1 # SOME DESCRIPTIVE TITLE
2 # Copyright (C) YEAR Free Software Foundation, Inc.
3 # This file is distributed under the same license as the PACKAGE package.
4 # FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
9 "Project-Id-Version: PACKAGE VERSION\n"
10 "POT-Creation-Date: 2012-04-25 05:36+0900\n"
11 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
12 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
13 "Language-Team: LANGUAGE <LL@li.org>\n"
16 "Content-Type: text/plain; charset=CHARSET\n"
17 "Content-Transfer-Encoding: 8bit\n"
20 #: build/C/man3/INFINITY.3:23
26 #: build/C/man3/INFINITY.3:23
32 #: build/C/man3/INFINITY.3:23 build/C/man3/abs.3:32 build/C/man3/acos.3:33 build/C/man3/acosh.3:33 build/C/man3/asin.3:33 build/C/man3/asinh.3:33 build/C/man3/atan.3:33 build/C/man3/atan2.3:33 build/C/man3/atanh.3:33 build/C/man3/cbrt.3:28 build/C/man3/ceil.3:25 build/C/man3/copysign.3:29 build/C/man3/cos.3:32 build/C/man3/cosh.3:34 build/C/man3/div.3:32 build/C/man3/erf.3:33 build/C/man3/erfc.3:24 build/C/man3/exp.3:34 build/C/man3/exp10.3:33 build/C/man3/exp2.3:34 build/C/man3/expm1.3:28 build/C/man3/fabs.3:30 build/C/man3/fdim.3:6 build/C/man3/fenv.3:26 build/C/man3/finite.3:23 build/C/man3/floor.3:25 build/C/man3/fma.3:8 build/C/man3/fmax.3:6 build/C/man3/fmin.3:6 build/C/man3/fmod.3:33 build/C/man3/fpclassify.3:6 build/C/man3/frexp.3:31 build/C/man3/gamma.3:6 build/C/man3/hypot.3:31 build/C/man3/ilogb.3:27 build/C/man3/infnan.3:31 build/C/man3/isgreater.3:6 build/C/man3/j0.3:34 build/C/man3/ldexp.3:32 build/C/man3/lgamma.3:7 build/C/man3/log.3:34 build/C/man3/log10.3:34 build/C/man3/log1p.3:27 build/C/man3/log2.3:34 build/C/man3/logb.3:27 build/C/man3/lrint.3:25 build/C/man3/lround.3:25 build/C/man7/math_error.7:24 build/C/man3/matherr.3:25 build/C/man3/modf.3:31 build/C/man3/nan.3:6 build/C/man3/nextafter.3:6 build/C/man3/pow.3:33 build/C/man3/pow10.3:23 build/C/man3/remainder.3:35 build/C/man3/remquo.3:7 build/C/man3/rint.3:25 build/C/man3/round.3:25 build/C/man3/scalb.3:25 build/C/man3/scalbln.3:25 build/C/man3/signbit.3:6 build/C/man3/significand.3:4 build/C/man3/sin.3:33 build/C/man3/sincos.3:6 build/C/man3/sinh.3:34 build/C/man3/sqrt.3:32 build/C/man3/tan.3:33 build/C/man3/tanh.3:33 build/C/man3/tgamma.3:8 build/C/man3/trunc.3:23 build/C/man3/y0.3:34
34 msgid "Linux Programmer's Manual"
38 #: build/C/man3/INFINITY.3:24 build/C/man3/abs.3:33 build/C/man3/acos.3:34 build/C/man3/acosh.3:34 build/C/man3/asin.3:34 build/C/man3/asinh.3:34 build/C/man3/atan.3:34 build/C/man3/atan2.3:34 build/C/man3/atanh.3:34 build/C/man3/cbrt.3:29 build/C/man3/ceil.3:26 build/C/man3/copysign.3:30 build/C/man3/cos.3:33 build/C/man3/cosh.3:35 build/C/man3/div.3:33 build/C/man3/erf.3:34 build/C/man3/erfc.3:25 build/C/man3/exp.3:35 build/C/man3/exp10.3:34 build/C/man3/exp2.3:35 build/C/man3/expm1.3:29 build/C/man3/fabs.3:31 build/C/man3/fdim.3:7 build/C/man3/fenv.3:27 build/C/man3/finite.3:24 build/C/man3/floor.3:26 build/C/man3/fma.3:9 build/C/man3/fmax.3:7 build/C/man3/fmin.3:7 build/C/man3/fmod.3:34 build/C/man3/fpclassify.3:7 build/C/man3/frexp.3:32 build/C/man3/gamma.3:7 build/C/man3/hypot.3:32 build/C/man3/ilogb.3:28 build/C/man3/infnan.3:32 build/C/man3/isgreater.3:7 build/C/man3/j0.3:35 build/C/man3/ldexp.3:33 build/C/man3/lgamma.3:8 build/C/man3/log.3:35 build/C/man3/log10.3:35 build/C/man3/log1p.3:28 build/C/man3/log2.3:35 build/C/man3/logb.3:28 build/C/man3/lrint.3:26 build/C/man3/lround.3:26 build/C/man7/math_error.7:25 build/C/man3/matherr.3:26 build/C/man3/modf.3:32 build/C/man3/nan.3:7 build/C/man3/nextafter.3:7 build/C/man3/pow.3:34 build/C/man3/pow10.3:24 build/C/man3/remainder.3:36 build/C/man3/remquo.3:8 build/C/man3/rint.3:26 build/C/man3/round.3:26 build/C/man3/scalb.3:26 build/C/man3/scalbln.3:26 build/C/man3/signbit.3:7 build/C/man3/significand.3:5 build/C/man3/sin.3:34 build/C/man3/sincos.3:7 build/C/man3/sinh.3:35 build/C/man3/sqrt.3:33 build/C/man3/tan.3:34 build/C/man3/tanh.3:34 build/C/man3/tgamma.3:9 build/C/man3/trunc.3:24 build/C/man3/y0.3:35
44 #: build/C/man3/INFINITY.3:26
45 msgid "INFINITY, NAN, HUGE_VAL, HUGE_VALF, HUGE_VALL - floating-point constants"
49 #: build/C/man3/INFINITY.3:26 build/C/man3/abs.3:35 build/C/man3/acos.3:36 build/C/man3/acosh.3:36 build/C/man3/asin.3:36 build/C/man3/asinh.3:36 build/C/man3/atan.3:36 build/C/man3/atan2.3:36 build/C/man3/atanh.3:36 build/C/man3/cbrt.3:31 build/C/man3/ceil.3:29 build/C/man3/copysign.3:32 build/C/man3/cos.3:35 build/C/man3/cosh.3:37 build/C/man3/div.3:36 build/C/man3/erf.3:36 build/C/man3/erfc.3:27 build/C/man3/exp.3:37 build/C/man3/exp10.3:36 build/C/man3/exp2.3:37 build/C/man3/expm1.3:31 build/C/man3/fabs.3:33 build/C/man3/fdim.3:9 build/C/man3/fenv.3:32 build/C/man3/finite.3:27 build/C/man3/floor.3:28 build/C/man3/fma.3:11 build/C/man3/fmax.3:9 build/C/man3/fmin.3:9 build/C/man3/fmod.3:36 build/C/man3/fpclassify.3:10 build/C/man3/frexp.3:35 build/C/man3/gamma.3:9 build/C/man3/hypot.3:34 build/C/man3/ilogb.3:30 build/C/man3/infnan.3:34 build/C/man3/isgreater.3:10 build/C/man3/j0.3:38 build/C/man3/ldexp.3:35 build/C/man3/lgamma.3:11 build/C/man3/log.3:37 build/C/man3/log10.3:37 build/C/man3/log1p.3:30 build/C/man3/log2.3:37 build/C/man3/logb.3:30 build/C/man3/lrint.3:28 build/C/man3/lround.3:29 build/C/man7/math_error.7:27 build/C/man3/matherr.3:28 build/C/man3/modf.3:35 build/C/man3/nan.3:9 build/C/man3/nextafter.3:10 build/C/man3/pow.3:36 build/C/man3/pow10.3:26 build/C/man3/remainder.3:39 build/C/man3/remquo.3:10 build/C/man3/rint.3:29 build/C/man3/round.3:28 build/C/man3/scalb.3:29 build/C/man3/scalbln.3:29 build/C/man3/signbit.3:9 build/C/man3/significand.3:8 build/C/man3/sin.3:36 build/C/man3/sincos.3:9 build/C/man3/sinh.3:37 build/C/man3/sqrt.3:35 build/C/man3/tan.3:36 build/C/man3/tanh.3:36 build/C/man3/tgamma.3:11 build/C/man3/trunc.3:26 build/C/man3/y0.3:38
55 #: build/C/man3/INFINITY.3:29
57 msgid "B<#define _ISOC99_SOURCE> /* See feature_test_macros(7) */\n"
61 #: build/C/man3/INFINITY.3:31 build/C/man3/acos.3:39 build/C/man3/acosh.3:39 build/C/man3/asin.3:39 build/C/man3/asinh.3:39 build/C/man3/atan.3:39 build/C/man3/atan2.3:39 build/C/man3/atanh.3:39 build/C/man3/cbrt.3:34 build/C/man3/ceil.3:32 build/C/man3/copysign.3:35 build/C/man3/cos.3:38 build/C/man3/cosh.3:40 build/C/man3/erf.3:39 build/C/man3/erfc.3:30 build/C/man3/exp.3:40 build/C/man3/exp10.3:41 build/C/man3/exp2.3:40 build/C/man3/expm1.3:34 build/C/man3/fabs.3:36 build/C/man3/finite.3:30 build/C/man3/floor.3:31 build/C/man3/fma.3:14 build/C/man3/fmod.3:39 build/C/man3/fpclassify.3:13 build/C/man3/frexp.3:38 build/C/man3/hypot.3:37 build/C/man3/infnan.3:37 build/C/man3/isgreater.3:13 build/C/man3/j0.3:41 build/C/man3/ldexp.3:38 build/C/man3/lgamma.3:14 build/C/man3/log.3:40 build/C/man3/log10.3:40 build/C/man3/log1p.3:33 build/C/man3/log2.3:40 build/C/man3/lrint.3:31 build/C/man3/lround.3:32 build/C/man3/modf.3:38 build/C/man3/pow.3:39 build/C/man3/remainder.3:42 build/C/man3/remquo.3:13 build/C/man3/rint.3:32 build/C/man3/round.3:31 build/C/man3/sin.3:39 build/C/man3/sinh.3:40 build/C/man3/sqrt.3:38 build/C/man3/tan.3:39 build/C/man3/tanh.3:39 build/C/man3/trunc.3:29 build/C/man3/y0.3:41
63 msgid "B<#include E<lt>math.hE<gt>>\n"
67 #: build/C/man3/INFINITY.3:33
73 #: build/C/man3/INFINITY.3:35
79 #: build/C/man3/INFINITY.3:37
85 #: build/C/man3/INFINITY.3:39
87 msgid "B<HUGE_VALF>\n"
91 #: build/C/man3/INFINITY.3:41
93 msgid "B<HUGE_VALL>\n"
97 #: build/C/man3/INFINITY.3:42 build/C/man3/abs.3:65 build/C/man3/acos.3:63 build/C/man3/acosh.3:75 build/C/man3/asin.3:65 build/C/man3/asinh.3:75 build/C/man3/atan.3:65 build/C/man3/atan2.3:63 build/C/man3/atanh.3:75 build/C/man3/cbrt.3:70 build/C/man3/ceil.3:58 build/C/man3/copysign.3:62 build/C/man3/cos.3:64 build/C/man3/cosh.3:66 build/C/man3/div.3:66 build/C/man3/erf.3:72 build/C/man3/erfc.3:63 build/C/man3/exp.3:66 build/C/man3/exp10.3:50 build/C/man3/exp2.3:66 build/C/man3/expm1.3:70 build/C/man3/fabs.3:62 build/C/man3/fdim.3:36 build/C/man3/fenv.3:60 build/C/man3/finite.3:88 build/C/man3/floor.3:57 build/C/man3/fma.3:41 build/C/man3/fmax.3:37 build/C/man3/fmin.3:37 build/C/man3/fmod.3:65 build/C/man3/fpclassify.3:62 build/C/man3/frexp.3:64 build/C/man3/gamma.3:37 build/C/man3/hypot.3:72 build/C/man3/ilogb.3:68 build/C/man3/infnan.3:42 build/C/man3/isgreater.3:44 build/C/man3/j0.3:87 build/C/man3/ldexp.3:64 build/C/man3/lgamma.3:67 build/C/man3/log.3:66 build/C/man3/log10.3:66 build/C/man3/log1p.3:69 build/C/man3/log2.3:66 build/C/man3/logb.3:67 build/C/man3/lrint.3:62 build/C/man3/lround.3:63 build/C/man7/math_error.7:33 build/C/man3/matherr.3:39 build/C/man3/modf.3:64 build/C/man3/nan.3:37 build/C/man3/nextafter.3:65 build/C/man3/pow.3:65 build/C/man3/pow10.3:39 build/C/man3/remainder.3:90 build/C/man3/remquo.3:40 build/C/man3/rint.3:84 build/C/man3/round.3:58 build/C/man3/scalb.3:58 build/C/man3/scalbln.3:74 build/C/man3/signbit.3:31 build/C/man3/significand.3:32 build/C/man3/sin.3:65 build/C/man3/sincos.3:22 build/C/man3/sinh.3:66 build/C/man3/sqrt.3:64 build/C/man3/tan.3:65 build/C/man3/tanh.3:65 build/C/man3/tgamma.3:39 build/C/man3/trunc.3:56 build/C/man3/y0.3:87
103 #: build/C/man3/INFINITY.3:46
105 "The macro B<INFINITY> expands to a I<float> constant representing positive "
110 #: build/C/man3/INFINITY.3:59
112 "The macro B<NAN> expands to a I<float> constant representing a quiet NaN "
113 "(when supported). A I<quiet> NaN is a NaN (\"not-a-number\") that does not "
114 "raise exceptions when it is used in arithmetic. The opposite is a "
115 "I<signaling> NaN. See IEC 60559:1989."
119 #: build/C/man3/INFINITY.3:67
121 "The macros B<HUGE_VAL>, B<HUGE_VALF>, B<HUGE_VALL> expand to constants of "
122 "types I<double>, I<float> and I<long double>, respectively, that represent a "
123 "large positive value, possibly plus infinity."
127 #: build/C/man3/INFINITY.3:67 build/C/man3/abs.3:80 build/C/man3/acos.3:108 build/C/man3/acosh.3:116 build/C/man3/asin.3:107 build/C/man3/asinh.3:101 build/C/man3/atan.3:92 build/C/man3/atan2.3:164 build/C/man3/atanh.3:135 build/C/man3/cbrt.3:87 build/C/man3/ceil.3:76 build/C/man3/copysign.3:83 build/C/man3/cos.3:98 build/C/man3/cosh.3:115 build/C/man3/div.3:89 build/C/man3/erf.3:122 build/C/man3/erfc.3:120 build/C/man3/exp.3:125 build/C/man3/exp10.3:74 build/C/man3/exp2.3:88 build/C/man3/expm1.3:132 build/C/man3/fabs.3:84 build/C/man3/fdim.3:78 build/C/man3/fenv.3:264 build/C/man3/floor.3:75 build/C/man3/fma.3:152 build/C/man3/fmax.3:55 build/C/man3/fmin.3:55 build/C/man3/fmod.3:130 build/C/man3/fpclassify.3:114 build/C/man3/frexp.3:96 build/C/man3/gamma.3:62 build/C/man3/hypot.3:156 build/C/man3/ilogb.3:144 build/C/man3/infnan.3:69 build/C/man3/isgreater.3:87 build/C/man3/j0.3:144 build/C/man3/ldexp.3:127 build/C/man3/lgamma.3:150 build/C/man3/log.3:125 build/C/man3/log10.3:86 build/C/man3/log1p.3:138 build/C/man3/log2.3:88 build/C/man3/logb.3:141 build/C/man3/lrint.3:107 build/C/man3/lround.3:110 build/C/man3/modf.3:88 build/C/man3/nan.3:70 build/C/man3/nextafter.3:163 build/C/man3/pow.3:311 build/C/man3/pow10.3:46 build/C/man3/remainder.3:178 build/C/man3/remquo.3:119 build/C/man3/rint.3:111 build/C/man3/round.3:81 build/C/man3/scalb.3:170 build/C/man3/scalbln.3:153 build/C/man3/signbit.3:56 build/C/man3/significand.3:46 build/C/man3/sin.3:102 build/C/man3/sincos.3:72 build/C/man3/sinh.3:120 build/C/man3/sqrt.3:104 build/C/man3/tan.3:127 build/C/man3/tanh.3:96 build/C/man3/tgamma.3:153 build/C/man3/trunc.3:67 build/C/man3/y0.3:215
129 msgid "CONFORMING TO"
133 #: build/C/man3/INFINITY.3:69
138 #: build/C/man3/INFINITY.3:69
144 #: build/C/man3/INFINITY.3:91
146 "On a glibc system, the macro B<HUGE_VAL> is always available. Availability "
147 "of the B<NAN> macro can be tested using B<#ifdef NAN>, and similarly for "
148 "B<INFINITY>, B<HUGE_VALF>, B<HUGE_VALL>. They will be defined by "
149 "I<E<lt>math.hE<gt>> if B<_ISOC99_SOURCE> or B<_GNU_SOURCE> is defined, or "
150 "B<__STDC_VERSION__> is defined and has a value not less than 199901L."
154 #: build/C/man3/INFINITY.3:91 build/C/man3/abs.3:124 build/C/man3/acos.3:114 build/C/man3/acosh.3:122 build/C/man3/asin.3:113 build/C/man3/asinh.3:107 build/C/man3/atan.3:98 build/C/man3/atan2.3:170 build/C/man3/atanh.3:152 build/C/man3/cbrt.3:91 build/C/man3/ceil.3:107 build/C/man3/copysign.3:93 build/C/man3/cos.3:111 build/C/man3/cosh.3:126 build/C/man3/div.3:104 build/C/man3/erf.3:128 build/C/man3/erfc.3:133 build/C/man3/exp.3:131 build/C/man3/exp10.3:76 build/C/man3/exp2.3:94 build/C/man3/expm1.3:166 build/C/man3/fabs.3:90 build/C/man3/fdim.3:80 build/C/man3/fenv.3:322 build/C/man3/finite.3:123 build/C/man3/floor.3:97 build/C/man3/fma.3:154 build/C/man3/fmax.3:57 build/C/man3/fmin.3:57 build/C/man3/fmod.3:144 build/C/man3/fpclassify.3:128 build/C/man3/frexp.3:135 build/C/man3/gamma.3:94 build/C/man3/hypot.3:162 build/C/man3/ilogb.3:146 build/C/man3/isgreater.3:94 build/C/man3/j0.3:157 build/C/man3/ldexp.3:133 build/C/man3/lgamma.3:169 build/C/man3/log.3:138 build/C/man3/log10.3:92 build/C/man3/log1p.3:141 build/C/man3/log2.3:94 build/C/man3/logb.3:143 build/C/man3/lrint.3:109 build/C/man3/lround.3:112 build/C/man7/math_error.7:254 build/C/man3/matherr.3:423 build/C/man3/modf.3:94 build/C/man3/nan.3:74 build/C/man3/nextafter.3:172 build/C/man3/pow.3:360 build/C/man3/pow10.3:51 build/C/man3/remainder.3:213 build/C/man3/remquo.3:121 build/C/man3/rint.3:133 build/C/man3/round.3:104 build/C/man3/scalb.3:193 build/C/man3/scalbln.3:171 build/C/man3/signbit.3:60 build/C/man3/significand.3:52 build/C/man3/sin.3:115 build/C/man3/sincos.3:74 build/C/man3/sinh.3:126 build/C/man3/sqrt.3:110 build/C/man3/tan.3:140 build/C/man3/tanh.3:102 build/C/man3/tgamma.3:179 build/C/man3/trunc.3:78 build/C/man3/y0.3:238
160 #: build/C/man3/INFINITY.3:93
161 msgid "B<fpclassify>(3), B<math_error>(7)"
165 #: build/C/man3/abs.3:32
171 #: build/C/man3/abs.3:32 build/C/man3/acos.3:33 build/C/man3/acosh.3:33 build/C/man3/asin.3:33 build/C/man3/asinh.3:33 build/C/man3/atan.3:33 build/C/man3/atan2.3:33 build/C/man3/cbrt.3:28 build/C/man3/ceil.3:25 build/C/man3/cosh.3:34 build/C/man3/erf.3:33 build/C/man3/erfc.3:24 build/C/man3/exp.3:34 build/C/man3/exp2.3:34 build/C/man3/fabs.3:30 build/C/man3/fdim.3:6 build/C/man3/floor.3:25 build/C/man3/fma.3:8 build/C/man3/fmax.3:6 build/C/man3/fmin.3:6 build/C/man3/fpclassify.3:6 build/C/man3/frexp.3:31 build/C/man3/hypot.3:31 build/C/man3/ilogb.3:27 build/C/man3/isgreater.3:6 build/C/man3/j0.3:34 build/C/man3/ldexp.3:32 build/C/man3/log.3:34 build/C/man3/log10.3:34 build/C/man3/log1p.3:27 build/C/man3/logb.3:27 build/C/man3/lrint.3:25 build/C/man3/lround.3:25 build/C/man3/modf.3:31 build/C/man3/nan.3:6 build/C/man3/nextafter.3:6 build/C/man3/remainder.3:35 build/C/man3/remquo.3:7 build/C/man3/rint.3:25 build/C/man3/round.3:25 build/C/man3/scalb.3:25 build/C/man3/scalbln.3:25 build/C/man3/signbit.3:6 build/C/man3/sinh.3:34 build/C/man3/sqrt.3:32 build/C/man3/tanh.3:33 build/C/man3/tgamma.3:8 build/C/man3/trunc.3:23
177 #: build/C/man3/abs.3:32 build/C/man3/asinh.3:33 build/C/man3/cbrt.3:28 build/C/man3/copysign.3:29 build/C/man3/erf.3:33 build/C/man3/erfc.3:24 build/C/man3/exp10.3:33 build/C/man3/gamma.3:6 build/C/man3/infnan.3:31 build/C/man3/nan.3:6 build/C/man3/nextafter.3:6 build/C/man3/remquo.3:7 build/C/man3/signbit.3:6 build/C/man3/significand.3:4 build/C/man3/sincos.3:6 build/C/man3/tgamma.3:8
183 #: build/C/man3/abs.3:35
184 msgid "abs, labs, llabs, imaxabs - compute the absolute value of an integer"
188 #: build/C/man3/abs.3:38 build/C/man3/div.3:39
190 msgid "B<#include E<lt>stdlib.hE<gt>>\n"
194 #: build/C/man3/abs.3:40
196 msgid "B<int abs(int >I<j>B<);>\n"
200 #: build/C/man3/abs.3:42
202 msgid "B<long int labs(long int >I<j>B<);>\n"
206 #: build/C/man3/abs.3:44
208 msgid "B<long long int llabs(long long int >I<j>B<);>\n"
212 #: build/C/man3/abs.3:46 build/C/man3/div.3:47
214 msgid "B<#include E<lt>inttypes.hE<gt>>\n"
218 #: build/C/man3/abs.3:48
220 msgid "B<intmax_t imaxabs(intmax_t >I<j>B<);>\n"
224 #: build/C/man3/abs.3:53 build/C/man3/acos.3:50 build/C/man3/acosh.3:52 build/C/man3/asin.3:52 build/C/man3/asinh.3:52 build/C/man3/atan.3:52 build/C/man3/atan2.3:50 build/C/man3/atanh.3:52 build/C/man3/cbrt.3:47 build/C/man3/ceil.3:45 build/C/man3/copysign.3:48 build/C/man3/cos.3:51 build/C/man3/cosh.3:53 build/C/man3/div.3:54 build/C/man3/erf.3:50 build/C/man3/erfc.3:41 build/C/man3/exp.3:53 build/C/man3/exp2.3:53 build/C/man3/expm1.3:47 build/C/man3/fabs.3:49 build/C/man3/fdim.3:23 build/C/man3/finite.3:53 build/C/man3/floor.3:44 build/C/man3/fma.3:27 build/C/man3/fmax.3:23 build/C/man3/fmin.3:23 build/C/man3/fmod.3:52 build/C/man3/fpclassify.3:30 build/C/man3/frexp.3:51 build/C/man3/gamma.3:23 build/C/man3/hypot.3:50 build/C/man3/ilogb.3:44 build/C/man3/isgreater.3:32 build/C/man3/j0.3:66 build/C/man3/ldexp.3:51 build/C/man3/lgamma.3:35 build/C/man3/log.3:53 build/C/man3/log10.3:53 build/C/man3/log1p.3:46 build/C/man3/log2.3:53 build/C/man3/logb.3:44 build/C/man3/lrint.3:50 build/C/man3/lround.3:51 build/C/man3/modf.3:51 build/C/man3/nan.3:23 build/C/man3/nextafter.3:30 build/C/man3/pow.3:52 build/C/man3/remainder.3:59 build/C/man3/remquo.3:26 build/C/man3/rint.3:51 build/C/man3/round.3:44 build/C/man3/scalb.3:43 build/C/man3/scalbln.3:49 build/C/man3/signbit.3:19 build/C/man3/significand.3:22 build/C/man3/sin.3:52 build/C/man3/sinh.3:53 build/C/man3/sqrt.3:51 build/C/man3/tan.3:52 build/C/man3/tanh.3:52 build/C/man3/tgamma.3:25 build/C/man3/trunc.3:42 build/C/man3/y0.3:66
225 msgid "Feature Test Macro Requirements for glibc (see B<feature_test_macros>(7)):"
229 #: build/C/man3/abs.3:57
234 #: build/C/man3/abs.3:60 build/C/man3/div.3:61 build/C/man3/exp2.3:61 build/C/man3/fdim.3:31 build/C/man3/fma.3:36 build/C/man3/fmax.3:32 build/C/man3/fmin.3:32 build/C/man3/fpclassify.3:41 build/C/man3/isgreater.3:39 build/C/man3/log2.3:61 build/C/man3/lrint.3:57 build/C/man3/lround.3:58 build/C/man3/nan.3:32 build/C/man3/nextafter.3:60 build/C/man3/remquo.3:35 build/C/man3/round.3:53 build/C/man3/scalbln.3:58 build/C/man3/signbit.3:26 build/C/man3/tgamma.3:34 build/C/man3/trunc.3:51
236 "_XOPEN_SOURCE\\ E<gt>=\\ 600 || _ISOC99_SOURCE || _POSIX_C_SOURCE\\ E<gt>=\\ "
241 #: build/C/man3/abs.3:63 build/C/man3/acos.3:61 build/C/man3/acosh.3:63 build/C/man3/acosh.3:73 build/C/man3/asin.3:63 build/C/man3/asinh.3:63 build/C/man3/asinh.3:73 build/C/man3/atan.3:63 build/C/man3/atan2.3:61 build/C/man3/atanh.3:63 build/C/man3/atanh.3:73 build/C/man3/cbrt.3:59 build/C/man3/cbrt.3:68 build/C/man3/ceil.3:56 build/C/man3/copysign.3:60 build/C/man3/cos.3:62 build/C/man3/cosh.3:64 build/C/man3/div.3:64 build/C/man3/erf.3:60 build/C/man3/erf.3:70 build/C/man3/erfc.3:51 build/C/man3/erfc.3:61 build/C/man3/exp.3:64 build/C/man3/exp2.3:64 build/C/man3/expm1.3:58 build/C/man3/expm1.3:68 build/C/man3/fabs.3:60 build/C/man3/fdim.3:34 build/C/man3/finite.3:68 build/C/man3/finite.3:81 build/C/man3/floor.3:55 build/C/man3/fma.3:39 build/C/man3/fmax.3:35 build/C/man3/fmin.3:35 build/C/man3/fmod.3:63 build/C/man3/fpclassify.3:44 build/C/man3/fpclassify.3:52 build/C/man3/fpclassify.3:60 build/C/man3/frexp.3:62 build/C/man3/hypot.3:60 build/C/man3/hypot.3:70 build/C/man3/ilogb.3:56 build/C/man3/ilogb.3:66 build/C/man3/isgreater.3:42 build/C/man3/ldexp.3:62 build/C/man3/lgamma.3:45 build/C/man3/lgamma.3:55 build/C/man3/log.3:64 build/C/man3/log10.3:64 build/C/man3/log1p.3:57 build/C/man3/log1p.3:67 build/C/man3/log2.3:64 build/C/man3/logb.3:55 build/C/man3/logb.3:65 build/C/man3/lrint.3:60 build/C/man3/lround.3:61 build/C/man3/modf.3:62 build/C/man3/nan.3:35 build/C/man3/nextafter.3:42 build/C/man3/nextafter.3:52 build/C/man3/nextafter.3:63 build/C/man3/pow.3:63 build/C/man3/remainder.3:71 build/C/man3/remainder.3:81 build/C/man3/remquo.3:38 build/C/man3/rint.3:62 build/C/man3/rint.3:72 build/C/man3/rint.3:82 build/C/man3/round.3:56 build/C/man3/scalbln.3:61 build/C/man3/scalbln.3:72 build/C/man3/signbit.3:29 build/C/man3/sin.3:63 build/C/man3/sinh.3:64 build/C/man3/sqrt.3:62 build/C/man3/tan.3:63 build/C/man3/tanh.3:63 build/C/man3/tgamma.3:37 build/C/man3/trunc.3:54
242 msgid "or I<cc\\ -std=c99>"
246 #: build/C/man3/abs.3:77
248 "The B<abs>() function computes the absolute value of the integer argument "
249 "I<j>. The B<labs>(), B<llabs>() and B<imaxabs>() functions compute the "
250 "absolute value of the argument I<j> of the appropriate integer type for the "
255 #: build/C/man3/abs.3:77 build/C/man3/acos.3:68 build/C/man3/acosh.3:80 build/C/man3/asin.3:70 build/C/man3/asinh.3:80 build/C/man3/atan.3:70 build/C/man3/atan2.3:70 build/C/man3/atanh.3:80 build/C/man3/cbrt.3:76 build/C/man3/ceil.3:67 build/C/man3/copysign.3:73 build/C/man3/cos.3:69 build/C/man3/cosh.3:75 build/C/man3/div.3:87 build/C/man3/erf.3:79 build/C/man3/erfc.3:69 build/C/man3/exp.3:71 build/C/man3/exp10.3:55 build/C/man3/exp2.3:71 build/C/man3/expm1.3:84 build/C/man3/fabs.3:67 build/C/man3/fdim.3:39 build/C/man3/fenv.3:257 build/C/man3/floor.3:66 build/C/man3/fma.3:49 build/C/man3/fmax.3:42 build/C/man3/fmin.3:42 build/C/man3/fmod.3:72 build/C/man3/frexp.3:69 build/C/man3/gamma.3:56 build/C/man3/hypot.3:90 build/C/man3/ilogb.3:76 build/C/man3/infnan.3:52 build/C/man3/isgreater.3:77 build/C/man3/j0.3:108 build/C/man3/ldexp.3:71 build/C/man3/lgamma.3:88 build/C/man3/log.3:70 build/C/man3/log10.3:70 build/C/man3/log1p.3:79 build/C/man3/log2.3:70 build/C/man3/logb.3:92 build/C/man3/lrint.3:71 build/C/man3/lround.3:74 build/C/man3/modf.3:70 build/C/man3/nextafter.3:88 build/C/man3/pow.3:70 build/C/man3/remainder.3:116 build/C/man3/remquo.3:68 build/C/man3/rint.3:103 build/C/man3/round.3:71 build/C/man3/scalb.3:77 build/C/man3/scalbln.3:93 build/C/man3/signbit.3:48 build/C/man3/sin.3:70 build/C/man3/sincos.3:46 build/C/man3/sinh.3:75 build/C/man3/sqrt.3:68 build/C/man3/tan.3:70 build/C/man3/tanh.3:74 build/C/man3/tgamma.3:58 build/C/man3/trunc.3:59 build/C/man3/y0.3:111
261 #: build/C/man3/abs.3:80
263 "Returns the absolute value of the integer argument, of the appropriate "
264 "integer type for the function."
