Legacy Document

log2

You can use the log2 function to compute the binary logarithm of a real number.

double_t log2 (double_t x);

x

Any positive floating-point number.

DESCRIPTION

The log2 function returns the binary (base 2) logarithm of its argument.

$log2(x)\; =log$₂x = y such that $x\; =\; 2y$

The exp2 function performs the inverse operation.

EXCEPTIONS

When x is finite and nonzero, the result of $log2(x)$ might raise one of the following exceptions:

• inexact (for all finite, nonzero values of x other than +1)
• invalid (when x is negative)

SPECIAL CASES

Table 10-17 shows the results when the argument to the log2 function is a zero, a NaN, or an Infinity, plus other special cases for the log2 function.

Special cases for the log2 function

Operation	Result	Exceptions raised
$log2(x)$ for x < 0	NaN	Invalid
$log2(+1)$	+0	None
$log2(+0)$	$-$	Divide-by-zero
$log2(-0)$	$-$	Divide-by-zero
$log2(NaN)$	NaN	None[38]
$log2(+$ )	+	None
$log2(-$ )	NaN	Invalid

EXAMPLES

z = log2(+1.0);   /* z = +0 because 20 = 1 */
z = log2(2.0);    /* z = 1 because 21 = 2. The inexact exception 
                     is raised. */
z = log2(-1.0);   /* z = NAN because negative arguments are not 
                     allowed. The invalid exception is raised. */

---

[38] If the NaN is a signaling NaN, the invalid exception is raised.