arm_scale_f32.c

00001 /* ----------------------------------------------------------------------    
00002 * Copyright (C) 2010-2014 ARM Limited. All rights reserved.    
00003 *    
00004 * $Date:        19. March 2015
00005 * $Revision:    V.1.4.5
00006 *    
00007 * Project:      CMSIS DSP Library    
00008 * Title:        arm_scale_f32.c    
00009 *    
00010 * Description:  Multiplies a floating-point vector by a scalar.    
00011 *    
00012 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0
00013 *  
00014 * Redistribution and use in source and binary forms, with or without 
00015 * modification, are permitted provided that the following conditions
00016 * are met:
00017 *   - Redistributions of source code must retain the above copyright
00018 *     notice, this list of conditions and the following disclaimer.
00019 *   - Redistributions in binary form must reproduce the above copyright
00020 *     notice, this list of conditions and the following disclaimer in
00021 *     the documentation and/or other materials provided with the 
00022 *     distribution.
00023 *   - Neither the name of ARM LIMITED nor the names of its contributors
00024 *     may be used to endorse or promote products derived from this
00025 *     software without specific prior written permission.
00026 *
00027 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
00028 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
00029 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
00030 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE 
00031 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
00032 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
00033 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
00034 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
00035 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
00036 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
00037 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
00038 * POSSIBILITY OF SUCH DAMAGE.    
00039 * ---------------------------------------------------------------------------- */
00040 
00041 #include "arm_math.h"
00042 
00043 /**        
00044  * @ingroup groupMath        
00045  */
00046 
00047 /**        
00048  * @defgroup scale Vector Scale        
00049  *        
00050  * Multiply a vector by a scalar value.  For floating-point data, the algorithm used is:        
00051  *        
00052  * <pre>        
00053  *     pDst[n] = pSrc[n] * scale,   0 <= n < blockSize.        
00054  * </pre>        
00055  *        
00056  * In the fixed-point Q7, Q15, and Q31 functions, <code>scale</code> is represented by        
00057  * a fractional multiplication <code>scaleFract</code> and an arithmetic shift <code>shift</code>.        
00058  * The shift allows the gain of the scaling operation to exceed 1.0.        
00059  * The algorithm used with fixed-point data is:        
00060  *        
00061  * <pre>        
00062  *     pDst[n] = (pSrc[n] * scaleFract) << shift,   0 <= n < blockSize.        
00063  * </pre>        
00064  *        
00065  * The overall scale factor applied to the fixed-point data is        
00066  * <pre>        
00067  *     scale = scaleFract * 2^shift.        
00068  * </pre>        
00069  *
00070  * The functions support in-place computation allowing the source and destination
00071  * pointers to reference the same memory buffer.
00072  */
00073 
00074 /**        
00075  * @addtogroup scale        
00076  * @{        
00077  */
00078 
00079 /**        
00080  * @brief Multiplies a floating-point vector by a scalar.        
00081  * @param[in]       *pSrc points to the input vector        
00082  * @param[in]       scale scale factor to be applied        
00083  * @param[out]      *pDst points to the output vector        
00084  * @param[in]       blockSize number of samples in the vector        
00085  * @return none.        
00086  */
00087 
00088 
00089 void arm_scale_f32(
00090   float32_t * pSrc,
00091   float32_t scale,
00092   float32_t * pDst,
00093   uint32_t blockSize)
00094 {
00095   uint32_t blkCnt;                               /* loop counter */
00096 #ifndef ARM_MATH_CM0_FAMILY
00097 
00098 /* Run the below code for Cortex-M4 and Cortex-M3 */
00099   float32_t in1, in2, in3, in4;                  /* temporary variabels */
00100 
00101   /*loop Unrolling */
00102   blkCnt = blockSize >> 2u;
00103 
00104   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.        
00105    ** a second loop below computes the remaining 1 to 3 samples. */
00106   while(blkCnt > 0u)
00107   {
00108     /* C = A * scale */
00109     /* Scale the input and then store the results in the destination buffer. */
00110     /* read input samples from source */
00111     in1 = *pSrc;
00112     in2 = *(pSrc + 1);
00113 
00114     /* multiply with scaling factor */
00115     in1 = in1 * scale;
00116 
00117     /* read input sample from source */
00118     in3 = *(pSrc + 2);
00119 
00120     /* multiply with scaling factor */
00121     in2 = in2 * scale;
00122 
00123     /* read input sample from source */
00124     in4 = *(pSrc + 3);
00125 
00126     /* multiply with scaling factor */
00127     in3 = in3 * scale;
00128     in4 = in4 * scale;
00129     /* store the result to destination */
00130     *pDst = in1;
00131     *(pDst + 1) = in2;
00132     *(pDst + 2) = in3;
00133     *(pDst + 3) = in4;
00134 
00135     /* update pointers to process next samples */
00136     pSrc += 4u;
00137     pDst += 4u;
00138 
00139     /* Decrement the loop counter */
00140     blkCnt--;
00141   }
00142 
00143   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.        
00144    ** No loop unrolling is used. */
00145   blkCnt = blockSize % 0x4u;
00146 
00147 #else
00148 
00149   /* Run the below code for Cortex-M0 */
00150 
00151   /* Initialize blkCnt with number of samples */
00152   blkCnt = blockSize;
00153 
00154 #endif /* #ifndef ARM_MATH_CM0_FAMILY */
00155 
00156   while(blkCnt > 0u)
00157   {
00158     /* C = A * scale */
00159     /* Scale the input and then store the result in the destination buffer. */
00160     *pDst++ = (*pSrc++) * scale;
00161 
00162     /* Decrement the loop counter */
00163     blkCnt--;
00164   }
00165 }
00166 
00167 /**        
00168  * @} end of scale group        
00169  */