| 提交 | 用户 | age | ||
| bfc108 | 1 | /* ---------------------------------------------------------------------------- |
| Q | 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_f32.c | |
| 9 | * | |
| 10 | * Description: Floating-point 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 | * @defgroup MatrixAdd Matrix Addition | |
| 49 | * | |
| 50 | * Adds two matrices. | |
| 51 | * \image html MatrixAddition.gif "Addition of two 3 x 3 matrices" | |
| 52 | * | |
| 53 | * The functions check to make sure that | |
| 54 | * <code>pSrcA</code>, <code>pSrcB</code>, and <code>pDst</code> have the same | |
| 55 | * number of rows and columns. | |
| 56 | */ | |
| 57 | ||
| 58 | /** | |
| 59 | * @addtogroup MatrixAdd | |
| 60 | * @{ | |
| 61 | */ | |
| 62 | ||
| 63 | ||
| 64 | /** | |
| 65 | * @brief Floating-point matrix addition. | |
| 66 | * @param[in] *pSrcA points to the first input matrix structure | |
| 67 | * @param[in] *pSrcB points to the second input matrix structure | |
| 68 | * @param[out] *pDst points to output matrix structure | |
| 69 | * @return The function returns either | |
| 70 | * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking. | |
| 71 | */ | |
| 72 | ||
| 73 | arm_status arm_mat_add_f32( | |
| 74 | const arm_matrix_instance_f32 * pSrcA, | |
| 75 | const arm_matrix_instance_f32 * pSrcB, | |
| 76 | arm_matrix_instance_f32 * pDst) | |
| 77 | { | |
| 78 | float32_t *pIn1 = pSrcA->pData; /* input data matrix pointer A */ | |
| 79 | float32_t *pIn2 = pSrcB->pData; /* input data matrix pointer B */ | |
| 80 | float32_t *pOut = pDst->pData; /* output data matrix pointer */ | |
| 81 | ||
| 82 | #ifndef ARM_MATH_CM0_FAMILY | |
| 83 | ||
| 84 | float32_t inA1, inA2, inB1, inB2, out1, out2; /* temporary variables */ | |
| 85 | ||
| 86 | #endif // #ifndef ARM_MATH_CM0_FAMILY | |
| 87 | ||
| 88 | uint32_t numSamples; /* total number of elements in the matrix */ | |
| 89 | uint32_t blkCnt; /* loop counters */ | |
| 90 | arm_status status; /* status of matrix addition */ | |
| 91 | ||
| 92 | #ifdef ARM_MATH_MATRIX_CHECK | |
| 93 | /* Check for matrix mismatch condition */ | |
| 94 | if((pSrcA->numRows != pSrcB->numRows) || | |
| 95 | (pSrcA->numCols != pSrcB->numCols) || | |
| 96 | (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols)) | |
| 97 | { | |
| 98 | /* Set status as ARM_MATH_SIZE_MISMATCH */ | |
| 99 | status = ARM_MATH_SIZE_MISMATCH; | |
| 100 | } | |
| 101 | else | |
| 102 | #endif | |
| 103 | { | |
| 104 | ||
| 105 | /* Total number of samples in the input matrix */ | |
| 106 | numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols; | |
| 107 | ||
| 108 | #ifndef ARM_MATH_CM0_FAMILY | |
| 109 | ||
| 110 | /* Loop unrolling */ | |
| 111 | blkCnt = numSamples >> 2u; | |
| 112 | ||
| 113 | /* First part of the processing with loop unrolling. Compute 4 outputs at a time. | |
| 114 | ** a second loop below computes the remaining 1 to 3 samples. */ | |
| 115 | while(blkCnt > 0u) | |
| 116 | { | |
| 117 | /* C(m,n) = A(m,n) + B(m,n) */ | |
| 118 | /* Add and then store the results in the destination buffer. */ | |
| 119 | /* Read values from source A */ | |
| 120 | inA1 = pIn1[0]; | |
| 121 | ||
| 122 | /* Read values from source B */ | |
| 123 | inB1 = pIn2[0]; | |
| 124 | ||
| 125 | /* Read values from source A */ | |
| 126 | inA2 = pIn1[1]; | |
| 127 | ||
| 128 | /* out = sourceA + sourceB */ | |
| 129 | out1 = inA1 + inB1; | |
| 130 | ||
| 131 | /* Read values from source B */ | |
| 132 | inB2 = pIn2[1]; | |
| 133 | ||
| 134 | /* Read values from source A */ | |
| 135 | inA1 = pIn1[2]; | |
| 136 | ||
| 137 | /* out = sourceA + sourceB */ | |
| 138 | out2 = inA2 + inB2; | |
| 139 | ||
| 140 | /* Read values from source B */ | |
| 141 | inB1 = pIn2[2]; | |
| 142 | ||
| 143 | /* Store result in destination */ | |
| 144 | pOut[0] = out1; | |
| 145 | pOut[1] = out2; | |
| 146 | ||
| 147 | /* Read values from source A */ | |
| 148 | inA2 = pIn1[3]; | |
| 149 | ||
| 150 | /* Read values from source B */ | |
| 151 | inB2 = pIn2[3]; | |
| 152 | ||
| 153 | /* out = sourceA + sourceB */ | |
| 154 | out1 = inA1 + inB1; | |
| 155 | ||
| 156 | /* out = sourceA + sourceB */ | |
| 157 | out2 = inA2 + inB2; | |
| 158 | ||
| 159 | /* Store result in destination */ | |
| 160 | pOut[2] = out1; | |
| 161 | ||
| 162 | /* Store result in destination */ | |
| 163 | pOut[3] = out2; | |
| 164 | ||
| 165 | ||
| 166 | /* update pointers to process next sampels */ | |
| 167 | pIn1 += 4u; | |
| 168 | pIn2 += 4u; | |
| 169 | pOut += 4u; | |
| 170 | /* Decrement the loop counter */ | |
| 171 | blkCnt--; | |
| 172 | } | |
| 173 | ||
| 174 | /* If the numSamples is not a multiple of 4, compute any remaining output samples here. | |
| 175 | ** No loop unrolling is used. */ | |
| 176 | blkCnt = numSamples % 0x4u; | |
| 177 | ||
| 178 | #else | |
| 179 | ||
| 180 | /* Run the below code for Cortex-M0 */ | |
| 181 | ||
| 182 | /* Initialize blkCnt with number of samples */ | |
| 183 | blkCnt = numSamples; | |
| 184 | ||
| 185 | #endif /* #ifndef ARM_MATH_CM0_FAMILY */ | |
| 186 | ||
| 187 | while(blkCnt > 0u) | |
| 188 | { | |
| 189 | /* C(m,n) = A(m,n) + B(m,n) */ | |
| 190 | /* Add and then store the results in the destination buffer. */ | |
| 191 | *pOut++ = (*pIn1++) + (*pIn2++); | |
| 192 | ||
| 193 | /* Decrement the loop counter */ | |
| 194 | blkCnt--; | |
| 195 | } | |
| 196 | ||
| 197 | /* set status as ARM_MATH_SUCCESS */ | |
| 198 | status = ARM_MATH_SUCCESS; | |
| 199 | ||
| 200 | } | |
| 201 | ||
| 202 | /* Return to application */ | |
| 203 | return (status); | |
| 204 | } | |
| 205 | ||
| 206 | /** | |
| 207 | * @} end of MatrixAdd group | |
| 208 | */ | |
