751 lines
23 KiB
C
751 lines
23 KiB
C
/* A GNU-like <math.h>.
|
|
|
|
Copyright (C) 2002-2003, 2007-2010 Free Software Foundation, Inc.
|
|
|
|
This program is free software: you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; either version 3 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 General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with this program. If not, see <http://www.gnu.org/licenses/>. */
|
|
|
|
#ifndef _GL_MATH_H
|
|
|
|
#if __GNUC__ >= 3
|
|
@PRAGMA_SYSTEM_HEADER@
|
|
#endif
|
|
|
|
/* The include_next requires a split double-inclusion guard. */
|
|
#@INCLUDE_NEXT_AS_FIRST_DIRECTIVE@ @NEXT_AS_FIRST_DIRECTIVE_MATH_H@
|
|
|
|
#ifndef _GL_MATH_H
|
|
#define _GL_MATH_H
|
|
|
|
|
|
/* The definitions of _GL_FUNCDECL_RPL etc. are copied here. */
|
|
|
|
/* The definition of _GL_ARG_NONNULL is copied here. */
|
|
|
|
/* The definition of _GL_WARN_ON_USE is copied here. */
|
|
|
|
/* Helper macros to define a portability warning for the
|
|
classification macro FUNC called with VALUE. POSIX declares the
|
|
classification macros with an argument of real-floating (that is,
|
|
one of float, double, or long double). */
|
|
#define _GL_WARN_REAL_FLOATING_DECL(func) \
|
|
static inline int \
|
|
rpl_ ## func ## f (float f) \
|
|
{ \
|
|
return func (f); \
|
|
} \
|
|
static inline int \
|
|
rpl_ ## func ## d (double d) \
|
|
{ \
|
|
return func (d); \
|
|
} \
|
|
static inline int \
|
|
rpl_ ## func ## l (long double l) \
|
|
{ \
|
|
return func (l); \
|
|
} \
|
|
_GL_WARN_ON_USE (rpl_ ## func ## f, #func " is unportable - " \
|
|
"use gnulib module " #func " for portability"); \
|
|
_GL_WARN_ON_USE (rpl_ ## func ## d, #func " is unportable - " \
|
|
"use gnulib module " #func " for portability"); \
|
|
_GL_WARN_ON_USE (rpl_ ## func ## l, #func " is unportable - " \
|
|
"use gnulib module " #func " for portability")
|
|
#define _GL_WARN_REAL_FLOATING_IMPL(func, value) \
|
|
(sizeof (value) == sizeof (float) ? rpl_ ## func ## f (value) \
|
|
: sizeof (value) == sizeof (double) ? rpl_ ## func ## d (value) \
|
|
: rpl_ ## func ## l (value))
|
|
|
|
|
|
/* POSIX allows platforms that don't support NAN. But all major
|
|
machines in the past 15 years have supported something close to
|
|
IEEE NaN, so we define this unconditionally. We also must define
|
|
it on platforms like Solaris 10, where NAN is present but defined
|
|
as a function pointer rather than a floating point constant. */
|
|
#if !defined NAN || @REPLACE_NAN@
|
|
# undef NAN
|
|
/* The Compaq (ex-DEC) C 6.4 compiler chokes on the expression 0.0 / 0.0. */
|
|
# ifdef __DECC
|
|
static float
|
|
_NaN ()
|
|
{
|
|
static float zero = 0.0f;
|
|
return zero / zero;
|
|
}
|
|
# define NAN (_NaN())
|
|
# else
|
|
# define NAN (0.0f / 0.0f)
|
|
# endif
|
|
#endif
|
|
|
|
/* Solaris 10 defines HUGE_VAL, but as a function pointer rather
|
|
than a floating point constant. */
|
|
#if @REPLACE_HUGE_VAL@
|
|
# undef HUGE_VAL
|
|
# define HUGE_VAL (1.0 / 0.0)
|
|
#endif
|
|
|
|
|
|
/* Write x as
|
|
x = mantissa * 2^exp
|
|
where
|
|
If x finite and nonzero: 0.5 <= |mantissa| < 1.0.
