Intel® Integrated Performance Primitives Developer Guide and Reference

ID 790148
Date 11/07/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

Expm1

Computes e raised to the power of each vector element and decreased by 1.

Syntax

IppStatus ippsExpm1_32f_A11 (const Ipp32f* pSrc, Ipp32f* pDst, Ipp32s len);

IppStatus ippsExpm1_32f_A21 (const Ipp32f* pSrc, Ipp32f* pDst, Ipp32s len);

IppStatus ippsExpm1_32f_A24 (const Ipp32f* pSrc, Ipp32f* pDst, Ipp32s len);

IppStatus ippsExpm1_64f_A26 (const Ipp64f* pSrc, Ipp64f* pDst, Ipp32s len);

IppStatus ippsExpm1_64f_A50 (const Ipp64f* pSrc, Ipp64f* pDst, Ipp32s len);

IppStatus ippsExpm1_64f_A53 (const Ipp64f* pSrc, Ipp64f* pDst, Ipp32s len);

Include Files

ippvm.h

Domain Dependencies

Headers: ippcore.h

Libraries: ippcore.lib

Parameters

pSrc

Pointer to the source vector.

pDst

Pointer to the destination vector.

len

Number of elements in the vectors.

Description

This function computes e raised to the power of each vector element of pSrc and decreased by 1, and stores the result in the corresponding element of the vector pDst.

For single precision data:

function flavor ippsExpm1_32f_A11 guarantees 11 correctly rounded bits of significand, or at least 3 exact decimal digits;

function flavor ippsExpm1_32f_A21 guarantees 21 correctly rounded bits of significand, or 4 ulps, or about 6 exact decimal digits;

function flavor ippsExpm1_32f_A24 guarantees 24 correctly rounded bits of significand, including the implied bit, with the maximum guaranteed error within 1 ulp.

For double precision data:

function flavor ippsExpm1_64f_A26 guarantees 26 correctly rounded bits of significand, or 6.7E+7 ulps, or approximately 8 exact decimal digits;

function flavor ippsExpm1_64f_A50 guarantees 50 correctly rounded bits of significand, or 4 ulps, or approximately 15 exact decimal digits;

function flavor ippsExpm1_64f_A53 guarantees 53 correctly rounded bits of significand, including the implied bit, with the maximum guaranteed error within 1 ulp.

Return Values

ippStsNoErr

Indicates no error.

ippStsNullPtrErr

Indicates an error when pSrc or pDst pointer is NULL.

ippStsSizeErr

Indicates an error when len is less than or equal to 0.

IppStsOverflow

Indicates a warning that the function overflows, that is, at least one of elements of pSrc is greater than Ln(FPMAX), where FPMAX is the maximum representable floating-point number.