mirror of
https://sourceware.org/git/glibc.git
synced 2024-12-25 12:11:10 +00:00
adac845094
1998-09-09 Benjamin Kosnik <bkoz@tintin.cygnus.com> * math/libm-test.c (exp10_test): Adapt epsilons for powerpc. (csqrt_test): Likewise. * sysdeps/powerpc/bits/fenv.h: Add parens. .
6085 lines
244 KiB
C
6085 lines
244 KiB
C
/* Copyright (C) 1997, 1998 Free Software Foundation, Inc.
|
|
This file is part of the GNU C Library.
|
|
Contributed by Andreas Jaeger <aj@arthur.rhein-neckar.de>, 1997.
|
|
|
|
The GNU C Library is free software; you can redistribute it and/or
|
|
modify it under the terms of the GNU Library General Public License as
|
|
published by the Free Software Foundation; either version 2 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
|
|
Library General Public License for more details.
|
|
|
|
You should have received a copy of the GNU Library General Public
|
|
License along with the GNU C Library; see the file COPYING.LIB. If not,
|
|
write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
Boston, MA 02111-1307, USA. */
|
|
|
|
|
|
/*
|
|
Part of testsuite for libm.
|
|
|
|
This file has to be included by a master file that defines:
|
|
|
|
Makros:
|
|
FUNC(function): converts general function name (like cos) to
|
|
name with correct suffix (e.g. cosl or cosf)
|
|
MATHCONST(x): like FUNC but for constants (e.g convert 0.0 to 0.0L)
|
|
MATHTYPE: floating point type to test
|
|
TEST_MSG: informal message to be displayed
|
|
CHOOSE(Clongdouble,Cdouble,Cfloat):
|
|
chooses one of the parameters as epsilon for testing
|
|
equality
|
|
PRINTF_EXPR Floating point conversion specification to print a variable
|
|
of type MATHTYPE with printf. PRINTF_EXPR just contains
|
|
the specifier, not the percent and width arguments,
|
|
e.g. "f".
|
|
PRINTF_XEXPR Like PRINTF_EXPR, but print in hexadecimal format.
|
|
*/
|
|
|
|
/* This program isn't finished yet.
|
|
It has tests for:
|
|
acos, acosh, asin, asinh, atan, atan2, atanh,
|
|
cbrt, ceil, copysign, cos, cosh, erf, erfc, exp, exp10, exp2, expm1,
|
|
fabs, fdim, floor, fma, fmax, fmin, fmod, fpclassify,
|
|
frexp, gamma, hypot,
|
|
ilogb, isfinite, isinf, isnan, isnormal,
|
|
isless, islessequal, isgreater, isgreaterequal, islessgreater, isunordered,
|
|
ldexp, lgamma, log, log10, log1p, log2, logb,
|
|
modf, nearbyint, nextafter,
|
|
pow, remainder, remquo, rint, lrint, llrint,
|
|
round, lround, llround,
|
|
scalb, scalbn, signbit, sin, sincos, sinh, sqrt, tan, tanh, tgamma, trunc
|
|
|
|
and for the following complex math functions:
|
|
cabs, cacos, cacosh, carg, casin, casinh, catan, catanh,
|
|
ccos, ccosh, cexp, clog, cpow, csin, csinh, csqrt, ctan, ctanh.
|
|
|
|
At the moment the following functions aren't tested:
|
|
OB conj, cproj, cimag, creal, drem,
|
|
j0, j1, jn, y0, y1, yn,
|
|
significand,
|
|
nan
|
|
|
|
The routines using random variables are still under construction. I don't
|
|
like it the way it's working now and will change it.
|
|
|
|
Parameter handling is primitive in the moment:
|
|
--verbose=[0..4] for different levels of output:
|
|
0: only error count
|
|
1: basic report on failed tests (default)
|
|
2: full report on failed tests
|
|
3: full report on failed and passed tests
|
|
4: additional report on exceptions
|
|
-v for full output (equals --verbose=4)
|
|
-s,--silent outputs only the error count (equals --verbose=0)
|
|
*/
|
|
|
|
/* "Philosophy":
|
|
|
|
This suite tests some aspects of the correct implementation of
|
|
mathematical functions in libm. Some simple, specific parameters
|
|
are tested for correctness but there's no exhaustive
|
|
testing. Handling of specific inputs (e.g. infinity, not-a-number)
|
|
is also tested. Correct handling of exceptions is checked
|
|
against. These implemented tests should check all cases that are
|
|
specified in ISO C 9X.
|
|
|
|
Exception testing: At the moment only divide-by-zero and invalid
|
|
exceptions are tested. Overflow/underflow and inexact exceptions
|
|
aren't checked at the moment.
|
|
|
|
NaN values: There exist signalling and quiet NaNs. This implementation
|
|
only uses signalling NaN as parameter but does not differenciate
|
|
between the two kinds of NaNs as result.
|
|
|
|
Inline functions: Inlining functions should give an improvement in
|
|
speed - but not in precission. The inlined functions return
|
|
reasonable values for a reasonable range of input values. The
|
|
result is not necessarily correct for all values and exceptions are
|
|
not correctly raised in all cases. Problematic input and return
|
|
values are infinity, not-a-number and minus zero. This suite
|
|
therefore does not check these specific inputs and the exception
|
|
handling for inlined mathematical functions - just the "reasonable"
|
|
values are checked.
|
|
|
|
Beware: The tests might fail for any of the following reasons:
|
|
- Tests are wrong
|
|
- Functions are wrong
|
|
- Floating Point Unit not working properly
|
|
- Compiler has errors
|
|
|
|
With e.g. gcc 2.7.2.2 the test for cexp fails because of a compiler error.
|
|
|
|
*/
|
|
|
|
#ifndef _GNU_SOURCE
|
|
# define _GNU_SOURCE
|
|
#endif
|
|
|
|
#include <complex.h>
|
|
#include <math.h>
|
|
#include <float.h>
|
|
#include <fenv.h>
|
|
|
|
#include <errno.h>
|
|
#include <stdlib.h>
|
|
#include <stdio.h>
|
|
#include <getopt.h>
|
|
|
|
/* Possible exceptions */
|
|
#define NO_EXCEPTION 0x0
|
|
#define INVALID_EXCEPTION 0x1
|
|
#define DIVIDE_BY_ZERO_EXCEPTION 0x2
|
|
|
|
#define PRINT 1
|
|
#define NO_PRINT 0
|
|
|
|
/* Various constants (we must supply them precalculated for accuracy). */
|
|
#define M_PI_6l .52359877559829887308L
|
|
#define M_E2l 7.389056098930650227230L
|
|
#define M_E3l 20.08553692318766774093L
|
|
|
|
static int noErrors; /* number of errors */
|
|
static int noTests; /* number of tests (without testing exceptions) */
|
|
static int noExcTests; /* number of tests for exception flags */
|
|
|
|
static int verbose = 3;
|
|
static MATHTYPE minus_zero, plus_zero;
|
|
static MATHTYPE plus_infty, minus_infty, nan_value;
|
|
|
|
typedef MATHTYPE (*mathfunc) (MATHTYPE);
|
|
|
|
#define BUILD_COMPLEX(real, imag) \
|
|
({ __complex__ MATHTYPE __retval; \
|
|
__real__ __retval = (real); \
|
|
__imag__ __retval = (imag); \
|
|
__retval; })
|
|
|
|
|
|
#define ISINF(x) \
|
|
(sizeof (x) == sizeof (float) ? \
|
|
isinff (x) \
|
|
: sizeof (x) == sizeof (double) ? \
|
|
isinf (x) : isinfl (x))
|
|
|
|
|
|
/*
|
|
Test if Floating-Point stack hasn't changed
|
|
*/
|
|
static void
|
|
fpstack_test (const char *test_name)
|
|
{
|
|
#ifdef i386
|
|
static int old_stack;
|
|
int sw;
|
|
asm ("fnstsw":"=a" (sw));
|
|
sw >>= 11;
|
|
sw &= 7;
|
|
if (sw != old_stack)
|
|
{
|
|
printf ("FP-Stack wrong after test %s\n", test_name);
|
|
if (verbose > 2)
|
|
printf ("=======> stack = %d\n", sw);
|
|
++noErrors;
|
|
old_stack = sw;
|
|
}
|
|
#endif
|
|
}
|
|
|
|
|
|
/*
|
|
Get a random value x with min_value < x < max_value
|
|
and min_value, max_value finite,
|
|
max_value and min_value shouldn't be too close together
|
|
*/
|
|
static MATHTYPE
|
|
random_value (MATHTYPE min_value, MATHTYPE max_value)
|
|
{
|
|
int r;
|
|
MATHTYPE x;
|
|
|
|
r = rand ();
|
|
|
|
x = (max_value - min_value) / RAND_MAX * (MATHTYPE) r + min_value;
|
|
|
|
if ((x <= min_value) || (x >= max_value) || !isfinite (x))
|
|
x = (max_value - min_value) / 2 + min_value;
|
|
|
|
/* Make sure the RNG has no influence on the exceptions. */
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
return x;
|
|
}
|
|
|
|
/* Get a random value x with x > min_value. */
|
|
static MATHTYPE
|
|
random_greater (MATHTYPE min_value)
|
|
{
|
|
return random_value (min_value, 1e6); /* CHOOSE (LDBL_MAX, DBL_MAX, FLT_MAX) */
|
|
}
|
|
|
|
/* Get a random value x with x < max_value. */
|
|
static MATHTYPE
|
|
random_less (MATHTYPE max_value)
|
|
{
|
|
return random_value (-1e6, max_value);
|
|
}
|
|
|
|
|
|
static void
|
|
output_new_test (const char *test_name)
|
|
{
|
|
if (verbose > 2)
|
|
printf ("\nTesting: %s\n", test_name);
|
|
}
|
|
|
|
|
|
static void
|
|
output_pass_value (void)
|
|
{
|
|
if (verbose > 2)
|
|
printf ("Pass: Value Ok.\n");
|
|
}
|
|
|
|
|
|
static void
|
|
output_fail_value (const char * test_name)
|
|
{
|
|
if (verbose > 0 && verbose < 3)
|
|
printf ("Fail: %s\n", test_name);
|
|
if (verbose >= 3)
|
|
printf ("Fail:\n");
|
|
}
|
|
|
|
|
|
/* Test whether a given exception was raised. */
|
|
static void
|
|
test_single_exception (const char *test_name,
|
|
short int exception,
|
|
short int exc_flag,
|
|
int fe_flag,
|
|
const char *flag_name)
|
|
{
|
|
#ifndef TEST_INLINE
|
|
if (exception & exc_flag)
|
|
{
|
|
if (fetestexcept (fe_flag))
|
|
{
|
|
if (verbose > 3)
|
|
printf ("Pass: Exception \"%s\" set\n", flag_name);
|
|
}
|
|
else
|
|
{
|
|
if (verbose && verbose < 3)
|
|
printf ("Fail: %s: Exception \"%s\" not set\n",
|
|
test_name, flag_name);
|
|
if (verbose >= 3)
|
|
printf ("Fail: Exception \"%s\" not set\n",
|
|
flag_name);
|
|
++noErrors;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
if (fetestexcept (fe_flag))
|
|
{
|
|
if (verbose && verbose < 3)
|
|
printf ("Fail: %s: Exception \"%s\" set\n",
|
|
test_name, flag_name);
|
|
if (verbose >= 3)
|
|
printf ("Fail: Exception \"%s\" set\n",
|
|
flag_name);
|
|
++noErrors;
|
|
}
|
|
else
|
|
{
|
|
if (verbose > 3)
|
|
printf ("Pass: Exception \"%s\" not set\n",
|
|
flag_name);
|
|
}
|
|
}
|
|
#endif
|
|
}
|
|
|
|
|
|
/* Test whether exception given by EXCEPTION are raised. */
|
|
static void
|
|
test_not_exception (const char *test_name, short int exception)
|
|
{
|
|
++noExcTests;
|
|
#ifdef FE_DIVBYZERO
|
|
if ((exception & DIVIDE_BY_ZERO_EXCEPTION) == 0)
|
|
test_single_exception (test_name, exception,
|
|
DIVIDE_BY_ZERO_EXCEPTION, FE_DIVBYZERO,
|
|
"Divide by zero");
|
|
#endif
|
|
#ifdef FE_INVALID
|
|
if ((exception & INVALID_EXCEPTION) == 0)
|
|
test_single_exception (test_name, exception, INVALID_EXCEPTION, FE_INVALID,
|
|
"Invalid operation");
|
|
#endif
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
}
|
|
|
|
|
|
/* Test whether exceptions given by EXCEPTION are raised. */
|
|
static void
|
|
test_exceptions (const char *test_name, short int exception)
|
|
{
|
|
++noExcTests;
|
|
#ifdef FE_DIVBYZERO
|
|
test_single_exception (test_name, exception,
|
|
DIVIDE_BY_ZERO_EXCEPTION, FE_DIVBYZERO,
|
|
"Divide by zero");
|
|
#endif
|
|
#ifdef FE_INVALID
|
|
test_single_exception (test_name, exception, INVALID_EXCEPTION, FE_INVALID,
|
|
"Invalid operation");
|
|
#endif
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
}
|
|
|
|
|
|
/* Test if two floating point numbers are equal. */
|
|
static int
|
|
check_equal (MATHTYPE computed, MATHTYPE supplied, MATHTYPE eps, MATHTYPE * diff)
|
|
{
|
|
int ret_value;
|
|
|
|
/* Both plus Infinity or both minus infinity. */
|
|
if (ISINF (computed) && (ISINF (computed) == ISINF (supplied)))
|
|
return 1;
|
|
|
|
if (isnan (computed) && isnan (supplied)) /* isnan works for all types */
|
|
return 1;
|
|
|
|
*diff = FUNC(fabs) (computed - supplied);
|
|
|
|
|
|
ret_value = (*diff <= eps &&
|
|
(signbit (computed) == signbit (supplied) || eps != 0.0));
|
|
|
|
/* Make sure the subtraction/comparison have no influence on the exceptions. */
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
return ret_value;
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
output_result_bool (const char *test_name, int result)
|
|
{
|
|
++noTests;
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1)
|
|
printf (" Value: %d\n", result);
|
|
++noErrors;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
|
|
static void
|
|
output_isvalue (const char *test_name, int result,
|
|
MATHTYPE value)
|
|
{
|
|
++noTests;
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1)
|
|
printf (" Value: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
value, value);
|
|
++noErrors;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
|
|
static void
|
|
output_isvalue_ext (const char *test_name, int result,
|
|
MATHTYPE value, MATHTYPE parameter)
|
|
{
|
|
++noTests;
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1)
|
|
{
|
|
printf (" Value: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
value, value);
|
|
printf (" Parameter: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
parameter, parameter);
|
|
}
|
|
noErrors++;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
|
|
static void
|
|
output_result (const char *test_name, int result,
|
|
MATHTYPE computed, MATHTYPE expected,
|
|
MATHTYPE difference,
|
|
int print_values, int print_diff)
|
|
{
|
|
++noTests;
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1 && print_values)
|
|
{
|
|
printf ("Result:\n");
|
|
printf (" is: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
computed, computed);
|
|
printf (" should be: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
expected, expected);
|
|
if (print_diff)
|
|
printf (" difference: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR
|
|
"\n", difference, difference);
|
|
}
|
|
++noErrors;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
|
|
static void
|
|
output_result_ext (const char *test_name, int result,
|
|
MATHTYPE computed, MATHTYPE expected,
|
|
MATHTYPE difference,
|
|
MATHTYPE parameter,
|
|
int print_values, int print_diff)
|
|
{
|
|
++noTests;
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1 && print_values)
|
|
{
|
|
printf ("Result:\n");
|
|
printf (" is: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
computed, computed);
|
|
printf (" should be: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
expected, expected);
|
|
if (print_diff)
|
|
printf (" difference: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR
|
|
"\n", difference, difference);
|
|
printf ("Parameter: % .20" PRINTF_EXPR " % .20" PRINTF_XEXPR "\n",
|
|
parameter, parameter);
|
|
}
|
|
++noErrors;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
/*
|
|
check that computed and expected values are the same
|
|
*/
|
|
static void
|
|
check (const char *test_name, MATHTYPE computed, MATHTYPE expected)
|
|
{
|
|
MATHTYPE diff;
|
|
int result;
|
|
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
result = check_equal (computed, expected, 0, &diff);
|
|
output_result (test_name, result,
|
|
computed, expected, diff, PRINT, PRINT);
|
|
}
|
|
|
|
|
|
/*
|
|
check that computed and expected values are the same,
|
|
outputs the parameter to the function
|
|
*/
|
|
static void
|
|
check_ext (const char *test_name, MATHTYPE computed, MATHTYPE expected,
|
|
MATHTYPE parameter)
|
|
{
|
|
MATHTYPE diff;
|
|
int result;
|
|
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
result = check_equal (computed, expected, 0, &diff);
|
|
output_result_ext (test_name, result,
|
|
computed, expected, diff, parameter, PRINT, PRINT);
|
|
}
|
|
|
|
|
|
/*
|
|
check that computed and expected values are the same and
|
|
checks also for exception flags
|
|
*/
|
|
static void
|
|
check_exc (const char *test_name, MATHTYPE computed, MATHTYPE expected,
|
|
short exception)
|
|
{
|
|
MATHTYPE diff;
|
|
int result;
|
|
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, exception);
|
|
result = check_equal (computed, expected, 0, &diff);
|
|
output_result (test_name, result,
|
|
computed, expected, diff, PRINT, PRINT);
|
|
}
|
|
|
|
/*
|
|
check that computed and expected values are close enough
|
|
*/
|
|
static void
|
|
check_eps (const char *test_name, MATHTYPE computed, MATHTYPE expected,
|
|
MATHTYPE epsilon)
|
|
{
|
|
MATHTYPE diff;
|
|
int result;
|
|
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
result = check_equal (computed, expected, epsilon, &diff);
|
|
output_result (test_name, result,
|
|
computed, expected, diff, PRINT, PRINT);
|
|
}
|
|
|
|
/*
|
|
check a boolean condition
|
|
*/
|
|
static void
|
|
check_bool (const char *test_name, int computed)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_result_bool (test_name, computed);
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
check that computed and expected values are equal (int values)
|
|
*/
|
|
static void
|
|
check_int (const char *test_name, int computed, int expected)
|
|
{
|
|
int diff = computed - expected;
|
|
int result = diff == 0;
|
|
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1)
|
|
{
|
|
printf ("Result:\n");
|
|
printf (" is: %d\n", computed);
|
|
printf (" should be: %d\n", expected);
|
|
}
|
|
noErrors++;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
|
|
/*
|
|
check that computed and expected values are equal (long int values)
|
|
*/
|
|
static void
|
|
check_long (const char *test_name, long int computed, long int expected)
|
|
{
|
|
long int diff = computed - expected;
|
|
int result = diff == 0;
|
|
|
|
++noTests;
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1)
|
|
{
|
|
printf ("Result:\n");
|
|
printf (" is: %ld\n", computed);
|
|
printf (" should be: %ld\n", expected);
|
|
}
|
|
noErrors++;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
/*
|
|
check that computed and expected values are equal (long long int values)
|
|
*/
|
|
static void
|
|
check_longlong (const char *test_name, long long int computed,
|
|
long long int expected)
|
|
{
|
|
long long int diff = computed - expected;
|
|
int result = diff == 0;
|
|
|
|
++noTests;
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
|
|
if (result)
|
|
{
|
|
output_pass_value ();
|
|
}
|
|
else
|
|
{
|
|
output_fail_value (test_name);
|
|
if (verbose > 1)
|
|
{
|
|
printf ("Result:\n");
|
|
printf (" is: %lld\n", computed);
|
|
printf (" should be: %lld\n", expected);
|
|
}
|
|
noErrors++;
|
|
}
|
|
|
|
fpstack_test (test_name);
|
|
}
|
|
|
|
/*
|
|
check that computed value is not-a-number
|
|
*/
|
|
static void
|
|
check_isnan (const char *test_name, MATHTYPE computed)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_isvalue (test_name, isnan (computed), computed);
|
|
}
|
|
|
|
|
|
/*
|
|
check that computed value is not-a-number and test for exceptions
|
|
*/
|
|
static void
|
|
check_isnan_exc (const char *test_name, MATHTYPE computed,
|
|
short exception)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, exception);
|
|
output_isvalue (test_name, isnan (computed), computed);
|
|
}
|
|
|
|
|
|
/*
|
|
check that computed value is not-a-number and test for exceptions
|
|
*/
|
|
static void
|
|
check_isnan_maybe_exc (const char *test_name, MATHTYPE computed,
|
|
short exception)
|
|
{
|
|
output_new_test (test_name);
|
|
test_not_exception (test_name, exception);
|
|
output_isvalue (test_name, isnan (computed), computed);
|
|
}
|
|
|
|
/*
|
|
check that computed value is not-a-number and supply parameter
|
|
*/
|
|
#ifndef TEST_INLINE
|
|
static void
|
|
check_isnan_ext (const char *test_name, MATHTYPE computed,
|
|
MATHTYPE parameter)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_isvalue_ext (test_name, isnan (computed), computed, parameter);
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
check that computed value is not-a-number, test for exceptions
|
|
and supply parameter
|
|
*/
|
|
static void
|
|
check_isnan_exc_ext (const char *test_name, MATHTYPE computed,
|
|
short exception, MATHTYPE parameter)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name,exception);
|
|
output_isvalue_ext (test_name, isnan (computed), computed, parameter);
|
|
}
|
|
|
|
|
|
/* Tests if computed is +Inf */
|
|
static void
|
|
check_isinfp (const char *test_name, MATHTYPE computed)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_isvalue (test_name, (ISINF (computed) == +1), computed);
|
|
}
|
|
|
|
|
|
static void
|
|
check_isinfp_ext (const char *test_name, MATHTYPE computed,
|
|
MATHTYPE parameter)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_isvalue_ext (test_name, (ISINF (computed) == +1), computed, parameter);
|
|
}
|
|
|
|
|
|
/* Tests if computed is +Inf */
|
|
static void
|
|
check_isinfp_exc (const char *test_name, MATHTYPE computed,
|
|
int exception)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, exception);
|
|
output_isvalue (test_name, (ISINF (computed) == +1), computed);
|
|
}
|
|
|
|
/* Tests if computed is -Inf */
|
|
static void
|
|
check_isinfn (const char *test_name, MATHTYPE computed)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_isvalue (test_name, (ISINF (computed) == -1), computed);
|
|
}
|
|
|
|
|
|
#ifndef TEST_INLINE
|
|
static void
|
|
check_isinfn_ext (const char *test_name, MATHTYPE computed,
|
|
MATHTYPE parameter)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, NO_EXCEPTION);
|
|
output_isvalue_ext (test_name, (ISINF (computed) == -1), computed, parameter);
|
|
}
|
|
#endif
|
|
|
|
|
|
/* Tests if computed is -Inf */
|
|
static void
|
|
check_isinfn_exc (const char *test_name, MATHTYPE computed,
|
|
int exception)
|
|
{
|
|
output_new_test (test_name);
|
|
test_exceptions (test_name, exception);
|
|
output_isvalue (test_name, (ISINF (computed) == -1), computed);
|
|
}
|
|
|
|
|
|
/* This is to prevent messages from the SVID libm emulation. */
|
|
int
|
|
matherr (struct exception *x __attribute__ ((unused)))
|
|
{
|
|
return 1;
|
|
}
|
|
|
|
|
|
/****************************************************************************
|
|
Test for single functions of libm
|
|
****************************************************************************/
|
|
|
|
static void
|
|
acos_test (void)
|
|
{
|
|
#ifndef TEST_INLINE
|
|
MATHTYPE x;
|
|
|
|
x = random_greater (1);
|
|
check_isnan_exc ("acos (x) == NaN plus invalid exception for |x| > 1",
|
|
FUNC(acos) (x),
|
|
INVALID_EXCEPTION);
|
|
|
|
x = random_less (1);
|
|
check_isnan_exc ("acos (x) == NaN plus invalid exception for |x| > 1",
|
|
FUNC(acos) (x),
|
|
INVALID_EXCEPTION);
|
|
#endif
|
|
check ("acos (0) == pi/2", FUNC(acos) (0), M_PI_2l);
|
|
check ("acos (-0) == pi/2", FUNC(acos) (minus_zero), M_PI_2l);
|
|
|
|
check ("acos (1) == 0", FUNC(acos) (1), 0);
|
|
check ("acos (-1) == pi", FUNC(acos) (-1), M_PIl);
|
|
|
|
check_eps ("acos (0.5) == pi/3", FUNC(acos) (0.5), M_PI_6l * 2.0,
|
|
CHOOSE (1e-18, 0, 0));
|
|
check_eps ("acos (-0.5) == 2*pi/3", FUNC(acos) (-0.5), M_PI_6l * 4.0,
|
|
CHOOSE (1e-17, 0, 0));
|
|
|
|
check_eps ("acos (0.7) == 0.795398830...", FUNC(acos) (0.7),
|
|
0.7953988301841435554L, CHOOSE(7e-17L, 0, 0));
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
acosh_test (void)
|
|
{
|
|
#ifndef TEST_INLINE
|
|
MATHTYPE x;
|
|
|
|
check_isinfp ("acosh(+inf) == +inf", FUNC(acosh) (plus_infty));
|
|
|
|
x = random_less (1);
|
|
check_isnan_exc ("acosh(x) == NaN plus invalid exception if x < 1",
|
|
FUNC(acosh) (x), INVALID_EXCEPTION);
|
|
#endif
|
|
|
|
check ("acosh(1) == 0", FUNC(acosh) (1), 0);
|
|
check_eps ("acosh(7) == 2.633915793...", FUNC(acosh) (7),
|
|
2.6339157938496334172L, CHOOSE (3e-19, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
asin_test (void)
|
|
{
|
|
#ifndef TEST_INLINE
|
|
MATHTYPE x;
|
|
|
|
x = random_greater (1);
|
|
check_isnan_exc ("asin x == NaN plus invalid exception for |x| > 1",
|
|
FUNC(asin) (x),
|
|
INVALID_EXCEPTION);
|
|
|
|
x = random_less (1);
|
|
check_isnan_exc ("asin x == NaN plus invalid exception for |x| > 1",
|
|
FUNC(asin) (x),
|
|
INVALID_EXCEPTION);
|
|
#endif
|
|
|
|
check ("asin (0) == 0", FUNC(asin) (0), 0);
|
|
check ("asin (-0) == -0", FUNC(asin) (minus_zero), minus_zero);
|
|
check_eps ("asin (0.5) == pi/6", FUNC(asin) (0.5), M_PI_6l,
|
|
CHOOSE(3.5e-18, 0, 2e-7));
|
|
check_eps ("asin (-0.5) == -pi/6", FUNC(asin) (-0.5), -M_PI_6l,
|
|
CHOOSE(3.5e-18, 0, 2e-7));
|
|
check ("asin (1.0) == pi/2", FUNC(asin) (1.0), M_PI_2l);
|
|
check ("asin (-1.0) == -pi/2", FUNC(asin) (-1.0), -M_PI_2l);
|
|
check_eps ("asin (0.7) == 0.775397496...", FUNC(asin) (0.7),
|
|
0.7753974966107530637L, CHOOSE(7e-17L, 2e-16, 2e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
asinh_test (void)
|
|
{
|
|
|
|
check ("asinh(+0) == +0", FUNC(asinh) (0), 0);
|
|
#ifndef TEST_INLINE
|
|
check ("asinh(-0) == -0", FUNC(asinh) (minus_zero), minus_zero);
|
|
check_isinfp ("asinh(+inf) == +inf", FUNC(asinh) (plus_infty));
|
|
check_isinfn ("asinh(-inf) == -inf", FUNC(asinh) (minus_infty));
|
|
#endif
|
|
check_eps ("asinh(0.7) == 0.652666566...", FUNC(asinh) (0.7),
|
|
0.652666566082355786L, CHOOSE(4e-17L, 0, 6e-8));
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
atan_test (void)
|
|
{
|
|
check ("atan (0) == 0", FUNC(atan) (0), 0);
|
|
check ("atan (-0) == -0", FUNC(atan) (minus_zero), minus_zero);
|
|
|
|
check ("atan (+inf) == pi/2", FUNC(atan) (plus_infty), M_PI_2l);
|
|
check ("atan (-inf) == -pi/2", FUNC(atan) (minus_infty), -M_PI_2l);
|
|
|
|
check_eps ("atan (1) == pi/4", FUNC(atan) (1), M_PI_4l,
|
|
CHOOSE (1e-18, 0, 0));
|
|
check_eps ("atan (-1) == -pi/4", FUNC(atan) (1), M_PI_4l,
|
|
CHOOSE (1e-18, 0, 0));
|
|
|
|
check_eps ("atan (0.7) == 0.610725964...", FUNC(atan) (0.7),
|
|
0.6107259643892086165L, CHOOSE(3e-17L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
atan2_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
x = random_greater (0);
|
|
check ("atan2 (0,x) == 0 for x > 0",
|
|
FUNC(atan2) (0, x), 0);
|
|
x = random_greater (0);
|
|
check ("atan2 (-0,x) == -0 for x > 0",
|
|
FUNC(atan2) (minus_zero, x), minus_zero);
|
|
|
|
check ("atan2 (+0,+0) == +0", FUNC(atan2) (0, 0), 0);
|
|
check ("atan2 (-0,+0) == -0", FUNC(atan2) (minus_zero, 0), minus_zero);
|
|
|
|
x = -random_greater (0);
|
|
check ("atan2 (+0,x) == +pi for x < 0", FUNC(atan2) (0, x), M_PIl);
|
|
|
|
x = -random_greater (0);
|
|
check ("atan2 (-0,x) == -pi for x < 0", FUNC(atan2) (minus_zero, x), -M_PIl);
|
|
|
|
check ("atan2 (+0,-0) == +pi", FUNC(atan2) (0, minus_zero), M_PIl);
|
|
check ("atan2 (-0,-0) == -pi", FUNC(atan2) (minus_zero, minus_zero), -M_PIl);
|
|
|
|
x = random_greater (0);
|
|
check ("atan2 (y,+0) == pi/2 for y > 0", FUNC(atan2) (x, 0), M_PI_2l);
|
|
|
|
x = random_greater (0);
|
|
check ("atan2 (y,-0) == pi/2 for y > 0", FUNC(atan2) (x, minus_zero),
|
|
M_PI_2l);
|
|
|
|
x = random_less (0);
|
|
check ("atan2 (y,+0) == -pi/2 for y < 0", FUNC(atan2) (x, 0), -M_PI_2l);
|
|
|
|
x = random_less (0);
|
|
check ("atan2 (y,-0) == -pi/2 for y < 0", FUNC(atan2) (x, minus_zero),
|
|
-M_PI_2l);
|
|
|
|
x = random_greater (0);
|
|
check ("atan2 (y,inf) == +0 for finite y > 0",
|
|
FUNC(atan2) (x, plus_infty), 0);
|
|
|
|
x = -random_greater (0);
|
|
check ("atan2 (y,inf) == -0 for finite y < 0",
|
|
FUNC(atan2) (x, plus_infty), minus_zero);
|
|
|
|
x = random_value (-1e4, 1e4);
|
|
check ("atan2(+inf, x) == pi/2 for finite x",
|
|
FUNC(atan2) (plus_infty, x), M_PI_2l);
|
|
|
|
x = random_value (-1e4, 1e4);
|
|
check ("atan2(-inf, x) == -pi/2 for finite x",
|
|
FUNC(atan2) (minus_infty, x), -M_PI_2l);
|
|
|
|
x = random_greater (0);
|
|
check ("atan2 (y,-inf) == +pi for finite y > 0",
|
|
FUNC(atan2) (x, minus_infty), M_PIl);
|
|
|
|
x = -random_greater (0);
|
|
check ("atan2 (y,-inf) == -pi for finite y < 0",
|
|
FUNC(atan2) (x, minus_infty), -M_PIl);
|
|
|
|
check ("atan2 (+inf,+inf) == +pi/4",
|
|
FUNC(atan2) (plus_infty, plus_infty), M_PI_4l);
|
|
|
|
check ("atan2 (-inf,+inf) == -pi/4",
|
|
FUNC(atan2) (minus_infty, plus_infty), -M_PI_4l);
|
|
|
|
check ("atan2 (+inf,-inf) == +3*pi/4",
|
|
FUNC(atan2) (plus_infty, minus_infty), 3 * M_PI_4l);
|
|
|
|
check ("atan2 (-inf,-inf) == -3*pi/4",
|
|
FUNC(atan2) (minus_infty, minus_infty), -3 * M_PI_4l);
|
|
|
|
/* FIXME: Add some specific tests */
|
|
check_eps ("atan2 (0.7,1) == 0.61072...", FUNC(atan2) (0.7,1),
|
|
0.6107259643892086165L, CHOOSE(3e-17L, 0, 0));
|
|
check_eps ("atan2 (0.4,0.0003) == 1.57004...", FUNC(atan2) (0.4, 0.0003),