|
|
If x is zero: mantissa = x, exp = 0.
|
|
If x is infinite or NaN: mantissa = x, exp unspecified.
|
|
Store exp in *EXPPTR and return mantissa. */
|
|
#if @GNULIB_FREXP@
|
|
# if @REPLACE_FREXP@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define frexp rpl_frexp
|
|
# endif
|
|
_GL_FUNCDECL_RPL (frexp, double, (double x, int *expptr) _GL_ARG_NONNULL ((2)));
|
|
_GL_CXXALIAS_RPL (frexp, double, (double x, int *expptr));
|
|
# else
|
|
_GL_CXXALIAS_SYS (frexp, double, (double x, int *expptr));
|
|
# endif
|
|
_GL_CXXALIASWARN (frexp);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef frexp
|
|
/* Assume frexp is always declared. */
|
|
_GL_WARN_ON_USE (frexp, "frexp is unportable - "
|
|
"use gnulib module frexp for portability");
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_LOGB@
|
|
# if !@HAVE_DECL_LOGB@
|
|
_GL_EXTERN_C double logb (double x);
|
|
# endif
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef logb
|
|
# if HAVE_RAW_DECL_LOGB
|
|
_GL_WARN_ON_USE (logb, "logb is unportable - "
|
|
"use gnulib module logb for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ACOSL@
|
|
# if !@HAVE_ACOSL@ || !@HAVE_DECL_ACOSL@
|
|
_GL_FUNCDECL_SYS (acosl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (acosl, long double, (long double x));
|
|
_GL_CXXALIASWARN (acosl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef acosl
|
|
# if HAVE_RAW_DECL_ACOSL
|
|
_GL_WARN_ON_USE (acosl, "acosl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ASINL@
|
|
# if !@HAVE_ASINL@ || !@HAVE_DECL_ASINL@
|
|
_GL_FUNCDECL_SYS (asinl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (asinl, long double, (long double x));
|
|
_GL_CXXALIASWARN (asinl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef asinl
|
|
# if HAVE_RAW_DECL_ASINL
|
|
_GL_WARN_ON_USE (asinl, "asinl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ATANL@
|
|
# if !@HAVE_ATANL@ || !@HAVE_DECL_ATANL@
|
|
_GL_FUNCDECL_SYS (atanl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (atanl, long double, (long double x));
|
|
_GL_CXXALIASWARN (atanl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef atanl
|
|
# if HAVE_RAW_DECL_ATANL
|
|
_GL_WARN_ON_USE (atanl, "atanl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_CEILF@
|
|
# if @REPLACE_CEILF@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define ceilf rpl_ceilf
|
|
# endif
|
|
_GL_FUNCDECL_RPL (ceilf, float, (float x));
|
|
_GL_CXXALIAS_RPL (ceilf, float, (float x));
|
|
# else
|
|
# if !@HAVE_DECL_CEILF@
|
|
_GL_FUNCDECL_SYS (ceilf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (ceilf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIASWARN (ceilf);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef ceilf
|
|
# if HAVE_RAW_DECL_CEILF
|
|
_GL_WARN_ON_USE (ceilf, "ceilf is unportable - "
|
|
"use gnulib module ceilf for portability");
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_CEILL@
|
|
# if @REPLACE_CEILL@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define ceill rpl_ceill
|
|
# endif
|
|
_GL_FUNCDECL_RPL (ceill, long double, (long double x));
|
|
_GL_CXXALIAS_RPL (ceill, long double, (long double x));
|
|
# else
|
|
# if !@HAVE_DECL_CEILL@
|
|
_GL_FUNCDECL_SYS (ceill, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (ceill, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIASWARN (ceill);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef ceill
|
|
# if HAVE_RAW_DECL_CEILL
|
|
_GL_WARN_ON_USE (ceill, "ceill is unportable - "
|
|
"use gnulib module ceill for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_COSL@
|
|
# if !@HAVE_COSL@ || !@HAVE_DECL_COSL@
|
|
_GL_FUNCDECL_SYS (cosl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (cosl, long double, (long double x));
