The

ippInit

function detects the processor type and sets the dispatcher to use the processor-specific code of the Intel® IPP library corresponding to the instruction set capabilities available. If your application does not call the

ippInit

function, initialization of the library to the available instruction set capabilities is performed automatically with the first call of any Intel IPP function from the domain different from

ippCore

.

In some cases like debugging and performance analysis, you may want to get the data on the difference between various processor-specific codes on the same machine. Use the

ippSetCpuFeatures

function for this. This function sets the dispatcher to use the processor-specific code according to the specified set of CPU features. You can obtain features supported by CPU using

ippGetCpuFeatures

and obtain features supported by the currently dispatched Intel IPP code using

ippGetEnabledCpuFeatures

. If you need to enable support of some CPU features without querying the system (without CPUID instruction call), you must set the

ippCPUID_NOCHECK

bit for

ippSetCpuFeatures

, otherwise, only supported by the current CPU features are set.

The

ippInit

,

ippGetCpuFeatures

,

ippGetEnabledCpuFeatures

, and

ippSetCpuFeatures

functions are a part of the

ippCore

library.

The

ippGetStatusString

function decodes the numeric status return value of Intel® IPP functions and converts them to a human readable text:

The

ippGetStatusString

function is a part of the

ippCore

library.

Each domain has its own

GetLibVersion

function that returns information about the library layer in use from the dispatcher. The code snippet below demonstrates the usage of the

ippiGetLibVersion

from the image processing domain:

Use this function in combination with

ippInitCpu

to compare the output of different implementations on the same machine.

Intel IPP functions provide better performance if they process data with aligned pointers. Intel IPP provides the following functions to ensure that data is aligned appropriately - 16-byte for CPU that does not support Intel® Advanced Vector Extensions (Intel® AVX) instruction set, 32-byte for Intel AVX and Intel® Advanced Vector Extensions 2 (Intel® AVX2), and 64-byte for Intel® Many Integrated Core instructions.

The

ippMalloc

function provides appropriately aligned buffer, and the

ippFree

function frees it.

The signal and image processing libraries provide

ippsMalloc

and

ippiMalloc

functions, respectively, to allocate appropriately aligned buffer that can be freed by the

ippsFree

and

ippiFree

functions.