arm_mat_add_q15.c

00001 /* ----------------------------------------------------------------------  
00002 * Copyright (C) 2010 ARM Limited. All rights reserved.  
00003 *  
00004 * $Date:        29. November 2010  
00005 * $Revision:    V1.0.3  
00006 *  
00007 * Project:      CMSIS DSP Library  
00008 * Title:        arm_mat_add_q15.c  
00009 *  
00010 * Description:  Q15 matrix addition  
00011 *  
00012 * Target Processor: Cortex-M4/Cortex-M3
00013 *  
00014 * Version 1.0.3 2010/11/29 
00015 *    Re-organized the CMSIS folders and updated documentation.  
00016 *   
00017 * Version 1.0.2 2010/11/11  
00018 *    Documentation updated.   
00019 *  
00020 * Version 1.0.1 2010/10/05   
00021 *    Production release and review comments incorporated.  
00022 *  
00023 * Version 1.0.0 2010/09/20   
00024 *    Production release and review comments incorporated.  
00025 *  
00026 * Version 0.0.5  2010/04/26   
00027 *    incorporated review comments and updated with latest CMSIS layer  
00028 *  
00029 * Version 0.0.3  2010/03/10   
00030 *    Initial version  
00031 * -------------------------------------------------------------------- */ 
00032  
00033 #include "arm_math.h" 
00034  
00035 /**  
00036  * @ingroup groupMatrix  
00037  */ 
00038  
00039 /**  
00040  * @addtogroup MatrixAdd  
00041  * @{  
00042  */ 
00043  
00044 /**  
00045  * @brief Q15 matrix addition.  
00046  * @param[in]       *pSrcA points to the first input matrix structure  
00047  * @param[in]       *pSrcB points to the second input matrix structure  
00048  * @param[out]      *pDst points to output matrix structure  
00049  * @return          The function returns either  
00050  * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.  
00051  *  
00052  * <b>Scaling and Overflow Behavior:</b>  
00053  * \par  
00054  * The function uses saturating arithmetic.  
00055  * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.  
00056  */ 
00057  
00058 arm_status arm_mat_add_q15( 
00059   const arm_matrix_instance_q15 * pSrcA, 
00060   const arm_matrix_instance_q15 * pSrcB, 
00061   arm_matrix_instance_q15 * pDst) 
00062 { 
00063   q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A  */ 
00064   q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B */ 
00065   q15_t *pOut = pDst->pData;                     /* output data matrix pointer */ 
00066   uint16_t numSamples;                           /* total number of elements in the matrix  */ 
00067   uint32_t blkCnt;                               /* loop counters  */ 
00068   arm_status status;                             /* status of matrix addition  */ 
00069  
00070 #ifdef ARM_MATH_MATRIX_CHECK 
00071   /* Check for matrix mismatch condition */ 
00072   if((pSrcA->numRows != pSrcB->numRows) || 
00073      (pSrcA->numCols != pSrcB->numCols) || 
00074      (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols)) 
00075   { 
00076     /* Set status as ARM_MATH_SIZE_MISMATCH */ 
00077     status = ARM_MATH_SIZE_MISMATCH; 
00078   } 
00079   else 
00080 #endif 
00081   { 
00082     /* Total number of samples in the input matrix */ 
00083     numSamples = (uint16_t) (pSrcA->numRows * pSrcA->numCols); 
00084  
00085     /* Loop unrolling */ 
00086     blkCnt = (uint32_t) numSamples >> 2u; 
00087  
00088     /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00089      ** a second loop below computes the remaining 1 to 3 samples. */ 
00090     while(blkCnt > 0u) 
00091     { 
00092       /* C(m,n) = A(m,n) + B(m,n) */ 
00093       /* Add, Saturate and then store the results in the destination buffer. */ 
00094       *__SIMD32(pOut)++ = __QADD16(*__SIMD32(pInA)++, *__SIMD32(pInB)++); 
00095       *__SIMD32(pOut)++ = __QADD16(*__SIMD32(pInA)++, *__SIMD32(pInB)++); 
00096  
00097       /* Decrement the loop counter */ 
00098       blkCnt--; 
00099     } 
00100  
00101     /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00102      ** No loop unrolling is used. */ 
00103     blkCnt = (uint32_t) numSamples % 0x4u; 
00104  
00105     /* q15 pointers of input and output are initialized */ 
00106  
00107     while(blkCnt > 0u) 
00108     { 
00109       /* C(m,n) = A(m,n) + B(m,n) */ 
00110       /* Add, Saturate and then store the results in the destination buffer. */ 
00111       *pOut++ = (q15_t) __QADD16(*pInA++, *pInB++); 
00112  
00113       /* Decrement the loop counter */ 
00114       blkCnt--; 
00115     } 
00116  
00117     /* set status as ARM_MATH_SUCCESS */ 
00118     status = ARM_MATH_SUCCESS; 
00119   } 
00120  
00121   /* Return to application */ 
00122   return (status); 
00123 } 
00124  
00125 /**  
00126  * @} end of MatrixAdd group  
00127  */