267 #. POSIX.1 (1996 edition) only requires the
271 #: build/C/man3/abs.3:95
273 "SVr4, POSIX.1-2001, 4.3BSD, C99. C89 only includes the B<abs>() and "
274 "B<labs>() functions; the functions B<llabs>() and B<imaxabs>() were added "
279 #: build/C/man3/abs.3:95 build/C/man3/ceil.3:82 build/C/man3/copysign.3:88 build/C/man3/erfc.3:126 build/C/man3/fenv.3:266 build/C/man3/finite.3:103 build/C/man3/floor.3:81 build/C/man3/fpclassify.3:121 build/C/man3/gamma.3:65 build/C/man3/infnan.3:71 build/C/man3/isgreater.3:89 build/C/man7/math_error.7:185 build/C/man3/pow10.3:48 build/C/man3/rint.3:113 build/C/man3/round.3:83 build/C/man3/scalbln.3:155 build/C/man3/tgamma.3:155 build/C/man3/trunc.3:69
285 #: build/C/man3/abs.3:98
287 "Trying to take the absolute value of the most negative integer is not "
292 #: build/C/man3/abs.3:107
294 "The B<llabs>() function is included in glibc since version 2.0, but is not "
295 "in libc5 or libc4. The B<imaxabs>() function is included in glibc since "
300 #: build/C/man3/abs.3:113
302 "For B<llabs>() to be declared, it may be necessary to define "
303 "B<_ISOC99_SOURCE> or B<_ISOC9X_SOURCE> (depending on the version of glibc) "
304 "before including any standard headers."
308 #: build/C/man3/abs.3:124
310 "GCC handles B<abs>() and B<labs>() as built-in functions. GCC 3.0 also "
311 "handles B<llabs>() and B<imaxabs>() as built-ins."
315 #: build/C/man3/abs.3:129
316 msgid "B<cabs>(3), B<ceil>(3), B<fabs>(3), B<floor>(3), B<rint>(3)"
320 #: build/C/man3/acos.3:33
326 #: build/C/man3/acos.3:36
327 msgid "acos, acosf, acosl - arc cosine function"
331 #: build/C/man3/acos.3:43
334 "B<double acos(double >I<x>B<);>\n"
335 "B<float acosf(float >I<x>B<);>\n"
336 "B<long double acosl(long double >I<x>B<);>\n"
340 #: build/C/man3/acos.3:46 build/C/man3/acosh.3:48 build/C/man3/asin.3:48 build/C/man3/asinh.3:48 build/C/man3/atan.3:48 build/C/man3/atan2.3:46 build/C/man3/atanh.3:48 build/C/man3/cbrt.3:43 build/C/man3/ceil.3:41 build/C/man3/copysign.3:44 build/C/man3/cos.3:47 build/C/man3/cosh.3:49 build/C/man3/erf.3:46 build/C/man3/erfc.3:37 build/C/man3/exp.3:49 build/C/man3/exp10.3:50 build/C/man3/exp2.3:49 build/C/man3/expm1.3:43 build/C/man3/fabs.3:45 build/C/man3/fdim.3:19 build/C/man3/fenv.3:60 build/C/man3/floor.3:40 build/C/man3/fma.3:23 build/C/man3/fmax.3:19 build/C/man3/fmin.3:19 build/C/man3/fmod.3:48 build/C/man3/fpclassify.3:26 build/C/man3/frexp.3:47 build/C/man3/gamma.3:19 build/C/man3/hypot.3:46 build/C/man3/ilogb.3:40 build/C/man3/infnan.3:42 build/C/man3/isgreater.3:28 build/C/man3/j0.3:62 build/C/man3/ldexp.3:47 build/C/man3/lgamma.3:31 build/C/man3/log.3:49 build/C/man3/log10.3:49 build/C/man3/log1p.3:42 build/C/man3/log2.3:49 build/C/man3/logb.3:40 build/C/man3/lrint.3:46 build/C/man3/lround.3:47 build/C/man3/matherr.3:39 build/C/man3/modf.3:47 build/C/man3/nan.3:19 build/C/man3/nextafter.3:26 build/C/man3/pow.3:48 build/C/man3/pow10.3:39 build/C/man3/remainder.3:55 build/C/man3/remquo.3:22 build/C/man3/rint.3:47 build/C/man3/round.3:40 build/C/man3/scalb.3:39 build/C/man3/scalbln.3:45 build/C/man3/signbit.3:15 build/C/man3/significand.3:18 build/C/man3/sin.3:48 build/C/man3/sincos.3:22 build/C/man3/sinh.3:49 build/C/man3/sqrt.3:47 build/C/man3/tan.3:48 build/C/man3/tanh.3:48 build/C/man3/tgamma.3:21 build/C/man3/trunc.3:38 build/C/man3/y0.3:62
341 msgid "Link with I<-lm>."
345 #: build/C/man3/acos.3:55
346 msgid "B<acosf>(), B<acosl>():"
350 #: build/C/man3/acos.3:58 build/C/man3/acosh.3:70 build/C/man3/asin.3:60 build/C/man3/asinh.3:70 build/C/man3/atan.3:60 build/C/man3/atan2.3:58 build/C/man3/atanh.3:70 build/C/man3/ceil.3:53 build/C/man3/cos.3:59 build/C/man3/cosh.3:61 build/C/man3/erf.3:67 build/C/man3/exp.3:61 build/C/man3/expm1.3:65 build/C/man3/fabs.3:57 build/C/man3/floor.3:52 build/C/man3/fmod.3:60 build/C/man3/fpclassify.3:57 build/C/man3/frexp.3:59 build/C/man3/hypot.3:67 build/C/man3/ilogb.3:63 build/C/man3/ldexp.3:59 build/C/man3/lgamma.3:52 build/C/man3/log.3:61 build/C/man3/log10.3:61 build/C/man3/log1p.3:64 build/C/man3/logb.3:62 build/C/man3/modf.3:59 build/C/man3/nextafter.3:49 build/C/man3/pow.3:60 build/C/man3/remainder.3:78 build/C/man3/rint.3:79 build/C/man3/scalbln.3:69 build/C/man3/sin.3:60 build/C/man3/sinh.3:61 build/C/man3/sqrt.3:59 build/C/man3/tan.3:60 build/C/man3/tanh.3:60
352 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600 || "
353 "_ISOC99_SOURCE || _POSIX_C_SOURCE\\ E<gt>=\\ 200112L;"
357 #: build/C/man3/acos.3:68
359 "The B<acos>() function calculates the arc cosine of I<x>; that is the value "
360 "whose cosine is I<x>."
364 #: build/C/man3/acos.3:72
366 "On success, these functions return the arc cosine of I<x> in radians; the "
367 "return value is in the range [0,\\ pi]."
371 #: build/C/man3/acos.3:76 build/C/man3/acosh.3:87 build/C/man3/asin.3:78 build/C/man3/asinh.3:87 build/C/man3/atan.3:78 build/C/man3/atanh.3:87 build/C/man3/cos.3:76 build/C/man3/cosh.3:82 build/C/man3/erf.3:87 build/C/man3/erfc.3:77 build/C/man3/exp.3:79 build/C/man3/expm1.3:92 build/C/man3/fabs.3:74 build/C/man3/j0.3:116 build/C/man3/ldexp.3:85 build/C/man3/lgamma.3:94 build/C/man3/log.3:78 build/C/man3/log1p.3:87 build/C/man3/logb.3:100 build/C/man3/scalbln.3:99 build/C/man3/sin.3:77 build/C/man3/sinh.3:82 build/C/man3/sqrt.3:75 build/C/man3/tan.3:77 build/C/man3/tanh.3:81 build/C/man3/tgamma.3:64 build/C/man3/y0.3:119
372 msgid "If I<x> is a NaN, a NaN is returned."
376 #: build/C/man3/acos.3:81 build/C/man3/acosh.3:91
377 msgid "If I<x> is +1, +0 is returned."
381 #: build/C/man3/acos.3:87 build/C/man3/cos.3:82 build/C/man3/sin.3:86 build/C/man3/tan.3:83
383 "If I<x> is positive infinity or negative infinity, a domain error occurs, "
384 "and a NaN is returned."
388 #. POSIX.1-2001 documents an optional range error for subnormal x;
389 #. glibc 2.8 does not do this.
391 #: build/C/man3/acos.3:93 build/C/man3/asin.3:92
393 "If I<x> is outside the range [-1,\\ 1], a domain error occurs, and a NaN is "
398 #: build/C/man3/acos.3:93 build/C/man3/acosh.3:101 build/C/man3/asin.3:92 build/C/man3/asinh.3:99 build/C/man3/atan.3:90 build/C/man3/atan2.3:160 build/C/man3/atanh.3:111 build/C/man3/cbrt.3:85 build/C/man3/ceil.3:73 build/C/man3/copysign.3:81 build/C/man3/cos.3:82 build/C/man3/cosh.3:100 build/C/man3/erf.3:102 build/C/man3/erfc.3:100 build/C/man3/exp.3:102 build/C/man3/exp10.3:62 build/C/man3/exp2.3:78 build/C/man3/expm1.3:113 build/C/man3/fabs.3:82 build/C/man3/fdim.3:56 build/C/man3/floor.3:72 build/C/man3/fma.3:113 build/C/man3/fmax.3:51 build/C/man3/fmin.3:51 build/C/man3/fmod.3:104 build/C/man3/frexp.3:94 build/C/man3/gamma.3:59 build/C/man3/hypot.3:126 build/C/man3/ilogb.3:108 build/C/man3/infnan.3:62 build/C/man3/isgreater.3:85 build/C/man3/j0.3:123 build/C/man3/ldexp.3:104 build/C/man3/lgamma.3:124 build/C/man3/log.3:102 build/C/man3/log10.3:78 build/C/man3/log1p.3:109 build/C/man3/log2.3:78 build/C/man3/logb.3:114 build/C/man3/lrint.3:85 build/C/man3/lround.3:88 build/C/man3/modf.3:86 build/C/man3/nextafter.3:133 build/C/man3/pow.3:251 build/C/man3/remainder.3:148 build/C/man3/remquo.3:96 build/C/man3/rint.3:108 build/C/man3/round.3:76 build/C/man3/scalb.3:131 build/C/man3/scalbln.3:123 build/C/man3/signbit.3:54 build/C/man3/sin.3:86 build/C/man3/sincos.3:49 build/C/man3/sinh.3:105 build/C/man3/sqrt.3:89 build/C/man3/tan.3:101 build/C/man3/tanh.3:94 build/C/man3/tgamma.3:98 build/C/man3/trunc.3:63 build/C/man3/y0.3:156
404 #: build/C/man3/acos.3:98 build/C/man3/acosh.3:106 build/C/man3/asin.3:97 build/C/man3/atanh.3:116 build/C/man3/cos.3:87 build/C/man3/cosh.3:105 build/C/man3/erf.3:107 build/C/man3/erfc.3:105 build/C/man3/exp.3:107 build/C/man3/exp10.3:67 build/C/man3/exp2.3:83 build/C/man3/expm1.3:118 build/C/man3/fdim.3:61 build/C/man3/fma.3:118 build/C/man3/fmod.3:109 build/C/man3/hypot.3:131 build/C/man3/ilogb.3:113 build/C/man3/j0.3:128 build/C/man3/ldexp.3:109 build/C/man3/lgamma.3:129 build/C/man3/log.3:107 build/C/man3/log10.3:83 build/C/man3/log1p.3:114 build/C/man3/log2.3:83 build/C/man3/logb.3:119 build/C/man3/lrint.3:90 build/C/man3/lround.3:93 build/C/man3/nextafter.3:138 build/C/man3/pow.3:276 build/C/man3/remainder.3:153 build/C/man3/remquo.3:101 build/C/man3/scalb.3:136 build/C/man3/scalbln.3:128 build/C/man3/sin.3:91 build/C/man3/sincos.3:54 build/C/man3/sinh.3:110 build/C/man3/sqrt.3:94 build/C/man3/tan.3:106 build/C/man3/tgamma.3:103 build/C/man3/y0.3:161
406 "See B<math_error>(7) for information on how to determine whether an error "
407 "has occurred when calling these functions."
411 #: build/C/man3/acos.3:100 build/C/man3/acosh.3:108 build/C/man3/asin.3:99 build/C/man3/atanh.3:118 build/C/man3/cos.3:89 build/C/man3/cosh.3:107 build/C/man3/erf.3:109 build/C/man3/erfc.3:107 build/C/man3/exp.3:109 build/C/man3/expm1.3:120 build/C/man3/fdim.3:63 build/C/man3/fma.3:120 build/C/man3/fmod.3:111 build/C/man3/hypot.3:133 build/C/man3/ilogb.3:115 build/C/man3/j0.3:130 build/C/man3/ldexp.3:111 build/C/man3/lgamma.3:131 build/C/man3/log.3:109 build/C/man3/log1p.3:116 build/C/man3/logb.3:121 build/C/man3/lrint.3:92 build/C/man3/lround.3:95 build/C/man3/nextafter.3:140 build/C/man3/pow.3:278 build/C/man3/remainder.3:155 build/C/man3/remquo.3:103 build/C/man3/scalb.3:138 build/C/man3/scalbln.3:130 build/C/man3/sin.3:93 build/C/man3/sincos.3:56 build/C/man3/sinh.3:112 build/C/man3/sqrt.3:96 build/C/man3/tan.3:108 build/C/man3/tgamma.3:105 build/C/man3/y0.3:163
412 msgid "The following errors can occur:"
416 #: build/C/man3/acos.3:100 build/C/man3/asin.3:99
418 msgid "Domain error: I<x> is outside the range [-1,\\ 1]"
422 #: build/C/man3/acos.3:108 build/C/man3/acosh.3:116 build/C/man3/asin.3:107 build/C/man3/atanh.3:126 build/C/man3/fmod.3:130 build/C/man3/log.3:117 build/C/man3/pow.3:286 build/C/man3/remainder.3:178 build/C/man3/sqrt.3:104 build/C/man3/y0.3:171
424 "I<errno> is set to B<EDOM>. An invalid floating-point exception "
425 "(B<FE_INVALID>) is raised."
429 #: build/C/man3/acos.3:114 build/C/man3/acosh.3:122 build/C/man3/asin.3:113 build/C/man3/asinh.3:107 build/C/man3/atan.3:98 build/C/man3/atan2.3:170 build/C/man3/atanh.3:141 build/C/man3/ceil.3:82 build/C/man3/exp.3:131 build/C/man3/exp2.3:94 build/C/man3/fabs.3:90 build/C/man3/floor.3:81 build/C/man3/fmod.3:136 build/C/man3/frexp.3:102 build/C/man3/ldexp.3:133 build/C/man3/log.3:131 build/C/man3/log10.3:92 build/C/man3/modf.3:94 build/C/man3/pow.3:317 build/C/man3/sin.3:108 build/C/man3/sinh.3:126 build/C/man3/sqrt.3:110 build/C/man3/tan.3:133 build/C/man3/tanh.3:102
431 "C99, POSIX.1-2001. The variant returning I<double> also conforms to SVr4, "
436 #: build/C/man3/acos.3:121
438 "B<asin>(3), B<atan>(3), B<atan2>(3), B<cacos>(3), B<cos>(3), B<sin>(3), "
443 #: build/C/man3/acosh.3:33
449 #: build/C/man3/acosh.3:36
450 msgid "acosh, acoshf, acoshl - inverse hyperbolic cosine function"
454 #: build/C/man3/acosh.3:41
456 msgid "B<double acosh(double >I<x>B<);>\n"
460 #: build/C/man3/acosh.3:43
462 msgid "B<float acoshf(float >I<x>B<);>\n"
466 #: build/C/man3/acosh.3:45
468 msgid "B<long double acoshl(long double >I<x>B<);>\n"
472 #: build/C/man3/acosh.3:56
477 #: build/C/man3/acosh.3:60 build/C/man3/asinh.3:60 build/C/man3/atanh.3:60 build/C/man3/cbrt.3:56 build/C/man3/expm1.3:55 build/C/man3/ilogb.3:53 build/C/man3/log1p.3:54 build/C/man3/logb.3:52 build/C/man3/nextafter.3:39 build/C/man3/rint.3:69
479 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 500 || "
480 "_XOPEN_SOURCE\\ &&\\ _XOPEN_SOURCE_EXTENDED || _ISOC99_SOURCE || "
481 "_POSIX_C_SOURCE\\ E<gt>=\\ 200112L;"
485 #: build/C/man3/acosh.3:67
486 msgid "B<acoshf>(), B<acoshl>():"
490 #: build/C/man3/acosh.3:80
492 "The B<acosh>() function calculates the inverse hyperbolic cosine of I<x>; "
493 "that is the value whose hyperbolic cosine is I<x>."
497 #: build/C/man3/acosh.3:83
498 msgid "On success, these functions return the inverse hyperbolic cosine of I<x>."
502 #: build/C/man3/acosh.3:95 build/C/man3/exp.3:84 build/C/man3/expm1.3:101 build/C/man3/log.3:87 build/C/man3/log1p.3:91 build/C/man3/sqrt.3:83 build/C/man3/tgamma.3:68
503 msgid "If I<x> is positive infinity, positive infinity is returned."
507 #: build/C/man3/acosh.3:101
509 "If I<x> is less than 1, a domain error occurs, and the functions return a "
514 #: build/C/man3/acosh.3:108
516 msgid "Domain error: I<x> is less than 1"
520 #: build/C/man3/acosh.3:128
521 msgid "B<asinh>(3), B<atanh>(3), B<cacosh>(3), B<cosh>(3), B<sinh>(3), B<tanh>(3)"
525 #: build/C/man3/asin.3:33
531 #: build/C/man3/asin.3:36
532 msgid "asin, asinf, asinl - arc sine function"
536 #: build/C/man3/asin.3:41
538 msgid "B<double asin(double >I<x>B<);>\n"
542 #: build/C/man3/asin.3:43
544 msgid "B<float asinf(float >I<x>B<);>\n"
548 #: build/C/man3/asin.3:45
550 msgid "B<long double asinl(long double >I<x>B<);>\n"
554 #: build/C/man3/asin.3:57
555 msgid "B<asinf>(), B<asinl>():"
559 #: build/C/man3/asin.3:70
561 "The B<asin>() function calculates the principal value of the arc sine of "
562 "I<x>; that is the value whose sine is I<x>."
566 #: build/C/man3/asin.3:74
568 "On success, these functions return the principal value of the arc sine of "
569 "I<x> in radians; the return value is in the range [-pi/2,\\ pi/2]."
573 #: build/C/man3/asin.3:83 build/C/man3/asinh.3:91 build/C/man3/atan.3:83 build/C/man3/atanh.3:91 build/C/man3/erf.3:91 build/C/man3/expm1.3:97 build/C/man3/scalbln.3:108 build/C/man3/sinh.3:86 build/C/man3/sqrt.3:79 build/C/man3/tanh.3:85
574 msgid "If I<x> is +0 (-0), +0 (-0) is returned."
578 #: build/C/man3/asin.3:120
580 "B<acos>(3), B<atan>(3), B<atan2>(3), B<casin>(3), B<cos>(3), B<sin>(3), "
585 #: build/C/man3/asinh.3:33
591 #: build/C/man3/asinh.3:36
592 msgid "asinh, asinhf, asinhl - inverse hyperbolic sine function"
596 #: build/C/man3/asinh.3:41
598 msgid "B<double asinh(double >I<x>B<);>\n"
602 #: build/C/man3/asinh.3:43
604 msgid "B<float asinhf(float >I<x>B<);>\n"
608 #: build/C/man3/asinh.3:45
610 msgid "B<long double asinhl(long double >I<x>B<);>\n"
614 #: build/C/man3/asinh.3:56
619 #: build/C/man3/asinh.3:67
620 msgid "B<asinhf>(), B<asinhl>():"
624 #: build/C/man3/asinh.3:80
626 "The B<asinh>() function calculates the inverse hyperbolic sine of I<x>; "
627 "that is the value whose hyperbolic sine is I<x>."
631 #: build/C/man3/asinh.3:83
632 msgid "On success, these functions return the inverse hyperbolic sine of I<x>."
636 #: build/C/man3/asinh.3:99 build/C/man3/ldexp.3:90 build/C/man3/scalbln.3:104 build/C/man3/sinh.3:91
638 "If I<x> is positive infinity (negative infinity), positive infinity "
639 "(negative infinity) is returned."
643 #: build/C/man3/asinh.3:101 build/C/man3/atan.3:92 build/C/man3/atan2.3:164 build/C/man3/cbrt.3:87 build/C/man3/copysign.3:83 build/C/man3/fabs.3:84 build/C/man3/fmax.3:53 build/C/man3/fmin.3:53 build/C/man3/frexp.3:96 build/C/man3/isgreater.3:87 build/C/man3/modf.3:88 build/C/man3/signbit.3:56 build/C/man3/tanh.3:96 build/C/man3/trunc.3:65
644 msgid "No errors occur."
648 #: build/C/man3/asinh.3:113
649 msgid "B<acosh>(3), B<atanh>(3), B<casinh>(3), B<cosh>(3), B<sinh>(3), B<tanh>(3)"
653 #: build/C/man3/atan.3:33
659 #: build/C/man3/atan.3:36
660 msgid "atan, atanf, atanl - arc tangent function"
664 #: build/C/man3/atan.3:41
666 msgid "B<double atan(double >I<x>B<);>\n"
670 #: build/C/man3/atan.3:43
672 msgid "B<float atanf(float >I<x>B<);>\n"
676 #: build/C/man3/atan.3:45
678 msgid "B<long double atanl( long double >I<x>B<);>\n"
682 #: build/C/man3/atan.3:57
683 msgid "B<atanf>(), B<atanl>():"
687 #: build/C/man3/atan.3:70
689 "The B<atan>() function calculates the principal value of the arc tangent of "
690 "I<x>; that is the value whose tangent is I<x>."
694 #: build/C/man3/atan.3:74
696 "On success, these functions return the principal value of the arc tangent of "
697 "I<x> in radians; the return value is in the range [-pi/2,\\ pi/2]."
701 #. POSIX.1-2001 documents an optional range error for subnormal x;
702 #. glibc 2.8 does not do this.
704 #: build/C/man3/atan.3:90
705 msgid "If I<x> is positive infinity (negative infinity), +pi/2 (-pi/2) is returned."
709 #: build/C/man3/atan.3:106
711 "B<acos>(3), B<asin>(3), B<atan2>(3), B<carg>(3), B<catan>(3), B<cos>(3), "
712 "B<sin>(3), B<tan>(3)"
716 #: build/C/man3/atan2.3:33
722 #: build/C/man3/atan2.3:36
723 msgid "atan2, atan2f, atan2l - arc tangent function of two variables"
727 #: build/C/man3/atan2.3:43
730 "B<double atan2(double >I<y>B<, double >I<x>B<);>\n"
731 "B<float atan2f(float >I<y>B<, float >I<x>B<);>\n"
732 "B<long double atan2l(long double >I<y>B<, long double >I<x>B<);>\n"
736 #: build/C/man3/atan2.3:55
737 msgid "B<atan2f>(), B<atan2l>():"
741 #: build/C/man3/atan2.3:70
743 "The B<atan2>() function calculates the principal value of the arc tangent "
744 "of I<y/x>, using the signs of the two arguments to determine the quadrant of "
749 #: build/C/man3/atan2.3:74
751 "On success, these functions return the principal value of the arc tangent of "
752 "I<y/x> in radians; the return value is in the range [-pi,\\ pi]."
756 #: build/C/man3/atan2.3:80
757 msgid "If I<y> is +0 (-0) and I<x> is less than 0, +pi (-pi) is returned."
761 #: build/C/man3/atan2.3:86
762 msgid "If I<y> is +0 (-0) and I<x> is greater than 0, +0 (-0) is returned."
766 #: build/C/man3/atan2.3:92
767 msgid "If I<y> is less than 0 and I<x> is +0 or -0, -pi/2 is returned."
771 #: build/C/man3/atan2.3:98
772 msgid "If I<y> is greater than 0 and I<x> is +0 or -0, pi/2 is returned."
778 #. is 0, a pole error shall not occur.
780 #: build/C/man3/atan2.3:109
781 msgid "If either I<x> or I<y> is NaN, a NaN is returned."
785 #. If the result underflows, a range error may occur and
787 #. should be returned.
789 #: build/C/man3/atan2.3:120
790 msgid "If I<y> is +0 (-0) and I<x> is -0, +pi (-pi) is returned."
794 #: build/C/man3/atan2.3:126
795 msgid "If I<y> is +0 (-0) and I<x> is +0, +0 (-0) is returned."
799 #: build/C/man3/atan2.3:132
801 "If I<y> is a finite value greater (less) than 0, and I<x> is negative "
802 "infinity, +pi (-pi) is returned."
806 #: build/C/man3/atan2.3:138
808 "If I<y> is a finite value greater (less) than 0, and I<x> is positive "
809 "infinity, +0 (-0) is returned."
813 #: build/C/man3/atan2.3:145
815 "If I<y> is positive infinity (negative infinity), and I<x> is finite, pi/2 "
816 "(-pi/2) is returned."
820 #: build/C/man3/atan2.3:151
822 "If I<y> is positive infinity (negative infinity) and I<x> is negative "
823 "infinity, +3*pi/4 (-3*pi/4) is returned."
828 #. If both arguments are 0, a domain error shall not occur.
830 #: build/C/man3/atan2.3:160
832 "If I<y> is positive infinity (negative infinity) and I<x> is positive "
833 "infinity, +pi/4 (-pi/4) is returned."
837 #: build/C/man3/atan2.3:177
839 "B<acos>(3), B<asin>(3), B<atan>(3), B<carg>(3), B<cos>(3), B<sin>(3), "
844 #: build/C/man3/atanh.3:33
850 #: build/C/man3/atanh.3:33 build/C/man3/cos.3:32 build/C/man3/lgamma.3:7 build/C/man3/sin.3:33 build/C/man3/tan.3:33
856 #: build/C/man3/atanh.3:36
857 msgid "atanh, atanhf, atanhl - inverse hyperbolic tangent function"
861 #: build/C/man3/atanh.3:41
863 msgid "B<double atanh(double >I<x>B<);>\n"
867 #: build/C/man3/atanh.3:43
869 msgid "B<float atanhf(float >I<x>B<);>\n"
873 #: build/C/man3/atanh.3:45
875 msgid "B<long double atanhl(long double >I<x>B<);>\n"
879 #: build/C/man3/atanh.3:56
884 #: build/C/man3/atanh.3:67
885 msgid "B<atanhf>(), B<atanhl>():"
889 #: build/C/man3/atanh.3:80
891 "The B<atanh>() function calculates the inverse hyperbolic tangent of I<x>; "
892 "that is the value whose hyperbolic tangent is I<x>."
896 #: build/C/man3/atanh.3:83
897 msgid "On success, these functions return the inverse hyperbolic tangent of I<x>."
901 #: build/C/man3/atanh.3:102
903 "If I<x> is +1 or -1, a pole error occurs, and the functions return "
904 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, with the "
905 "mathematically correct sign."
909 #. POSIX.1-2001 documents an optional range error for subnormal x;
910 #. glibc 2.8 does not do this.
912 #: build/C/man3/atanh.3:111
914 "If the absolute value of I<x> is greater than 1, a domain error occurs, and "
919 #: build/C/man3/atanh.3:118
921 msgid "Domain error: I<x> less than -1 or greater than +1"
925 #: build/C/man3/atanh.3:126
927 msgid "Pole error: I<x> is +1 or -1"
931 #: build/C/man3/atanh.3:135 build/C/man3/lgamma.3:140 build/C/man3/pow.3:295
933 "I<errno> is set to B<ERANGE> (but see BUGS). A divide-by-zero "
934 "floating-point exception (B<FE_DIVBYZERO>) is raised."
