arm_fir_sparse_f32.c

Go to the documentation of this file.

/* ----------------------------------------------------------------------    
* Copyright (C) 2010-2014 ARM Limited. All rights reserved.    
*    
* $Date:        19. March 2015
* $Revision:    V.1.4.5
*    
* Project:      CMSIS DSP Library    
* Title:        arm_fir_sparse_f32.c    
*    
* Description:  Floating-point sparse FIR filter processing function.   
*    
* Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
*  
* Redistribution and use in source and binary forms, with or without 
* modification, are permitted provided that the following conditions
* are met:
*   - Redistributions of source code must retain the above copyright
*     notice, this list of conditions and the following disclaimer.
*   - Redistributions in binary form must reproduce the above copyright
*     notice, this list of conditions and the following disclaimer in
*     the documentation and/or other materials provided with the 
*     distribution.
*   - Neither the name of ARM LIMITED nor the names of its contributors
*     may be used to endorse or promote products derived from this
*     software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE 
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.    
* ------------------------------------------------------------------- */
#include "arm_math.h"

void arm_fir_sparse_f32(
  arm_fir_sparse_instance_f32 * S,
  float32_t * pSrc,
  float32_t * pDst,
  float32_t * pScratchIn,
  uint32_t blockSize)
{

  float32_t *pState = S->pState;                 /* State pointer */
  float32_t *pCoeffs = S->pCoeffs;               /* Coefficient pointer */
  float32_t *px;                                 /* Scratch buffer pointer */
  float32_t *py = pState;                        /* Temporary pointers for state buffer */
  float32_t *pb = pScratchIn;                    /* Temporary pointers for scratch buffer */
  float32_t *pOut;                               /* Destination pointer */
  int32_t *pTapDelay = S->pTapDelay;             /* Pointer to the array containing offset of the non-zero tap values. */
  uint32_t delaySize = S->maxDelay + blockSize;  /* state length */
  uint16_t numTaps = S->numTaps;                 /* Number of filter coefficients in the filter  */
  int32_t readIndex;                             /* Read index of the state buffer */
  uint32_t tapCnt, blkCnt;                       /* loop counters */
  float32_t coeff = *pCoeffs++;                  /* Read the first coefficient value */



  /* BlockSize of Input samples are copied into the state buffer */
  /* StateIndex points to the starting position to write in the state buffer */
  arm_circularWrite_f32((int32_t *) py, delaySize, &S->stateIndex, 1,
                        (int32_t *) pSrc, 1, blockSize);


  /* Read Index, from where the state buffer should be read, is calculated. */
  readIndex = ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;

  /* Wraparound of readIndex */
  if(readIndex < 0)
  {
    readIndex += (int32_t) delaySize;
  }

  /* Working pointer for state buffer is updated */
  py = pState;

  /* blockSize samples are read from the state buffer */
  arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
                       (int32_t *) pb, (int32_t *) pb, blockSize, 1,
                       blockSize);

  /* Working pointer for the scratch buffer */
  px = pb;

  /* Working pointer for destination buffer */
  pOut = pDst;


#ifndef ARM_MATH_CM0_FAMILY

  /* Run the below code for Cortex-M4 and Cortex-M3 */

  /* Loop over the blockSize. Unroll by a factor of 4.    
   * Compute 4 Multiplications at a time. */
  blkCnt = blockSize >> 2u;

  while(blkCnt > 0u)
  {
    /* Perform Multiplications and store in destination buffer */
    *pOut++ = *px++ * coeff;
    *pOut++ = *px++ * coeff;
    *pOut++ = *px++ * coeff;
    *pOut++ = *px++ * coeff;

    /* Decrement the loop counter */
    blkCnt--;
  }

  /* If the blockSize is not a multiple of 4,    
   * compute the remaining samples */
  blkCnt = blockSize % 0x4u;

  while(blkCnt > 0u)
  {
    /* Perform Multiplications and store in destination buffer */
    *pOut++ = *px++ * coeff;

    /* Decrement the loop counter */
    blkCnt--;
  }

  /* Load the coefficient value and    
   * increment the coefficient buffer for the next set of state values */
  coeff = *pCoeffs++;

  /* Read Index, from where the state buffer should be read, is calculated. */
  readIndex = ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;

  /* Wraparound of readIndex */
  if(readIndex < 0)
  {
    readIndex += (int32_t) delaySize;
  }

  /* Loop over the number of taps. */
  tapCnt = (uint32_t) numTaps - 2u;

  while(tapCnt > 0u)
  {

    /* Working pointer for state buffer is updated */
    py = pState;

    /* blockSize samples are read from the state buffer */
    arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
                         (int32_t *) pb, (int32_t *) pb, blockSize, 1,
                         blockSize);

    /* Working pointer for the scratch buffer */
    px = pb;

    /* Working pointer for destination buffer */
    pOut = pDst;