|
|
1.5700463269355215718L, CHOOSE(2e-19L, 0, 0));
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
atanh_test (void)
|
|
{
|
|
#ifndef TEST_INLINE
|
|
MATHTYPE x;
|
|
#endif
|
|
|
|
check ("atanh(+0) == +0", FUNC(atanh) (0), 0);
|
|
#ifndef TEST_INLINE
|
|
check ("atanh(-0) == -0", FUNC(atanh) (minus_zero), minus_zero);
|
|
|
|
check_isinfp_exc ("atanh(+1) == +inf plus divide-by-zero exception",
|
|
FUNC(atanh) (1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfn_exc ("atanh(-1) == -inf plus divide-by-zero exception",
|
|
FUNC(atanh) (-1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
x = random_greater (1.0);
|
|
check_isnan_exc_ext ("atanh (x) == NaN plus invalid exception if |x| > 1",
|
|
FUNC(atanh) (x), INVALID_EXCEPTION, x);
|
|
|
|
x = random_less (1.0);
|
|
check_isnan_exc_ext ("atanh (x) == NaN plus invalid exception if |x| > 1",
|
|
FUNC(atanh) (x), INVALID_EXCEPTION, x);
|
|
|
|
#endif
|
|
check_eps ("atanh(0.7) == 0.867300527...", FUNC(atanh) (0.7),
|
|
0.8673005276940531944L, CHOOSE(9e-17L, 2e-16, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
cbrt_test (void)
|
|
{
|
|
check ("cbrt (+0) == +0", FUNC(cbrt) (0.0), 0.0);
|
|
check ("cbrt (-0) == -0", FUNC(cbrt) (minus_zero), minus_zero);
|
|
|
|
#ifndef TEST_INLINE
|
|
check_isinfp ("cbrt (+inf) == +inf", FUNC(cbrt) (plus_infty));
|
|
check_isinfn ("cbrt (-inf) == -inf", FUNC(cbrt) (minus_infty));
|
|
check_isnan ("cbrt (NaN) == NaN", FUNC(cbrt) (nan_value));
|
|
#endif
|
|
check_eps ("cbrt (-0.001) == -0.1", FUNC(cbrt) (-0.001), -0.1,
|
|
CHOOSE (5e-18L, 0, 0));
|
|
check_eps ("cbrt (8) == 2", FUNC(cbrt) (8), 2, CHOOSE (5e-17L, 0, 0));
|
|
check_eps ("cbrt (-27) == -3", FUNC(cbrt) (-27.0), -3.0,
|
|
CHOOSE (3e-16L, 5e-16, 0));
|
|
check_eps ("cbrt (0.970299) == 0.99", FUNC(cbrt) (0.970299), 0.99,
|
|
CHOOSE (2e-17L, 2e-16, 0));
|
|
check_eps ("cbrt (0.7) == .8879040017...", FUNC(cbrt) (0.7),
|
|
0.8879040017426007084L, CHOOSE(2e-17L, 6e-16, 0));
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
ceil_test (void)
|
|
{
|
|
check ("ceil (+0) == +0", FUNC(ceil) (0.0), 0.0);
|
|
check ("ceil (-0) == -0", FUNC(ceil) (minus_zero), minus_zero);
|
|
check_isinfp ("ceil (+inf) == +inf", FUNC(ceil) (plus_infty));
|
|
check_isinfn ("ceil (-inf) == -inf", FUNC(ceil) (minus_infty));
|
|
|
|
check ("ceil (pi) == 4", FUNC(ceil) (M_PIl), 4.0);
|
|
check ("ceil (-pi) == -3", FUNC(ceil) (-M_PIl), -3.0);
|
|
}
|
|
|
|
|
|
static void
|
|
cos_test (void)
|
|
{
|
|
|
|
check ("cos (+0) == 1", FUNC(cos) (0), 1);
|
|
check ("cos (-0) == 1", FUNC(cos) (minus_zero), 1);
|
|
check_isnan_exc ("cos (+inf) == NaN plus invalid exception",
|
|
FUNC(cos) (plus_infty),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_exc ("cos (-inf) == NaN plus invalid exception",
|
|
FUNC(cos) (minus_infty),
|
|
INVALID_EXCEPTION);
|
|
|
|
check_eps ("cos (pi/3) == 0.5", FUNC(cos) (M_PI_6l * 2.0),
|
|
0.5, CHOOSE (4e-18L, 1e-15L, 1e-7L));
|
|
check_eps ("cos (2*pi/3) == -0.5", FUNC(cos) (M_PI_6l * 4.0),
|
|
-0.5, CHOOSE (4e-18L, 1e-15L, 1e-7L));
|
|
check_eps ("cos (pi/2) == 0", FUNC(cos) (M_PI_2l),
|
|
0, CHOOSE (1e-19L, 1e-16L, 1e-7L));
|
|
|
|
check_eps ("cos (0.7) == 0.7648421872...", FUNC(cos) (0.7),
|
|
0.7648421872844884262L, CHOOSE(3e-17, 2e-16, 6e-8));
|
|
}
|
|
|
|
static void
|
|
cosh_test (void)
|
|
{
|
|
check ("cosh (+0) == 1", FUNC(cosh) (0), 1);
|
|
check ("cosh (-0) == 1", FUNC(cosh) (minus_zero), 1);
|
|
|
|
#ifndef TEST_INLINE
|
|
check_isinfp ("cosh (+inf) == +inf", FUNC(cosh) (plus_infty));
|
|
check_isinfp ("cosh (-inf) == +inf", FUNC(cosh) (minus_infty));
|
|
#endif
|
|
|
|
check_eps ("cosh (0.7) == 1.2551690056...", FUNC(cosh) (0.7),
|
|
1.255169005630943018L, CHOOSE(4e-17L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
erf_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(erf) (0);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
|
|
check ("erf (+0) == +0", FUNC(erf) (0), 0);
|
|
check ("erf (-0) == -0", FUNC(erf) (minus_zero), minus_zero);
|
|
check ("erf (+inf) == +1", FUNC(erf) (plus_infty), 1);
|
|
check ("erf (-inf) == -1", FUNC(erf) (minus_infty), -1);
|
|
|
|
check_eps ("erf (0.7) == 0.6778011938...", FUNC(erf) (0.7),
|
|
0.67780119383741847297L, CHOOSE(0, 2e-16, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
erfc_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(erfc) (0);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
|
|
check ("erfc (+inf) == 0", FUNC(erfc) (plus_infty), 0.0);
|
|
check ("erfc (-inf) == 2", FUNC(erfc) (minus_infty), 2.0);
|
|
check ("erfc (+0) == 1", FUNC(erfc) (0.0), 1.0);
|
|
check ("erfc (-0) == 1", FUNC(erfc) (minus_zero), 1.0);
|
|
|
|
check_eps ("erfc (0.7) == 0.3221988061...", FUNC(erfc) (0.7),
|
|
0.32219880616258152702L, CHOOSE(0, 6e-17, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
exp_test (void)
|
|
{
|
|
check ("exp (+0) == 1", FUNC(exp) (0), 1);
|
|
check ("exp (-0) == 1", FUNC(exp) (minus_zero), 1);
|
|
|
|
#ifndef TEST_INLINE
|
|
check_isinfp ("exp (+inf) == +inf", FUNC(exp) (plus_infty));
|
|
check ("exp (-inf) == 0", FUNC(exp) (minus_infty), 0);
|
|
#endif
|
|
check_eps ("exp (1) == e", FUNC(exp) (1), M_El, CHOOSE (4e-18L, 0, 0));
|
|
|
|
check_eps ("exp (2) == e^2", FUNC(exp) (2), M_E2l,
|
|
CHOOSE (1e-18, 0, 0));
|
|
check_eps ("exp (3) == e^3", FUNC(exp) (3), M_E3l,
|
|
CHOOSE (1.5e-17, 0, 0));
|
|
check_eps ("exp (0.7) == 2.0137527074...", FUNC(exp) (0.7),
|
|
2.0137527074704765216L, CHOOSE(9e-17L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
exp10_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(exp10) (0);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
|
|
check ("exp10 (+0) == 1", FUNC(exp10) (0), 1);
|
|
check ("exp10 (-0) == 1", FUNC(exp10) (minus_zero), 1);
|
|
|
|
check_isinfp ("exp10 (+inf) == +inf", FUNC(exp10) (plus_infty));
|
|
check ("exp10 (-inf) == 0", FUNC(exp10) (minus_infty), 0);
|
|
check_eps ("exp10 (3) == 1000", FUNC(exp10) (3), 1000,
|
|
CHOOSE(5e-16, 7e-13, 2e-4));
|
|
check_eps ("exp10 (-1) == 0.1", FUNC(exp10) (-1), 0.1,
|
|
CHOOSE(6e-18, 3e-17, 8e-09));
|
|
check_isinfp ("exp10 (1e6) == +inf", FUNC(exp10) (1e6));
|
|
check ("exp10 (-1e6) == 0", FUNC(exp10) (-1e6), 0);
|
|
check_eps ("exp10 (0.7) == 5.0118723...", FUNC(exp10) (0.7),
|
|
5.0118723362727228500L, CHOOSE(6e-16, 9e-16, 5e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
exp2_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(exp2) (0);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
|
|
check ("exp2 (+0) == 1", FUNC(exp2) (0), 1);
|
|
check ("exp2 (-0) == 1", FUNC(exp2) (minus_zero), 1);
|
|
|
|
check_isinfp ("exp2 (+inf) == +inf", FUNC(exp2) (plus_infty));
|
|
check ("exp2 (-inf) == 0", FUNC(exp2) (minus_infty), 0);
|
|
check ("exp2 (10) == 1024", FUNC(exp2) (10), 1024);
|
|
check ("exp2 (-1) == 0.5", FUNC(exp2) (-1), 0.5);
|
|
check_isinfp ("exp2 (1e6) == +inf", FUNC(exp2) (1e6));
|
|
check ("exp2 (-1e6) == 0", FUNC(exp2) (-1e6), 0);
|
|
check_eps ("exp2 (0.7) == 1.6245047927...", FUNC(exp2) (0.7),
|
|
1.6245047927124710452L, CHOOSE(6e-17L, 0, 6e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
expm1_test (void)
|
|
{
|
|
check ("expm1 (+0) == 0", FUNC(expm1) (0), 0);
|
|
#ifndef TEST_INLINE
|
|
check ("expm1 (-0) == -0", FUNC(expm1) (minus_zero), minus_zero);
|
|
|
|
check_isinfp ("expm1 (+inf) == +inf", FUNC(expm1) (plus_infty));
|
|
check ("expm1 (-inf) == -1", FUNC(expm1) (minus_infty), -1);
|
|
#endif
|
|
|
|
check_eps ("expm1 (1) == e-1", FUNC(expm1) (1), M_El - 1.0,
|
|
CHOOSE (4e-18L, 0, 2e-7));
|
|
|
|
check_eps ("expm1 (0.7) == 1.01375...", FUNC(expm1) (0.7),
|
|
1.0137527074704765216L, CHOOSE(9e-17L, 0, 0));
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
check_frexp (const char *test_name, MATHTYPE computed, MATHTYPE expected,
|
|
int comp_int, int exp_int)
|
|
{
|
|
MATHTYPE diff;
|
|
int result;
|
|
|
|
result = (check_equal (computed, expected, 0, &diff)
|
|
&& (comp_int == exp_int));
|
|
|
|
if (result)
|
|
{
|
|
if (verbose > 2)
|
|
printf ("Pass: %s\n", test_name);
|
|
}
|
|
else
|
|
{
|
|
if (verbose)
|
|
printf ("Fail: %s\n", test_name);
|
|
if (verbose > 1)
|
|
{
|
|
printf ("Result:\n");
|
|
printf (" is: %.20" PRINTF_EXPR " *2^%d %.20"
|
|
PRINTF_XEXPR "*2^%d\n",
|
|
computed, comp_int, computed, comp_int);
|
|
printf (" should be: %.20" PRINTF_EXPR " *2^%d %.20"
|
|
PRINTF_XEXPR "*2^%d\n",
|
|
expected, exp_int, expected, exp_int);
|
|
printf (" difference: %.20" PRINTF_EXPR " %.20" PRINTF_XEXPR "\n",
|
|
diff, diff);
|
|
}
|
|
noErrors++;
|
|
}
|
|
fpstack_test (test_name);
|
|
output_result (test_name, result,
|
|
computed, expected, diff, PRINT, PRINT);
|
|
}
|
|
|
|
|
|
static void
|
|
frexp_test (void)
|
|
{
|
|
int x_int;
|
|
MATHTYPE result;
|
|
|
|
result = FUNC(frexp) (plus_infty, &x_int);
|
|
check_isinfp ("frexp (+inf, expr) == +inf", result);
|
|
|
|
result = FUNC(frexp) (minus_infty, &x_int);
|
|
check_isinfn ("frexp (-inf, expr) == -inf", result);
|
|
|
|
result = FUNC(frexp) (nan_value, &x_int);
|
|
check_isnan ("frexp (Nan, expr) == NaN", result);
|
|
|
|
result = FUNC(frexp) (0, &x_int);
|
|
check_frexp ("frexp: +0 == 0 * 2^0", result, 0, x_int, 0);
|
|
|
|
result = FUNC(frexp) (minus_zero, &x_int);
|
|
check_frexp ("frexp: -0 == -0 * 2^0", result, minus_zero, x_int, 0);
|
|
|
|
result = FUNC(frexp) (12.8L, &x_int);
|
|
check_frexp ("frexp: 12.8 == 0.8 * 2^4", result, 0.8L, x_int, 4);
|
|
|
|
result = FUNC(frexp) (-27.34L, &x_int);
|
|
check_frexp ("frexp: -27.34 == -0.854375 * 2^5", result, -0.854375L, x_int, 5);
|
|
|
|
}
|
|
|
|
|
|
#if __GLIBC__ < 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ < 1)
|
|
/* All floating-point numbers can be put in one of these categories. */
|
|
enum
|
|
{
|
|
FP_NAN,
|
|
#define FP_NAN FP_NAN
|
|
FP_INFINITE,
|
|
#define FP_INFINITE FP_INFINITE
|
|
FP_ZERO,
|
|
#define FP_ZERO FP_ZERO
|
|
FP_SUBNORMAL,
|
|
#define FP_SUBNORMAL FP_SUBNORMAL
|
|
FP_NORMAL
|
|
#define FP_NORMAL FP_NORMAL
|
|
};
|
|
#endif
|
|
|
|
|
|
static void
|
|
fpclassify_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
/* fpclassify is a macro, don't give it constants as parameter */
|
|
check_bool ("fpclassify (NaN) == FP_NAN", fpclassify (nan_value) == FP_NAN);
|
|
check_bool ("fpclassify (+inf) == FP_INFINITE",
|
|
fpclassify (plus_infty) == FP_INFINITE);
|
|
check_bool ("fpclassify (-inf) == FP_INFINITE",
|
|
fpclassify (minus_infty) == FP_INFINITE);
|
|
check_bool ("fpclassify (+0) == FP_ZERO",
|
|
fpclassify (plus_zero) == FP_ZERO);
|
|
check_bool ("fpclassify (-0) == FP_ZERO",
|
|
fpclassify (minus_zero) == FP_ZERO);
|
|
|
|
x = 1000.0;
|
|
check_bool ("fpclassify (1000) == FP_NORMAL",
|
|
fpclassify (x) == FP_NORMAL);
|
|
}
|
|
|
|
|
|
static void
|
|
isfinite_test (void)
|
|
{
|
|
check_bool ("isfinite (0) != 0", isfinite (0));
|
|
check_bool ("isfinite (-0) != 0", isfinite (minus_zero));
|
|
check_bool ("isfinite (10) != 0", isfinite (10));
|
|
check_bool ("isfinite (+inf) == 0", isfinite (plus_infty) == 0);
|
|
check_bool ("isfinite (-inf) == 0", isfinite (minus_infty) == 0);
|
|
check_bool ("isfinite (NaN) == 0", isfinite (nan_value) == 0);
|
|
}
|
|
|
|
|
|
static void
|
|
isnormal_test (void)
|
|
{
|
|
check_bool ("isnormal (0) == 0", isnormal (0) == 0);
|
|
check_bool ("isnormal (-0) == 0", isnormal (minus_zero) == 0);
|
|
check_bool ("isnormal (10) != 0", isnormal (10));
|
|
check_bool ("isnormal (+inf) == 0", isnormal (plus_infty) == 0);
|
|
check_bool ("isnormal (-inf) == 0", isnormal (minus_infty) == 0);
|
|
check_bool ("isnormal (NaN) == 0", isnormal (nan_value) == 0);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
signbit_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
check_bool ("signbit (+0) == 0", signbit (0) == 0);
|
|
check_bool ("signbit (-0) != 0", signbit (minus_zero));
|
|
check_bool ("signbit (+inf) == 0", signbit (plus_infty) == 0);
|
|
check_bool ("signbit (-inf) != 0", signbit (minus_infty));
|
|
|
|
x = random_less (0);
|
|
check_bool ("signbit (x) != 0 for x < 0", signbit (x));
|
|
|
|
x = random_greater (0);
|
|
check_bool ("signbit (x) == 0 for x > 0", signbit (x) == 0);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
gamma_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(gamma) (1);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
|
|
check_isinfp ("gamma (+inf) == +inf", FUNC(gamma) (plus_infty));
|
|
check_isinfp_exc ("gamma (0) == +inf plus divide by zero exception",
|
|
FUNC(gamma) (0), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check_isinfp_exc ("gamma (x) == +inf plus divide by zero exception for integer x <= 0",
|
|
FUNC(gamma) (-3), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isnan_exc ("gamma (-inf) == NaN plus invalid exception",
|
|
FUNC(gamma) (minus_infty), INVALID_EXCEPTION);
|
|
|
|
signgam = 0;
|
|
check ("gamma (1) == 0", FUNC(gamma) (1), 0);
|
|
check_int ("gamma (1) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check ("gamma (3) == M_LN2", FUNC(gamma) (3), M_LN2l);
|
|
check_int ("gamma (3) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check_eps ("gamma (0.5) == log(sqrt(pi))", FUNC(gamma) (0.5),
|
|
FUNC(log) (FUNC(sqrt) (M_PIl)), CHOOSE (0, 1e-15, 1e-7));
|
|
check_int ("gamma (0.5) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check_eps ("gamma (-0.5) == log(2*sqrt(pi))", FUNC(gamma) (-0.5),
|
|
FUNC(log) (2*FUNC(sqrt) (M_PIl)), CHOOSE (0, 1e-15, 0));
|
|
|
|
check_int ("gamma (-0.5) sets signgam to -1", signgam, -1);
|
|
}
|
|
|
|
|
|
static void
|
|
tgamma_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(tgamma) (1);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
|
|
check_isinfp ("tgamma (+inf) == +inf", FUNC(tgamma) (plus_infty));
|
|
check_isnan_exc ("tgamma (0) == NaN plus invalid exception",
|
|
FUNC(tgamma) (0), INVALID_EXCEPTION);
|
|
|
|
check_isnan_exc_ext ("tgamma (x) == NaN plus invalid exception for integer x <= 0",
|
|
FUNC(tgamma) (-2), INVALID_EXCEPTION, -2);
|
|
check_isnan_exc ("tgamma (-inf) == NaN plus invalid exception",
|
|
FUNC(tgamma) (minus_infty), INVALID_EXCEPTION);
|
|
|
|
#ifdef TODO
|
|
check_eps ("tgamma (0.5) == sqrt(pi)", FUNC(tgamma) (0.5),
|
|
FUNC(sqrt) (M_PIl), CHOOSE (0, 5e-16, 2e-7));
|
|
#endif
|
|
check_eps ("tgamma (-0.5) == -2*sqrt(pi)", FUNC(tgamma) (-0.5),
|
|
-2*FUNC(sqrt) (M_PIl), CHOOSE (0, 5e-16, 3e-7));
|
|
|
|
check ("tgamma (1) == 1", FUNC(tgamma) (1), 1);
|
|
check ("tgamma (4) == 6", FUNC(tgamma) (4), 6);
|
|
|
|
check_eps ("tgamma (0.7) == 1.29805...", FUNC(tgamma) (0.7),
|
|
1.29805533264755778568L, CHOOSE(0, 3e-16, 2e-7));
|
|
check ("tgamma (1.2) == 0.91816...", FUNC(tgamma) (1.2),
|
|
0.91816874239976061064L);
|
|
}
|
|
|
|
|
|
static void
|
|
lgamma_test (void)
|
|
{
|
|
errno = 0;
|
|
FUNC(lgamma) (0);
|
|
if (errno == ENOSYS)
|
|
/* Function not implemented. */
|
|
return;
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
check_isinfp ("lgamma (+inf) == +inf", FUNC(lgamma) (plus_infty));
|
|
check_isinfp_exc ("lgamma (0) == +inf plus divide by zero exception",
|
|
FUNC(lgamma) (0), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check_isinfp_exc ("lgamma (x) == +inf plus divide by zero exception for integer x <= 0",
|
|
FUNC(lgamma) (-3), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isnan_exc ("lgamma (-inf) == NaN plus invalid exception",
|
|
FUNC(lgamma) (minus_infty), INVALID_EXCEPTION);
|
|
|
|
signgam = 0;
|
|
check ("lgamma (1) == 0", FUNC(lgamma) (1), 0);
|
|
check_int ("lgamma (0) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check ("lgamma (3) == M_LN2", FUNC(lgamma) (3), M_LN2l);
|
|
check_int ("lgamma (3) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check_eps ("lgamma (0.5) == log(sqrt(pi))", FUNC(lgamma) (0.5),
|
|
FUNC(log) (FUNC(sqrt) (M_PIl)), CHOOSE (0, 1e-15, 1e-7));
|
|
check_int ("lgamma (0.5) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check_eps ("lgamma (-0.5) == log(2*sqrt(pi))", FUNC(lgamma) (-0.5),
|
|
FUNC(log) (2*FUNC(sqrt) (M_PIl)), CHOOSE (0, 1e-15, 0));
|
|
|
|
check_int ("lgamma (-0.5) sets signgam to -1", signgam, -1);
|
|
|
|
signgam = 0;
|
|
check_eps ("lgamma (0.7) == 0.26086...", FUNC(lgamma) (0.7),
|
|
0.26086724653166651439L, CHOOSE(0, 6e-17, 3e-8));
|
|
check_int ("lgamma (0.7) sets signgam to 1", signgam, 1);
|
|
|
|
signgam = 0;
|
|
check_eps ("lgamma (1.2) == -0.08537...", FUNC(lgamma) (1.2),
|
|
-0.853740900033158497197e-1L, CHOOSE(0, 2e-17, 2e-8));
|
|
check_int ("lgamma (1.2) sets signgam to 1", signgam, 1);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
ilogb_test (void)
|
|
{
|
|
int i;
|
|
|
|
check_int ("ilogb (1) == 0", FUNC(ilogb) (1), 0);
|
|
check_int ("ilogb (e) == 1", FUNC(ilogb) (M_El), 1);
|
|
check_int ("ilogb (1024) == 10", FUNC(ilogb) (1024), 10);
|
|
check_int ("ilogb (-2000) == 10", FUNC(ilogb) (-2000), 10);
|
|
|
|
/* XXX We have a problem here: the standard does not tell us whether
|
|
exceptions are allowed/required. ignore them for now. */
|
|
i = FUNC (ilogb) (0.0);
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
check_int ("ilogb (0) == FP_ILOGB0", i, FP_ILOGB0);
|
|
i = FUNC(ilogb) (nan_value);
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
check_int ("ilogb (NaN) == FP_ILOGBNAN", i, FP_ILOGBNAN);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
ldexp_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
check ("ldexp (0, 0) == 0", FUNC(ldexp) (0, 0), 0);
|
|
|
|
check_isinfp ("ldexp (+inf, 1) == +inf", FUNC(ldexp) (plus_infty, 1));
|
|
check_isinfn ("ldexp (-inf, 1) == -inf", FUNC(ldexp) (minus_infty, 1));
|
|
check_isnan ("ldexp (NaN, 1) == NaN", FUNC(ldexp) (nan_value, 1));
|
|
|
|
check ("ldexp (0.8, 4) == 12.8", FUNC(ldexp) (0.8L, 4), 12.8L);
|
|
check ("ldexp (-0.854375, 5) == -27.34", FUNC(ldexp) (-0.854375L, 5), -27.34L);
|
|
|
|
x = random_greater (0.0);
|
|
check_ext ("ldexp (x, 0) == x", FUNC(ldexp) (x, 0L), x, x);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
log_test (void)
|
|
{
|
|
check_isinfn_exc ("log (+0) == -inf plus divide-by-zero exception",
|
|
FUNC(log) (0), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfn_exc ("log (-0) == -inf plus divide-by-zero exception",
|
|
FUNC(log) (minus_zero), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check ("log (1) == 0", FUNC(log) (1), 0);
|
|
|
|
check_isnan_exc ("log (x) == NaN plus invalid exception if x < 0",
|
|
FUNC(log) (-1), INVALID_EXCEPTION);
|
|
check_isinfp ("log (+inf) == +inf", FUNC(log) (plus_infty));
|
|
|
|
check_eps ("log (e) == 1", FUNC(log) (M_El), 1, CHOOSE (1e-18L, 0, 9e-8L));
|
|
check_eps ("log (1/e) == -1", FUNC(log) (1.0 / M_El), -1,
|
|
CHOOSE (2e-18L, 0, 0));
|
|
check ("log (2) == M_LN2", FUNC(log) (2), M_LN2l);
|
|
check_eps ("log (10) == M_LN10", FUNC(log) (10), M_LN10l,
|
|
CHOOSE (1e-18L, 0, 0));
|
|
check_eps ("log (0.7) == -0.3566749439...", FUNC(log) (0.7),
|
|
-0.35667494393873237891L, CHOOSE(7e-17L, 6e-17, 3e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
log10_test (void)
|
|
{
|
|
check_isinfn_exc ("log10 (+0) == -inf plus divide-by-zero exception",
|
|
FUNC(log10) (0), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfn_exc ("log10 (-0) == -inf plus divide-by-zero exception",
|
|
FUNC(log10) (minus_zero), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check ("log10 (1) == +0", FUNC(log10) (1), 0);
|
|
|
|
check_isnan_exc ("log10 (x) == NaN plus invalid exception if x < 0",
|
|
FUNC(log10) (-1), INVALID_EXCEPTION);
|
|
|
|
check_isinfp ("log10 (+inf) == +inf", FUNC(log10) (plus_infty));
|
|
|
|
check_eps ("log10 (0.1) == -1", FUNC(log10) (0.1L), -1,
|
|
CHOOSE (1e-18L, 0, 0));
|
|
check_eps ("log10 (10) == 1", FUNC(log10) (10.0), 1,
|
|
CHOOSE (1e-18L, 0, 0));
|
|
check_eps ("log10 (100) == 2", FUNC(log10) (100.0), 2,
|
|
CHOOSE (1e-18L, 0, 0));
|
|
check ("log10 (10000) == 4", FUNC(log10) (10000.0), 4);
|
|
check_eps ("log10 (e) == M_LOG10E", FUNC(log10) (M_El), M_LOG10El,
|
|
CHOOSE (1e-18, 0, 9e-8));
|
|
check_eps ("log10 (0.7) == -0.1549019599...", FUNC(log10) (0.7),
|
|
-0.15490195998574316929L, CHOOSE(3e-17L, 3e-17, 2e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
log1p_test (void)
|
|
{
|
|
check ("log1p (+0) == +0", FUNC(log1p) (0), 0);
|
|
check ("log1p (-0) == -0", FUNC(log1p) (minus_zero), minus_zero);
|
|
|
|
check_isinfn_exc ("log1p (-1) == -inf plus divide-by-zero exception",
|
|
FUNC(log1p) (-1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isnan_exc ("log1p (x) == NaN plus invalid exception if x < -1",
|
|
FUNC(log1p) (-2), INVALID_EXCEPTION);
|
|
|
|
check_isinfp ("log1p (+inf) == +inf", FUNC(log1p) (plus_infty));
|
|
|
|
check_eps ("log1p (e-1) == 1", FUNC(log1p) (M_El - 1.0), 1,
|
|
CHOOSE (1e-18L, 0, 6e-8));
|
|
|
|
check_eps ("log1p (-0.3) == -0.35667...", FUNC(log1p) (-0.3),
|
|
-0.35667494393873237891L, CHOOSE(2e-17L, 6e-17, 3e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
log2_test (void)
|
|
{
|
|
check_isinfn_exc ("log2 (+0) == -inf plus divide-by-zero exception",
|
|
FUNC(log2) (0), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfn_exc ("log2 (-0) == -inf plus divide-by-zero exception",
|
|
FUNC(log2) (minus_zero), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check ("log2 (1) == +0", FUNC(log2) (1), 0);
|
|
|
|
check_isnan_exc ("log2 (x) == NaN plus invalid exception if x < 0",
|
|
FUNC(log2) (-1), INVALID_EXCEPTION);
|
|
|
|
check_isinfp ("log2 (+inf) == +inf", FUNC(log2) (plus_infty));
|
|
|
|
check_eps ("log2 (e) == M_LOG2E", FUNC(log2) (M_El), M_LOG2El,
|
|
CHOOSE (1e-18L, 0, 0));
|
|
check ("log2 (2) == 1", FUNC(log2) (2.0), 1);
|
|
check_eps ("log2 (16) == 4", FUNC(log2) (16.0), 4, CHOOSE (1e-18L, 0, 0));
|
|
check ("log2 (256) == 8", FUNC(log2) (256.0), 8);
|
|
check_eps ("log2 (0.7) == -0.5145731728...", FUNC(log2) (0.7),
|
|
-0.51457317282975824043L, CHOOSE(1e-16L, 2e-16, 6e-8));
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
logb_test (void)
|
|
{
|
|
check_isinfp ("logb (+inf) == +inf", FUNC(logb) (plus_infty));
|
|
check_isinfp ("logb (-inf) == +inf", FUNC(logb) (minus_infty));
|
|
|
|
check_isinfn_exc ("logb (+0) == -inf plus divide-by-zero exception",
|
|
FUNC(logb) (0), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check_isinfn_exc ("logb (-0) == -inf plus divide-by-zero exception",
|
|
FUNC(logb) (minus_zero), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check ("logb (1) == 0", FUNC(logb) (1), 0);
|
|
check ("logb (e) == 1", FUNC(logb) (M_El), 1);
|
|
check ("logb (1024) == 10", FUNC(logb) (1024), 10);
|
|
check ("logb (-2000) == 10", FUNC(logb) (-2000), 10);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
modf_test (void)
|
|
{
|
|
MATHTYPE result, intpart;
|
|
|
|
result = FUNC(modf) (plus_infty, &intpart);
|
|
check ("modf (+inf, &x) returns +0", result, 0);
|
|
check_isinfp ("modf (+inf, &x) set x to +inf", intpart);
|
|
|
|
result = FUNC(modf) (minus_infty, &intpart);
|
|
check ("modf (-inf, &x) returns -0", result, minus_zero);
|
|
check_isinfn ("modf (-inf, &x) sets x to -inf", intpart);
|
|
|
|
result = FUNC(modf) (nan_value, &intpart);
|
|
check_isnan ("modf (NaN, &x) returns NaN", result);
|
|
check_isnan ("modf (NaN, &x) sets x to NaN", intpart);
|
|
|
|
result = FUNC(modf) (0, &intpart);
|
|
check ("modf (0, &x) returns 0", result, 0);
|
|
check ("modf (0, &x) sets x to 0", intpart, 0);
|
|
|
|
result = FUNC(modf) (minus_zero, &intpart);
|
|
check ("modf (-0, &x) returns -0", result, minus_zero);
|
|
check ("modf (-0, &x) sets x to -0", intpart, minus_zero);
|
|
|
|
result = FUNC(modf) (1.5, &intpart);
|
|
check ("modf (1.5, &x) returns 0.5", result, 0.5);
|
|
check ("modf (1.5, &x) sets x to 1", intpart, 1);
|
|
|
|
result = FUNC(modf) (2.5, &intpart);
|
|
check ("modf (2.5, &x) returns 0.5", result, 0.5);
|
|
check ("modf (2.5, &x) sets x to 2", intpart, 2);
|
|
|
|
result = FUNC(modf) (-2.5, &intpart);
|
|
check ("modf (-2.5, &x) returns -0.5", result, -0.5);
|
|
check ("modf (-2.5, &x) sets x to -2", intpart, -2);
|
|
|
|
result = FUNC(modf) (20, &intpart);
|
|
check ("modf (20, &x) returns 0", result, 0);
|
|
check ("modf (20, &x) sets x to 20", intpart, 20);
|
|
|
|
result = FUNC(modf) (21, &intpart);
|
|
check ("modf (21, &x) returns 0", result, 0);
|
|
check ("modf (21, &x) sets x to 21", intpart, 21);
|
|
|
|
result = FUNC(modf) (89.6, &intpart);
|
|
check_eps ("modf (89.6, &x) returns 0.6", result, 0.6,
|
|
CHOOSE(6e-15L, 6e-15, 2e-6));
|
|
check ("modf (89.6, &x) sets x to 89", intpart, 89);
|
|
}
|
|
|
|
|
|
static void
|
|
scalb_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
check_isnan ("scalb (2, 0.5) == NaN", FUNC(scalb) (2, 0.5));
|
|
check_isnan ("scalb (3, -2.5) == NaN", FUNC(scalb) (3, -2.5));
|
|
|
|
check_isnan ("scalb (0, NaN) == NaN", FUNC(scalb) (0, nan_value));
|
|
check_isnan ("scalb (1, NaN) == NaN", FUNC(scalb) (1, nan_value));
|
|
|
|
x = random_greater (0.0);
|
|
check ("scalb (x, 0) == 0", FUNC(scalb) (x, 0), x);
|
|
x = random_greater (0.0);
|
|
check ("scalb (-x, 0) == 0", FUNC(scalb) (-x, 0), -x);
|
|
|
|
check_isnan_exc ("scalb (+0, +inf) == NaN plus invalid exception",
|
|
FUNC(scalb) (0, plus_infty), INVALID_EXCEPTION);
|
|
check_isnan_exc ("scalb (-0, +inf) == NaN plus invalid exception",
|
|
FUNC(scalb) (minus_zero, plus_infty), INVALID_EXCEPTION);
|
|
|
|
check ("scalb (+0, 2) == +0", FUNC(scalb) (0, 2), 0);
|
|
check ("scalb (-0, 4) == -0", FUNC(scalb) (minus_zero, -4), minus_zero);
|
|
check ("scalb (+0, 0) == +0", FUNC(scalb) (0, 0), 0);
|
|
check ("scalb (-0, 0) == -0", FUNC(scalb) (minus_zero, 0), minus_zero);
|
|
check ("scalb (+0, -1) == +0", FUNC(scalb) (0, -1), 0);
|
|
check ("scalb (-0, -10) == -0", FUNC(scalb) (minus_zero, -10), minus_zero);
|
|
check ("scalb (+0, -inf) == +0", FUNC(scalb) (0, minus_infty), 0);
|
|
check ("scalb (-0, -inf) == -0", FUNC(scalb) (minus_zero, minus_infty),
|
|
minus_zero);
|
|
|
|
check_isinfp ("scalb (+inf, -1) == +inf", FUNC(scalb) (plus_infty, -1));
|
|
check_isinfn ("scalb (-inf, -10) == -inf", FUNC(scalb) (minus_infty, -10));
|
|
check_isinfp ("scalb (+inf, 0) == +inf", FUNC(scalb) (plus_infty, 0));
|
|
check_isinfn ("scalb (-inf, 0) == -inf", FUNC(scalb) (minus_infty, 0));
|
|
check_isinfp ("scalb (+inf, 2) == +inf", FUNC(scalb) (plus_infty, 2));
|
|
check_isinfn ("scalb (-inf, 100) == -inf", FUNC(scalb) (minus_infty, 100));
|
|
|
|
x = random_greater (0.0);
|
|
check ("scalb (x, -inf) == 0", FUNC(scalb) (x, minus_infty), 0.0);
|
|
check ("scalb (-x, -inf) == -0", FUNC(scalb) (-x, minus_infty), minus_zero);
|
|
|
|
x = random_greater (0.0);
|
|
check_isinfp ("scalb (x, +inf) == +inf", FUNC(scalb) (x, plus_infty));
|
|
x = random_greater (0.0);
|
|
check_isinfn ("scalb (-x, +inf) == -inf", FUNC(scalb) (-x, plus_infty));
|
|
check_isinfp ("scalb (+inf, +inf) == +inf",
|
|
FUNC(scalb) (plus_infty, plus_infty));
|
|
check_isinfn ("scalb (-inf, +inf) == -inf",
|
|
FUNC(scalb) (minus_infty, plus_infty));
|
|
|
|
check_isnan ("scalb (+inf, -inf) == NaN",
|
|
FUNC(scalb) (plus_infty, minus_infty));
|
|
check_isnan ("scalb (-inf, -inf) == NaN",
|
|
FUNC(scalb) (minus_infty, minus_infty));
|
|
|
|
check_isnan ("scalb (NaN, 1) == NaN", FUNC(scalb) (nan_value, 1));
|
|
check_isnan ("scalb (1, NaN) == NaN", FUNC(scalb) (1, nan_value));
|
|
check_isnan ("scalb (NaN, 0) == NaN", FUNC(scalb) (nan_value, 0));
|
|
check_isnan ("scalb (0, NaN) == NaN", FUNC(scalb) (0, nan_value));
|
|
check_isnan ("scalb (NaN, +inf) == NaN",
|
|
FUNC(scalb) (nan_value, plus_infty));
|
|
check_isnan ("scalb (+inf, NaN) == NaN",
|
|
FUNC(scalb) (plus_infty, nan_value));
|
|
check_isnan ("scalb (NaN, NaN) == NaN", FUNC(scalb) (nan_value, nan_value));