938 #: build/C/man3/atanh.3:141 build/C/man3/cos.3:104 build/C/man3/cosh.3:121 build/C/man3/expm1.3:135 build/C/man3/fenv.3:311 build/C/man3/fmod.3:136 build/C/man3/j0.3:150 build/C/man3/lgamma.3:159 build/C/man3/log.3:131 build/C/man3/nextafter.3:167 build/C/man3/pow.3:317 build/C/man3/remainder.3:202 build/C/man3/sin.3:108 build/C/man3/tan.3:133 build/C/man3/tgamma.3:162 build/C/man3/y0.3:221
943 #. Bug: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6759
944 #. This can be seen in sysdeps/ieee754/k_standard.c
946 #: build/C/man3/atanh.3:152
948 "In glibc 2.9 and earlier, when a pole error occurs, I<errno> as set to "
949 "B<EDOM> instead of the POSIX-mandated B<ERANGE>. Since version 2.10, glibc "
950 "does the right thing."
954 #: build/C/man3/atanh.3:158
955 msgid "B<acosh>(3), B<asinh>(3), B<catanh>(3), B<cosh>(3), B<sinh>(3), B<tanh>(3)"
959 #: build/C/man3/cbrt.3:28
965 #: build/C/man3/cbrt.3:31
966 msgid "cbrt, cbrtf, cbrtl - cube root function"
970 #: build/C/man3/cbrt.3:36
972 msgid "B<double cbrt(double >I<x>B<);>\n"
976 #: build/C/man3/cbrt.3:38
978 msgid "B<float cbrtf(float >I<x>B<);>\n"
982 #: build/C/man3/cbrt.3:40
984 msgid "B<long double cbrtl(long double >I<x>B<);>\n"
988 #: build/C/man3/cbrt.3:51
993 #: build/C/man3/cbrt.3:62
994 msgid "B<cbrtf>(), B<cbrtl>():"
998 #: build/C/man3/cbrt.3:65
1000 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600 || "
1001 "_ISOC99_SOURCE || _POSIX_C_SOUCE\\ E<gt>=\\ 200112L;"
1005 #: build/C/man3/cbrt.3:76
1007 "The B<cbrt>() function returns the (real) cube root of I<x>. This function "
1008 "cannot fail; every representable real value has a representable real cube "
1013 #: build/C/man3/cbrt.3:79
1014 msgid "These functions return the cube root of I<x>."
1018 #: build/C/man3/cbrt.3:85
1020 "If I<x> is +0, -0, positive infinity, negative infinity, or NaN, I<x> is "
1025 #: build/C/man3/cbrt.3:91 build/C/man3/expm1.3:135 build/C/man3/fdim.3:80 build/C/man3/fma.3:154 build/C/man3/fmax.3:57 build/C/man3/fmin.3:57 build/C/man3/ilogb.3:146 build/C/man3/isgreater.3:89 build/C/man3/log1p.3:141 build/C/man3/logb.3:143 build/C/man3/lrint.3:109 build/C/man3/lround.3:112 build/C/man3/remquo.3:121 build/C/man3/rint.3:113 build/C/man3/round.3:83 build/C/man3/scalbln.3:155 build/C/man3/tgamma.3:155 build/C/man3/trunc.3:69
1026 msgid "C99, POSIX.1-2001."
1030 #: build/C/man3/cbrt.3:93
1031 msgid "B<pow>(3), B<sqrt>(3)"
1035 #: build/C/man3/ceil.3:25
1041 #: build/C/man3/ceil.3:29
1043 "ceil, ceilf, ceill - ceiling function: smallest integral value not less than "
1048 #: build/C/man3/ceil.3:34
1050 msgid "B<double ceil(double >I<x>B<);>\n"
1054 #: build/C/man3/ceil.3:36
1056 msgid "B<float ceilf(float >I<x>B<);>\n"
1060 #: build/C/man3/ceil.3:38
1062 msgid "B<long double ceill(long double >I<x>B<);>\n"
1066 #: build/C/man3/ceil.3:50
1067 msgid "B<ceilf>(), B<ceill>():"
1071 #: build/C/man3/ceil.3:61
1073 "These functions return the smallest integral value that is not less than "
1078 #: build/C/man3/ceil.3:67
1079 msgid "For example, I<ceil(0.5)> is 1.0, and I<ceil(-0.5)> is 0.0."
1083 #: build/C/man3/ceil.3:70
1084 msgid "These functions return the ceiling of I<x>."
1088 #: build/C/man3/ceil.3:73 build/C/man3/rint.3:108 build/C/man3/round.3:76
1089 msgid "If I<x> is integral, +0, -0, NaN, or infinite, I<x> itself is returned."
1093 #: build/C/man3/ceil.3:76 build/C/man3/floor.3:75 build/C/man3/rint.3:111 build/C/man3/round.3:79
1095 "No errors occur. POSIX.1-2001 documents a range error for overflows, but "
1100 #: build/C/man3/ceil.3:98 build/C/man3/floor.3:97 build/C/man3/rint.3:128
1102 "SUSv2 and POSIX.1-2001 contain text about overflow (which might set I<errno> "
1103 "to B<ERANGE>, or raise an B<FE_OVERFLOW> exception). In practice, the "
1104 "result cannot overflow on any current machine, so this error-handling stuff "
1105 "is just nonsense. (More precisely, overflow can happen only when the "
1106 "maximum value of the exponent is smaller than the number of mantissa bits. "
1107 "For the IEEE-754 standard 32-bit and 64-bit floating-point numbers the "
1108 "maximum value of the exponent is 128 (respectively, 1024), and the number of "
1109 "mantissa bits is 24 (respectively, 53).)"
1113 #: build/C/man3/ceil.3:107 build/C/man3/trunc.3:78
1115 "The integral value returned by these functions may be too large to store in "
1116 "an integer type (I<int>, I<long>, etc.). To avoid an overflow, which will "
1117 "produce undefined results, an application should perform a range check on "
1118 "the returned value before assigning it to an integer type."
1122 #: build/C/man3/ceil.3:113
1124 "B<floor>(3), B<lrint>(3), B<nearbyint>(3), B<rint>(3), B<round>(3), "
1129 #: build/C/man3/copysign.3:29
1135 #: build/C/man3/copysign.3:29
1141 #: build/C/man3/copysign.3:32
1142 msgid "copysign, copysignf, copysignl - copy sign of a number"
1146 #: build/C/man3/copysign.3:37
1148 msgid "B<double copysign(double >I<x>B<, double >I<y>B<);>\n"
1152 #: build/C/man3/copysign.3:39
1154 msgid "B<float copysignf(float >I<x>B<, float >I<y>B<);>\n"
1158 #: build/C/man3/copysign.3:41
1160 msgid "B<long double copysignl(long double >I<x>B<, long double >I<y>B<);>\n"
1164 #: build/C/man3/copysign.3:54
1165 msgid "B<copysign>(), B<copysignf>(), B<copysignl>():"
1169 #: build/C/man3/copysign.3:57
1171 "_SVID_SOURCE || _BSD_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600 || "
1172 "_ISOC99_SOURCE || _POSIX_C_SOURCE\\ E<gt>=\\ 200112L;"
1176 #: build/C/man3/copysign.3:67
1178 "The B<copysign>() functions return a value whose absolute value matches "
1179 "that of I<x>, but whose sign bit matches that of I<y>."
1183 #: build/C/man3/copysign.3:73
1185 "For example, I<copysign(42.0,\\ -1.0)> and I<copysign(-42.0, -1.0)> both "
1190 #: build/C/man3/copysign.3:78
1192 "On success, these functions return a value whose magnitude is taken from "
1193 "I<x> and whose sign is taken from I<y>."
1197 #: build/C/man3/copysign.3:81
1198 msgid "If I<x> is a NaN, a NaN with the sign bit of I<y> is returned."
1202 #: build/C/man3/copysign.3:88 build/C/man3/nextafter.3:167 build/C/man3/signbit.3:60
1204 "C99, POSIX.1-2001. This function is defined in IEC 559 (and the appendix "
1205 "with recommended functions in IEEE 754/IEEE 854)."
1209 #: build/C/man3/copysign.3:93
1211 "On architectures where the floating-point formats are not IEEE 754 "
1212 "compliant, the B<copysign>() functions may treat a negative zero as "
1217 #: build/C/man3/copysign.3:94
1218 msgid "B<signbit>(3)"
1222 #: build/C/man3/cos.3:32
1228 #: build/C/man3/cos.3:35
1229 msgid "cos, cosf, cosl - cosine function"
1233 #: build/C/man3/cos.3:40
1235 msgid "B<double cos(double >I<x>B<);>\n"
1239 #: build/C/man3/cos.3:42
1241 msgid "B<float cosf(float >I<x>B<);>\n"
1245 #: build/C/man3/cos.3:44
1247 msgid "B<long double cosl(long double >I<x>B<);>\n"
1251 #: build/C/man3/cos.3:56
1252 msgid "B<cosf>(), B<cosl>():"
1256 #: build/C/man3/cos.3:69
1258 "The B<cos>() function returns the cosine of I<x>, where I<x> is given in "
1263 #: build/C/man3/cos.3:72
1264 msgid "On success, these functions return the cosine of I<x>."
1268 #: build/C/man3/cos.3:89 build/C/man3/fmod.3:111 build/C/man3/ilogb.3:128 build/C/man3/sin.3:93 build/C/man3/sincos.3:56 build/C/man3/tan.3:108
1270 msgid "Domain error: I<x> is an infinity"
1274 #: build/C/man3/cos.3:98 build/C/man3/fmod.3:120 build/C/man3/sin.3:102 build/C/man3/tan.3:117
1276 "I<errno> is set to B<EDOM> (but see BUGS). An invalid floating-point "
1277 "exception (B<FE_INVALID>) is raised."
1281 #: build/C/man3/cos.3:104 build/C/man3/cosh.3:121 build/C/man3/erf.3:128 build/C/man3/erfc.3:126 build/C/man3/hypot.3:162 build/C/man3/log2.3:94
1283 "C99, POSIX.1-2001. The variant returning I<double> also conforms to SVr4, "
1287 #. http://sourceware.org/bugzilla/show_bug.cgi?id=6782
1289 #: build/C/man3/cos.3:111 build/C/man3/sin.3:115 build/C/man3/tan.3:140
1291 "Before version 2.10, the glibc implementation did not set I<errno> to "
1292 "B<EDOM> when a domain error occurred."
1296 #: build/C/man3/cos.3:119
1298 "B<acos>(3), B<asin>(3), B<atan>(3), B<atan2>(3), B<ccos>(3), B<sin>(3), "
1299 "B<sincos>(3), B<tan>(3)"
1303 #: build/C/man3/cosh.3:34
1309 #: build/C/man3/cosh.3:37
1310 msgid "cosh, coshf, coshl - hyperbolic cosine function"
1314 #: build/C/man3/cosh.3:42
1316 msgid "B<double cosh(double >I<x>B<);>\n"
1320 #: build/C/man3/cosh.3:44
1322 msgid "B<float coshf(float >I<x>B<);>\n"
1326 #: build/C/man3/cosh.3:46
1328 msgid "B<long double coshl(long double >I<x>B<);>\n"
1332 #: build/C/man3/cosh.3:58
1333 msgid "B<coshf>(), B<coshl>():"
1337 #: build/C/man3/cosh.3:71
1339 "The B<cosh>() function returns the hyperbolic cosine of I<x>, which is "
1340 "defined mathematically as:"
1344 #: build/C/man3/cosh.3:74
1346 msgid " cosh(x) = (exp(x) + exp(-x)) / 2\n"
1350 #: build/C/man3/cosh.3:78
1351 msgid "On success, these functions return the hyperbolic cosine of I<x>."
1355 #: build/C/man3/cosh.3:86 build/C/man3/erfc.3:81
1356 msgid "If I<x> is +0 or -0, 1 is returned."
1360 #: build/C/man3/cosh.3:91 build/C/man3/lgamma.3:103
1362 "If I<x> is positive infinity or negative infinity, positive infinity is "
1367 #: build/C/man3/cosh.3:100 build/C/man3/exp.3:102
1369 "If the result overflows, a range error occurs, and the functions return "
1370 "+B<HUGE_VAL>, +B<HUGE_VALF>, or +B<HUGE_VALL>, respectively."
1374 #: build/C/man3/cosh.3:107 build/C/man3/fdim.3:63 build/C/man3/fma.3:129 build/C/man3/hypot.3:133 build/C/man3/lgamma.3:140 build/C/man3/nextafter.3:140 build/C/man3/sinh.3:112 build/C/man3/tan.3:117 build/C/man3/tgamma.3:123 build/C/man3/y0.3:202
1376 msgid "Range error: result overflow"
1380 #: build/C/man3/cosh.3:115 build/C/man3/exp.3:117 build/C/man3/hypot.3:141 build/C/man3/ldexp.3:119 build/C/man3/lgamma.3:150 build/C/man3/pow.3:303 build/C/man3/sinh.3:120 build/C/man3/tgamma.3:131
1382 "I<errno> is set to B<ERANGE>. An overflow floating-point exception "
1383 "(B<FE_OVERFLOW>) is raised."
1387 #: build/C/man3/cosh.3:126
1389 "In glibc version 2.3.4 and earlier, an overflow floating-point "
1390 "(B<FE_OVERFLOW>) exception is not raised when an overflow occurs."
1394 #: build/C/man3/cosh.3:132
1395 msgid "B<acosh>(3), B<asinh>(3), B<atanh>(3), B<ccos>(3), B<sinh>(3), B<tanh>(3)"
1399 #: build/C/man3/div.3:32
1405 #: build/C/man3/div.3:32
1411 #: build/C/man3/div.3:36
1413 "div, ldiv, lldiv, imaxdiv - compute quotient and remainder of an integer "
1418 #: build/C/man3/div.3:41
1420 msgid "B<div_t div(int >I<numerator>B<, int >I<denominator>B<);>\n"
1424 #: build/C/man3/div.3:43
1426 msgid "B<ldiv_t ldiv(long >I<numerator>B<, long >I<denominator>B<);>\n"
1430 #: build/C/man3/div.3:45
1432 msgid "B<lldiv_t lldiv(long long >I<numerator>B<, long long >I<denominator>B<);>\n"
1436 #: build/C/man3/div.3:49
1439 "B<imaxdiv_t imaxdiv(intmax_t >I<numerator>B<, intmax_t "
1440 ">I<denominator>B<);>\n"
1444 #: build/C/man3/div.3:58
1449 #: build/C/man3/div.3:76
1451 "The B<div>() function computes the value I<numerator>/I<denominator> and "
1452 "returns the quotient and remainder in a structure named I<div_t> that "
1453 "contains two integer members (in unspecified order) named I<quot> and "
1454 "I<rem>. The quotient is rounded toward zero. The result satisfies "
1455 "I<quot>*I<denominator>+I<rem> = I<numerator>."
1459 #: build/C/man3/div.3:87
1461 "The B<ldiv>(), B<lldiv>(), and B<imaxdiv>() functions do the same, dividing "
1462 "numbers of the indicated type and returning the result in a structure of the "
1463 "indicated name, in all cases with fields I<quot> and I<rem> of the same type "
1464 "as the function arguments."
1468 #: build/C/man3/div.3:89
1469 msgid "The I<div_t> (etc.) structure."
1473 #: build/C/man3/div.3:96
1475 "SVr4, 4.3BSD, C89. C99. The functions B<lldiv>() and B<imaxdiv>() were "
1480 #: build/C/man3/div.3:96 build/C/man3/frexp.3:102 build/C/man3/matherr.3:273 build/C/man3/remainder.3:211
1486 #: build/C/man3/div.3:98
1491 #: build/C/man3/div.3:101
1493 msgid " div_t q = div(-5, 3);\n"
1497 #: build/C/man3/div.3:104
1498 msgid "the values I<q.quot> and I<q.rem> are -1 and -2, respectively."
1502 #: build/C/man3/div.3:106
1503 msgid "B<abs>(3), B<remainder>(3)"
1507 #: build/C/man3/erf.3:33
1513 #: build/C/man3/erf.3:36
1514 msgid "erf, erff, erfl, - error function"
1518 #: build/C/man3/erf.3:43
1521 "B<double erf(double >I<x>B<);>\n"
1522 "B<float erff(float >I<x>B<);>\n"
1523 "B<long double erfl(long double >I<x>B<);>\n"
1527 #: build/C/man3/erf.3:54
1532 #: build/C/man3/erf.3:57 build/C/man3/erfc.3:48 build/C/man3/fpclassify.3:49 build/C/man3/hypot.3:57 build/C/man3/lgamma.3:42
1534 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE || _ISOC99_SOURCE || "
1535 "_POSIX_C_SOURCE\\ E<gt>=\\ 200112L;"
1539 #: build/C/man3/erf.3:64
1540 msgid "B<erff>(), B<erfl>():"
1544 #: build/C/man3/erf.3:77
1545 msgid "The B<erf>() function returns the error function of I<x>, defined as"
1549 #: build/C/man3/erf.3:77
1551 msgid " erf(x) = 2/sqrt(pi)* integral from 0 to x of exp(-t*t) dt"
1555 #: build/C/man3/erf.3:83
1557 "On success, these functions return the error function of I<x>, a value in "
1558 "the range [-1,\\ 1]."
1562 #. POSIX.1-2001 documents an optional range error (underflow)
1564 #. glibc 2.8 does not do this.
1566 #: build/C/man3/erf.3:96 build/C/man3/tanh.3:94
1567 msgid "If I<x> is positive infinity (negative infinity), +1 (-1) is returned."
1571 #: build/C/man3/erf.3:102
1573 "If I<x> is subnormal, a range error occurs, and the return value is "
1578 #: build/C/man3/erf.3:109
1580 msgid "Range error: result underflow (I<x> is subnormal)"
1583 #. e.g., tgamma(-172.5) on glibc 2.8/x86-32
1588 #: build/C/man3/erf.3:117 build/C/man3/erfc.3:115 build/C/man3/fma.3:145 build/C/man3/hypot.3:149 build/C/man3/nextafter.3:158 build/C/man3/scalb.3:164 build/C/man3/scalbln.3:146 build/C/man3/tgamma.3:143
1589 msgid "An underflow floating-point exception (B<FE_UNDERFLOW>) is raised."
1592 #. FIXME . Is it intentional that these functions do not set errno?
1593 #. sin() and cos() also don't set errno; bugs have been raised for
1596 #: build/C/man3/erf.3:122 build/C/man3/erfc.3:120 build/C/man3/fdim.3:76 build/C/man3/fma.3:150 build/C/man3/log1p.3:138 build/C/man3/logb.3:141 build/C/man3/lrint.3:105 build/C/man3/lround.3:108 build/C/man3/nextafter.3:163 build/C/man3/remquo.3:117 build/C/man3/scalb.3:170 build/C/man3/scalbln.3:151 build/C/man3/sincos.3:70
1597 msgid "These functions do not set I<errno>."
1601 #: build/C/man3/erf.3:131
1602 msgid "B<cerf>(3), B<erfc>(3), B<exp>(3)"
1606 #: build/C/man3/erfc.3:24
1612 #: build/C/man3/erfc.3:27
1613 msgid "erfc, erfcf, erfcl - complementary error function"
1617 #: build/C/man3/erfc.3:34
1620 "B<double erfc(double >I<x>B<);>\n"
1621 "B<float erfcf(float >I<x>B<);>\n"
1622 "B<long double erfcl(long double >I<x>B<);>\n"
1626 #: build/C/man3/erfc.3:45
1631 #: build/C/man3/erfc.3:55
1632 msgid "B<erfcf>(), B<erfcl>():"
1636 #: build/C/man3/erfc.3:58
1638 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600|| _ISOC99_SOURCE "
1639 "|| _POSIX_C_SOURCE\\ E<gt>=\\ 200112L;"
1643 #: build/C/man3/erfc.3:69
1645 "The B<erfc>() function returns the complementary error function of I<x>, "
1646 "that is, 1.0 - erf(x)."
1650 #: build/C/man3/erfc.3:73
1652 "On success, these functions return the complementary error function of I<x>, "
1653 "a value in the range [0,2]."
1657 #: build/C/man3/erfc.3:86
1658 msgid "If I<x> is positive infinity, +0 is returned."
1662 #: build/C/man3/erfc.3:91
1663 msgid "If I<x> is negative infinity, +2 is returned."
1667 #: build/C/man3/erfc.3:94
1669 "If the function result underflows and produces an unrepresentable value, the "
1670 "return value is 0.0."
1673 #. e.g., erfc(27) on x86-32
1675 #: build/C/man3/erfc.3:100
1677 "If the function result underflows but produces a representable (i.e., "
1678 "subnormal) value, that value is returned, and a range error occurs."
1682 #: build/C/man3/erfc.3:107
1684 msgid "Range error: result underflow (result is subnormal)"
1688 #: build/C/man3/erfc.3:133
1690 "The B<erfc>() functions are provided to avoid the loss accuracy that would "
1691 "occur for the calculation 1-erf(x) for large values of I<x> (for which the "
1692 "value of erf(x) approaches 1)."
1696 #: build/C/man3/erfc.3:136
1697 msgid "B<cerf>(3), B<erf>(3), B<exp>(3)"
1701 #: build/C/man3/exp.3:34
1707 #: build/C/man3/exp.3:37
1708 msgid "exp, expf, expl - base-e exponential function"
1712 #: build/C/man3/exp.3:42
1714 msgid "B<double exp(double >I<x>B<);>\n"
1718 #: build/C/man3/exp.3:44
1720 msgid "B<float expf(float >I<x>B<);>\n"
1724 #: build/C/man3/exp.3:46
1726 msgid "B<long double expl(long double >I<x>B<);>\n"
1730 #: build/C/man3/exp.3:58
1731 msgid "B<expf>(), B<expl>():"
1735 #: build/C/man3/exp.3:71
1737 "The B<exp>() function returns the value of e (the base of natural "
1738 "logarithms) raised to the power of I<x>."
1742 #: build/C/man3/exp.3:74
1743 msgid "On success, these functions return the exponential value of I<x>."
1747 #: build/C/man3/exp.3:89
1748 msgid "If I<x> is negative infinity, +0 is returned."
1752 #: build/C/man3/exp.3:93 build/C/man3/ldexp.3:94
1753 msgid "If the result underflows, a range error occurs, and zero is returned."
1757 #: build/C/man3/exp.3:109 build/C/man3/expm1.3:120 build/C/man3/ldexp.3:111 build/C/man3/scalb.3:148 build/C/man3/scalbln.3:130
1759 msgid "Range error, overflow"
1763 #: build/C/man3/exp.3:117 build/C/man3/ldexp.3:119 build/C/man3/scalb.3:156 build/C/man3/scalbln.3:138
1765 msgid "Range error, underflow"
1769 #: build/C/man3/exp.3:125 build/C/man3/ldexp.3:127 build/C/man3/pow.3:311
1771 "I<errno> is set to B<ERANGE>. An underflow floating-point exception "
1772 "(B<FE_UNDERFLOW>) is raised."
1776 #: build/C/man3/exp.3:136
1777 msgid "B<cbrt>(3), B<cexp>(3), B<exp10>(3), B<exp2>(3), B<sqrt>(3)"
1781 #: build/C/man3/exp10.3:33
1787 #: build/C/man3/exp10.3:33 build/C/man7/math_error.7:24 build/C/man3/pow10.3:23 build/C/man3/sincos.3:6
1793 #: build/C/man3/exp10.3:36
1794 msgid "exp10, exp10f, exp10l - base-10 exponential function"
1798 #: build/C/man3/exp10.3:39 build/C/man3/fenv.3:289
1800 msgid "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
1804 #: build/C/man3/exp10.3:43
1806 msgid "B<double exp10(double >I<x>B<);>\n"
1810 #: build/C/man3/exp10.3:45
1812 msgid "B<float exp10f(float >I<x>B<);>\n"
1816 #: build/C/man3/exp10.3:47
1818 msgid "B<long double exp10l(long double >I<x>B<);>\n"
1822 #: build/C/man3/exp10.3:55
1824 "The B<exp10>() function returns the value of 10 raised to the power of "
1829 #: build/C/man3/exp10.3:58
1830 msgid "On success, these functions return the base-10 exponential value of I<x>."
1834 #: build/C/man3/exp10.3:62 build/C/man3/exp2.3:78
1836 "For various special cases, including the handling of infinity and NaN, as "
1837 "well as overflows and underflows, see B<exp>(3)."
1841 #: build/C/man3/exp10.3:72 build/C/man3/exp2.3:86
1843 "For a discussion of the errors that can occur for these functions, see "
1848 #: build/C/man3/exp10.3:72 build/C/man3/exp2.3:86 build/C/man3/fdim.3:76 build/C/man3/fenv.3:262 build/C/man3/fma.3:150 build/C/man3/fmax.3:53 build/C/man3/fmin.3:53 build/C/man3/log2.3:86 build/C/man3/lrint.3:105 build/C/man3/lround.3:108 build/C/man3/nan.3:68 build/C/man3/pow10.3:44 build/C/man3/remquo.3:117 build/C/man3/round.3:79 build/C/man3/scalbln.3:151 build/C/man3/sincos.3:70 build/C/man3/tgamma.3:151 build/C/man3/trunc.3:65
1854 #: build/C/man3/exp10.3:74 build/C/man3/exp2.3:88 build/C/man3/fdim.3:78 build/C/man3/fenv.3:264 build/C/man3/fma.3:152 build/C/man3/fmax.3:55 build/C/man3/fmin.3:55 build/C/man3/log2.3:88 build/C/man3/lrint.3:107 build/C/man3/lround.3:110 build/C/man3/nan.3:70 build/C/man3/pow10.3:46 build/C/man3/remquo.3:119 build/C/man3/round.3:81 build/C/man3/scalbln.3:153 build/C/man3/sincos.3:72 build/C/man3/tgamma.3:153 build/C/man3/trunc.3:67
1855 msgid "These functions first appeared in glibc in version 2.1."
1859 #: build/C/man3/exp10.3:76
1860 msgid "These functions are GNU extensions."
1864 #: build/C/man3/exp10.3:81
1865 msgid "B<cbrt>(3), B<exp>(3), B<exp2>(3), B<log10>(3), B<sqrt>(3)"
1869 #: build/C/man3/exp2.3:34
1875 #: build/C/man3/exp2.3:37
1876 msgid "exp2, exp2f, exp2l - base-2 exponential function"
1880 #: build/C/man3/exp2.3:42
1882 msgid "B<double exp2(double >I<x>B<);>\n"
1886 #: build/C/man3/exp2.3:44
1888 msgid "B<float exp2f(float >I<x>B<);>\n"
1892 #: build/C/man3/exp2.3:46
1894 msgid "B<long double exp2l(long double >I<x>B<);>\n"
1898 #: build/C/man3/exp2.3:59
1899 msgid "B<exp2>(), B<exp2f>(), B<exp2l>():"
1903 #: build/C/man3/exp2.3:71
1904 msgid "The B<exp2>() function returns the value of 2 raised to the power of I<x>."
1908 #: build/C/man3/exp2.3:74
1909 msgid "On success, these functions return the base-2 exponential value of I<x>."
1913 #: build/C/man3/exp2.3:99
1914 msgid "B<cbrt>(3), B<cexp2>(3), B<exp>(3), B<exp10>(3), B<sqrt>(3)"
1918 #: build/C/man3/expm1.3:28
1924 #: build/C/man3/expm1.3:28 build/C/man3/log2.3:34 build/C/man3/pow.3:33
1930 #: build/C/man3/expm1.3:31
1931 msgid "expm1, expm1f, expm1l - exponential minus 1"
1935 #: build/C/man3/expm1.3:36
1937 msgid "B<double expm1(double >I<x>B<);>\n"
1941 #: build/C/man3/expm1.3:38
1943 msgid "B<float expm1f(float >I<x>B<);>\n"
1947 #: build/C/man3/expm1.3:40
1949 msgid "B<long double expm1l(long double >I<x>B<);>\n"
1953 #: build/C/man3/expm1.3:51
1958 #: build/C/man3/expm1.3:62
1959 msgid "B<expm1f>(), B<expm1l>():"
1963 #: build/C/man3/expm1.3:73
1964 msgid "I<expm1(x)> returns a value equivalent to"
1968 #: build/C/man3/expm1.3:76
1970 msgid " exp(x) - 1\n"
1974 #: build/C/man3/expm1.3:84
1976 "It is computed in a way that is accurate even if the value of I<x> is near "
1977 "zero\\(ema case where I<exp(x) - 1> would be inaccurate due to subtraction "
1978 "of two numbers that are nearly equal."
1982 #: build/C/man3/expm1.3:87
1983 msgid "On success, these functions return I<exp(x)\\ -\\ 1>."
1987 #: build/C/man3/expm1.3:105
1988 msgid "If I<x> is negative infinity, -1 is returned."
1992 #: build/C/man3/expm1.3:113
1994 "If the result overflows, a range error occurs, and the functions return "
1995 "-B<HUGE_VAL>, -B<HUGE_VALF>, or -B<HUGE_VALL>, respectively."
1999 #. POSIX.1 specifies an optional range error (underflow) if
2000 #. x is subnormal. Glibc does not implement this.
2002 #: build/C/man3/expm1.3:132
2004 "I<errno> is set to B<ERANGE> (but see BUGS). An overflow floating-point "
2005 "exception (B<FE_OVERFLOW>) is raised."
2009 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6778
2011 #: build/C/man3/expm1.3:143
2013 "For some large negative I<x> values (where the function result approaches "
2014 "-1), B<expm1>() raises a bogus underflow floating-point exception."
2018 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6814
2019 #. e.g., expm1(1e5) through expm1(1.00199970127e5),
2020 #. but not expm1(1.00199970128e5) and beyond.
2022 #: build/C/man3/expm1.3:154
2024 "For some large positive I<x> values, B<expm1>() raises a bogus invalid "
2025 "floating-point exception in addition to the expected overflow exception, and "
2026 "returns a NaN instead of positive infinity."
2029 #. It looks like the fix was in 2.11, or possibly 2.12.
2030 #. I have no test system for 2.11, but 2.12 passes.
2031 #. From the source (sysdeps/i386/fpu/s_expm1.S) it looks
2032 #. like the changes were in 2.11.
2033 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=6788
2035 #: build/C/man3/expm1.3:166
2037 "Before version 2.11, the glibc implementation did not set I<errno> to "
2038 "B<ERANGE> when a range error occurred."
