lib/main/CMSIS/DSP/Source/BasicMathFunctions/arm_sub_q15.c

   1 /* ----------------------------------------------------------------------
   2  * Project:      CMSIS DSP Library
   3  * Title:        arm_sub_q15.c
   4  * Description:  Q15 vector subtraction
   5  *
   6  * $Date:        27. January 2017
   7  * $Revision:    V.1.5.1
   8  *
   9  * Target Processor: Cortex-M cores
  10  * -------------------------------------------------------------------- */
  11 /*
  12  * Copyright (C) 2010-2017 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 "arm_math.h"
  30
  31 /**
  32  * @ingroup groupMath
  33  */
  34
  35 /**
  36  * @addtogroup BasicSub
  37  * @{
  38  */
  39
  40 /**
  41  * @brief Q15 vector subtraction.
  42  * @param[in]       *pSrcA points to the first input vector
  43  * @param[in]       *pSrcB points to the second input vector
  44  * @param[out]      *pDst points to the output vector
  45  * @param[in]       blockSize number of samples in each vector
  46  * @return none.
  47  *
  48  * <b>Scaling and Overflow Behavior:</b>
  49  * \par
  50  * The function uses saturating arithmetic.
  51  * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.
  52  */
  53
  54 void arm_sub_q15(
  55   q15_t * pSrcA,
  56   q15_t * pSrcB,
  57   q15_t * pDst,
  58   uint32_t blockSize)
  59 {
  60   uint32_t blkCnt;                               /* loop counter */
  61
  62
  63 #if defined (ARM_MATH_DSP)
  64
  65 /* Run the below code for Cortex-M4 and Cortex-M3 */
  66   q31_t inA1, inA2;
  67   q31_t inB1, inB2;
  68
  69   /*loop Unrolling */
  70   blkCnt = blockSize >> 2U;
  71
  72   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
  73    ** a second loop below computes the remaining 1 to 3 samples. */
  74   while (blkCnt > 0U)
  75   {
  76     /* C = A - B */
  77     /* Subtract and then store the results in the destination buffer two samples at a time. */
  78     inA1 = *__SIMD32(pSrcA)++;
  79     inA2 = *__SIMD32(pSrcA)++;
  80     inB1 = *__SIMD32(pSrcB)++;
  81     inB2 = *__SIMD32(pSrcB)++;
  82
  83     *__SIMD32(pDst)++ = __QSUB16(inA1, inB1);
  84     *__SIMD32(pDst)++ = __QSUB16(inA2, inB2);
  85
  86     /* Decrement the loop counter */
  87     blkCnt--;
  88   }
  89
  90   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
  91    ** No loop unrolling is used. */
  92   blkCnt = blockSize % 0x4U;
  93
  94   while (blkCnt > 0U)
  95   {
  96     /* C = A - B */
  97     /* Subtract and then store the result in the destination buffer. */
  98     *pDst++ = (q15_t) __QSUB16(*pSrcA++, *pSrcB++);
  99
 100     /* Decrement the loop counter */
 101     blkCnt--;
 102   }
 103
 104 #else
 105
 106   /* Run the below code for Cortex-M0 */
 107
 108   /* Initialize blkCnt with number of samples */
 109   blkCnt = blockSize;
 110
 111   while (blkCnt > 0U)
 112   {
 113     /* C = A - B */
 114     /* Subtract and then store the result in the destination buffer. */
 115     *pDst++ = (q15_t) __SSAT(((q31_t) * pSrcA++ - *pSrcB++), 16);
 116
 117     /* Decrement the loop counter */
 118     blkCnt--;
 119   }
 120
 121 #endif /* #if defined (ARM_MATH_DSP) */
 122
 123
 124 }
 125
 126 /**
 127  * @} end of BasicSub group
 128  */