arm_sub_f32.c

00001 /* ----------------------------------------------------------------------
00002  * Project:      CMSIS DSP Library
00003  * Title:        arm_sub_f32.c
00004  * Description:  Floating-point vector subtraction.
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 groupMath
00033  */
00034 
00035 /**
00036  * @defgroup BasicSub Vector Subtraction
00037  *
00038  * Element-by-element subtraction of two vectors.
00039  *
00040  * <pre>
00041  *     pDst[n] = pSrcA[n] - pSrcB[n],   0 <= n < blockSize.
00042  * </pre>
00043  *
00044  * There are separate functions for floating-point, Q7, Q15, and Q31 data types.
00045  */
00046 
00047 /**
00048  * @addtogroup BasicSub
00049  * @{
00050  */
00051 
00052 
00053 /**
00054  * @brief Floating-point vector subtraction.
00055  * @param[in]       *pSrcA points to the first input vector
00056  * @param[in]       *pSrcB points to the second input vector
00057  * @param[out]      *pDst points to the output vector
00058  * @param[in]       blockSize number of samples in each vector
00059  * @return none.
00060  */
00061 
00062 void arm_sub_f32(
00063   float32_t * pSrcA,
00064   float32_t * pSrcB,
00065   float32_t * pDst,
00066   uint32_t blockSize)
00067 {
00068   uint32_t blkCnt;                               /* loop counter */
00069 
00070 #if defined (ARM_MATH_DSP)
00071 
00072 /* Run the below code for Cortex-M4 and Cortex-M3 */
00073   float32_t inA1, inA2, inA3, inA4;              /* temporary variables */
00074   float32_t inB1, inB2, inB3, inB4;              /* temporary variables */
00075 
00076   /*loop Unrolling */
00077   blkCnt = blockSize >> 2U;
00078 
00079   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
00080    ** a second loop below computes the remaining 1 to 3 samples. */
00081   while (blkCnt > 0U)
00082   {
00083     /* C = A - B */
00084     /* Subtract and then store the results in the destination buffer. */
00085     /* Read 4 input samples from sourceA and sourceB */
00086     inA1 = *pSrcA;
00087     inB1 = *pSrcB;
00088     inA2 = *(pSrcA + 1);
00089     inB2 = *(pSrcB + 1);
00090     inA3 = *(pSrcA + 2);
00091     inB3 = *(pSrcB + 2);
00092     inA4 = *(pSrcA + 3);
00093     inB4 = *(pSrcB + 3);
00094 
00095     /* dst = srcA - srcB */
00096     /* subtract and store the result */
00097     *pDst = inA1 - inB1;
00098     *(pDst + 1) = inA2 - inB2;
00099     *(pDst + 2) = inA3 - inB3;
00100     *(pDst + 3) = inA4 - inB4;
00101 
00102 
00103     /* Update pointers to process next sampels */
00104     pSrcA += 4U;
00105     pSrcB += 4U;
00106     pDst += 4U;
00107 
00108     /* Decrement the loop counter */
00109     blkCnt--;
00110   }
00111 
00112   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
00113    ** No loop unrolling is used. */
00114   blkCnt = blockSize % 0x4U;
00115 
00116 #else
00117 
00118   /* Run the below code for Cortex-M0 */
00119 
00120   /* Initialize blkCnt with number of samples */
00121   blkCnt = blockSize;
00122 
00123 #endif /* #if defined (ARM_MATH_DSP) */
00124 
00125   while (blkCnt > 0U)
00126   {
00127     /* C = A - B */
00128     /* Subtract and then store the results in the destination buffer. */
00129     *pDst++ = (*pSrcA++) - (*pSrcB++);
00130 
00131     /* Decrement the loop counter */
00132     blkCnt--;
00133   }
00134 }
00135 
00136 /**
00137  * @} end of BasicSub group
00138  */
00139