The following fragment of C code describes a regular matrix array shown in Figure “Standard Description”, A:
// Allocate memory for the array of continuous matrices
Ipp32f pMatrices[5*5*3];
// Set stride2
int stride2 = sizeof(Ipp32f)*2;
// Set stride1
int stride1 = sizeof(Ipp32f)*10;
// Set stride0
int stride0 = sizeof(Ipp32f)*25;
* * *
// Call IPP MX function
ippm<Operation>_ma_32f(..., pMatrices, stride0, stride1, stride2, ...);
The following code fragment represents description for a regular array of vectors shown in Figure “Standard
Description”, B:
// Allocate memory for the array of continuous vectors
Ipp32f pVectors[5*3];
// Set stride2
int stride2 = sizeof(Ipp32f)*2;
// Set stride0
int stride0 = sizeof(Ipp32f)*5;
* * *
// Call IPP MX function
ippm<Operation>_va_32f(...,pVectors, stride0, stride2, ...);
Single matrices and vectors are described in the same way without the stride 0 specification.
The following code fragment represents description for a regular array of constants shown in Figure
“Standard Description”, C:
// Allocate memory for the array of continuous constants
// example is for count=3
Ipp32f pVal[5*3];
// Set stride0
int valStride0 = sizeof(Ipp32f)*5;
* * *
// Call IPP MX function
ippm<Operation>_ca_32f(..., pVal, valStride0, ...);
2 Intel
®
Integrated Performance Primitives Reference Manual, Volume 3: Small Matrices and Realistic Rendering
16
剩余278页未读，继续阅读
