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_add_q15.c |
||
| 9 | * |
||
| 10 | * Description: Q15 matrix addition |
||
| 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 MatrixAdd |
||
| 49 | * @{ |
||
| 50 | */ |
||
| 51 | |||
| 52 | /** |
||
| 53 | * @brief Q15 matrix addition. |
||
| 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 | |||
| 66 | arm_status arm_mat_add_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 | uint16_t numSamples; /* total number of elements in the matrix */ |
||
| 75 | uint32_t blkCnt; /* loop counters */ |
||
| 76 | arm_status status; /* status of matrix addition */ |
||
| 77 | |||
| 78 | #ifdef ARM_MATH_MATRIX_CHECK |
||
| 79 | |||
| 80 | |||
| 81 | /* Check for matrix mismatch condition */ |
||
| 82 | if((pSrcA->numRows != pSrcB->numRows) || |
||
| 83 | (pSrcA->numCols != pSrcB->numCols) || |
||
| 84 | (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols)) |
||
| 85 | { |
||
| 86 | /* Set status as ARM_MATH_SIZE_MISMATCH */ |
||
| 87 | status = ARM_MATH_SIZE_MISMATCH; |
||
| 88 | } |
||
| 89 | else |
||
| 90 | #endif /* #ifdef ARM_MATH_MATRIX_CHECK */ |
||
| 91 | |||
| 92 | { |
||
| 93 | /* Total number of samples in the input matrix */ |
||
| 94 | numSamples = (uint16_t) (pSrcA->numRows * pSrcA->numCols); |
||
| 95 | |||
| 96 | #ifndef ARM_MATH_CM0_FAMILY |
||
| 97 | |||
| 98 | /* Run the below code for Cortex-M4 and Cortex-M3 */ |
||
| 99 | |||
| 100 | /* Loop unrolling */ |
||
| 101 | blkCnt = (uint32_t) numSamples >> 2u; |
||
| 102 | |||
| 103 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. |
||
| 104 | ** a second loop below computes the remaining 1 to 3 samples. */ |
||
| 105 | while(blkCnt > 0u) |
||
| 106 | { |
||
| 107 | /* C(m,n) = A(m,n) + B(m,n) */ |
||
| 108 | /* Add, Saturate and then store the results in the destination buffer. */ |
||
| 109 | *__SIMD32(pOut)++ = __QADD16(*__SIMD32(pInA)++, *__SIMD32(pInB)++); |
||
| 110 | *__SIMD32(pOut)++ = __QADD16(*__SIMD32(pInA)++, *__SIMD32(pInB)++); |
||
| 111 | |||
| 112 | /* Decrement the loop counter */ |
||
| 113 | blkCnt--; |
||
| 114 | } |
||
| 115 | |||
| 116 | /* If the blockSize is not a multiple of 4, compute any remaining output samples here. |
||
| 117 | ** No loop unrolling is used. */ |
||
| 118 | blkCnt = (uint32_t) numSamples % 0x4u; |
||
| 119 | |||
| 120 | /* q15 pointers of input and output are initialized */ |
||
| 121 | |||
| 122 | while(blkCnt > 0u) |
||
| 123 | { |
||
| 124 | /* C(m,n) = A(m,n) + B(m,n) */ |
||
| 125 | /* Add, Saturate and then store the results in the destination buffer. */ |
||
| 126 | *pOut++ = (q15_t) __QADD16(*pInA++, *pInB++); |
||
| 127 | |||
| 128 | /* Decrement the loop counter */ |
||
| 129 | blkCnt--; |
||
| 130 | } |
||
| 131 | |||
| 132 | #else |
||
| 133 | |||
| 134 | /* Run the below code for Cortex-M0 */ |
||
| 135 | |||
| 136 | /* Initialize blkCnt with number of samples */ |
||
| 137 | blkCnt = (uint32_t) numSamples; |
||
| 138 | |||
| 139 | |||
| 140 | /* q15 pointers of input and output are initialized */ |
||
| 141 | while(blkCnt > 0u) |
||
| 142 | { |
||
| 143 | /* C(m,n) = A(m,n) + B(m,n) */ |
||
| 144 | /* Add, Saturate and then store the results in the destination buffer. */ |
||
| 145 | *pOut++ = (q15_t) __SSAT(((q31_t) * pInA++ + *pInB++), 16); |
||
| 146 | |||
| 147 | /* Decrement the loop counter */ |
||
| 148 | blkCnt--; |
||
| 149 | } |
||
| 150 | |||
| 151 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ |
||
| 152 | |||
| 153 | /* set status as ARM_MATH_SUCCESS */ |
||
| 154 | status = ARM_MATH_SUCCESS; |
||
| 155 | } |
||
| 156 | |||
| 157 | /* Return to application */ |
||
| 158 | return (status); |
||
| 159 | } |
||
| 160 | |||
| 161 | /** |
||
| 162 | * @} end of MatrixAdd group |
||
| 163 | */ |