man3/signbit.3

   1 .\" Copyright 2002 Walter Harms (walter.harms@informatik.uni-oldenburg.de)
   2 .\" and Copyright 2008, Linux Foundation, written by Michael Kerrisk
   3 .\"     <mtk.manpages@gmail.com>
   4 .\" Distributed under GPL
   5 .\" Based on glibc infopages
   6 .TH SIGNBIT 3 2008-08-05 "GNU" "Linux Programmer's Manual"
   7 .SH NAME
   8 signbit \- test sign of a real floating-point number
   9 .SH SYNOPSIS
  10 .B "#include <math.h>"
  11 .sp
  12 .BI  "int signbit(" x ");"
  13 .sp
  14 Link with \fI\-lm\fP.
  15 .sp
  16 .in -4n
  17 Feature Test Macro Requirements for glibc (see
  18 .BR feature_test_macros (7)):
  19 .in
  20 .sp
  21 .ad l
  22 .BR signbit ():
  23 _XOPEN_SOURCE\ >=\ 600 || _ISOC99_SOURCE; or
  24 .I cc\ -std=c99
  25 .ad b
  26 .SH DESCRIPTION
  27 .BR signbit ()
  28 is a generic macro which can work on all real floating-point types.
  29 It returns a non-zero value if the value of
  30 .I x
  31 has its sign bit set.
  32 .PP
  33 This is not the same as
  34 .IR "x < 0.0" ,
  35 because IEEE 754 floating point allows zero to be signed.
  36 The comparison
  37 .IR "-0.0 < 0.0"
  38 is false, but
  39 .IR "signbit(\-0.0)"
  40 will return a non-zero value.
  41
  42 NaNs and infinities have a sign bit.
  43 .SH RETURN VALUE
  44 The
  45 .BR signbit ()
  46 macro returns non-zero if the sign of
  47 .x
  48 is negative; otherwise it returns zero.
  49 .SH ERRORS
  50 No errors occur.
  51 .SH "CONFORMING TO"
  52 C99, POSIX.1-2001.
  53 This function is defined in IEC 559 (and the appendix with
  54 recommended functions in IEEE 754/IEEE 854).
  55 .SH "SEE ALSO"
  56 .BR copysign (3)