|
|
_GL_CXXALIASWARN (cosl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef cosl
|
|
# if HAVE_RAW_DECL_COSL
|
|
_GL_WARN_ON_USE (cosl, "cosl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_EXPL@
|
|
# if !@HAVE_EXPL@ || !@HAVE_DECL_EXPL@
|
|
_GL_FUNCDECL_SYS (expl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (expl, long double, (long double x));
|
|
_GL_CXXALIASWARN (expl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef expl
|
|
# if HAVE_RAW_DECL_EXPL
|
|
_GL_WARN_ON_USE (expl, "expl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_FLOORF@
|
|
# if @REPLACE_FLOORF@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define floorf rpl_floorf
|
|
# endif
|
|
_GL_FUNCDECL_RPL (floorf, float, (float x));
|
|
_GL_CXXALIAS_RPL (floorf, float, (float x));
|
|
#else
|
|
# if !@HAVE_DECL_FLOORF@
|
|
_GL_FUNCDECL_SYS (floorf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (floorf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIASWARN (floorf);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef floorf
|
|
# if HAVE_RAW_DECL_FLOORF
|
|
_GL_WARN_ON_USE (floorf, "floorf is unportable - "
|
|
"use gnulib module floorf for portability");
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_FLOORL@
|
|
# if @REPLACE_FLOORL@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define floorl rpl_floorl
|
|
# endif
|
|
_GL_FUNCDECL_RPL (floorl, long double, (long double x));
|
|
_GL_CXXALIAS_RPL (floorl, long double, (long double x));
|
|
# else
|
|
# if !@HAVE_DECL_FLOORL@
|
|
_GL_FUNCDECL_SYS (floorl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (floorl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIASWARN (floorl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef floorl
|
|
# if HAVE_RAW_DECL_FLOORL
|
|
_GL_WARN_ON_USE (floorl, "floorl is unportable - "
|
|
"use gnulib module floorl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
/* Write x as
|
|
x = mantissa * 2^exp
|
|
where
|
|
If x finite and nonzero: 0.5 <= |mantissa| < 1.0.
|
|
If x is zero: mantissa = x, exp = 0.
|
|
If x is infinite or NaN: mantissa = x, exp unspecified.
|
|
Store exp in *EXPPTR and return mantissa. */
|
|
#if @GNULIB_FREXPL@ && @REPLACE_FREXPL@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define frexpl rpl_frexpl
|
|
# endif
|
|
_GL_FUNCDECL_RPL (frexpl, long double,
|
|
(long double x, int *expptr) _GL_ARG_NONNULL ((2)));
|
|
_GL_CXXALIAS_RPL (frexpl, long double, (long double x, int *expptr));
|
|
#else
|
|
# if !@HAVE_DECL_FREXPL@
|
|
_GL_FUNCDECL_SYS (frexpl, long double,
|
|
(long double x, int *expptr) _GL_ARG_NONNULL ((2)));
|
|
# endif
|
|
# if @GNULIB_FREXPL@
|
|
_GL_CXXALIAS_SYS (frexpl, long double, (long double x, int *expptr));
|
|
# endif
|
|
#endif
|
|
#if @GNULIB_FREXPL@ && !(@REPLACE_FREXPL@ && !@HAVE_DECL_FREXPL@)
|
|
_GL_CXXALIASWARN (frexpl);
|
|
#endif
|
|
#if !@GNULIB_FREXPL@ && defined GNULIB_POSIXCHECK
|
|
# undef frexpl
|
|
# if HAVE_RAW_DECL_FREXPL
|
|
_GL_WARN_ON_USE (frexpl, "frexpl is unportable - "
|
|
"use gnulib module frexpl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
/* Return x * 2^exp. */
|
|
#if @GNULIB_LDEXPL@ && @REPLACE_LDEXPL@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# define ldexpl rpl_ldexpl
|
|
# endif
|
|
_GL_FUNCDECL_RPL (ldexpl, long double, (long double x, int exp));
|
|
_GL_CXXALIAS_RPL (ldexpl, long double, (long double x, int exp));
|
|
#else
|
|
# if !@HAVE_DECL_LDEXPL@
|
|
_GL_FUNCDECL_SYS (ldexpl, long double, (long double x, int exp));
|
|
# endif
|
|
# if @GNULIB_LDEXPL@
|
|
_GL_CXXALIAS_SYS (ldexpl, long double, (long double x, int exp));
|
|
# endif
|
|
#endif
|
|
#if @GNULIB_LDEXPL@
|
|
_GL_CXXALIASWARN (ldexpl);
