Inside Macintosh: PowerPC Numerics / Part 2 - The PowerPC Numerics C Implementation

Chapter 9 - Conversion Functions

Rounding Floating-Point Numbers to Integers

MathLib provides six functions that convert floating-point numbers to integers and return the integer in a floating-point type. The first is the rint function, which performs the round-to-integer operation as described in Chapter 6, "Numeric Operations and Functions." The other functions either round in a specific direction or perform a variation of the rint operation.

$ceil(x)$	Returns the nearest integer not less than x.
$floor(x)$	Returns the nearest integer not greater than x.
$nearbyint(x)$	Returns the nearest integer to x in the current rounding direction.
$round(x)$	Adds 1/2 to the magnitude of x and chops to an integer.
$trunc(x)$	Truncates the fractional part of x.