|
|
|
|
check ("scalb (0.8, 4) == 12.8", FUNC(scalb) (0.8L, 4), 12.8L);
|
|
check ("scalb (-0.854375, 5) == -27.34", FUNC(scalb) (-0.854375L, 5), -27.34L);
|
|
}
|
|
|
|
|
|
static void
|
|
scalbn_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
check ("scalbn (0, 0) == 0", FUNC(scalbn) (0, 0), 0);
|
|
|
|
check_isinfp ("scalbn (+inf, 1) == +inf", FUNC(scalbn) (plus_infty, 1));
|
|
check_isinfn ("scalbn (-inf, 1) == -inf", FUNC(scalbn) (minus_infty, 1));
|
|
check_isnan ("scalbn (NaN, 1) == NaN", FUNC(scalbn) (nan_value, 1));
|
|
|
|
check ("scalbn (0.8, 4) == 12.8", FUNC(scalbn) (0.8L, 4), 12.8L);
|
|
check ("scalbn (-0.854375, 5) == -27.34", FUNC(scalbn) (-0.854375L, 5), -27.34L);
|
|
|
|
x = random_greater (0.0);
|
|
check_ext ("scalbn (x, 0) == x", FUNC(scalbn) (x, 0L), x, x);
|
|
}
|
|
|
|
|
|
static void
|
|
sin_test (void)
|
|
{
|
|
check ("sin (+0) == +0", FUNC(sin) (0), 0);
|
|
check ("sin (-0) == -0", FUNC(sin) (minus_zero), minus_zero);
|
|
check_isnan_exc ("sin (+inf) == NaN plus invalid exception",
|
|
FUNC(sin) (plus_infty),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_exc ("sin (-inf) == NaN plus invalid exception",
|
|
FUNC(sin) (minus_infty),
|
|
INVALID_EXCEPTION);
|
|
|
|
check_eps ("sin (pi/6) == 0.5", FUNC(sin) (M_PI_6l),
|
|
0.5, CHOOSE (4e-18L, 0, 0));
|
|
check_eps ("sin (-pi/6) == -0.5", FUNC(sin) (-M_PI_6l),
|
|
-0.5, CHOOSE (4e-18L, 0, 0));
|
|
check ("sin (pi/2) == 1", FUNC(sin) (M_PI_2l), 1);
|
|
check ("sin (-pi/2) == -1", FUNC(sin) (-M_PI_2l), -1);
|
|
check_eps ("sin (0.7) == 0.6442176872...", FUNC(sin) (0.7),
|
|
0.64421768723769105367L, CHOOSE(4e-17L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
sinh_test (void)
|
|
{
|
|
check ("sinh (+0) == +0", FUNC(sinh) (0), 0);
|
|
|
|
#ifndef TEST_INLINE
|
|
check ("sinh (-0) == -0", FUNC(sinh) (minus_zero), minus_zero);
|
|
|
|
check_isinfp ("sinh (+inf) == +inf", FUNC(sinh) (plus_infty));
|
|
check_isinfn ("sinh (-inf) == -inf", FUNC(sinh) (minus_infty));
|
|
#endif
|
|
|
|
check_eps ("sinh (0.7) == 0.7585837018...", FUNC(sinh) (0.7),
|
|
0.75858370183953350346L, CHOOSE(6e-17L, 2e-16, 6e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
sincos_test (void)
|
|
{
|
|
MATHTYPE sin_res, cos_res;
|
|
fenv_t fenv;
|
|
|
|
FUNC(sincos) (0, &sin_res, &cos_res);
|
|
fegetenv (&fenv);
|
|
check ("sincos (+0, &sin, &cos) puts +0 in sin", sin_res, 0);
|
|
fesetenv (&fenv);
|
|
check ("sincos (+0, &sin, &cos) puts 1 in cos", cos_res, 1);
|
|
|
|
FUNC(sincos) (minus_zero, &sin_res, &cos_res);
|
|
fegetenv (&fenv);
|
|
check ("sincos (-0, &sin, &cos) puts -0 in sin", sin_res, minus_zero);
|
|
fesetenv (&fenv);
|
|
check ("sincos (-0, &sin, &cos) puts 1 in cos", cos_res, 1);
|
|
|
|
FUNC(sincos) (plus_infty, &sin_res, &cos_res);
|
|
fegetenv (&fenv);
|
|
check_isnan_exc ("sincos (+inf, &sin, &cos) puts NaN in sin plus invalid exception",
|
|
sin_res, INVALID_EXCEPTION);
|
|
fesetenv (&fenv);
|
|
check_isnan_exc ("sincos (+inf, &sin, &cos) puts NaN in cos plus invalid exception",
|
|
cos_res, INVALID_EXCEPTION);
|
|
|
|
FUNC(sincos) (minus_infty, &sin_res, &cos_res);
|
|
fegetenv (&fenv);
|
|
check_isnan_exc ("sincos (-inf,&sin, &cos) puts NaN in sin plus invalid exception",
|
|
sin_res, INVALID_EXCEPTION);
|
|
fesetenv (&fenv);
|
|
check_isnan_exc ("sincos (-inf,&sin, &cos) puts NaN in cos plus invalid exception",
|
|
cos_res, INVALID_EXCEPTION);
|
|
|
|
FUNC(sincos) (M_PI_2l, &sin_res, &cos_res);
|
|
fegetenv (&fenv);
|
|
check ("sincos (pi/2, &sin, &cos) puts 1 in sin", sin_res, 1);
|
|
fesetenv (&fenv);
|
|
check_eps ("sincos (pi/2, &sin, &cos) puts 0 in cos", cos_res, 0,
|
|
CHOOSE (1e-18L, 1e-16, 1e-7));
|
|
|
|
FUNC(sincos) (M_PI_6l, &sin_res, &cos_res);
|
|
check_eps ("sincos (pi/6, &sin, &cos) puts 0.5 in sin", sin_res, 0.5,
|
|
CHOOSE (5e-18L, 0, 0));
|
|
|
|
FUNC(sincos) (M_PI_6l*2.0, &sin_res, &cos_res);
|
|
check_eps ("sincos (pi/3, &sin, &cos) puts 0.5 in cos", cos_res, 0.5,
|
|
CHOOSE (5e-18L, 1e-15, 1e-7));
|
|
|
|
FUNC(sincos) (0.7, &sin_res, &cos_res);
|
|
check_eps ("sincos (0.7, &sin, &cos) puts 0.6442176872... in sin", sin_res,
|
|
0.64421768723769105367L, CHOOSE(4e-17L, 0, 0));
|
|
check_eps ("sincos (0.7, &sin, &cos) puts 0.7648421872... in cos", cos_res,
|
|
0.76484218728448842626L, CHOOSE(3e-17L, 2e-16, 6e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
tan_test (void)
|
|
{
|
|
check ("tan (+0) == +0", FUNC(tan) (0), 0);
|
|
check ("tan (-0) == -0", FUNC(tan) (minus_zero), minus_zero);
|
|
check_isnan_exc ("tan (+inf) == NaN plus invalid exception",
|
|
FUNC(tan) (plus_infty), INVALID_EXCEPTION);
|
|
check_isnan_exc ("tan (-inf) == NaN plus invalid exception",
|
|
FUNC(tan) (minus_infty), INVALID_EXCEPTION);
|
|
|
|
check_eps ("tan (pi/4) == 1", FUNC(tan) (M_PI_4l), 1,
|
|
CHOOSE (2e-18L, 1e-15L, 2e-7));
|
|
check_eps ("tan (0.7) == 0.8422883804...", FUNC(tan) (0.7),
|
|
0.84228838046307944813L, CHOOSE(8e-17L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
tanh_test (void)
|
|
{
|
|
check ("tanh (+0) == +0", FUNC(tanh) (0), 0);
|
|
#ifndef TEST_INLINE
|
|
check ("tanh (-0) == -0", FUNC(tanh) (minus_zero), minus_zero);
|
|
|
|
check ("tanh (+inf) == +1", FUNC(tanh) (plus_infty), 1);
|
|
check ("tanh (-inf) == -1", FUNC(tanh) (minus_infty), -1);
|
|
#endif
|
|
check_eps ("tanh (0.7) == 0.6043677771...", FUNC(tanh) (0.7),
|
|
0.60436777711716349631L, CHOOSE(3e-17L, 2e-16, 6e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
fabs_test (void)
|
|
{
|
|
check ("fabs (+0) == +0", FUNC(fabs) (0), 0);
|
|
check ("fabs (-0) == +0", FUNC(fabs) (minus_zero), 0);
|
|
|
|
check_isinfp ("fabs (+inf) == +inf", FUNC(fabs) (plus_infty));
|
|
check_isinfp ("fabs (-inf) == +inf", FUNC(fabs) (minus_infty));
|
|
|
|
check ("fabs (+38) == 38", FUNC(fabs) (38.0), 38.0);
|
|
check ("fabs (-e) == e", FUNC(fabs) (-M_El), M_El);
|
|
}
|
|
|
|
|
|
static void
|
|
floor_test (void)
|
|
{
|
|
check ("floor (+0) == +0", FUNC(floor) (0.0), 0.0);
|
|
check ("floor (-0) == -0", FUNC(floor) (minus_zero), minus_zero);
|
|
check_isinfp ("floor (+inf) == +inf", FUNC(floor) (plus_infty));
|
|
check_isinfn ("floor (-inf) == -inf", FUNC(floor) (minus_infty));
|
|
|
|
check ("floor (pi) == 3", FUNC(floor) (M_PIl), 3.0);
|
|
check ("floor (-pi) == -4", FUNC(floor) (-M_PIl), -4.0);
|
|
}
|
|
|
|
|
|
static void
|
|
hypot_test (void)
|
|
{
|
|
MATHTYPE a;
|
|
|
|
a = random_greater (0);
|
|
check_isinfp_ext ("hypot (+inf, x) == +inf", FUNC(hypot) (plus_infty, a), a);
|
|
check_isinfp_ext ("hypot (-inf, x) == +inf", FUNC(hypot) (minus_infty, a), a);
|
|
|
|
#ifndef TEST_INLINE
|
|
check_isinfp ("hypot (+inf, NaN) == +inf", FUNC(hypot) (plus_infty, nan_value));
|
|
check_isinfp ("hypot (-inf, NaN) == +inf", FUNC(hypot) (minus_infty, nan_value));
|
|
#endif
|
|
|
|
check_isnan ("hypot (NaN, NaN) == NaN", FUNC(hypot) (nan_value, nan_value));
|
|
|
|
a = FUNC(hypot) (12.4L, 0.7L);
|
|
check ("hypot (x,y) == hypot (y,x)", FUNC(hypot) (0.7L, 12.4L), a);
|
|
check ("hypot (x,y) == hypot (-x,y)", FUNC(hypot) (-12.4L, 0.7L), a);
|
|
check ("hypot (x,y) == hypot (-y,x)", FUNC(hypot) (-0.7L, 12.4L), a);
|
|
check ("hypot (x,y) == hypot (-x,-y)", FUNC(hypot) (-12.4L, -0.7L), a);
|
|
check ("hypot (x,y) == hypot (-y,-x)", FUNC(hypot) (-0.7L, -12.4L), a);
|
|
check ("hypot (x,0) == fabs (x)", FUNC(hypot) (-0.7L, 0), 0.7L);
|
|
check ("hypot (x,0) == fabs (x)", FUNC(hypot) (0.7L, 0), 0.7L);
|
|
check ("hypot (x,0) == fabs (x)", FUNC(hypot) (-1.0L, 0), 1.0L);
|
|
check ("hypot (x,0) == fabs (x)", FUNC(hypot) (1.0L, 0), 1.0L);
|
|
check ("hypot (x,0) == fabs (x)", FUNC(hypot) (-5.7e7L, 0), 5.7e7L);
|
|
check ("hypot (x,0) == fabs (x)", FUNC(hypot) (5.7e7L, 0), 5.7e7L);
|
|
|
|
check_eps ("hypot (0.7,1.2) == 1.38924...", FUNC(hypot) (0.7, 1.2),
|
|
1.3892443989449804508L, CHOOSE(7e-17L, 3e-16, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
pow_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
check ("pow (+0, +0) == 1", FUNC(pow) (0, 0), 1);
|
|
check ("pow (+0, -0) == 1", FUNC(pow) (0, minus_zero), 1);
|
|
check ("pow (-0, +0) == 1", FUNC(pow) (minus_zero, 0), 1);
|
|
check ("pow (-0, -0) == 1", FUNC(pow) (minus_zero, minus_zero), 1);
|
|
|
|
check ("pow (+10, +0) == 1", FUNC(pow) (10, 0), 1);
|
|
check ("pow (+10, -0) == 1", FUNC(pow) (10, minus_zero), 1);
|
|
check ("pow (-10, +0) == 1", FUNC(pow) (-10, 0), 1);
|
|
check ("pow (-10, -0) == 1", FUNC(pow) (-10, minus_zero), 1);
|
|
|
|
check ("pow (NaN, +0) == 1", FUNC(pow) (nan_value, 0), 1);
|
|
check ("pow (NaN, -0) == 1", FUNC(pow) (nan_value, minus_zero), 1);
|
|
|
|
#ifndef TEST_INLINE
|
|
check_isinfp ("pow (+1.1, +inf) == +inf", FUNC(pow) (1.1, plus_infty));
|
|
check_isinfp ("pow (+inf, +inf) == +inf", FUNC(pow) (plus_infty, plus_infty));
|
|
check_isinfp ("pow (-1.1, +inf) == +inf", FUNC(pow) (-1.1, plus_infty));
|
|
check_isinfp ("pow (-inf, +inf) == +inf", FUNC(pow) (minus_infty, plus_infty));
|
|
|
|
check ("pow (0.9, +inf) == +0", FUNC(pow) (0.9L, plus_infty), 0);
|
|
check ("pow (1e-7, +inf) == +0", FUNC(pow) (1e-7L, plus_infty), 0);
|
|
check ("pow (-0.9, +inf) == +0", FUNC(pow) (-0.9L, plus_infty), 0);
|
|
check ("pow (-1e-7, +inf) == +0", FUNC(pow) (-1e-7L, plus_infty), 0);
|
|
|
|
check ("pow (+1.1, -inf) == 0", FUNC(pow) (1.1, minus_infty), 0);
|
|
check ("pow (+inf, -inf) == 0", FUNC(pow) (plus_infty, minus_infty), 0);
|
|
check ("pow (-1.1, -inf) == 0", FUNC(pow) (-1.1, minus_infty), 0);
|
|
check ("pow (-inf, -inf) == 0", FUNC(pow) (minus_infty, minus_infty), 0);
|
|
|
|
check_isinfp ("pow (0.9, -inf) == +inf", FUNC(pow) (0.9L, minus_infty));
|
|
check_isinfp ("pow (1e-7, -inf) == +inf", FUNC(pow) (1e-7L, minus_infty));
|
|
check_isinfp ("pow (-0.9, -inf) == +inf", FUNC(pow) (-0.9L, minus_infty));
|
|
check_isinfp ("pow (-1e-7, -inf) == +inf", FUNC(pow) (-1e-7L, minus_infty));
|
|
|
|
check_isinfp ("pow (+inf, 1e-7) == +inf", FUNC(pow) (plus_infty, 1e-7L));
|
|
check_isinfp ("pow (+inf, 1) == +inf", FUNC(pow) (plus_infty, 1));
|
|
check_isinfp ("pow (+inf, 1e7) == +inf", FUNC(pow) (plus_infty, 1e7L));
|
|
|
|
check ("pow (+inf, -1e-7) == 0", FUNC(pow) (plus_infty, -1e-7L), 0);
|
|
check ("pow (+inf, -1) == 0", FUNC(pow) (plus_infty, -1), 0);
|
|
check ("pow (+inf, -1e7) == 0", FUNC(pow) (plus_infty, -1e7L), 0);
|
|
|
|
check_isinfn ("pow (-inf, 1) == -inf", FUNC(pow) (minus_infty, 1));
|
|
check_isinfn ("pow (-inf, 11) == -inf", FUNC(pow) (minus_infty, 11));
|
|
check_isinfn ("pow (-inf, 1001) == -inf", FUNC(pow) (minus_infty, 1001));
|
|
|
|
check_isinfp ("pow (-inf, 2) == +inf", FUNC(pow) (minus_infty, 2));
|
|
check_isinfp ("pow (-inf, 12) == +inf", FUNC(pow) (minus_infty, 12));
|
|
check_isinfp ("pow (-inf, 1002) == +inf", FUNC(pow) (minus_infty, 1002));
|
|
check_isinfp ("pow (-inf, 0.1) == +inf", FUNC(pow) (minus_infty, 0.1));
|
|
check_isinfp ("pow (-inf, 1.1) == +inf", FUNC(pow) (minus_infty, 1.1));
|
|
check_isinfp ("pow (-inf, 11.1) == +inf", FUNC(pow) (minus_infty, 11.1));
|
|
check_isinfp ("pow (-inf, 1001.1) == +inf", FUNC(pow) (minus_infty, 1001.1));
|
|
|
|
check ("pow (-inf, -1) == -0", FUNC(pow) (minus_infty, -1), minus_zero);
|
|
check ("pow (-inf, -11) == -0", FUNC(pow) (minus_infty, -11), minus_zero);
|
|
check ("pow (-inf, -1001) == -0", FUNC(pow) (minus_infty, -1001), minus_zero);
|
|
|
|
check ("pow (-inf, -2) == +0", FUNC(pow) (minus_infty, -2), 0);
|
|
check ("pow (-inf, -12) == +0", FUNC(pow) (minus_infty, -12), 0);
|
|
check ("pow (-inf, -1002) == +0", FUNC(pow) (minus_infty, -1002), 0);
|
|
check ("pow (-inf, -0.1) == +0", FUNC(pow) (minus_infty, -0.1), 0);
|
|
check ("pow (-inf, -1.1) == +0", FUNC(pow) (minus_infty, -1.1), 0);
|
|
check ("pow (-inf, -11.1) == +0", FUNC(pow) (minus_infty, -11.1), 0);
|
|
check ("pow (-inf, -1001.1) == +0", FUNC(pow) (minus_infty, -1001.1), 0);
|
|
|
|
check_isnan ("pow (NaN, NaN) == NaN", FUNC(pow) (nan_value, nan_value));
|
|
check_isnan ("pow (0, NaN) == NaN", FUNC(pow) (0, nan_value));
|
|
check_isnan ("pow (1, NaN) == NaN", FUNC(pow) (1, nan_value));
|
|
check_isnan ("pow (-1, NaN) == NaN", FUNC(pow) (-1, nan_value));
|
|
check_isnan ("pow (NaN, 1) == NaN", FUNC(pow) (nan_value, 1));
|
|
check_isnan ("pow (NaN, -1) == NaN", FUNC(pow) (nan_value, -1));
|
|
|
|
x = random_greater (0.0);
|
|
check_isnan_ext ("pow (x, NaN) == NaN", FUNC(pow) (x, nan_value), x);
|
|
|
|
check_isnan_exc ("pow (+1, +inf) == NaN plus invalid exception",
|
|
FUNC(pow) (1, plus_infty), INVALID_EXCEPTION);
|
|
check_isnan_exc ("pow (-1, +inf) == NaN plus invalid exception",
|
|
FUNC(pow) (-1, plus_infty), INVALID_EXCEPTION);
|
|
check_isnan_exc ("pow (+1, -inf) == NaN plus invalid exception",
|
|
FUNC(pow) (1, minus_infty), INVALID_EXCEPTION);
|
|
check_isnan_exc ("pow (-1, -inf) == NaN plus invalid exception",
|
|
FUNC(pow) (-1, minus_infty), INVALID_EXCEPTION);
|
|
|
|
check_isnan_exc ("pow (-0.1, 1.1) == NaN plus invalid exception",
|
|
FUNC(pow) (-0.1, 1.1), INVALID_EXCEPTION);
|
|
check_isnan_exc ("pow (-0.1, -1.1) == NaN plus invalid exception",
|
|
FUNC(pow) (-0.1, -1.1), INVALID_EXCEPTION);
|
|
check_isnan_exc ("pow (-10.1, 1.1) == NaN plus invalid exception",
|
|
FUNC(pow) (-10.1, 1.1), INVALID_EXCEPTION);
|
|
check_isnan_exc ("pow (-10.1, -1.1) == NaN plus invalid exception",
|
|
FUNC(pow) (-10.1, -1.1), INVALID_EXCEPTION);
|
|
|
|
check_isinfp_exc ("pow (+0, -1) == +inf plus divide-by-zero exception",
|
|
FUNC(pow) (0, -1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfp_exc ("pow (+0, -11) == +inf plus divide-by-zero exception",
|
|
FUNC(pow) (0, -11), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfn_exc ("pow (-0, -1) == -inf plus divide-by-zero exception",
|
|
FUNC(pow) (minus_zero, -1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfn_exc ("pow (-0, -11) == -inf plus divide-by-zero exception",
|
|
FUNC(pow) (minus_zero, -11), DIVIDE_BY_ZERO_EXCEPTION);
|
|
|
|
check_isinfp_exc ("pow (+0, -2) == +inf plus divide-by-zero exception",
|
|
FUNC(pow) (0, -2), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfp_exc ("pow (+0, -11.1) == +inf plus divide-by-zero exception",
|
|
FUNC(pow) (0, -11.1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfp_exc ("pow (-0, -2) == +inf plus divide-by-zero exception",
|
|
FUNC(pow) (minus_zero, -2), DIVIDE_BY_ZERO_EXCEPTION);
|
|
check_isinfp_exc ("pow (-0, -11.1) == +inf plus divide-by-zero exception",
|
|
FUNC(pow) (minus_zero, -11.1), DIVIDE_BY_ZERO_EXCEPTION);
|
|
#endif
|
|
|
|
check ("pow (+0, 1) == +0", FUNC(pow) (0, 1), 0);
|
|
check ("pow (+0, 11) == +0", FUNC(pow) (0, 11), 0);
|
|
#ifndef TEST_INLINE
|
|
check ("pow (-0, 1) == -0", FUNC(pow) (minus_zero, 1), minus_zero);
|
|
check ("pow (-0, 11) == -0", FUNC(pow) (minus_zero, 11), minus_zero);
|
|
#endif
|
|
|
|
check ("pow (+0, 2) == +0", FUNC(pow) (0, 2), 0);
|
|
check ("pow (+0, 11.1) == +0", FUNC(pow) (0, 11.1), 0);
|
|
|
|
#ifndef TEST_INLINE
|
|
check ("pow (-0, 2) == +0", FUNC(pow) (minus_zero, 2), 0);
|
|
check ("pow (-0, 11.1) == +0", FUNC(pow) (minus_zero, 11.1), 0);
|
|
|
|
x = random_greater (1.0);
|
|
check_isinfp_ext ("pow (x, +inf) == +inf for |x| > 1",
|
|
FUNC(pow) (x, plus_infty), x);
|
|
|
|
x = random_value (-1.0, 1.0);
|
|
check_ext ("pow (x, +inf) == +0 for |x| < 1",
|
|
FUNC(pow) (x, plus_infty), 0.0, x);
|
|
|
|
x = random_greater (1.0);
|
|
check_ext ("pow (x, -inf) == +0 for |x| > 1",
|
|
FUNC(pow) (x, minus_infty), 0.0, x);
|
|
|
|
x = random_value (-1.0, 1.0);
|
|
check_isinfp_ext ("pow (x, -inf) == +inf for |x| < 1",
|
|
FUNC(pow) (x, minus_infty), x);
|
|
|
|
x = random_greater (0.0);
|
|
check_isinfp_ext ("pow (+inf, y) == +inf for y > 0",
|
|
FUNC(pow) (plus_infty, x), x);
|
|
|
|
x = random_less (0.0);
|
|
check_ext ("pow (+inf, y) == +0 for y < 0",
|
|
FUNC(pow) (plus_infty, x), 0.0, x);
|
|
|
|
x = (rand () % 1000000) * 2.0 + 1; /* Get random odd integer > 0 */
|
|
check_isinfn_ext ("pow (-inf, y) == -inf for y an odd integer > 0",
|
|
FUNC(pow) (minus_infty, x), x);
|
|
|
|
x = ((rand () % 1000000) + 1) * 2.0; /* Get random even integer > 1 */
|
|
check_isinfp_ext ("pow (-inf, y) == +inf for y > 0 and not an odd integer",
|
|
FUNC(pow) (minus_infty, x), x);
|
|
|
|
x = -((rand () % 1000000) * 2.0 + 1); /* Get random odd integer < 0 */
|
|
check_ext ("pow (-inf, y) == -0 for y an odd integer < 0",
|
|
FUNC(pow) (minus_infty, x), minus_zero, x);
|
|
|
|
x = ((rand () % 1000000) + 1) * -2.0; /* Get random even integer < 0 */
|
|
check_ext ("pow (-inf, y) == +0 for y < 0 and not an odd integer",
|
|
FUNC(pow) (minus_infty, x), 0.0, x);
|
|
#endif
|
|
|
|
x = (rand () % 1000000) * 2.0 + 1; /* Get random odd integer > 0 */
|
|
check_ext ("pow (+0, y) == +0 for y an odd integer > 0",
|
|
FUNC(pow) (0.0, x), 0.0, x);
|
|
#ifndef TEST_INLINE
|
|
x = (rand () % 1000000) * 2.0 + 1; /* Get random odd integer > 0 */
|
|
check_ext ("pow (-0, y) == -0 for y an odd integer > 0",
|
|
FUNC(pow) (minus_zero, x), minus_zero, x);
|
|
#endif
|
|
|
|
x = ((rand () % 1000000) + 1) * 2.0; /* Get random even integer > 1 */
|
|
check_ext ("pow (+0, y) == +0 for y > 0 and not an odd integer",
|
|
FUNC(pow) (0.0, x), 0.0, x);
|
|
|
|
x = ((rand () % 1000000) + 1) * 2.0; /* Get random even integer > 1 */
|
|
check_ext ("pow (-0, y) == +0 for y > 0 and not an odd integer",
|
|
FUNC(pow) (minus_zero, x), 0.0, x);
|
|
|
|
check_eps ("pow (0.7, 1.2) == 0.65180...", FUNC(pow) (0.7, 1.2),
|
|
0.65180494056638638188L, CHOOSE(4e-17L, 0, 0));
|
|
|
|
#ifdef TEST_DOUBLE
|
|
check ("pow (-7.49321e+133, -9.80818e+16) == 0",
|
|
FUNC(pow) (-7.49321e+133, -9.80818e+16), 0);
|
|
#endif
|
|
}
|
|
|
|
|
|
static void
|
|
fdim_test (void)
|
|
{
|
|
check ("fdim (+0, +0) = +0", FUNC(fdim) (0, 0), 0);
|
|
check ("fdim (9, 0) = 9", FUNC(fdim) (9, 0), 9);
|
|
check ("fdim (0, 9) = 0", FUNC(fdim) (0, 9), 0);
|
|
check ("fdim (-9, 0) = 9", FUNC(fdim) (-9, 0), 0);
|
|
check ("fdim (0, -9) = 9", FUNC(fdim) (0, -9), 9);
|
|
|
|
check_isinfp ("fdim (+inf, 9) = +inf", FUNC(fdim) (plus_infty, 9));
|
|
check_isinfp ("fdim (+inf, -9) = +inf", FUNC(fdim) (plus_infty, -9));
|
|
check ("fdim (-inf, 9) = 0", FUNC(fdim) (minus_infty, 9), 0);
|
|
check ("fdim (-inf, -9) = 0", FUNC(fdim) (minus_infty, -9), 0);
|
|
check_isinfp ("fdim (+9, -inf) = +inf", FUNC(fdim) (9, minus_infty));
|
|
check_isinfp ("fdim (-9, -inf) = +inf", FUNC(fdim) (-9, minus_infty));
|
|
check ("fdim (9, inf) = 0", FUNC(fdim) (9, plus_infty), 0);
|
|
check ("fdim (-9, inf) = 0", FUNC(fdim) (-9, plus_infty), 0);
|
|
|
|
check_isnan ("fdim (0, NaN) = NaN", FUNC(fdim) (0, nan_value));
|
|
check_isnan ("fdim (9, NaN) = NaN", FUNC(fdim) (9, nan_value));
|
|
check_isnan ("fdim (-9, NaN) = NaN", FUNC(fdim) (-9, nan_value));
|
|
check_isnan ("fdim (NaN, 9) = NaN", FUNC(fdim) (nan_value, 9));
|
|
check_isnan ("fdim (NaN, -9) = NaN", FUNC(fdim) (nan_value, -9));
|
|
check_isnan ("fdim (+inf, NaN) = NaN", FUNC(fdim) (plus_infty, nan_value));
|
|
check_isnan ("fdim (-inf, NaN) = NaN", FUNC(fdim) (minus_infty, nan_value));
|
|
check_isnan ("fdim (NaN, +inf) = NaN", FUNC(fdim) (nan_value, plus_infty));
|
|
check_isnan ("fdim (NaN, -inf) = NaN", FUNC(fdim) (nan_value, minus_infty));
|
|
check_isnan ("fdim (NaN, NaN) = NaN", FUNC(fdim) (nan_value, nan_value));
|
|
}
|
|
|
|
|
|
static void
|
|
fmin_test (void)
|
|
{
|
|
check ("fmin (+0, +0) = +0", FUNC(fmin) (0, 0), 0);
|
|
check ("fmin (9, 0) = 0", FUNC(fmin) (9, 0), 0);
|
|
check ("fmin (0, 9) = 0", FUNC(fmin) (0, 9), 0);
|
|
check ("fmin (-9, 0) = -9", FUNC(fmin) (-9, 0), -9);
|
|
check ("fmin (0, -9) = -9", FUNC(fmin) (0, -9), -9);
|
|
|
|
check ("fmin (+inf, 9) = 9", FUNC(fmin) (plus_infty, 9), 9);
|
|
check ("fmin (9, +inf) = 9", FUNC(fmin) (9, plus_infty), 9);
|
|
check ("fmin (+inf, -9) = -9", FUNC(fmin) (plus_infty, -9), -9);
|
|
check ("fmin (-9, +inf) = -9", FUNC(fmin) (-9, plus_infty), -9);
|
|
check_isinfn ("fmin (-inf, 9) = -inf", FUNC(fmin) (minus_infty, 9));
|
|
check_isinfn ("fmin (-inf, -9) = -inf", FUNC(fmin) (minus_infty, -9));
|
|
check_isinfn ("fmin (+9, -inf) = -inf", FUNC(fmin) (9, minus_infty));
|
|
check_isinfn ("fmin (-9, -inf) = -inf", FUNC(fmin) (-9, minus_infty));
|
|
|
|
check ("fmin (0, NaN) = 0", FUNC(fmin) (0, nan_value), 0);
|
|
check ("fmin (9, NaN) = 9", FUNC(fmin) (9, nan_value), 9);
|
|
check ("fmin (-9, NaN) = 9", FUNC(fmin) (-9, nan_value), -9);
|
|
check ("fmin (NaN, 0) = 0", FUNC(fmin) (nan_value, 0), 0);
|
|
check ("fmin (NaN, 9) = NaN", FUNC(fmin) (nan_value, 9), 9);
|
|
check ("fmin (NaN, -9) = NaN", FUNC(fmin) (nan_value, -9), -9);
|
|
check_isinfp ("fmin (+inf, NaN) = +inf", FUNC(fmin) (plus_infty, nan_value));
|
|
check_isinfn ("fmin (-inf, NaN) = -inf", FUNC(fmin) (minus_infty, nan_value));
|
|
check_isinfp ("fmin (NaN, +inf) = +inf", FUNC(fmin) (nan_value, plus_infty));
|
|
check_isinfn ("fmin (NaN, -inf) = -inf", FUNC(fmin) (nan_value, minus_infty));
|
|
check_isnan ("fmin (NaN, NaN) = NaN", FUNC(fmin) (nan_value, nan_value));
|
|
}
|
|
|
|
|
|
static void
|
|
fmax_test (void)
|
|
{
|
|
check ("fmax (+0, +0) = +0", FUNC(fmax) (0, 0), 0);
|
|
check ("fmax (9, 0) = 9", FUNC(fmax) (9, 0), 9);
|
|
check ("fmax (0, 9) = 9", FUNC(fmax) (0, 9), 9);
|
|
check ("fmax (-9, 0) = 0", FUNC(fmax) (-9, 0), 0);
|
|
check ("fmax (0, -9) = 0", FUNC(fmax) (0, -9), 0);
|
|
|
|
check_isinfp ("fmax (+inf, 9) = +inf", FUNC(fmax) (plus_infty, 9));
|
|
check_isinfp ("fmax (9, +inf) = +inf", FUNC(fmax) (0, plus_infty));
|
|
check_isinfp ("fmax (-9, +inf) = +inf", FUNC(fmax) (-9, plus_infty));
|
|
check_isinfp ("fmax (+inf, -9) = +inf", FUNC(fmax) (plus_infty, -9));
|
|
check ("fmax (-inf, 9) = 9", FUNC(fmax) (minus_infty, 9), 9);
|
|
check ("fmax (-inf, -9) = -9", FUNC(fmax) (minus_infty, -9), -9);
|
|
check ("fmax (+9, -inf) = 9", FUNC(fmax) (9, minus_infty), 9);
|
|
check ("fmax (-9, -inf) = -9", FUNC(fmax) (-9, minus_infty), -9);
|
|
|
|
check ("fmax (0, NaN) = 0", FUNC(fmax) (0, nan_value), 0);
|
|
check ("fmax (9, NaN) = 9", FUNC(fmax) (9, nan_value), 9);
|
|
check ("fmax (-9, NaN) = 9", FUNC(fmax) (-9, nan_value), -9);
|
|
check ("fmax (NaN, 0) = 0", FUNC(fmax) (nan_value, 0), 0);
|
|
check ("fmax (NaN, 9) = NaN", FUNC(fmax) (nan_value, 9), 9);
|
|
check ("fmax (NaN, -9) = NaN", FUNC(fmax) (nan_value, -9), -9);
|
|
check_isinfp ("fmax (+inf, NaN) = +inf", FUNC(fmax) (plus_infty, nan_value));
|
|
check_isinfn ("fmax (-inf, NaN) = -inf", FUNC(fmax) (minus_infty, nan_value));
|
|
check_isinfp ("fmax (NaN, +inf) = +inf", FUNC(fmax) (nan_value, plus_infty));
|
|
check_isinfn ("fmax (NaN, -inf) = -inf", FUNC(fmax) (nan_value, minus_infty));
|
|
check_isnan ("fmax (NaN, NaN) = NaN", FUNC(fmax) (nan_value, nan_value));
|
|
}
|
|
|
|
|
|
static void
|
|
fmod_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
x = random_greater (0);
|
|
check_ext ("fmod (+0, y) == +0 for y != 0", FUNC(fmod) (0, x), 0, x);
|
|
|
|
x = random_greater (0);
|
|
check_ext ("fmod (-0, y) == -0 for y != 0", FUNC(fmod) (minus_zero, x),
|
|
minus_zero, x);
|
|
|
|
check_isnan_exc_ext ("fmod (+inf, y) == NaN plus invalid exception",
|
|
FUNC(fmod) (plus_infty, x), INVALID_EXCEPTION, x);
|
|
check_isnan_exc_ext ("fmod (-inf, y) == NaN plus invalid exception",
|
|
FUNC(fmod) (minus_infty, x), INVALID_EXCEPTION, x);
|
|
check_isnan_exc_ext ("fmod (x, +0) == NaN plus invalid exception",
|
|
FUNC(fmod) (x, 0), INVALID_EXCEPTION, x);
|
|
check_isnan_exc_ext ("fmod (x, -0) == NaN plus invalid exception",
|
|
FUNC(fmod) (x, minus_zero), INVALID_EXCEPTION, x);
|
|
|
|
x = random_greater (0);
|
|
check_ext ("fmod (x, +inf) == x for x not infinite",
|
|
FUNC(fmod) (x, plus_infty), x, x);
|
|
x = random_greater (0);
|
|
check_ext ("fmod (x, -inf) == x for x not infinite",
|
|
FUNC(fmod) (x, minus_infty), x, x);
|
|
|
|
check_eps ("fmod (6.5, 2.3) == 1.9", FUNC(fmod) (6.5, 2.3), 1.9,
|
|
CHOOSE(5e-16, 1e-15, 2e-7));
|
|
check_eps ("fmod (-6.5, 2.3) == -1.9", FUNC(fmod) (-6.5, 2.3), -1.9,
|
|
CHOOSE(5e-16, 1e-15, 2e-7));
|
|
check_eps ("fmod (6.5, -2.3) == 1.9", FUNC(fmod) (6.5, -2.3), 1.9,
|
|
CHOOSE(5e-16, 1e-15, 2e-7));
|
|
check_eps ("fmod (-6.5, -2.3) == -1.9", FUNC(fmod) (-6.5, -2.3), -1.9,
|
|
CHOOSE(5e-16, 1e-15, 2e-7));
|
|
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
nextafter_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
check ("nextafter (+0, +0) = +0", FUNC(nextafter) (0, 0), 0);
|
|
check ("nextafter (-0, +0) = +0", FUNC(nextafter) (minus_zero, 0), 0);
|
|
check ("nextafter (+0, -0) = -0", FUNC(nextafter) (0, minus_zero),
|
|
minus_zero);
|
|
check ("nextafter (-0, -0) = -0", FUNC(nextafter) (minus_zero, minus_zero),
|
|
minus_zero);
|
|
|
|
check ("nextafter (9, 9) = 9", FUNC(nextafter) (9, 9), 9);
|
|
check ("nextafter (-9, -9) = -9", FUNC(nextafter) (-9, -9), -9);
|
|
check_isinfp ("nextafter (+inf, +inf) = +inf",
|
|
FUNC(nextafter) (plus_infty, plus_infty));
|
|
check_isinfn ("nextafter (-inf, -inf) = -inf",
|
|
FUNC(nextafter) (minus_infty, minus_infty));
|
|
|
|
x = rand () * 1.1;
|
|
check_isnan ("nextafter (NaN, x) = NaN", FUNC(nextafter) (nan_value, x));
|
|
check_isnan ("nextafter (x, NaN) = NaN", FUNC(nextafter) (x, nan_value));
|
|
check_isnan ("nextafter (NaN, NaN) = NaN", FUNC(nextafter) (nan_value,
|
|
nan_value));
|
|
|
|
/* XXX We need the hexadecimal FP number representation here for further
|
|
tests. */
|
|
}
|
|
|
|
|
|
static void
|
|
copysign_test (void)
|
|
{
|
|
check ("copysign (0, 4) = 0", FUNC(copysign) (0, 4), 0);
|
|
check ("copysign (0, -4) = -0", FUNC(copysign) (0, -4), minus_zero);
|
|
check ("copysign (-0, 4) = 0", FUNC(copysign) (minus_zero, 4), 0);
|
|
check ("copysign (-0, -4) = -0", FUNC(copysign) (minus_zero, -4),
|
|
minus_zero);
|
|
|
|
check_isinfp ("copysign (+inf, 0) = +inf", FUNC(copysign) (plus_infty, 0));
|
|
check_isinfn ("copysign (+inf, -0) = -inf", FUNC(copysign) (plus_infty,
|
|
minus_zero));
|
|
check_isinfp ("copysign (-inf, 0) = +inf", FUNC(copysign) (minus_infty, 0));
|
|
check_isinfn ("copysign (-inf, -0) = -inf", FUNC(copysign) (minus_infty,
|
|
minus_zero));
|
|
|
|
check ("copysign (0, +inf) = 0", FUNC(copysign) (0, plus_infty), 0);
|
|
check ("copysign (0, -inf) = -0", FUNC(copysign) (0, minus_zero),
|
|
minus_zero);
|
|
check ("copysign (-0, +inf) = 0", FUNC(copysign) (minus_zero, plus_infty),
|
|
0);
|
|
check ("copysign (-0, -inf) = -0", FUNC(copysign) (minus_zero, minus_zero),
|
|
minus_zero);
|
|
|
|
/* XXX More correctly we would have to check the sign of the NaN. */
|
|
check_isnan ("copysign (+NaN, 0) = +NaN", FUNC(copysign) (nan_value, 0));
|
|
check_isnan ("copysign (+NaN, -0) = -NaN", FUNC(copysign) (nan_value,
|
|
minus_zero));
|
|
check_isnan ("copysign (-NaN, 0) = +NaN", FUNC(copysign) (-nan_value, 0));
|
|
check_isnan ("copysign (-NaN, -0) = -NaN", FUNC(copysign) (-nan_value,
|
|
minus_zero));
|
|
}
|
|
|
|
|
|
static void
|
|
trunc_test (void)
|
|
{
|
|
check ("trunc(0) = 0", FUNC(trunc) (0), 0);
|
|
check ("trunc(-0) = -0", FUNC(trunc) (minus_zero), minus_zero);
|
|
check ("trunc(0.625) = 0", FUNC(trunc) (0.625), 0);
|
|
check ("trunc(-0.625) = -0", FUNC(trunc) (-0.625), minus_zero);
|
|
check ("trunc(1) = 1", FUNC(trunc) (1), 1);
|
|
check ("trunc(-1) = -1", FUNC(trunc) (-1), -1);
|
|
check ("trunc(1.625) = 1", FUNC(trunc) (1.625), 1);
|
|
check ("trunc(-1.625) = -1", FUNC(trunc) (-1.625), -1);
|
|
|
|
check ("trunc(1048580.625) = 1048580", FUNC(trunc) (1048580.625L),
|
|
1048580L);
|
|
check ("trunc(-1048580.625) = -1048580", FUNC(trunc) (-1048580.625L),
|
|
-1048580L);
|
|
|
|
check ("trunc(8388610.125) = 8388610", FUNC(trunc) (8388610.125L),
|
|
8388610.0L);
|
|
check ("trunc(-8388610.125) = -8388610", FUNC(trunc) (-8388610.125L),
