1 /* ----------------------------------------------------------------------
2 * Project: CMSIS DSP Library
3 * Title: arm_mult_f32.c
4 * Description: Floating-point vector multiplication
6 * $Date: 27. January 2017
9 * Target Processor: Cortex-M cores
10 * -------------------------------------------------------------------- */
12 * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
14 * SPDX-License-Identifier: Apache-2.0
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
20 * www.apache.org/licenses/LICENSE-2.0
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.
36 * @defgroup BasicMult Vector Multiplication
38 * Element-by-element multiplication of two vectors.
41 * pDst[n] = pSrcA[n] * pSrcB[n], 0 <= n < blockSize.
44 * There are separate functions for floating-point, Q7, Q15, and Q31 data types.
48 * @addtogroup BasicMult
53 * @brief Floating-point vector multiplication.
54 * @param[in] *pSrcA points to the first input vector
55 * @param[in] *pSrcB points to the second input vector
56 * @param[out] *pDst points to the output vector
57 * @param[in] blockSize number of samples in each vector
67 uint32_t blkCnt
; /* loop counters */
68 #if defined (ARM_MATH_DSP)
70 /* Run the below code for Cortex-M4 and Cortex-M3 */
71 float32_t inA1
, inA2
, inA3
, inA4
; /* temporary input variables */
72 float32_t inB1
, inB2
, inB3
, inB4
; /* temporary input variables */
73 float32_t out1
, out2
, out3
, out4
; /* temporary output variables */
76 blkCnt
= blockSize
>> 2U;
78 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
79 ** a second loop below computes the remaining 1 to 3 samples. */
83 /* Multiply the inputs and store the results in output buffer */
84 /* read sample from sourceA */
86 /* read sample from sourceB */
88 /* read sample from sourceA */
90 /* read sample from sourceB */
93 /* out = sourceA * sourceB */
96 /* read sample from sourceA */
98 /* read sample from sourceB */
101 /* out = sourceA * sourceB */
104 /* read sample from sourceA */
107 /* store result to destination buffer */
110 /* read sample from sourceB */
113 /* out = sourceA * sourceB */
116 /* store result to destination buffer */
119 /* out = sourceA * sourceB */
121 /* store result to destination buffer */
123 /* store result to destination buffer */
127 /* update pointers to process next samples */
132 /* Decrement the blockSize loop counter */
136 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
137 ** No loop unrolling is used. */
138 blkCnt
= blockSize
% 0x4U
;
142 /* Run the below code for Cortex-M0 */
144 /* Initialize blkCnt with number of samples */
147 #endif /* #if defined (ARM_MATH_DSP) */
152 /* Multiply the inputs and store the results in output buffer */
153 *pDst
++ = (*pSrcA
++) * (*pSrcB
++);
155 /* Decrement the blockSize loop counter */
161 * @} end of BasicMult group