/*******************************************************************************
* Copyright 2015-2018 Intel Corporation.
*
* This software and the related documents are Intel copyrighted  materials,  and
* your use of  them is  governed by the  express license  under which  they were
* provided to you (License).  Unless the License provides otherwise, you may not
* use, modify, copy, publish, distribute,  disclose or transmit this software or
* the related documents without Intel's prior written permission.
*
* This software and the related documents  are provided as  is,  with no express
* or implied  warranties,  other  than those  that are  expressly stated  in the
* License.
*******************************************************************************/

//   An example of performing The code example shows how these functions can be used to organize the separable convolution as a step of
// image processing pipeline.implemented with Intel(R) Integrated Primitives (Intel(R) IPP) functions :
//     ippiFilterRowBorderPipelineGetBufferSize_Low_16s_C1R
//     ippiFilterColumnPipeline_Low_16s_C1R
//     ippiFilterColumnPipelineGetBufferSize_Low_16s_C1R
//     ippiFilterRowBorderPipeline_Low_16s_C1R


#include <stdio.h>
#include "ipp.h"

#define WIDTH  32  /* image width  */
#define HEIGHT 16  /* image height */
#define kernelSize 5

/* Next two defines are created to simplify code reading and understanding */
#define EXIT_MAIN exitLine:                                  /* Label for Exit */
#define check_sts(st) if((st) != ippStsNoErr) goto exitLine; /* Go to Exit if Intel(R) IPP function returned status different from ippStsNoErr */

/* Results of ippMalloc() are not validated because Intel(R) IPP functions perform bad arguments check and will return an appropriate status  */

int main(void)
{
    IppStatus status = ippStsNoErr;
    const Ipp16s **pGet = NULL;
    Ipp16s* src = NULL, *dst = NULL;
    int xAnchor = kernelSize >> 1; /* The anchor value, (0 <= xAnchor < kernelSize) */
    Ipp16s pKerX[kernelSize] = { 1, 2, 3, 2, 1 };
    Ipp16s pKerY[kernelSize] = { 1, 2, 0, 2, 1 }; /* The pointer to the kernels */
    Ipp16s* pSrc = NULL, *pDst = NULL; /* Pointers to source/destination images */
    int srcStep = 0, dstStep = 0;      /* Steps, in bytes, through the source/destination images */
    IppiSize roiSize = { WIDTH, HEIGHT }; /* Size of source/destination ROI in pixels */
    IppiSize roi = { WIDTH, 1 }; /* Size of destination ROI in pixels */
    int divisor = 1; /* The value to divide output pixels by */
    Ipp8u borderValue = 0;
    Ipp8u *pBufRow = NULL, *pBufCol = NULL;/* Pointer to the work buffer */
    int sizeRow = 0, sizeCol = 0;    /* Common work buffer size */
    int todo = roiSize.height, bufLen;
    int mStep = (roiSize.width + 7)&(~7);
    int sStep = 0, dStep = 0;
    int i, j, n;
    Ipp16s* pCycleBuf;
    pGet      = (const Ipp16s**)ippsMalloc_8u(kernelSize*sizeof(Ipp16s*));
    pCycleBuf = (const Ipp16s**)ippsMalloc_8u(mStep * kernelSize *sizeof(Ipp16s));
    
    pSrc = ippiMalloc_16s_C1(roiSize.width, roiSize.height, &srcStep);
    pDst = ippiMalloc_16s_C1(roiSize.width, roiSize.height, &dstStep);

    dst = pDst; src = pSrc;
    sStep = srcStep >> 1, dStep = dstStep >> 1;
    for (n = 0; n < kernelSize; n++)
    {
        pGet[n] = (Ipp16s*)pCycleBuf + n * mStep;
    }
    for (i = 0; i < roiSize.height; i++)
    {
        for (j = 0; j < roiSize.width; j++)
        {
            *(pSrc + i*sStep + j) = i;
        }
    }
    check_sts( status = ippiFilterRowBorderPipelineGetBufferSize_Low_16s_C1R(roiSize, 3, &sizeRow) )
    check_sts( status = ippiFilterColumnPipelineGetBufferSize_Low_16s_C1R(roiSize, 3, &sizeCol) )

    pBufRow = ippsMalloc_8u(sizeRow);
    pBufCol = ippsMalloc_8u(sizeCol);

    src = pSrc;
    for (j = 0; j < kernelSize - 1; j++){
        check_sts(status = ippiFilterRowBorderPipeline_Low_16s_C1R(src, srcStep, &pGet[j], roi, pKerX, kernelSize, xAnchor, ippBorderRepl, borderValue, divisor, pBufRow))
        src += sStep;
    }
    for (j = 0; j < roiSize.height; j++){
        check_sts( status = ippiFilterRowBorderPipeline_Low_16s_C1R(src, srcStep, &pGet[kernelSize-1], roi, pKerX, kernelSize, xAnchor, ippBorderRepl, borderValue, divisor, pBufRow) )
        check_sts( status = ippiFilterColumnPipeline_Low_16s_C1R(pGet, dst, dstStep, roi, pKerY, kernelSize, divisor, pBufCol) )
        src += sStep;
        dst += dStep;
        {
            Ipp16s* tGet= pGet[0];
            for (n = 0; n < kernelSize - 1; n++){
                pGet[n] = pGet[n + 1];
            }
            pGet[kernelSize - 1] = tGet;
        }

    }

    EXIT_MAIN
    ippsFree(pGet);
    ippsFree(pCycleBuf);
    ippiFree(pSrc);
    ippsFree(pBufRow);
    ippsFree(pBufCol);
    ippiFree(pDst);
    printf("Exit status %d (%s)\n", (int)status, ippGetStatusString(status));
    return (int)status;
}