|
|
-8388610.0L);
|
|
|
|
check ("trunc(4294967296.625) = 4294967296", FUNC(trunc) (4294967296.625L),
|
|
4294967296.0L);
|
|
check ("trunc(-4294967296.625) = -4294967296",
|
|
FUNC(trunc) (-4294967296.625L), -4294967296.0L);
|
|
|
|
check_isinfp ("trunc(+inf) = +inf", FUNC(trunc) (plus_infty));
|
|
check_isinfn ("trunc(-inf) = -inf", FUNC(trunc) (minus_infty));
|
|
check_isnan ("trunc(NaN) = NaN", FUNC(trunc) (nan_value));
|
|
}
|
|
|
|
|
|
static void
|
|
sqrt_test (void)
|
|
{
|
|
MATHTYPE x;
|
|
|
|
|
|
/* XXX Tests fuer negative x are missing */
|
|
check ("sqrt (0) == 0", FUNC(sqrt) (0), 0);
|
|
check_isnan ("sqrt (NaN) == NaN", FUNC(sqrt) (nan_value));
|
|
check_isinfp ("sqrt (+inf) == +inf", FUNC(sqrt) (plus_infty));
|
|
|
|
check ("sqrt (-0) == -0", FUNC(sqrt) (0), 0);
|
|
|
|
x = random_less (0.0);
|
|
check_isnan_exc_ext ("sqrt (x) == NaN plus invalid exception for x < 0",
|
|
FUNC(sqrt) (x), INVALID_EXCEPTION, x);
|
|
|
|
x = random_value (0, 10000);
|
|
check_ext ("sqrt (x*x) == x", FUNC(sqrt) (x*x), x, x);
|
|
check ("sqrt (4) == 2", FUNC(sqrt) (4), 2);
|
|
check ("sqrt (2) == 1.14142...", FUNC(sqrt) (2), M_SQRT2l);
|
|
check ("sqrt (0.25) == 0.5", FUNC(sqrt) (0.25), 0.5);
|
|
check ("sqrt (6642.25) == 81.5", FUNC(sqrt) (6642.25), 81.5);
|
|
check_eps ("sqrt (15239.903) == 123.45", FUNC(sqrt) (15239.903), 123.45,
|
|
CHOOSE (3e-6L, 3e-6, 8e-6));
|
|
check_eps ("sqrt (0.7) == 0.8366600265", FUNC(sqrt) (0.7),
|
|
0.83666002653407554798L, CHOOSE(3e-17L, 0, 0));
|
|
}
|
|
|
|
static void
|
|
remainder_test (void)
|
|
{
|
|
MATHTYPE result;
|
|
|
|
result = FUNC(remainder) (1, 0);
|
|
check_isnan_exc ("remainder(1, +0) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remainder) (1, minus_zero);
|
|
check_isnan_exc ("remainder(1, -0) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remainder) (plus_infty, 1);
|
|
check_isnan_exc ("remainder(+inf, 1) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remainder) (minus_infty, 1);
|
|
check_isnan_exc ("remainder(-inf, 1) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remainder) (1.625, 1.0);
|
|
check ("remainder(1.625, 1.0) == -0.375", result, -0.375);
|
|
|
|
result = FUNC(remainder) (-1.625, 1.0);
|
|
check ("remainder(-1.625, 1.0) == 0.375", result, 0.375);
|
|
|
|
result = FUNC(remainder) (1.625, -1.0);
|
|
check ("remainder(1.625, -1.0) == -0.375", result, -0.375);
|
|
|
|
result = FUNC(remainder) (-1.625, -1.0);
|
|
check ("remainder(-1.625, -1.0) == 0.375", result, 0.375);
|
|
|
|
result = FUNC(remainder) (5.0, 2.0);
|
|
check ("remainder(5.0, 2.0) == 1.0", result, 1.0);
|
|
|
|
result = FUNC(remainder) (3.0, 2.0);
|
|
check ("remainder(3.0, 2.0) == -1.0", result, -1.0);
|
|
}
|
|
|
|
|
|
static void
|
|
remquo_test (void)
|
|
{
|
|
int quo;
|
|
MATHTYPE result;
|
|
|
|
result = FUNC(remquo) (1, 0, &quo);
|
|
check_isnan_exc ("remquo(1, +0, &x) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remquo) (1, minus_zero, &quo);
|
|
check_isnan_exc ("remquo(1, -0, &x) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remquo) (plus_infty, 1, &quo);
|
|
check_isnan_exc ("remquo(+inf, 1, &x) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remquo) (minus_infty, 1, &quo);
|
|
check_isnan_exc ("remquo(-inf, 1, &x) == NaN plus invalid exception",
|
|
result, INVALID_EXCEPTION);
|
|
|
|
result = FUNC(remquo) (1.625, 1.0, &quo);
|
|
check ("remquo(1.625, 1.0, &x) == -0.375", result, -0.375);
|
|
check_long ("remquo(1.625, 1.0, &x) puts 2 in x", quo, 2);
|
|
|
|
result = FUNC(remquo) (-1.625, 1.0, &quo);
|
|
check ("remquo(-1.625, 1.0, &x) == 0.375", result, 0.375);
|
|
check_long ("remquo(-1.625, 1.0, &x) puts -2 in x", quo, -2);
|
|
|
|
result = FUNC(remquo) (1.625, -1.0, &quo);
|
|
check ("remquo(1.625, -1.0, &x) == -0.375", result, -0.375);
|
|
check_long ("remquo(1.625, -1.0, &x) puts -2 in x", quo, -2);
|
|
|
|
result = FUNC(remquo) (-1.625, -1.0, &quo);
|
|
check ("remquo(-1.625, -1.0, &x) == 0.375", result, 0.375);
|
|
check_long ("remquo(-1.625, -1.0, &x) puts 2 in x", quo, 2);
|
|
|
|
result = FUNC(remquo) (5.0, 2.0, &quo);
|
|
check ("remquo(5.0, 2.0, &x) == 1.0", result, 1.0);
|
|
check_long ("remquo (5.0, 2.0, &x) puts 2 in x", quo, 2);
|
|
|
|
result = FUNC(remquo) (3.0, 2.0, &quo);
|
|
check ("remquo(3.0, 2.0, &x) == -1.0", result, -1.0);
|
|
check_long ("remquo (3.0, 2.0, &x) puts 2 in x", quo, 2);
|
|
}
|
|
|
|
|
|
static void
|
|
cexp_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_zero, plus_zero));
|
|
check ("real(cexp(0 + 0i)) = 1", __real__ result, 1);
|
|
check ("imag(cexp(0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_zero, plus_zero));
|
|
check ("real(cexp(-0 + 0i)) = 1", __real__ result, 1);
|
|
check ("imag(cexp(-0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_zero, minus_zero));
|
|
check ("real(cexp(0 - 0i)) = 1", __real__ result, 1);
|
|
check ("imag(cexp(0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(cexp(-0 - 0i)) = 1", __real__ result, 1);
|
|
check ("imag(cexp(-0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, plus_zero));
|
|
check_isinfp ("real(cexp(+inf + 0i)) = +inf", __real__ result);
|
|
check ("imag(cexp(+inf + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(cexp(+inf - 0i)) = +inf", __real__ result);
|
|
check ("imag(cexp(+inf - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, plus_zero));
|
|
check ("real(cexp(-inf + 0i)) = 0", __real__ result, 0);
|
|
check ("imag(cexp(-inf + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(cexp(-inf - 0i)) = 0", __real__ result, 0);
|
|
check ("imag(cexp(-inf - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (0.0, plus_infty));
|
|
check_isnan_exc ("real(cexp(0 + i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(0 + i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
#if defined __GNUC__ && __GNUC__ <= 2 && __GNUC_MINOR__ <= 7
|
|
if (verbose)
|
|
printf ("The following test for cexp might fail due to a gcc compiler error!\n");
|
|
#endif
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isnan_exc ("real(cexp(-0 + i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(-0 + i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (0.0, minus_infty));
|
|
check_isnan_exc ("real(cexp(0 - i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(0 - i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isnan_exc ("real(cexp(-0 - i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(-0 - i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (100.0, plus_infty));
|
|
check_isnan_exc ("real(cexp(100.0 + i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(100.0 + i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (-100.0, plus_infty));
|
|
check_isnan_exc ("real(cexp(-100.0 + i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(-100.0 + i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (100.0, minus_infty));
|
|
check_isnan_exc ("real(cexp(100.0 - i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(100.0 - i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (-100.0, minus_infty));
|
|
check_isnan_exc ("real(cexp(-100.0 - i inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(-100.0 - i inf)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, 2.0));
|
|
check ("real(cexp(-inf + 2.0i)) = -0", __real__ result, minus_zero);
|
|
check ("imag(cexp(-inf + 2.0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, 4.0));
|
|
check ("real(cexp(-inf + 4.0i)) = -0", __real__ result, minus_zero);
|
|
check ("imag(cexp(-inf + 4.0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, 2.0));
|
|
check_isinfn ("real(cexp(+inf + 2.0i)) = -inf", __real__ result);
|
|
check_isinfp ("imag(cexp(+inf + 2.0i)) = +inf", __imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, 4.0));
|
|
check_isinfn ("real(cexp(+inf + 4.0i)) = -inf", __real__ result);
|
|
check_isinfn ("imag(cexp(+inf + 4.0i)) = -inf", __imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp_exc ("real(cexp(+inf + i inf)) = +inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(+inf + i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp_exc ("real(cexp(+inf - i inf)) = +inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(+inf - i inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check ("real(cexp(-inf + i inf)) = 0", __real__ result, 0);
|
|
check ("imag(cexp(-inf + i inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check ("real(cexp(-inf - i inf)) = 0", __real__ result, 0);
|
|
check ("imag(cexp(-inf - i inf)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check ("real(cexp(-inf + i NaN)) = 0", __real__ result, 0);
|
|
check ("imag(cexp(-inf + i NaN)) = 0", fabs (__imag__ result), 0);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(cexp(+inf + i NaN)) = +inf", __real__ result);
|
|
check_isnan ("imag(cexp(+inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (nan_value, 0.0));
|
|
check_isnan_maybe_exc ("real(cexp(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (nan_value, 1.0));
|
|
check_isnan_maybe_exc ("real(cexp(NaN + 1i)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(NaN + 1i)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan_maybe_exc ("real(cexp(NaN + i inf)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(NaN + i inf)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (0, nan_value));
|
|
check_isnan_maybe_exc ("real(cexp(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cexp) (BUILD_COMPLEX (1, nan_value));
|
|
check_isnan_maybe_exc ("real(cexp(1 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cexp(1 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(cexp(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(cexp(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(cexp(0.7 + i 1.2)) == 0.72969...", __real__ result,
|
|
0.7296989091503236012L, CHOOSE(6e-17L, 2e-16, 2e-7));
|
|
check_eps ("imag(cexp(0.7 + i 1.2)) == 1.87689...", __imag__ result,
|
|
1.8768962328348102821L, CHOOSE(2e-16L, 0, 3e-7));
|
|
|
|
result = FUNC(cexp) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(cexp(-2 - i 3)) == -0.13398...", __real__ result,
|
|
-0.1339809149295426134L, CHOOSE(6e-20L, 0, 2e-8));
|
|
check_eps ("imag(cexp(-2 - i 3)) == -0.01909...", __imag__ result,
|
|
-0.0190985162611351964L, CHOOSE(4e-20L, 0, 2e-9));
|
|
}
|
|
|
|
|
|
static void
|
|
csin_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (0.0, 0.0));
|
|
check ("real(csin(0 + 0i)) = 0", __real__ result, 0);
|
|
check ("imag(csin(0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_zero, 0.0));
|
|
check ("real(csin(-0 + 0i)) = -0", __real__ result, minus_zero);
|
|
check ("imag(csin(-0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(csin) (BUILD_COMPLEX (0.0, minus_zero));
|
|
check ("real(csin(0 - 0i)) = 0", __real__ result, 0);
|
|
check ("imag(csin(0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(csin(-0 - 0i)) = -0", __real__ result, minus_zero);
|
|
check ("imag(csin(-0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (0.0, plus_infty));
|
|
check ("real(csin(0 + i Inf)) = 0", __real__ result, 0);
|
|
check_isinfp ("imag(csin(0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check ("real(csin(-0 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check_isinfp ("imag(csin(-0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (0.0, minus_infty));
|
|
check ("real(csin(0 - i Inf)) = 0", __real__ result, 0);
|
|
check_isinfn ("imag(csin(0 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check ("real(csin(-0 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check_isinfn("imag(csin(-0 - i Inf)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, 0.0));
|
|
check_isnan_exc ("real(csin(+Inf + 0i)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(csin(+Inf + 0i)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, 0.0));
|
|
check_isnan_exc ("real(csin(-Inf + 0i)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(csin(-Inf + 0i)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isnan_exc ("real(csin(+Inf - 0i)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(csin(+Inf - 0i)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0.0);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isnan_exc ("real(csin(-Inf - 0i)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(csin(-Inf - 0i)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0.0);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isnan_exc ("real(csin(+Inf + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isinfp ("imag(csin(+Inf + i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__imag__ result));
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isnan_exc ("real(csin(-Inf + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isinfp ("imag(csin(-Inf + i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__imag__ result));
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isnan_exc ("real(csin(Inf - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isinfp ("imag(csin(Inf - i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__imag__ result));
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isnan_exc ("real(csin(-Inf - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isinfp ("imag(csin(-Inf - i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__imag__ result));
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, 6.75));
|
|
check_isnan_exc ("real(csin(+Inf + i 6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(+Inf + i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, -6.75));
|
|
check_isnan_exc ("real(csin(+Inf - i 6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(+Inf - i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, 6.75));
|
|
check_isnan_exc ("real(csin(-Inf + i6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(-Inf + i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, -6.75));
|
|
check_isnan_exc ("real(csin(-Inf - i6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(-Inf - i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (4.625, plus_infty));
|
|
check_isinfn ("real(csin(4.625 + i Inf)) = -Inf", __real__ result);
|
|
check_isinfn ("imag(csin(4.625 + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (4.625, minus_infty));
|
|
check_isinfn ("real(csin(4.625 - i Inf)) = -Inf", __real__ result);
|
|
check_isinfp ("imag(csin(4.625 - i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (-4.625, plus_infty));
|
|
check_isinfp ("real(csin(-4.625 + i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csin(-4.625 + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (-4.625, minus_infty));
|
|
check_isinfp ("real(csin(-4.625 - i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csin(-4.625 - i Inf)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, 0.0));
|
|
check_isnan ("real(csin(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(csin(NaN + i0)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(csin(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(csin(NaN - i0)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan ("real(csin(NaN + i Inf)) = NaN", __real__ result);
|
|
check_isinfp ("imag(csin(NaN + i Inf)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isnan ("real(csin(NaN - i Inf)) = NaN", __real__ result);
|
|
check_isinfp ("real(csin(NaN - i Inf)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, 9.0));
|
|
check_isnan_maybe_exc ("real(csin(NaN + i9.0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(NaN + i9.0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, -9.0));
|
|
check_isnan_maybe_exc ("real(csin(NaN - i9.0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(NaN - i9.0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (0.0, nan_value));
|
|
check ("real(csin(0 + i NaN))", __real__ result, 0.0);
|
|
check_isnan ("imag(csin(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check ("real(csin(-0 + i NaN)) = -0", __real__ result, minus_zero);
|
|
check_isnan ("imag(csin(-0 + NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (10.0, nan_value));
|
|
check_isnan_maybe_exc ("real(csin(10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, -10.0));
|
|
check_isnan_maybe_exc ("real(csin(-10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(-10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isnan_maybe_exc ("real(csin(+Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(+Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csin) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isnan_maybe_exc ("real(csin(-Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csin(-Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(csin(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(csin(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(csin(0.7 + i 1.2)) = 1.166456341...", __real__ result,
|
|
1.1664563419657581376L, CHOOSE(2e-16L, 0, 0));
|
|
check_eps ("imag(csin(0.7 + i 1.2)) = 1.154499724...", __imag__ result,
|
|
1.1544997246948547371L, CHOOSE(2e-17L, 0, 2e-7));
|
|
|
|
result = FUNC(csin) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(csin(-2 - i 3)) == -9.15449...", __real__ result,
|
|
-9.1544991469114295734L, CHOOSE(4e-18L, 0, 1e-6));
|
|
check_eps ("imag(csin(-2 - i 3)) == -4.16890...", __imag__ result,
|
|
4.1689069599665643507L, CHOOSE(2e-17L, 0, 5e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
csinh_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (0.0, 0.0));
|
|
check ("real(csinh(0 + 0i)) = 0", __real__ result, 0);
|
|
check ("imag(csinh(0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_zero, 0.0));
|
|
check ("real(csinh(-0 + 0i)) = -0", __real__ result, minus_zero);
|
|
check ("imag(csinh(-0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (0.0, minus_zero));
|
|
check ("real(csinh(0 - 0i)) = 0", __real__ result, 0);
|
|
check ("imag(csinh(0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(csinh(-0 - 0i)) = -0", __real__ result, minus_zero);
|
|
check ("imag(csinh(-0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (0.0, plus_infty));
|
|
check_exc ("real(csinh(0 + i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__real__ result), 0, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(0 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_exc ("real(csinh(-0 + i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__real__ result), 0, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-0 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (0.0, minus_infty));
|
|
check_exc ("real(csinh(0 - i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__real__ result), 0, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(0 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_exc ("real(csinh(-0 - i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__real__ result), 0, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-0 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, 0.0));
|
|
check_isinfp ("real(csinh(+Inf + 0i)) = +Inf", __real__ result);
|
|
check ("imag(csinh(+Inf + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, 0.0));
|
|
check_isinfn ("real(csinh(-Inf + 0i)) = -Inf", __real__ result);
|
|
check ("imag(csinh(-Inf + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(csinh(+Inf - 0i)) = +Inf", __real__ result);
|
|
check ("imag(csinh(+Inf - 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isinfn ("real(csinh(-Inf - 0i)) = -Inf", __real__ result);
|
|
check ("imag(csinh(-Inf - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp_exc ("real(csinh(+Inf + i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__real__ result), INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(+Inf + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp_exc ("real(csinh(-Inf + i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__real__ result), INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-Inf + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp_exc ("real(csinh(Inf - i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__real__ result), INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(Inf - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp_exc ("real(csinh(-Inf - i Inf)) = +-Inf plus invalid exception",
|
|
FUNC(fabs) (__real__ result), INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-Inf - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, 4.625));
|
|
check_isinfn ("real(csinh(+Inf + i4.625)) = -Inf", __real__ result);
|
|
check_isinfn ("imag(csinh(+Inf + i4.625)) = -Inf", __imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, 4.625));
|
|
check_isinfp ("real(csinh(-Inf + i4.625)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csinh(-Inf + i4.625)) = -Inf", __imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, -4.625));
|
|
check_isinfn ("real(csinh(+Inf - i4.625)) = -Inf", __real__ result);
|
|
check_isinfp ("imag(csinh(+Inf - i4.625)) = +Inf", __imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, -4.625));
|
|
check_isinfp ("real(csinh(-Inf - i4.625)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csinh(-Inf - i4.625)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (6.75, plus_infty));
|
|
check_isnan_exc ("real(csinh(6.75 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(6.75 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (-6.75, plus_infty));
|
|
check_isnan_exc ("real(csinh(-6.75 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-6.75 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (6.75, minus_infty));
|
|
check_isnan_exc ("real(csinh(6.75 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(6.75 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (-6.75, minus_infty));
|
|
check_isnan_exc ("real(csinh(-6.75 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-6.75 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (0.0, nan_value));
|
|
check ("real(csinh(0 + i NaN)) = +-0", FUNC(fabs) (__real__ result), 0);
|
|
check_isnan ("imag(csinh(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check ("real(csinh(-0 + i NaN)) = +-0", FUNC(fabs) (__real__ result), 0);
|
|
check_isnan ("imag(csinh(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(csinh(+Inf + i NaN)) = +-Inf",
|
|
FUNC(fabs) (__real__ result));
|
|
check_isnan ("imag(csinh(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isinfp ("real(csinh(-Inf + i NaN)) = +-Inf",
|
|
FUNC(fabs) (__real__ result));
|
|
check_isnan ("imag(csinh(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (9.0, nan_value));
|
|
check_isnan_maybe_exc ("real(csinh(9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (-9.0, nan_value));
|
|
check_isnan_maybe_exc ("real(csinh(-9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(-9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, 0.0));
|
|
check_isnan ("real(csinh(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(csinh(NaN + i0)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(csinh(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(csinh(NaN - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, 10.0));
|
|
check_isnan_maybe_exc ("real(csinh(NaN + i10)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(NaN + i10)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, -10.0));
|
|
check_isnan_maybe_exc ("real(csinh(NaN - i10)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(NaN - i10)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan_maybe_exc ("real(csinh(NaN + i Inf)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(NaN + i Inf)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isnan_maybe_exc ("real(csinh(NaN - i Inf)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csinh(NaN - i Inf)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(csinh(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(csinh(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(csinh(0.7 + i 1.2)) = 0.274878686...", __real__ result,
|
|
0.27487868678117583582L, CHOOSE(2e-17L, 6e-17, 3e-8));
|
|
check_eps ("imag(csinh(0.7 + i 1.2)) = 1.169866572...", __imag__ result,
|
|
1.1698665727426565139L, CHOOSE(6e-17L, 0, 2e-7));
|
|
|
|
result = FUNC(csinh) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(csinh(-2 - i 3)) == -3.59056...", __real__ result,
|
|
3.5905645899857799520L, CHOOSE(7e-19L, 5e-16, 3e-7));
|
|
check_eps ("imag(csinh(-2 - i 3)) == -0.53092...", __imag__ result,
|
|
-0.5309210862485198052L, CHOOSE(3e-19L, 2e-16, 6e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
ccos_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (0.0, 0.0));
|
|
check ("real(ccos(0 + 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccos(0 + 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_zero, 0.0));
|
|
check ("real(ccos(-0 + 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccos(-0 + 0i)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (0.0, minus_zero));
|
|
check ("real(ccos(0 - 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccos(0 - 0i)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(ccos(-0 - 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccos(-0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, 0.0));
|
|
check_isnan_exc ("real(ccos(+Inf + i0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccos(Inf + i0)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isnan_exc ("real(ccos(Inf - i0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccos(Inf - i0)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, 0.0));
|
|
check_isnan_exc ("real(ccos(-Inf + i0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccos(-Inf + i0)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isnan_exc ("real(ccos(-Inf - i0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccos(-Inf - i0)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (0.0, plus_infty));
|
|
check_isinfp ("real(ccos(0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(ccos(0 + i Inf)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (0.0, minus_infty));
|
|
check_isinfp ("real(ccos(0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(ccos(0 - i Inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isinfp ("real(ccos(-0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(ccos(-0 + i Inf)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isinfp ("real(ccos(-0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(ccos(-0 - i Inf)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp_exc ("real(ccos(+Inf + i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(+Inf + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp_exc ("real(ccos(-Inf + i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(-Inf + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp_exc ("real(ccos(Inf - i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(Inf - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp_exc ("real(ccos(-Inf - i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(-Inf - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (4.625, plus_infty));
|
|
check_isinfn ("real(ccos(4.625 + i Inf)) = -Inf", __real__ result);
|
|
check_isinfp ("imag(ccos(4.625 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (4.625, minus_infty));
|
|
check_isinfn ("real(ccos(4.625 - i Inf)) = -Inf", __real__ result);
|
|
check_isinfn ("imag(ccos(4.625 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (-4.625, plus_infty));
|
|
check_isinfn ("real(ccos(-4.625 + i Inf)) = -Inf", __real__ result);
|
|
check_isinfn ("imag(ccos(-4.625 + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (-4.625, minus_infty));
|
|
check_isinfn ("real(ccos(-4.625 - i Inf)) = -Inf", __real__ result);
|
|
check_isinfp ("imag(ccos(-4.625 - i Inf)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, 6.75));