2042 #: build/C/man3/expm1.3:169
2043 msgid "B<exp>(3), B<log>(3), B<log1p>(3)"
2047 #: build/C/man3/fabs.3:30
2053 #: build/C/man3/fabs.3:33
2054 msgid "fabs, fabsf, fabsl - absolute value of floating-point number"
2058 #: build/C/man3/fabs.3:38
2060 msgid "B<double fabs(double >I<x>B<);>\n"
2064 #: build/C/man3/fabs.3:40
2066 msgid "B<float fabsf(float >I<x>B<);>\n"
2070 #: build/C/man3/fabs.3:42
2072 msgid "B<long double fabsl(long double >I<x>B<);>\n"
2076 #: build/C/man3/fabs.3:54
2077 msgid "B<fabsf>(), B<fabsl>():"
2081 #: build/C/man3/fabs.3:67
2083 "The B<fabs>() functions return the absolute value of the floating-point "
2088 #: build/C/man3/fabs.3:70
2089 msgid "These functions return the absolute value of I<x>."
2093 #: build/C/man3/fabs.3:78
2094 msgid "If I<x> is -0, +0 is returned."
2098 #: build/C/man3/fabs.3:82
2100 "If I<x> is negative infinity or positive infinity, positive infinity is "
2105 #: build/C/man3/fabs.3:96
2106 msgid "B<abs>(3), B<cabs>(3), B<ceil>(3), B<floor>(3), B<labs>(3), B<rint>(3)"
2110 #: build/C/man3/fdim.3:6
2116 #: build/C/man3/fdim.3:9
2117 msgid "fdim, fdimf, fdiml - positive difference"
2121 #: build/C/man3/fdim.3:11 build/C/man3/fmax.3:11 build/C/man3/fmin.3:11 build/C/man3/gamma.3:11 build/C/man3/ilogb.3:32 build/C/man3/logb.3:32 build/C/man3/nan.3:11 build/C/man3/nextafter.3:12 build/C/man3/scalb.3:31 build/C/man3/scalbln.3:31 build/C/man3/signbit.3:11 build/C/man3/significand.3:10 build/C/man3/tgamma.3:13
2122 msgid "B<#include E<lt>math.hE<gt>>"
2126 #: build/C/man3/fdim.3:13
2127 msgid "B<double fdim(double >I<x>B<, double >I<y>B<);>"
2131 #: build/C/man3/fdim.3:15
2132 msgid "B<float fdimf(float >I<x>B<, float >I<y>B<);>"
2136 #: build/C/man3/fdim.3:17
2137 msgid "B<long double fdiml(long double >I<x>B<, long double >I<y>B<);>"
2141 #: build/C/man3/fdim.3:28
2142 msgid "B<fdimf>(), B<fdiml>():"
2146 #: build/C/man3/fdim.3:39
2148 "These functions return the positive difference, max(I<x>-I<y>,0), between "
2153 #: build/C/man3/fdim.3:41
2154 msgid "On success, these functions return the positive difference."
2158 #: build/C/man3/fdim.3:47 build/C/man3/fma.3:59 build/C/man3/fmod.3:86 build/C/man3/nextafter.3:111 build/C/man3/remainder.3:128 build/C/man3/remquo.3:78
2159 msgid "If I<x> or I<y> is a NaN, a NaN is returned."
2163 #: build/C/man3/fdim.3:56 build/C/man3/hypot.3:120
2165 "If the result overflows, a range error occurs, and the functions return "
2166 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively."
2169 #. Unable to test this case, since the best approximation of
2170 #. pi/2 in double precision only yields a tan() value of 1.633e16.
2175 #: build/C/man3/fdim.3:71 build/C/man3/fma.3:137 build/C/man3/nextafter.3:149 build/C/man3/scalb.3:156 build/C/man3/scalbln.3:138 build/C/man3/tan.3:127
2176 msgid "An overflow floating-point exception (B<FE_OVERFLOW>) is raised."
2180 #: build/C/man3/fdim.3:81 build/C/man3/fmin.3:58
2185 #: build/C/man3/fenv.3:26
2191 #: build/C/man3/fenv.3:26
2197 #: build/C/man3/fenv.3:26 build/C/man7/math_error.7:24 build/C/man3/matherr.3:25
2203 #: build/C/man3/fenv.3:32
2205 "feclearexcept, fegetexceptflag, feraiseexcept, fesetexceptflag, "
2206 "fetestexcept, fegetenv, fegetround, feholdexcept, fesetround, fesetenv, "
2207 "feupdateenv, feenableexcept, fedisableexcept, fegetexcept - floating-point "
2208 "rounding and exception handling"
2212 #: build/C/man3/fenv.3:35 build/C/man3/fenv.3:291
2214 msgid "B<#include E<lt>fenv.hE<gt>>\n"
2218 #: build/C/man3/fenv.3:37
2220 msgid "B<int feclearexcept(int >I<excepts>B<);>\n"
2224 #: build/C/man3/fenv.3:39
2226 msgid "B<int fegetexceptflag(fexcept_t *>I<flagp>B<, int >I<excepts>B<);>\n"
2230 #: build/C/man3/fenv.3:41
2232 msgid "B<int feraiseexcept(int >I<excepts>B<);>\n"
2236 #: build/C/man3/fenv.3:43
2238 msgid "B<int fesetexceptflag(const fexcept_t *>I<flagp>B<, int >I<excepts>B<);>\n"
2242 #: build/C/man3/fenv.3:45
2244 msgid "B<int fetestexcept(int >I<excepts>B<);>\n"
2248 #: build/C/man3/fenv.3:47
2250 msgid "B<int fegetround(void);>\n"
2254 #: build/C/man3/fenv.3:49
2256 msgid "B<int fesetround(int >I<rounding_mode>B<);>\n"
2260 #: build/C/man3/fenv.3:51
2262 msgid "B<int fegetenv(fenv_t *>I<envp>B<);>\n"
2266 #: build/C/man3/fenv.3:53
2268 msgid "B<int feholdexcept(fenv_t *>I<envp>B<);>\n"
2272 #: build/C/man3/fenv.3:55
2274 msgid "B<int fesetenv(const fenv_t *>I<envp>B<);>\n"
2278 #: build/C/man3/fenv.3:57
2280 msgid "B<int feupdateenv(const fenv_t *>I<envp>B<);>\n"
2284 #: build/C/man3/fenv.3:63
2286 "These eleven functions were defined in C99, and describe the handling of "
2287 "floating-point rounding and exceptions (overflow, zero-divide etc.)."
2291 #: build/C/man3/fenv.3:63
2297 #: build/C/man3/fenv.3:68
2299 "The I<divide-by-zero> exception occurs when an operation on finite numbers "
2300 "produces infinity as exact answer."
2304 #: build/C/man3/fenv.3:74
2306 "The I<overflow> exception occurs when a result has to be represented as a "
2307 "floating-point number, but has (much) larger absolute value than the largest "
2308 "(finite) floating-point number that is representable."
2312 #: build/C/man3/fenv.3:81
2314 "The I<underflow> exception occurs when a result has to be represented as a "
2315 "floating-point number, but has smaller absolute value than the smallest "
2316 "positive normalized floating-point number (and would lose much accuracy when "
2317 "represented as a denormalized number)."
2321 #: build/C/man3/fenv.3:91
2323 "The I<inexact> exception occurs when the rounded result of an operation is "
2324 "not equal to the infinite precision result. It may occur whenever "
2325 "I<overflow> or I<underflow> occurs."
2329 #: build/C/man3/fenv.3:96
2331 "The I<invalid> exception occurs when there is no well-defined result for an "
2332 "operation, as for 0/0 or infinity - infinity or sqrt(-1)."
2336 #: build/C/man3/fenv.3:96
2338 msgid "Exception handling"
2342 #: build/C/man3/fenv.3:102
2344 "Exceptions are represented in two ways: as a single bit (exception "
2345 "present/absent), and these bits correspond in some implementation-defined "
2346 "way with bit positions in an integer, and also as an opaque structure that "
2347 "may contain more information about the exception (perhaps the code address "
2348 "where it occurred)."
2352 #: build/C/man3/fenv.3:118
2354 "Each of the macros B<FE_DIVBYZERO>, B<FE_INEXACT>, B<FE_INVALID>, "
2355 "B<FE_OVERFLOW>, B<FE_UNDERFLOW> is defined when the implementation supports "
2356 "handling of the corresponding exception, and if so then defines the "
2357 "corresponding bit(s), so that one can call exception handling functions, for "
2358 "example, using the integer argument B<FE_OVERFLOW>|B<FE_UNDERFLOW>. Other "
2359 "exceptions may be supported. The macro B<FE_ALL_EXCEPT> is the bitwise OR "
2360 "of all bits corresponding to supported exceptions."
2364 #: build/C/man3/fenv.3:123
2366 "The B<feclearexcept>() function clears the supported exceptions represented "
2367 "by the bits in its argument."
2371 #: build/C/man3/fenv.3:131
2373 "The B<fegetexceptflag>() function stores a representation of the state of "
2374 "the exception flags represented by the argument I<excepts> in the opaque "
2379 #: build/C/man3/fenv.3:136
2381 "The B<feraiseexcept>() function raises the supported exceptions represented "
2382 "by the bits in I<excepts>."
2386 #: build/C/man3/fenv.3:147
2388 "The B<fesetexceptflag>() function sets the complete status for the "
2389 "exceptions represented by I<excepts> to the value I<*flagp>. This value "
2390 "must have been obtained by an earlier call of B<fegetexceptflag>() with a "
2391 "last argument that contained all bits in I<excepts>."
2395 #: build/C/man3/fenv.3:154
2397 "The B<fetestexcept>() function returns a word in which the bits are set "
2398 "that were set in the argument I<excepts> and for which the corresponding "
2399 "exception is currently set."
2403 #: build/C/man3/fenv.3:154
2405 msgid "Rounding mode"
2409 #: build/C/man3/fenv.3:162
2411 "The rounding mode determines how the result of floating-point operations is "
2412 "treated when the result cannot be exactly represented in the significand. "
2413 "Various rounding modes may be provided: round to nearest (the default), "
2414 "round up (toward positive infinity), round down (toward negative infinity), "
2415 "and round toward zero."
2419 #: build/C/man3/fenv.3:171
2421 "Each of the macros B<FE_TONEAREST>, B<FE_UPWARD>, B<FE_DOWNWARD>, and "
2422 "B<FE_TOWARDZERO> is defined when the implementation supports getting and "
2423 "setting the corresponding rounding direction."
2427 #: build/C/man3/fenv.3:176
2429 "The B<fegetround>() function returns the macro corresponding to the current "
2434 #: build/C/man3/fenv.3:181
2436 "The B<fesetround>() function sets the rounding mode as specified by its "
2437 "argument and returns zero when it was successful."
2441 #: build/C/man3/fenv.3:189
2443 "C99 and POSIX.1-2008 specify an identifier, B<FLT_ROUNDS>, defined in "
2444 "I<E<lt>float.hE<gt>>, which indicates the implementation-defined rounding "
2445 "behavior for floating-point addition. This identifier has one of the "
2450 #: build/C/man3/fenv.3:189
2456 #: build/C/man3/fenv.3:191
2457 msgid "The rounding mode is not determinable."
2461 #: build/C/man3/fenv.3:191
2467 #: build/C/man3/fenv.3:193
2468 msgid "Rounding is toward 0."
2472 #: build/C/man3/fenv.3:193
2478 #: build/C/man3/fenv.3:195
2479 msgid "Rounding is toward nearest number."
2483 #: build/C/man3/fenv.3:195
2489 #: build/C/man3/fenv.3:197
2490 msgid "Rounding is toward positive infinity."
2494 #: build/C/man3/fenv.3:197
2500 #: build/C/man3/fenv.3:199
2501 msgid "Rounding is toward negative infinity."
2505 #: build/C/man3/fenv.3:201
2506 msgid "Other values represent machine-dependent, nonstandard rounding modes."
2510 #: build/C/man3/fenv.3:207
2512 "The value of B<FLT_ROUNDS> should reflect the current rounding mode as set "
2513 "by B<fesetround>() (but see BUGS)."
2517 #: build/C/man3/fenv.3:207
2519 msgid "Floating-point environment"
2523 #: build/C/man3/fenv.3:219
2525 "The entire floating-point environment, including control modes and status "
2526 "flags, can be handled as one opaque object, of type I<fenv_t>. The default "
2527 "environment is denoted by B<FE_DFL_ENV> (of type I<const fenv_t *>). This "
2528 "is the environment setup at program start and it is defined by ISO C to have "
2529 "round to nearest, all exceptions cleared and a nonstop (continue on "
2534 #: build/C/man3/fenv.3:224
2536 "The B<fegetenv>() function saves the current floating-point environment in "
2537 "the object I<*envp>."
2541 #: build/C/man3/fenv.3:231
2543 "The B<feholdexcept>() function does the same, then clears all exception "
2544 "flags, and sets a nonstop (continue on exceptions) mode, if available. It "
2545 "returns zero when successful."
2549 #: build/C/man3/fenv.3:244
2551 "The B<fesetenv>() function restores the floating-point environment from the "
2552 "object I<*envp>. This object must be known to be valid, for example, the "
2553 "result of a call to B<fegetenv>() or B<feholdexcept>() or equal to "
2554 "B<FE_DFL_ENV>. This call does not raise exceptions."
2558 #: build/C/man3/fenv.3:257
2560 "The B<feupdateenv>() function installs the floating-point environment "
2561 "represented by the object I<*envp>, except that currently raised exceptions "
2562 "are not cleared. After calling this function, the raised exceptions will be "
2563 "a bitwise OR of those previously set with those in I<*envp>. As before, the "
2564 "object I<*envp> must be known to be valid."
2567 #. Earlier seven of these functions were listed as returning void.
2568 #. This was corrected in Corrigendum 1 (ISO/IEC 9899:1999/Cor.1:2001(E))
2569 #. of the C99 Standard.
2571 #: build/C/man3/fenv.3:262
2572 msgid "These functions return zero on success and nonzero if an error occurred."
2576 #: build/C/man3/fenv.3:266
2577 msgid "IEC 60559 (IEC 559:1989), ANSI/IEEE 854, C99, POSIX.1-2001."
2581 #: build/C/man3/fenv.3:267
2587 #: build/C/man3/fenv.3:286
2589 "If possible, the GNU C Library defines a macro B<FE_NOMASK_ENV> which "
2590 "represents an environment where every exception raised causes a trap to "
2591 "occur. You can test for this macro using B<#ifdef>. It is only defined if "
2592 "B<_GNU_SOURCE> is defined. The C99 standard does not define a way to set "
2593 "individual bits in the floating-point mask, for example, to trap on specific "
2594 "flags. Since version 2.2, glibc supports the functions B<feenableexcept>() "
2595 "and B<fedisableexcept>() to set individual floating-point traps, and "
2596 "B<fegetexcept>() to query the state."
2600 #: build/C/man3/fenv.3:293
2602 msgid "B<int feenableexcept(int >I<excepts>B<);>\n"
2606 #: build/C/man3/fenv.3:295
2608 msgid "B<int fedisableexcept(int >I<excepts>B<);>\n"
2612 #: build/C/man3/fenv.3:297
2614 msgid "B<int fegetexcept(void);>\n"
2618 #: build/C/man3/fenv.3:311
2620 "The B<feenableexcept>() and B<fedisableexcept>() functions enable "
2621 "(disable) traps for each of the exceptions represented by I<excepts> and "
2622 "return the previous set of enabled exceptions when successful, and -1 "
2623 "otherwise. The B<fegetexcept>() function returns the set of all currently "
2624 "enabled exceptions."
2627 #. Aug 08, glibc 2.8
2628 #. See http://gcc.gnu.org/ml/gcc/2002-02/msg01535.html
2630 #: build/C/man3/fenv.3:322
2632 "C99 specifies that the value of B<FLT_ROUNDS> should reflect changes to the "
2633 "current rounding mode, as set by B<fesetround>(). Currently, this does not "
2634 "occur: B<FLT_ROUNDS> always has the value 1."
2638 #: build/C/man3/fenv.3:323
2639 msgid "B<math_error>(7)"
2643 #: build/C/man3/finite.3:23
2649 #: build/C/man3/finite.3:23 build/C/man3/gamma.3:6
2655 #: build/C/man3/finite.3:27
2657 "finite, finitef, finitel, isinf, isinff, isinfl, isnan, isnanf, isnanl - BSD "
2658 "floating-point classification functions"
2662 #: build/C/man3/finite.3:32
2664 msgid "B<int finite(double >I<x>B<);>\n"
2668 #: build/C/man3/finite.3:34
2670 msgid "B<int finitef(float >I<x>B<);>\n"
2674 #: build/C/man3/finite.3:36
2676 msgid "B<int finitel(long double >I<x>B<);>\n"
2680 #: build/C/man3/finite.3:38
2682 msgid "B<int isinf(double >I<x>B<);>\n"
2686 #: build/C/man3/finite.3:40
2688 msgid "B<int isinff(float >I<x>B<);>\n"
2692 #: build/C/man3/finite.3:42
2694 msgid "B<int isinfl(long double >I<x>B<);>\n"
2698 #: build/C/man3/finite.3:44
2700 msgid "B<int isnan(double >I<x>B<);>\n"
2704 #: build/C/man3/finite.3:46
2706 msgid "B<int isnanf(float >I<x>B<);>\n"
2710 #: build/C/man3/finite.3:48
2712 msgid "B<int isnanl(long double >I<x>B<);>\n"
2716 #: build/C/man3/finite.3:59
2717 msgid "B<finite>(), B<finitef>(), B<finitel>():"
2721 #: build/C/man3/finite.3:61 build/C/man3/finite.3:74 build/C/man3/lgamma.3:61
2722 msgid "_BSD_SOURCE || _SVID_SOURCE"
2726 #: build/C/man3/finite.3:63 build/C/man3/fpclassify.3:54
2731 #: build/C/man3/finite.3:65
2733 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600 || "
2738 #: build/C/man3/finite.3:72
2739 msgid "B<isinff>(), B<isinfl>():"
2743 #: build/C/man3/finite.3:76 build/C/man3/fpclassify.3:46
2748 #: build/C/man3/finite.3:78
2749 msgid "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE || _ISOC99_SOURCE;"
2753 #: build/C/man3/finite.3:84
2754 msgid "B<isnanf>(), B<isnanl>():"
2758 #: build/C/man3/finite.3:86 build/C/man3/gamma.3:35 build/C/man3/scalb.3:56
2759 msgid "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600"
2763 #: build/C/man3/finite.3:93
2765 "The B<finite>() functions return a nonzero value if I<x> is neither "
2766 "infinite nor a \"not-a-number\" (NaN) value, and 0 otherwise."
2770 #: build/C/man3/finite.3:98
2772 "The B<isnan>() functions return a nonzero value if I<x> is a NaN value, and "
2777 #: build/C/man3/finite.3:103
2779 "The B<isinf>() functions return 1 if I<x> is positive infinity, -1 if I<x> "
2780 "is negative infinity, and 0 otherwise."
2784 #. finite* not on HP-UX; they exist on Tru64.
2788 #. function occurs in 4.3BSD.
2789 #. see IEEE.3 in the 4.3BSD manual
2791 #: build/C/man3/finite.3:123
2793 "Note that these functions are obsolete. C99 defines macros B<isfinite>(), "
2794 "B<isinf>(), and B<isnan>() (for all types) replacing them. Further note "
2795 "that the C99 B<isinf>() has weaker guarantees on the return value. See "
2800 #: build/C/man3/finite.3:124
2801 msgid "B<fpclassify>(3)"
2805 #: build/C/man3/floor.3:25
2811 #: build/C/man3/floor.3:28
2812 msgid "floor, floorf, floorl - largest integral value not greater than argument"
2816 #: build/C/man3/floor.3:33
2818 msgid "B<double floor(double >I<x>B<);>\n"
2822 #: build/C/man3/floor.3:35
2824 msgid "B<float floorf(float >I<x>B<);>\n"
2828 #: build/C/man3/floor.3:37
2830 msgid "B<long double floorl(long double >I<x>B<);>\n"
2834 #: build/C/man3/floor.3:49
2835 msgid "B<floorf>(), B<floorl>():"
2839 #: build/C/man3/floor.3:60
2841 "These functions return the largest integral value that is not greater than "
2846 #: build/C/man3/floor.3:66
2847 msgid "For example, I<floor(0.5)> is 0.0, and I<floor(-0.5)> is -1.0."
2851 #: build/C/man3/floor.3:69
2852 msgid "These functions return the floor of I<x>."
2856 #: build/C/man3/floor.3:72
2857 msgid "If I<x> is integral, +0, -0, NaN, or an infinity, I<x> itself is returned."
2861 #: build/C/man3/floor.3:103
2863 "B<ceil>(3), B<lrint>(3), B<nearbyint>(3), B<rint>(3), B<round>(3), "
2868 #: build/C/man3/fma.3:8
2874 #: build/C/man3/fma.3:11
2875 msgid "fma, fmaf, fmal - floating-point multiply and add"
2879 #: build/C/man3/fma.3:16
2881 msgid "B<double fma(double >I<x>B<, double >I<y>B<, double >I<z>B<);>\n"
2885 #: build/C/man3/fma.3:18
2887 msgid "B<float fmaf(float >I<x>B<, float >I<y>B<, float >I<z>B<);>\n"
2891 #: build/C/man3/fma.3:20
2894 "B<long double fmal(long double >I<x>B<, long double >I<y>B<, long double "
2899 #: build/C/man3/fma.3:33
2900 msgid "B<fma>(), B<fmaf>(), B<fmal>():"
2904 #: build/C/man3/fma.3:49
2906 "The B<fma>() function computes I<x> * I<y> + I<z>. The result is rounded "
2907 "as one ternary operation according to the current rounding mode (see "
2912 #: build/C/man3/fma.3:53
2914 "These functions return the value of I<x> * I<y> + I<z>, rounded as one "
2915 "ternary operation."
2919 #: build/C/man3/fma.3:69
2921 "If I<x> times I<y> is an exact infinity, and I<z> is an infinity with the "
2922 "opposite sign, a domain error occurs, and a NaN is returned."
2925 #. POSIX.1-2008 allows some possible differences for the following two
2926 #. domain error cases, but on Linux they are treated the same (AFAICS).
2927 #. Nevertheless, we'll mirror POSIX.1 and describe the two cases
2929 #. POSIX.1 says that a NaN or an implementation-defined value shall
2930 #. be returned for this case.
2932 #: build/C/man3/fma.3:85
2934 "If one of I<x> or I<y> is an infinity, the other is 0, and I<z> is not a "
2935 "NaN, a domain error occurs, and a NaN is returned."
2938 #. POSIX.1 makes the domain error optional for this case.
2940 #: build/C/man3/fma.3:96
2942 "If one of I<x> or I<y> is an infinity, and the other is 0, and I<z> is a "
2943 "NaN, a domain error occurs, and a NaN is returned."
2947 #: build/C/man3/fma.3:105
2949 "If I<x> times I<y> is not an infinity times zero (or vice versa), and I<z> "
2950 "is a NaN, a NaN is returned."
2954 #: build/C/man3/fma.3:109
2956 "If the result overflows, a range error occurs, and an infinity with the "
2957 "correct sign is returned."
2961 #: build/C/man3/fma.3:113
2962 msgid "If the result underflows, a range error occurs, and a signed 0 is returned."
2966 #: build/C/man3/fma.3:120
2969 "Domain error: I<x> * I<y> + I<z>, or I<x> * I<y> is invalid and I<z> is not "
2977 #: build/C/man3/fma.3:129 build/C/man3/ilogb.3:123 build/C/man3/log1p.3:124 build/C/man3/lrint.3:100 build/C/man3/lround.3:103 build/C/man3/remainder.3:163 build/C/man3/remquo.3:112 build/C/man3/scalb.3:148 build/C/man3/sincos.3:64
2978 msgid "An invalid floating-point exception (B<FE_INVALID>) is raised."
2982 #: build/C/man3/fma.3:137 build/C/man3/hypot.3:141 build/C/man3/tgamma.3:134 build/C/man3/y0.3:186
2984 msgid "Range error: result underflow"
2988 #: build/C/man3/fma.3:156
2989 msgid "B<remainder>(3), B<remquo>(3)"
2993 #: build/C/man3/fmax.3:6
2999 #: build/C/man3/fmax.3:9
3000 msgid "fmax, fmaxf, fmaxl - determine maximum of two floating-point numbers"
3004 #: build/C/man3/fmax.3:13
3005 msgid "B<double fmax(double >I<x>B<, double >I<y>B<);>"
3009 #: build/C/man3/fmax.3:15
3010 msgid "B<float fmaxf(float >I<x>B<, float >I<y>B<);>"
3014 #: build/C/man3/fmax.3:17
3015 msgid "B<long double fmaxl(long double >I<x>B<, long double >I<y>B<);>"
3019 #: build/C/man3/fmax.3:29
3020 msgid "B<fmax>(), B<fmaxf>(), B<fmaxl>():"
3024 #: build/C/man3/fmax.3:42
3025 msgid "These functions return the larger value of I<x> and I<y>."
3029 #: build/C/man3/fmax.3:47
3030 msgid "These functions return the maximum of I<x> and I<y>."
3034 #: build/C/man3/fmax.3:49 build/C/man3/fmin.3:49
3035 msgid "If one argument is a NaN, the other argument is returned."
3039 #: build/C/man3/fmax.3:51 build/C/man3/fmin.3:51
3040 msgid "If both arguments are NaN, a NaN is returned."
3044 #: build/C/man3/fmax.3:58
3049 #: build/C/man3/fmin.3:6
3055 #: build/C/man3/fmin.3:9
3056 msgid "fmin, fminf, fminl - determine minimum of two floating-point numbers"
3060 #: build/C/man3/fmin.3:13
3061 msgid "B<double fmin(double >I<x>B<, double >I<y>B<);>"
3065 #: build/C/man3/fmin.3:15
3066 msgid "B<float fminf(float >I<x>B<, float >I<y>B<);>"
3070 #: build/C/man3/fmin.3:17
3071 msgid "B<long double fminl(long double >I<x>B<, long double >I<y>B<);>"
3075 #: build/C/man3/fmin.3:29
3076 msgid "B<fmin>(), B<fminf>(), B<fminl>():"
3080 #: build/C/man3/fmin.3:42
3081 msgid "These functions the lesser value of I<x> and I<y>."
3085 #: build/C/man3/fmin.3:47
3086 msgid "These functions return the minimum of I<x> and I<y>."
3090 #: build/C/man3/fmod.3:33
3096 #: build/C/man3/fmod.3:33
3102 #: build/C/man3/fmod.3:36
3103 msgid "fmod, fmodf, fmodl - floating-point remainder function"
3107 #: build/C/man3/fmod.3:41
3109 msgid "B<double fmod(double >I<x>B<, double >I<y>B<);>\n"
3113 #: build/C/man3/fmod.3:43
3115 msgid "B<float fmodf(float >I<x>B<, float >I<y>B<);>\n"
3119 #: build/C/man3/fmod.3:45
3121 msgid "B<long double fmodl(long double >I<x>B<, long double >I<y>B<);>\n"
3125 #: build/C/man3/fmod.3:57
3126 msgid "B<fmodf>(), B<fmodl>():"
3130 #: build/C/man3/fmod.3:72
3132 "The B<fmod>() function computes the floating-point remainder of dividing "
3133 "I<x> by I<y>. The return value is I<x> - I<n> * I<y>, where I<n> is the "
3134 "quotient of I<x> / I<y>, rounded toward zero to an integer."
3138 #: build/C/man3/fmod.3:80
3140 "On success, these functions return the value I<x>\\ -\\ I<n>*I<y>, for some "
3141 "integer I<n>, such that the returned value has the same sign as I<x> and a "
3142 "magnitude less than the magnitude of I<y>."
3146 #: build/C/man3/fmod.3:92
3147 msgid "If I<x> is an infinity, a domain error occurs, and a NaN is returned."
3151 #: build/C/man3/fmod.3:98
3152 msgid "If I<y> is zero, a domain error occurs, and a NaN is returned."
3156 #: build/C/man3/fmod.3:104
3157 msgid "If I<x> is +0 (-0), and I<y> is not zero, +0 (-0) is returned."
3160 #. [XXX see bug above] and \fIx\fP is not a NaN
3162 #: build/C/man3/fmod.3:120 build/C/man3/remainder.3:170
3164 msgid "Domain error: I<y> is zero"
3167 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=6784
3169 #: build/C/man3/fmod.3:144
3171 "Before version 2.10, the glibc implementation did not set I<errno> to "
3172 "B<EDOM> when a domain error occurred for an infinite I<x>."
3176 #: build/C/man3/fmod.3:145
3177 msgid "B<remainder>(3)"
3181 #: build/C/man3/fpclassify.3:6
3187 #: build/C/man3/fpclassify.3:10
3189 "fpclassify, isfinite, isnormal, isnan, isinf - floating-point classification "
3194 #: build/C/man3/fpclassify.3:15
3196 msgid "B<int fpclassify(>I<x>B<);>\n"
3200 #: build/C/man3/fpclassify.3:17
3202 msgid "B<int isfinite(>I<x>B<);>\n"
3206 #: build/C/man3/fpclassify.3:19
3208 msgid "B<int isnormal(>I<x>B<);>\n"
3212 #: build/C/man3/fpclassify.3:21
3214 msgid "B<int isnan(>I<x>B<);>\n"
3218 #: build/C/man3/fpclassify.3:23
3220 msgid "B<int isinf(>I<x>B<);>\n"
3224 #: build/C/man3/fpclassify.3:38
3225 msgid "B<fpclassify>(), B<isfinite>(), B<isnormal>():"
3229 #: build/C/man3/fpclassify.3:72
3231 "Floating point numbers can have special values, such as infinite or NaN. "
3232 "With the macro B<fpclassify(>I<x>B<)> you can find out what type I<x> is. "
3233 "The macro takes any floating-point expression as argument. The result is "
3234 "one of the following values:"
3238 #: build/C/man3/fpclassify.3:72
3244 #: build/C/man3/fpclassify.3:76
3245 msgid "I<x> is \"Not a Number\"."
