frexp, frexpf, frexpl

Gets the mantissa and exponent of a floating-point number.

Syntax

double frexp(
   double x,
   int *expptr
);
float frexpf(
   float x,
   int * expptr
);
long double frexpl(
   long double x,
   int * expptr
);
#define frexpl(X, INT_PTR) // Requires C11 or higher
float frexp(
   float x,
   int * expptr
);  // C++ only
long double frexp(
   long double x,
   int * expptr
);  // C++ only

Parameters

x
Floating-point value.

expptr
Pointer to stored integer exponent.

Return value

frexp returns the mantissa. If x is 0, the function returns 0 for both the mantissa and the exponent. If expptr is NULL, the invalid parameter handler is invoked as described in Parameter validation. If execution is allowed to continue, this function sets errno to EINVAL and returns 0.

Remarks

The frexp function breaks down the floating-point value (x) into a mantissa (m) and an exponent (n), such that the absolute value of m is greater than or equal to 0.5 and less than 1.0, and x = m * 2n. The integer exponent n is stored at the location pointed to by expptr.

C++ allows overloading, so you can call overloads of frexp. In a C program, unless you're using the <tgmath.h> macro to call this function, frexp always takes a double and an int pointer and returns a double.

If you use the <tgmath.h> frexp() macro, the type of the argument determines which version of the function is selected. See Type-generic math for details.

By default, this function's global state is scoped to the application. To change this behavior, see Global state in the CRT.

Requirements

Function Required header
frexp, frexpf, frexpl <math.h>
frexp macro <tgmath.h>

For more compatibility information, see Compatibility.

Example

// crt_frexp.c
// This program calculates frexp( 16.4, &n )
// then displays y and n.

#include <math.h>
#include <stdio.h>

int main( void )
{
   double x, y;
   int n;

   x = 16.4;
   y = frexp( x, &n );
   printf( "frexp( %f, &n ) = %f, n = %d\n", x, y, n );
}
frexp( 16.400000, &n ) = 0.512500, n = 5

See also

Math and floating-point support
ldexp
modf, modff, modfl