|
|
#endif
|
|
#if !@GNULIB_LDEXPL@ && defined GNULIB_POSIXCHECK
|
|
# undef ldexpl
|
|
# if HAVE_RAW_DECL_LDEXPL
|
|
_GL_WARN_ON_USE (ldexpl, "ldexpl is unportable - "
|
|
"use gnulib module ldexpl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_LOGL@
|
|
# if !@HAVE_LOGL@ || !@HAVE_DECL_LOGL@
|
|
_GL_FUNCDECL_SYS (logl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (logl, long double, (long double x));
|
|
_GL_CXXALIASWARN (logl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef logl
|
|
# if HAVE_RAW_DECL_LOGL
|
|
_GL_WARN_ON_USE (logl, "logl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ROUNDF@
|
|
# if @REPLACE_ROUNDF@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# undef roundf
|
|
# define roundf rpl_roundf
|
|
# endif
|
|
_GL_FUNCDECL_RPL (roundf, float, (float x));
|
|
_GL_CXXALIAS_RPL (roundf, float, (float x));
|
|
# else
|
|
# if !@HAVE_DECL_ROUNDF@
|
|
_GL_FUNCDECL_SYS (roundf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (roundf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIASWARN (roundf);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef roundf
|
|
# if HAVE_RAW_DECL_ROUNDF
|
|
_GL_WARN_ON_USE (roundf, "roundf is unportable - "
|
|
"use gnulib module roundf for portability");
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_ROUND@
|
|
# if @REPLACE_ROUND@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# undef round
|
|
# define round rpl_round
|
|
# endif
|
|
_GL_FUNCDECL_RPL (round, double, (double x));
|
|
_GL_CXXALIAS_RPL (round, double, (double x));
|
|
# else
|
|
# if !@HAVE_DECL_ROUND@
|
|
_GL_FUNCDECL_SYS (round, double, (double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (round, double, (double x));
|
|
# endif
|
|
_GL_CXXALIASWARN (round);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef round
|
|
# if HAVE_RAW_DECL_ROUND
|
|
_GL_WARN_ON_USE (round, "round is unportable - "
|
|
"use gnulib module round for portability");
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_ROUNDL@
|
|
# if @REPLACE_ROUNDL@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# undef roundl
|
|
# define roundl rpl_roundl
|
|
# endif
|
|
_GL_FUNCDECL_RPL (roundl, long double, (long double x));
|
|
_GL_CXXALIAS_RPL (roundl, long double, (long double x));
|
|
# else
|
|
# if !@HAVE_DECL_ROUNDL@
|
|
_GL_FUNCDECL_SYS (roundl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (roundl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIASWARN (roundl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef roundl
|
|
# if HAVE_RAW_DECL_ROUNDL
|
|
_GL_WARN_ON_USE (roundl, "roundl is unportable - "
|
|
"use gnulib module roundl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_SINL@
|
|
# if !@HAVE_SINL@ || !@HAVE_DECL_SINL@
|
|
_GL_FUNCDECL_SYS (sinl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (sinl, long double, (long double x));
|
|
_GL_CXXALIASWARN (sinl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef sinl
|
|
# if HAVE_RAW_DECL_SINL
|
|
_GL_WARN_ON_USE (sinl, "sinl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_SQRTL@
|
|
# if !@HAVE_SQRTL@ || !@HAVE_DECL_SQRTL@
|
|
_GL_FUNCDECL_SYS (sqrtl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (sqrtl, long double, (long double x));
|
|
_GL_CXXALIASWARN (sqrtl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef sqrtl
|
|
# if HAVE_RAW_DECL_SQRTL
|
|
_GL_WARN_ON_USE (sqrtl, "sqrtl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_TANL@
|
|
# if !@HAVE_TANL@ || !@HAVE_DECL_TANL@
|
|
_GL_FUNCDECL_SYS (tanl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (tanl, long double, (long double x));
|
|
_GL_CXXALIASWARN (tanl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef tanl
|
|
