1 /* ----------------------------------------------------------------------
2 * Copyright (C) 2010 ARM Limited. All rights reserved.
7 * Project: CMSIS DSP Library
8 * Title: arm_mat_add_q31.c
10 * Description: Q31 matrix addition
12 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
14 * Version 1.0.10 2011/7/15
15 * Big Endian support added and Merged M0 and M3/M4 Source code.
17 * Version 1.0.3 2010/11/29
18 * Re-organized the CMSIS folders and updated documentation.
20 * Version 1.0.2 2010/11/11
21 * Documentation updated.
23 * Version 1.0.1 2010/10/05
24 * Production release and review comments incorporated.
26 * Version 1.0.0 2010/09/20
27 * Production release and review comments incorporated.
29 * Version 0.0.5 2010/04/26
30 * incorporated review comments and updated with latest CMSIS layer
32 * Version 0.0.3 2010/03/10
34 * -------------------------------------------------------------------- */
39 * @ingroup groupMatrix
43 * @addtogroup MatrixAdd
48 * @brief Q31 matrix addition.
49 * @param[in] *pSrcA points to the first input matrix structure
50 * @param[in] *pSrcB points to the second input matrix structure
51 * @param[out] *pDst points to output matrix structure
52 * @return The function returns either
53 * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
55 * <b>Scaling and Overflow Behavior:</b>
57 * The function uses saturating arithmetic.
58 * Results outside of the allowable Q31 range [0x80000000 0x7FFFFFFF] will be saturated.
61 arm_status arm_mat_add_q31(
62 const arm_matrix_instance_q31 * pSrcA,
63 const arm_matrix_instance_q31 * pSrcB,
64 arm_matrix_instance_q31 * pDst)
66 q31_t *pIn1 = pSrcA->pData; /* input data matrix pointer A */
67 q31_t *pIn2 = pSrcB->pData; /* input data matrix pointer B */
68 q31_t *pOut = pDst->pData; /* output data matrix pointer */
69 uint32_t numSamples; /* total number of elements in the matrix */
70 uint32_t blkCnt; /* loop counters */
71 arm_status status; /* status of matrix addition */
73 #ifdef ARM_MATH_MATRIX_CHECK
76 /* Check for matrix mismatch condition */
77 if((pSrcA->numRows != pSrcB->numRows) ||
78 (pSrcA->numCols != pSrcB->numCols) ||
79 (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
81 /* Set status as ARM_MATH_SIZE_MISMATCH */
82 status = ARM_MATH_SIZE_MISMATCH;
85 #endif /* #ifdef ARM_MATH_MATRIX_CHECK */
88 /* Total number of samples in the input matrix */
89 numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
93 /* Run the below code for Cortex-M4 and Cortex-M3 */
96 blkCnt = numSamples >> 2u;
99 /* First part of the processing with loop unrolling. Compute 4 outputs at a time.
100 ** a second loop below computes the remaining 1 to 3 samples. */
103 /* C(m,n) = A(m,n) + B(m,n) */
104 /* Add, saturate and then store the results in the destination buffer. */
105 *pOut++ = __QADD(*pIn1++, *pIn2++);
106 *pOut++ = __QADD(*pIn1++, *pIn2++);
107 *pOut++ = __QADD(*pIn1++, *pIn2++);
108 *pOut++ = __QADD(*pIn1++, *pIn2++);
110 /* Decrement the loop counter */
114 /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
115 ** No loop unrolling is used. */
116 blkCnt = numSamples % 0x4u;
120 /* C(m,n) = A(m,n) + B(m,n) */
121 /* Add, saturate and then store the results in the destination buffer. */
122 *pOut++ = __QADD(*pIn1++, *pIn2++);
124 /* Decrement the loop counter */
130 /* Run the below code for Cortex-M0 */
132 /* Initialize blkCnt with number of samples */
137 /* C(m,n) = A(m,n) + B(m,n) */
138 /* Add, saturate and then store the results in the destination buffer. */
139 *pOut++ = clip_q63_to_q31(((q63_t) (*pIn1++)) + (*pIn2++));
141 /* Decrement the loop counter */
145 #endif /* #ifndef ARM_MATH_CM0 */
147 /* set status as ARM_MATH_SUCCESS */
148 status = ARM_MATH_SUCCESS;
151 /* Return to application */
156 * @} end of MatrixAdd group