time-to-botec

Benchmark sampling in different programming languages
Log | Files | Refs | README

sapxsumkbn.c (1977B)


      1 /**
      2 * @license Apache-2.0
      3 *
      4 * Copyright (c) 2020 The Stdlib Authors.
      5 *
      6 * Licensed under the Apache License, Version 2.0 (the "License");
      7 * you may not use this file except in compliance with the License.
      8 * You may obtain a copy of the License at
      9 *
     10 *    http://www.apache.org/licenses/LICENSE-2.0
     11 *
     12 * Unless required by applicable law or agreed to in writing, software
     13 * distributed under the License is distributed on an "AS IS" BASIS,
     14 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     15 * See the License for the specific language governing permissions and
     16 * limitations under the License.
     17 */
     18 
     19 #include "stdlib/blas/ext/base/sapxsumkbn.h"
     20 #include <stdint.h>
     21 #include <math.h>
     22 
     23 /**
     24 * Adds a constant to each single-precision floating-point strided array element and computes the sum using an improved Kahan–Babuška algorithm.
     25 *
     26 * ## Method
     27 *
     28 * -   This implementation uses an "improved Kahan–Babuška algorithm", as described by Neumaier (1974).
     29 *
     30 * ## References
     31 *
     32 * -   Neumaier, Arnold. 1974. "Rounding Error Analysis of Some Methods for Summing Finite Sums." _Zeitschrift Für Angewandte Mathematik Und Mechanik_ 54 (1): 39–51. doi:[10.1002/zamm.19740540106](https://doi.org/10.1002/zamm.19740540106).
     33 *
     34 * @param N       number of indexed elements
     35 * @param alpha   constant
     36 * @param X       input array
     37 * @param stride  stride length
     38 * @return        output value
     39 */
     40 float stdlib_strided_sapxsumkbn( const int64_t N, const float alpha, const float *X, const int64_t stride ) {
     41 	int64_t ix;
     42 	int64_t i;
     43 	float sum;
     44 	float v;
     45 	float t;
     46 	float c;
     47 
     48 	if ( N <= 0 ) {
     49 		return 0.0f;
     50 	}
     51 	if ( N == 1 || stride == 0 ) {
     52 		return alpha + X[ 0 ];
     53 	}
     54 	if ( stride < 0 ) {
     55 		ix = (1-N) * stride;
     56 	} else {
     57 		ix = 0;
     58 	}
     59 	sum = 0.0f;
     60 	c = 0.0f;
     61 	for ( i = 0; i < N; i++ ) {
     62 		v = alpha + X[ ix ];
     63 		t = sum + v;
     64 		if ( fabsf( sum ) >= fabsf( v ) ) {
     65 			c += (sum-t) + v;
     66 		} else {
     67 			c += (v-t) + sum;
     68 		}
     69 		sum = t;
     70 		ix += stride;
     71 	}
     72 	return sum + c;
     73 }