k60-keil

K60-Keil版本（下载安装MDK4.23）

        /* Read y[1] sample */
        c0 = *(py++);

        /* Read x[4] sample */
        x0 = *(px++);

        /* Perform the multiply-accumulates */
        /* acc0 +=  x[1] * y[1] */
        acc0 = (q31_t) ((((q63_t) acc0 << 32) + ((q63_t) x1 * c0)) >> 32);
        /* acc1 +=  x[2] * y[1] */
        acc1 = (q31_t) ((((q63_t) acc1 << 32) + ((q63_t) x2 * c0)) >> 32);
        /* acc2 +=  x[3] * y[1] */
        acc2 = (q31_t) ((((q63_t) acc2 << 32) + ((q63_t) x3 * c0)) >> 32);
        /* acc3 +=  x[4] * y[1] */
        acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x0 * c0)) >> 32);

        /* Read y[2] sample */
        c0 = *(py++);

        /* Read x[5] sample */
        x1 = *(px++);

        /* Perform the multiply-accumulates */
        /* acc0 +=  x[2] * y[2] */
        acc0 = (q31_t) ((((q63_t) acc0 << 32) + ((q63_t) x2 * c0)) >> 32);
        /* acc1 +=  x[3] * y[2] */
        acc1 = (q31_t) ((((q63_t) acc1 << 32) + ((q63_t) x3 * c0)) >> 32);
        /* acc2 +=  x[4] * y[2] */
        acc2 = (q31_t) ((((q63_t) acc2 << 32) + ((q63_t) x0 * c0)) >> 32);
        /* acc3 +=  x[5] * y[2] */
        acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x1 * c0)) >> 32);

        /* Read y[3] sample */
        c0 = *(py++);

        /* Read x[6] sample */
        x2 = *(px++);

        /* Perform the multiply-accumulates */
        /* acc0 +=  x[3] * y[3] */
        acc0 = (q31_t) ((((q63_t) acc0 << 32) + ((q63_t) x3 * c0)) >> 32);
        /* acc1 +=  x[4] * y[3] */
        acc1 = (q31_t) ((((q63_t) acc1 << 32) + ((q63_t) x0 * c0)) >> 32);
        /* acc2 +=  x[5] * y[3] */
        acc2 = (q31_t) ((((q63_t) acc2 << 32) + ((q63_t) x1 * c0)) >> 32);
        /* acc3 +=  x[6] * y[3] */
        acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x2 * c0)) >> 32);


      } while(--k);

      /* If the srcBLen is not a multiple of 4, compute any remaining MACs here.   
       ** No loop unrolling is used. */
      k = srcBLen % 0x4u;

      while(k > 0u)
      {
        /* Read y[4] sample */
        c0 = *(py++);

        /* Read x[7] sample */
        x3 = *(px++);

        /* Perform the multiply-accumulates */
        /* acc0 +=  x[4] * y[4] */
        acc0 = (q31_t) ((((q63_t) acc0 << 32) + ((q63_t) x0 * c0)) >> 32);
        /* acc1 +=  x[5] * y[4] */
        acc1 = (q31_t) ((((q63_t) acc1 << 32) + ((q63_t) x1 * c0)) >> 32);
        /* acc2 +=  x[6] * y[4] */
        acc2 = (q31_t) ((((q63_t) acc2 << 32) + ((q63_t) x2 * c0)) >> 32);
        /* acc3 +=  x[7] * y[4] */
        acc3 = (q31_t) ((((q63_t) acc3 << 32) + ((q63_t) x3 * c0)) >> 32);

        /* Reuse the present samples for the next MAC */
        x0 = x1;
        x1 = x2;
        x2 = x3;

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

      /* Store the result in the accumulator in the destination buffer. */
      *pOut = (q31_t) (acc0 << 1);
      /* Destination pointer is updated according to the address modifier, inc */
      pOut += inc;

      *pOut = (q31_t) (acc1 << 1);
      pOut += inc;

      *pOut = (q31_t) (acc2 << 1);
      pOut += inc;

      *pOut = (q31_t) (acc3 << 1);
      pOut += inc;

      /* Update the inputA and inputB pointers for next MAC calculation */
      px = pIn1 + (count * 4u);
      py = pIn2;

      /* Increment the pointer pIn1 index, count by 1 */
      count++;

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

    /* If the blockSize2 is not a multiple of 4, compute any remaining output samples here.   
     ** No loop unrolling is used. */
    blkCnt = blockSize2 % 0x4u;

    while(blkCnt > 0u)
    {
      /* Accumulator is made zero for every iteration */
      sum = 0;

      /* Apply loop unrolling and compute 4 MACs simultaneously. */
      k = srcBLen >> 2u;

      /* First part of the processing with loop unrolling.  Compute 4 MACs at a time.   
       ** a second loop below computes MACs for the remaining 1 to 3 samples. */
      while(k > 0u)
      {
        /* Perform the multiply-accumulates */
        sum = (q31_t) ((((q63_t) sum << 32) +
                        ((q63_t) * px++ * (*py++))) >> 32);
        sum = (q31_t) ((((q63_t) sum << 32) +
                        ((q63_t) * px++ * (*py++))) >> 32);
        sum = (q31_t) ((((q63_t) sum << 32) +
                        ((q63_t) * px++ * (*py++))) >> 32);
        sum = (q31_t) ((((q63_t) sum << 32) +
                        ((q63_t) * px++ * (*py++))) >> 32);

