1 /* ----------------------------------------------------------------------
2 * Project: CMSIS DSP Library
3 * Title: arm_offset_q31.c
4 * Description: Q31 vector offset
5 *
6 * $Date: 23 April 2021
7 * $Revision: V1.9.0
8 *
9 * Target Processor: Cortex-M and Cortex-A cores
10 * -------------------------------------------------------------------- */
11 /*
12 * Copyright (C) 2010-2021 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 "dsp/basic_math_functions.h"
30
31 /**
32 @ingroup groupMath
33 */
34
35 /**
36 @addtogroup BasicOffset
37 @{
38 */
39
40 /**
41 @brief Adds a constant offset to a Q31 vector.
42 @param[in] pSrc points to the input vector
43 @param[in] offset is the offset to be added
44 @param[out] pDst points to the output vector
45 @param[in] blockSize number of samples in each vector
46 @return none
47
48 @par Scaling and Overflow Behavior
49 The function uses saturating arithmetic.
50 Results outside of the allowable Q31 range [0x80000000 0x7FFFFFFF] are saturated.
51 */
52
53 #if defined(ARM_MATH_MVEI) && !defined(ARM_MATH_AUTOVECTORIZE)
54
55 #include "arm_helium_utils.h"
56
arm_offset_q31(const q31_t * pSrc,q31_t offset,q31_t * pDst,uint32_t blockSize)57 void arm_offset_q31(
58 const q31_t * pSrc,
59 q31_t offset,
60 q31_t * pDst,
61 uint32_t blockSize)
62 {
63 uint32_t blkCnt; /* loop counters */
64 q31x4_t vecSrc;
65
66 /* Compute 4 outputs at a time */
67 blkCnt = blockSize >> 2;
68 while (blkCnt > 0U)
69 {
70 /*
71 * C = A + offset
72 * Add offset and then store the result in the destination buffer.
73 */
74 vecSrc = vld1q(pSrc);
75 vst1q(pDst, vqaddq(vecSrc, offset));
76 /*
77 * Decrement the blockSize loop counter
78 */
79 blkCnt--;
80 /*
81 * advance vector source and destination pointers
82 */
83 pSrc += 4;
84 pDst += 4;
85 }
86 /*
87 * tail
88 */
89 blkCnt = blockSize & 3;
90 if (blkCnt > 0U)
91 {
92 mve_pred16_t p0 = vctp32q(blkCnt);
93 vecSrc = vld1q(pSrc);
94 vstrwq_p(pDst, vqaddq(vecSrc, offset), p0);
95 }
96 }
97
98 #else
arm_offset_q31(const q31_t * pSrc,q31_t offset,q31_t * pDst,uint32_t blockSize)99 void arm_offset_q31(
100 const q31_t * pSrc,
101 q31_t offset,
102 q31_t * pDst,
103 uint32_t blockSize)
104 {
105 uint32_t blkCnt; /* Loop counter */
106
107 #if defined (ARM_MATH_LOOPUNROLL)
108
109 /* Loop unrolling: Compute 4 outputs at a time */
110 blkCnt = blockSize >> 2U;
111
112 while (blkCnt > 0U)
113 {
114 /* C = A + offset */
115
116 /* Add offset and store result in destination buffer. */
117 *pDst++ = __QADD(*pSrc++, offset);
118
119 *pDst++ = __QADD(*pSrc++, offset);
120
121 *pDst++ = __QADD(*pSrc++, offset);
122
123 *pDst++ = __QADD(*pSrc++, offset);
124
125 /* Decrement loop counter */
126 blkCnt--;
127 }
128
129 /* Loop unrolling: Compute remaining outputs */
130 blkCnt = blockSize % 0x4U;
131
132 #else
133
134 /* Initialize blkCnt with number of samples */
135 blkCnt = blockSize;
136
137 #endif /* #if defined (ARM_MATH_LOOPUNROLL) */
138
139 while (blkCnt > 0U)
140 {
141 /* C = A + offset */
142
143 /* Add offset and store result in destination buffer. */
144 #if defined (ARM_MATH_DSP)
145 *pDst++ = __QADD(*pSrc++, offset);
146 #else
147 *pDst++ = (q31_t) clip_q63_to_q31((q63_t) * pSrc++ + offset);
148 #endif
149
150 /* Decrement loop counter */
151 blkCnt--;
152 }
153
154 }
155 #endif /* defined(ARM_MATH_MVEI) */
156
157 /**
158 @} end of BasicOffset group
159 */
160