Rev 49 | Details | Compare with Previous | Last modification | View Log | RSS feed
Rev | Author | Line No. | Line |
---|---|---|---|
2 | mjames | 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_mat_sub_q31.c |
||
9 | * |
||
10 | * Description: Q31 matrix subtraction |
||
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 groupMatrix |
||
45 | */ |
||
46 | |||
47 | /** |
||
48 | * @addtogroup MatrixSub |
||
49 | * @{ |
||
50 | */ |
||
51 | |||
52 | /** |
||
53 | * @brief Q31 matrix subtraction. |
||
54 | * @param[in] *pSrcA points to the first input matrix structure |
||
55 | * @param[in] *pSrcB points to the second input matrix structure |
||
56 | * @param[out] *pDst points to output matrix structure |
||
57 | * @return The function returns either |
||
58 | * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking. |
||
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 | |||
67 | arm_status arm_mat_sub_q31( |
||
68 | const arm_matrix_instance_q31 * pSrcA, |
||
69 | const arm_matrix_instance_q31 * pSrcB, |
||
70 | arm_matrix_instance_q31 * pDst) |
||
71 | { |
||
72 | q31_t *pIn1 = pSrcA->pData; /* input data matrix pointer A */ |
||
73 | q31_t *pIn2 = pSrcB->pData; /* input data matrix pointer B */ |
||
74 | q31_t *pOut = pDst->pData; /* output data matrix pointer */ |
||
75 | q31_t inA1, inB1; /* temporary variables */ |
||
76 | |||
77 | #ifndef ARM_MATH_CM0_FAMILY |
||
78 | |||
79 | q31_t inA2, inB2; /* temporary variables */ |
||
80 | q31_t out1, out2; /* temporary variables */ |
||
81 | |||
82 | #endif // #ifndef ARM_MATH_CM0_FAMILY |
||
83 | |||
84 | uint32_t numSamples; /* total number of elements in the matrix */ |
||
85 | uint32_t blkCnt; /* loop counters */ |
||
86 | arm_status status; /* status of matrix subtraction */ |
||
87 | |||
88 | |||
89 | #ifdef ARM_MATH_MATRIX_CHECK |
||
90 | /* Check for matrix mismatch condition */ |
||
91 | if((pSrcA->numRows != pSrcB->numRows) || |
||
92 | (pSrcA->numCols != pSrcB->numCols) || |
||
93 | (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols)) |
||
94 | { |
||
95 | /* Set status as ARM_MATH_SIZE_MISMATCH */ |
||
96 | status = ARM_MATH_SIZE_MISMATCH; |
||
97 | } |
||
98 | else |
||
99 | #endif |
||
100 | { |
||
101 | /* Total number of samples in the input matrix */ |
||
102 | numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols; |
||
103 | |||
104 | #ifndef ARM_MATH_CM0_FAMILY |
||
105 | |||
106 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
||
107 | |||
108 | /* Loop Unrolling */ |
||
109 | blkCnt = numSamples >> 2u; |
||
110 | |||
111 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
||
112 | ** a second loop below computes the remaining 1 to 3 samples. */ |
||
113 | while(blkCnt > 0u) |
||
114 | { |
||
115 | /* C(m,n) = A(m,n) - B(m,n) */ |
||
116 | /* Subtract, saturate and then store the results in the destination buffer. */ |
||
117 | /* Read values from source A */ |
||
118 | inA1 = pIn1[0]; |
||
119 | |||
120 | /* Read values from source B */ |
||
121 | inB1 = pIn2[0]; |
||
122 | |||
123 | /* Read values from source A */ |
||
124 | inA2 = pIn1[1]; |
||
125 | |||
126 | /* Subtract and saturate */ |
||
127 | out1 = __QSUB(inA1, inB1); |
||
128 | |||
129 | /* Read values from source B */ |
||
130 | inB2 = pIn2[1]; |
||
131 | |||
132 | /* Read values from source A */ |
||
133 | inA1 = pIn1[2]; |
||
134 | |||
135 | /* Subtract and saturate */ |
||
136 | out2 = __QSUB(inA2, inB2); |
||
137 | |||
138 | /* Read values from source B */ |
||
139 | inB1 = pIn2[2]; |
||
140 | |||
141 | /* Store result in destination */ |
||
142 | pOut[0] = out1; |
||
143 | pOut[1] = out2; |
||
144 | |||
145 | /* Read values from source A */ |
||
146 | inA2 = pIn1[3]; |
||
147 | |||
148 | /* Read values from source B */ |
||
149 | inB2 = pIn2[3]; |
||
150 | |||
151 | /* Subtract and saturate */ |
||
152 | out1 = __QSUB(inA1, inB1); |
||
153 | |||
154 | /* Subtract and saturate */ |
||
155 | out2 = __QSUB(inA2, inB2); |
||
156 | |||
157 | /* Store result in destination */ |
||
158 | pOut[2] = out1; |
||
159 | pOut[3] = out2; |
||
160 | |||
161 | /* update pointers to process next samples */ |
||
162 | pIn1 += 4u; |
||
163 | pIn2 += 4u; |
||
164 | pOut += 4u; |
||
165 | |||
166 | /* Decrement the loop counter */ |
||
167 | blkCnt--; |
||
168 | } |
||
169 | |||
170 | /* If the numSamples is not a multiple of 4, compute any remaining output samples here. |
||
171 | ** No loop unrolling is used. */ |
||
172 | blkCnt = numSamples % 0x4u; |
||
173 | |||
174 | #else |
||
175 | |||
176 | /* Run the below code for Cortex-M0 */ |
||
177 | |||
178 | /* Initialize blkCnt with number of samples */ |
||
179 | blkCnt = numSamples; |
||
180 | |||
181 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
||
182 | |||
183 | while(blkCnt > 0u) |
||
184 | { |
||
185 | /* C(m,n) = A(m,n) - B(m,n) */ |
||
186 | /* Subtract, saturate and then store the results in the destination buffer. */ |
||
187 | inA1 = *pIn1++; |
||
188 | inB1 = *pIn2++; |
||
189 | |||
190 | inA1 = __QSUB(inA1, inB1); |
||
191 | |||
192 | *pOut++ = inA1; |
||
193 | |||
194 | /* Decrement the loop counter */ |
||
195 | blkCnt--; |
||
196 | } |
||
197 | |||
198 | /* Set status as ARM_MATH_SUCCESS */ |
||
199 | status = ARM_MATH_SUCCESS; |
||
200 | } |
||
201 | |||
202 | /* Return to application */ |
||
203 | return (status); |
||
204 | } |
||
205 | |||
206 | /** |
||
207 | * @} end of MatrixSub group |
||
208 | */ |