arm_mat_scale_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_scale_q15.c  
00009 *  
00010 * Description:  Multiplies a Q15 matrix by a scalar.  
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 MatrixScale  
00041  * @{  
00042  */ 
00043  
00044 /**  
00045  * @brief Q15 matrix scaling.  
00046  * @param[in]       *pSrc points to input matrix  
00047  * @param[in]       scaleFract fractional portion of the scale factor  
00048  * @param[in]       shift number of bits to shift the result by  
00049  * @param[out]      *pDst points to output matrix structure  
00050  * @return          The function returns either  
00051  * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.  
00052  *  
00053  * @details  
00054  * <b>Scaling and Overflow Behavior:</b>  
00055  * \par  
00056  * The input data <code>*pSrc</code> and <code>scaleFract</code> are in 1.15 format.  
00057  * These are multiplied to yield a 2.30 intermediate result and this is shifted with saturation to 1.15 format.  
00058  */ 
00059  
00060 arm_status arm_mat_scale_q15( 
00061   const arm_matrix_instance_q15 * pSrc, 
00062   q15_t scaleFract, 
00063   int32_t shift, 
00064   arm_matrix_instance_q15 * pDst) 
00065 { 
00066   q15_t *pIn = pSrc->pData;                      /* input data matrix pointer */ 
00067   q15_t *pOut = pDst->pData;                     /* output data matrix pointer */ 
00068   uint32_t numSamples;                           /* total number of elements in the matrix */ 
00069   int32_t totShift = 15 - shift;                 /* total shift to apply after scaling */ 
00070   uint32_t blkCnt;                               /* loop counters */ 
00071   arm_status status;                             /* status of matrix scaling     */ 
00072  
00073 #ifdef ARM_MATH_MATRIX_CHECK 
00074   /* Check for matrix mismatch */ 
00075   if((pSrc->numRows != pDst->numRows) || (pSrc->numCols != pDst->numCols)) 
00076   { 
00077     /* Set status as ARM_MATH_SIZE_MISMATCH */ 
00078     status = ARM_MATH_SIZE_MISMATCH; 
00079   } 
00080   else 
00081 #endif 
00082   { 
00083     /* Total number of samples in the input matrix */ 
00084     numSamples = (uint32_t) pSrc->numRows * pSrc->numCols; 
00085  
00086     /* Loop Unrolling */ 
00087     blkCnt = numSamples >> 2; 
00088  
00089     /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00090      ** a second loop below computes the remaining 1 to 3 samples. */ 
00091     while(blkCnt > 0u) 
00092     { 
00093       /* C(m,n) = A(m,n) * k */ 
00094       /* Scale, saturate and then store the results in the destination buffer. */ 
00095       *pOut++ = 
00096         (q15_t) (__SSAT(((q31_t) (*pIn++) * scaleFract) >> totShift, 16)); 
00097       *pOut++ = 
00098         (q15_t) (__SSAT(((q31_t) (*pIn++) * scaleFract) >> totShift, 16)); 
00099       *pOut++ = 
00100         (q15_t) (__SSAT(((q31_t) (*pIn++) * scaleFract) >> totShift, 16)); 
00101       *pOut++ = 
00102         (q15_t) (__SSAT(((q31_t) (*pIn++) * scaleFract) >> totShift, 16)); 
00103  
00104       /* Decrement the numSamples loop counter */ 
00105       blkCnt--; 
00106     } 
00107  
00108     /* If the numSamples is not a multiple of 4, compute any remaining output samples here.  
00109      ** No loop unrolling is used. */ 
00110     blkCnt = numSamples % 0x4u; 
00111  
00112     while(blkCnt > 0u) 
00113     { 
00114       /* C(m,n) = A(m,n) * k */ 
00115       /* Scale, saturate and then store the results in the destination buffer. */ 
00116       *pOut++ = 
00117         (q15_t) (__SSAT(((q31_t) (*pIn++) * scaleFract) >> totShift, 16)); 
00118  
00119       /* Decrement the numSamples loop counter */ 
00120       blkCnt--; 
00121     } 
00122  
00123     /* Set status as ARM_MATH_SUCCESS */ 
00124     status = ARM_MATH_SUCCESS; 
00125   } 
00126  
00127   /* Return to application */ 
00128   return (status); 
00129 } 
00130  
00131 /**  
00132  * @} end of MatrixScale group  
00133  */