# if HAVE_RAW_DECL_TANL
|
|
_GL_WARN_ON_USE (tanl, "tanl is unportable - "
|
|
"use gnulib module mathl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_TRUNCF@
|
|
# if !@HAVE_DECL_TRUNCF@
|
|
_GL_FUNCDECL_SYS (truncf, float, (float x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (truncf, float, (float x));
|
|
_GL_CXXALIASWARN (truncf);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef truncf
|
|
# if HAVE_RAW_DECL_TRUNCF
|
|
_GL_WARN_ON_USE (truncf, "truncf is unportable - "
|
|
"use gnulib module truncf for portability");
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_TRUNC@
|
|
# if !@HAVE_DECL_TRUNC@
|
|
_GL_FUNCDECL_SYS (trunc, double, (double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (trunc, double, (double x));
|
|
_GL_CXXALIASWARN (trunc);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef trunc
|
|
# if HAVE_RAW_DECL_TRUNC
|
|
_GL_WARN_ON_USE (trunc, "trunc is unportable - "
|
|
"use gnulib module trunc for portability");
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_TRUNCL@
|
|
# if @REPLACE_TRUNCL@
|
|
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
|
|
# undef truncl
|
|
# define truncl rpl_truncl
|
|
# endif
|
|
_GL_FUNCDECL_RPL (truncl, long double, (long double x));
|
|
_GL_CXXALIAS_RPL (truncl, long double, (long double x));
|
|
# else
|
|
# if !@HAVE_DECL_TRUNCL@
|
|
_GL_FUNCDECL_SYS (truncl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIAS_SYS (truncl, long double, (long double x));
|
|
# endif
|
|
_GL_CXXALIASWARN (truncl);
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# undef truncl
|
|
# if HAVE_RAW_DECL_TRUNCL
|
|
_GL_WARN_ON_USE (truncl, "truncl is unportable - "
|
|
"use gnulib module truncl for portability");
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ISFINITE@
|
|
# if @REPLACE_ISFINITE@
|
|
_GL_EXTERN_C int gl_isfinitef (float x);
|
|
_GL_EXTERN_C int gl_isfinited (double x);
|
|
_GL_EXTERN_C int gl_isfinitel (long double x);
|
|
# undef isfinite
|
|
# define isfinite(x) \
|
|
(sizeof (x) == sizeof (long double) ? gl_isfinitel (x) : \
|
|
sizeof (x) == sizeof (double) ? gl_isfinited (x) : \
|
|
gl_isfinitef (x))
|
|
# endif
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# if defined isfinite
|
|
_GL_WARN_REAL_FLOATING_DECL (isfinite);
|
|
# undef isfinite
|
|
# define isfinite(x) _GL_WARN_REAL_FLOATING_IMPL (isfinite, x)
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ISINF@
|
|
# if @REPLACE_ISINF@
|
|
_GL_EXTERN_C int gl_isinff (float x);
|
|
_GL_EXTERN_C int gl_isinfd (double x);
|
|
_GL_EXTERN_C int gl_isinfl (long double x);
|
|
# undef isinf
|
|
# define isinf(x) \
|
|
(sizeof (x) == sizeof (long double) ? gl_isinfl (x) : \
|
|
sizeof (x) == sizeof (double) ? gl_isinfd (x) : \
|
|
gl_isinff (x))
|
|
# endif
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# if defined isinf
|
|
_GL_WARN_REAL_FLOATING_DECL (isinf);
|
|
# undef isinf
|
|
# define isinf(x) _GL_WARN_REAL_FLOATING_IMPL (isinf, x)
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_ISNANF@
|
|
/* Test for NaN for 'float' numbers. */
|
|
# if @HAVE_ISNANF@
|
|
/* The original <math.h> included above provides a declaration of isnan macro
|
|
or (older) isnanf function. */
|
|
# if __GNUC__ >= 4
|
|
/* GCC 4.0 and newer provides three built-ins for isnan. */
|
|
# undef isnanf
|
|
# define isnanf(x) __builtin_isnanf ((float)(x))
|
|
# elif defined isnan
|
|
# undef isnanf
|
|
# define isnanf(x) isnan ((float)(x))
|
|
# endif
|
|
# else
|
|
/* Test whether X is a NaN. */
|
|
# undef isnanf
|
|
# define isnanf rpl_isnanf
|
|
_GL_EXTERN_C int isnanf (float x);
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_ISNAND@
|
|
/* Test for NaN for 'double' numbers.