3249 #: build/C/man3/fpclassify.3:76
3251 msgid "B<FP_INFINITE>"
3255 #: build/C/man3/fpclassify.3:80
3256 msgid "I<x> is either positive infinity or negative infinity."
3260 #: build/C/man3/fpclassify.3:80
3266 #: build/C/man3/fpclassify.3:84
3267 msgid "I<x> is zero."
3271 #: build/C/man3/fpclassify.3:84
3273 msgid "B<FP_SUBNORMAL>"
3277 #: build/C/man3/fpclassify.3:88
3278 msgid "I<x> is too small to be represented in normalized format."
3282 #: build/C/man3/fpclassify.3:88
3284 msgid "B<FP_NORMAL>"
3288 #: build/C/man3/fpclassify.3:92
3290 "if nothing of the above is correct then it must be a normal floating-point "
3295 #: build/C/man3/fpclassify.3:94
3296 msgid "The other macros provide a short answer to some standard questions."
3300 #: build/C/man3/fpclassify.3:94
3302 msgid "B<isfinite(>I<x>B<)>"
3306 #: build/C/man3/fpclassify.3:97
3307 msgid "returns a nonzero value if"
3311 #: build/C/man3/fpclassify.3:99
3312 msgid "(fpclassify(x) != FP_NAN && fpclassify(x) != FP_INFINITE)"
3316 #: build/C/man3/fpclassify.3:99
3318 msgid "B<isnormal(>I<x>B<)>"
3322 #: build/C/man3/fpclassify.3:103
3323 msgid "returns a nonzero value if (fpclassify(x) == FP_NORMAL)"
3327 #: build/C/man3/fpclassify.3:103
3329 msgid "B<isnan(>I<x>B<)>"
3333 #: build/C/man3/fpclassify.3:107
3334 msgid "returns a nonzero value if (fpclassify(x) == FP_NAN)"
3338 #: build/C/man3/fpclassify.3:107
3340 msgid "B<isinf(>I<x>B<)>"
3344 #: build/C/man3/fpclassify.3:114
3345 msgid "returns 1 if I<x> is positive infinity, and -1 if I<x> is negative infinity."
3349 #: build/C/man3/fpclassify.3:116
3350 msgid "C99, POSIX.1."
3354 #: build/C/man3/fpclassify.3:121
3356 "For B<isinf>(), the standards merely say that the return value is nonzero if "
3357 "and only if the argument has an infinite value."
3361 #: build/C/man3/fpclassify.3:128
3363 "In glibc 2.01 and earlier, B<isinf>() returns a nonzero value (actually: 1) "
3364 "if I<x> is positive infinity or negative infinity. (This is all that C99 "
3369 #: build/C/man3/fpclassify.3:132
3370 msgid "B<finite>(3), B<INFINITY>(3), B<isgreater>(3), B<signbit>(3)"
3374 #: build/C/man3/frexp.3:31
3380 #: build/C/man3/frexp.3:35
3382 "frexp, frexpf, frexpl - convert floating-point number to fractional and "
3383 "integral components"
3387 #: build/C/man3/frexp.3:40
3389 msgid "B<double frexp(double >I<x>B<, int *>I<exp>B<);>\n"
3393 #: build/C/man3/frexp.3:42
3395 msgid "B<float frexpf(float >I<x>B<, int *>I<exp>B<);>\n"
3399 #: build/C/man3/frexp.3:44
3401 msgid "B<long double frexpl(long double >I<x>B<, int *>I<exp>B<);>\n"
3405 #: build/C/man3/frexp.3:56
3406 msgid "B<frexpf>(), B<frexpl>():"
3410 #: build/C/man3/frexp.3:69
3412 "The B<frexp>() function is used to split the number I<x> into a normalized "
3413 "fraction and an exponent which is stored in I<exp>."
3417 #: build/C/man3/frexp.3:77
3419 "The B<frexp>() function returns the normalized fraction. If the argument "
3420 "I<x> is not zero, the normalized fraction is I<x> times a power of two, and "
3421 "its absolute value is always in the range 1/2 (inclusive) to 1 (exclusive), "
3426 #: build/C/man3/frexp.3:80
3428 "If I<x> is zero, then the normalized fraction is zero and zero is stored in "
3433 #: build/C/man3/frexp.3:87
3435 "If I<x> is a NaN, a NaN is returned, and the value of I<*exp> is "
3440 #: build/C/man3/frexp.3:94
3442 "If I<x> is positive infinity (negative infinity), positive infinity "
3443 "(negative infinity) is returned, and the value of I<*exp> is unspecified."
3447 #: build/C/man3/frexp.3:104
3448 msgid "The program below produces results such as the following:"
3452 #: build/C/man3/frexp.3:111
3455 "$B< ./a.out 2560>\n"
3456 "frexp(2560, &e) = 0.625: 0.625 * 2^12 = 2560\n"
3458 "frexp(-4, &e) = -0.5: -0.5 * 2^3 = -4\n"
3462 #: build/C/man3/frexp.3:113 build/C/man3/matherr.3:358
3464 msgid "Program source"
3468 #: build/C/man3/frexp.3:120
3471 "#include E<lt>math.hE<gt>\n"
3472 "#include E<lt>float.hE<gt>\n"
3473 "#include E<lt>stdio.hE<gt>\n"
3474 "#include E<lt>stdlib.hE<gt>\n"
3478 #: build/C/man3/frexp.3:126
3482 "main(int argc, char *argv[])\n"
3489 #: build/C/man3/frexp.3:129
3492 " x = strtod(argv[1], NULL);\n"
3493 " r = frexp(x, &exp);\n"
3497 #: build/C/man3/frexp.3:134
3500 " printf(\"frexp(%g, &e) = %g: %g * %d^%d = %g\\en\",\n"
3501 " x, r, r, FLT_RADIX, exp, x);\n"
3502 " exit(EXIT_SUCCESS);\n"
3507 #: build/C/man3/frexp.3:137
3508 msgid "B<ldexp>(3), B<modf>(3)"
3512 #: build/C/man3/gamma.3:6
3518 #: build/C/man3/gamma.3:9
3519 msgid "gamma, gammaf, gammal - (logarithm of the) gamma function"
3523 #: build/C/man3/gamma.3:13
3524 msgid "B<double gamma(double >I<x>B<);>"
3528 #: build/C/man3/gamma.3:15
3529 msgid "B<float gammaf(float >I<x>B<);>"
3533 #: build/C/man3/gamma.3:17
3534 msgid "B<long double gammal(long double >I<x>B<);>"
3538 #: build/C/man3/gamma.3:27
3543 #: build/C/man3/gamma.3:29 build/C/man3/lgamma.3:65
3544 msgid "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE"
3548 #: build/C/man3/gamma.3:32
3549 msgid "B<gammaf>(), B<gammal>():"
3553 #: build/C/man3/gamma.3:43
3555 "These functions are deprecated: instead, use either the B<tgamma>(3) or the "
3556 "B<lgamma>(3) functions, as appropriate."
3560 #: build/C/man3/gamma.3:46 build/C/man3/lgamma.3:70
3561 msgid "For the definition of the Gamma function, see B<tgamma>(3)."
3565 #: build/C/man3/gamma.3:46
3567 msgid "*BSD version"
3571 #: build/C/man3/gamma.3:50
3573 "The libm in 4.4BSD and some versions of FreeBSD had a B<gamma>() function "
3574 "that computes the Gamma function, as one would expect."
3578 #: build/C/man3/gamma.3:50
3580 msgid "glibc version"
3584 #: build/C/man3/gamma.3:56
3586 "Glibc has a B<gamma>() function that is equivalent to B<lgamma>(3) and "
3587 "computes the natural logarithm of the Gamma function."
3591 #: build/C/man3/gamma.3:59 build/C/man3/gamma.3:62
3592 msgid "See B<lgamma>(3)."
3596 #: build/C/man3/gamma.3:65
3598 "Because of historical variations in behavior across systems, this function "
3599 "is not specified in any standard."
3603 #: build/C/man3/gamma.3:66
3609 #: build/C/man3/gamma.3:78
3611 "4.2BSD had a B<gamma>() that computed ln(|Gamma(|I<x>|)|), leaving the sign "
3612 "of Gamma(|I<x>|) in the external integer I<signgam>. In 4.3BSD the name "
3613 "was changed to B<lgamma>(3), and the man page promises"
3617 #: build/C/man3/gamma.3:82
3619 "At some time in the future the name gamma will be rehabilitated and used for "
3620 "the Gamma function"
3623 #. The FreeBSD man page says about gamma() that it is like lgamma()
3624 #. except that is does not set signgam.
3625 #. Also, that 4.4BSD has a gamma() that computes the true gamma function.
3627 #: build/C/man3/gamma.3:94
3629 "This did indeed happen in 4.4BSD, where B<gamma>() computes the Gamma "
3630 "function (with no effect on I<signgam>). However, this came too late, and "
3631 "we now have B<tgamma>(3), the \"true gamma\" function."
3635 #: build/C/man3/gamma.3:97
3636 msgid "B<lgamma>(3), B<signgam>(3), B<tgamma>(3)"
3640 #: build/C/man3/hypot.3:31
3646 #: build/C/man3/hypot.3:34
3647 msgid "hypot, hypotf, hypotl - Euclidean distance function"
3651 #: build/C/man3/hypot.3:39
3653 msgid "B<double hypot(double >I<x>B<, double >I<y>B<);>\n"
3657 #: build/C/man3/hypot.3:41
3659 msgid "B<float hypotf(float >I<x>B<, float >I<y>B<);>\n"
3663 #: build/C/man3/hypot.3:43
3665 msgid "B<long double hypotl(long double >I<x>B<, long double >I<y>B<);>\n"
3669 #: build/C/man3/hypot.3:54
3674 #: build/C/man3/hypot.3:64
3675 msgid "B<hypotf>(), B<hypotl>():"
3679 #: build/C/man3/hypot.3:85
3681 "The B<hypot>() function returns sqrt(I<x>*I<x>+I<y>*I<y>). This is the "
3682 "length of the hypotenuse of a right-angled triangle with sides of length "
3683 "I<x> and I<y>, or the distance of the point (I<x>,I<y>) from the origin."
3686 #. e.g., hypot(DBL_MIN, DBL_MIN) does the right thing, as does, say
3687 #. hypot(DBL_MAX/2.0, DBL_MAX/2.0).
3689 #: build/C/man3/hypot.3:90
3691 "The calculation is performed without undue overflow or underflow during the "
3692 "intermediate steps of the calculation."
3696 #: build/C/man3/hypot.3:96
3698 "On success, these functions return the length of a right-angled triangle "
3699 "with sides of length I<x> and I<y>."
3703 #: build/C/man3/hypot.3:103
3704 msgid "If I<x> or I<y> is an infinity, positive infinity is returned."
3708 #: build/C/man3/hypot.3:111
3710 "If I<x> or I<y> is a NaN, and the other argument is not an infinity, a NaN "
3714 #. Actually, could the result not be subnormal if both arguments
3715 #. are subnormal? I think not -- mtk, Jul 2008
3717 #: build/C/man3/hypot.3:126
3719 "If both arguments are subnormal, and the result is subnormal, a range error "
3720 "occurs, and the correct result is returned."
3723 #. FIXME . Is it intentional that these functions do not set errno?
3724 #. They do set errno for the y == 0 case, below.
3725 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6783
3727 #: build/C/man3/hypot.3:156 build/C/man3/ilogb.3:128 build/C/man3/remainder.3:170
3728 msgid "These functions do not set I<errno> for this case."
3732 #: build/C/man3/hypot.3:164
3733 msgid "B<cabs>(3), B<sqrt>(3)"
3737 #: build/C/man3/ilogb.3:27
3743 #: build/C/man3/ilogb.3:30
3744 msgid "ilogb, ilogbf, ilogbl - get integer exponent of a floating-point value"
3748 #: build/C/man3/ilogb.3:34
3749 msgid "B<int ilogb(double >I<x>B<);>"
3753 #: build/C/man3/ilogb.3:36
3754 msgid "B<int ilogbf(float >I<x>B<);>"
3758 #: build/C/man3/ilogb.3:38
3759 msgid "B<int ilogbl(long double >I<x>B<);>"
3763 #: build/C/man3/ilogb.3:48
3768 #: build/C/man3/ilogb.3:60
3769 msgid "B<ilogbf>(), B<ilogbl>():"
3773 #: build/C/man3/ilogb.3:76
3775 "These functions return the exponent part of their argument as a signed "
3776 "integer. When no error occurs, these functions are equivalent to the "
3777 "corresponding B<logb>(3) functions, cast to I<int>."
3781 #: build/C/man3/ilogb.3:80
3783 "On success, these functions return the exponent of I<x>, as a signed "
3787 #. the POSIX.1 spec for logb() says logb() gives pole error for this
3788 #. case, but for ilogb() it says domain error.
3789 #. glibc: The numeric value is either `INT_MIN' or `-INT_MAX'.
3791 #: build/C/man3/ilogb.3:88
3793 "If I<x> is zero, then a domain error occurs, and the functions return "
3797 #. glibc: The numeric value is either `INT_MIN' or `INT_MAX'.
3798 #. On i386, FP_ILOGB0 and FP_ILOGBNAN have the same value.
3800 #: build/C/man3/ilogb.3:95
3802 "If I<x> is a NaN, then a domain error occurs, and the functions return "
3807 #. POSIX.1-2001 also says:
3808 #. If the correct value is greater than {INT_MAX}, {INT_MAX}
3809 #. shall be returned and a domain error shall occur.
3811 #. If the correct value is less than {INT_MIN}, {INT_MIN}
3812 #. shall be returned and a domain error shall occur.
3814 #: build/C/man3/ilogb.3:108
3816 "If I<x> is negative infinity or positive infinity, then a domain error "
3817 "occurs, and the functions return B<INT_MAX>."
3821 #: build/C/man3/ilogb.3:115
3823 msgid "Domain error: I<x> is 0 or a NaN"
3826 #. FIXME . Is it intentional that these functions do not set errno,
3827 #. or raise an exception?
3828 #. log(), log2(), log10() do set errno
3829 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6794
3831 #: build/C/man3/ilogb.3:144
3832 msgid "These functions do not set I<errno> or raise an exception for this case."
3836 #: build/C/man3/ilogb.3:149
3837 msgid "B<log>(3), B<logb>(3), B<significand>(3)"
3841 #: build/C/man3/infnan.3:31
3847 #: build/C/man3/infnan.3:31
3853 #: build/C/man3/infnan.3:34
3854 msgid "infnan - deal with infinite or not-a-number (NaN) result"
3858 #: build/C/man3/infnan.3:39
3860 msgid "B<double infnan(int >I<error>B<);>\n"
3864 #: build/C/man3/infnan.3:52
3866 "The B<infnan>() function returns a suitable value for infinity and "
3867 "\"not-a-number\" (NaN) results. The value of I<error> can be B<ERANGE> to "
3868 "represent infinity or anything else to represent NaN. I<errno> is also set."
3872 #: build/C/man3/infnan.3:58
3873 msgid "If I<error> is B<ERANGE> (Infinity), B<HUGE_VAL> is returned."
3877 #: build/C/man3/infnan.3:60
3878 msgid "If I<error> is -B<ERANGE> (-Infinity), -B<HUGE_VAL> is returned."
3882 #: build/C/man3/infnan.3:62
3883 msgid "If I<error> is anything else, NAN is returned."
3887 #: build/C/man3/infnan.3:63
3893 #: build/C/man3/infnan.3:66
3894 msgid "The value of I<error> is \"not-a-number\" (NaN)."
3898 #: build/C/man3/infnan.3:66
3904 #: build/C/man3/infnan.3:69
3905 msgid "The value of I<error> is positive infinity or negative infinity."
3909 #: build/C/man3/infnan.3:71
3914 #: build/C/man3/infnan.3:73
3916 "This obsolete function was provided in libc4 and libc5, but is not available "
3921 #: build/C/man3/isgreater.3:6
3927 #: build/C/man3/isgreater.3:10
3929 "isgreater, isgreaterequal, isless, islessequal, islessgreater, isunordered - "
3930 "floating-point relational tests without exception for NaN"
3934 #: build/C/man3/isgreater.3:15
3936 msgid "B<int isgreater(>I<x>B<, >I<y>B<);>\n"
3940 #: build/C/man3/isgreater.3:17
3942 msgid "B<int isgreaterequal(>I<x>B<, >I<y>B<);>\n"
3946 #: build/C/man3/isgreater.3:19
3948 msgid "B<int isless(>I<x>B<, >I<y>B<);>\n"
3952 #: build/C/man3/isgreater.3:21
3954 msgid "B<int islessequal(>I<x>B<, >I<y>B<);>\n"
3958 #: build/C/man3/isgreater.3:23
3960 msgid "B<int islessgreater(>I<x>B<, >I<y>B<);>\n"
3964 #: build/C/man3/isgreater.3:25
3966 msgid "B<int isunordered(>I<x>B<, >I<y>B<);>\n"
3970 #: build/C/man3/isgreater.3:36
3971 msgid "All functions described here:"
3975 #: build/C/man3/isgreater.3:51
3977 "The normal relation operations (like B<E<lt>>, \"less than\") will fail if "
3978 "one of the operands is NaN. This will cause an exception. To avoid this, "
3979 "C99 defines these macros. The macros are guaranteed to evaluate their "
3980 "operands only once. The operands can be of any real floating-point type."
3984 #: build/C/man3/isgreater.3:51
3986 msgid "B<isgreater>()"
3990 #: build/C/man3/isgreater.3:55
3991 msgid "determines I<(x)\\ E<gt>\\ (y)> without an exception if I<x> or I<y> is NaN."
3995 #: build/C/man3/isgreater.3:55
3997 msgid "B<isgreaterequal>()"
4001 #: build/C/man3/isgreater.3:59
4003 "determines I<(x)\\ E<gt>=\\ (y)> without an exception if I<x> or I<y> is "
4008 #: build/C/man3/isgreater.3:59
4014 #: build/C/man3/isgreater.3:63
4015 msgid "determines I<(x)\\ E<lt>\\ (y)> without an exception if I<x> or I<y> is NaN."
4019 #: build/C/man3/isgreater.3:63
4021 msgid "B<islessequal>()"
4025 #: build/C/man3/isgreater.3:67
4027 "determines I<(x)\\ E<lt>=\\ (y)> without an exception if I<x> or I<y> is "
4032 #: build/C/man3/isgreater.3:67
4034 msgid "B<islessgreater>()"
4038 #: build/C/man3/isgreater.3:73
4040 "determines I<(x)\\ E<lt> (y) || (x) E<gt>\\ (y)> without an exception if "
4041 "I<x> or I<y> is NaN. This macro is not equivalent to I<x\\ !=\\ y> because "
4042 "that expression is true if I<x> or I<y> is NaN."
4046 #: build/C/man3/isgreater.3:73
4048 msgid "B<isunordered>()"
4052 #: build/C/man3/isgreater.3:77
4054 "determines whether its arguments are unordered, that is, whether at least "
4055 "one of the arguments is a NaN."
4059 #: build/C/man3/isgreater.3:82
4061 "The macros other than B<isunordered>() return the result of the relational "
4062 "comparison; these macros return 0 if either argument is a NaN."
4066 #: build/C/man3/isgreater.3:85
4067 msgid "B<isunordered>() returns 1 if I<x> or I<y> is NaN and 0 otherwise."
4071 #: build/C/man3/isgreater.3:94
4073 "Not all hardware supports these functions, and where hardware support isn't "
4074 "provided, they will be emulated by macros. This will result in a "
4075 "performance penalty. Don't use these functions if NaN is of no concern for "
4080 #: build/C/man3/isgreater.3:96
4081 msgid "B<fpclassify>(3), B<isnan>(3)"
4085 #: build/C/man3/j0.3:34
4091 #: build/C/man3/j0.3:38
4093 "j0, j0f, j0l, j1, j1f, j1l, jn, jnf, jnl - Bessel functions of the first "
4098 #: build/C/man3/j0.3:44
4099 msgid "B<double j0(double >I<x>B<);>"
4103 #: build/C/man3/j0.3:46
4104 msgid "B<double j1(double >I<x>B<);>"
4108 #: build/C/man3/j0.3:48
4109 msgid "B<double jn(int >I<n>B<, double >I<x>B<);>"
4113 #: build/C/man3/j0.3:50
4114 msgid "B<float j0f(float >I<x>B<);>"
4118 #: build/C/man3/j0.3:52
4119 msgid "B<float j1f(float >I<x>B<);>"
4123 #: build/C/man3/j0.3:54
4124 msgid "B<float jnf(int >I<n>B<, float >I<x>B<);>"
4128 #: build/C/man3/j0.3:56
4129 msgid "B<long double j0l(long double >I<x>B<);>"
4133 #: build/C/man3/j0.3:58
4134 msgid "B<long double j1l(long double >I<x>B<);>"
4138 #: build/C/man3/j0.3:60
4139 msgid "B<long double jnl(int >I<n>B<, long double >I<x>B<);>"
4143 #: build/C/man3/j0.3:72
4144 msgid "B<j0>(), B<j1>(), B<jn>():"
4148 #: build/C/man3/j0.3:74 build/C/man3/y0.3:74
4149 msgid "_SVID_SOURCE || _BSD_SOURCE || _XOPEN_SOURCE"
4153 #: build/C/man3/j0.3:82
4154 msgid "B<j0f>(), B<j0l>(), B<j1f>(), B<j1l>(), B<jnf>(), B<jnl>():"
4157 #. Also seems to work: -std=c99 -D_XOPEN_SOURCE
4159 #: build/C/man3/j0.3:85 build/C/man3/y0.3:85
4160 msgid "_SVID_SOURCE || _BSD_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 600"
4164 #: build/C/man3/j0.3:98
4166 "The B<j0>() and B<j1>() functions return Bessel functions of I<x> of the "
4167 "first kind of orders 0 and 1, respectively. The B<jn>() function returns "
4168 "the Bessel function of I<x> of the first kind of order I<n>."
4172 #: build/C/man3/j0.3:108
4174 "The B<j0f>() etc. and B<j0l>() etc. functions are versions that take and "
4175 "return I<float> and I<long double> values, respectively."
4179 #: build/C/man3/j0.3:112
4181 "On success, these functions return the appropriate Bessel value of the first "
4186 #: build/C/man3/j0.3:123
4188 "If I<x> is too large in magnitude, or the result underflows, a range error "
4189 "occurs, and the return value is 0."
4193 #: build/C/man3/j0.3:130
4195 msgid "Range error: result underflow, or I<x> is too large in magnitude"
4198 #. An underflow floating-point exception
4199 #. .RB ( FE_UNDERFLOW )
4202 #: build/C/man3/j0.3:138
4203 msgid "I<errno> is set to B<ERANGE>."
4206 #. FIXME . Is it intentional that these functions do not raise exceptions?
4208 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6805
4210 #: build/C/man3/j0.3:144
4211 msgid "These functions do not raise exceptions for B<fetestexcept>(3)."
4215 #: build/C/man3/j0.3:150 build/C/man3/y0.3:221
4217 "The functions returning I<double> conform to SVr4, 4.3BSD, POSIX.1-2001. "
4218 "The others are nonstandard functions that also exist on the BSDs."
4222 #: build/C/man3/j0.3:157
4224 "There are errors of up to 2e-16 in the values returned by B<j0>(), B<j1>() "
4225 "and B<jn>() for values of I<x> between -8 and 8."
4229 #: build/C/man3/j0.3:158
4234 #: build/C/man3/ldexp.3:32
4240 #: build/C/man3/ldexp.3:35
4242 "ldexp, ldexpf, ldexpl - multiply floating-point number by integral power of "
4247 #: build/C/man3/ldexp.3:40
4249 msgid "B<double ldexp(double >I<x>B<, int >I<exp>B<);>\n"
4253 #: build/C/man3/ldexp.3:42
4255 msgid "B<float ldexpf(float >I<x>B<, int >I<exp>B<);>\n"
4259 #: build/C/man3/ldexp.3:44
4261 msgid "B<long double ldexpl(long double >I<x>B<, int >I<exp>B<);>\n"
4265 #: build/C/man3/ldexp.3:56
4266 msgid "B<ldexpf>(), B<ldexpl>():"
4270 #: build/C/man3/ldexp.3:71
4272 "The B<ldexp>() function returns the result of multiplying the "
4273 "floating-point number I<x> by 2 raised to the power I<exp>."
4277 #: build/C/man3/ldexp.3:74
4278 msgid "On success, these functions return I<x * (2^exp)>."
4282 #: build/C/man3/ldexp.3:80
4283 msgid "If I<exp> is zero, then I<x> is returned."
4287 #: build/C/man3/ldexp.3:104 build/C/man3/scalb.3:126 build/C/man3/scalbln.3:118
4289 "If the result overflows, a range error occurs, and the functions return "
4290 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, with a sign the "
4295 #: build/C/man3/ldexp.3:136
4296 msgid "B<frexp>(3), B<modf>(3), B<scalbln>(3)"
4300 #: build/C/man3/lgamma.3:7
4306 #: build/C/man3/lgamma.3:11
4308 "lgamma, lgammaf, lgammal, lgamma_r, lgammaf_r, lgammal_r, signgam - log "
4313 #: build/C/man3/lgamma.3:16
4315 msgid "B<double lgamma(double >I<x>B<);>\n"
4319 #: build/C/man3/lgamma.3:18
4321 msgid "B<float lgammaf(float >I<x>B<);>\n"
4325 #: build/C/man3/lgamma.3:20
4327 msgid "B<long double lgammal(long double >I<x>B<);>\n"
4331 #: build/C/man3/lgamma.3:22
4333 msgid "B<double lgamma_r(double >I<x>B<, int *>I<signp>B<);>\n"
4337 #: build/C/man3/lgamma.3:24
4339 msgid "B<float lgammaf_r(float >I<x>B<, int *>I<signp>B<);>\n"
4343 #: build/C/man3/lgamma.3:26
4345 msgid "B<long double lgammal_r(long double >I<x>B<, int *>I<signp>B<);>\n"
4349 #: build/C/man3/lgamma.3:28
4351 msgid "B<extern int >I<signgam>B<;>\n"
4355 #: build/C/man3/lgamma.3:39
4356 msgid "B<lgamma>():"
4360 #: build/C/man3/lgamma.3:49
4361 msgid "B<lgammaf>(), B<lgammal>():"
4365 #: build/C/man3/lgamma.3:59
4366 msgid "B<lgamma_r>(), B<lgammaf_r>(), B<lgammal_r>():"
4370 #: build/C/man3/lgamma.3:63
4375 #: build/C/man3/lgamma.3:80
4377 "The B<lgamma>() function returns the natural logarithm of the absolute "
4378 "value of the Gamma function. The sign of the Gamma function is returned in "
4379 "the external integer I<signgam> declared in I<E<lt>math.hE<gt>>. It is 1 "
4380 "when the Gamma function is positive or zero, -1 when it is negative."
4384 #: build/C/man3/lgamma.3:88
4386 "Since using a constant location I<signgam> is not thread-safe, the functions "
4387 "B<lgamma_r>() etc. have been introduced; they return the sign via the "
4388 "argument I<signp>."
4392 #: build/C/man3/lgamma.3:90
4393 msgid "On success, these functions return the natural logarithm of Gamma(x)."
4397 #: build/C/man3/lgamma.3:98
4398 msgid "If I<x> is 1 or 2, +0 is returned."
4402 #: build/C/man3/lgamma.3:114
4404 "If I<x> is a nonpositive integer, a pole error occurs, and the functions "
4405 "return +B<HUGE_VAL>, +B<HUGE_VALF>, or +B<HUGE_VALL>, respectively."
4409 #: build/C/man3/lgamma.3:124 build/C/man3/tgamma.3:83
4411 "If the result overflows, a range error occurs, and the functions return "
4412 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, with the correct "
4413 "mathematical sign."
4417 #: build/C/man3/lgamma.3:131
4419 msgid "Pole error: I<x> is a nonpositive integer"
4423 #: build/C/man3/lgamma.3:159
4425 "The B<lgamma>() functions are specified in C99 and POSIX.1-2001. "
4426 "I<signgam> is specified in POSIX.1-2001, but not in C99. The B<lgamma_r>() "
4427 "functions are nonstandard, but present on several other systems."
4430 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=6777
4432 #: build/C/man3/lgamma.3:169
4434 "In glibc 2.9 and earlier, when a pole error occurs, I<errno> is set to "
4435 "B<EDOM>; instead of the POSIX-mandated B<ERANGE>. Since version 2.10, glibc "
4436 "does the right thing."
4440 #: build/C/man3/lgamma.3:170
4441 msgid "B<tgamma>(3)"
4445 #: build/C/man3/log.3:34
4451 #: build/C/man3/log.3:37
4452 msgid "log, logf, logl - natural logarithmic function"
4456 #: build/C/man3/log.3:42
4458 msgid "B<double log(double >I<x>B<);>\n"
4462 #: build/C/man3/log.3:44
4464 msgid "B<float logf(float >I<x>B<);>\n"
4468 #: build/C/man3/log.3:46
4470 msgid "B<long double logl(long double >I<x>B<);>\n"
4474 #: build/C/man3/log.3:58
4475 msgid "B<logf>(), B<logl>():"
4479 #: build/C/man3/log.3:70
4480 msgid "The B<log>() function returns the natural logarithm of I<x>."
4484 #: build/C/man3/log.3:73
4485 msgid "On success, these functions return the natural logarithm of I<x>."
4489 #: build/C/man3/log.3:82
4490 msgid "If I<x> is 1, the result is +0."
4494 #: build/C/man3/log.3:97 build/C/man3/logb.3:109
4496 "If I<x> is zero, then a pole error occurs, and the functions return "
4497 "-B<HUGE_VAL>, -B<HUGE_VALF>, or -B<HUGE_VALL>, respectively."
4501 #: build/C/man3/log.3:102
4503 "If I<x> is negative (including negative infinity), then a domain error "
4504 "occurs, and a NaN (not a number) is returned."
