arm_dot_prod_q31.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_dot_prod_q31.c  
00009 *  
00010 * Description:  Q31 dot product.  
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 0.0.7  2010/06/10   
00027 *    Misra-C changes done  
00028 * -------------------------------------------------------------------- */ 
00029  
00030 #include "arm_math.h" 
00031  
00032 /**  
00033  * @ingroup groupMath  
00034  */ 
00035  
00036 /**  
00037  * @addtogroup dot_prod  
00038  * @{  
00039  */ 
00040  
00041 /**  
00042  * @brief Dot product of Q31 vectors.  
00043  * @param[in]       *pSrcA points to the first input vector  
00044  * @param[in]       *pSrcB points to the second input vector  
00045  * @param[in]       blockSize number of samples in each vector  
00046  * @param[out]      *result output result returned here  
00047  * @return none.  
00048  *  
00049  * <b>Scaling and Overflow Behavior:</b>  
00050  * \par  
00051  * The intermediate multiplications are in 1.31 x 1.31 = 2.62 format and these  
00052  * are truncated to 2.48 format by discarding the lower 14 bits.  
00053  * The 2.48 result is then added without saturation to a 64-bit accumulator in 16.48 format.  
00054  * There are 15 guard bits in the accumulator and there is no risk of overflow as long as  
00055  * the length of the vectors is less than 2^16 elements.  
00056  * The return result is in 16.48 format.  
00057  */ 
00058  
00059 void arm_dot_prod_q31( 
00060   q31_t * pSrcA, 
00061   q31_t * pSrcB, 
00062   uint32_t blockSize, 
00063   q63_t * result) 
00064 { 
00065   q63_t sum = 0;                                 /* Temporary result storage */ 
00066   uint32_t blkCnt;                               /* loop counter */ 
00067  
00068  
00069   /*loop Unrolling */ 
00070   blkCnt = blockSize >> 2u; 
00071  
00072   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00073    ** a second loop below computes the remaining 1 to 3 samples. */ 
00074   while(blkCnt > 0u) 
00075   { 
00076     /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */ 
00077     /* Calculate dot product and then store the result in a temporary buffer. */ 
00078     sum += ((q63_t) * pSrcA++ * *pSrcB++) >> 14u; 
00079     sum += ((q63_t) * pSrcA++ * *pSrcB++) >> 14u; 
00080     sum += ((q63_t) * pSrcA++ * *pSrcB++) >> 14u; 
00081     sum += ((q63_t) * pSrcA++ * *pSrcB++) >> 14u; 
00082  
00083     /* Decrement the loop counter */ 
00084     blkCnt--; 
00085   } 
00086  
00087   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00088    ** No loop unrolling is used. */ 
00089   blkCnt = blockSize % 0x4u; 
00090  
00091   while(blkCnt > 0u) 
00092   { 
00093     /* C = A[0]* B[0] + A[1]* B[1] + A[2]* B[2] + .....+ A[blockSize-1]* B[blockSize-1] */ 
00094     /* Calculate dot product and then store the result in a temporary buffer. */ 
00095     sum += ((q63_t) * pSrcA++ * *pSrcB++) >> 14u; 
00096  
00097     /* Decrement the loop counter */ 
00098     blkCnt--; 
00099   } 
00100  
00101   /* Store the result in the destination buffer in 16.48 format */ 
00102   *result = sum; 
00103 } 
00104  
00105 /**  
00106  * @} end of dot_prod group  
00107  */