/ Drivers / CMSIS / DSP / Source / BasicMathFunctions / arm_offset_q7.c
arm_offset_q7.c
  1  /* ----------------------------------------------------------------------
  2   * Project:      CMSIS DSP Library
  3   * Title:        arm_offset_q7.c
  4   * Description:  Q7 vector offset
  5   *
  6   * $Date:        23 April 2021
  7   * $Revision:    V1.9.0
  8   *
  9   * Target Processor: Cortex-M and Cortex-A cores
 10   * -------------------------------------------------------------------- */
 11  /*
 12   * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
 13   *
 14   * SPDX-License-Identifier: Apache-2.0
 15   *
 16   * Licensed under the Apache License, Version 2.0 (the License); you may
 17   * not use this file except in compliance with the License.
 18   * You may obtain a copy of the License at
 19   *
 20   * www.apache.org/licenses/LICENSE-2.0
 21   *
 22   * Unless required by applicable law or agreed to in writing, software
 23   * distributed under the License is distributed on an AS IS BASIS, WITHOUT
 24   * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 25   * See the License for the specific language governing permissions and
 26   * limitations under the License.
 27   */
 28  
 29  #include "dsp/basic_math_functions.h"
 30  
 31  /**
 32    @ingroup groupMath
 33   */
 34  
 35  /**
 36    @addtogroup BasicOffset
 37    @{
 38   */
 39  
 40  /**
 41    @brief         Adds a constant offset to a Q7 vector.
 42    @param[in]     pSrc       points to the input vector
 43    @param[in]     offset     is the offset to be added
 44    @param[out]    pDst       points to the output vector
 45    @param[in]     blockSize  number of samples in each vector
 46    @return        none
 47  
 48    @par           Scaling and Overflow Behavior
 49                     The function uses saturating arithmetic.
 50                     Results outside of the allowable Q7 range [0x80 0x7F] are saturated.
 51   */
 52  #if defined(ARM_MATH_MVEI) && !defined(ARM_MATH_AUTOVECTORIZE)
 53  
 54  #include "arm_helium_utils.h"
 55  
 56  void arm_offset_q7(
 57      const q7_t * pSrc,
 58      q7_t   offset,
 59      q7_t * pDst,
 60      uint32_t blockSize)
 61  {
 62      uint32_t  blkCnt;           /* loop counters */
 63      q7x16_t vecSrc;
 64  
 65      /* Compute 16 outputs at a time */
 66      blkCnt = blockSize >> 4;
 67      while (blkCnt > 0U)
 68      {
 69          /*
 70           * C = A + offset
 71           * Add offset and then store the result in the destination buffer.
 72           */
 73          vecSrc = vld1q(pSrc);
 74          vst1q(pDst, vqaddq(vecSrc, offset));
 75          /*
 76           * Decrement the blockSize loop counter
 77           */
 78          blkCnt--;
 79          /*
 80           * advance vector source and destination pointers
 81           */
 82          pSrc += 16;
 83          pDst += 16;
 84      }
 85      /*
 86       * tail
 87       */
 88      blkCnt = blockSize & 0xF;
 89      if (blkCnt > 0U)
 90      {
 91          mve_pred16_t p0 = vctp8q(blkCnt);
 92          vecSrc = vld1q(pSrc);
 93          vstrbq_p(pDst, vqaddq(vecSrc, offset), p0);
 94      }
 95  }
 96  
 97  #else
 98  void arm_offset_q7(
 99    const q7_t * pSrc,
100          q7_t offset,
101          q7_t * pDst,
102          uint32_t blockSize)
103  {
104          uint32_t blkCnt;                               /* Loop counter */
105  
106  #if defined (ARM_MATH_LOOPUNROLL)
107  
108  #if defined (ARM_MATH_DSP)
109    q31_t offset_packed;                           /* Offset packed to 32 bit */
110  
111    /* Offset is packed to 32 bit in order to use SIMD32 for addition */
112    offset_packed = __PACKq7(offset, offset, offset, offset);
113  #endif
114  
115    /* Loop unrolling: Compute 4 outputs at a time */
116    blkCnt = blockSize >> 2U;
117  
118    while (blkCnt > 0U)
119    {
120      /* C = A + offset */
121  
122  #if defined (ARM_MATH_DSP)
123      /* Add offset and store result in destination buffer (4 samples at a time). */
124      write_q7x4_ia (&pDst, __QADD8(read_q7x4_ia (&pSrc), offset_packed));
125  #else
126      *pDst++ = (q7_t) __SSAT((q15_t) *pSrc++ + offset, 8);
127      *pDst++ = (q7_t) __SSAT((q15_t) *pSrc++ + offset, 8);
128      *pDst++ = (q7_t) __SSAT((q15_t) *pSrc++ + offset, 8);
129      *pDst++ = (q7_t) __SSAT((q15_t) *pSrc++ + offset, 8);
130  #endif
131  
132      /* Decrement loop counter */
133      blkCnt--;
134    }
135  
136    /* Loop unrolling: Compute remaining outputs */
137    blkCnt = blockSize % 0x4U;
138  
139  #else
140  
141    /* Initialize blkCnt with number of samples */
142    blkCnt = blockSize;
143  
144  #endif /* #if defined (ARM_MATH_LOOPUNROLL) */
145  
146    while (blkCnt > 0U)
147    {
148      /* C = A + offset */
149  
150      /* Add offset and store result in destination buffer. */
151      *pDst++ = (q7_t) __SSAT((q15_t) *pSrc++ + offset, 8);
152  
153      /* Decrement loop counter */
154      blkCnt--;
155    }
156  
157  }
158  #endif /* defined(ARM_MATH_MVEI) */
159  
160  /**
161    @} end of BasicOffset group
162   */