4508 #: build/C/man3/log.3:109 build/C/man3/y0.3:163
4510 msgid "Domain error: I<x> is negative"
4514 #: build/C/man3/log.3:117
4516 msgid "Pole error: I<x> is zero"
4520 #: build/C/man3/log.3:125 build/C/man3/tgamma.3:123
4522 "I<errno> is set to B<ERANGE>. A divide-by-zero floating-point exception "
4523 "(B<FE_DIVBYZERO>) is raised."
4527 #: build/C/man3/log.3:138
4529 "In glibc 2.5 and earlier, taking the B<log>() of a NaN produces a bogus "
4530 "invalid floating-point (B<FE_INVALID>) exception."
4534 #: build/C/man3/log.3:144
4535 msgid "B<cbrt>(3), B<clog>(3), B<log10>(3), B<log1p>(3), B<log2>(3), B<sqrt>(3)"
4539 #: build/C/man3/log10.3:34
4545 #: build/C/man3/log10.3:37
4546 msgid "log10, log10f, log10l - base-10 logarithmic function"
4550 #: build/C/man3/log10.3:42
4552 msgid "B<double log10(double >I<x>B<);>\n"
4556 #: build/C/man3/log10.3:44
4558 msgid "B<float log10f(float >I<x>B<);>\n"
4562 #: build/C/man3/log10.3:46
4564 msgid "B<long double log10l(long double >I<x>B<);>\n"
4568 #: build/C/man3/log10.3:58
4569 msgid "B<log10f>(), B<log10l>():"
4573 #: build/C/man3/log10.3:70
4574 msgid "The B<log10>() function returns the base 10 logarithm of I<x>."
4578 #: build/C/man3/log10.3:73
4579 msgid "On success, these functions return the base 10 logarithm of I<x>."
4583 #: build/C/man3/log10.3:78 build/C/man3/log2.3:78
4585 "For special cases, including where I<x> is 0, 1, negative, infinity, or NaN, "
4590 #: build/C/man3/log10.3:86 build/C/man3/log2.3:86
4592 "For a discussion of the errors that can occur for these functions, see "
4597 #: build/C/man3/log10.3:98
4598 msgid "B<cbrt>(3), B<clog10>(3), B<exp10>(3), B<log>(3), B<log2>(3), B<sqrt>(3)"
4602 #: build/C/man3/log1p.3:27
4608 #: build/C/man3/log1p.3:30
4609 msgid "log1p, log1pf, log1pl - logarithm of 1 plus argument"
4613 #: build/C/man3/log1p.3:35
4615 msgid "B<double log1p(double >I<x>B<);>\n"
4619 #: build/C/man3/log1p.3:37
4621 msgid "B<float log1pf(float >I<x>B<);>\n"
4625 #: build/C/man3/log1p.3:39
4627 msgid "B<long double log1pl(long double >I<x>B<);>\n"
4631 #: build/C/man3/log1p.3:50
4636 #: build/C/man3/log1p.3:61
4637 msgid "B<log1pf>(), B<log1pl>():"
4641 #: build/C/man3/log1p.3:72
4642 msgid "I<log1p(x)> returns a value equivalent to"
4646 #: build/C/man3/log1p.3:75
4648 msgid " log (1 + I<x>)\n"
4652 #: build/C/man3/log1p.3:79
4654 "It is computed in a way that is accurate even if the value of I<x> is near "
4659 #: build/C/man3/log1p.3:82
4660 msgid "On success, these functions return the natural logarithm of I<(1\\ +\\ x)>."
4664 #: build/C/man3/log1p.3:101
4666 "If I<x> is -1, a pole error occurs, and the functions return -B<HUGE_VAL>, "
4667 "-B<HUGE_VALF>, or -B<HUGE_VALL>, respectively."
4670 #. POSIX.1 specifies a possible range error if x is subnormal
4671 #. glibc 2.8 doesn't do this
4673 #: build/C/man3/log1p.3:109
4675 "If I<x> is less than -1 (including negative infinity), a domain error "
4676 "occurs, and a NaN (not a number) is returned."
4680 #: build/C/man3/log1p.3:116
4682 msgid "Domain error: I<x> is less than -1"
4686 #: build/C/man3/log1p.3:124
4688 msgid "Pole error: I<x> is -1"
4695 #: build/C/man3/log1p.3:132 build/C/man3/logb.3:129
4696 msgid "A divide-by-zero floating-point exception (B<FE_DIVBYZERO>) is raised."
4700 #: build/C/man3/log1p.3:144
4701 msgid "B<exp>(3), B<expm1>(3), B<log>(3)"
4705 #: build/C/man3/log2.3:34
4711 #: build/C/man3/log2.3:37
4712 msgid "log2, log2f, log2l - base-2 logarithmic function"
4716 #: build/C/man3/log2.3:42
4718 msgid "B<double log2(double >I<x>B<);>\n"
4722 #: build/C/man3/log2.3:44
4724 msgid "B<float log2f(float >I<x>B<);>\n"
4728 #: build/C/man3/log2.3:46
4730 msgid "B<long double log2l(long double >I<x>B<);>\n"
4734 #: build/C/man3/log2.3:59
4735 msgid "B<log2>(), B<log2f>(), B<log2l>():"
4739 #: build/C/man3/log2.3:70
4740 msgid "The B<log2>() function returns the base 2 logarithm of I<x>."
4744 #: build/C/man3/log2.3:73
4745 msgid "On success, these functions return the base 2 logarithm of I<x>."
4749 #: build/C/man3/log2.3:99
4750 msgid "B<cbrt>(3), B<clog2>(3), B<log>(3), B<log10>(3), B<sqrt>(3)"
4754 #: build/C/man3/logb.3:27
4760 #: build/C/man3/logb.3:30
4761 msgid "logb, logbf, logbl - get exponent of a floating-point value"
4765 #: build/C/man3/logb.3:34
4766 msgid "B<double logb(double >I<x>B<);>"
4770 #: build/C/man3/logb.3:36
4771 msgid "B<float logbf(float >I<x>B<);>"
4775 #: build/C/man3/logb.3:38
4776 msgid "B<long double logbl(long double >I<x>B<);>"
4780 #: build/C/man3/logb.3:48
4785 #: build/C/man3/logb.3:59
4786 msgid "B<logbf>(), B<logbl>():"
4790 #: build/C/man3/logb.3:84
4792 "These functions extract the exponent from the internal floating-point "
4793 "representation of I<x> and return it as a floating-point value. The integer "
4794 "constant B<FLT_RADIX>, defined in I<E<lt>float.hE<gt>>, indicates the radix "
4795 "used for the system's floating-point representation. If B<FLT_RADIX> is 2, "
4796 "B<logb(>I<x>B<)> is equal to B<floor(log2(>I<x>B<))>, except that it is "
4801 #: build/C/man3/logb.3:92
4803 "If I<x> is subnormal, B<logb>() returns the exponent I<x> would have if it "
4808 #: build/C/man3/logb.3:95
4809 msgid "On success, these functions return the exponent of I<x>."
4813 #: build/C/man3/logb.3:114
4815 "If I<x> is negative infinity or positive infinity, then positive infinity is "
4820 #: build/C/man3/logb.3:121
4822 msgid "Pole error: I<x> is 0"
4826 #: build/C/man3/logb.3:145
4827 msgid "B<ilogb>(3), B<log>(3)"
4831 #: build/C/man3/lrint.3:25
4837 #: build/C/man3/lrint.3:28
4838 msgid "lrint, lrintf, lrintl, llrint, llrintf, llrintl - round to nearest integer"
4842 #: build/C/man3/lrint.3:33
4844 msgid "B<long int lrint(double >I<x>B<);>\n"
4848 #: build/C/man3/lrint.3:35
4850 msgid "B<long int lrintf(float >I<x>B<);>\n"
4854 #: build/C/man3/lrint.3:37
4856 msgid "B<long int lrintl(long double >I<x>B<);>\n"
4860 #: build/C/man3/lrint.3:39
4862 msgid "B<long long int llrint(double >I<x>B<);>\n"
4866 #: build/C/man3/lrint.3:41
4868 msgid "B<long long int llrintf(float >I<x>B<);>\n"
4872 #: build/C/man3/lrint.3:43
4874 msgid "B<long long int llrintl(long double >I<x>B<);>\n"
4878 #: build/C/man3/lrint.3:54 build/C/man3/lround.3:55
4879 msgid "All functions shown above:"
4883 #: build/C/man3/lrint.3:66
4885 "These functions round their argument to the nearest integer value, using the "
4886 "current rounding direction (see B<fesetround>(3))."
4890 #: build/C/man3/lrint.3:71
4892 "Note that unlike B<rint>(3), etc., the return type of these functions "
4893 "differs from that of their arguments."
4897 #: build/C/man3/lrint.3:73 build/C/man3/lround.3:76 build/C/man3/rint.3:105 build/C/man3/round.3:73 build/C/man3/trunc.3:61
4898 msgid "These functions return the rounded integer value."
4901 #. The return value is -(LONG_MAX - 1) or -(LLONG_MAX -1)
4903 #: build/C/man3/lrint.3:85 build/C/man3/lround.3:88
4905 "If I<x> is a NaN or an infinity, or the rounded value is too large to be "
4906 "stored in a I<long> (I<long long> in the case of the B<ll*> functions) then "
4907 "a domain error occurs, and the return value is unspecified."
4911 #: build/C/man3/lrint.3:92 build/C/man3/lround.3:95
4913 msgid "Domain error: I<x> is a NaN or infinite, or the rounded value is too large"
4917 #: build/C/man3/lrint.3:115
4919 "B<ceil>(3), B<floor>(3), B<lround>(3), B<nearbyint>(3), B<rint>(3), "
4924 #: build/C/man3/lround.3:25
4930 #: build/C/man3/lround.3:29
4932 "lround, lroundf, lroundl, llround, llroundf, llroundl - round to nearest "
4933 "integer, away from zero"
4937 #: build/C/man3/lround.3:34
4939 msgid "B<long int lround(double >I<x>B<);>\n"
4943 #: build/C/man3/lround.3:36
4945 msgid "B<long int lroundf(float >I<x>B<);>\n"
4949 #: build/C/man3/lround.3:38
4951 msgid "B<long int lroundl(long double >I<x>B<);>\n"
4955 #: build/C/man3/lround.3:40
4957 msgid "B<long long int llround(double >I<x>B<);>\n"
4961 #: build/C/man3/lround.3:42
4963 msgid "B<long long int llroundf(float >I<x>B<);>\n"
4967 #: build/C/man3/lround.3:44
4969 msgid "B<long long int llroundl(long double >I<x>B<);>\n"
4973 #: build/C/man3/lround.3:68
4975 "These functions round their argument to the nearest integer value, rounding "
4976 "away from zero, regardless of the current rounding direction (see "
4981 #: build/C/man3/lround.3:74
4983 "Note that unlike B<round>(3), B<ceil>(3), etc., the return type of these "
4984 "functions differs from that of their arguments."
4988 #: build/C/man3/lround.3:118 build/C/man3/trunc.3:84
4990 "B<ceil>(3), B<floor>(3), B<lrint>(3), B<nearbyint>(3), B<rint>(3), "
4995 #: build/C/man7/math_error.7:24
5001 #: build/C/man7/math_error.7:27
5002 msgid "math_error - detecting errors from mathematical functions"
5006 #: build/C/man7/math_error.7:32
5009 "B<#include E<lt>math.hE<gt>>\n"
5010 "B<#include E<lt>errno.hE<gt>>\n"
5011 "B<#include E<lt>fenv.hE<gt>>\n"
5015 #: build/C/man7/math_error.7:51
5017 "When an error occurs, most library functions indicate this fact by returning "
5018 "a special value (e.g., -1 or NULL). Because they typically return a "
5019 "floating-point number, the mathematical functions declared in "
5020 "I<E<lt>math.hE<gt>> indicate an error using other mechanisms. There are two "
5021 "error-reporting mechanisms: the older one sets I<errno>; the newer one uses "
5022 "the floating-point exception mechanism (the use of B<feclearexcept>(3) and "
5023 "B<fetestexcept>(3), as outlined below) described in B<fenv>(3)."
5027 #: build/C/man7/math_error.7:56
5029 "A portable program that needs to check for an error from a mathematical "
5030 "function should set I<errno> to zero, and make the following call"
5034 #: build/C/man7/math_error.7:60
5036 msgid "feclearexcept(FE_ALL_EXCEPT);\n"
5040 #: build/C/man7/math_error.7:64
5041 msgid "before calling a mathematical function."
5045 #: build/C/man7/math_error.7:70
5047 "Upon return from the mathematical function, if I<errno> is nonzero, or the "
5048 "following call (see B<fenv>(3)) returns nonzero"
5052 #: build/C/man7/math_error.7:75
5055 "fetestexcept(FE_INVALID | FE_DIVBYZERO | FE_OVERFLOW |\n"
5061 #. FE_INVALID = 0x01,
5062 #. __FE_DENORM = 0x02,
5063 #. FE_DIVBYZERO = 0x04,
5064 #. FE_OVERFLOW = 0x08,
5065 #. FE_UNDERFLOW = 0x10,
5066 #. FE_INEXACT = 0x20
5069 #: build/C/man7/math_error.7:88
5070 msgid "then an error occurred in the mathematical function."
5074 #: build/C/man7/math_error.7:91
5076 "The error conditions that can occur for mathematical functions are described "
5081 #: build/C/man7/math_error.7:91
5083 msgid "Domain Error"
5087 #: build/C/man7/math_error.7:107
5089 "A I<domain error> occurs when a mathematical function is supplied with an "
5090 "argument whose value falls outside the domain for which the function is "
5091 "defined (e.g., giving a negative argument to B<log>(3)). When a domain "
5092 "error occurs, math functions commonly return a NaN (though some functions "
5093 "return a different value in this case); I<errno> is set to B<EDOM>, and an "
5094 "\"invalid\" (B<FE_INVALID>) floating-point exception is raised."
5098 #: build/C/man7/math_error.7:107
5104 #: build/C/man7/math_error.7:131
5106 "A I<pole error> occurs when the mathematical result of a function is an "
5107 "exact infinity (e.g., the logarithm of 0 is negative infinity). When a pole "
5108 "error occurs, the function returns the (signed) value B<HUGE_VAL>, "
5109 "B<HUGE_VALF>, or B<HUGE_VALL>, depending on whether the function result type "
5110 "is I<double>, I<float>, or I<long double>. The sign of the result is that "
5111 "which is mathematically correct for the function. I<errno> is set to "
5112 "B<ERANGE>, and a \"divide-by-zero\" (B<FE_DIVBYZERO>) floating-point "
5113 "exception is raised."
5117 #: build/C/man7/math_error.7:131
5123 #: build/C/man7/math_error.7:138
5125 "A I<range error> occurs when the magnitude of the function result means that "
5126 "it cannot be represented in the result type of the function. The return "
5127 "value of the function depends on whether the range error was an overflow or "
5132 #: build/C/man7/math_error.7:160
5134 "A floating result I<overflows> if the result is finite, but is too large to "
5135 "represented in the result type. When an overflow occurs, the function "
5136 "returns the value B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, depending on "
5137 "whether the function result type is I<double>, I<float>, or I<long double>. "
5138 "I<errno> is set to B<ERANGE>, and an \"overflow\" (B<FE_OVERFLOW>) "
5139 "floating-point exception is raised."
5143 #: build/C/man7/math_error.7:175
5145 "A floating result I<underflows> if the result is too small to be represented "
5146 "in the result type. If an underflow occurs, a mathematical function "
5147 "typically returns 0.0 (C99 says a function shall return \"an "
5148 "implementation-defined value whose magnitude is no greater than the smallest "
5149 "normalized positive number in the specified type\"). I<errno> may be set to "
5150 "B<ERANGE>, and an \"overflow\" (B<FE_UNDERFLOW>) floating-point exception "
5155 #: build/C/man7/math_error.7:185
5157 "Some functions deliver a range error if the supplied argument value, or the "
5158 "correct function result, would be I<subnormal>. A subnormal value is one "
5159 "that is nonzero, but with a magnitude that is so small that it can't be "
5160 "presented in normalized form (i.e., with a 1 in the most significant bit of "
5161 "the significand). The representation of a subnormal number will contain one "
5162 "or more leading zeros in the significand."
5165 #. See CONFORMANCE in the glibc 2.8 (and earlier) source.
5167 #: build/C/man7/math_error.7:207
5169 "The I<math_errhandling> identifier specified by C99 and POSIX.1-2001 is not "
5170 "supported by glibc. This identifier is supposed to indicate which of the "
5171 "two error-notification mechanisms (I<errno>, exceptions retrievable via "
5172 "B<fettestexcept>(3)) is in use. The standards require that at least one be "
5173 "in use, but permit both to be available. The current (version 2.8) "
5174 "situation under glibc is messy. Most (but not all) functions raise "
5175 "exceptions on errors. Some also set I<errno>. A few functions set "
5176 "I<errno>, but don't raise an exception. A very few functions do neither. "
5177 "See the individual manual pages for details."
5180 #. http://www.securecoding.cert.org/confluence/display/seccode/FLP32-C.+Prevent+or+detect+domain+and+range+errors+in+math+functions
5182 #: build/C/man7/math_error.7:220
5184 "To avoid the complexities of using I<errno> and B<fetestexcept>(3) for "
5185 "error checking, it is often advised that one should instead check for bad "
5186 "argument values before each call. For example, the following code ensures "
5187 "that B<log>(3)'s argument is not a NaN and is not zero (a pole error) or "
5188 "less than zero (a domain error):"
5192 #: build/C/man7/math_error.7:224
5194 msgid "double x, r;\n"
5198 #: build/C/man7/math_error.7:228
5201 "if (isnan(x) || islessequal(x, 0)) {\n"
5202 " /* Deal with NaN / pole error / domain error */\n"
5207 #: build/C/man7/math_error.7:230
5209 msgid "r = log(x);\n"
5213 #: build/C/man7/math_error.7:238
5215 "The discussion on this page does not apply to the complex mathematical "
5216 "functions (i.e., those declared by I<E<lt>complex.hE<gt>>), which in general "
5217 "are not required to return errors by C99 and POSIX.1-2001."
5221 #: build/C/man7/math_error.7:254
5223 "The B<gcc>(1) I<-fno-math-errno> option causes the executable to employ "
5224 "implementations of some mathematical functions that are faster than the "
5225 "standard implementations, but do not set I<errno> on error. (The B<gcc>(1) "
5226 "I<-ffast-math> option also enables I<-fno-math-errno>.) An error can still "
5227 "be tested for using B<fetestexcept>(3)."
5231 #: build/C/man7/math_error.7:263
5233 "B<gcc>(1), B<errno>(3), B<fenv>(3), B<fpclassify>(3), B<INFINITY>(3), "
5234 "B<isgreater>(3), B<matherr>(3), B<nan>(3)"
5238 #: build/C/man7/math_error.7:264
5239 msgid "I<info libc>"
5243 #: build/C/man3/matherr.3:25
5249 #: build/C/man3/matherr.3:25
5255 #: build/C/man3/matherr.3:28
5256 msgid "matherr - SVID math library exception handling"
5260 #: build/C/man3/matherr.3:32
5263 "B<#define _SVID_SOURCE> /* See feature_test_macros(7) */\n"
5264 "B<#include E<lt>math.hE<gt>>\n"
5268 #: build/C/man3/matherr.3:34
5270 msgid "B<int matherr(struct exception *>I<exc>B<);>\n"
5274 #: build/C/man3/matherr.3:36
5276 msgid "B<extern _LIB_VERSION_TYPE _LIB_VERSION;>\n"
5280 #: build/C/man3/matherr.3:49
5282 "The System V Interface Definition (SVID) specifies that various math "
5283 "functions should invoke a function called B<matherr>() if a math exception "
5284 "is detected. This function is called before the math function returns; "
5285 "after B<matherr>() returns, the system then returns to the math function, "
5286 "which in turn returns to the caller."
5290 #: build/C/man3/matherr.3:60
5292 "The B<matherr>() mechanism is supported by glibc, but is now obsolete: new "
5293 "applications should use the techniques described in B<math_error>(7) and "
5294 "B<fenv>(3). This page documents the glibc B<matherr>() mechanism as an aid "
5295 "for maintaining and porting older applications."
5299 #: build/C/man3/matherr.3:73
5301 "To employ B<matherr>(), the programmer must define the B<_SVID_SOURCE> "
5302 "feature test macro (before including I<any> header files), and assign the "
5303 "value B<_SVID_> to the external variable B<_LIB_VERSION>."
5307 #: build/C/man3/matherr.3:85
5309 "The system provides a default version of B<matherr>(). This version does "
5310 "nothing, and returns zero (see below for the significance of this). The "
5311 "default B<matherr>() can be overridden by a programmer-defined version, "
5312 "which will be invoked when an exception occurs. The function is invoked "
5313 "with one argument, a pointer to an I<exception> structure, defined as "
5318 #: build/C/man3/matherr.3:95
5321 "struct exception {\n"
5322 " int type; /* Exception type */\n"
5323 " char *name; /* Name of function causing exception */\n"
5324 " double arg1; /* 1st argument to function */\n"
5325 " double arg2; /* 2nd argument to function */\n"
5326 " double retval; /* Function return value */\n"
5331 #: build/C/man3/matherr.3:101
5332 msgid "The I<type> field has one of the following values:"
5336 #: build/C/man3/matherr.3:101
5342 #: build/C/man3/matherr.3:109
5344 "A domain error occurred (the function argument was outside the range for "
5345 "which the function is defined). The return value depends on the function; "
5346 "I<errno> is set to B<EDOM>."
5350 #: build/C/man3/matherr.3:109
5356 #: build/C/man3/matherr.3:120
5358 "A pole error occurred (the function result is an infinity). The return "
5359 "value in most cases is B<HUGE> (the largest single precision floating-point "
5360 "number), appropriately signed. In most cases, I<errno> is set to B<EDOM>."
5364 #: build/C/man3/matherr.3:120
5370 #: build/C/man3/matherr.3:129
5372 "An overflow occurred. In most cases, the value B<HUGE> is returned, and "
5373 "I<errno> is set to B<ERANGE>."
5377 #: build/C/man3/matherr.3:129
5379 msgid "B<UNDERFLOW>"
5383 #: build/C/man3/matherr.3:136
5384 msgid "An underflow occurred. 0.0 is returned, and I<errno> is set to B<ERANGE>."
5388 #: build/C/man3/matherr.3:136
5394 #: build/C/man3/matherr.3:143
5396 "Total loss of significance. 0.0 is returned, and I<errno> is set to "
5401 #: build/C/man3/matherr.3:143
5407 #: build/C/man3/matherr.3:148
5409 "Partial loss of significance. This value is unused on glibc (and many other "
5414 #: build/C/man3/matherr.3:156
5416 "The I<arg1> and I<arg2> fields are the arguments supplied to the function "
5417 "(I<arg2> is undefined for functions that take only one argument)."
5421 #: build/C/man3/matherr.3:164
5423 "The I<retval> field specifies the return value that the math function will "
5424 "return to its caller. The programmer-defined B<matherr>() can modify this "
5425 "field to change the return value of the math function."
5429 #: build/C/man3/matherr.3:171
5431 "If the B<matherr>() function returns zero, then the system sets I<errno> as "
5432 "described above, and may print an error message on standard error (see "
5437 #: build/C/man3/matherr.3:177
5439 "If the B<matherr>() function returns a nonzero value, then the system does "
5440 "not set I<errno>, and doesn't print an error message."
5444 #: build/C/man3/matherr.3:177
5446 msgid "Math functions that employ matherr()"
5450 #: build/C/man3/matherr.3:187
5452 "The table below lists the functions and circumstances in which B<matherr>() "
5453 "is called. The \"Type\" column indicates the value assigned to "
5454 "I<exc-E<gt>type> when calling B<matherr>(). The \"Result\" column is the "
5455 "default return value assigned to I<exc-E<gt>retval>."
5459 #: build/C/man3/matherr.3:193
5461 "The \"Msg?\" and \"errno\" columns describe the default behavior if "
5462 "B<matherr>() returns zero. If the \"Msg?\" columns contains \"y\", then "
5463 "the system prints an error message on standard error."
5467 #: build/C/man3/matherr.3:195
5468 msgid "The table uses the following notations and abbreviations:"
5472 #: build/C/man3/matherr.3:207
5475 "x first argument to function\n"
5476 "y second argument to function\n"
5477 "fin finite value for argument\n"
5478 "neg negative value for argument\n"
5479 "int integral value for argument\n"
5480 "o/f result overflowed\n"
5481 "u/f result underflowed\n"
5482 "|x| absolute value of x\n"
5483 "X_TLOSS is a constant defined in I<E<lt>math.hE<gt>>\n"
5487 #: build/C/man3/matherr.3:214
5489 msgid "Function\tType\tResult\tMsg?\terrno\n"
5493 #: build/C/man3/matherr.3:215
5495 msgid "acos(|x|E<gt>1)\tDOMAIN\tHUGE\ty\tEDOM\n"
5499 #: build/C/man3/matherr.3:216
5501 msgid "asin(|x|E<gt>1)\tDOMAIN\tHUGE\ty\tEDOM\n"
5505 #: build/C/man3/matherr.3:217
5507 msgid "atan2(0,0)\tDOMAIN\tHUGE\ty\tEDOM\n"
5510 #. retval is 0.0/0.0
5512 #: build/C/man3/matherr.3:218
5514 msgid "acosh(xE<lt>1)\tDOMAIN\tNAN\ty\tEDOM \n"
5517 #. retval is 0.0/0.0
5519 #: build/C/man3/matherr.3:219
5521 msgid "atanh(|x|E<gt>1)\tDOMAIN\tNAN\ty\tEDOM \n"
5526 #: build/C/man3/matherr.3:220
5528 msgid "atanh(|x|==1)\tSING\t(xE<gt>0.0)?\ty\tEDOM \n"
5532 #: build/C/man3/matherr.3:221 build/C/man3/matherr.3:266
5534 msgid "\\ \t\\ \tHUGE_VAL :\n"
5538 #: build/C/man3/matherr.3:222 build/C/man3/matherr.3:267
5540 msgid "\\ \t\\ \t-HUGE_VAL\n"
5544 #: build/C/man3/matherr.3:223
5546 msgid "cosh(fin) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5550 #: build/C/man3/matherr.3:224
5552 msgid "sinh(fin) o/f\tOVERFLOW\t(xE<gt>0.0) ?\tn\tERANGE\n"
5556 #: build/C/man3/matherr.3:225
5558 msgid "\\ \t\\ \tHUGE : -HUGE\n"
5562 #: build/C/man3/matherr.3:226
5564 msgid "sqrt(xE<lt>0)\tDOMAIN\t0.0\ty\tEDOM\n"
5568 #: build/C/man3/matherr.3:227
5570 msgid "hypot(fin,fin) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5574 #: build/C/man3/matherr.3:228
5576 msgid "exp(fin) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5580 #: build/C/man3/matherr.3:229
5582 msgid "exp(fin) u/f\tUNDERFLOW\t0.0\tn\tERANGE\n"
5586 #: build/C/man3/matherr.3:230
5588 msgid "exp2(fin) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5592 #: build/C/man3/matherr.3:231
5594 msgid "exp2(fin) u/f\tUNDERFLOW\t0.0\tn\tERANGE\n"
5598 #: build/C/man3/matherr.3:232
5600 msgid "exp10(fin) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5604 #: build/C/man3/matherr.3:233
5606 msgid "exp10(fin) u/f\tUNDERFLOW\t0.0\tn\tERANGE\n"
5610 #: build/C/man3/matherr.3:234
5612 msgid "j0(|x|E<gt>X_TLOSS)\tTLOSS\t0.0\ty\tERANGE\n"
5616 #: build/C/man3/matherr.3:235
5618 msgid "j1(|x|E<gt>X_TLOSS)\tTLOSS\t0.0\ty\tERANGE\n"
5622 #: build/C/man3/matherr.3:236
5624 msgid "jn(|x|E<gt>X_TLOSS)\tTLOSS\t0.0\ty\tERANGE\n"
5628 #: build/C/man3/matherr.3:237
5630 msgid "y0(xE<gt>X_TLOSS)\tTLOSS\t0.0\ty\tERANGE\n"
5634 #: build/C/man3/matherr.3:238
5636 msgid "y1(xE<gt>X_TLOSS)\tTLOSS\t0.0\ty\tERANGE\n"
5640 #: build/C/man3/matherr.3:239
5642 msgid "yn(xE<gt>X_TLOSS)\tTLOSS\t0.0\ty\tERANGE\n"
5646 #: build/C/man3/matherr.3:240
5648 msgid "y0(0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5652 #: build/C/man3/matherr.3:241
5654 msgid "y0(xE<lt>0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5658 #: build/C/man3/matherr.3:242
5660 msgid "y1(0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5664 #: build/C/man3/matherr.3:243
5666 msgid "y1(xE<lt>0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5670 #: build/C/man3/matherr.3:244
5672 msgid "yn(n,0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5676 #: build/C/man3/matherr.3:245
5678 msgid "yn(xE<lt>0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5682 #: build/C/man3/matherr.3:246
5684 msgid "lgamma(fin) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5688 #: build/C/man3/matherr.3:247
5690 msgid "lgamma(-int) or\tSING\tHUGE\ty\tEDOM\n"
5694 #: build/C/man3/matherr.3:248
5696 msgid "\\ \\ lgamma(0)\n"
5700 #: build/C/man3/matherr.3:249
5702 msgid "tgamma(fin) o/f\tOVERFLOW\tHUGE_VAL\tn\tERANGE\n"
5706 #: build/C/man3/matherr.3:250
5708 msgid "tgamma(-int)\tSING\tNAN\ty\tEDOM\n"
5712 #: build/C/man3/matherr.3:251
5714 msgid "tgamma(0)\tSING\tcopysign(\ty\tERANGE\n"
5718 #: build/C/man3/matherr.3:252
5720 msgid "\\ \t\\ \tHUGE_VAL,x)\n"
5724 #: build/C/man3/matherr.3:253
5726 msgid "log(0)\tSING\t-HUGE\ty\tEDOM\n"
5730 #: build/C/man3/matherr.3:254
5732 msgid "log(xE<lt>0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5735 #. different from log()
5737 #: build/C/man3/matherr.3:255
5739 msgid "log2(0)\tSING\t-HUGE\tn\tEDOM\t\n"
5742 #. different from log()
5744 #: build/C/man3/matherr.3:256
5746 msgid "log2(xE<lt>0)\tDOMAIN\t-HUGE\tn\tEDOM\t\n"
5750 #: build/C/man3/matherr.3:257
5752 msgid "log10(0)\tSING\t-HUGE\ty\tEDOM\n"
5756 #: build/C/man3/matherr.3:258
5758 msgid "log10(xE<lt>0)\tDOMAIN\t-HUGE\ty\tEDOM\n"
5762 #: build/C/man3/matherr.3:259
5764 msgid "pow(0.0,0.0)\tDOMAIN\t0.0\ty\tEDOM\n"
5768 #: build/C/man3/matherr.3:260
5770 msgid "pow(x,y) o/f\tOVERFLOW\tHUGE\tn\tERANGE\n"
5774 #: build/C/man3/matherr.3:261
5776 msgid "pow(x,y) u/f\tUNDERFLOW\t0.0\tn\tERANGE\n"
5780 #: build/C/man3/matherr.3:262
5782 msgid "pow(NaN,0.0)\tDOMAIN\tx\tn\tEDOM\n"
5787 #: build/C/man3/matherr.3:263
5789 msgid "0**neg\tDOMAIN\t0.0\ty\tEDOM\t\n"
5793 #: build/C/man3/matherr.3:264
5795 msgid "neg**non-int\tDOMAIN\t0.0\ty\tEDOM\n"
5799 #: build/C/man3/matherr.3:265
5801 msgid "scalb() o/f\tOVERFLOW\t(xE<gt>0.0) ?\tn\tERANGE\n"
5805 #: build/C/man3/matherr.3:268
5807 msgid "scalb() u/f\tUNDERFLOW\tcopysign(\tn\tERANGE\n"
5811 #: build/C/man3/matherr.3:269
5813 msgid "\\ \t\\ \t\\ \\ 0.0,x)\n"
5817 #: build/C/man3/matherr.3:270
5819 msgid "fmod(x,0)\tDOMAIN\tx\ty\tEDOM\n"
5822 #. retval is 0.0/0.0
5824 #: build/C/man3/matherr.3:271
5826 msgid "remainder(x,0)\tDOMAIN\tNAN\ty\tEDOM \n"
5830 #: build/C/man3/matherr.3:294
5832 "The example program demonstrates the use of B<matherr>() when calling "
5833 "B<log>(3). The program takes up to three command-line arguments. The first "
5834 "argument is the floating-point number to be given to B<log>(3). If the "
5835 "optional second argument is provided, then B<_LIB_VERSION> is set to "
5836 "B<_SVID_> so that B<matherr>() is called, and the integer supplied in the "
5837 "command-line argument is used as the return value from B<matherr>(). If the "
5838 "optional third command-line argument is supplied, then it specifies an "
5839 "alternative return value that B<matherr>() should assign as the return "
5840 "value of the math function."