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

      /* If the srcBLen is not a multiple of 4, compute any remaining MACs here.   
       ** No loop unrolling is used. */
      k = srcBLen % 0x4u;

      while(k > 0u)
      {
        /* Perform the multiply-accumulate */
        sum = (q31_t) ((((q63_t) sum << 32) +
                        ((q63_t) * px++ * (*py++))) >> 32);

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

      /* Store the result in the accumulator in the destination buffer. */
      *pOut = sum << 1;
      /* Destination pointer is updated according to the address modifier, inc */
      pOut += inc;

      /* Update the inputA and inputB pointers for next MAC calculation */
      px = pIn1 + count;
      py = pIn2;

      /* Increment the MAC count */
      count++;

      /* Decrement the loop counter */
      blkCnt--;
    }
  }
  else
  {
    /* If the srcBLen is not a multiple of 4,   
     * the blockSize2 loop cannot be unrolled by 4 */
    blkCnt = blockSize2;

    while(blkCnt > 0u)
    {
      /* Accumulator is made zero for every iteration */
      sum = 0;

      /* Loop over srcBLen */
      k = srcBLen;

      while(k > 0u)
      {
        /* Perform the multiply-accumulate */
        sum = (q31_t) ((((q63_t) sum << 32) +
                        ((q63_t) * px++ * (*py++))) >> 32);

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

      /* Store the result in the accumulator in the destination buffer. */
      *pOut = sum << 1;
      /* Destination pointer is updated according to the address modifier, inc */
      pOut += inc;

      /* Update the inputA and inputB pointers for next MAC calculation */
      px = pIn1 + count;
      py = pIn2;

      /* Increment the MAC count */
      count++;

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

  /* --------------------------   
   * Initializations of stage3   
   * -------------------------*/

  /* sum += x[srcALen-srcBLen+1] * y[0] + x[srcALen-srcBLen+2] * y[1] +...+ x[srcALen-1] * y[srcBLen-1]   
   * sum += x[srcALen-srcBLen+2] * y[0] + x[srcALen-srcBLen+3] * y[1] +...+ x[srcALen-1] * y[srcBLen-1]   
   * ....   
   * sum +=  x[srcALen-2] * y[0] + x[srcALen-1] * y[1]   
   * sum +=  x[srcALen-1] * y[0]   
   */

  /* In this stage the MAC operations are decreased by 1 for every iteration.   
     The count variable holds the number of MAC operations performed */
  count = srcBLen - 1u;

  /* Working pointer of inputA */
  pSrc1 = ((pIn1 + srcALen) - srcBLen) + 1u;
  px = pSrc1;

  /* Working pointer of inputB */
  py = pIn2;

  /* -------------------   
   * Stage3 process   
   * ------------------*/

  while(blockSize3 > 0u)
  {
    /* Accumulator is made zero for every iteration */
    sum = 0;

    /* Apply loop unrolling and compute 4 MACs simultaneously. */
    k = count >> 2u;

    /* First part of the processing with loop unrolling.  Compute 4 MACs at a time.   
     ** a second loop below computes MACs for the remaining 1 to 3 samples. */
    while(k > 0u)
    {
      /* Perform the multiply-accumulates */
      /* sum += x[srcALen - srcBLen + 4] * y[3] */
      sum = (q31_t) ((((q63_t) sum << 32) +
                      ((q63_t) * px++ * (*py++))) >> 32);
      /* sum += x[srcALen - srcBLen + 3] * y[2] */
      sum = (q31_t) ((((q63_t) sum << 32) +
                      ((q63_t) * px++ * (*py++))) >> 32);
      /* sum += x[srcALen - srcBLen + 2] * y[1] */
      sum = (q31_t) ((((q63_t) sum << 32) +
                      ((q63_t) * px++ * (*py++))) >> 32);
      /* sum += x[srcALen - srcBLen + 1] * y[0] */
      sum = (q31_t) ((((q63_t) sum << 32) +
                      ((q63_t) * px++ * (*py++))) >> 32);

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

    /* If the count is not a multiple of 4, compute any remaining MACs here.   
     ** No loop unrolling is used. */
    k = count % 0x4u;

    while(k > 0u)
    {
      /* Perform the multiply-accumulates */
      sum = (q31_t) ((((q63_t) sum << 32) +
                      ((q63_t) * px++ * (*py++))) >> 32);

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

    /* Store the result in the accumulator in the destination buffer. */
    *pOut = sum << 1;
    /* Destination pointer is updated according to the address modifier, inc */
    pOut += inc;

    /* Update the inputA and inputB pointers for next MAC calculation */
    px = ++pSrc1;
    py = pIn2;

    /* Decrement the MAC count */
    count--;

    /* Decrement the loop counter */
    blockSize3--;
  }

}

/**   
 * @} end of Corr group   
 */