glibc/math/math.h
Joseph Myers b59ad2db99 Fix iszero for excess precision.
Floating-point classification macros are supposed to remove any excess
range or precision from their arguments.  This patch fixes the
non-sNaN version of iszero to do so, by casting the argument to its
own type.  (This will of course work only for standard-conforming
excess precision, not for what GCC does on 32-bit x86 by default where
the back end hides excess precision from the front end; the same
applies to most of the classification macros in that case, as showed
up when we made them use GCC built-in functions.)

(iseqsig will have the reverse issue, needing to ensure that when an
underlying function is used it's for a type wide enough not to remove
any excess precision, since comparison macros must not remove excess
precision.)

Tested for x86_64 and x86.

	* math/math.h
	[__GLIBC_USE (IEC_60559_BFP_EXT) && !__SUPPORT_SNAN__] (iszero):
	Cast argument to its own type.
	* math/test-iszero-excess-precision.c: New file.
	* math/Makefile (tests): Add test-iszero-excess-precision.
	(CFLAGS-test-iszero-excess-precision.c): New variable.
2016-09-28 21:11:58 +00:00

536 lines
18 KiB
C

/* Declarations for math functions.
Copyright (C) 1991-2016 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C 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.
The GNU C 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
License along with the GNU C Library; if not, see
<http://www.gnu.org/licenses/>. */
/*
* ISO C99 Standard: 7.12 Mathematics <math.h>
*/
#ifndef _MATH_H
#define _MATH_H 1
#define __GLIBC_INTERNAL_STARTING_HEADER_IMPLEMENTATION
#include <bits/libc-header-start.h>
__BEGIN_DECLS
/* Get machine-dependent vector math functions declarations. */
#include <bits/math-vector.h>
/* Get machine-dependent HUGE_VAL value (returned on overflow).
On all IEEE754 machines, this is +Infinity. */
#include <bits/huge_val.h>
#ifdef __USE_ISOC99
# include <bits/huge_valf.h>
# include <bits/huge_vall.h>
/* Get machine-dependent INFINITY value. */
# include <bits/inf.h>
/* Get machine-dependent NAN value (returned for some domain errors). */
# include <bits/nan.h>
#endif /* __USE_ISOC99 */
/* Get the architecture specific values describing the floating-point
evaluation. The following symbols will get defined:
float_t floating-point type at least as wide as `float' used
to evaluate `float' expressions
double_t floating-point type at least as wide as `double' used
to evaluate `double' expressions
FP_FAST_FMA
FP_FAST_FMAF
FP_FAST_FMAL
If defined it indicates that the `fma' function
generally executes about as fast as a multiply and an add.
This macro is defined only iff the `fma' function is
implemented directly with a hardware multiply-add instructions.
FP_ILOGB0 Expands to a value returned by `ilogb (0.0)'.
FP_ILOGBNAN Expands to a value returned by `ilogb (NAN)'.
*/
#include <bits/mathdef.h>
/* The file <bits/mathcalls.h> contains the prototypes for all the
actual math functions. These macros are used for those prototypes,
so we can easily declare each function as both `name' and `__name',
and can declare the float versions `namef' and `__namef'. */
#define __SIMD_DECL(function) __CONCAT (__DECL_SIMD_, function)
#define __MATHCALL_VEC(function, suffix, args) \
__SIMD_DECL (__MATH_PRECNAME (function, suffix)) \
__MATHCALL (function, suffix, args)
#define __MATHDECL_VEC(type, function,suffix, args) \
__SIMD_DECL (__MATH_PRECNAME (function, suffix)) \
__MATHDECL(type, function,suffix, args)
#define __MATHCALL(function,suffix, args) \