5844 #: build/C/man3/matherr.3:299
5846 "The following example run, where B<log>(3) is given an argument of 0.0, "
5847 "does not use B<matherr>():"
5851 #: build/C/man3/matherr.3:305
5854 "$B< ./a.out 0.0>\n"
5855 "errno: Numerical result out of range\n"
5860 #: build/C/man3/matherr.3:311
5861 msgid "In the following run, B<matherr>() is called, and returns 0:"
5865 #: build/C/man3/matherr.3:321
5868 "$B< ./a.out 0.0 0>\n"
5869 "matherr SING exception in log() function\n"
5870 " args: 0.000000, 0.000000\n"
5871 " retval: -340282346638528859811704183484516925440.000000\n"
5873 "errno: Numerical argument out of domain\n"
5874 "x=-340282346638528859811704183484516925440.000000\n"
5878 #: build/C/man3/matherr.3:325
5879 msgid "The message \"log: SING error\" was printed by the C library."
5883 #: build/C/man3/matherr.3:329
5884 msgid "In the following run, B<matherr>() is called, and returns a nonzero value:"
5888 #: build/C/man3/matherr.3:337
5891 "$B< ./a.out 0.0 1>\n"
5892 "matherr SING exception in log() function\n"
5893 " args: 0.000000, 0.000000\n"
5894 " retval: -340282346638528859811704183484516925440.000000\n"
5895 "x=-340282346638528859811704183484516925440.000000\n"
5899 #: build/C/man3/matherr.3:343
5901 "In this case, the C library did not print a message, and I<errno> was not "
5906 #: build/C/man3/matherr.3:348
5908 "In the following run, B<matherr>() is called, changes the return value of "
5909 "the math function, and returns a nonzero value:"
5913 #: build/C/man3/matherr.3:356
5916 "$B< ./a.out 0.0 1 12345.0>\n"
5917 "matherr SING exception in log() function\n"
5918 " args: 0.000000, 0.000000\n"
5919 " retval: -340282346638528859811704183484516925440.000000\n"
5924 #: build/C/man3/matherr.3:366
5927 "#define _SVID_SOURCE\n"
5928 "#include E<lt>errno.hE<gt>\n"
5929 "#include E<lt>math.hE<gt>\n"
5930 "#include E<lt>stdio.hE<gt>\n"
5931 "#include E<lt>stdlib.hE<gt>\n"
5935 #: build/C/man3/matherr.3:372
5938 "static int matherr_ret = 0; /* Value that matherr()\n"
5939 " should return */\n"
5940 "static int change_retval = 0; /* Should matherr() change\n"
5941 " function\\(aqs return value? */\n"
5942 "static double new_retval; /* New function return value */\n"
5946 #: build/C/man3/matherr.3:387
5950 "matherr(struct exception *exc)\n"
5952 " fprintf(stderr, \"matherr %s exception in %s() function\\en\",\n"
5953 " (exc-E<gt>type == DOMAIN) ? \"DOMAIN\" :\n"
5954 " (exc-E<gt>type == OVERFLOW) ? \"OVERFLOW\" :\n"
5955 " (exc-E<gt>type == UNDERFLOW) ? \"UNDERFLOW\" :\n"
5956 " (exc-E<gt>type == SING) ? \"SING\" :\n"
5957 " (exc-E<gt>type == TLOSS) ? \"TLOSS\" :\n"
5958 " (exc-E<gt>type == PLOSS) ? \"PLOSS\" : \"???\",\n"
5959 " exc-E<gt>name);\n"
5960 " fprintf(stderr, \" args: %f, %f\\en\",\n"
5961 " exc-E<gt>arg1, exc-E<gt>arg2);\n"
5962 " fprintf(stderr, \" retval: %f\\en\", exc-E<gt>retval);\n"
5966 #: build/C/man3/matherr.3:390
5969 " if (change_retval)\n"
5970 " exc-E<gt>retval = new_retval;\n"
5974 #: build/C/man3/matherr.3:393
5977 " return matherr_ret;\n"
5982 #: build/C/man3/matherr.3:398
5986 "main(int argc, char *argv[])\n"
5992 #: build/C/man3/matherr.3:404
5995 " if (argc E<lt> 2) {\n"
5996 " fprintf(stderr, \"Usage: %s E<lt>argvalE<gt>\"\n"
5997 " \" [E<lt>matherr-retE<gt> "
5998 "[E<lt>new-func-retvalE<gt>]]\\en\", argv[0]);\n"
5999 " exit(EXIT_FAILURE);\n"
6004 #: build/C/man3/matherr.3:409
6007 " if (argc E<gt> 2) {\n"
6008 " _LIB_VERSION = _SVID_;\n"
6009 " matherr_ret = atoi(argv[2]);\n"
6014 #: build/C/man3/matherr.3:414
6017 " if (argc E<gt> 3) {\n"
6018 " change_retval = 1;\n"
6019 " new_retval = atof(argv[3]);\n"
6024 #: build/C/man3/matherr.3:418
6027 " x = log(atof(argv[1]));\n"
6028 " if (errno != 0)\n"
6029 " perror(\"errno\");\n"
6033 #: build/C/man3/matherr.3:422
6036 " printf(\"x=%f\\en\", x);\n"
6037 " exit(EXIT_SUCCESS);\n"
6042 #: build/C/man3/matherr.3:426
6043 msgid "B<fenv>(3), B<math_error>(7), B<standards>(7)"
6047 #: build/C/man3/modf.3:31
6053 #: build/C/man3/modf.3:35
6055 "modf, modff, modfl - extract signed integral and fractional values from "
6056 "floating-point number"
6060 #: build/C/man3/modf.3:40
6062 msgid "B<double modf(double >I<x>B<, double *>I<iptr>B<);>\n"
6066 #: build/C/man3/modf.3:42
6068 msgid "B<float modff(float >I<x>B<, float *>I<iptr>B<);>\n"
6072 #: build/C/man3/modf.3:44
6074 msgid "B<long double modfl(long double >I<x>B<, long double *>I<iptr>B<);>\n"
6078 #: build/C/man3/modf.3:56
6079 msgid "B<modf>(), B<modfl>():"
6083 #: build/C/man3/modf.3:70
6085 "The B<modf>() function breaks the argument I<x> into an integral part and a "
6086 "fractional part, each of which has the same sign as I<x>. The integral part "
6087 "is stored in the location pointed to by I<iptr>."
6091 #: build/C/man3/modf.3:74
6092 msgid "The B<modf>() function returns the fractional part of I<x>."
6096 #: build/C/man3/modf.3:80
6097 msgid "If I<x> is a NaN, a NaN is returned, and I<*iptr> is set to a NaN."
6101 #: build/C/man3/modf.3:86
6103 "If I<x> is positive infinity (negative infinity), +0 (-0) is returned, and "
6104 "I<*iptr> is set to positive infinity (negative infinity)."
6108 #: build/C/man3/modf.3:96
6109 msgid "B<frexp>(3), B<ldexp>(3)"
6113 #: build/C/man3/nan.3:6
6119 #: build/C/man3/nan.3:9
6120 msgid "nan, nanf, nanl - return 'Not a Number'"
6124 #: build/C/man3/nan.3:13
6125 msgid "B<double nan(const char *>I<tagp>B<);>"
6129 #: build/C/man3/nan.3:15
6130 msgid "B<float nanf(const char *>I<tagp>B<);>"
6134 #: build/C/man3/nan.3:17
6135 msgid "B<long double nanl(const char *>I<tagp>B<);>"
6139 #: build/C/man3/nan.3:29
6140 msgid "B<nan>(), B<nanf>(), B<nanl>():"
6144 #: build/C/man3/nan.3:43
6146 "These functions return a representation (determined by I<tagp>) of a quiet "
6147 "NaN. If the implementation does not support quiet NaNs, these functions "
6152 #: build/C/man3/nan.3:47
6153 msgid "The call I<nan(\"char-sequence\")> is equivalent to:"
6157 #: build/C/man3/nan.3:50
6159 msgid " strtod(\"NAN(char-sequence)\", NULL);\n"
6163 #: build/C/man3/nan.3:60
6165 "Similarly, calls to B<nanf>() and B<nanl>() are equivalent to analogous "
6166 "calls to B<strtof>(3) and B<strtold>(3)."
6170 #: build/C/man3/nan.3:68
6172 "The argument I<tagp> is used in an unspecified manner. On IEEE 754 systems, "
6173 "there are many representations of NaN, and I<tagp> selects one. On other "
6174 "systems it may do nothing."
6178 #: build/C/man3/nan.3:74
6180 "C99, POSIX.1-2001. See also IEC 559 and the appendix with recommended "
6181 "functions in IEEE 754/IEEE 854."
6185 #: build/C/man3/nan.3:77
6186 msgid "B<isnan>(3), B<strtod>(3), B<math_error>(7)"
6190 #: build/C/man3/nextafter.3:6
6196 #: build/C/man3/nextafter.3:10
6198 "nextafter, nextafterf, nextafterl, nexttoward, nexttowardf, nexttowardl - "
6199 "floating-point number manipulation"
6203 #: build/C/man3/nextafter.3:14
6204 msgid "B<double nextafter(double >I<x>B<, double >I<y>B<);>"
6208 #: build/C/man3/nextafter.3:16
6209 msgid "B<float nextafterf(float >I<x>B<, float >I<y>B<);>"
6213 #: build/C/man3/nextafter.3:18
6214 msgid "B<long double nextafterl(long double >I<x>B<, long double >I<y>B<);>"
6218 #: build/C/man3/nextafter.3:20
6219 msgid "B<double nexttoward(double >I<x>B<, long double >I<y>B<);>"
6223 #: build/C/man3/nextafter.3:22
6224 msgid "B<float nexttowardf(float >I<x>B<, long double >I<y>B<);>"
6228 #: build/C/man3/nextafter.3:24
6229 msgid "B<long double nexttowardl(long double >I<x>B<, long double >I<y>B<);>"
6233 #: build/C/man3/nextafter.3:34
6234 msgid "B<nextafter>():"
6238 #: build/C/man3/nextafter.3:46
6239 msgid "B<nextafterf>(), B<nextafterl>():"
6243 #: build/C/man3/nextafter.3:57
6244 msgid "B<nexttoward>(), B<nexttowardf>(), B<nexttowardl>():"
6248 #: build/C/man3/nextafter.3:78
6250 "The B<nextafter>() functions return the next representable floating-point "
6251 "value following I<x> in the direction of I<y>. If I<y> is less than I<x>, "
6252 "these functions will return the largest representable number less than I<x>."
6256 #: build/C/man3/nextafter.3:80
6257 msgid "If I<x> equals I<y>, the functions return I<y>."
6261 #: build/C/man3/nextafter.3:88
6263 "The B<nexttoward>() functions do the same as the B<nextafter>() functions, "
6264 "except that they have a I<long double> second argument."
6268 #: build/C/man3/nextafter.3:94
6270 "On success, these functions return the next representable floating-point "
6271 "value after I<x> in the direction of I<y>."
6275 #: build/C/man3/nextafter.3:104
6276 msgid "If I<x> equals I<y>, then I<y> (cast to the same type as I<x>) is returned."
6281 #: build/C/man3/nextafter.3:124
6283 "If I<x> is finite, and the result would overflow, a range error occurs, and "
6284 "the functions return B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, "
6285 "respectively, with the correct mathematical sign."
6289 #: build/C/man3/nextafter.3:133
6291 "If I<x> is not equal to I<y>, and the correct function result would be "
6292 "subnormal, zero, or underflow, a range error occurs, and either the correct "
6293 "value (if it can be represented), or 0.0, is returned."
6297 #: build/C/man3/nextafter.3:149
6299 msgid "Range error: result is subnormal or underflows"
6303 #: build/C/man3/nextafter.3:172
6305 "In glibc version 2.5 and earlier, these functions do not raise an underflow "
6306 "floating-point (B<FE_UNDERFLOW>) exception when an underflow occurs."
6310 #: build/C/man3/nextafter.3:173
6311 msgid "B<nearbyint>(3)"
6315 #: build/C/man3/pow.3:33
6321 #: build/C/man3/pow.3:36
6322 msgid "pow, powf, powl - power functions"
6326 #: build/C/man3/pow.3:41
6328 msgid "B<double pow(double >I<x>B<, double >I<y>B<);>\n"
6332 #: build/C/man3/pow.3:43
6334 msgid "B<float powf(float >I<x>B<, float >I<y>B<);>\n"
6338 #: build/C/man3/pow.3:45
6340 msgid "B<long double powl(long double >I<x>B<, long double >I<y>B<);>\n"
6344 #: build/C/man3/pow.3:57
6345 msgid "B<powf>(), B<powl>():"
6349 #: build/C/man3/pow.3:70
6351 "The B<pow>() function returns the value of I<x> raised to the power of "
6356 #: build/C/man3/pow.3:75
6357 msgid "On success, these functions return the value of I<x> to the power of I<y>."
6360 #. The domain error is generated at least as far back as glibc 2.4
6362 #: build/C/man3/pow.3:83
6364 "If I<x> is a finite value less than 0, and I<y> is a finite noninteger, a "
6365 "domain error occurs, and a NaN is returned."
6368 #. The range error is generated at least as far back as glibc 2.4
6370 #: build/C/man3/pow.3:93
6372 "If the result overflows, a range error occurs, and the functions return "
6373 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, with the "
6374 "mathematically correct sign."
6377 #. POSIX.1 does not specify the sign of the zero,
6378 #. but http://sources.redhat.com/bugzilla/show_bug.cgi?id=2678
6379 #. points out that the zero has the wrong sign in some cases.
6381 #: build/C/man3/pow.3:100
6383 "If result underflows, and is not representable, a range error occurs, and "
6388 #: build/C/man3/pow.3:106
6389 msgid "Except as specified below, if I<x> or I<y> is a NaN, the result is a NaN."
6393 #: build/C/man3/pow.3:112
6394 msgid "If I<x> is +1, the result is 1.0 (even if I<y> is a NaN)."
6398 #: build/C/man3/pow.3:118
6399 msgid "If I<y> is 0, the result is 1.0 (even if I<x> is a NaN)."
6403 #: build/C/man3/pow.3:126
6405 "If I<x> is +0 (-0), and I<y> is an odd integer greater than 0, the result is "
6410 #: build/C/man3/pow.3:134
6412 "If I<x> is 0, and I<y> greater than 0 and not an odd integer, the result is "
6417 #: build/C/man3/pow.3:142
6419 "If I<x> is -1, and I<y> is positive infinity or negative infinity, the "
6424 #: build/C/man3/pow.3:150
6426 "If the absolute value of I<x> is less than 1, and I<y> is negative infinity, "
6427 "the result is positive infinity."
6431 #: build/C/man3/pow.3:158
6433 "If the absolute value of I<x> is greater than 1, and I<y> is negative "
6434 "infinity, the result is +0."
6438 #: build/C/man3/pow.3:166
6440 "If the absolute value of I<x> is less than 1, and I<y> is positive infinity, "
6445 #: build/C/man3/pow.3:174
6447 "If the absolute value of I<x> is greater than 1, and I<y> is positive "
6448 "infinity, the result is positive infinity."
6452 #: build/C/man3/pow.3:182
6454 "If I<x> is negative infinity, and I<y> is an odd integer less than 0, the "
6459 #: build/C/man3/pow.3:190
6461 "If I<x> is negative infinity, and I<y> less than 0 and not an odd integer, "
6466 #: build/C/man3/pow.3:198
6468 "If I<x> is negative infinity, and I<y> is an odd integer greater than 0, the "
6469 "result is negative infinity."
6473 #: build/C/man3/pow.3:206
6475 "If I<x> is negative infinity, and I<y> greater than 0 and not an odd "
6476 "integer, the result is positive infinity."
6480 #: build/C/man3/pow.3:214
6481 msgid "If I<x> is positive infinity, and I<y> less than 0, the result is +0."
6485 #: build/C/man3/pow.3:222
6487 "If I<x> is positive infinity, and I<y> greater than 0, the result is "
6488 "positive infinity."
6492 #: build/C/man3/pow.3:237
6494 "If I<x> is +0 or -0, and I<y> is an odd integer less than 0, a pole error "
6495 "occurs and B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, is returned, with the "
6496 "same sign as I<x>."
6499 #. The pole error is generated at least as far back as glibc 2.4
6501 #: build/C/man3/pow.3:251
6503 "If I<x> is +0 or -0, and I<y> is less than 0 and not an odd integer, a pole "
6504 "error occurs and +B<HUGE_VAL>, +B<HUGE_VALF>, or +B<HUGE_VALL>, is returned."
6508 #: build/C/man3/pow.3:278
6510 msgid "Domain error: I<x> is negative, and I<y> is a finite noninteger"
6514 #: build/C/man3/pow.3:286
6516 msgid "Pole error: I<x> is zero, and I<y> is negative"
6520 #: build/C/man3/pow.3:295
6522 msgid "Range error: the result overflows"
6526 #: build/C/man3/pow.3:303
6528 msgid "Range error: the result underflows"
6532 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=6776
6533 #. or possibly 2.9, I haven't found the source code change
6534 #. and I don't have a 2.9 system to test
6536 #: build/C/man3/pow.3:331
6538 "In glibc 2.9 and earlier, when a pole error occurs, I<errno> is set to "
6539 "B<EDOM> instead of the POSIX-mandated B<ERANGE>. Since version 2.10, glibc "
6540 "does the right thing."
6543 #. see bug http://sources.redhat.com/bugzilla/show_bug.cgi?id=3866
6544 #. and http://sources.redhat.com/bugzilla/show_bug.cgi?id=369
6546 #: build/C/man3/pow.3:351
6548 "If I<x> is negative, then large negative or positive I<y> values yield a NaN "
6549 "as the function result, with I<errno> set to B<EDOM>, and an invalid "
6550 "(B<FE_INVALID>) floating-point exception. For example, with B<pow>(), one "
6551 "sees this behavior when the absolute value of I<y> is greater than about "
6555 #. FIXME . Actually, 2.3.2 is the earliest test result I have; so yet
6556 #. to confirm if this error occurs only in 2.3.2.
6558 #: build/C/man3/pow.3:360
6560 "In version 2.3.2 and earlier, when an overflow or underflow error occurs, "
6561 "glibc's B<pow>() generates a bogus invalid floating-point exception "
6562 "(B<FE_INVALID>) in addition to the overflow or underflow exception."
6566 #: build/C/man3/pow.3:363
6567 msgid "B<cbrt>(3), B<cpow>(3), B<sqrt>(3)"
6571 #: build/C/man3/pow10.3:23
6577 #: build/C/man3/pow10.3:26
6578 msgid "pow10, pow10f, pow10l - base-10 power functions"
6582 #: build/C/man3/pow10.3:30 build/C/man3/sincos.3:13
6585 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
6586 "B<#include E<lt>math.hE<gt>>\n"
6590 #: build/C/man3/pow10.3:32
6592 msgid "B<double pow10(double >I<x>B<);>\n"
6596 #: build/C/man3/pow10.3:34
6598 msgid "B<float pow10f(float >I<x>B<);>\n"
6602 #: build/C/man3/pow10.3:36
6604 msgid "B<long double pow10l(long double >I<x>B<);>\n"
6608 #: build/C/man3/pow10.3:44
6609 msgid "The B<pow10>() function returns the value of 10 raised to the power I<x>."
6613 #: build/C/man3/pow10.3:48
6614 msgid "This is a GNU extension."
6618 #: build/C/man3/pow10.3:51
6619 msgid "These functions are identical to the functions listed under B<exp10>(3)."
6623 #: build/C/man3/pow10.3:53
6624 msgid "B<exp10>(3), B<pow>(3)"
6628 #: build/C/man3/remainder.3:35
6634 #: build/C/man3/remainder.3:39
6636 "drem, dremf, dreml, remainder, remainderf, remainderl - floating-point "
6637 "remainder function"
6641 #: build/C/man3/remainder.3:47
6644 "/* The C99 versions */\n"
6645 "B<double remainder(double >I<x>B<, double >I<y>B<);>\n"
6646 "B<float remainderf(float >I<x>B<, float >I<y>B<);>\n"
6647 "B<long double remainderl(long double >I<x>B<, long double >I<y>B<);>\n"
6651 #: build/C/man3/remainder.3:52
6654 "/* Obsolete synonyms */\n"
6655 "B<double drem(double >I<x>B<, double >I<y>B<);>\n"
6656 "B<float dremf(float >I<x>B<, float >I<y>B<);>\n"
6657 "B<long double dreml(long double >I<x>B<, long double >I<y>B<);>\n"
6661 #: build/C/man3/remainder.3:63
6662 msgid "B<remainder>():"
6666 #: build/C/man3/remainder.3:68
6668 "_SVID_SOURCE || _BSD_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 500 || "
6669 "_XOPEN_SOURCE\\ &&\\ _XOPEN_SOURCE_EXTENDED || _ISOC99_SOURCE || "
6670 "_POSIX_C_SOURCE\\ E<gt>=\\ 200112L;"
6674 #: build/C/man3/remainder.3:75
6675 msgid "B<remainderf>(), B<remainderl>():"
6679 #: build/C/man3/remainder.3:86
6680 msgid "B<drem>(), B<dremf>(), B<dreml>():"
6684 #: build/C/man3/remainder.3:88 build/C/man3/significand.3:30
6685 msgid "_SVID_SOURCE || _BSD_SOURCE"
6689 #: build/C/man3/remainder.3:109
6691 "The B<remainder>() function computes the remainder of dividing I<x> by "
6692 "I<y>. The return value is I<x>-I<n>*I<y>, where I<n> is the value I<x\\ /\\ "
6693 "y>, rounded to the nearest integer. If the absolute value of I<x>-I<n>*I<y> "
6694 "is 0.5, I<n> is chosen to be even."
6698 #: build/C/man3/remainder.3:112
6700 "These functions are unaffected by the current rounding mode (see "
6705 #: build/C/man3/remainder.3:116
6706 msgid "The B<drem>() function does precisely the same thing."
6710 #: build/C/man3/remainder.3:122
6712 "On success, these functions return the floating-point remainder, "
6713 "I<x>-I<n>*I<y>. If the return value is 0, it has the sign of I<x>."
6717 #: build/C/man3/remainder.3:137 build/C/man3/remquo.3:87
6719 "If I<x> is an infinity, and I<y> is not a NaN, a domain error occurs, and a "
6724 #: build/C/man3/remainder.3:148 build/C/man3/remquo.3:96
6726 "If I<y> is zero, and I<x> is not a NaN, a domain error occurs, and a NaN is "
6731 #: build/C/man3/remainder.3:155
6733 msgid "Domain error: I<x> is an infinity and I<y> is not a NaN"
6738 #: build/C/man3/remainder.3:186
6740 "The functions B<remainder>(), B<remainderf>(), and B<remainderl>() are "
6741 "specified in C99 and POSIX.1-2001."
6745 #: build/C/man3/remainder.3:202
6747 "The function B<drem>() is from 4.3BSD. The I<float> and I<long double> "
6748 "variants B<dremf>() and B<dreml>() exist on some systems, such as Tru64 "
6749 "and glibc2. Avoid the use of these functions in favor of B<remainder>() "
6754 #: build/C/man3/remainder.3:204
6759 #: build/C/man3/remainder.3:206
6761 msgid " remainder(nan(\"\"), 0);\n"
6764 #. FIXME . this bug occurs as at glibc 2.8.
6765 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6779
6767 #: build/C/man3/remainder.3:211
6769 "returns a NaN, as expected, but wrongly causes a domain error; it should "
6770 "yield a silent NaN."
6774 #: build/C/man3/remainder.3:213
6775 msgid "The call \"remainder(29.0, 3.0)\" returns -1."
6779 #: build/C/man3/remainder.3:216
6780 msgid "B<div>(3), B<fmod>(3), B<remquo>(3)"
6784 #: build/C/man3/remquo.3:7
6790 #: build/C/man3/remquo.3:10
6791 msgid "remquo, remquof, remquol - remainder and part of quotient"
6795 #: build/C/man3/remquo.3:15
6797 msgid "B<double remquo(double >I<x>B<, double >I<y>B<, int *>I<quo>B<);>\n"
6801 #: build/C/man3/remquo.3:17
6803 msgid "B<float remquof(float >I<x>B<, float >I<y>B<, int *>I<quo>B<);>\n"
6807 #: build/C/man3/remquo.3:19
6810 "B<long double remquol(long double >I<x>B<, long double >I<y>B<, int "
6815 #: build/C/man3/remquo.3:32
6816 msgid "B<remquo>(), B<remquof>(), B<remquol>():"
6820 #: build/C/man3/remquo.3:50
6822 "These functions compute the remainder and part of the quotient upon division "
6823 "of I<x> by I<y>. A few bits of the quotient are stored via the I<quo> "
6824 "pointer. The remainder is returned as the function result."
6828 #: build/C/man3/remquo.3:54
6830 "The value of the remainder is the same as that computed by the "
6831 "B<remainder>(3) function."
6835 #: build/C/man3/remquo.3:60
6837 "The value stored via the I<quo> pointer has the sign of I<x\\ /\\ y> and "
6838 "agrees with the quotient in at least the low order 3 bits."
6841 #. A possible application of this function might be the computation
6842 #. of sin(x). Compute remquo(x, pi/2, &quo) or so.
6844 #. glibc, UnixWare: return 3 bits
6845 #. MacOS 10: return 7 bits
6847 #: build/C/man3/remquo.3:68
6849 "For example, I<remquo(29.0,\\ 3.0)> returns -1.0 and might store 2. Note "
6850 "that the actual quotient might not fit in an integer."
6854 #: build/C/man3/remquo.3:72
6856 "On success, these functions return the same value as the analogous functions "
6857 "described in B<remainder>(3)."
6861 #: build/C/man3/remquo.3:103
6864 "Domain error: I<x> is an infinity or I<y> is 0, and the other argument is "
6869 #: build/C/man3/remquo.3:124
6870 msgid "B<fmod>(3), B<logb>(3), B<remainder>(3)"
6874 #: build/C/man3/rint.3:25
6880 #: build/C/man3/rint.3:29
6882 "nearbyint, nearbyintf, nearbyintl, rint, rintf, rintl - round to nearest "
6887 #: build/C/man3/rint.3:34
6889 msgid "B<double nearbyint(double >I<x>B<);>\n"
6893 #: build/C/man3/rint.3:36
6895 msgid "B<float nearbyintf(float >I<x>B<);>\n"
6899 #: build/C/man3/rint.3:38
6901 msgid "B<long double nearbyintl(long double >I<x>B<);>\n"
6905 #: build/C/man3/rint.3:40
6907 msgid "B<double rint(double >I<x>B<);>\n"
6911 #: build/C/man3/rint.3:42
6913 msgid "B<float rintf(float >I<x>B<);>\n"
6917 #: build/C/man3/rint.3:44
6919 msgid "B<long double rintl(long double >I<x>B<);>\n"
6923 #: build/C/man3/rint.3:57
6924 msgid "B<nearbyint>(), B<nearbyintf>(), B<nearbyintl>():"
6928 #: build/C/man3/rint.3:59
6930 "_XOPEN_SOURCE\\ E<gt>=\\ 600 || _POSIX_C_SOURCE\\ E<gt>=\\ 200112L || "
6935 #: build/C/man3/rint.3:65
6940 #: build/C/man3/rint.3:76
6941 msgid "B<rintf>(), B<rintl>():"
6945 #: build/C/man3/rint.3:93
6947 "The B<nearbyint>() functions round their argument to an integer value in "
6948 "floating-point format, using the current rounding direction (see "
6949 "B<fesetround>(3)) and without raising the I<inexact> exception."