|
|
This function is a gnulib extension, unlike isnan() which applied only
|
|
to 'double' numbers earlier but now is a type-generic macro. */
|
|
# if @HAVE_ISNAND@
|
|
/* The original <math.h> included above provides a declaration of isnan macro. */
|
|
# if __GNUC__ >= 4
|
|
/* GCC 4.0 and newer provides three built-ins for isnan. */
|
|
# undef isnand
|
|
# define isnand(x) __builtin_isnan ((double)(x))
|
|
# else
|
|
# undef isnand
|
|
# define isnand(x) isnan ((double)(x))
|
|
# endif
|
|
# else
|
|
/* Test whether X is a NaN. */
|
|
# undef isnand
|
|
# define isnand rpl_isnand
|
|
_GL_EXTERN_C int isnand (double x);
|
|
# endif
|
|
#endif
|
|
|
|
#if @GNULIB_ISNANL@
|
|
/* Test for NaN for 'long double' numbers. */
|
|
# if @HAVE_ISNANL@
|
|
/* The original <math.h> included above provides a declaration of isnan macro or (older) isnanl function. */
|
|
# if __GNUC__ >= 4
|
|
/* GCC 4.0 and newer provides three built-ins for isnan. */
|
|
# undef isnanl
|
|
# define isnanl(x) __builtin_isnanl ((long double)(x))
|
|
# elif defined isnan
|
|
# undef isnanl
|
|
# define isnanl(x) isnan ((long double)(x))
|
|
# endif
|
|
# else
|
|
/* Test whether X is a NaN. */
|
|
# undef isnanl
|
|
# define isnanl rpl_isnanl
|
|
_GL_EXTERN_C int isnanl (long double x);
|
|
# endif
|
|
#endif
|
|
|
|
/* This must come *after* the snippets for GNULIB_ISNANF and GNULIB_ISNANL! */
|
|
#if @GNULIB_ISNAN@
|
|
# if @REPLACE_ISNAN@
|
|
/* We can't just use the isnanf macro (e.g.) as exposed by
|
|
isnanf.h (e.g.) here, because those may end up being macros
|
|
that recursively expand back to isnan. So use the gnulib
|
|
replacements for them directly. */
|
|
# if @HAVE_ISNANF@ && __GNUC__ >= 4
|
|
# define gl_isnan_f(x) __builtin_isnan ((float)(x))
|
|
# else
|
|
_GL_EXTERN_C int rpl_isnanf (float x);
|
|
# define gl_isnan_f(x) rpl_isnanf (x)
|
|
# endif
|
|
# if @HAVE_ISNAND@ && __GNUC__ >= 4
|
|
# define gl_isnan_d(x) __builtin_isnan ((double)(x))
|
|
# else
|
|
_GL_EXTERN_C int rpl_isnand (double x);
|
|
# define gl_isnan_d(x) rpl_isnand (x)
|
|
# endif
|
|
# if @HAVE_ISNANL@ && __GNUC__ >= 4
|
|
# define gl_isnan_l(x) __builtin_isnan ((long double)(x))
|
|
# else
|
|
_GL_EXTERN_C int rpl_isnanl (long double x);
|
|
# define gl_isnan_l(x) rpl_isnanl (x)
|
|
# endif
|
|
# undef isnan
|
|
# define isnan(x) \
|
|
(sizeof (x) == sizeof (long double) ? gl_isnan_l (x) : \
|
|
sizeof (x) == sizeof (double) ? gl_isnan_d (x) : \
|
|
gl_isnan_f (x))
|
|
