Vector Optimized Library of Kernels  2.5.0
Architecture-tuned implementations of math kernels
volk_32f_8u_polarbutterflypuppet_32f.h
Go to the documentation of this file.
1 /* -*- c++ -*- */
2 /*
3  * Copyright 2015 Free Software Foundation, Inc.
4  *
5  * This file is part of GNU Radio
6  *
7  * GNU Radio is free software; you can redistribute it and/or modify
8  * it under the terms of the GNU General Public License as published by
9  * the Free Software Foundation; either version 3, or (at your option)
10  * any later version.
11  *
12  * GNU Radio is distributed in the hope that it will be useful,
13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15  * GNU General Public License for more details.
16  *
17  * You should have received a copy of the GNU General Public License
18  * along with GNU Radio; see the file COPYING. If not, write to
19  * the Free Software Foundation, Inc., 51 Franklin Street,
20  * Boston, MA 02110-1301, USA.
21  */
22 
23 /*
24  * This puppet is for VOLK tests only.
25  * For documentation see 'kernels/volk/volk_32f_8u_polarbutterfly_32f.h'
26  */
27 
28 #ifndef VOLK_KERNELS_VOLK_VOLK_32F_8U_POLARBUTTERFLYPUPPET_32F_H_
29 #define VOLK_KERNELS_VOLK_VOLK_32F_8U_POLARBUTTERFLYPUPPET_32F_H_
30 
34 
35 
36 static inline void sanitize_bytes(unsigned char* u, const int elements)
37 {
38  int i;
39  unsigned char* u_ptr = u;
40  for (i = 0; i < elements; i++) {
41  *u_ptr = (*u_ptr & 0x01);
42  u_ptr++;
43  }
44 }
45 
46 static inline void clean_up_intermediate_values(float* llrs,
47  unsigned char* u,
48  const int frame_size,
49  const int elements)
50 {
51  memset(u + frame_size, 0, sizeof(unsigned char) * (elements - frame_size));
52  memset(llrs + frame_size, 0, sizeof(float) * (elements - frame_size));
53 }
54 
55 static inline void
56 generate_error_free_input_vector(float* llrs, unsigned char* u, const int frame_size)
57 {
58  memset(u, 0, frame_size);
59  unsigned char* target = u + frame_size;
60  volk_8u_x2_encodeframepolar_8u_generic(target, u + 2 * frame_size, frame_size);
61  float* ft = llrs;
62  int i;
63  for (i = 0; i < frame_size; i++) {
64  *ft = (-2 * ((float)*target++)) + 1.0f;
65  ft++;
66  }
67 }
68 
69 static inline void
70 print_llr_tree(const float* llrs, const int frame_size, const int frame_exp)
71 {
72  int s, e;
73  for (s = 0; s < frame_size; s++) {
74  for (e = 0; e < frame_exp + 1; e++) {
75  printf("%+4.2f ", llrs[e * frame_size + s]);
76  }
77  printf("\n");
78  if ((s + 1) % 8 == 0) {
79  printf("\n");
80  }
81  }
82 }
83 
84 static inline int maximum_frame_size(const int elements)
85 {
86  unsigned int frame_size = next_lower_power_of_two(elements);
87  unsigned int frame_exp = log2_of_power_of_2(frame_size);
88  return next_lower_power_of_two(frame_size / frame_exp);
89 }
90 
91 #ifdef LV_HAVE_GENERIC
92 static inline void volk_32f_8u_polarbutterflypuppet_32f_generic(float* llrs,
93  const float* input,
94  unsigned char* u,
95  const int elements)
96 {
97  unsigned int frame_size = maximum_frame_size(elements);
98  unsigned int frame_exp = log2_of_power_of_2(frame_size);
99 
100  sanitize_bytes(u, elements);
101  clean_up_intermediate_values(llrs, u, frame_size, elements);
102  generate_error_free_input_vector(llrs + frame_exp * frame_size, u, frame_size);
103 
104  unsigned int u_num = 0;
105  for (; u_num < frame_size; u_num++) {
106  volk_32f_8u_polarbutterfly_32f_generic(llrs, u, frame_exp, 0, u_num, u_num);
107  u[u_num] = llrs[u_num] > 0 ? 0 : 1;
108  }
109 
110  clean_up_intermediate_values(llrs, u, frame_size, elements);
111 }