|
|
check_isnan_exc ("real(ccos(+Inf + i6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(+Inf + i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, -6.75));
|
|
check_isnan_exc ("real(ccos(+Inf - i6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(+Inf - i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, 6.75));
|
|
check_isnan_exc ("real(ccos(-Inf + i6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(-Inf + i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, -6.75));
|
|
check_isnan_exc ("real(ccos(-Inf - i6.75)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(-Inf - i6.75)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, 0.0));
|
|
check_isnan ("real(ccos(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(ccos(NaN + i0)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(ccos(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(ccos(NaN - i0)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isinfp ("real(ccos(NaN + i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(ccos(NaN + i Inf)) = NaN", __imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isinfp ("real(ccos(NaN - i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(ccos(NaN - i Inf)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, 9.0));
|
|
check_isnan_maybe_exc ("real(ccos(NaN + i9.0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(NaN + i9.0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, -9.0));
|
|
check_isnan_maybe_exc ("real(ccos(NaN - i9.0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(NaN - i9.0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (0.0, nan_value));
|
|
check_isnan ("real(ccos(0 + i NaN)) = NaN", __real__ result);
|
|
check ("imag(ccos(0 + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0.0);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan ("real(ccos(-0 + i NaN)) = NaN", __real__ result);
|
|
check ("imag(ccos(-0 + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0.0);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (10.0, nan_value));
|
|
check_isnan_maybe_exc ("real(ccos(10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (-10.0, nan_value));
|
|
check_isnan_maybe_exc ("real(ccos(-10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(-10 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isnan_maybe_exc ("real(ccos(+Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(+Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccos) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isnan_maybe_exc ("real(ccos(-Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccos(-Inf + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(ccos(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(ccos(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(ccos(0.7 + i 1.2)) = 1.384865764...", __real__ result,
|
|
1.3848657645312111080L, CHOOSE(4e-18L, 3e-16, 2e-7));
|
|
check_eps ("imag(ccos(0.7 + i 1.2)) = -0.972421703...", __imag__ result,
|
|
-0.97242170335830028619L, CHOOSE(2e-16L, 2e-16, 0));
|
|
|
|
result = FUNC(ccos) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(ccos(-2 - i 3)) == -4.18962...", __real__ result,
|
|
-4.1896256909688072301L, CHOOSE(2e-17L, 0, 5e-7));
|
|
check_eps ("imag(ccos(-2 - i 3)) == -9.10922...", __imag__ result,
|
|
-9.1092278937553365979L, CHOOSE(3e-18L, 0, 1e-6));
|
|
}
|
|
|
|
|
|
static void
|
|
ccosh_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (0.0, 0.0));
|
|
check ("real(ccosh(0 + 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccosh(0 + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_zero, 0.0));
|
|
check ("real(ccosh(-0 + 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccosh(-0 + 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (0.0, minus_zero));
|
|
check ("real(ccosh(0 - 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccosh(0 - 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(ccosh(-0 - 0i)) = 1.0", __real__ result, 1.0);
|
|
check ("imag(ccosh(-0 - 0i)) = 0", __imag__ result, 0.0);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (0.0, plus_infty));
|
|
check_isnan_exc ("real(ccosh(0 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccosh(0 + i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isnan_exc ("real(ccosh(-0 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccosh(-0 + i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (0.0, minus_infty));
|
|
check_isnan_exc ("real(ccosh(0 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccosh(0 - i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isnan_exc ("real(ccosh(-0 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check ("imag(ccosh(-0 - i Inf)) = +-0 plus invalid exception",
|
|
FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, 0.0));
|
|
check_isinfp ("real(ccosh(+Inf + 0i)) = +Inf", __real__ result);
|
|
check ("imag(ccosh(+Inf + 0i)) = 0", __imag__ result, 0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, 0.0));
|
|
check_isinfp ("real(ccosh(-Inf + 0i)) = +Inf", __real__ result);
|
|
check ("imag(ccosh(-Inf + 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(ccosh(+Inf - 0i)) = +Inf", __real__ result);
|
|
check ("imag(ccosh(+Inf - 0i)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isinfp ("real(ccosh(-Inf - 0i)) = +Inf", __real__ result);
|
|
check ("imag(ccosh(-Inf - 0i)) = 0", __imag__ result, 0.0);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp_exc ("real(ccosh(+Inf + i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(+Inf + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp_exc ("real(ccosh(-Inf + i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(-Inf + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp_exc ("real(ccosh(Inf - i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(Inf - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp_exc ("real(ccosh(-Inf - i Inf)) = +Inf plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(-Inf - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, 4.625));
|
|
check_isinfn ("real(ccosh(+Inf + i4.625)) = -Inf", __real__ result);
|
|
check_isinfn ("imag(ccosh(+Inf + i4.625)) = -Inf", __imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, 4.625));
|
|
check_isinfn ("real(ccosh(-Inf + i4.625)) = -Inf", __real__ result);
|
|
check_isinfp ("imag(ccosh(-Inf + i4.625)) = Inf", __imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, -4.625));
|
|
check_isinfn ("real(ccosh(+Inf - i4.625)) = -Inf", __real__ result);
|
|
check_isinfp ("imag(ccosh(+Inf - i4.625)) = +Inf", __imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, -4.625));
|
|
check_isinfn ("real(ccosh(-Inf - i4.625)) = -Inf", __real__ result);
|
|
check_isinfn ("imag(ccosh(-Inf - i4.625)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (6.75, plus_infty));
|
|
check_isnan_exc ("real(ccosh(6.75 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(6.75 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (-6.75, plus_infty));
|
|
check_isnan_exc ("real(ccosh(-6.75 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(-6.75 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (6.75, minus_infty));
|
|
check_isnan_exc ("real(ccosh(6.75 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(6.75 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (-6.75, minus_infty));
|
|
check_isnan_exc ("real(ccosh(-6.75 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(-6.75 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (0.0, nan_value));
|
|
check_isnan ("real(ccosh(0 + i NaN)) = NaN", __real__ result);
|
|
check ("imag(ccosh(0 + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan ("real(ccosh(-0 + i NaN)) = NaN", __real__ result);
|
|
check ("imag(ccosh(-0 + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(ccosh(+Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(ccosh(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isinfp ("real(ccosh(-Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(ccosh(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (9.0, nan_value));
|
|
check_isnan_maybe_exc ("real(ccosh(9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (-9.0, nan_value));
|
|
check_isnan_maybe_exc ("real(ccosh(-9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(-9.0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, 0.0));
|
|
check_isnan ("real(ccosh(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(ccosh(NaN + i0)) = +-0", FUNC(fabs) (__imag__ result), 0.0);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(ccosh(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(ccosh(NaN - i0)) = +-0", FUNC(fabs) (__imag__ result), 0.0);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, 10.0));
|
|
check_isnan_maybe_exc ("real(ccosh(NaN + i10)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(NaN + i10)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, -10.0));
|
|
check_isnan_maybe_exc ("real(ccosh(NaN - i10)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(NaN - i10)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan_maybe_exc ("real(ccosh(NaN + i Inf)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(NaN + i Inf)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isnan_maybe_exc ("real(ccosh(NaN - i Inf)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ccosh(NaN - i Inf)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(ccosh(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(ccosh(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(ccosh(0.7 + i 1.2)) == 0.45482...", __real__ result,
|
|
0.4548202223691477654L, CHOOSE(5e-17L, 6e-17, 9e-8));
|
|
check_eps ("imag(ccosh(0.7 + i 1.2)) == 0.70702...", __imag__ result,
|
|
0.7070296600921537682L, CHOOSE(7e-17L, 2e-16, 0));
|
|
|
|
result = FUNC(ccosh) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(ccosh(-2 - i 3)) == -3.72454...", __real__ result,
|
|
-3.7245455049153225654L, CHOOSE(7e-19L, 0, 3e-7));
|
|
check_eps ("imag(ccosh(-2 - i 3)) == -0.51182...", __imag__ result,
|
|
0.5118225699873846088L, CHOOSE(3e-19L, 2e-16, 6e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
cacos_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0, 0));
|
|
check ("real(cacos(0 + i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(cacos(0 + i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(cacos(-0 + i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(cacos(-0 + i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(cacos(0 - i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(cacos(0 - i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(cacos(-0 - i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(cacos(-0 - i0)) = 0", __imag__ result, 0);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check ("real(cacos(-Inf + i Inf)) = 3*pi/4", __real__ result,
|
|
M_PIl - M_PI_4l);
|
|
check_isinfn ("imag(cacos(-Inf + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check ("real(cacos(-Inf - i Inf)) = 3*pi/4", __real__ result,
|
|
M_PIl - M_PI_4l);
|
|
check_isinfp ("imag(cacos(-Inf - i Inf)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check ("real(cacos(+Inf + i Inf)) = pi/4", __real__ result, M_PI_4l);
|
|
check_isinfn ("imag(cacos(+Inf + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check ("real(cacos(+Inf - i Inf)) = pi/4", __real__ result, M_PI_4l);
|
|
check_isinfp ("imag(cacos(+Inf - i Inf)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (-10.0, plus_infty));
|
|
check ("real(cacos(-10.0 + i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfn ("imag(cacos(-10.0 + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (-10.0, minus_infty));
|
|
check ("real(cacos(-10.0 - i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfp ("imag(cacos(-10.0 - i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0, plus_infty));
|
|
check ("real(cacos(0 + i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfn ("imag(cacos(0 + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0, minus_infty));
|
|
check ("real(cacos(0 - i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfp ("imag(cacos(0 - i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0.1, plus_infty));
|
|
check ("real(cacos(0.1 + i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfn ("imag(cacos(0.1 + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0.1, minus_infty));
|
|
check ("real(cacos(0.1 - i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfp ("imag(cacos(0.1 - i Inf)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, 0));
|
|
check ("real(cacos(-Inf + i0)) = pi", __real__ result, M_PIl);
|
|
check_isinfn ("imag(cacos(-Inf + i0)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(cacos(-Inf - i0)) = pi", __real__ result, M_PIl);
|
|
check_isinfp ("imag(cacos(-Inf - i0)) = +Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, 100));
|
|
check ("real(cacos(-Inf + i100)) = pi", __real__ result, M_PIl);
|
|
check_isinfn ("imag(cacos(-Inf + i100)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, -100));
|
|
check ("real(cacos(-Inf - i100)) = pi", __real__ result, M_PIl);
|
|
check_isinfp ("imag(cacos(-Inf - i100)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, 0));
|
|
check ("real(cacos(+Inf + i0)) = 0", __real__ result, 0);
|
|
check_isinfn ("imag(cacos(+Inf + i0)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check ("real(cacos(+Inf - i0)) = 0", __real__ result, 0);
|
|
check_isinfp ("imag(cacos(+Inf - i0)) = +Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, 0.5));
|
|
check ("real(cacos(+Inf + i0.5)) = 0", __real__ result, 0);
|
|
check_isinfn ("imag(cacos(+Inf + i0.5)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, -0.5));
|
|
check ("real(cacos(+Inf - i0.5)) = 0", __real__ result, 0);
|
|
check_isinfp ("imag(cacos(+Inf - i0.5)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isnan ("real(cacos(+Inf + i NaN)) = NaN", __real__ result);
|
|
check_isinfp ("imag(cacos(+Inf + i NaN)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isnan ("real(cacos(-Inf + i NaN)) = NaN", __real__ result);
|
|
check_isinfp ("imag(cacos(-Inf + i NaN)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0, nan_value));
|
|
check ("real(cacos(0 + i NaN)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isnan ("imag(cacos(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check ("real(cacos(-0 + i NaN)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isnan ("imag(cacos(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan ("real(cacos(NaN + i Inf)) = NaN", __real__ result);
|
|
check_isinfn ("imag(cacos(NaN + i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isnan ("real(cacos(NaN - i Inf)) = NaN", __real__ result);
|
|
check_isinfp ("imag(cacos(NaN - i Inf)) = +Inf", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(cacos(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacos(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (-10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(cacos(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacos(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (nan_value, 0.75));
|
|
check_isnan_maybe_exc ("real(cacos(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacos(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cacos) (BUILD_COMPLEX (-10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(cacos(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacos(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(cacos(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(cacos(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(cacos(0.7 + i 1.2)) == 1.13518...", __real__ result,
|
|
1.1351827477151551089L, CHOOSE(2e-17L, 3e-16, 2e-7));
|
|
check_eps ("imag(cacos(0.7 + i 1.2)) == -1.09276...", __imag__ result,
|
|
-1.0927647857577371459L, CHOOSE(4e-17L, 3e-16, 3e-7));
|
|
|
|
result = FUNC(cacos) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(cacos(-2 - i 3)) == 2.14144...", __real__ result,
|
|
2.1414491111159960199L, CHOOSE(3e-19L, 0, 0));
|
|
check_eps ("imag(cacos(-2 - i 3)) == -1.98338...", __imag__ result,
|
|
1.9833870299165354323L, CHOOSE(3e-19L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
cacosh_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0, 0));
|
|
check ("real(cacosh(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(cacosh(0 + i0)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(cacosh(-0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(cacosh(-0 + i0)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(cacosh(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(cacosh(0 - i0)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(cacosh(-0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(cacosh(-0 - i0)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp ("real(cacosh(-Inf + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-Inf + i Inf)) = 3*pi/4", __imag__ result,
|
|
M_PIl - M_PI_4l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp ("real(cacosh(-Inf - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-Inf - i Inf)) = -3*pi/4", __imag__ result,
|
|
M_PI_4l - M_PIl);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp ("real(cacosh(+Inf + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(+Inf + i Inf)) = pi/4", __imag__ result, M_PI_4l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp ("real(cacosh(+Inf - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(+Inf - i Inf)) = -pi/4", __imag__ result, -M_PI_4l);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (-10.0, plus_infty));
|
|
check_isinfp ("real(cacosh(-10.0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-10.0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (-10.0, minus_infty));
|
|
check_isinfp ("real(cacosh(-10.0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-10.0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0, plus_infty));
|
|
check_isinfp ("real(cacosh(0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0, minus_infty));
|
|
check_isinfp ("real(cacosh(0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0.1, plus_infty));
|
|
check_isinfp ("real(cacosh(0.1 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(0.1 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0.1, minus_infty));
|
|
check_isinfp ("real(cacosh(0.1 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(0.1 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, 0));
|
|
check_isinfp ("real(cacosh(-Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-Inf + i0)) = pi", __imag__ result, M_PIl);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isinfp ("real(cacosh(-Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-Inf - i0)) = -pi", __imag__ result, -M_PIl);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, 100));
|
|
check_isinfp ("real(cacosh(-Inf + i100)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-Inf + i100)) = pi", __imag__ result, M_PIl);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, -100));
|
|
check_isinfp ("real(cacosh(-Inf - i100)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(-Inf - i100)) = -pi", __imag__ result, -M_PIl);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, 0));
|
|
check_isinfp ("real(cacosh(+Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(+Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(cacosh(+Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(+Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, 0.5));
|
|
check_isinfp ("real(cacosh(+Inf + i0.5)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(+Inf + i0.5)) = 0", __imag__ result, 0);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, -0.5));
|
|
check_isinfp ("real(cacosh(+Inf - i0.5)) = +Inf", __real__ result);
|
|
check ("imag(cacosh(+Inf - i0.5)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(cacosh(+Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(cacosh(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isinfp ("real(cacosh(-Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(cacosh(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0, nan_value));
|
|
check_isnan ("real(cacosh(0 + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(cacosh(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan ("real(cacosh(-0 + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(cacosh(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isinfp ("real(cacosh(NaN + i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(cacosh(NaN + i Inf)) = NaN", __imag__ result);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isinfp ("real(cacosh(NaN - i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(cacosh(NaN - i Inf)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(cacosh(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacosh(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (-10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(cacosh(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacosh(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (nan_value, 0.75));
|
|
check_isnan_maybe_exc ("real(cacosh(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacosh(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (-10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(cacosh(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(cacosh(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(cacosh(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(cacosh(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(cacosh(0.7 + i 1.2)) == 1.09276...", __real__ result,
|
|
1.0927647857577371459L, CHOOSE(4e-17L, 3e-16, 2e-7));
|
|
check_eps ("imag(cacosh(0.7 + i 1.2)) == 1.13518...", __imag__ result,
|
|
1.1351827477151551089L, CHOOSE(2e-17L, 0, 0));
|
|
|
|
result = FUNC(cacosh) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(cacosh(-2 - i 3)) == -1.98338...", __real__ result,
|
|
-1.9833870299165354323L, CHOOSE (2e-18L, 3e-16, 9e-7));
|
|
check_eps ("imag(cacosh(-2 - i 3)) == 2.14144...", __imag__ result,
|
|
2.1414491111159960199L, CHOOSE (3e-19, 5e-16, 1e-6));
|
|
}
|
|
|
|
|
|
static void
|
|
casin_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (0, 0));
|
|
check ("real(casin(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(casin(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(casin(-0 + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(casin(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casin) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(casin(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(casin(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(casin(-0 - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(casin(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check ("real(casin(+Inf + i Inf)) = pi/4", __real__ result, M_PI_4l);
|
|
check_isinfp ("imag(casin(+Inf + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check ("real(casin(+Inf - i Inf)) = pi/4", __real__ result, M_PI_4l);
|
|
check_isinfn ("imag(casin(+Inf - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check ("real(casin(-Inf + i Inf)) = -pi/4", __real__ result, -M_PI_4l);
|
|
check_isinfp ("imag(casin(-Inf + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check ("real(casin(-Inf - i Inf)) = -pi/4", __real__ result, -M_PI_4l);
|
|
check_isinfn ("imag(casin(-Inf - i Inf)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (-10.0, plus_infty));
|
|
check ("real(casin(-10.0 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check_isinfp ("imag(casin(-10.0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (-10.0, minus_infty));
|
|
check ("real(casin(-10.0 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check_isinfn ("imag(casin(-10.0 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (0, plus_infty));
|
|
check ("real(casin(0 + i Inf)) = 0", __real__ result, 0.0);
|
|
check_isinfp ("imag(casin(0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (0, minus_infty));
|
|
check ("real(casin(0 - i Inf)) = 0", __real__ result, 0.0);
|
|
check_isinfn ("imag(casin(0 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check ("real(casin(-0 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check_isinfp ("imag(casin(-0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check ("real(casin(-0 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check_isinfn ("imag(casin(-0 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (0.1, plus_infty));
|
|
check ("real(casin(0.1 + i Inf)) = 0", __real__ result, 0);
|
|
check_isinfp ("imag(casin(0.1 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (0.1, minus_infty));
|
|
check ("real(casin(0.1 - i Inf)) = 0", __real__ result, 0);
|
|
check_isinfn ("imag(casin(0.1 - i Inf)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, 0));
|
|
check ("real(casin(-Inf + i0)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check_isinfp ("imag(casin(-Inf + i0)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(casin(-Inf - i0)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check_isinfn ("imag(casin(-Inf - i0)) = -Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, 100));
|
|
check ("real(casin(-Inf + i100)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check_isinfp ("imag(casin(-Inf + i100)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, -100));
|
|
check ("real(casin(-Inf - i100)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check_isinfn ("imag(casin(-Inf - i100)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, 0));
|
|
check ("real(casin(+Inf + i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfp ("imag(casin(+Inf + i0)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check ("real(casin(+Inf - i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfn ("imag(casin(+Inf - i0)) = -Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, 0.5));
|
|
check ("real(casin(+Inf + i0.5)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfp ("imag(casin(+Inf + i0.5)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, -0.5));
|
|
check ("real(casin(+Inf - i0.5)) = pi/2", __real__ result, M_PI_2l);
|
|
check_isinfn ("imag(casin(+Inf - i0.5)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan ("real(casin(NaN + i Inf)) = NaN", __real__ result);
|
|
check_isinfp ("imag(casin(NaN + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isnan ("real(casin(NaN - i Inf)) = NaN", __real__ result);
|
|
check_isinfn ("imag(casin(NaN - i Inf)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (0.0, nan_value));
|
|
check ("real(casin(0 + i NaN)) = 0", __real__ result, 0.0);
|
|
check_isnan ("imag(casin(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check ("real(casin(-0 + i NaN)) = -0", __real__ result, minus_zero);
|
|
check_isnan ("imag(casin(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isnan ("real(casin(+Inf + i NaN)) = NaN", __real__ result);
|
|
check_isinfp ("imag(casin(+Inf + i NaN)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
result = FUNC(casin) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isnan ("real(casin(-Inf + i NaN)) = NaN", __real__ result);
|
|
check_isinfp ("imag(casin(-Inf + NaN)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (nan_value, 10.5));
|
|
check_isnan_maybe_exc ("real(casin(NaN + i10.5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casin(NaN + i10.5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (nan_value, -10.5));
|
|
check_isnan_maybe_exc ("real(casin(NaN - i10.5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casin(NaN - i10.5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (0.75, nan_value));
|
|
