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_q15.c
9 *
10 * Description: Q15 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 Q15 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 Q15 range [0x8000 0x7FFF] will be saturated.
64 */
65
arm_mat_sub_q15(const arm_matrix_instance_q15 * pSrcA,const arm_matrix_instance_q15 * pSrcB,arm_matrix_instance_q15 * pDst)66 arm_status arm_mat_sub_q15(
67 const arm_matrix_instance_q15 * pSrcA,
68 const arm_matrix_instance_q15 * pSrcB,
69 arm_matrix_instance_q15 * pDst)
70 {
71 q15_t *pInA = pSrcA->pData; /* input data matrix pointer A */
72 q15_t *pInB = pSrcB->pData; /* input data matrix pointer B */
73 q15_t *pOut = pDst->pData; /* output data matrix pointer */
74 uint32_t numSamples; /* total number of elements in the matrix */
75 uint32_t blkCnt; /* loop counters */
76 arm_status status; /* status of matrix subtraction */
77
78
79 #ifdef ARM_MATH_MATRIX_CHECK
80
81
82 /* Check for matrix mismatch condition */
83 if((pSrcA->numRows != pSrcB->numRows) ||
84 (pSrcA->numCols != pSrcB->numCols) ||
85 (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
86 {
87 /* Set status as ARM_MATH_SIZE_MISMATCH */
88 status = ARM_MATH_SIZE_MISMATCH;
89 }
90 else
91 #endif /* #ifdef ARM_MATH_MATRIX_CHECK */
92
93 {
94 /* Total number of samples in the input matrix */
95 numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
96
97 #ifndef ARM_MATH_CM0_FAMILY
98
99 /* Run the below code for Cortex-M4 and Cortex-M3 */
100
101 /* Apply loop unrolling */
102 blkCnt = numSamples >> 2u;
103
104 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
105 ** a second loop below computes the remaining 1 to 3 samples. */
106 while(blkCnt > 0u)
107 {
108 /* C(m,n) = A(m,n) - B(m,n) */
109 /* Subtract, Saturate and then store the results in the destination buffer. */
110 *__SIMD32(pOut)++ = __QSUB16(*__SIMD32(pInA)++, *__SIMD32(pInB)++);
111 *__SIMD32(pOut)++ = __QSUB16(*__SIMD32(pInA)++, *__SIMD32(pInB)++);
112
113 /* Decrement the loop counter */
114 blkCnt--;
115 }
116
117 /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
118 ** No loop unrolling is used. */
119 blkCnt = numSamples % 0x4u;
120
121 while(blkCnt > 0u)
122 {
123 /* C(m,n) = A(m,n) - B(m,n) */
124 /* Subtract and then store the results in the destination buffer. */
125 *pOut++ = (q15_t) __QSUB16(*pInA++, *pInB++);
126
127 /* Decrement the loop counter */
128 blkCnt--;
129 }
130
131 #else
132
133 /* Run the below code for Cortex-M0 */
134
135 /* Initialize blkCnt with number of samples */
136 blkCnt = numSamples;
137
138 while(blkCnt > 0u)
139 {
140 /* C(m,n) = A(m,n) - B(m,n) */
141 /* Subtract and then store the results in the destination buffer. */
142 *pOut++ = (q15_t) __SSAT(((q31_t) * pInA++ - *pInB++), 16);
143
144 /* Decrement the loop counter */
145 blkCnt--;
146 }
147
148 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
149
150 /* Set status as ARM_MATH_SUCCESS */
151 status = ARM_MATH_SUCCESS;
152 }
153
154 /* Return to application */
155 return (status);
156 }
157
158 /**
159 * @} end of MatrixSub group
160 */
161