FPCLASSIFY(3)             OpenBSD Programmer's Manual            FPCLASSIFY(3)

NAME
     fpclassify, isfinite, isinf, isnan, isnormal, signbit - classify a
     floating-point number

SYNOPSIS
     #include <math.h>

     int
     fpclassify(real-floating x);

     int
     isfinite(real-floating x);

     int
     isinf(real-floating x);

     int
     isnan(real-floating x);

     int
     isnormal(real-floating x);

     int
     signbit(real-floating x);

DESCRIPTION
     The fpclassify() macro takes an argument of x and returns one of the
     following manifest constants:

           FP_INFINITE      Indicates that x is an infinite number.
           FP_NAN           Indicates that x is not a number (NaN).
           FP_NORMAL        Indicates that x is a normalized number.
           FP_SUBNORMAL     Indicates that x is a denormalized number.
           FP_ZERO          Indicates that x is zero (0 or -0).

     The isfinite() macro returns a non-zero value if and only if its argument
     has a finite (zero, subnormal, or normal) value.

     The isinf(), isnan(), and isnormal() macros return non-zero if and only
     if x is an infinity, NaN, or a non-zero normalized number, respectively.

     The signbit() macro takes an argument of x and returns non-zero if the
     value of its sign is negative, otherwise 0.

SEE ALSO
     isgreater(3), math(3)

STANDARDS
     The fpclassify(), isfinite(), isinf(), isnan(), isnormal(), and signbit()
     macros conform to ISO/IEC 9899:1999 (``ISO C99'').

     The symbols isinff(), and isnanf() are provided as compatibility aliases
     to isinf(), and isnan(), respectively, and their uses are deprecated.
     Similarly, finite() and finitef() are deprecated versions of isfinite().

HISTORY
     The fpclassify(), isfinite(), isinf(), isnan(), isnormal(), and signbit()
     macros were added in OpenBSD 4.4.  3BSD introduced isinf() and isnan()
     functions, which accepted double arguments; these have been superseded by
     the macros described above.

BUGS
     VAX floating point does not have distinguished values for either infinity
     or not-a-number.  isinf(), isinff(), isnan(), and isnanf() routines
     always return 0 on that architecture.  Similarly, finite(), and
     isfinite() always return 1.

OpenBSD 4.9                      July 29, 2008                     OpenBSD 4.9