lib/main/CMSIS/DSP/Source/BasicMathFunctions/arm_mult_q7.c

   1 /* ----------------------------------------------------------------------
   2  * Project:      CMSIS DSP Library
   3  * Title:        arm_mult_q7.c
   4  * Description:  Q7 vector multiplication
   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 BasicMult
  37  * @{
  38  */
  39
  40 /**
  41  * @brief           Q7 vector multiplication
  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 Q7 range [0x80 0x7F] will be saturated.
  52  */
  53
  54 void arm_mult_q7(
  55   q7_t * pSrcA,
  56   q7_t * pSrcB,
  57   q7_t * pDst,
  58   uint32_t blockSize)
  59 {
  60   uint32_t blkCnt;                               /* loop counters */
  61
  62 #if defined (ARM_MATH_DSP)
  63
  64 /* Run the below code for Cortex-M4 and Cortex-M3 */
  65   q7_t out1, out2, out3, out4;                   /* Temporary variables to store the product */
  66
  67   /* loop Unrolling */
  68   blkCnt = blockSize >> 2U;
  69
  70   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
  71    ** a second loop below computes the remaining 1 to 3 samples. */
  72   while (blkCnt > 0U)
  73   {
  74     /* C = A * B */
  75     /* Multiply the inputs and store the results in temporary variables */
  76     out1 = (q7_t) __SSAT((((q15_t) (*pSrcA++) * (*pSrcB++)) >> 7), 8);
  77     out2 = (q7_t) __SSAT((((q15_t) (*pSrcA++) * (*pSrcB++)) >> 7), 8);
  78     out3 = (q7_t) __SSAT((((q15_t) (*pSrcA++) * (*pSrcB++)) >> 7), 8);
  79     out4 = (q7_t) __SSAT((((q15_t) (*pSrcA++) * (*pSrcB++)) >> 7), 8);
  80
  81     /* Store the results of 4 inputs in the destination buffer in single cycle by packing */
  82     *__SIMD32(pDst)++ = __PACKq7(out1, out2, out3, out4);
  83
  84     /* Decrement the blockSize loop counter */
  85     blkCnt--;
  86   }
  87
  88   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
  89    ** No loop unrolling is used. */
  90   blkCnt = blockSize % 0x4U;
  91
  92 #else
  93
  94   /* Run the below code for Cortex-M0 */
  95
  96   /* Initialize blkCnt with number of samples */
  97   blkCnt = blockSize;
  98
  99 #endif /* #if defined (ARM_MATH_DSP) */
 100
 101
 102   while (blkCnt > 0U)
 103   {
 104     /* C = A * B */
 105     /* Multiply the inputs and store the result in the destination buffer */
 106     *pDst++ = (q7_t) __SSAT((((q15_t) (*pSrcA++) * (*pSrcB++)) >> 7), 8);
 107
 108     /* Decrement the blockSize loop counter */
 109     blkCnt--;
 110   }
 111 }
 112
 113 /**
 114  * @} end of BasicMult group
 115  */