# endif
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# if defined isnan
|
|
_GL_WARN_REAL_FLOATING_DECL (isnan);
|
|
# undef isnan
|
|
# define isnan(x) _GL_WARN_REAL_FLOATING_IMPL (isnan, x)
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#if @GNULIB_SIGNBIT@
|
|
# if @REPLACE_SIGNBIT_USING_GCC@
|
|
# undef signbit
|
|
/* GCC 4.0 and newer provides three built-ins for signbit. */
|
|
# define signbit(x) \
|
|
(sizeof (x) == sizeof (long double) ? __builtin_signbitl (x) : \
|
|
sizeof (x) == sizeof (double) ? __builtin_signbit (x) : \
|
|
__builtin_signbitf (x))
|
|
# endif
|
|
# if @REPLACE_SIGNBIT@
|
|
# undef signbit
|
|
_GL_EXTERN_C int gl_signbitf (float arg);
|
|
_GL_EXTERN_C int gl_signbitd (double arg);
|
|
_GL_EXTERN_C int gl_signbitl (long double arg);
|
|
# if __GNUC__ >= 2 && !__STRICT_ANSI__
|
|
# if defined FLT_SIGNBIT_WORD && defined FLT_SIGNBIT_BIT && !defined gl_signbitf
|
|
# define gl_signbitf_OPTIMIZED_MACRO
|
|
# define gl_signbitf(arg) \
|
|
({ union { float _value; \
|
|
unsigned int _word[(sizeof (float) + sizeof (unsigned int) - 1) / sizeof (unsigned int)]; \
|
|
} _m; \
|
|
_m._value = (arg); \
|
|
(_m._word[FLT_SIGNBIT_WORD] >> FLT_SIGNBIT_BIT) & 1; \
|
|
})
|
|
# endif
|
|
# if defined DBL_SIGNBIT_WORD && defined DBL_SIGNBIT_BIT && !defined gl_signbitd
|
|
# define gl_signbitd_OPTIMIZED_MACRO
|
|
# define gl_signbitd(arg) \
|
|
({ union { double _value; \
|
|
unsigned int _word[(sizeof (double) + sizeof (unsigned int) - 1) / sizeof (unsigned int)]; \
|
|
} _m; \
|
|
_m._value = (arg); \
|
|
(_m._word[DBL_SIGNBIT_WORD] >> DBL_SIGNBIT_BIT) & 1; \
|
|
})
|
|
# endif
|
|
# if defined LDBL_SIGNBIT_WORD && defined LDBL_SIGNBIT_BIT && !defined gl_signbitl
|
|
# define gl_signbitl_OPTIMIZED_MACRO
|
|
# define gl_signbitl(arg) \
|
|
({ union { long double _value; \
|
|
unsigned int _word[(sizeof (long double) + sizeof (unsigned int) - 1) / sizeof (unsigned int)]; \
|
|
} _m; \
|
|
_m._value = (arg); \
|
|
(_m._word[LDBL_SIGNBIT_WORD] >> LDBL_SIGNBIT_BIT) & 1; \
|
|
})
|
|
# endif
|
|
# endif
|
|
# define signbit(x) \
|
|
(sizeof (x) == sizeof (long double) ? gl_signbitl (x) : \
|
|
sizeof (x) == sizeof (double) ? gl_signbitd (x) : \
|
|
gl_signbitf (x))
|
|
# endif
|
|
#elif defined GNULIB_POSIXCHECK
|
|
# if defined signbit
|
|
_GL_WARN_REAL_FLOATING_DECL (signbit);
|
|
# undef signbit
|
|
# define signbit(x) _GL_WARN_REAL_FLOATING_IMPL (signbit, x)
|
|
# endif
|
|
#endif
|
|
|
|
|
|
#endif /* _GL_MATH_H */
|
|
#endif /* _GL_MATH_H */
|