327037251069E8L,
|
|
|
|
Q3 = -2.019684072836541751428967854947019415698E7L,
|
|
|
|
Q4 = 1.682912729190313538934190635536631941751E6L,
|
|
|
|
Q5 = -9.615511549171441430850103489315371768998E4L,
|
|
|
|
Q6 = 3.697714952261803935521187272204485251835E3L,
|
|
|
|
Q7 = -8.802340681794263968892934703309274564037E1L,
|
|
|
|
/* Q8 = 1.000000000000000000000000000000000000000E0 */
|
|
|
|
/* C1 + C2 = ln 2 */
|
|
|
|
|
|
|
|
C1 = 6.93145751953125E-1L,
|
|
|
|
C2 = 1.428606820309417232121458176568075500134E-6L,
|
|
|
|
/* ln 2^-114 */
|
2012-11-22 14:55:17 +00:00
|
|
|
minarg = -7.9018778583833765273564461846232128760607E1L, big = 1e290L;
|
2006-01-28 00:15:15 +00:00
|
|
|
|
|
|
|
|
|
|
|
long double
|
|
|
|
__expm1l (long double x)
|
|
|
|
{
|
|
|
|
long double px, qx, xx;
|
2013-08-17 08:54:58 +00:00
|
|
|
int32_t ix, lx, sign;
|
2006-01-28 00:15:15 +00:00
|
|
|
int k;
|
2013-08-17 08:54:58 +00:00
|
|
|
double xhi;
|
2006-01-28 00:15:15 +00:00
|
|
|
|
|
|
|
/* Detect infinity and NaN. */
|
2013-08-17 08:54:58 +00:00
|
|
|
xhi = ldbl_high (x);
|
|
|
|
EXTRACT_WORDS (ix, lx, xhi);
|
2006-01-28 00:15:15 +00:00
|
|
|
sign = ix & 0x80000000;
|
|
|
|
ix &= 0x7fffffff;
|
Fix ldbl-128ibm expm1l on large arguments (bug 16408).
This patch fixes bug 16408, ldbl-128ibm expm1l returning NaN for some
large arguments.
The basic problem is that the approach of converting the exponent to
the form n * log(2) + y, where -0.5 <= y <= 0.5, then computing 2^n *
expm1(y) + (2^n - 1) falls over when 2^n overflows (starting slightly
before the point where expm1 overflows, when y is negative and n is
the least integer for which 2^n overflows). The ldbl-128 code, and
the x86/x86_64 code, make expm1l fall back to expl for large positive
arguments to avoid this issue. This patch makes the ldbl-128ibm code
do the same. (The problem appears for the particular argument in the
testsuite because the ldbl-128ibm code also uses an overflow threshold
that's for ldbl-128 and is too big for ldbl-128ibm, but the problem
described applies for large non-overflowing cases as well, although
during the freeze is not a suitable time for making the expm1 tests
cover cases close to overflow more thoroughly.)
This leaves some code for large positive arguments in expm1l that is
now dead. To keep the code for ldbl-128 and ldbl-128ibm similar, and
to avoid unnecessary changes during the freeze, the patch doesn't
remove it; instead I propose to file a bug in Bugzilla as a reminder
that this code (for overflow, including errno setting, and for
arguments of +Inf) is no longer needed and should be removed from both
those expm1l implementations.
Tested powerpc32.
* sysdeps/ieee754/ldbl-128ibm/s_expm1l.c (__expm1l): Use __expl
for large positive arguments.
2014-01-08 13:32:39 +00:00
|
|
|
if (!sign && ix >= 0x40600000)
|
|
|
|
return __expl (x);
|
2006-01-28 00:15:15 +00:00
|
|
|
if (ix >= 0x7ff00000)
|
|
|
|
{
|
2015-09-16 16:42:46 +00:00
|
|
|
/* Infinity (which must be negative infinity). */
|
2013-08-17 08:54:58 +00:00
|
|
|
if (((ix - 0x7ff00000) | lx) == 0)
|
2015-09-16 16:42:46 +00:00
|
|
|
return -1.0L;
|
2016-06-09 17:24:52 +00:00
|
|
|
/* NaN. Invalid exception if signaling. */
|
|
|
|
return x + x;
|
2006-01-28 00:15:15 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* expm1(+- 0) = +- 0. */
|
2013-08-17 08:54:58 +00:00
|
|
|
if ((ix | lx) == 0)
|
2006-01-28 00:15:15 +00:00
|
|
|
return x;
|
|
|
|
|
|
|
|
/* Minimum value. */
|
|
|
|
if (x < minarg)
|
|
|
|
return (4.0/big - 1.0L);
|
|
|
|
|
|
|
|
/* Express x = ln 2 (k + remainder), remainder not exceeding 1/2. */
|
|
|
|
xx = C1 + C2; /* ln 2. */
|
Use floor functions not __floor functions in glibc libm.