__MATHDECL (_Mdouble_,function,suffix, args)
#define __MATHDECL(type, function,suffix, args) \
__MATHDECL_1(type, function,suffix, args); \
__MATHDECL_1(type, __CONCAT(__,function),suffix, args)
#define __MATHCALLX(function,suffix, args, attrib) \
__MATHDECLX (_Mdouble_,function,suffix, args, attrib)
#define __MATHDECLX(type, function,suffix, args, attrib) \
__MATHDECL_1(type, function,suffix, args) __attribute__ (attrib); \
__MATHDECL_1(type, __CONCAT(__,function),suffix, args) __attribute__ (attrib)
#define __MATHDECL_1(type, function,suffix, args) \
extern type __MATH_PRECNAME(function,suffix) args __THROW
#define _Mdouble_ double
#define __MATH_PRECNAME(name,r) __CONCAT(name,r)
#define __MATH_DECLARING_DOUBLE 1
#define _Mdouble_BEGIN_NAMESPACE __BEGIN_NAMESPACE_STD
#define _Mdouble_END_NAMESPACE __END_NAMESPACE_STD
#include <bits/mathcalls.h>
#undef _Mdouble_
#undef _Mdouble_BEGIN_NAMESPACE
#undef _Mdouble_END_NAMESPACE
#undef __MATH_PRECNAME
#undef __MATH_DECLARING_DOUBLE
#ifdef __USE_ISOC99
/* Include the file of declarations again, this time using `float'
instead of `double' and appending f to each function name. */
# ifndef _Mfloat_
# define _Mfloat_ float
# endif
# define _Mdouble_ _Mfloat_
# define __MATH_PRECNAME(name,r) name##f##r
# define __MATH_DECLARING_DOUBLE 0
# define _Mdouble_BEGIN_NAMESPACE __BEGIN_NAMESPACE_C99
# define _Mdouble_END_NAMESPACE __END_NAMESPACE_C99
# include <bits/mathcalls.h>
# undef _Mdouble_
# undef _Mdouble_BEGIN_NAMESPACE
# undef _Mdouble_END_NAMESPACE
# undef __MATH_PRECNAME
# undef __MATH_DECLARING_DOUBLE
# if !(defined __NO_LONG_DOUBLE_MATH && defined _LIBC) \
|| defined __LDBL_COMPAT \
|| defined _LIBC_TEST
# ifdef __LDBL_COMPAT
# ifdef __USE_ISOC99
extern float __nldbl_nexttowardf (float __x, long double __y)
__THROW __attribute__ ((__const__));
# ifdef __REDIRECT_NTH
extern float __REDIRECT_NTH (nexttowardf, (float __x, long double __y),
__nldbl_nexttowardf)
__attribute__ ((__const__));
extern double __REDIRECT_NTH (nexttoward, (double __x, long double __y),
nextafter) __attribute__ ((__const__));
extern long double __REDIRECT_NTH (nexttowardl,
(long double __x, long double __y),
nextafter) __attribute__ ((__const__));
# endif
# endif
# undef __MATHDECL_1
# define __MATHDECL_2(type, function,suffix, args, alias) \
extern type __REDIRECT_NTH(__MATH_PRECNAME(function,suffix), \
args, alias)
# define __MATHDECL_1(type, function,suffix, args) \
__MATHDECL_2(type, function,suffix, args, __CONCAT(function,suffix))
# endif
/* Include the file of declarations again, this time using `long double'
instead of `double' and appending l to each function name. */
# ifndef _Mlong_double_
# define _Mlong_double_ long double
# endif
# define _Mdouble_ _Mlong_double_
# define __MATH_PRECNAME(name,r) name##l##r
# define __MATH_DECLARING_DOUBLE 0
# define _Mdouble_BEGIN_NAMESPACE __BEGIN_NAMESPACE_C99
# define _Mdouble_END_NAMESPACE __END_NAMESPACE_C99
# define __MATH_DECLARE_LDOUBLE 1
# include <bits/mathcalls.h>
# undef _Mdouble_
# undef _Mdouble_BEGIN_NAMESPACE
# undef _Mdouble_END_NAMESPACE
# undef __MATH_PRECNAME
# undef __MATH_DECLARING_DOUBLE
# endif /* !(__NO_LONG_DOUBLE_MATH && _LIBC) || __LDBL_COMPAT */
#endif /* Use ISO C99. */
#undef __MATHDECL_1
#undef __MATHDECL
#undef __MATHCALL
#if defined __USE_MISC || defined __USE_XOPEN
/* This variable is used by `gamma' and `lgamma'. */
extern int signgam;
#endif
