arm_shift_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_shift_q15.c  
00009 *  
00010 * Description:  Shifts elements of a Q15 vector a specified number of bits.  
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 shift  
00038  * @{  
00039  */ 
00040  
00041 /**  
00042  * @brief  Shifts the elements of a Q15 vector a specified number of bits.  
00043  * @param  *pSrc points to the input vector  
00044  * @param  shiftBits number of bits to shift.  A positive value shifts left; a negative value shifts right.  
00045  * @param  *pDst points to the output vector  
00046  * @param  blockSize number of samples in the vector  
00047  * @return none.  
00048  *  
00049  * <b>Scaling and Overflow Behavior:</b>  
00050  * \par  
00051  * The function uses saturating arithmetic.  
00052  * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.  
00053  */ 
00054  
00055 void arm_shift_q15( 
00056   q15_t * pSrc, 
00057   int8_t shiftBits, 
00058   q15_t * pDst, 
00059   uint32_t blockSize) 
00060 { 
00061   uint32_t blkCnt;                               /* loop counter */ 
00062   uint8_t sign;                                  /* Sign of shiftBits */ 
00063   q15_t in1, in2;                                /* Temporary variables */ 
00064  
00065  
00066   /*loop Unrolling */ 
00067   blkCnt = blockSize >> 2u; 
00068  
00069   /* Getting the sign of shiftBits */ 
00070   sign = (shiftBits & 0x80); 
00071  
00072   /* If the shift value is positive then do right shift else left shift */ 
00073   if(sign == 0u) 
00074   { 
00075     /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00076      ** a second loop below computes the remaining 1 to 3 samples. */ 
00077     while(blkCnt > 0u) 
00078     { 
00079       /* Read 2 inputs */ 
00080       in1 = *pSrc++; 
00081       in2 = *pSrc++; 
00082       /* C = A << shiftBits */ 
00083       /* Shift the inputs and then store the results in the destination buffer. */ 
00084       *__SIMD32(pDst)++ = __PKHBT(__SSAT((in1 << shiftBits), 16), 
00085                                   __SSAT((in2 << shiftBits), 16), 16); 
00086  
00087       in1 = *pSrc++; 
00088       in2 = *pSrc++; 
00089  
00090       *__SIMD32(pDst)++ = __PKHBT(__SSAT((in1 << shiftBits), 16), 
00091                                   __SSAT((in2 << shiftBits), 16), 16); 
00092  
00093       /* Decrement the loop counter */ 
00094       blkCnt--; 
00095     } 
00096  
00097     /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00098      ** No loop unrolling is used. */ 
00099     blkCnt = blockSize % 0x4u; 
00100  
00101     while(blkCnt > 0u) 
00102     { 
00103       /* C = A << shiftBits */ 
00104       /* Shift and then store the results in the destination buffer. */ 
00105       *pDst++ = __SSAT((*pSrc++ << shiftBits), 16); 
00106  
00107       /* Decrement the loop counter */ 
00108       blkCnt--; 
00109     } 
00110   } 
00111   else 
00112   { 
00113     /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00114      ** a second loop below computes the remaining 1 to 3 samples. */ 
00115     while(blkCnt > 0u) 
00116     { 
00117       /* Read 2 inputs */ 
00118       in1 = *pSrc++; 
00119       in2 = *pSrc++; 
00120       /* C = A >> shiftBits */ 
00121       /* Shift the inputs and then store the results in the destination buffer. */ 
00122       *__SIMD32(pDst)++ = __PKHBT((in1 >> -shiftBits), 
00123                                   (in2 >> -shiftBits), 16); 
00124       in1 = *pSrc++; 
00125       in2 = *pSrc++; 
00126  
00127       *__SIMD32(pDst)++ = __PKHBT((in1 >> -shiftBits), 
00128                                   (in2 >> -shiftBits), 16); 
00129  
00130       /* Decrement the loop counter */ 
00131       blkCnt--; 
00132     } 
00133  
00134     /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00135      ** No loop unrolling is used. */ 
00136     blkCnt = blockSize % 0x4u; 
00137  
00138     while(blkCnt > 0u) 
00139     { 
00140       /* C = A >> shiftBits */ 
00141       /* Shift the inputs and then store the results in the destination buffer. */ 
00142       *pDst++ = (*pSrc++ >> -shiftBits); 
00143  
00144       /* Decrement the loop counter */ 
00145       blkCnt--; 
00146     } 
00147   } 
00148 } 
00149  
00150 /**  
00151  * @} end of shift group  
00152  */