Similar to the changes that were made to call sqrt functions directly
in glibc, instead of __ieee754_sqrt variants, so that the compiler
could inline them automatically without needing special inline
definitions in lots of math_private.h headers, this patch makes libm
code call floor functions directly instead of __floor variants,
removing the inlines / macros for x86_64 (SSE4.1) and powerpc
(POWER5).
The redirection used to ensure that __ieee754_sqrt does still get
called when the compiler doesn't inline a built-in function expansion
is refactored so it can be applied to other functions; the refactoring
is arranged so it's not limited to unary functions either (it would be
reasonable to use this mechanism for copysign - removing the inline in
math_private_calls.h but also eliminating unnecessary local PLT entry
use in the cases (powerpc soft-float and e500v1, for IBM long double)
where copysign calls don't get inlined).
The point of this change is that more architectures can get floor
calls inlined where they weren't previously (AArch64, for example),
without needing special inline definitions in their math_private.h,
and existing such definitions in math_private.h headers can be
removed.
Note that it's possible that in some cases an inline may be used where
an IFUNC call was previously used - this is the case on x86_64, for
example. I think the direct calls to floor are still appropriate; if
there's any significant performance cost from inline SSE2 floor
instead of an IFUNC call ending up with SSE4.1 floor, that indicates
that either the function should be doing something else that's faster
than using floor at all, or it should itself have IFUNC variants, or
that the compiler choice of inlining for generic tuning should change
to allow for the possibility that, by not inlining, an SSE4.1 IFUNC
might be called at runtime - but not that glibc should avoid calling
floor internally. (After all, all the same considerations would apply
to any user program calling floor, where it might either be inlined or
left as an out-of-line call allowing for a possible IFUNC.)
Tested for x86_64, and with build-many-glibcs.py.
* include/math.h [!_ISOMAC && !(__FINITE_MATH_ONLY__ &&
__FINITE_MATH_ONLY__ > 0) && !NO_MATH_REDIRECT] (MATH_REDIRECT):
New macro.
[!_ISOMAC && !(__FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
&& !NO_MATH_REDIRECT] (MATH_REDIRECT_LDBL): Likewise.
[!_ISOMAC && !(__FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
&& !NO_MATH_REDIRECT] (MATH_REDIRECT_F128): Likewise.
[!_ISOMAC && !(__FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
&& !NO_MATH_REDIRECT] (MATH_REDIRECT_UNARY_ARGS): Likewise.
[!_ISOMAC && !(__FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
&& !NO_MATH_REDIRECT] (sqrt): Redirect using MATH_REDIRECT.
[!_ISOMAC && !(__FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0)
&& !NO_MATH_REDIRECT] (floor): Likewise.
* sysdeps/aarch64/fpu/s_floor.c: Define NO_MATH_REDIRECT before
header inclusion.
* sysdeps/aarch64/fpu/s_floorf.c: Likewise.
* sysdeps/ieee754/dbl-64/s_floor.c: Likewise.
* sysdeps/ieee754/dbl-64/wordsize-64/s_floor.c: Likewise.
* sysdeps/ieee754/float128/s_floorf128.c: Likewise.
* sysdeps/ieee754/flt-32/s_floorf.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_floorl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_floorl.c: Likewise.
* sysdeps/m68k/m680x0/fpu/s_floor_template.c: Likewise.
* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_floor.c: Likewise.
* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_floorf.c: Likewise.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_floor.c: Likewise.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_floorf.c: Likewise.
* sysdeps/riscv/rv64/rvd/s_floor.c: Likewise.
* sysdeps/riscv/rvf/s_floorf.c: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floor.c: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floorf.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_floor.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/s_floorf.c: Likewise.
* sysdeps/powerpc/fpu/math_private.h [_ARCH_PWR5X] (__floor):
Remove macro.