/* ISO C99 defines some generic macros which work on any data type. */
#ifdef __USE_ISOC99
/* All floating-point numbers can be put in one of these categories. */
enum
{
FP_NAN =
# define FP_NAN 0
FP_NAN,
FP_INFINITE =
# define FP_INFINITE 1
FP_INFINITE,
FP_ZERO =
# define FP_ZERO 2
FP_ZERO,
FP_SUBNORMAL =
# define FP_SUBNORMAL 3
FP_SUBNORMAL,
FP_NORMAL =
# define FP_NORMAL 4
FP_NORMAL
};
/* GCC bug 66462 means we cannot use the math builtins with -fsignaling-nan,
so disable builtins if this is enabled. When fixed in a newer GCC,
the __SUPPORT_SNAN__ check may be skipped for those versions. */
/* Return number of classification appropriate for X. */
# if __GNUC_PREREQ (4,4) && !defined __SUPPORT_SNAN__ \
&& !defined __OPTIMIZE_SIZE__
# define fpclassify(x) __builtin_fpclassify (FP_NAN, FP_INFINITE, \
FP_NORMAL, FP_SUBNORMAL, FP_ZERO, x)
# elif defined __NO_LONG_DOUBLE_MATH
# define fpclassify(x) \
(sizeof (x) == sizeof (float) ? __fpclassifyf (x) : __fpclassify (x))
# else
# define fpclassify(x) \
(sizeof (x) == sizeof (float) \
? __fpclassifyf (x) \
: sizeof (x) == sizeof (double) \
? __fpclassify (x) : __fpclassifyl (x))
# endif
/* Return nonzero value if sign of X is negative. */
# if __GNUC_PREREQ (4,0)
# define signbit(x) \
(sizeof (x) == sizeof (float) \
? __builtin_signbitf (x) \
: sizeof (x) == sizeof (double) \
? __builtin_signbit (x) : __builtin_signbitl (x))
# else
# ifdef __NO_LONG_DOUBLE_MATH
# define signbit(x) \
(sizeof (x) == sizeof (float) ? __signbitf (x) : __signbit (x))
# else
# define signbit(x) \
(sizeof (x) == sizeof (float) \
? __signbitf (x) \
: sizeof (x) == sizeof (double) \
? __signbit (x) : __signbitl (x))
# endif
# endif
/* Return nonzero value if X is not +-Inf or NaN. */
# if __GNUC_PREREQ (4,4) && !defined __SUPPORT_SNAN__
# define isfinite(x) __builtin_isfinite (x)
# elif defined __NO_LONG_DOUBLE_MATH
# define isfinite(x) \
(sizeof (x) == sizeof (float) ? __finitef (x) : __finite (x))
# else
# define isfinite(x) \
(sizeof (x) == sizeof (float) \
? __finitef (x) \
: sizeof (x) == sizeof (double) \
? __finite (x) : __finitel (x))
# endif
/* Return nonzero value if X is neither zero, subnormal, Inf, nor NaN. */
# if __GNUC_PREREQ (4,4) && !defined __SUPPORT_SNAN__
# define isnormal(x) __builtin_isnormal (x)
# else
# define isnormal(x) (fpclassify (x) == FP_NORMAL)
# endif
/* Return nonzero value if X is a NaN. We could use `fpclassify' but
we already have this functions `__isnan' and it is faster. */
# if __GNUC_PREREQ (4,4) && !defined __SUPPORT_SNAN__
# define isnan(x) __builtin_isnan (x)
# elif defined __NO_LONG_DOUBLE_MATH
# define isnan(x) \
(sizeof (x) == sizeof (float) ? __isnanf (x) : __isnan (x))
# else
# define isnan(x) \
(sizeof (x) == sizeof (float) \
? __isnanf (x) \
: sizeof (x) == sizeof (double) \
? __isnan (x) : __isnanl (x))
# endif
/* Return nonzero value if X is positive or negative infinity. */
# if __GNUC_PREREQ (4,4) && !defined __SUPPORT_SNAN__
# define isinf(x) __builtin_isinf_sign (x)
# elif defined __NO_LONG_DOUBLE_MATH
# define isinf(x) \
(sizeof (x) == sizeof (float) ? __isinff (x) : __isinf (x))
# else
# define isinf(x) \
(sizeof (x) == sizeof (float) \
? __isinff (x) \
: sizeof (x) == sizeof (double) \
? __isinf (x) : __isinfl (x))
# endif
/* Bitmasks for the math_errhandling macro. */
# define MATH_ERRNO 1 /* errno set by math functions. */
# define MATH_ERREXCEPT 2 /* Exceptions raised by math functions. */
/* By default all functions support both errno and exception handling.