check_isnan_maybe_exc ("real(casin(0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casin(0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(casin) (BUILD_COMPLEX (-0.75, nan_value));
|
|
check_isnan_maybe_exc ("real(casin(-0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casin(-0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(casin(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(casin(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(casin(0.7 + i 1.2)) == 0.43561...", __real__ result,
|
|
0.4356135790797415103L, CHOOSE(2e-17L, 2e-16, 2e-7));
|
|
check_eps ("imag(casin(0.7 + i 1.2)) == 1.09276...", __imag__ result,
|
|
1.0927647857577371459L, CHOOSE(4e-17L, 3e-16, 3e-7));
|
|
|
|
result = FUNC(casin) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(casin(-2 - i 3)) == -0.57065...", __real__ result,
|
|
-0.5706527843210994007L, CHOOSE(4e-19L, 0, 0));
|
|
check_eps ("imag(casin(-2 - i 3)) == -1.98338...", __imag__ result,
|
|
-1.9833870299165354323L, CHOOSE(3e-19L, 0, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
casinh_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0, 0));
|
|
check ("real(casinh(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(casinh(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(casinh(-0 + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(casinh(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(casinh(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(casinh(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(casinh(-0 - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(casinh(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp ("real(casinh(+Inf + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(casinh(+Inf + i Inf)) = pi/4", __imag__ result, M_PI_4l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp ("real(casinh(+Inf - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(casinh(+Inf - i Inf)) = -pi/4", __imag__ result, -M_PI_4l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfn ("real(casinh(-Inf + i Inf)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-Inf + i Inf)) = pi/4", __imag__ result, M_PI_4l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfn ("real(casinh(-Inf - i Inf)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-Inf - i Inf)) = -pi/4", __imag__ result, -M_PI_4l);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (-10.0, plus_infty));
|
|
check_isinfn ("real(casinh(-10.0 + i Inf)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-10.0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (-10.0, minus_infty));
|
|
check_isinfn ("real(casinh(-10.0 - i Inf)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-10.0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0, plus_infty));
|
|
check_isinfp ("real(casinh(0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(casinh(0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0, minus_infty));
|
|
check_isinfp ("real(casinh(0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(casinh(0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isinfn ("real(casinh(-0 + i Inf)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isinfn ("real(casinh(-0 - i Inf)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0.1, plus_infty));
|
|
check_isinfp ("real(casinh(0.1 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(casinh(0.1 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0.1, minus_infty));
|
|
check_isinfp ("real(casinh(0.1 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(casinh(0.1 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, 0));
|
|
check_isinfn ("real(casinh(-Inf + i0)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isinfn ("real(casinh(-Inf - i0)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, 100));
|
|
check_isinfn ("real(casinh(-Inf + i100)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-Inf + i100)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, -100));
|
|
check_isinfn ("real(casinh(-Inf - i100)) = -Inf", __real__ result);
|
|
check ("imag(casinh(-Inf - i100)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, 0));
|
|
check_isinfp ("real(casinh(+Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(casinh(+Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(casinh(+Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(casinh(+Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, 0.5));
|
|
check_isinfp ("real(casinh(+Inf + i0.5)) = +Inf", __real__ result);
|
|
check ("imag(casinh(+Inf + i0.5)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, -0.5));
|
|
check_isinfp ("real(casinh(+Inf - i0.5)) = +Inf", __real__ result);
|
|
check ("imag(casinh(+Inf - i0.5)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(casinh(+Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(casinh(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isinfn ("real(casinh(-Inf + i NaN)) = -Inf", __real__ result);
|
|
check_isnan ("imag(casinh(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan ("real(casinh(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(casinh(NaN + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(casinh(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(casinh(NaN - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isinfp ("real(casinh(NaN + i Inf)) = +-Inf",
|
|
FUNC(fabs) (__real__ result));
|
|
check_isnan ("imag(casinh(NaN + i Inf)) = NaN", __imag__ result);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isinfp ("real(casinh(NaN - i Inf)) = +-Inf",
|
|
FUNC(fabs) (__real__ result));
|
|
check_isnan ("imag(casinh(NaN - i Inf)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(casinh(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casinh(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (-10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(casinh(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casinh(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (nan_value, 0.75));
|
|
check_isnan_maybe_exc ("real(casinh(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casinh(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(casinh) (BUILD_COMPLEX (-0.75, nan_value));
|
|
check_isnan_maybe_exc ("real(casinh(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(casinh(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(casinh(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(casinh(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(casinh(0.7 + i 1.2)) == 0.97865...", __real__ result,
|
|
0.9786545955936738768L, CHOOSE(5e-17L, 2e-16, 0));
|
|
check_eps ("imag(casinh(0.7 + i 1.2)) == 0.91135...", __imag__ result,
|
|
0.9113541895315601156L, CHOOSE(7e-19L, 2e-16, 2e-7));
|
|
|
|
result = FUNC(casinh) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(casinh(-2 - i 3)) == -1.96863...", __real__ result,
|
|
-1.9686379257930962917L, CHOOSE(7e-19L, 2e-15, 3e-6));
|
|
check_eps ("imag(casinh(-2 - i 3)) == -0.96465...", __imag__ result,
|
|
-0.9646585044076027920L, CHOOSE(4e-19L, 2e-15, 4e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
catan_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (0, 0));
|
|
check ("real(catan(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(catan(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(catan(-0 + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catan(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(catan(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(catan(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(catan(-0 - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catan(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check ("real(catan(+Inf + i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(+Inf + i Inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check ("real(catan(+Inf - i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(+Inf - i Inf)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check ("real(catan(-Inf + i Inf)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf + i Inf)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check ("real(catan(-Inf - i Inf)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf - i Inf)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, -10.0));
|
|
check ("real(catan(+Inf - i10.0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(+Inf - i10.0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, -10.0));
|
|
check ("real(catan(-Inf - i10.0)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf - i10.0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check ("real(catan(Inf - i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(catan(-Inf - i0)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, 0.0));
|
|
check ("real(catan(Inf + i0)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(Inf + i0)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, 0.0));
|
|
check ("real(catan(-Inf + i0)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf + i0)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, 0.1));
|
|
check ("real(catan(+Inf + i0.1)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(+Inf + i0.1)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, 0.1));
|
|
check ("real(catan(-Inf + i0.1)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf + i0.1)) = 0", __imag__ result, 0);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (0.0, minus_infty));
|
|
check ("real(catan(0 - i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(0 - i Inf)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check ("real(catan(-0 - i Inf)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-0 - i Inf)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (100.0, minus_infty));
|
|
check ("real(catan(100 - i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(100 - i Inf)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catan) (BUILD_COMPLEX (-100.0, minus_infty));
|
|
check ("real(catan(-100 - i Inf)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-100 - i Inf)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (0.0, plus_infty));
|
|
check ("real(catan(0 + i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(0 + i Inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check ("real(catan(-0 + i Inf)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-0 + i Inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (0.5, plus_infty));
|
|
check ("real(catan(0.5 + i Inf)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(0.5 + i Inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (-0.5, plus_infty));
|
|
check ("real(catan(-0.5 + i Inf)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-0.5 + i Inf)) = 0", __imag__ result, 0);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, 0.0));
|
|
check_isnan ("real(catan(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(catan(NaN + i0)) = 0", __imag__ result, 0.0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(catan(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(catan(NaN - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isnan ("real(catan(NaN + i Inf)) = NaN", __real__ result);
|
|
check ("imag(catan(NaN + i Inf)) = 0", __imag__ result, 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isnan ("real(catan(NaN - i Inf)) = NaN", __real__ result);
|
|
check ("imag(catan(NaN - i Inf)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (0.0, nan_value));
|
|
check_isnan ("real(catan(0 + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(catan(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan ("real(catan(-0 + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(catan(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check ("real(catan(+Inf + i NaN)) = pi/2", __real__ result, M_PI_2l);
|
|
check ("imag(catan(+Inf + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(catan) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check ("real(catan(-Inf + i NaN)) = -pi/2", __real__ result, -M_PI_2l);
|
|
check ("imag(catan(-Inf + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, 10.5));
|
|
check_isnan_maybe_exc ("real(catan(NaN + i10.5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catan(NaN + i10.5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, -10.5));
|
|
check_isnan_maybe_exc ("real(catan(NaN - i10.5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catan(NaN - i10.5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (0.75, nan_value));
|
|
check_isnan_maybe_exc ("real(catan(0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catan(0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(catan) (BUILD_COMPLEX (-0.75, nan_value));
|
|
check_isnan_maybe_exc ("real(catan(-0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catan(-0.75 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(catan(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(catan(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(catan(0.7 + i 1.2)) == 1.07857...", __real__ result,
|
|
1.0785743834118921877L, CHOOSE (3e-17, 0, 5e-7));
|
|
check_eps ("imag(catan(0.7 + i 1.2)) == 0.57705...", __imag__ result,
|
|
0.5770573776534306764L, CHOOSE(3e-17L, 2e-16, 6e-8));
|
|
|
|
result = FUNC(catan) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(catan(-2 - i 3)) == -1.40992...", __real__ result,
|
|
-1.4099210495965755225L, CHOOSE(0, 0, 4e-7));
|
|
check_eps ("imag(catan(-2 - i 3)) == -0.22907...", __imag__ result,
|
|
-0.2290726829685387662L, CHOOSE(1.1e-19L, 3e-17, 2e-8));
|
|
}
|
|
|
|
|
|
static void
|
|
catanh_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0, 0));
|
|
check ("real(catanh(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(catanh(-0 + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(catanh(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(catanh(-0 - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check ("real(catanh(+Inf + i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(+Inf + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check ("real(catanh(+Inf - i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(+Inf - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check ("real(catanh(-Inf + i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-Inf + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check ("real(catanh(-Inf - i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-Inf - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (-10.0, plus_infty));
|
|
check ("real(catanh(-10.0 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-10.0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (-10.0, minus_infty));
|
|
check ("real(catanh(-10.0 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-10.0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check ("real(catanh(-0 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check ("real(catanh(-0 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0, plus_infty));
|
|
check ("real(catanh(0 + i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0, minus_infty));
|
|
check ("real(catanh(0 - i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0.1, plus_infty));
|
|
check ("real(catanh(0.1 + i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(0.1 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0.1, minus_infty));
|
|
check ("real(catanh(0.1 - i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(0.1 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, 0));
|
|
check ("real(catanh(-Inf + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-Inf + i0)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(catanh(-Inf - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-Inf - i0)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, 100));
|
|
check ("real(catanh(-Inf + i100)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-Inf + i100)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, -100));
|
|
check ("real(catanh(-Inf - i100)) = -0", __real__ result, minus_zero);
|
|
check ("imag(catanh(-Inf - i100)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, 0));
|
|
check ("real(catanh(+Inf + i0)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(+Inf + i0)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check ("real(catanh(+Inf - i0)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(+Inf - i0)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, 0.5));
|
|
check ("real(catanh(+Inf + i0.5)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(+Inf + i0.5)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, -0.5));
|
|
check ("real(catanh(+Inf - i0.5)) = 0", __real__ result, 0);
|
|
check ("imag(catanh(+Inf - i0.5)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0, nan_value));
|
|
check ("real(catanh(0 + i NaN)) = 0", __real__ result, 0);
|
|
check_isnan ("imag(catanh(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check ("real(catanh(-0 + i NaN)) = -0", __real__ result, minus_zero);
|
|
check_isnan ("imag(catanh(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check ("real(catanh(+Inf + i NaN)) = 0", __real__ result, 0);
|
|
check_isnan ("imag(catanh(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check ("real(catanh(-Inf + i NaN)) = -0", __real__ result, minus_zero);
|
|
check_isnan ("imag(catanh(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan ("real(catanh(NaN + i0)) = NaN", __real__ result);
|
|
check_isnan ("imag(catanh(NaN + i0)) = NaN", __imag__ result);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(catanh(NaN - i0)) = NaN", __real__ result);
|
|
check_isnan ("imag(catanh(NaN - i0)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check ("real(catanh(NaN + i Inf)) = +-0", FUNC(fabs) (__real__ result), 0);
|
|
check ("imag(catanh(NaN + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check ("real(catanh(NaN - i Inf)) = +-0", FUNC(fabs) (__real__ result), 0);
|
|
check ("imag(catanh(NaN - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(catanh(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catanh(10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (-10.5, nan_value));
|
|
check_isnan_maybe_exc ("real(catanh(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catanh(-10.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, 0.75));
|
|
check_isnan_maybe_exc ("real(catanh(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catanh(NaN + i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, -0.75));
|
|
check_isnan_maybe_exc ("real(catanh(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(catanh(NaN - i0.75)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(catanh(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(catanh(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(catanh(0.7 + i 1.2)) == 0.26007...", __real__ result,
|
|
0.2600749516525135959L, CHOOSE (2e-18, 6e-17, 0));
|
|
check_eps ("imag(catanh(0.7 + i 1.2)) == 0.97024...", __imag__ result,
|
|
0.9702403077950989849L, CHOOSE (3e-17, 2e-16, 4e-7));
|
|
|
|
result = FUNC(catanh) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(catanh(-2 - i 3)) == -0.14694...", __real__ result,
|
|
-0.1469466662255297520L, CHOOSE (9e-20L, 2e-16, 2e-8));
|
|
check_eps ("imag(catanh(-2 - i 3)) == -1.33897...", __imag__ result,
|
|
-1.3389725222944935611L, CHOOSE (7e-19L, 0, 5e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
ctan_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0, 0));
|
|
check ("real(ctan(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(ctan(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(ctan(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(ctan(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(ctan(-0 + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctan(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(ctan(-0 - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctan(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0, plus_infty));
|
|
check ("real(ctan(0 + i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(ctan(0 + i Inf)) = 1", __imag__ result, 1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (1, plus_infty));
|
|
check ("real(ctan(1 + i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(ctan(1 + i Inf)) = 1", __imag__ result, 1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check ("real(ctan(-0 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctan(-0 + i Inf)) = 1", __imag__ result, 1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (-1, plus_infty));
|
|
check ("real(ctan(-1 + i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctan(-1 + i Inf)) = 1", __imag__ result, 1);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0, minus_infty));
|
|
check ("real(ctan(0 - i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(ctan(0 - i Inf)) = -1", __imag__ result, -1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (1, minus_infty));
|
|
check ("real(ctan(1 - i Inf)) = 0", __real__ result, 0);
|
|
check ("imag(ctan(1 - i Inf)) = -1", __imag__ result, -1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check ("real(ctan(-0 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctan(-0 - i Inf)) = -1", __imag__ result, -1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (-1, minus_infty));
|
|
check ("real(ctan(-1 - i Inf)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctan(-1 - i Inf)) = -1", __imag__ result, -1);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (plus_infty, 0));
|
|
check_isnan_exc ("real(ctan(Inf + i 0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(Inf + i 0)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (plus_infty, 2));
|
|
check_isnan_exc ("real(ctan(Inf + i 2)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(Inf + i 2)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_infty, 0));
|
|
check_isnan_exc ("real(ctan(-Inf + i 0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(-Inf + i 0)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_infty, 2));
|
|
check_isnan_exc ("real(ctan(- Inf + i 2)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(- Inf + i 2)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isnan_exc ("real(ctan(Inf - i 0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(Inf - i 0)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (plus_infty, -2));
|
|
check_isnan_exc ("real(ctan(Inf - i 2)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(Inf - i 2)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isnan_exc ("real(ctan(-Inf - i 0)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(-Inf - i 0)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_infty, -2));
|
|
check_isnan_exc ("real(ctan(-Inf - i 2)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(-Inf - i 2)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check ("real(ctan(NaN + i Inf)) = +-0", FUNC(fabs) (__real__ result), 0);
|
|
check ("imag(ctan(NaN + i Inf)) = 1", __imag__ result, 1);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check ("real(ctan(NaN - i Inf)) = +-0", FUNC(fabs) (__real__ result), 0);
|
|
check ("imag(ctan(NaN - i Inf)) = -1", __imag__ result, -1);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0, nan_value));
|
|
check ("real(ctan(0 + i NaN)) = 0", __real__ result, 0);
|
|
check_isnan ("imag(ctan(0 + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check ("real(ctan(-0 + i NaN)) = -0", __real__ result, minus_zero);
|
|
check_isnan ("imag(ctan(-0 + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0.5, nan_value));
|
|
check_isnan_maybe_exc ("real(ctan(0.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(0.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (-4.5, nan_value));
|
|
check_isnan_maybe_exc ("real(ctan(-4.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(-4.5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan_maybe_exc ("real(ctan(NaN + i 0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(NaN + i 0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, 5));
|
|
check_isnan_maybe_exc ("real(ctan(NaN + i 5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(NaN + i 5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan_maybe_exc ("real(ctan(NaN - i 0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(NaN - i 0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, -0.25));
|
|
check_isnan_maybe_exc ("real(ctan(NaN -i 0.25)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctan(NaN -i 0.25)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(ctan(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(ctan(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(ctan(0.7 + i 1.2)) == 0.17207...", __real__ result,
|
|
0.1720734197630349001L, CHOOSE(1e-17L, 3e-17, 2e-8));
|
|
check_eps ("imag(ctan(0.7 + i 1.2)) == 0.95448...", __imag__ result,
|
|
0.9544807059989405538L, CHOOSE(2e-17L, 2e-16, 6e-8));
|
|
|
|
result = FUNC(ctan) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(ctan(-2 - i 3)) == -0.00376...", __real__ result,
|
|
0.0037640256415042482L, CHOOSE(1e-19L, 5e-19, 0));
|
|
check_eps ("imag(ctan(-2 - i 3)) == -1.00323...", __imag__ result,
|
|
-1.0032386273536098014L, CHOOSE(2e-19L, 0, 2e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
ctanh_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0, 0));
|
|
check ("real(ctanh(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(ctanh(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(ctanh(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(ctanh(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(ctanh(-0 + i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctanh(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(ctanh(-0 - i0)) = -0", __real__ result, minus_zero);
|
|
check ("imag(ctanh(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (plus_infty, 0));
|
|
check ("real(ctanh(+Inf + i0)) = 1", __real__ result, 1);
|
|
check ("imag(ctanh(+Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (plus_infty, 1));
|
|
check ("real(ctanh(+Inf + i1)) = 1", __real__ result, 1);
|
|
check ("imag(ctanh(+Inf + i1)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check ("real(ctanh(+Inf - i0)) = 1", __real__ result, 1);
|
|
check ("imag(ctanh(+Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (plus_infty, -1));
|
|
check ("real(ctanh(+Inf - i1)) = 1", __real__ result, 1);
|
|
check ("imag(ctanh(+Inf - i1)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_infty, 0));
|
|
check ("real(ctanh(-Inf + i0)) = -1", __real__ result, -1);
|
|
check ("imag(ctanh(-Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_infty, 1));
|
|
check ("real(ctanh(-Inf + i1)) = -1", __real__ result, -1);
|
|
check ("imag(ctanh(-Inf + i1)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(ctanh(-Inf - i0)) = -1", __real__ result, -1);
|
|
check ("imag(ctanh(-Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_infty, -1));
|
|
check ("real(ctanh(-Inf - i1)) = -1", __real__ result, -1);
|
|
check ("imag(ctanh(-Inf - i1)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0, plus_infty));
|
|
check_isnan_exc ("real(ctanh(0 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(0 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (2, plus_infty));
|
|
check_isnan_exc ("real(ctanh(2 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(2 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0, minus_infty));
|
|
check_isnan_exc ("real(ctanh(0 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(0 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (2, minus_infty));
|
|
check_isnan_exc ("real(ctanh(2 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(2 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isnan_exc ("real(ctanh(-0 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(-0 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (-2, plus_infty));
