ilogbf4

NAME

ilogbf4 - return integer exponents of float elements

SYNOPSIS

Procedure call syntax:

#include <simdmath.h>

#include <math.h>

vector signed int ilogbf4(vector float x);

Link with -lsimdmath

Inline call syntax:

#include <simdmath.h>

#include <ilogbf4.h>

vector signed int _ilogbf4(vector float x);

Parameters
x	input vector

DESCRIPTION

The ilogbf4 function extracts the exponents of the input vector as signed integers.

Because the SPU treats single-precision Inf and NaN codes as regular floating point numbers, ilogbf4 returns a result of 128 for these values. However, FP_ILOGBNAN is set to INT_MAX for compatibility with the double function ilogbd2.

RETURN VALUE

The function ilogbf4 returns a signed int vector in which each element is defined as:

the macro FP_ILOGBNAN if the corresponding element of x is not a number (NaN),
the macro FP_ILOGB0 if the corresponding element of x is equal to 0 or Inf, or
the value of (int)logb(x) for the corresponding element of x otherwise.

ENVIRONMENT

SPU and PPU

CONFORMING TO

SIMD Math library specification for the Cell Broadband Engine™ Architecture.

NOTES

Basis

ISO9899 (C99) ilogb functions.