In gcc's fast math mode and if inline functions are defined this
might not be true. */
# ifndef __FAST_MATH__
# define math_errhandling (MATH_ERRNO | MATH_ERREXCEPT)
# endif
#endif /* Use ISO C99. */
#if __GLIBC_USE (IEC_60559_BFP_EXT)
/* Return nonzero value if X is a signaling NaN. */
# ifdef __NO_LONG_DOUBLE_MATH
# define issignaling(x) \
(sizeof (x) == sizeof (float) ? __issignalingf (x) : __issignaling (x))
# else
# define issignaling(x) \
(sizeof (x) == sizeof (float) \
? __issignalingf (x) \
: sizeof (x) == sizeof (double) \
? __issignaling (x) : __issignalingl (x))
# endif
/* Return nonzero value if X is subnormal. */
# define issubnormal(x) (fpclassify (x) == FP_SUBNORMAL)
/* Return nonzero value if X is zero. */
# ifdef __SUPPORT_SNAN__
# define iszero(x) (fpclassify (x) == FP_ZERO)
# else
# define iszero(x) (((__typeof (x)) (x)) == 0)
# endif
#endif /* Use IEC_60559_BFP_EXT. */
#ifdef __USE_MISC
/* Support for various different standard error handling behaviors. */
typedef enum
{
_IEEE_ = -1, /* According to IEEE 754/IEEE 854. */
_SVID_, /* According to System V, release 4. */
_XOPEN_, /* Nowadays also Unix98. */
_POSIX_,
_ISOC_ /* Actually this is ISO C99. */
} _LIB_VERSION_TYPE;
/* This variable can be changed at run-time to any of the values above to
affect floating point error handling behavior (it may also be necessary
to change the hardware FPU exception settings). */
extern _LIB_VERSION_TYPE _LIB_VERSION;
#endif
#ifdef __USE_MISC
/* In SVID error handling, `matherr' is called with this description
of the exceptional condition.
We have a problem when using C++ since `exception' is a reserved
name in C++. */
# ifdef __cplusplus
struct __exception
# else
struct exception
# endif
{
int type;
char *name;
double arg1;
double arg2;
double retval;
};
# ifdef __cplusplus
extern int matherr (struct __exception *__exc) throw ();
# else
extern int matherr (struct exception *__exc);
# endif
# define X_TLOSS 1.41484755040568800000e+16
/* Types of exceptions in the `type' field. */
# define DOMAIN 1
# define SING 2
# define OVERFLOW 3
# define UNDERFLOW 4
# define TLOSS 5
# define PLOSS 6
/* SVID mode specifies returning this large value instead of infinity. */
# define HUGE 3.40282347e+38F
#else /* !Misc. */
# ifdef __USE_XOPEN
/* X/Open wants another strange constant. */
# define MAXFLOAT 3.40282347e+38F
# endif
#endif /* Misc. */
/* Some useful constants. */
#if defined __USE_MISC || defined __USE_XOPEN
# define M_E 2.7182818284590452354 /* e */
# define M_LOG2E 1.4426950408889634074 /* log_2 e */
# define M_LOG10E 0.43429448190325182765 /* log_10 e */
# define M_LN2 0.69314718055994530942 /* log_e 2 */
# define M_LN10 2.30258509299404568402 /* log_e 10 */
# define M_PI 3.14159265358979323846 /* pi */
# define M_PI_2 1.57079632679489661923 /* pi/2 */
# define M_PI_4 0.78539816339744830962 /* pi/4 */
# define M_1_PI 0.31830988618379067154 /* 1/pi */
# define M_2_PI 0.63661977236758134308 /* 2/pi */
# define M_2_SQRTPI 1.12837916709551257390 /* 2/sqrt(pi) */
# define M_SQRT2 1.41421356237309504880 /* sqrt(2) */
# define M_SQRT1_2 0.70710678118654752440 /* 1/sqrt(2) */
#endif
/* The above constants are not adequate for computation using `long double's.
