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
47 @par Scaling and Overflow Behavior
48 The function uses saturating arithmetic.
49 Results outside of the allowable Q31 range [0x80000000 0x7FFFFFFF] are saturated.
50 */
51
52 #if defined(ARM_MATH_MVEI) && !defined(ARM_MATH_AUTOVECTORIZE)
53
54 #include "arm_helium_utils.h"
55
arm_offset_q31(const q31_t * pSrc,q31_t offset,q31_t * pDst,uint32_t blockSize)56 void arm_offset_q31(
57 const q31_t * pSrc,
58 q31_t offset,
59 q31_t * pDst,
60 uint32_t blockSize)
61 {
62 uint32_t blkCnt; /* loop counters */
63 q31x4_t vecSrc;
64
65 /* Compute 4 outputs at a time */
66 blkCnt = blockSize >> 2;
67 while (blkCnt > 0U)
68 {
69 /*
70 * C = A + offset
71 * Add offset and then store the result in the destination buffer.
72 */
73 vecSrc = vld1q(pSrc);
74 vst1q(pDst, vqaddq(vecSrc, offset));
75 /*
76 * Decrement the blockSize loop counter
77 */
78 blkCnt--;
79 /*
80 * advance vector source and destination pointers
81 */
82 pSrc += 4;
83 pDst += 4;
84 }
85 /*
86 * tail
87 */
88 blkCnt = blockSize & 3;
89 if (blkCnt > 0U)
90 {
91 mve_pred16_t p0 = vctp32q(blkCnt);
92 vecSrc = vld1q(pSrc);
93 vstrwq_p(pDst, vqaddq(vecSrc, offset), p0);
94 }
95 }
96
97 #else
arm_offset_q31(const q31_t * pSrc,q31_t offset,q31_t * pDst,uint32_t blockSize)98 void arm_offset_q31(
99 const q31_t * pSrc,
100 q31_t offset,
101 q31_t * pDst,
102 uint32_t blockSize)
103 {
104 uint32_t blkCnt; /* Loop counter */
105
106 #if defined (ARM_MATH_LOOPUNROLL)
107
108 /* Loop unrolling: Compute 4 outputs at a time */
109 blkCnt = blockSize >> 2U;
110
111 while (blkCnt > 0U)
112 {
113 /* C = A + offset */
114
115 /* Add offset and store result in destination buffer. */
116 *pDst++ = __QADD(*pSrc++, offset);
117
118 *pDst++ = __QADD(*pSrc++, offset);
119
120 *pDst++ = __QADD(*pSrc++, offset);
121
122 *pDst++ = __QADD(*pSrc++, offset);
123
124 /* Decrement loop counter */
125 blkCnt--;
126 }
127
128 /* Loop unrolling: Compute remaining outputs */
129 blkCnt = blockSize % 0x4U;
130
131 #else
132
133 /* Initialize blkCnt with number of samples */
134 blkCnt = blockSize;
135
136 #endif /* #if defined (ARM_MATH_LOOPUNROLL) */
137
138 while (blkCnt > 0U)
139 {
140 /* C = A + offset */
141
142 /* Add offset and store result in destination buffer. */
143 #if defined (ARM_MATH_DSP)
144 *pDst++ = __QADD(*pSrc++, offset);
145 #else
146 *pDst++ = (q31_t) clip_q63_to_q31((q63_t) * pSrc++ + offset);
147 #endif
148
149 /* Decrement loop counter */
150 blkCnt--;
151 }
152
153 }
154 #endif /* defined(ARM_MATH_MVEI) */
155
156 /**
157 @} end of BasicOffset group
158 */
159
