tmk_core/tool/mbed/mbed-sdk/libraries/dsp/cmsis_dsp/MatrixFunctions/arm_mat_add_q31.c

   1 /* ----------------------------------------------------------------------
   2 * Copyright (C) 2010-2013 ARM Limited. All rights reserved.
   3 *
   4 * $Date:        17. January 2013
   5 * $Revision:    V1.4.1
   6 *
   7 * Project:          CMSIS DSP Library
   8 * Title:            arm_mat_add_q31.c
   9 *
  10 * Description:  Q31 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 Q31 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 Q31 range [0x80000000 0x7FFFFFFF] will be saturated.
  64  */
  65
  66 arm_status arm_mat_add_q31(
  67   const arm_matrix_instance_q31 * pSrcA,
  68   const arm_matrix_instance_q31 * pSrcB,
  69   arm_matrix_instance_q31 * pDst)
  70 {
  71   q31_t *pIn1 = pSrcA->pData;                    /* input data matrix pointer A */
  72   q31_t *pIn2 = pSrcB->pData;                    /* input data matrix pointer B */
  73   q31_t *pOut = pDst->pData;                     /* output data matrix pointer */
  74   q31_t inA1, inB1;                              /* temporary variables */
  75
  76 #ifndef ARM_MATH_CM0_FAMILY
  77
  78   q31_t inA2, inB2;                              /* temporary variables */
  79   q31_t out1, out2;                              /* temporary variables */
  80
  81 #endif //      #ifndef ARM_MATH_CM0_FAMILY
  82
  83   uint32_t numSamples;                           /* total number of elements in the matrix  */
  84   uint32_t blkCnt;                               /* loop counters */
  85   arm_status status;                             /* status of matrix addition */
  86
  87 #ifdef ARM_MATH_MATRIX_CHECK
  88   /* Check for matrix mismatch condition */
  89   if((pSrcA->numRows != pSrcB->numRows) ||
  90      (pSrcA->numCols != pSrcB->numCols) ||
  91      (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
  92   {
  93     /* Set status as ARM_MATH_SIZE_MISMATCH */
  94     status = ARM_MATH_SIZE_MISMATCH;
  95   }
  96   else
  97 #endif
  98   {
  99     /* Total number of samples in the input matrix */
 100     numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
 101
 102 #ifndef ARM_MATH_CM0_FAMILY
 103
 104     /* Run the below code for Cortex-M4 and Cortex-M3 */
 105
 106     /* Loop Unrolling */
 107     blkCnt = numSamples >> 2u;
 108
 109
 110     /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
 111      ** a second loop below computes the remaining 1 to 3 samples. */
 112     while(blkCnt > 0u)
 113     {
 114       /* C(m,n) = A(m,n) + B(m,n) */
 115       /* Add, saturate and then store the results in the destination buffer. */
 116       /* Read values from source A */
 117       inA1 = pIn1[0];
 118
 119       /* Read values from source B */
 120       inB1 = pIn2[0];
 121
 122       /* Read values from source A */
 123       inA2 = pIn1[1];
 124
 125       /* Add and saturate */
 126       out1 = __QADD(inA1, inB1);
 127
 128       /* Read values from source B */
 129       inB2 = pIn2[1];
 130
 131       /* Read values from source A */
 132       inA1 = pIn1[2];
 133
 134       /* Add and saturate */
 135       out2 = __QADD(inA2, inB2);
 136
 137       /* Read values from source B */
 138       inB1 = pIn2[2];
 139
 140       /* Store result in destination */
 141       pOut[0] = out1;
 142       pOut[1] = out2;
 143
 144       /* Read values from source A */
 145       inA2 = pIn1[3];
 146
 147       /* Read values from source B */
 148       inB2 = pIn2[3];
 149
 150       /* Add and saturate */
 151       out1 = __QADD(inA1, inB1);
 152       out2 = __QADD(inA2, inB2);
 153
 154       /* Store result in destination */
 155       pOut[2] = out1;
 156       pOut[3] = out2;
 157
 158       /* update pointers to process next sampels */
 159       pIn1 += 4u;
 160       pIn2 += 4u;
 161       pOut += 4u;
 162
 163       /* Decrement the loop counter */
 164       blkCnt--;
 165     }
 166
 167     /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
 168      ** No loop unrolling is used. */
 169     blkCnt = numSamples % 0x4u;
 170
 171 #else
 172
 173     /* Run the below code for Cortex-M0 */
 174
 175     /* Initialize blkCnt with number of samples */
 176     blkCnt = numSamples;
 177
 178
 179 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
 180
 181     while(blkCnt > 0u)
 182     {
 183       /* C(m,n) = A(m,n) + B(m,n) */
 184       /* Add, saturate and then store the results in the destination buffer. */
 185       inA1 = *pIn1++;
 186       inB1 = *pIn2++;
 187
 188       inA1 = __QADD(inA1, inB1);
 189
 190       /* Decrement the loop counter */
 191       blkCnt--;
 192
 193       *pOut++ = inA1;
 194
 195     }
 196
 197     /* set status as ARM_MATH_SUCCESS */
 198     status = ARM_MATH_SUCCESS;
 199   }
 200
 201   /* Return to application */
 202   return (status);
 203 }
 204
 205 /**
 206  * @} end of MatrixAdd group
 207  */