112 #endif /* LV_HAVE_GENERIC */
113 
114 #ifdef LV_HAVE_AVX
115 static inline void volk_32f_8u_polarbutterflypuppet_32f_u_avx(float* llrs,
116  const float* input,
117  unsigned char* u,
118  const int elements)
119 {
120  unsigned int frame_size = maximum_frame_size(elements);
121  unsigned int frame_exp = log2_of_power_of_2(frame_size);
122 
123  sanitize_bytes(u, elements);
124  clean_up_intermediate_values(llrs, u, frame_size, elements);
125  generate_error_free_input_vector(llrs + frame_exp * frame_size, u, frame_size);
126 
127  unsigned int u_num = 0;
128  for (; u_num < frame_size; u_num++) {
129  volk_32f_8u_polarbutterfly_32f_u_avx(llrs, u, frame_exp, 0, u_num, u_num);
130  u[u_num] = llrs[u_num] > 0 ? 0 : 1;
131  }
132 
133  clean_up_intermediate_values(llrs, u, frame_size, elements);
134 }
135 #endif /* LV_HAVE_AVX */
136 
137 #ifdef LV_HAVE_AVX2
138 static inline void volk_32f_8u_polarbutterflypuppet_32f_u_avx2(float* llrs,
139  const float* input,
140  unsigned char* u,
141  const int elements)
142 {
143  unsigned int frame_size = maximum_frame_size(elements);
144  unsigned int frame_exp = log2_of_power_of_2(frame_size);
145 
146  sanitize_bytes(u, elements);
147  clean_up_intermediate_values(llrs, u, frame_size, elements);
148  generate_error_free_input_vector(llrs + frame_exp * frame_size, u, frame_size);
149 
150  unsigned int u_num = 0;
151  for (; u_num < frame_size; u_num++) {
152  volk_32f_8u_polarbutterfly_32f_u_avx2(llrs, u, frame_exp, 0, u_num, u_num);
153  u[u_num] = llrs[u_num] > 0 ? 0 : 1;
154  }
155 
156  clean_up_intermediate_values(llrs, u, frame_size, elements);
157 }
158 #endif /* LV_HAVE_AVX2 */
159 
160 
161 #endif /* VOLK_KERNELS_VOLK_VOLK_32F_8U_POLARBUTTERFLYPUPPET_32F_H_ */
static void volk_32f_8u_polarbutterfly_32f_generic(float *llrs, unsigned char *u, const int frame_exp, const int stage, const int u_num, const int row)
Definition: volk_32f_8u_polarbutterfly_32f.h:157
static void volk_32f_8u_polarbutterfly_32f_u_avx(float *llrs, unsigned char *u, const int frame_exp, const int stage, const int u_num, const int row)
Definition: volk_32f_8u_polarbutterfly_32f.h:213
static void clean_up_intermediate_values(float *llrs, unsigned char *u, const int frame_size, const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:46
static void print_llr_tree(const float *llrs, const int frame_size, const int frame_exp)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:70
static void volk_32f_8u_polarbutterflypuppet_32f_generic(float *llrs, const float *input, unsigned char *u, const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:92
static int maximum_frame_size(const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:84
static void generate_error_free_input_vector(float *llrs, unsigned char *u, const int frame_size)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:56
static void sanitize_bytes(unsigned char *u, const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:36
static void volk_32f_8u_polarbutterflypuppet_32f_u_avx(float *llrs, const float *input, unsigned char *u, const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:115
static void volk_8u_x2_encodeframepolar_8u_generic(unsigned char *frame, unsigned char *temp, unsigned int frame_size)
Definition: volk_8u_x2_encodeframepolar_8u.h:65
static unsigned int log2_of_power_of_2(unsigned int val)
Definition: volk_8u_x2_encodeframepolar_8u.h:31
static unsigned int next_lower_power_of_two(const unsigned int val)
Definition: volk_8u_x3_encodepolarpuppet_8u.h:32
for i
Definition: volk_config_fixed.tmpl.h:25