NAME
fmax(), fmaxf(), fmaxl(), fmaxw(), fmaxq() — maximum value functions
SYNOPSIS
#include <math.h>
double fmax(double x, double y);
Itanium(R)-based System Only
float fmaxf(float x, float y);
long double fmaxl(long double x, long double y);
extended fmaxw(extended x, extended y);
quad fmaxq(quad x, quad y);
DESCRIPTION
The
fmax()
function determines the maximum numeric value of its arguments.
Itanium-based System Only
fmaxf()
is a
float
version of
fmax();
it takes
float
arguments and returns a
float
result.
fmaxl()
is a
long double
version of
fmax();
it takes
long double
arguments and returns a
long double
result.
fmaxw()
is an
extended
version of
fmax();
it takes
extended
arguments and returns an
extended
result.
fmaxq()
is equivalent to
fmaxl()
on HP-UX systems.
USAGE
To use this function, compile either with the default
-Ae
option or with the
-Aa
and
-D_HPUX_SOURCE
options. To use (for Itanium-based systems)
fmaxw()
or
fmaxq(),
compile with the
-fpwidetypes
option. Make sure your program includes
<math.h>.
Link in the math library by specifying
-lm
on the compiler or linker command line.
RETURN VALUE
The
fmax()
function returns the maximum numeric value of its arguments.
If one argument is a NaN and the other is numeric,
fmax()
returns the numeric argument.
If both arguments are NaNs,
fmax()
returns NaN.
ERRORS
No errors are defined.
STANDARDS CONFORMANCE
fmax(),
fmaxf(),
fmaxl()
: ISO/IEC C99 (including Annex F, ``IEC 60559 floating-point arithmetic'')