|
|
check_isnan_exc ("real(ctanh(-2 + i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(-2 + i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isnan_exc ("real(ctanh(-0 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(-0 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (-2, minus_infty));
|
|
check_isnan_exc ("real(ctanh(-2 - i Inf)) = NaN plus invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(-2 - i Inf)) = NaN plus invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check ("real(ctanh(+Inf + i NaN)) = 1", __real__ result, 1);
|
|
check ("imag(ctanh(+Inf + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check ("real(ctanh(-Inf + i NaN)) = -1", __real__ result, -1);
|
|
check ("imag(ctanh(-Inf + i NaN)) = +-0", FUNC(fabs) (__imag__ result), 0);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan ("real(ctanh(NaN + i0)) = NaN", __real__ result);
|
|
check ("imag(ctanh(NaN + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan ("real(ctanh(NaN - i0)) = NaN", __real__ result);
|
|
check ("imag(ctanh(NaN - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (nan_value, 0.5));
|
|
check_isnan_maybe_exc ("real(ctanh(NaN + i0.5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(NaN + i0.5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (nan_value, -4.5));
|
|
check_isnan_maybe_exc ("real(ctanh(NaN - i4.5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(NaN - i4.5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0, nan_value));
|
|
check_isnan_maybe_exc ("real(ctanh(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (5, nan_value));
|
|
check_isnan_maybe_exc ("real(ctanh(5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(5 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan_maybe_exc ("real(ctanh(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (-0.25, nan_value));
|
|
check_isnan_maybe_exc ("real(ctanh(-0.25 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(ctanh(-0.25 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(ctanh(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(ctanh(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0, M_PI_4l));
|
|
check ("real(ctanh (0 + i pi/4)) == 0", __real__ result, 0);
|
|
check_eps ("imag(ctanh (0 + i pi/4)) == 1", __imag__ result, 1,
|
|
CHOOSE (0, 2e-16, 2e-7));
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(ctanh(0.7 + i 1.2)) == 1.34721...", __real__ result,
|
|
1.3472197399061191630L, CHOOSE(4e-17L, 5e-16, 2e-7));
|
|
check_eps ("imag(ctanh(0.7 + i 1.2)) == -0.47786...", __imag__ result,
|
|
0.4778641038326365540L, CHOOSE(9e-17L, 2e-16, 9e-8));
|
|
|
|
result = FUNC(ctanh) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(ctanh(-2 - i 3)) == -0.96538...", __real__ result,
|
|
-0.9653858790221331242L, CHOOSE(2e-19L, 2e-16, 2e-7));
|
|
check_eps ("imag(ctanh(-2 - i 3)) == 0.00988...", __imag__ result,
|
|
0.0098843750383224937L, CHOOSE(7e-20L, 2e-16, 1e-9));
|
|
}
|
|
|
|
|
|
static void
|
|
clog_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_zero, 0));
|
|
check_isinfn_exc ("real(clog(-0 + i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog(-0 + i0)) = pi plus divide-by-zero exception",
|
|
__imag__ result, M_PIl);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check_isinfn_exc ("real(clog(-0 - i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog(-0 - i0)) = -pi plus divide-by-zero exception",
|
|
__imag__ result, -M_PIl);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (0, 0));
|
|
check_isinfn_exc ("real(clog(0 + i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog(0 + i0)) = 0 plus divide-by-zero exception",
|
|
__imag__ result, 0);
|
|
result = FUNC(clog) (BUILD_COMPLEX (0, minus_zero));
|
|
check_isinfn_exc ("real(clog(0 - i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog(0 - i0)) = -0 plus divide-by-zero exception",
|
|
__imag__ result, minus_zero);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp ("real(clog(-Inf + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(-Inf + i Inf)) = 3*pi/4", __imag__ result,
|
|
M_PIl - M_PI_4l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp ("real(clog(-Inf - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(-Inf - i Inf)) = -3*pi/4", __imag__ result,
|
|
M_PI_4l - M_PIl);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp ("real(clog(+Inf + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(+Inf + i Inf)) = pi/4", __imag__ result, M_PI_4l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp ("real(clog(+Inf - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(+Inf - i Inf)) = -pi/4", __imag__ result, -M_PI_4l);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (0, plus_infty));
|
|
check_isinfp ("real(clog(0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (3, plus_infty));
|
|
check_isinfp ("real(clog(3 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(3 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isinfp ("real(clog(-0 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(-0 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (-3, plus_infty));
|
|
check_isinfp ("real(clog(-3 + i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(-3 + i Inf)) = pi/2", __imag__ result, M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (0, minus_infty));
|
|
check_isinfp ("real(clog(0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (3, minus_infty));
|
|
check_isinfp ("real(clog(3 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(3 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isinfp ("real(clog(-0 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(-0 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
result = FUNC(clog) (BUILD_COMPLEX (-3, minus_infty));
|
|
check_isinfp ("real(clog(-3 - i Inf)) = +Inf", __real__ result);
|
|
check ("imag(clog(-3 - i Inf)) = -pi/2", __imag__ result, -M_PI_2l);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, 0));
|
|
check_isinfp ("real(clog(-Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(clog(-Inf + i0)) = pi", __imag__ result, M_PIl);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, 1));
|
|
check_isinfp ("real(clog(-Inf + i1)) = +Inf", __real__ result);
|
|
check ("imag(clog(-Inf + i1)) = pi", __imag__ result, M_PIl);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isinfp ("real(clog(-Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(clog(-Inf - i0)) = -pi", __imag__ result, -M_PIl);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, -1));
|
|
check_isinfp ("real(clog(-Inf - i1)) = +Inf", __real__ result);
|
|
check ("imag(clog(-Inf - i1)) = -pi", __imag__ result, -M_PIl);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, 0));
|
|
check_isinfp ("real(clog(+Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(clog(+Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, 1));
|
|
check_isinfp ("real(clog(+Inf + i1)) = +Inf", __real__ result);
|
|
check ("imag(clog(+Inf + i1)) = 0", __imag__ result, 0);
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(clog(+Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(clog(+Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, -1));
|
|
check_isinfp ("real(clog(+Inf - i1)) = +Inf", __real__ result);
|
|
check ("imag(clog(+Inf - i1)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(clog(+Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isinfp ("real(clog(-Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isinfp ("real(clog(NaN + i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog(NaN + i Inf)) = NaN", __imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isinfp ("real(clog(NaN - i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog(NaN - i Inf)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (0, nan_value));
|
|
check_isnan_maybe_exc ("real(clog(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (3, nan_value));
|
|
check_isnan_maybe_exc ("real(clog(3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan_maybe_exc ("real(clog(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (-3, nan_value));
|
|
check_isnan_maybe_exc ("real(clog(-3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(-3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan_maybe_exc ("real(clog(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, 5));
|
|
check_isnan_maybe_exc ("real(clog(NaN + i5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(NaN + i5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan_maybe_exc ("real(clog(NaN - i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(NaN - i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, -5));
|
|
check_isnan_maybe_exc ("real(clog(NaN - i5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog(NaN - i5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(clog(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(clog(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(clog(0.7 + i 1.2)) == 0.32876...", __real__ result,
|
|
0.3287600014583970919L, CHOOSE(5e-17L, 6e-17, 3e-8));
|
|
check_eps ("imag(clog(0.7 + i 1.2)) == 1.04272...", __imag__ result,
|
|
1.0427218783685369524L, CHOOSE(2e-17L, 0, 0));
|
|
|
|
result = FUNC(clog) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(clog(-2 - i 3)) == 1.28247...", __real__ result,
|
|
1.2824746787307683680L, CHOOSE(3e-19L, 0, 0));
|
|
check_eps ("imag(clog(-2 - i 3)) == -2.15879...", __imag__ result,
|
|
-2.1587989303424641704L, CHOOSE(2e-18L, 5e-16, 8e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
clog10_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_zero, 0));
|
|
check_isinfn_exc ("real(clog10(-0 + i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog10(-0 + i0)) = pi plus divide-by-zero exception",
|
|
__imag__ result, M_PIl);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check_isinfn_exc ("real(clog10(-0 - i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog10(-0 - i0)) = -pi plus divide-by-zero exception",
|
|
__imag__ result, -M_PIl);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (0, 0));
|
|
check_isinfn_exc ("real(clog10(0 + i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog10(0 + i0)) = 0 plus divide-by-zero exception",
|
|
__imag__ result, 0);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (0, minus_zero));
|
|
check_isinfn_exc ("real(clog10(0 - i0)) = -Inf plus divide-by-zero exception",
|
|
__real__ result, DIVIDE_BY_ZERO_EXCEPTION);
|
|
check ("imag(clog10(0 - i0)) = -0 plus divide-by-zero exception",
|
|
__imag__ result, minus_zero);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp ("real(clog10(-Inf + i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-Inf + i Inf)) = 3*pi/4*M_LOG10E", __imag__ result,
|
|
(M_PIl - M_PI_4l) * M_LOG10El, CHOOSE (0, 3e-16, 0));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp ("real(clog10(-Inf - i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-Inf - i Inf)) = -3*pi/4*M_LOG10E", __imag__ result,
|
|
(M_PI_4l - M_PIl) * M_LOG10El, CHOOSE (0, 3e-16, 0));
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp ("real(clog10(+Inf + i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(+Inf + i Inf)) = pi/4*M_LOG10E", __imag__ result,
|
|
M_PI_4l * M_LOG10El, CHOOSE (0, 6e-17, 3e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp ("real(clog10(+Inf - i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(+Inf - i Inf)) = -pi/4*M_LOG10E", __imag__ result,
|
|
-M_PI_4l * M_LOG10El, CHOOSE (0, 6e-17, 3e-8));
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (0, plus_infty));
|
|
check_isinfp ("real(clog10(0 + i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(0 + i Inf)) = pi/2*M_LOG10E", __imag__ result,
|
|
M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (3, plus_infty));
|
|
check_isinfp ("real(clog10(3 + i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(3 + i Inf)) = pi/2*M_LOG10E", __imag__ result,
|
|
M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isinfp ("real(clog10(-0 + i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-0 + i Inf)) = pi/2*M_LOG10E", __imag__ result,
|
|
M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (-3, plus_infty));
|
|
check_isinfp ("real(clog10(-3 + i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-3 + i Inf)) = pi/2*M_LOG10E", __imag__ result,
|
|
M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (0, minus_infty));
|
|
check_isinfp ("real(clog10(0 - i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(0 - i Inf)) = -pi/2*M_LOG10E", __imag__ result,
|
|
-M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (3, minus_infty));
|
|
check_isinfp ("real(clog10(3 - i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(3 - i Inf)) = -pi/2*M_LOG10E", __imag__ result,
|
|
-M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isinfp ("real(clog10(-0 - i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-0 - i Inf)) = -pi/2*M_LOG10E", __imag__ result,
|
|
-M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (-3, minus_infty));
|
|
check_isinfp ("real(clog10(-3 - i Inf)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-3 - i Inf)) = -pi/2*M_LOG10E", __imag__ result,
|
|
-M_PI_2l * M_LOG10El, CHOOSE (0, 2e-16, 6e-8));
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, 0));
|
|
check_isinfp ("real(clog10(-Inf + i0)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-Inf + i0)) = pi*M_LOG10E", __imag__ result,
|
|
M_PIl * M_LOG10El, CHOOSE (0, 3e-16, 2e-7));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, 1));
|
|
check_isinfp ("real(clog10(-Inf + i1)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-Inf + i1)) = pi*M_LOG10E", __imag__ result,
|
|
M_PIl * M_LOG10El, CHOOSE (0, 3e-16, 2e-7));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check_isinfp ("real(clog10(-Inf - i0)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-Inf - i0)) = -pi*M_LOG10E", __imag__ result,
|
|
-M_PIl * M_LOG10El, CHOOSE (0, 3e-16, 2e-7));
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, -1));
|
|
check_isinfp ("real(clog10(-Inf - i1)) = +Inf", __real__ result);
|
|
check_eps ("imag(clog10(-Inf - i1)) = -pi*M_LOG10E", __imag__ result,
|
|
-M_PIl * M_LOG10El, CHOOSE (0, 3e-16, 2e-7));
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, 0));
|
|
check_isinfp ("real(clog10(+Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(clog10(+Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, 1));
|
|
check_isinfp ("real(clog10(+Inf + i1)) = +Inf", __real__ result);
|
|
check ("imag(clog10(+Inf + i1)) = 0", __imag__ result, 0);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(clog10(+Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(clog10(+Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, -1));
|
|
check_isinfp ("real(clog10(+Inf - i1)) = +Inf", __real__ result);
|
|
check ("imag(clog10(+Inf - i1)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(clog10(+Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog10(+Inf + i NaN)) = NaN", __imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isinfp ("real(clog10(-Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog10(-Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, plus_infty));
|
|
check_isinfp ("real(clog10(NaN + i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog10(NaN + i Inf)) = NaN", __imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, minus_infty));
|
|
check_isinfp ("real(clog10(NaN - i Inf)) = +Inf", __real__ result);
|
|
check_isnan ("imag(clog10(NaN - i Inf)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (0, nan_value));
|
|
check_isnan_maybe_exc ("real(clog10(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (3, nan_value));
|
|
check_isnan_maybe_exc ("real(clog10(3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan_maybe_exc ("real(clog10(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (-3, nan_value));
|
|
check_isnan_maybe_exc ("real(clog10(-3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(-3 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan_maybe_exc ("real(clog10(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, 5));
|
|
check_isnan_maybe_exc ("real(clog10(NaN + i5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(NaN + i5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan_maybe_exc ("real(clog10(NaN - i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(NaN - i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, -5));
|
|
check_isnan_maybe_exc ("real(clog10(NaN - i5)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(clog10(NaN - i5)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(clog10(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(clog10(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(clog10(0.7 + i 1.2)) == 0.14277...", __real__ result,
|
|
0.1427786545038868803L, CHOOSE(2e-17L, 6e-17, 2e-8));
|
|
check_eps ("imag(clog10(0.7 + i 1.2)) == 0.45284...", __imag__ result,
|
|
0.4528483579352493248L, CHOOSE(6e-18, 6e-17, 3e-8));
|
|
|
|
result = FUNC(clog10) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(clog10(-2 - i 3)) == 0.55697...", __real__ result,
|
|
0.5569716761534183846L, CHOOSE(6e-20L, 0, 0));
|
|
check_eps ("imag(clog10(-2 - i 3)) == -0.93755...", __imag__ result,
|
|
-0.9375544629863747085L, CHOOSE (7e-19L, 2e-16, 3e-7));
|
|
}
|
|
|
|
|
|
static void
|
|
csqrt_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0, 0));
|
|
check ("real(csqrt(0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(csqrt(0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0, minus_zero));
|
|
check ("real(csqrt(0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(csqrt(0 - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_zero, 0));
|
|
check ("real(csqrt(-0 + i0)) = 0", __real__ result, 0);
|
|
check ("imag(csqrt(-0 + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_zero, minus_zero));
|
|
check ("real(csqrt(-0 - i0)) = 0", __real__ result, 0);
|
|
check ("imag(csqrt(-0 - i0)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, 0));
|
|
check ("real(csqrt(-Inf + i0)) = 0", __real__ result, 0);
|
|
check_isinfp ("imag(csqrt(-Inf + i0)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, 6));
|
|
check ("real(csqrt(-Inf + i6)) = 0", __real__ result, 0);
|
|
check_isinfp ("imag(csqrt(-Inf + i6)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, minus_zero));
|
|
check ("real(csqrt(-Inf - i0)) = 0", __real__ result, 0);
|
|
check_isinfn ("imag(csqrt(-Inf - i0)) = -Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, -6));
|
|
check ("real(csqrt(-Inf - i6)) = 0", __real__ result, 0);
|
|
check_isinfn ("imag(csqrt(-Inf - i6)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, 0));
|
|
check_isinfp ("real(csqrt(+Inf + i0)) = +Inf", __real__ result);
|
|
check ("imag(csqrt(+Inf + i0)) = 0", __imag__ result, 0);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, 6));
|
|
check_isinfp ("real(csqrt(+Inf + i6)) = +Inf", __real__ result);
|
|
check ("imag(csqrt(+Inf + i6)) = 0", __imag__ result, 0);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, minus_zero));
|
|
check_isinfp ("real(csqrt(+Inf - i0)) = +Inf", __real__ result);
|
|
check ("imag(csqrt(+Inf - i0)) = -0", __imag__ result, minus_zero);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, -6));
|
|
check_isinfp ("real(csqrt(+Inf - i6)) = +Inf", __real__ result);
|
|
check ("imag(csqrt(+Inf - i6)) = -0", __imag__ result, minus_zero);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0, plus_infty));
|
|
check_isinfp ("real(csqrt(0 + i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csqrt(0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (4, plus_infty));
|
|
check_isinfp ("real(csqrt(4 + i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csqrt(4 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, plus_infty));
|
|
check_isinfp ("real(csqrt(+Inf + i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csqrt(+Inf + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_zero, plus_infty));
|
|
check_isinfp ("real(csqrt(-0 + i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csqrt(-0 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (-4, plus_infty));
|
|
check_isinfp ("real(csqrt(-4 + i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csqrt(-4 + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, plus_infty));
|
|
check_isinfp ("real(csqrt(-Inf + i Inf)) = +Inf", __real__ result);
|
|
check_isinfp ("imag(csqrt(-Inf + i Inf)) = +Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0, minus_infty));
|
|
check_isinfp ("real(csqrt(0 - i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csqrt(0 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (4, minus_infty));
|
|
check_isinfp ("real(csqrt(4 - i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csqrt(4 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, minus_infty));
|
|
check_isinfp ("real(csqrt(+Inf - i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csqrt(+Inf - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_zero, minus_infty));
|
|
check_isinfp ("real(csqrt(-0 - i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csqrt(-0 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (-4, minus_infty));
|
|
check_isinfp ("real(csqrt(-4 - i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csqrt(-4 - i Inf)) = -Inf", __imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, minus_infty));
|
|
check_isinfp ("real(csqrt(-Inf - i Inf)) = +Inf", __real__ result);
|
|
check_isinfn ("imag(csqrt(-Inf - i Inf)) = -Inf", __imag__ result);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_infty, nan_value));
|
|
check_isnan ("real(csqrt(-Inf + i NaN)) = NaN", __real__ result);
|
|
check_isinfp ("imag(csqrt(-Inf + i NaN)) = +-Inf",
|
|
FUNC(fabs) (__imag__ result));
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (plus_infty, nan_value));
|
|
check_isinfp ("real(csqrt(+Inf + i NaN)) = +Inf", __real__ result);
|
|
check_isnan ("imag(csqrt(+Inf + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0, nan_value));
|
|
check_isnan_maybe_exc ("real(csqrt(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (1, nan_value));
|
|
check_isnan_maybe_exc ("real(csqrt(1 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(1 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (minus_zero, nan_value));
|
|
check_isnan_maybe_exc ("real(csqrt(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(-0 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (-1, nan_value));
|
|
check_isnan_maybe_exc ("real(csqrt(-1 + i NaN)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(-1 + i NaN)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (nan_value, 0));
|
|
check_isnan_maybe_exc ("real(csqrt(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(NaN + i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (nan_value, 8));
|
|
check_isnan_maybe_exc ("real(csqrt(NaN + i8)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(NaN + i8)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (nan_value, minus_zero));
|
|
check_isnan_maybe_exc ("real(csqrt(NaN - i0)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(NaN - i0)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (nan_value, -8));
|
|
check_isnan_maybe_exc ("real(csqrt(NaN - i8)) = NaN plus maybe invalid exception",
|
|
__real__ result, INVALID_EXCEPTION);
|
|
check_isnan ("imag(csqrt(NaN - i8)) = NaN plus maybe invalid exception",
|
|
__imag__ result);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (nan_value, nan_value));
|
|
check_isnan ("real(csqrt(NaN + i NaN)) = NaN", __real__ result);
|
|
check_isnan ("imag(csqrt(NaN + i NaN)) = NaN", __imag__ result);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (16.0, -30.0));
|
|
check ("real(csqrt(16 - 30i)) = 5", __real__ result, 5.0);
|
|
check ("imag(csqrt(16 - 30i)) = -3", __imag__ result, -3.0);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (-1, 0));
|
|
check ("real(csqrt(1 + i0) = 0", __real__ result, 0);
|
|
check ("imag(csqrt(1 + i0) = 1", __imag__ result, 1);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0, 2));
|
|
check ("real(csqrt(0 + i 2) = 1", __real__ result, 1);
|
|
check ("imag(csqrt(0 + i 2) = 1", __imag__ result, 1);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (119, 120));
|
|
check ("real(csqrt(119 + i 120) = 12", __real__ result, 12);
|
|
check ("imag(csqrt(119 + i 120) = 5", __imag__ result, 5);
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (0.7, 1.2));
|
|
check_eps ("real(csqrt(0.7 + i 1.2)) == 1.02206...", __real__ result,
|
|
1.0220676100300264507L, CHOOSE(3e-17L, 3e-16, 2e-7));
|
|
check_eps ("imag(csqrt(0.7 + i 1.2)) == 0.58704...", __imag__ result,
|
|
0.5870453129635652115L, CHOOSE(7e-18L, 0, 6e-8));
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (-2, -3));
|
|
check_eps ("real(csqrt(-2 - i 3)) == 0.89597...", __real__ result,
|
|
0.8959774761298381247L, CHOOSE(6e-16L, 4e-16, 6e-8));
|
|
check_eps ("imag(csqrt(-2 - i 3)) == -1.67414...", __imag__ result,
|
|
-1.6741492280355400404L, CHOOSE(0, 5e-16, 0));
|
|
|
|
result = FUNC(csqrt) (BUILD_COMPLEX (-2, 3));
|
|
check_eps ("real(csqrt(-2 + i 3)) == 0.89597...", __real__ result,
|
|
0.8959774761298381247L, CHOOSE(6e-20L, 4e-16, 6e-8));
|
|
check_eps ("imag(csqrt(-2 + i 3)) == 1.67414...", __imag__ result,
|
|
1.6741492280355400404L, CHOOSE(0, 5e-16, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
cpow_test (void)
|
|
{
|
|
__complex__ MATHTYPE result;
|
|
|
|
result = FUNC (cpow) (BUILD_COMPLEX (1, 0), BUILD_COMPLEX (0, 0));
|
|
check ("real(cpow (1 + i0), (0 + i0)) == 0", __real__ result, 1);
|
|
check ("imag(cpow (1 + i0), (0 + i0)) == 0", __imag__ result, 0);
|
|
|
|
result = FUNC (cpow) (BUILD_COMPLEX (2, 0), BUILD_COMPLEX (10, 0));
|
|
check_eps ("real(cpow (2 + i0), (10 + i0)) == 1024", __real__ result, 1024,
|
|
CHOOSE (2e-16L, 0, 0));
|
|
check ("imag(cpow (2 + i0), (10 + i0)) == 0", __imag__ result, 0);
|
|
|
|
result = FUNC (cpow) (BUILD_COMPLEX (M_El, 0), BUILD_COMPLEX (0, 2 * M_PIl));
|
|
check_eps ("real(cpow (e + i0), (0 + i 2*PI)) == 1", __real__ result, 1,
|
|
CHOOSE (0, 0, 6e-8));
|
|
check_eps ("imag(cpow (e + i0), (0 + i 2*PI)) == 0", __imag__ result, 0,
|
|
CHOOSE (3e-18L, 3e-16, 4e-7));
|
|
|
|
result = FUNC (cpow) (BUILD_COMPLEX (2, 3), BUILD_COMPLEX (4, 0));
|
|
check_eps ("real(cpow (2 + i3), (4 + i0)) == -119", __real__ result, -119,
|
|
CHOOSE (9e-16L, 2e-14, 4e-5));
|
|
check_eps ("imag(cpow (2 + i3), (4 + i0)) == -120", __imag__ result, -120,
|
|
CHOOSE (1e-15L, 0, 5e-5));
|
|
}
|
|
|
|
|
|
static void
|
|
cabs_test (void)
|
|
{
|
|
/* cabs (x + iy) is specified as hypot (x,y) */
|
|
MATHTYPE a;
|
|
a = random_greater (0);
|
|
check_isinfp_ext ("cabs (+inf + i x) == +inf",
|
|
FUNC(cabs) (BUILD_COMPLEX (plus_infty, a)), a);
|
|
check_isinfp_ext ("cabs (-inf + i x) == +inf",
|
|
FUNC(cabs) (BUILD_COMPLEX (minus_infty, a)), a);
|
|
|
|
check_isinfp ("cabs (+inf+ iNaN) == +inf",
|
|
FUNC(cabs) (BUILD_COMPLEX(minus_infty, nan_value)));
|
|
check_isinfp ("cabs (-inf+ iNaN) == +inf",
|
|
FUNC(cabs) (BUILD_COMPLEX(minus_infty, nan_value)));
|
|
|
|
check_isnan ("cabs (NaN+ iNaN) == NaN",
|
|
FUNC(cabs) (BUILD_COMPLEX(nan_value, nan_value)));
|
|
|
|
a = FUNC(cabs) (BUILD_COMPLEX (12.4L, 0.7L));
|
|
check ("cabs (x,y) == cabs (y,x)",
|
|
FUNC(cabs) (BUILD_COMPLEX(0.7L, 12.4L)), a);
|
|
check ("cabs (x,y) == cabs (-x,y)",
|
|
FUNC(cabs) (BUILD_COMPLEX(-12.4L, 0.7L)), a);
|
|
check ("cabs (x,y) == cabs (-y,x)",
|
|
FUNC(cabs) (BUILD_COMPLEX(-0.7L, 12.4L)), a);
|
|
