man3/lgamma.3

   1 .\" Copyright 2002 Walter Harms (walter.harms@informatik.uni-oldenburg.de)
   2 .\" Distributed under GPL
   3 .\" based on glibc infopages
   4 .\"
   5 .TH LGAMMA 3 2002-08-10 "" "Linux Programmer's Manual"
   6 .SH NAME
   7 lgamma, lgammaf, lgammal, lgamma_r, lgammaf_r, lgammal_r \- log gamma function
   8 .SH SYNOPSIS
   9 .nf
  10 .B #include <math.h>
  11 .sp
  12 .BI "double lgamma(double " x );
  13 .br
  14 .BI "float lgammaf(float " x );
  15 .br
  16 .BI "long double lgammal(long double " x );
  17 .sp
  18 .BI "double lgamma_r(double " x ", int *" signp );
  19 .br
  20 .BI "float lgammaf_r(float " x ", int *" signp );
  21 .br
  22 .BI "long double lgammal_r(long double " x ", int *" signp );
  23 .fi
  24 .sp
  25 Compile with \fI\-std=c99\fP; link with \fI\-lm\fP.
  26 .SH DESCRIPTION
  27 For the definition of the Gamma function, see
  28 .BR tgamma (3).
  29 .PP
  30 The
  31 .BR lgamma ()
  32 function returns the natural logarithm of
  33 the absolute value of the Gamma function.
  34 The sign of the Gamma function is returned in the
  35 external integer \fIsigngam\fP declared in
  36 .IR <math.h> .
  37 It is 1 when the Gamma function is positive or zero, \-1
  38 when it is negative.
  39 .PP
  40 Since using a constant location
  41 .I signgam
  42 is not thread-safe, the functions
  43 .BR lgamma_r ()
  44 etc. have
  45 been introduced; they return this sign via the parameter
  46 .IR signp .
  47 .PP
  48 For non-positive integer values of \fIx\fP,
  49 .BR lgamma ()
  50 returns
  51 .BR HUGE_VAL ,
  52 sets \fIerrno\fP to
  53 .B ERANGE
  54 and raises the zero divide exception.
  55 (Similarly,
  56 .BR lgammaf ()
  57 returns
  58 .B HUGE_VALF
  59 and
  60 .BR lgammal ()
  61 returns
  62 .BR HUGE_VALL .)
  63 .SH ERRORS
  64 In order to check for errors, set
  65 .I errno
  66 to zero and call
  67 .I feclearexcept(FE_ALL_EXCEPT)
  68 before calling these functions.
  69 On return, if
  70 .I errno
  71 is non-zero or
  72 .I fetestexcept(FE_INVALID | FE_DIVBYZERO | FE_OVERFLOW | FE_UNDERFLOW)
  73 is non-zero, an error has occurred.
  74 .LP
  75 A range error occurs if x is too large.
  76 A pole error occurs if x is a negative integer or zero.
  77 .SH "CONFORMING TO"
  78 C99, SVr4, 4.3BSD
  79 .SH "SEE ALSO"
  80 .BR tgamma (3)