    /* Loop over the blockSize. Unroll by a factor of 4.    
     * Compute 4 MACS at a time. */
    blkCnt = blockSize >> 2u;

    while(blkCnt > 0u)
    {
      /* Perform Multiply-Accumulate */
      *pOut++ += *px++ * coeff;
      *pOut++ += *px++ * coeff;
      *pOut++ += *px++ * coeff;
      *pOut++ += *px++ * coeff;

      /* Decrement the loop counter */
      blkCnt--;
    }

    /* If the blockSize is not a multiple of 4,    
     * compute the remaining samples */
    blkCnt = blockSize % 0x4u;

    while(blkCnt > 0u)
    {
      /* Perform Multiply-Accumulate */
      *pOut++ += *px++ * coeff;

      /* Decrement the loop counter */
      blkCnt--;
    }

    /* Load the coefficient value and    
     * increment the coefficient buffer for the next set of state values */
    coeff = *pCoeffs++;

    /* Read Index, from where the state buffer should be read, is calculated. */
    readIndex = ((int32_t) S->stateIndex -
                 (int32_t) blockSize) - *pTapDelay++;

    /* Wraparound of readIndex */
    if(readIndex < 0)
    {
      readIndex += (int32_t) delaySize;
    }

    /* Decrement the tap loop counter */
    tapCnt--;
  }
    
    /* Compute last tap without the final read of pTapDelay */

    /* Working pointer for state buffer is updated */
    py = pState;

    /* blockSize samples are read from the state buffer */
    arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
                                             (int32_t *) pb, (int32_t *) pb, blockSize, 1,
                                             blockSize);

    /* Working pointer for the scratch buffer */
    px = pb;

    /* Working pointer for destination buffer */
    pOut = pDst;

    /* Loop over the blockSize. Unroll by a factor of 4.    
     * Compute 4 MACS at a time. */
    blkCnt = blockSize >> 2u;

    while(blkCnt > 0u)
    {
        /* Perform Multiply-Accumulate */
        *pOut++ += *px++ * coeff;
        *pOut++ += *px++ * coeff;
        *pOut++ += *px++ * coeff;
        *pOut++ += *px++ * coeff;

        /* Decrement the loop counter */
        blkCnt--;
    }

    /* If the blockSize is not a multiple of 4,    
     * compute the remaining samples */
    blkCnt = blockSize % 0x4u;

    while(blkCnt > 0u)
    {
        /* Perform Multiply-Accumulate */
        *pOut++ += *px++ * coeff;

        /* Decrement the loop counter */
        blkCnt--;
    }

#else

/* Run the below code for Cortex-M0 */

  blkCnt = blockSize;

  while(blkCnt > 0u)
  {
    /* Perform Multiplications and store in destination buffer */
    *pOut++ = *px++ * coeff;

    /* Decrement the loop counter */
    blkCnt--;
  }

  /* Load the coefficient value and           
   * increment the coefficient buffer for the next set of state values */
  coeff = *pCoeffs++;

  /* Read Index, from where the state buffer should be read, is calculated. */
  readIndex = ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;

  /* Wraparound of readIndex */
  if(readIndex < 0)
  {
    readIndex += (int32_t) delaySize;
  }

  /* Loop over the number of taps. */
  tapCnt = (uint32_t) numTaps - 2u;

  while(tapCnt > 0u)
  {

    /* Working pointer for state buffer is updated */
    py = pState;

    /* blockSize samples are read from the state buffer */
    arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
                         (int32_t *) pb, (int32_t *) pb, blockSize, 1,
                         blockSize);

    /* Working pointer for the scratch buffer */
    px = pb;

    /* Working pointer for destination buffer */
    pOut = pDst;

    blkCnt = blockSize;

    while(blkCnt > 0u)
    {
      /* Perform Multiply-Accumulate */
      *pOut++ += *px++ * coeff;

      /* Decrement the loop counter */
      blkCnt--;
    }

    /* Load the coefficient value and           
     * increment the coefficient buffer for the next set of state values */
    coeff = *pCoeffs++;

    /* Read Index, from where the state buffer should be read, is calculated. */
    readIndex =
      ((int32_t) S->stateIndex - (int32_t) blockSize) - *pTapDelay++;

    /* Wraparound of readIndex */
    if(readIndex < 0)
    {
      readIndex += (int32_t) delaySize;
    }

    /* Decrement the tap loop counter */
    tapCnt--;
  }
    
    /* Compute last tap without the final read of pTapDelay */  
    
    /* Working pointer for state buffer is updated */
    py = pState;

    /* blockSize samples are read from the state buffer */
    arm_circularRead_f32((int32_t *) py, delaySize, &readIndex, 1,
                                             (int32_t *) pb, (int32_t *) pb, blockSize, 1,
                                             blockSize);

    /* Working pointer for the scratch buffer */
    px = pb;

    /* Working pointer for destination buffer */
    pOut = pDst;

    blkCnt = blockSize;

    while(blkCnt > 0u)
    {
        /* Perform Multiply-Accumulate */
        *pOut++ += *px++ * coeff;

        /* Decrement the loop counter */
        blkCnt--;
    }

#endif /*   #ifndef ARM_MATH_CM0_FAMILY        */

}