check ("cabs (x,y) == cabs (-x,-y)",
|
|
FUNC(cabs) (BUILD_COMPLEX(-12.4L, -0.7L)), a);
|
|
check ("cabs (x,y) == cabs (-y,-x)",
|
|
FUNC(cabs) (BUILD_COMPLEX(-0.7L, -12.4L)), a);
|
|
check ("cabs (x,0) == fabs (x)", FUNC(cabs) (BUILD_COMPLEX(-0.7L, 0)), 0.7L);
|
|
check ("cabs (x,0) == fabs (x)", FUNC(cabs) (BUILD_COMPLEX(0.7L, 0)), 0.7L);
|
|
check ("cabs (x,0) == fabs (x)", FUNC(cabs) (BUILD_COMPLEX(-1.0L, 0)), 1.0L);
|
|
check ("cabs (x,0) == fabs (x)", FUNC(cabs) (BUILD_COMPLEX(1.0L, 0)), 1.0L);
|
|
check ("cabs (x,0) == fabs (x)", FUNC(cabs) (BUILD_COMPLEX(-5.7e7L, 0)),
|
|
5.7e7L);
|
|
check ("cabs (x,0) == fabs (x)", FUNC(cabs) (BUILD_COMPLEX(5.7e7L, 0)),
|
|
5.7e7L);
|
|
|
|
check_eps ("cabs (0.7 + i 1.2) == 1.38924...", FUNC(cabs) (BUILD_COMPLEX(0.7, 1.2)),
|
|
1.3892443989449804508L, CHOOSE(7e-17L, 3e-16, 0));
|
|
}
|
|
|
|
|
|
static void
|
|
carg_test (void)
|
|
{
|
|
/* carg (x + iy) is specified as atan2 (y, x) */
|
|
MATHTYPE x;
|
|
|
|
x = random_greater (0);
|
|
check ("carg (x + i 0) == 0 for x > 0",
|
|
FUNC(carg) (BUILD_COMPLEX(x, 0)), 0);
|
|
x = random_greater (0);
|
|
check ("carg (x - i 0) == -0 for x > 0",
|
|
FUNC(carg) (BUILD_COMPLEX(x, minus_zero)), minus_zero);
|
|
|
|
check ("carg (+0 + i 0) == +0", FUNC(carg) (BUILD_COMPLEX(0, 0)), 0);
|
|
check ("carg (+0 - i 0) == -0", FUNC(carg) (BUILD_COMPLEX(0, minus_zero)),
|
|
minus_zero);
|
|
|
|
x = -random_greater (0);
|
|
check ("carg (x + i 0) == +pi for x < 0", FUNC(carg) (BUILD_COMPLEX(x, 0)),
|
|
M_PIl);
|
|
|
|
x = -random_greater (0);
|
|
check ("carg (x - i 0) == -pi for x < 0",
|
|
FUNC(carg) (BUILD_COMPLEX(x, minus_zero)), -M_PIl);
|
|
|
|
check ("carg (-0 + i 0) == +pi", FUNC(carg) (BUILD_COMPLEX(minus_zero, 0)),
|
|
M_PIl);
|
|
check ("carg (-0 - i 0) == -pi",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_zero, minus_zero)), -M_PIl);
|
|
|
|
x = random_greater (0);
|
|
check ("carg (+0 + i y) == pi/2 for y > 0", FUNC(carg) (BUILD_COMPLEX(0, x)),
|
|
M_PI_2l);
|
|
|
|
x = random_greater (0);
|
|
check ("carg (-0 + i y) == pi/2 for y > 0",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_zero, x)), M_PI_2l);
|
|
|
|
x = random_less (0);
|
|
check ("carg (+0 + i y) == -pi/2 for y < 0", FUNC(carg) (BUILD_COMPLEX(0, x)),
|
|
-M_PI_2l);
|
|
|
|
x = random_less (0);
|
|
check ("carg (-0 + i y) == -pi/2 for y < 0",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_zero, x)), -M_PI_2l);
|
|
|
|
x = random_greater (0);
|
|
check ("carg (inf + i y) == +0 for finite y > 0",
|
|
FUNC(carg) (BUILD_COMPLEX(plus_infty, x)), 0);
|
|
|
|
x = -random_greater (0);
|
|
check ("carg (inf + i y) == -0 for finite y < 0",
|
|
FUNC(carg) (BUILD_COMPLEX(plus_infty, x)), minus_zero);
|
|
|
|
x = random_value (-1e4, 1e4);
|
|
check ("carg(x + i inf) == pi/2 for finite x",
|
|
FUNC(carg) (BUILD_COMPLEX(x, plus_infty)), M_PI_2l);
|
|
|
|
x = random_value (-1e4, 1e4);
|
|
check ("carg(x - i inf) == -pi/2 for finite x",
|
|
FUNC(carg) (BUILD_COMPLEX(x, minus_infty)), -M_PI_2l);
|
|
|
|
x = random_greater (0);
|
|
check ("carg (-inf + i y) == +pi for finite y > 0",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_infty, x)), M_PIl);
|
|
|
|
x = -random_greater (0);
|
|
check ("carg (-inf + i y) == -pi for finite y < 0",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_infty, x)), -M_PIl);
|
|
|
|
check ("carg (+inf + i inf) == +pi/4",
|
|
FUNC(carg) (BUILD_COMPLEX(plus_infty, plus_infty)), M_PI_4l);
|
|
|
|
check ("carg (+inf -i inf) == -pi/4",
|
|
FUNC(carg) (BUILD_COMPLEX(plus_infty, minus_infty)), -M_PI_4l);
|
|
|
|
check ("carg (-inf +i inf) == +3*pi/4",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_infty, plus_infty)), 3 * M_PI_4l);
|
|
|
|
check ("carg (-inf -i inf) == -3*pi/4",
|
|
FUNC(carg) (BUILD_COMPLEX(minus_infty, minus_infty)), -3 * M_PI_4l);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
nearbyint_test (void)
|
|
{
|
|
check ("nearbyint(+0) = 0", FUNC(nearbyint) (0.0), 0.0);
|
|
check ("nearbyint(-0) = -0", FUNC(nearbyint) (minus_zero), minus_zero);
|
|
check_isinfp ("nearbyint(+Inf) = +Inf", FUNC(nearbyint) (plus_infty));
|
|
check_isinfn ("nearbyint(-Inf) = -Inf", FUNC(nearbyint) (minus_infty));
|
|
}
|
|
|
|
|
|
static void
|
|
rint_test (void)
|
|
{
|
|
check ("rint(0) = 0", FUNC(rint) (0.0), 0.0);
|
|
check ("rint(-0) = -0", FUNC(rint) (minus_zero), minus_zero);
|
|
check_isinfp ("rint(+Inf) = +Inf", FUNC(rint) (plus_infty));
|
|
check_isinfn ("rint(-Inf) = -Inf", FUNC(rint) (minus_infty));
|
|
}
|
|
|
|
|
|
static void
|
|
lrint_test (void)
|
|
{
|
|
/* XXX this test is incomplete. We need to have a way to specifiy
|
|
the rounding method and test the critical cases. So far, only
|
|
unproblematic numbers are tested. */
|
|
|
|
check_long ("lrint(0) = 0", FUNC(lrint) (0.0), 0);
|
|
check_long ("lrint(-0) = 0", FUNC(lrint) (minus_zero), 0);
|
|
check_long ("lrint(0.2) = 0", FUNC(lrint) (0.2), 0);
|
|
check_long ("lrint(-0.2) = 0", FUNC(lrint) (-0.2), 0);
|
|
|
|
check_long ("lrint(1.4) = 1", FUNC(lrint) (1.4), 1);
|
|
check_long ("lrint(-1.4) = -1", FUNC(lrint) (-1.4), -1);
|
|
|
|
check_long ("lrint(8388600.3) = 8388600", FUNC(lrint) (8388600.3), 8388600);
|
|
check_long ("lrint(-8388600.3) = -8388600", FUNC(lrint) (-8388600.3),
|
|
-8388600);
|
|
}
|
|
|
|
|
|
static void
|
|
llrint_test (void)
|
|
{
|
|
/* XXX this test is incomplete. We need to have a way to specifiy
|
|
the rounding method and test the critical cases. So far, only
|
|
unproblematic numbers are tested. */
|
|
|
|
check_longlong ("llrint(0) = 0", FUNC(llrint) (0.0), 0);
|
|
check_longlong ("llrint(-0) = 0", FUNC(llrint) (minus_zero), 0);
|
|
check_longlong ("llrint(0.2) = 0", FUNC(llrint) (0.2), 0);
|
|
check_longlong ("llrint(-0.2) = 0", FUNC(llrint) (-0.2), 0);
|
|
|
|
check_longlong ("llrint(1.4) = 1", FUNC(llrint) (1.4), 1);
|
|
check_longlong ("llrint(-1.4) = -1", FUNC(llrint) (-1.4), -1);
|
|
|
|
check_longlong ("llrint(8388600.3) = 8388600", FUNC(llrint) (8388600.3),
|
|
8388600);
|
|
check_longlong ("llrint(-8388600.3) = -8388600", FUNC(llrint) (-8388600.3),
|
|
-8388600);
|
|
|
|
/* Test boundary conditions. */
|
|
/* 0x1FFFFF */
|
|
check_longlong ("llrint(2097151.0) = 2097151", FUNC(llrint) (2097151.0),
|
|
2097151LL);
|
|
/* 0x800000 */
|
|
check_longlong ("llrint(8388608.0) = 8388608", FUNC(llrint) (8388608.0),
|
|
8388608LL);
|
|
/* 0x1000000 */
|
|
check_longlong ("llrint(16777216.0) = 16777216",
|
|
FUNC(llrint) (16777216.0), 16777216LL);
|
|
/* 0x20000000000 */
|
|
check_longlong ("llrint(2199023255552.0) = 2199023255552",
|
|
FUNC(llrint) (2199023255552.0), 2199023255552LL);
|
|
/* 0x40000000000 */
|
|
check_longlong ("llrint(4398046511104.0) = 4398046511104",
|
|
FUNC(llrint) (4398046511104.0), 4398046511104LL);
|
|
/* 0x10000000000000 */
|
|
check_longlong ("llrint(4503599627370496.0) = 4503599627370496",
|
|
FUNC(llrint) (4503599627370496.0), 4503599627370496LL);
|
|
/* 0x10000080000000 */
|
|
check_longlong ("llrint(4503601774854144.0) = 4503601774854144",
|
|
FUNC(llrint) (4503601774854144.0), 4503601774854144LL);
|
|
/* 0x20000000000000 */
|
|
check_longlong ("llrint(9007199254740992.0) = 9007199254740992",
|
|
FUNC(llrint) (9007199254740992.0), 9007199254740992LL);
|
|
/* 0x80000000000000 */
|
|
check_longlong ("llrint(36028797018963968.0) = 36028797018963968",
|
|
FUNC(llrint) (36028797018963968.0), 36028797018963968LL);
|
|
/* 0x100000000000000 */
|
|
check_longlong ("llrint(72057594037927936.0) = 72057594037927936",
|
|
FUNC(llrint) (72057594037927936.0), 72057594037927936LL);
|
|
}
|
|
|
|
|
|
static void
|
|
round_test (void)
|
|
{
|
|
check ("round(0) = 0", FUNC(round) (0), 0);
|
|
check ("round(-0) = -0", FUNC(round) (minus_zero), minus_zero);
|
|
check ("round(0.2) = 0", FUNC(round) (0.2), 0.0);
|
|
check ("round(-0.2) = -0", FUNC(round) (-0.2), minus_zero);
|
|
check ("round(0.5) = 1", FUNC(round) (0.5), 1.0);
|
|
check ("round(-0.5) = -1", FUNC(round) (-0.5), -1.0);
|
|
check ("round(0.8) = 1", FUNC(round) (0.8), 1.0);
|
|
check ("round(-0.8) = -1", FUNC(round) (-0.8), -1.0);
|
|
check ("round(1.5) = 2", FUNC(round) (1.5), 2.0);
|
|
check ("round(-1.5) = -2", FUNC(round) (-1.5), -2.0);
|
|
check ("round(2097152.5) = 2097153", FUNC(round) (2097152.5), 2097153);
|
|
check ("round(-2097152.5) = -2097153", FUNC(round) (-2097152.5), -2097153);
|
|
}
|
|
|
|
|
|
static void
|
|
lround_test (void)
|
|
{
|
|
check_long ("lround(0) = 0", FUNC(lround) (0), 0);
|
|
check_long ("lround(-0) = 0", FUNC(lround) (minus_zero), 0);
|
|
check_long ("lround(0.2) = 0", FUNC(lround) (0.2), 0.0);
|
|
check_long ("lround(-0.2) = 0", FUNC(lround) (-0.2), 0);
|
|
check_long ("lround(0.5) = 1", FUNC(lround) (0.5), 1);
|
|
check_long ("lround(-0.5) = -1", FUNC(lround) (-0.5), -1);
|
|
check_long ("lround(0.8) = 1", FUNC(lround) (0.8), 1);
|
|
check_long ("lround(-0.8) = -1", FUNC(lround) (-0.8), -1);
|
|
check_long ("lround(1.5) = 2", FUNC(lround) (1.5), 2);
|
|
check_long ("lround(-1.5) = -2", FUNC(lround) (-1.5), -2);
|
|
check_long ("lround(22514.5) = 22515", FUNC(lround) (22514.5), 22515);
|
|
check_long ("lround(-22514.5) = -22515", FUNC(lround) (-22514.5), -22515);
|
|
#ifndef TEST_FLOAT
|
|
check_long ("lround(2097152.5) = 2097153", FUNC(lround) (2097152.5),
|
|
2097153);
|
|
check_long ("lround(-2097152.5) = -2097153", FUNC(lround) (-2097152.5),
|
|
-2097153);
|
|
#endif
|
|
}
|
|
|
|
|
|
static void
|
|
llround_test (void)
|
|
{
|
|
check_longlong ("llround(0) = 0", FUNC(llround) (0), 0);
|
|
check_longlong ("llround(-0) = 0", FUNC(llround) (minus_zero), 0);
|
|
check_longlong ("llround(0.2) = 0", FUNC(llround) (0.2), 0.0);
|
|
check_longlong ("llround(-0.2) = 0", FUNC(llround) (-0.2), 0);
|
|
check_longlong ("llround(0.5) = 1", FUNC(llround) (0.5), 1);
|
|
check_longlong ("llround(-0.5) = -1", FUNC(llround) (-0.5), -1);
|
|
check_longlong ("llround(0.8) = 1", FUNC(llround) (0.8), 1);
|
|
check_longlong ("llround(-0.8) = -1", FUNC(llround) (-0.8), -1);
|
|
check_longlong ("llround(1.5) = 2", FUNC(llround) (1.5), 2);
|
|
check_longlong ("llround(-1.5) = -2", FUNC(llround) (-1.5), -2);
|
|
check_longlong ("llround(22514.5) = 22515", FUNC(llround) (22514.5), 22515);
|
|
check_longlong ("llround(-22514.5) = -22515", FUNC(llround) (-22514.5),
|
|
-22515);
|
|
#ifndef TEST_FLOAT
|
|
check_longlong ("llround(2097152.5) = 2097153",
|
|
FUNC(llround) (2097152.5), 2097153);
|
|
check_longlong ("llround(-2097152.5) = -2097153",
|
|
FUNC(llround) (-2097152.5), -2097153);
|
|
check_longlong ("llround(34359738368.5) = 34359738369",
|
|
FUNC(llround) (34359738368.5), 34359738369ll);
|
|
check_longlong ("llround(-34359738368.5) = -34359738369",
|
|
FUNC(llround) (-34359738368.5), -34359738369ll);
|
|
#endif
|
|
|
|
/* Test boundary conditions. */
|
|
/* 0x1FFFFF */
|
|
check_longlong ("llround(2097151.0) = 2097151", FUNC(llround) (2097151.0),
|
|
2097151LL);
|
|
/* 0x800000 */
|
|
check_longlong ("llround(8388608.0) = 8388608", FUNC(llround) (8388608.0),
|
|
8388608LL);
|
|
/* 0x1000000 */
|
|
check_longlong ("llround(16777216.0) = 16777216",
|
|
FUNC(llround) (16777216.0), 16777216LL);
|
|
/* 0x20000000000 */
|
|
check_longlong ("llround(2199023255552.0) = 2199023255552",
|
|
FUNC(llround) (2199023255552.0), 2199023255552LL);
|
|
/* 0x40000000000 */
|
|
check_longlong ("llround(4398046511104.0) = 4398046511104",
|
|
FUNC(llround) (4398046511104.0), 4398046511104LL);
|
|
/* 0x10000000000000 */
|
|
check_longlong ("llround(4503599627370496.0) = 4503599627370496",
|
|
FUNC(llround) (4503599627370496.0), 4503599627370496LL);
|
|
/* 0x10000080000000 */
|
|
check_longlong ("llrint(4503601774854144.0) = 4503601774854144",
|
|
FUNC(llrint) (4503601774854144.0), 4503601774854144LL);
|
|
/* 0x20000000000000 */
|
|
check_longlong ("llround(9007199254740992.0) = 9007199254740992",
|
|
FUNC(llround) (9007199254740992.0), 9007199254740992LL);
|
|
/* 0x80000000000000 */
|
|
check_longlong ("llround(36028797018963968.0) = 36028797018963968",
|
|
FUNC(llround) (36028797018963968.0), 36028797018963968LL);
|
|
/* 0x100000000000000 */
|
|
check_longlong ("llround(72057594037927936.0) = 72057594037927936",
|
|
FUNC(llround) (72057594037927936.0), 72057594037927936LL);
|
|
}
|
|
|
|
|
|
static void
|
|
fma_test (void)
|
|
{
|
|
check ("fma(1.0, 2.0, 3.0) = 5.0", FUNC(fma) (1.0, 2.0, 3.0), 5.0);
|
|
check_isnan ("fma(NaN, 2.0, 3.0) = NaN", FUNC(fma) (nan_value, 2.0, 3.0));
|
|
check_isnan ("fma(1.0, NaN, 3.0) = NaN", FUNC(fma) (1.0, nan_value, 3.0));
|
|
check_isnan_maybe_exc ("fma(1.0, 2.0, NaN) = NaN",
|
|
FUNC(fma) (1.0, 2.0, nan_value), INVALID_EXCEPTION);
|
|
check_isnan_maybe_exc ("fma(+Inf, 0.0, NaN) = NaN",
|
|
FUNC(fma) (plus_infty, 0.0, nan_value),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_maybe_exc ("fma(-Inf, 0.0, NaN) = NaN",
|
|
FUNC(fma) (minus_infty, 0.0, nan_value),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_maybe_exc ("fma(0.0, +Inf, NaN) = NaN",
|
|
FUNC(fma) (0.0, plus_infty, nan_value),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_maybe_exc ("fma(0.0, -Inf, NaN) = NaN",
|
|
FUNC(fma) (0.0, minus_infty, nan_value),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(+Inf, 0.0, 1.0) = NaN",
|
|
FUNC(fma) (plus_infty, 0.0, 1.0), INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(-Inf, 0.0, 1.0) = NaN",
|
|
FUNC(fma) (minus_infty, 0.0, 1.0), INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(0.0, +Inf, 1.0) = NaN",
|
|
FUNC(fma) (0.0, plus_infty, 1.0), INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(0.0, -Inf, 1.0) = NaN",
|
|
FUNC(fma) (0.0, minus_infty, 1.0), INVALID_EXCEPTION);
|
|
|
|
check_isnan_exc ("fma(+Inf, +Inf, -Inf) = NaN",
|
|
FUNC(fma) (plus_infty, plus_infty, minus_infty),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(-Inf, +Inf, +Inf) = NaN",
|
|
FUNC(fma) (minus_infty, plus_infty, plus_infty),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(+Inf, -Inf, +Inf) = NaN",
|
|
FUNC(fma) (plus_infty, minus_infty, plus_infty),
|
|
INVALID_EXCEPTION);
|
|
check_isnan_exc ("fma(-Inf, -Inf, -Inf) = NaN",
|
|
FUNC(fma) (minus_infty, minus_infty, minus_infty),
|
|
INVALID_EXCEPTION);
|
|
}
|
|
|
|
|
|
/*
|
|
Tests for the comparison macros
|
|
*/
|
|
typedef enum {is_less, is_equal, is_greater, is_unordered} comp_result;
|
|
|
|
|
|
static void
|
|
comparison2_test (MATHTYPE x, MATHTYPE y, comp_result comp)
|
|
{
|
|
char buf[255];
|
|
int result;
|
|
int expected;
|
|
|
|
expected = (comp == is_greater);
|
|
sprintf (buf, "isgreater (%" PRINTF_EXPR ", %" PRINTF_EXPR ") == %d", x, y,
|
|
expected);
|
|
result = (isgreater (x, y) == expected);
|
|
check_bool (buf, result);
|
|
|
|
expected = (comp == is_greater || comp == is_equal);
|
|
sprintf (buf, "isgreaterequal (%" PRINTF_EXPR ", %" PRINTF_EXPR ") == %d", x, y,
|
|
expected);
|
|
result = (isgreaterequal (x, y) == expected);
|
|
check_bool (buf, result);
|
|
|
|
expected = (comp == is_less);
|
|
sprintf (buf, "isless (%" PRINTF_EXPR ", %" PRINTF_EXPR ") == %d", x, y,
|
|
expected);
|
|
result = (isless (x, y) == expected);
|
|
check_bool (buf, result);
|
|
|
|
expected = (comp == is_less || comp == is_equal);
|
|
sprintf (buf, "islessequal (%" PRINTF_EXPR ", %" PRINTF_EXPR ") == %d", x, y,
|
|
expected);
|
|
result = (islessequal (x, y) == expected);
|
|
check_bool (buf, result);
|
|
|
|
expected = (comp == is_greater || comp == is_less);
|
|
sprintf (buf, "islessgreater (%" PRINTF_EXPR ", %" PRINTF_EXPR ") == %d", x, y,
|
|
expected);
|
|
result = (islessgreater (x, y) == expected);
|
|
check_bool (buf, result);
|
|
|
|
expected = (comp == is_unordered);
|
|
sprintf (buf, "isunordered (%" PRINTF_EXPR ", %" PRINTF_EXPR ") == %d", x, y,
|
|
expected);
|
|
result = (isunordered (x, y) == expected);
|
|
check_bool (buf, result);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
comparison1_test (MATHTYPE x, MATHTYPE y, comp_result comp)
|
|
{
|
|
comp_result comp_swap;
|
|
switch (comp)
|
|
{
|
|
case is_less:
|
|
comp_swap = is_greater;
|
|
break;
|
|
case is_greater:
|
|
comp_swap = is_less;
|
|
break;
|
|
default:
|
|
comp_swap = comp;
|
|
break;
|
|
}
|
|
comparison2_test (x, y, comp);
|
|
comparison2_test (y, x, comp_swap);
|
|
}
|
|
|
|
|
|
static void
|
|
comparisons_test (void)
|
|
{
|
|
comparison1_test (1, 2, is_less);
|
|
comparison1_test (-30, 30, is_less);
|
|
comparison1_test (42, 42, is_equal);
|
|
comparison1_test (1, plus_infty, is_less);
|
|
comparison1_test (35, minus_infty, is_greater);
|
|
comparison1_test (1, nan_value, is_unordered);
|
|
comparison1_test (nan_value, nan_value, is_unordered);
|
|
comparison1_test (plus_infty, nan_value, is_unordered);
|
|
comparison1_test (minus_infty, nan_value, is_unordered);
|
|
comparison1_test (plus_infty, minus_infty, is_greater);
|
|
}
|
|
|
|
|
|
static void
|
|
inverse_func_pair_test (const char *test_name,
|
|
mathfunc f1, mathfunc inverse,
|
|
MATHTYPE x, MATHTYPE epsilon)
|
|
{
|
|
MATHTYPE a, b, difference;
|
|
int result;
|
|
|
|
a = f1 (x);
|
|
(void) &a;
|
|
b = inverse (a);
|
|
(void) &b;
|
|
|
|
output_new_test (test_name);
|
|
result = check_equal (b, x, epsilon, &difference);
|
|
output_result (test_name, result,
|
|
b, x, difference, PRINT, PRINT);
|
|
}
|
|
|
|
|
|
static void
|
|
inverse_functions (void)
|
|
{
|
|
inverse_func_pair_test ("asin(sin(x)) == x",
|
|
FUNC(sin), FUNC(asin), 1.0,
|
|
CHOOSE (2e-18L, 0, 3e-7L));
|
|
inverse_func_pair_test ("sin(asin(x)) == x",
|
|
FUNC(asin), FUNC(sin), 1.0, 0.0);
|
|
|
|
inverse_func_pair_test ("acos(cos(x)) == x",
|
|
FUNC(cos), FUNC(acos), 1.0,
|
|
CHOOSE (4e-18L, 1e-15L, 0));
|
|
inverse_func_pair_test ("cos(acos(x)) == x",
|
|
FUNC(acos), FUNC(cos), 1.0, 0.0);
|
|
inverse_func_pair_test ("atan(tan(x)) == x",
|
|
FUNC(tan), FUNC(atan), 1.0, CHOOSE (2e-18L, 0, 0));
|
|
inverse_func_pair_test ("tan(atan(x)) == x",
|
|
FUNC(atan), FUNC(tan), 1.0,
|
|
CHOOSE (2e-18L, 1e-15L, 2e-7));
|
|
|
|
inverse_func_pair_test ("asinh(sinh(x)) == x",
|
|
FUNC(sinh), FUNC(asinh), 1.0, CHOOSE (1e-18L, 0, 1e-7));
|
|
inverse_func_pair_test ("sinh(asinh(x)) == x",
|
|
FUNC(asinh), FUNC(sinh), 1.0,
|
|
CHOOSE (2e-18L, 2e-16L, 2e-7));
|
|
|
|
inverse_func_pair_test ("acosh(cosh(x)) == x",
|
|
FUNC(cosh), FUNC(acosh), 1.0,
|
|
CHOOSE (1e-18L, 1e-15L, 6e-8));
|
|
inverse_func_pair_test ("cosh(acosh(x)) == x",
|
|
FUNC(acosh), FUNC(cosh), 1.0, 0.0);
|
|
|
|
inverse_func_pair_test ("atanh(tanh(x)) == x",
|
|
FUNC(tanh), FUNC(atanh), 1.0, CHOOSE (1e-18L, 1e-15L, 0));
|
|
inverse_func_pair_test ("tanh(atanh(x)) == x",
|
|
FUNC(atanh), FUNC(tanh), 1.0, 0.0);
|
|
|
|
}
|
|
|
|
/* Test sin and cos with the identity: sin(x)^2 + cos(x)^2 = 1. */
|
|
static void
|
|
identities1_test (MATHTYPE x, MATHTYPE epsilon)
|
|
{
|
|
MATHTYPE res1, res2, res3, diff;
|
|
int result;
|
|
|
|
res1 = FUNC(sin) (x);
|
|
(void) &res1;
|
|
res2 = FUNC(cos) (x);
|
|
(void) &res2;
|
|
res3 = res1 * res1 + res2 * res2;
|
|
(void) &res3;
|
|
|
|
output_new_test ("sin^2 + cos^2 == 1");
|
|
result = check_equal (res3, 1.0, epsilon, &diff);
|
|
output_result_ext ("sin^2 + cos^2 == 1", result,
|
|
res3, 1.0, diff, x, PRINT, PRINT);
|
|
}
|
|
|
|
|
|
/* Test sin, cos, tan with the following relation: tan = sin/cos. */
|
|
static void
|
|
identities2_test (MATHTYPE x, MATHTYPE epsilon)
|
|
{
|
|
#ifndef TEST_INLINE
|
|
MATHTYPE res1, res2, res3, res4, diff;
|
|
int result;
|
|
|
|
res1 = FUNC(sin) (x);
|
|
(void) &res1;
|
|
res2 = FUNC(cos) (x);
|
|
(void) &res2;
|
|
res3 = FUNC(tan) (x);
|
|
(void) &res3;
|
|
res4 = res1 / res2;
|
|
(void) &res4;
|
|
|
|
output_new_test ("sin/cos == tan");
|
|
result = check_equal (res4, res3, epsilon, &diff);
|
|
output_result_ext ("sin/cos == tan", result,
|
|
res4, res3, diff, x, PRINT, PRINT);
|
|
#endif
|
|
}
|
|
|
|
|
|
/* Test cosh and sinh with the identity cosh^2 - sinh^2 = 1. */
|
|
static void
|
|
identities3_test (MATHTYPE x, MATHTYPE epsilon)
|
|
{
|
|
MATHTYPE res1, res2, res3, diff;
|
|
int result;
|
|
|
|
res1 = FUNC(sinh) (x);
|
|
(void) &res1;
|
|
res2 = FUNC(cosh) (x);
|
|
(void) &res2;
|
|
res3 = res2 * res2 - res1 * res1;
|
|
(void) &res3;
|
|
|
|
output_new_test ("cosh^2 - sinh^2 == 1");
|
|
result = check_equal (res3, 1.0, epsilon, &diff);
|
|
output_result_ext ("cosh^2 - sinh^2 == 1", result,
|
|
res3, 1.0, diff, x, PRINT, PRINT);
|
|
}
|
|
|
|
|
|
static void
|
|
identities (void)
|
|
{
|
|
identities1_test (0.2L, CHOOSE (1e-18L, 0, 2e-7));
|
|
identities1_test (0.9L, CHOOSE (1e-18L, 0, 2e-7));
|
|
identities1_test (0, 0);
|
|
identities1_test (-1, CHOOSE (1e-18L, 0, 1e-7));
|
|
|
|
identities2_test (0.2L, CHOOSE (1e-19L, 1e-16, 0));
|
|
identities2_test (0.9L, CHOOSE (3e-19L, 1e-15, 2e-7));
|
|
identities2_test (0, 0);
|
|
identities2_test (-1, CHOOSE (1e-18L, 1e-15, 2e-7));
|
|
|
|
identities3_test (0.2L, CHOOSE (1e-18L, 0, 1e-7));
|
|
identities3_test (0.9L, CHOOSE (1e-18L, 1e-15, 1e-6));
|
|
identities3_test (0, CHOOSE (0, 0, 1e-6));
|
|
identities3_test (-1, CHOOSE (1e-18L, 7e-16, 1e-6));
|
|
}
|
|
|
|
|
|
/*
|
|
Let's test that basic arithmetic is working
|
|
tests: Infinity and NaN
|
|
*/
|
|
static void
|
|
basic_tests (void)
|
|
{
|
|
/* variables are declared volatile to forbid some compiler
|
|
optimizations */
|
|
volatile MATHTYPE Inf_var, NaN_var, zero_var, one_var;
|
|
MATHTYPE x1, x2;
|
|
|
|
zero_var = 0.0;
|
|
one_var = 1.0;
|
|
NaN_var = nan_value;
|
|
Inf_var = one_var / zero_var;
|
|
|
|
(void) &zero_var;
|
|
(void) &one_var;
|
|
(void) &NaN_var;
|
|
(void) &Inf_var;
|
|
|
|
/* Clear all exceptions. The previous computations raised exceptions. */
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
check_isinfp ("isinf (inf) == +1", Inf_var);
|
|
check_isinfn ("isinf (-inf) == -1", -Inf_var);
|
|
check_bool ("!isinf (1)", !(FUNC(isinf) (one_var)));
|
|
check_bool ("!isinf (NaN)", !(FUNC(isinf) (NaN_var)));
|
|
|
|
check_isnan ("isnan (NaN)", NaN_var);
|
|
check_isnan ("isnan (-NaN)", -NaN_var);
|
|
check_bool ("!isnan (1)", !(FUNC(isnan) (one_var)));
|
|
check_bool ("!isnan (inf)", !(FUNC(isnan) (Inf_var)));
|
|
|
|
check_bool ("inf == inf", Inf_var == Inf_var);
|
|
check_bool ("-inf == -inf", -Inf_var == -Inf_var);
|
|
check_bool ("inf != -inf", Inf_var != -Inf_var);
|
|
check_bool ("NaN != NaN", NaN_var != NaN_var);
|
|
|
|
/*
|
|
the same tests but this time with NAN from <bits/nan.h>
|
|
NAN is a double const
|
|
*/
|
|
check_bool ("isnan (NAN)", isnan (NAN));
|
|
check_bool ("isnan (-NAN)", isnan (-NAN));
|
|
check_bool ("!isinf (NAN)", !(isinf (NAN)));
|
|
check_bool ("!isinf (-NAN)", !(isinf (-NAN)));
|
|
check_bool ("NAN != NAN", NAN != NAN);
|
|
|
|
/*
|
|
And again with the value returned by the `nan' function.
|
|
*/
|
|
check_bool ("isnan (NAN)", FUNC(isnan) (FUNC(nan) ("")));
|
|
check_bool ("isnan (-NAN)", FUNC(isnan) (-FUNC(nan) ("")));
|
|
check_bool ("!isinf (NAN)", !(FUNC(isinf) (FUNC(nan) (""))));
|
|
check_bool ("!isinf (-NAN)", !(FUNC(isinf) (-FUNC(nan) (""))));
|
|
check_bool ("NAN != NAN", FUNC(nan) ("") != FUNC(nan) (""));
|
|
|
|
/* test if EPSILON is ok */
|
|
x1 = MATHCONST (1.0);
|
|
x2 = x1 + CHOOSE (LDBL_EPSILON, DBL_EPSILON, FLT_EPSILON);
|
|
check_bool ("1 != 1+EPSILON", x1 != x2);
|
|
|
|
x1 = MATHCONST (1.0);
|
|
x2 = x1 - CHOOSE (LDBL_EPSILON, DBL_EPSILON, FLT_EPSILON);
|
|
check_bool ("1 != 1-EPSILON", x1 != x2);
|
|
|
|
/* test if HUGE_VALx is ok */
|
|
x1 = CHOOSE (HUGE_VALL, HUGE_VAL, HUGE_VALF);
|
|
check_bool ("isinf (HUGE_VALx) == +1", ISINF (x1) == +1);
|
|
x1 = -CHOOSE (HUGE_VALL, HUGE_VAL, HUGE_VALF);
|
|
check_bool ("isinf (-HUGE_VALx) == -1", ISINF (x1) == -1);
|
|
|
|
}
|
|
|
|
|
|
static void
|
|
initialize (void)
|
|
{
|
|
fpstack_test ("start *init*");
|
|
plus_zero = 0.0;
|
|
nan_value = plus_zero / plus_zero; /* Suppress GCC warning */
|
|
|
|
minus_zero = FUNC (copysign) (0.0, -1.0);
|
|
plus_infty = CHOOSE (HUGE_VALL, HUGE_VAL, HUGE_VALF);
|
|
minus_infty = -CHOOSE (HUGE_VALL, HUGE_VAL, HUGE_VALF);
|
|
|
|
(void) &plus_zero;
|
|
(void) &nan_value;
|
|
(void) &minus_zero;
|
|
(void) &plus_infty;
|
|
(void) &minus_infty;
|
|
|
|
/* Clear all exceptions. From now on we must not get random exceptions. */
|
|
feclearexcept (FE_ALL_EXCEPT);
|
|
|
|
/* Test to make sure we start correctly. */
|
|
fpstack_test ("end *init*");
|
|
}
|
|
|
|
|
|
static struct option long_options[] =
|
|
{
|
|
{"verbose", optional_argument, NULL, 'v'},
|
|
{"silent", no_argument, NULL, 's'},
|
|
{0, 0, 0, 0}
|
|
};
|
|
|
|
|
|
static void
|
|
parse_options (int argc, char *argv[])
|
|
{
|
|
int c;
|
|
int option_index;
|
|
|
|
verbose = 1;
|
|
|
|
while (1)
|
|
{
|
|
c = getopt_long (argc, argv, "v::s",
|
|
long_options, &option_index);
|
|
|
|
/* Detect the end of the options. */
|
|
if (c == -1)
|
|
break;
|
|
|
|
switch (c)
|
|
{
|
|
case 'v':
|
|
if (optarg)
|
|
verbose = (unsigned int) strtoul (optarg, NULL, 0);
|
|
else
|
|
verbose = 4;
|
|
break;
|
|
case 's':
|
|
verbose = 0;
|
|
default:
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
int
|
|
main (int argc, char *argv[])
|
|
{
|
|
|
|
parse_options (argc, argv);
|
|
|
|
initialize ();
|
|
printf (TEST_MSG);
|
|
|
|
basic_tests ();
|
|
|
|
/* keep the tests a wee bit ordered (according to ISO 9X) */
|
|
/* classification functions */
|
|
fpclassify_test ();
|
|
isfinite_test ();
|
|
isnormal_test ();
|
|
signbit_test ();
|
|
|
|
comparisons_test ();
|
|
|
|
/* trigonometric functions */
|
|
acos_test ();
|
|
asin_test ();
|
|
atan_test ();
|
|
atan2_test ();
|
|
cos_test ();
|
|
sin_test ();
|
|
sincos_test ();
|
|
tan_test ();
|
|
|
|
/* hyperbolic functions */
|
|
acosh_test ();
|
|
asinh_test ();
|
|
atanh_test ();
|
|
cosh_test ();
|
|
sinh_test ();
|
|
tanh_test ();
|
|
|
|
/* exponential and logarithmic functions */
|
|
exp_test ();
|
|
exp10_test ();
|
|
exp2_test ();
|
|
expm1_test ();
|
|
frexp_test ();
|
|
ldexp_test ();
|
|
log_test ();
|
|
log10_test ();
|
|
log1p_test ();
|
|
log2_test ();
|
|
logb_test ();
|
|
modf_test ();
|
|
ilogb_test ();
|
|
scalb_test ();
|
|
scalbn_test ();
|
|
|
|
/* power and absolute value functions */
|
|
cbrt_test ();
|
|
fabs_test ();
|
|
hypot_test ();
|
|
pow_test ();
|
|
sqrt_test ();
|
|
|
|
/* error and gamma functions */
|
|
erf_test ();
|
|
erfc_test ();
|
|
gamma_test ();
|
|
tgamma_test ();
|
|
lgamma_test ();
|
|
|
|
/* nearest integer functions */
|
|
ceil_test ();
|
|
floor_test ();
|
|
nearbyint_test ();
|
|
rint_test ();
|
|
lrint_test ();
|
|
llrint_test ();
|
|
round_test ();
|
|
lround_test ();
|
|
llround_test ();
|
|
trunc_test ();
|
|
|
|
/* remainder functions */
|
|
fmod_test ();
|
|
remainder_test ();
|
|
remquo_test ();
|
|
|
|
/* manipulation functions */
|
|
copysign_test ();
|
|
nextafter_test ();
|
|
|
|
/* maximum, minimum and positive difference functions */
|
|
fdim_test ();
|
|
fmin_test ();
|
|
fmax_test ();
|
|
|
|
/* complex functions */
|
|
cabs_test ();
|
|
carg_test ();
|
|
cexp_test ();
|
|
csin_test ();
|
|
csinh_test ();
|
|
ccos_test ();
|
|
ccosh_test ();
|
|
clog_test ();
|
|
clog10_test ();
|
|
cacos_test ();
|
|
cacosh_test ();
|
|
casin_test ();
|
|
casinh_test ();
|
|
catan_test ();
|
|
catanh_test ();
|
|
ctan_test ();
|
|
ctanh_test ();
|
|
csqrt_test ();
|
|
cpow_test ();
|
|
|
|
/* multiply and add */
|
|
fma_test ();
|
|
|
|
/* special tests */
|
|
identities ();
|
|
inverse_functions ();
|
|
|
|
printf ("\nTest suite completed:\n");
|
|
printf (" %d test cases plus %d tests for exception flags executed.\n",
|
|
noTests, noExcTests);
|
|
if (noErrors)
|
|
{
|
|
printf (" %d errors occured.\n", noErrors);
|
|
exit (1);
|
|
}
|
|
printf (" All tests passed successfully.\n");
|
|
exit (0);
|
|
}
|