Therefore we provide as an extension constants with similar names as a
GNU extension. Provide enough digits for the 128-bit IEEE quad. */
#ifdef __USE_GNU
# define M_El 2.718281828459045235360287471352662498L /* e */
# define M_LOG2El 1.442695040888963407359924681001892137L /* log_2 e */
# define M_LOG10El 0.434294481903251827651128918916605082L /* log_10 e */
# define M_LN2l 0.693147180559945309417232121458176568L /* log_e 2 */
# define M_LN10l 2.302585092994045684017991454684364208L /* log_e 10 */
# define M_PIl 3.141592653589793238462643383279502884L /* pi */
# define M_PI_2l 1.570796326794896619231321691639751442L /* pi/2 */
# define M_PI_4l 0.785398163397448309615660845819875721L /* pi/4 */
# define M_1_PIl 0.318309886183790671537767526745028724L /* 1/pi */
# define M_2_PIl 0.636619772367581343075535053490057448L /* 2/pi */
# define M_2_SQRTPIl 1.128379167095512573896158903121545172L /* 2/sqrt(pi) */
# define M_SQRT2l 1.414213562373095048801688724209698079L /* sqrt(2) */
# define M_SQRT1_2l 0.707106781186547524400844362104849039L /* 1/sqrt(2) */
#endif
/* When compiling in strict ISO C compatible mode we must not use the
inline functions since they, among other things, do not set the
`errno' variable correctly. */
#if defined __STRICT_ANSI__ && !defined __NO_MATH_INLINES
# define __NO_MATH_INLINES 1
#endif
#if defined __USE_ISOC99 && __GNUC_PREREQ(2,97)
/* ISO C99 defines some macros to compare number while taking care for
unordered numbers. Many FPUs provide special instructions to support
these operations. Generic support in GCC for these as builtins went
in before 3.0.0, but not all cpus added their patterns. We define
versions that use the builtins here, and <bits/mathinline.h> will
undef/redefine as appropriate for the specific GCC version in use. */
# define isgreater(x, y) __builtin_isgreater(x, y)
# define isgreaterequal(x, y) __builtin_isgreaterequal(x, y)
# define isless(x, y) __builtin_isless(x, y)
# define islessequal(x, y) __builtin_islessequal(x, y)
# define islessgreater(x, y) __builtin_islessgreater(x, y)
# define isunordered(u, v) __builtin_isunordered(u, v)
#endif
/* Get machine-dependent inline versions (if there are any). */
#ifdef __USE_EXTERN_INLINES
# include <bits/mathinline.h>
#endif
/* Define special entry points to use when the compiler got told to
only expect finite results. */
#if defined __FINITE_MATH_ONLY__ && __FINITE_MATH_ONLY__ > 0
# include <bits/math-finite.h>
#endif
#ifdef __USE_ISOC99
/* If we've still got undefined comparison macros, provide defaults. */
/* Return nonzero value if X is greater than Y. */
# ifndef isgreater
# define isgreater(x, y) \
(__extension__ \
({ __typeof__(x) __x = (x); __typeof__(y) __y = (y); \
!isunordered (__x, __y) && __x > __y; }))
# endif
/* Return nonzero value if X is greater than or equal to Y. */
# ifndef isgreaterequal
# define isgreaterequal(x, y) \
(__extension__ \
({ __typeof__(x) __x = (x); __typeof__(y) __y = (y); \
!isunordered (__x, __y) && __x >= __y; }))
# endif
/* Return nonzero value if X is less than Y. */
# ifndef isless
# define isless(x, y) \
(__extension__ \
({ __typeof__(x) __x = (x); __typeof__(y) __y = (y); \
!isunordered (__x, __y) && __x < __y; }))
# endif
/* Return nonzero value if X is less than or equal to Y. */
# ifndef islessequal
# define islessequal(x, y) \
(__extension__ \
({ __typeof__(x) __x = (x); __typeof__(y) __y = (y); \
!isunordered (__x, __y) && __x <= __y; }))
# endif
/* Return nonzero value if either X is less than Y or Y is less than X. */
# ifndef islessgreater
# define islessgreater(x, y) \
(__extension__ \
({ __typeof__(x) __x = (x); __typeof__(y) __y = (y); \
!isunordered (__x, __y) && (__x < __y || __y < __x); }))
# endif
/* Return nonzero value if arguments are unordered. */
# ifndef isunordered
# define isunordered(u, v) \
(__extension__ \
({ __typeof__(u) __u = (u); __typeof__(v) __v = (v); \
fpclassify (__u) == FP_NAN || fpclassify (__v) == FP_NAN; }))
# endif
#endif
__END_DECLS
#endif /* math.h */