6953 #: build/C/man3/rint.3:103
6955 "The B<rint>() functions do the same, but will raise the I<inexact> "
6956 "exception (B<FE_INEXACT>, checkable via B<fetestexcept>(3)) when the result "
6957 "differs in value from the argument."
6961 #: build/C/man3/rint.3:133
6963 "If you want to store the rounded value in an integer type, you probably want "
6964 "to use one of the functions described in B<lrint>(3) instead."
6968 #: build/C/man3/rint.3:138
6969 msgid "B<ceil>(3), B<floor>(3), B<lrint>(3), B<round>(3), B<trunc>(3)"
6973 #: build/C/man3/round.3:25
6979 #: build/C/man3/round.3:28
6980 msgid "round, roundf, roundl - round to nearest integer, away from zero"
6984 #: build/C/man3/round.3:33
6986 msgid "B<double round(double >I<x>B<);>\n"
6990 #: build/C/man3/round.3:35
6992 msgid "B<float roundf(float >I<x>B<);>\n"
6996 #: build/C/man3/round.3:37
6998 msgid "B<long double roundl(long double >I<x>B<);>\n"
7002 #: build/C/man3/round.3:50
7003 msgid "B<round>(), B<roundf>(), B<roundl>():"
7007 #: build/C/man3/round.3:65
7009 "These functions round I<x> to the nearest integer, but round halfway cases "
7010 "away from zero (regardless of the current rounding direction, see "
7011 "B<fenv>(3)), instead of to the nearest even integer like B<rint>(3)."
7015 #: build/C/man3/round.3:71
7016 msgid "For example, I<round(0.5)> is 1.0, and I<round(-0.5)> is -1.0."
7019 #. The POSIX.1-2001 APPLICATION USAGE SECTION discusses this point.
7021 #: build/C/man3/round.3:99
7023 "POSIX.1-2001 contains text about overflow (which might set I<errno> to "
7024 "B<ERANGE>, or raise an B<FE_OVERFLOW> exception). In practice, the result "
7025 "cannot overflow on any current machine, so this error-handling stuff is just "
7026 "nonsense. (More precisely, overflow can happen only when the maximum value "
7027 "of the exponent is smaller than the number of mantissa bits. For the "
7028 "IEEE-754 standard 32-bit and 64-bit floating-point numbers the maximum value "
7029 "of the exponent is 128 (respectively, 1024), and the number of mantissa bits "
7030 "is 24 (respectively, 53).)"
7034 #: build/C/man3/round.3:104
7036 "If you want to store the rounded value in an integer type, you probably want "
7037 "to use one of the functions described in B<lround>(3) instead."
7041 #: build/C/man3/round.3:110
7043 "B<ceil>(3), B<floor>(3), B<lround>(3), B<nearbyint>(3), B<rint>(3), "
7048 #: build/C/man3/scalb.3:25
7054 #: build/C/man3/scalb.3:29
7056 "scalb, scalbf, scalbl - multiply floating-point number by integral power of "
7061 #: build/C/man3/scalb.3:33
7062 msgid "B<double scalb(double >I<x>B<, double >I<exp>B<);>"
7066 #: build/C/man3/scalb.3:35
7067 msgid "B<float scalbf(float >I<x>B<, double >I<exp>B<);>"
7071 #: build/C/man3/scalb.3:37
7072 msgid "B<long double scalbl(long double >I<x>B<, double >I<exp>B<);>"
7076 #: build/C/man3/scalb.3:47
7081 #: build/C/man3/scalb.3:50
7083 "_BSD_SOURCE || _SVID_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 500 || "
7084 "_XOPEN_SOURCE\\ &&\\ _XOPEN_SOURCE_EXTENDED"
7088 #: build/C/man3/scalb.3:54
7089 msgid "B<scalbf>(), B<scalbl>():"
7093 #: build/C/man3/scalb.3:67 build/C/man3/scalbln.3:83
7095 "These functions multiply their first argument I<x> by B<FLT_RADIX> (probably "
7096 "2) to the power of I<exp>, that is:"
7100 #: build/C/man3/scalb.3:70 build/C/man3/scalbln.3:86
7102 msgid " x * FLT_RADIX ** exp\n"
7105 #. not in /usr/include but in a gcc lib
7107 #: build/C/man3/scalb.3:77 build/C/man3/scalbln.3:93
7109 "The definition of B<FLT_RADIX> can be obtained by including "
7110 "I<E<lt>float.hE<gt>>."
7114 #: build/C/man3/scalb.3:79 build/C/man3/scalbln.3:95
7115 msgid "On success, these functions return I<x> * B<FLT_RADIX> ** I<exp>."
7119 #: build/C/man3/scalb.3:85
7120 msgid "If I<x> or I<exp> is a NaN, a NaN is returned."
7124 #: build/C/man3/scalb.3:93
7126 "If I<x> is positive infinity (negative infinity), and I<exp> is not negative "
7127 "infinity, positive infinity (negative infinity) is returned."
7131 #: build/C/man3/scalb.3:99
7133 "If I<x> is +0 (-0), and I<exp> is not positive infinity, +0 (-0) is "
7138 #: build/C/man3/scalb.3:107
7140 "If I<x> is zero, and I<exp> is positive infinity, a domain error occurs, and "
7141 "a NaN is returned."
7145 #: build/C/man3/scalb.3:116
7147 "If I<x> is an infinity, and I<exp> is negative infinity, a domain error "
7148 "occurs, and a NaN is returned."
7152 #: build/C/man3/scalb.3:131 build/C/man3/scalbln.3:123
7154 "If the result underflows, a range error occurs, and the functions return "
7155 "zero, with a sign the same as I<x>."
7159 #: build/C/man3/scalb.3:138
7162 "Domain error: I<x> is 0, and I<exp> is positive infinity, or I<x> is "
7163 "positive infinity and I<exp> is negative infinity and the other argument is "
7168 #: build/C/man3/scalb.3:184
7170 "B<scalb>() is specified in POSIX.1-2001, but marked obsolescent. "
7171 "POSIX.1-2008 removes the specification of B<scalb>(), recommending the use "
7172 "of B<scalbln>(3), B<scalblnf>(3), or B<scalblnl>(3) instead. The "
7173 "B<scalb>() function is from 4.3BSD."
7176 #. Looking at header files: scalbf() is present on the
7177 #. BSDs, Tru64, HP-UX 11, Irix 6.5; scalbl() is on HP-UX 11 and Tru64.
7179 #: build/C/man3/scalb.3:193
7181 "B<scalbf>() and B<scalbl>() are unstandardized; B<scalbf>() is "
7182 "nevertheless present on several other systems"
7186 #: build/C/man3/scalb.3:195
7187 msgid "B<ldexp>(3), B<scalbln>(3)"
7191 #: build/C/man3/scalbln.3:25
7197 #: build/C/man3/scalbln.3:29
7199 "scalbn, scalbnf, scalbnl, scalbln, scalblnf, scalblnl - multiply "
7200 "floating-point number by integral power of radix"
7204 #: build/C/man3/scalbln.3:33
7205 msgid "B<double scalbln(double >I<x>B<, long int >I<exp>B<);>"
7209 #: build/C/man3/scalbln.3:35
7210 msgid "B<float scalblnf(float >I<x>B<, long int >I<exp>B<);>"
7214 #: build/C/man3/scalbln.3:37
7215 msgid "B<long double scalblnl(long double >I<x>B<, long int >I<exp>B<);>"
7219 #: build/C/man3/scalbln.3:39
7220 msgid "B<double scalbn(double >I<x>B<, int >I<exp>B<);>"
7224 #: build/C/man3/scalbln.3:41
7225 msgid "B<float scalbnf(float >I<x>B<, int >I<exp>B<);>"
7229 #: build/C/man3/scalbln.3:43
7230 msgid "B<long double scalbnl(long double >I<x>B<, int >I<exp>B<);>"
7234 #: build/C/man3/scalbln.3:55
7235 msgid "B<scalbln>(), B<scalblnf>(), B<scalblnl>():"
7239 #: build/C/man3/scalbln.3:66
7240 msgid "B<scalbn>(), B<scalbnf>(), B<scalbnl>():"
7244 #: build/C/man3/scalbln.3:164
7246 "These functions differ from the obsolete functions described in B<scalb>(3) "
7247 "in the type of their second argument. The functions described on this page "
7248 "have a second argument of an integral type, while those in B<scalb>(3) have "
7249 "a second argument of type I<double>."
7253 #: build/C/man3/scalbln.3:171
7255 "If B<FLT_RADIX> equals 2 (which is usual), then B<scalbn>() is equivalent "
7260 #: build/C/man3/scalbln.3:173
7261 msgid "B<ldexp>(3), B<scalb>(3)"
7265 #: build/C/man3/signbit.3:6
7271 #: build/C/man3/signbit.3:9
7272 msgid "signbit - test sign of a real floating-point number"
7276 #: build/C/man3/signbit.3:13
7277 msgid "B<int signbit(>I<x>B<);>"
7281 #: build/C/man3/signbit.3:23
7282 msgid "B<signbit>():"
7286 #: build/C/man3/signbit.3:37
7288 "B<signbit>() is a generic macro which can work on all real floating-point "
7289 "types. It returns a nonzero value if the value of I<x> has its sign bit "
7294 #: build/C/man3/signbit.3:46
7296 "This is not the same as I<x E<lt> 0.0>, because IEEE 754 floating point "
7297 "allows zero to be signed. The comparison I<-0.0 E<lt> 0.0> is false, but "
7298 "I<signbit(-0.0)> will return a nonzero value."
7302 #: build/C/man3/signbit.3:48
7303 msgid "NaNs and infinities have a sign bit."
7307 #: build/C/man3/signbit.3:54
7309 "The B<signbit>() macro returns nonzero if the sign of I<x> is negative; "
7310 "otherwise it returns zero."
7314 #: build/C/man3/signbit.3:61
7315 msgid "B<copysign>(3)"
7319 #: build/C/man3/significand.3:4
7325 #: build/C/man3/significand.3:4
7331 #: build/C/man3/significand.3:8
7333 "significand, significandf, significandl - get mantissa of floating-point "
7338 #: build/C/man3/significand.3:12
7339 msgid "B<double significand(double >I<x>B<);>"
7343 #: build/C/man3/significand.3:14
7344 msgid "B<float significandf(float >I<x>B<);>"
7348 #: build/C/man3/significand.3:16
7349 msgid "B<long double significandl(long double >I<x>B<);>"
7353 #: build/C/man3/significand.3:28
7354 msgid "B<significand>(), B<significandf>(), B<significandl>():"
7358 #: build/C/man3/significand.3:39
7360 "The B<significand>() function returns the mantissa of I<x> scaled to the "
7361 "range [1,2). It is equivalent to"
7365 #: build/C/man3/significand.3:42
7366 msgid "scalb(x, (double) -ilogb(x))"
7370 #: build/C/man3/significand.3:46
7372 "This function exists mainly for use in certain standardized tests for IEEE "
7377 #. This function came from BSD.
7379 #: build/C/man3/significand.3:52
7381 "These functions are nonstandard; the I<double> version is available on a "
7382 "number of other systems."
7386 #: build/C/man3/significand.3:54
7387 msgid "B<ilogb>(3), B<scalb>(3)"
7391 #: build/C/man3/sin.3:33
7397 #: build/C/man3/sin.3:36
7398 msgid "sin, sinf, sinl - sine function"
7402 #: build/C/man3/sin.3:41
7404 msgid "B<double sin(double >I<x>B<);>\n"
7408 #: build/C/man3/sin.3:43
7410 msgid "B<float sinf(float >I<x>B<);>\n"
7414 #: build/C/man3/sin.3:45
7416 msgid "B<long double sinl(long double >I<x>B<);>\n"
7420 #: build/C/man3/sin.3:57
7421 msgid "B<sinf>(), B<sinl>():"
7425 #: build/C/man3/sin.3:70
7427 "The B<sin>() function returns the sine of I<x>, where I<x> is given in "
7432 #: build/C/man3/sin.3:73
7433 msgid "On success, these functions return the sine of I<x>."
7437 #: build/C/man3/sin.3:123
7439 "B<acos>(3), B<asin>(3), B<atan>(3), B<atan2>(3), B<cos>(3), B<csin>(3), "
7440 "B<sincos>(3), B<tan>(3)"
7444 #: build/C/man3/sincos.3:6
7450 #: build/C/man3/sincos.3:9
7451 msgid "sincos, sincosf, sincosl - calculate sin and cos simultaneously"
7455 #: build/C/man3/sincos.3:15
7457 msgid "B<void sincos(double >I<x>B<, double *>I<sin>B<, double *>I<cos>B<);>\n"
7461 #: build/C/man3/sincos.3:17
7463 msgid "B<void sincosf(float >I<x>B<, float *>I<sin>B<, float *>I<cos>B<);>\n"
7467 #: build/C/man3/sincos.3:19
7470 "B<void sincosl(long double >I<x>B<, long double *>I<sin>B<, long double "
7475 #: build/C/man3/sincos.3:29
7477 "Several applications need sine and cosine of the same angle I<x>. This "
7478 "function computes both at the same time, and stores the results in I<*sin> "
7483 #: build/C/man3/sincos.3:37
7484 msgid "If I<x> is a NaN, a NaN is returned in I<*sin> and I<*cos>."
7488 #: build/C/man3/sincos.3:46
7490 "If I<x> is positive infinity or negative infinity, a domain error occurs, "
7491 "and a NaN is returned in I<*sin> and I<*cos>."
7495 #: build/C/man3/sincos.3:49
7496 msgid "These functions return I<void>."
7500 #: build/C/man3/sincos.3:74
7501 msgid "This function is a GNU extension."
7505 #: build/C/man3/sincos.3:77
7506 msgid "B<cos>(3), B<sin>(3), B<tan>(3)"
7510 #: build/C/man3/sinh.3:34
7516 #: build/C/man3/sinh.3:37
7517 msgid "sinh, sinhf, sinhl - hyperbolic sine function"
7521 #: build/C/man3/sinh.3:42
7523 msgid "B<double sinh(double >I<x>B<);>\n"
7527 #: build/C/man3/sinh.3:44
7529 msgid "B<float sinhf(float >I<x>B<);>\n"
7533 #: build/C/man3/sinh.3:46
7535 msgid "B<long double sinhl(long double >I<x>B<);>\n"
7539 #: build/C/man3/sinh.3:58
7540 msgid "B<sinhf>(), B<sinhl>():"
7544 #: build/C/man3/sinh.3:71
7546 "The B<sinh>() function returns the hyperbolic sine of I<x>, which is "
7547 "defined mathematically as:"
7551 #: build/C/man3/sinh.3:74
7553 msgid " sinh(x) = (exp(x) - exp(-x)) / 2\n"
7557 #: build/C/man3/sinh.3:78
7558 msgid "On success, these functions return the hyperbolic sine of I<x>."
7562 #. POSIX.1-2001 documents an optional range error (underflow)
7564 #. glibc 2.8 does not do this.
7566 #: build/C/man3/sinh.3:105
7568 "If the result overflows, a range error occurs, and the functions return "
7569 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, with the same sign "
7574 #: build/C/man3/sinh.3:132
7575 msgid "B<acosh>(3), B<asinh>(3), B<atanh>(3), B<cosh>(3), B<csinh>(3), B<tanh>(3)"
7579 #: build/C/man3/sqrt.3:32
7585 #: build/C/man3/sqrt.3:35
7586 msgid "sqrt, sqrtf, sqrtl - square root function"
7590 #: build/C/man3/sqrt.3:40
7592 msgid "B<double sqrt(double >I<x>B<);>\n"
7596 #: build/C/man3/sqrt.3:42
7598 msgid "B<float sqrtf(float >I<x>B<);>\n"
7602 #: build/C/man3/sqrt.3:44
7604 msgid "B<long double sqrtl(long double >I<x>B<);>\n"
7608 #: build/C/man3/sqrt.3:56
7609 msgid "B<sqrtf>(), B<sqrtl>():"
7613 #: build/C/man3/sqrt.3:68
7614 msgid "The B<sqrt>() function returns the nonnegative square root of I<x>."
7618 #: build/C/man3/sqrt.3:71
7619 msgid "On success, these functions return the square root of I<x>."
7623 #: build/C/man3/sqrt.3:89
7624 msgid "If I<x> is less than -0, a domain error occurs, and a NaN is returned."
7628 #: build/C/man3/sqrt.3:96
7630 msgid "Domain error: I<x> less than -0"
7634 #: build/C/man3/sqrt.3:113
7635 msgid "B<cbrt>(3), B<csqrt>(3), B<hypot>(3)"
7639 #: build/C/man3/tan.3:33
7645 #: build/C/man3/tan.3:36
7646 msgid "tan, tanf, tanl - tangent function"
7650 #: build/C/man3/tan.3:41
7652 msgid "B<double tan(double >I<x>B<);>\n"
7656 #: build/C/man3/tan.3:43
7658 msgid "B<float tanf(float >I<x>B<);>\n"
7662 #: build/C/man3/tan.3:45
7664 msgid "B<long double tanl(long double >I<x>B<);>\n"
7668 #: build/C/man3/tan.3:57
7669 msgid "B<tanf>(), B<tanl>():"
7673 #: build/C/man3/tan.3:70
7675 "The B<tan>() function returns the tangent of I<x>, where I<x> is given in "
7680 #: build/C/man3/tan.3:73
7681 msgid "On success, these functions return the tangent of I<x>."
7684 #. I think overflow can't occur, because the closest floating-point
7685 #. representation of pi/2 is still not close enough to pi/2 to
7686 #. produce a large enough value to overflow.
7687 #. Testing certainly seems to bear this out. -- mtk, Jul 08
7689 #. POSIX.1 allows an optional underflow error;
7690 #. glibc 2.8 doesn't do this
7691 #. POSIX.1 an optional range error for subnormal x;
7692 #. glibc 2.8 doesn't do this
7694 #: build/C/man3/tan.3:101
7696 "If the correct result would overflow, a range error occurs, and the "
7697 "functions return B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, "
7698 "with the mathematically correct sign."
7702 #: build/C/man3/tan.3:147
7704 "B<acos>(3), B<asin>(3), B<atan>(3), B<atan2>(3), B<cos>(3), B<ctan>(3), "
7709 #: build/C/man3/tanh.3:33
7715 #: build/C/man3/tanh.3:36
7716 msgid "tanh, tanhf, tanhl - hyperbolic tangent function"
7720 #: build/C/man3/tanh.3:41
7722 msgid "B<double tanh(double >I<x>B<);>\n"
7726 #: build/C/man3/tanh.3:43
7728 msgid "B<float tanhf(float >I<x>B<);>\n"
7732 #: build/C/man3/tanh.3:45
7734 msgid "B<long double tanhl(long double >I<x>B<);>\n"
7738 #: build/C/man3/tanh.3:57
7739 msgid "B<tanhf>(), B<tanhl>():"
7743 #: build/C/man3/tanh.3:70
7745 "The B<tanh>() function returns the hyperbolic tangent of I<x>, which is "
7746 "defined mathematically as:"
7750 #: build/C/man3/tanh.3:73
7752 msgid " tanh(x) = sinh(x) / cosh(x)\n"
7756 #: build/C/man3/tanh.3:77
7757 msgid "On success, these functions return the hyperbolic tangent of I<x>."
7761 #: build/C/man3/tanh.3:108
7762 msgid "B<acosh>(3), B<asinh>(3), B<atanh>(3), B<cosh>(3), B<ctanh>(3), B<sinh>(3)"
7766 #: build/C/man3/tgamma.3:8
7772 #: build/C/man3/tgamma.3:11
7773 msgid "tgamma, tgammaf, tgammal - true gamma function"
7777 #: build/C/man3/tgamma.3:15
7778 msgid "B<double tgamma(double >I<x>B<);>"
7782 #: build/C/man3/tgamma.3:17
7783 msgid "B<float tgammaf(float >I<x>B<);>"
7787 #: build/C/man3/tgamma.3:19
7788 msgid "B<long double tgammal(long double >I<x>B<);>"
7792 #: build/C/man3/tgamma.3:31
7793 msgid "B<tgamma>(), B<tgammaf>(), B<tgammal>():"
7797 #: build/C/man3/tgamma.3:41
7798 msgid "The Gamma function is defined by"
7802 #: build/C/man3/tgamma.3:43
7804 msgid " Gamma(x) = integral from 0 to infinity of t^(x-1) e^-t dt\n"
7808 #: build/C/man3/tgamma.3:46
7810 "It is defined for every real number except for nonpositive integers. For "
7811 "nonnegative integral I<m> one has"
7815 #: build/C/man3/tgamma.3:48
7817 msgid " Gamma(m+1) = m!\n"
7821 #: build/C/man3/tgamma.3:50
7822 msgid "and, more generally, for all I<x>:"
7826 #: build/C/man3/tgamma.3:52
7828 msgid " Gamma(x+1) = x * Gamma(x)\n"
7832 #: build/C/man3/tgamma.3:55
7834 "Furthermore, the following is valid for all values of I<x> outside the "
7839 #: build/C/man3/tgamma.3:57
7841 msgid " Gamma(x) * Gamma(1 - x) = PI / sin(PI * x)\n"
7845 #: build/C/man3/tgamma.3:60
7846 msgid "On success, these functions return Gamma(x)."
7850 #: build/C/man3/tgamma.3:74
7852 "If I<x> is a negative integer, or is negative infinity, a domain error "
7853 "occurs, and a NaN is returned."
7857 #: build/C/man3/tgamma.3:87
7859 "If the result underflows, a range error occurs, and the functions return 0, "
7860 "with the correct mathematical sign."
7864 #: build/C/man3/tgamma.3:98
7866 "If I<x> is -0 or +0, a pole error occurs, and the functions return "
7867 "B<HUGE_VAL>, B<HUGE_VALF>, or B<HUGE_VALL>, respectively, with the same sign "
7872 #: build/C/man3/tgamma.3:105
7874 msgid "Domain error: I<x> is a negative integer, or negative infinity"
7877 #. FIXME . errno is not set to EDOM for x == -inf
7878 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6809
7880 #: build/C/man3/tgamma.3:115
7882 "I<errno> is set to B<EDOM>. An invalid floating-point exception "
7883 "(B<FE_INVALID>) is raised (but see BUGS)."
7887 #: build/C/man3/tgamma.3:115
7889 msgid "Pole error: I<x> is +0 or -0"
7893 #: build/C/man3/tgamma.3:134
7895 "glibc also gives the following error which is not specified in C99 or "
7899 #. FIXME . Is it intentional that errno is not set:
7900 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6810
7902 #. glibc (as at 2.8) also supports and an inexact
7903 #. exception for various cases.
7905 #: build/C/man3/tgamma.3:151
7906 msgid "I<errno> is not set for this case."
7910 #: build/C/man3/tgamma.3:162
7912 "This function had to be called \"true gamma function\" since there is "
7913 "already a function B<gamma>(3) that returns something else (see B<gamma>(3) "
7917 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6809
7919 #: build/C/man3/tgamma.3:170
7921 "If I<x> is negative infinity, I<errno> is not set (it should be set to "
7926 #: build/C/man3/tgamma.3:179
7928 "In glibc versions 2.3.3 and earlier, an argument of +0 or -0 incorrectly "
7929 "produced a domain error (I<errno> set to B<EDOM> and an B<FE_INVALID> "
7930 "exception raised), rather than a pole error."
7934 #: build/C/man3/tgamma.3:181
7935 msgid "B<gamma>(3), B<lgamma>(3)"
7939 #: build/C/man3/trunc.3:23
7945 #: build/C/man3/trunc.3:26
7946 msgid "trunc, truncf, truncl - round to integer, toward zero"
7950 #: build/C/man3/trunc.3:31
7952 msgid "B<double trunc(double >I<x>B<);>\n"
7956 #: build/C/man3/trunc.3:33
7958 msgid "B<float truncf(float >I<x>B<);>\n"
7962 #: build/C/man3/trunc.3:35
7964 msgid "B<long double truncl(long double >I<x>B<);>\n"
7968 #: build/C/man3/trunc.3:48
7969 msgid "B<trunc>(), B<truncf>(), B<truncl>():"
7973 #: build/C/man3/trunc.3:59
7975 "These functions round I<x> to the nearest integer not larger in absolute "
7980 #: build/C/man3/trunc.3:63
7981 msgid "If I<x> is integral, infinite, or NaN, I<x> itself is returned."
7985 #: build/C/man3/y0.3:34
7991 #: build/C/man3/y0.3:34
7997 #: build/C/man3/y0.3:38
7999 "y0, y0f, y0l, y1, y1f, y1l, yn, ynf, ynl - Bessel functions of the second "
8004 #: build/C/man3/y0.3:43
8006 msgid "B<double y0(double >I<x>B<);>\n"
8010 #: build/C/man3/y0.3:45
8012 msgid "B<double y1(double >I<x>B<);>\n"
8016 #: build/C/man3/y0.3:47
8018 msgid "B<double yn(int >I<n>B<, double >I<x>B<);>\n"
8022 #: build/C/man3/y0.3:49
8024 msgid "B<float y0f(float >I<x>B<);>\n"
8028 #: build/C/man3/y0.3:51
8030 msgid "B<float y1f(float >I<x>B<);>\n"
8034 #: build/C/man3/y0.3:53
8036 msgid "B<float ynf(int >I<n>B<, float >I<x>B<);>\n"
8040 #: build/C/man3/y0.3:55
8042 msgid "B<long double y0l(long double >I<x>B<);>\n"
8046 #: build/C/man3/y0.3:57
8048 msgid "B<long double y1l(long double >I<x>B<);>\n"
8052 #: build/C/man3/y0.3:59
8054 msgid "B<long double ynl(int >I<n>B<, long double >I<x>B<);>\n"
8058 #: build/C/man3/y0.3:72
8059 msgid "B<y0>(), B<y1>(), B<yn>():"
8063 #: build/C/man3/y0.3:82
8064 msgid "B<y0f>(), B<y0l>(), B<y1f>(), B<y1l>(), B<ynf>(), B<ynl>():"
8068 #: build/C/man3/y0.3:99
8070 "The B<y0>() and B<y1>() functions return Bessel functions of I<x> of the "
8071 "second kind of orders 0 and 1, respectively. The B<yn>() function returns "
8072 "the Bessel function of I<x> of the second kind of order I<n>."
8076 #: build/C/man3/y0.3:101
8077 msgid "The value of I<x> must be positive."
8081 #: build/C/man3/y0.3:111
8083 "The B<y0f>() etc. and B<y0l>() etc. functions are versions that take and "
8084 "return I<float> and I<long double> values, respectively."
8088 #: build/C/man3/y0.3:115
8090 "On success, these functions return the appropriate Bessel value of the "
8091 "second kind for I<x>."
8095 #: build/C/man3/y0.3:131
8097 "If I<x> is negative, a domain error occurs, and the functions return "
8098 "-B<HUGE_VAL>, -B<HUGE_VALF>, or -B<HUGE_VALL>, respectively. (POSIX.1-2001 "
8099 "also allows a NaN return for this case.)"
8103 #: build/C/man3/y0.3:142
8105 "If I<x> is 0.0, a pole error occurs, and the functions return -B<HUGE_VAL>, "
8106 "-B<HUGE_VALF>, or -B<HUGE_VALL>, respectively."
8110 #: build/C/man3/y0.3:146
8111 msgid "If the result underflows, a range error occurs, and the functions return 0.0"
8115 #: build/C/man3/y0.3:156
8117 "If the result overflows, a range error occurs, and the functions return "
8118 "-B<HUGE_VAL>, -B<HUGE_VALF>, or -B<HUGE_VALL>, respectively. (POSIX.1-2001 "
8119 "also allows a 0.0 return for this case.)"
8123 #: build/C/man3/y0.3:171
8125 msgid "Pole error: I<x> is 0.0"
8128 #. Before POSIX.1-2001 TC2, this was (inconsistently) specified
8129 #. as a range error.
8130 #. FIXME . y0(0.0) gives EDOM
8131 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6808
8133 #: build/C/man3/y0.3:186
8135 "I<errno> is set to B<ERANGE> (but see BUGS). No B<FE_DIVBYZERO> exception "
8136 "is returned by B<fetestexcept>(3) for this case."
8139 #. e.g., y0(1e33) on glibc 2.8/x86-32
8140 #. An underflow floating-point exception
8141 #. .RB ( FE_UNDERFLOW )
8143 #. FIXME . Is it intentional that these functions do not use FE_*?
8144 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6806
8146 #: build/C/man3/y0.3:202
8148 "I<errno> is set to B<ERANGE>. No B<FE_UNDERFLOW> exception is returned by "
8149 "B<fetestexcept>(3) for this case."
8152 #. e.g., yn(10, 1e-40) on glibc 2.8/x86-32
8156 #. FIXME . Is it intentional that errno is not set?
8157 #. Bug raised: http://sources.redhat.com/bugzilla/show_bug.cgi?id=6808
8159 #: build/C/man3/y0.3:215
8161 "I<errno> is not set for this case. An overflow floating-point exception "
8162 "(B<FE_OVERFLOW>) is raised."
8166 #. Bug raised: http://sourceware.org/bugzilla/show_bug.cgi?id=6807
8168 #: build/C/man3/y0.3:231
8170 "On a pole error, these functions set I<errno> to B<EDOM>, instead of "
8171 "B<ERANGE> as POSIX.1-2004 requires."
8174 #. FIXME . Actually, 2.3.2 is the earliest test result I have; so yet
8175 #. to confirm if this error occurs only in 2.3.2.
8177 #: build/C/man3/y0.3:238
8179 "In glibc version 2.3.2 and earlier, these functions do not raise an invalid "
8180 "floating-point exception (B<FE_INVALID>) when a domain error occurs."
8184 #: build/C/man3/y0.3:239