|
2
|
1 /* ----------------------------------------------------------------------
|
|
|
2 * Copyright (C) 2010-2014 ARM Limited. All rights reserved.
|
|
|
3 *
|
|
|
4 * $Date: 19. March 2015
|
|
|
5 * $Revision: V.1.4.5
|
|
|
6 *
|
|
|
7 * Project: CMSIS DSP Library
|
|
|
8 * Title: arm_mult_q31.c
|
|
|
9 *
|
|
|
10 * Description: Q31 vector multiplication.
|
|
|
11 *
|
|
|
12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
|
|
|
13 *
|
|
|
14 * Redistribution and use in source and binary forms, with or without
|
|
|
15 * modification, are permitted provided that the following conditions
|
|
|
16 * are met:
|
|
|
17 * - Redistributions of source code must retain the above copyright
|
|
|
18 * notice, this list of conditions and the following disclaimer.
|
|
|
19 * - Redistributions in binary form must reproduce the above copyright
|
|
|
20 * notice, this list of conditions and the following disclaimer in
|
|
|
21 * the documentation and/or other materials provided with the
|
|
|
22 * distribution.
|
|
|
23 * - Neither the name of ARM LIMITED nor the names of its contributors
|
|
|
24 * may be used to endorse or promote products derived from this
|
|
|
25 * software without specific prior written permission.
|
|
|
26 *
|
|
|
27 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
|
28 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
|
29 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
|
|
|
30 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
|
|
|
31 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
|
32 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
|
|
|
33 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
|
|
|
34 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
|
|
35 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
|
|
|
37 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
|
|
38 * POSSIBILITY OF SUCH DAMAGE.
|
|
|
39 * -------------------------------------------------------------------- */
|
|
|
40
|
|
|
41 #include "arm_math.h"
|
|
|
42
|
|
|
43 /**
|
|
|
44 * @ingroup groupMath
|
|
|
45 */
|
|
|
46
|
|
|
47 /**
|
|
|
48 * @addtogroup BasicMult
|
|
|
49 * @{
|
|
|
50 */
|
|
|
51
|
|
|
52 /**
|
|
|
53 * @brief Q31 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
|
|
|
58 * @return none.
|
|
|
59 *
|
|
|
60 * <b>Scaling and Overflow Behavior:</b>
|
|
|
61 * \par
|
|
|
62 * The function uses saturating arithmetic.
|
|
|
63 * Results outside of the allowable Q31 range[0x80000000 0x7FFFFFFF] will be saturated.
|
|
|
64 */
|
|
|
65
|
|
|
66 void arm_mult_q31(
|
|
|
67 q31_t * pSrcA,
|
|
|
68 q31_t * pSrcB,
|
|
|
69 q31_t * pDst,
|
|
|
70 uint32_t blockSize)
|
|
|
71 {
|
|
|
72 uint32_t blkCnt; /* loop counters */
|
|
|
73
|
|
|
74 #ifndef ARM_MATH_CM0_FAMILY
|
|
|
75
|
|
|
76 /* Run the below code for Cortex-M4 and Cortex-M3 */
|
|
|
77 q31_t inA1, inA2, inA3, inA4; /* temporary input variables */
|
|
|
78 q31_t inB1, inB2, inB3, inB4; /* temporary input variables */
|
|
|
79 q31_t out1, out2, out3, out4; /* temporary output variables */
|
|
|
80
|
|
|
81 /* loop Unrolling */
|
|
|
82 blkCnt = blockSize >> 2u;
|
|
|
83
|
|
|
84 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
|
|
|
85 ** a second loop below computes the remaining 1 to 3 samples. */
|
|
|
86 while(blkCnt > 0u)
|
|
|
87 {
|
|
|
88 /* C = A * B */
|
|
|
89 /* Multiply the inputs and then store the results in the destination buffer. */
|
|
|
90 inA1 = *pSrcA++;
|
|
|
91 inA2 = *pSrcA++;
|
|
|
92 inA3 = *pSrcA++;
|
|
|
93 inA4 = *pSrcA++;
|
|
|
94 inB1 = *pSrcB++;
|
|
|
95 inB2 = *pSrcB++;
|
|
|
96 inB3 = *pSrcB++;
|
|
|
97 inB4 = *pSrcB++;
|
|
|
98
|
|
|
99 out1 = ((q63_t) inA1 * inB1) >> 32;
|
|
|
100 out2 = ((q63_t) inA2 * inB2) >> 32;
|
|
|
101 out3 = ((q63_t) inA3 * inB3) >> 32;
|
|
|
102 out4 = ((q63_t) inA4 * inB4) >> 32;
|
|
|
103
|
|
|
104 out1 = __SSAT(out1, 31);
|
|
|
105 out2 = __SSAT(out2, 31);
|
|
|
106 out3 = __SSAT(out3, 31);
|
|
|
107 out4 = __SSAT(out4, 31);
|
|
|
108
|
|
|
109 *pDst++ = out1 << 1u;
|
|
|
110 *pDst++ = out2 << 1u;
|
|
|
111 *pDst++ = out3 << 1u;
|
|
|
112 *pDst++ = out4 << 1u;
|
|
|
113
|
|
|
114 /* Decrement the blockSize loop counter */
|
|
|
115 blkCnt--;
|
|
|
116 }
|
|
|
117
|
|
|
118 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
|
|
|
119 ** No loop unrolling is used. */
|
|
|
120 blkCnt = blockSize % 0x4u;
|
|
|
121
|
|
|
122 while(blkCnt > 0u)
|
|
|
123 {
|
|
|
124 /* C = A * B */
|
|
|
125 /* Multiply the inputs and then store the results in the destination buffer. */
|
|
|
126 inA1 = *pSrcA++;
|
|
|
127 inB1 = *pSrcB++;
|
|
|
128 out1 = ((q63_t) inA1 * inB1) >> 32;
|
|
|
129 out1 = __SSAT(out1, 31);
|
|
|
130 *pDst++ = out1 << 1u;
|
|
|
131
|
|
|
132 /* Decrement the blockSize loop counter */
|
|
|
133 blkCnt--;
|
|
|
134 }
|
|
|
135
|
|
|
136 #else
|
|
|
137
|
|
|
138 /* Run the below code for Cortex-M0 */
|
|
|
139
|
|
|
140 /* Initialize blkCnt with number of samples */
|
|
|
141 blkCnt = blockSize;
|
|
|
142
|
|
|
143
|
|
|
144 while(blkCnt > 0u)
|
|
|
145 {
|
|
|
146 /* C = A * B */
|
|
|
147 /* Multiply the inputs and then store the results in the destination buffer. */
|
|
|
148 *pDst++ =
|
|
|
149 (q31_t) clip_q63_to_q31(((q63_t) (*pSrcA++) * (*pSrcB++)) >> 31);
|
|
|
150
|
|
|
151 /* Decrement the blockSize loop counter */
|
|
|
152 blkCnt--;
|
|
|
153 }
|
|
|
154
|
|
|
155 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
|
|
|
156 }
|
|
|
157
|
|
|
158 /**
|
|
|
159 * @} end of BasicMult group
|
|
|
160 */
|