[_ARCH_PWR5X] (__floorf): Likewise.
* sysdeps/x86_64/fpu/math_private.h [__SSE4_1__] (__floor): Remove
inline function.
[__SSE4_1__] (__floorf): Likewise.
* math/w_lgamma_main.c (LGFUNC (__lgamma)): Use floor functions
instead of __floor variants.
* math/w_lgamma_r_compat.c (__lgamma_r): Likewise.
* math/w_lgammaf_main.c (LGFUNC (__lgammaf)): Likewise.
* math/w_lgammaf_r_compat.c (__lgammaf_r): Likewise.
* math/w_lgammal_main.c (LGFUNC (__lgammal)): Likewise.
* math/w_lgammal_r_compat.c (__lgammal_r): Likewise.
* math/w_tgamma_compat.c (__tgamma): Likewise.
* math/w_tgamma_template.c (M_DECL_FUNC (__tgamma)): Likewise.
* math/w_tgammaf_compat.c (__tgammaf): Likewise.
* math/w_tgammal_compat.c (__tgammal): Likewise.
* sysdeps/ieee754/dbl-64/e_lgamma_r.c (sin_pi): Likewise.
* sysdeps/ieee754/dbl-64/k_rem_pio2.c (__kernel_rem_pio2):
Likewise.
* sysdeps/ieee754/dbl-64/lgamma_neg.c (__lgamma_neg): Likewise.
* sysdeps/ieee754/flt-32/e_lgammaf_r.c (sin_pif): Likewise.
* sysdeps/ieee754/flt-32/lgamma_negf.c (__lgamma_negf): Likewise.
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c (__ieee754_lgammal_r):
Likewise.
* sysdeps/ieee754/ldbl-128/e_powl.c (__ieee754_powl): Likewise.
* sysdeps/ieee754/ldbl-128/lgamma_negl.c (__lgamma_negl):
Likewise.
* sysdeps/ieee754/ldbl-128/s_expm1l.c (__expm1l): Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_lgammal_r.c (__ieee754_lgammal_r):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_powl.c (__ieee754_powl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/lgamma_negl.c (__lgamma_negl):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_expm1l.c (__expm1l): Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_truncl.c (__truncl): Likewise.
* sysdeps/ieee754/ldbl-96/e_lgammal_r.c (sin_pi): Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_negl.c (__lgamma_negl): Likewise.
* sysdeps/powerpc/power5+/fpu/s_modf.c (__modf): Likewise.
* sysdeps/powerpc/power5+/fpu/s_modff.c (__modff): Likewise.
2018-09-14 13:09:01 +00:00
|
|
|
px = floorl (0.5 + x / xx);
|
2006-01-28 00:15:15 +00:00
|
|
|
k = px;
|
|
|
|
/* remainder times ln 2 */
|
|
|
|
x -= px * C1;
|
|
|
|
x -= px * C2;
|
|
|
|
|
|
|
|
/* Approximate exp(remainder ln 2). */
|
|
|
|
px = (((((((P7 * x
|
|
|
|
+ P6) * x
|
|
|
|
+ P5) * x + P4) * x + P3) * x + P2) * x + P1) * x + P0) * x;
|
|
|
|
|
|
|
|
qx = (((((((x
|
|
|
|
+ Q7) * x
|
|
|
|
+ Q6) * x + Q5) * x + Q4) * x + Q3) * x + Q2) * x + Q1) * x + Q0;
|
|
|
|
|
|
|
|
xx = x * x;
|
|
|
|
qx = x + (0.5 * xx + xx * px / qx);
|
|
|
|
|
|
|
|
/* exp(x) = exp(k ln 2) exp(remainder ln 2) = 2^k exp(remainder ln 2).
|
|
|
|
|
|
|
|
We have qx = exp(remainder ln 2) - 1, so
|
|
|
|
exp(x) - 1 = 2^k (qx + 1) - 1
|
|
|
|
= 2^k qx + 2^k - 1. */
|
|
|
|
|
2008-04-12 00:51:34 +00:00
|
|
|
px = __ldexpl (1.0L, k);
|
2006-01-28 00:15:15 +00:00
|
|
|
x = px * qx + (px - 1.0);
|
|
|
|
return x;
|
|
|
|
}
|
|
|
|
libm_hidden_def (__expm1l)
|
|
|
|
long_double_symbol (libm, __expm1l, expm1l);
|