00001 /* ---------------------------------------------------------------------- 00002 * Copyright (C) 2010 ARM Limited. All rights reserved. 00003 * 00004 * $Date: 15. July 2011 00005 * $Revision: V1.0.10 00006 * 00007 * Project: CMSIS DSP Library 00008 * Title: arm_power_f32.c 00009 * 00010 * Description: Sum of the squares of the elements of a floating-point vector. 00011 * 00012 * Target Processor: Cortex-M4/Cortex-M3/Cortex-M0 00013 * 00014 * Version 1.0.10 2011/7/15 00015 * Big Endian support added and Merged M0 and M3/M4 Source code. 00016 * 00017 * Version 1.0.3 2010/11/29 00018 * Re-organized the CMSIS folders and updated documentation. 00019 * 00020 * Version 1.0.2 2010/11/11 00021 * Documentation updated. 00022 * 00023 * Version 1.0.1 2010/10/05 00024 * Production release and review comments incorporated. 00025 * 00026 * Version 1.0.0 2010/09/20 00027 * Production release and review comments incorporated. 00028 * 00029 * Version 0.0.7 2010/06/10 00030 * Misra-C changes done 00031 * ---------------------------------------------------------------------------- */ 00032 00033 #include "arm_math.h" 00034 00068 void arm_power_f32( 00069 float32_t * pSrc, 00070 uint32_t blockSize, 00071 float32_t * pResult) 00072 { 00073 float32_t sum = 0.0f; /* accumulator */ 00074 float32_t in; /* Temporary variable to store input value */ 00075 uint32_t blkCnt; /* loop counter */ 00076 00077 #ifndef ARM_MATH_CM0 00078 00079 /* Run the below code for Cortex-M4 and Cortex-M3 */ 00080 00081 /*loop Unrolling */ 00082 blkCnt = blockSize >> 2u; 00083 00084 /* First part of the processing with loop unrolling. Compute 4 outputs at a time. 00085 ** a second loop below computes the remaining 1 to 3 samples. */ 00086 while(blkCnt > 0u) 00087 { 00088 /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 00089 /* Compute Power and then store the result in a temporary variable, sum. */ 00090 in = *pSrc++; 00091 sum += in * in; 00092 in = *pSrc++; 00093 sum += in * in; 00094 in = *pSrc++; 00095 sum += in * in; 00096 in = *pSrc++; 00097 sum += in * in; 00098 00099 /* Decrement the loop counter */ 00100 blkCnt--; 00101 } 00102 00103 /* If the blockSize is not a multiple of 4, compute any remaining output samples here. 00104 ** No loop unrolling is used. */ 00105 blkCnt = blockSize % 0x4u; 00106 00107 00108 #else 00109 00110 /* Run the below code for Cortex-M0 */ 00111 00112 /* Loop over blockSize number of values */ 00113 blkCnt = blockSize; 00114 00115 #endif /* #ifndef ARM_MATH_CM0 */ 00116 00117 00118 while(blkCnt > 0u) 00119 { 00120 /* C = A[0] * A[0] + A[1] * A[1] + A[2] * A[2] + ... + A[blockSize-1] * A[blockSize-1] */ 00121 /* compute power and then store the result in a temporary variable, sum. */ 00122 in = *pSrc++; 00123 sum += in * in; 00124 00125 /* Decrement the loop counter */ 00126 blkCnt--; 00127 } 00128 00129 /* Store the result to the destination */ 00130 *pResult = sum; 00131 } 00132