arm_copy_q7.c

00001 /* ----------------------------------------------------------------------
00002  * Project:      CMSIS DSP Library
00003  * Title:        arm_copy_q7.c
00004  * Description:  Copies the elements of a Q7 vector
00005  *
00006  * $Date:        27. January 2017
00007  * $Revision:    V.1.5.1
00008  *
00009  * Target Processor: Cortex-M cores
00010  * -------------------------------------------------------------------- */
00011 /*
00012  * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
00013  *
00014  * SPDX-License-Identifier: Apache-2.0
00015  *
00016  * Licensed under the Apache License, Version 2.0 (the License); you may
00017  * not use this file except in compliance with the License.
00018  * You may obtain a copy of the License at
00019  *
00020  * www.apache.org/licenses/LICENSE-2.0
00021  *
00022  * Unless required by applicable law or agreed to in writing, software
00023  * distributed under the License is distributed on an AS IS BASIS, WITHOUT
00024  * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
00025  * See the License for the specific language governing permissions and
00026  * limitations under the License.
00027  */
00028 
00029 #include "arm_math.h"
00030 
00031 /**
00032  * @ingroup groupSupport
00033  */
00034 
00035 /**
00036  * @addtogroup copy
00037  * @{
00038  */
00039 
00040 /**
00041  * @brief Copies the elements of a Q7 vector.
00042  * @param[in]       *pSrc points to input vector
00043  * @param[out]      *pDst points to output vector
00044  * @param[in]       blockSize length of the input vector
00045  * @return none.
00046  *
00047  */
00048 
00049 void arm_copy_q7(
00050   q7_t * pSrc,
00051   q7_t * pDst,
00052   uint32_t blockSize)
00053 {
00054   uint32_t blkCnt;                               /* loop counter */
00055 
00056 #if defined (ARM_MATH_DSP)
00057 
00058   /* Run the below code for Cortex-M4 and Cortex-M3 */
00059 
00060   /*loop Unrolling */
00061   blkCnt = blockSize >> 2U;
00062 
00063   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
00064    ** a second loop below computes the remaining 1 to 3 samples. */
00065   while (blkCnt > 0U)
00066   {
00067     /* C = A */
00068     /* Copy and then store the results in the destination buffer */
00069     /* 4 samples are copied and stored at a time using SIMD */
00070     *__SIMD32(pDst)++ = *__SIMD32(pSrc)++;
00071 
00072     /* Decrement the loop counter */
00073     blkCnt--;
00074   }
00075 
00076   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
00077    ** No loop unrolling is used. */
00078   blkCnt = blockSize % 0x4U;
00079 
00080 #else
00081 
00082   /* Run the below code for Cortex-M0 */
00083 
00084   /* Loop over blockSize number of values */
00085   blkCnt = blockSize;
00086 
00087 #endif /* #if defined (ARM_MATH_DSP) */
00088 
00089 
00090   while (blkCnt > 0U)
00091   {
00092     /* C = A */
00093     /* Copy and then store the results in the destination buffer */
00094     *pDst++ = *pSrc++;
00095 
00096     /* Decrement the loop counter */
00097     blkCnt--;
00098   }
00099 }
00100 
00101 /**
00102  * @} end of BasicCopy group
00103  */
00104