Vector Optimized Library of Kernels  2.5.1
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 
35 
36 
37 static inline void sanitize_bytes(unsigned char* u, const int elements)
38 {
39  int i;
40  unsigned char* u_ptr = u;
41  for (i = 0; i < elements; i++) {
42  *u_ptr = (*u_ptr & 0x01);
43  u_ptr++;
44  }
45 }
46 
47 static inline void clean_up_intermediate_values(float* llrs,
48  unsigned char* u,
49  const int frame_size,
50  const int elements)
51 {
52  memset(u + frame_size, 0, sizeof(unsigned char) * (elements - frame_size));
53  memset(llrs + frame_size, 0, sizeof(float) * (elements - frame_size));
54 }
55 
56 static inline void
57 generate_error_free_input_vector(float* llrs, unsigned char* u, const int frame_size)
58 {
59  memset(u, 0, frame_size);
60  unsigned char* target = u + frame_size;
61  volk_8u_x2_encodeframepolar_8u_generic(target, u + 2 * frame_size, frame_size);
62  float* ft = llrs;
63  int i;
64  for (i = 0; i < frame_size; i++) {
65  *ft = (-2 * ((float)*target++)) + 1.0f;
66  ft++;
67  }
68 }
69 
70 static inline void
71 print_llr_tree(const float* llrs, const int frame_size, const int frame_exp)
72 {
73  int s, e;
74  for (s = 0; s < frame_size; s++) {
75  for (e = 0; e < frame_exp + 1; e++) {
76  printf("%+4.2f ", llrs[e * frame_size + s]);
77  }
78  printf("\n");
79  if ((s + 1) % 8 == 0) {
80  printf("\n");
81  }
82  }
83 }
84 
85 static inline int maximum_frame_size(const int elements)
86 {
87  unsigned int frame_size = next_lower_power_of_two(elements);
88  unsigned int frame_exp = log2_of_power_of_2(frame_size);
89  return next_lower_power_of_two(frame_size / frame_exp);
90 }
91 
92 #ifdef LV_HAVE_GENERIC
93 static inline void volk_32f_8u_polarbutterflypuppet_32f_generic(float* llrs,
94  const float* input,
95  unsigned char* u,
96  const int elements)
97 {
98  unsigned int frame_size = maximum_frame_size(elements);
99  unsigned int frame_exp = log2_of_power_of_2(frame_size);
100 
101  sanitize_bytes(u, elements);
102  clean_up_intermediate_values(llrs, u, frame_size, elements);
103  generate_error_free_input_vector(llrs + frame_exp * frame_size, u, frame_size);
104 
105  unsigned int u_num = 0;
106  for (; u_num < frame_size; u_num++) {
107  volk_32f_8u_polarbutterfly_32f_generic(llrs, u, frame_exp, 0, u_num, u_num);
108  u[u_num] = llrs[u_num] > 0 ? 0 : 1;
109  }
110 
111  clean_up_intermediate_values(llrs, u, frame_size, elements);
112 }
113 #endif /* LV_HAVE_GENERIC */
114 
115 #ifdef LV_HAVE_AVX
116 static inline void volk_32f_8u_polarbutterflypuppet_32f_u_avx(float* llrs,
117  const float* input,
118  unsigned char* u,
119  const int elements)
120 {
121  unsigned int frame_size = maximum_frame_size(elements);
122  unsigned int frame_exp = log2_of_power_of_2(frame_size);
123 
124  sanitize_bytes(u, elements);
125  clean_up_intermediate_values(llrs, u, frame_size, elements);
126  generate_error_free_input_vector(llrs + frame_exp * frame_size, u, frame_size);
127 
128  unsigned int u_num = 0;
129  for (; u_num < frame_size; u_num++) {
130  volk_32f_8u_polarbutterfly_32f_u_avx(llrs, u, frame_exp, 0, u_num, u_num);
131  u[u_num] = llrs[u_num] > 0 ? 0 : 1;
132  }
133 
134  clean_up_intermediate_values(llrs, u, frame_size, elements);
135 }
136 #endif /* LV_HAVE_AVX */
137 
138 #ifdef LV_HAVE_AVX2
139 static inline void volk_32f_8u_polarbutterflypuppet_32f_u_avx2(float* llrs,
140  const float* input,
141  unsigned char* u,
142  const int elements)
143 {
144  unsigned int frame_size = maximum_frame_size(elements);
145  unsigned int frame_exp = log2_of_power_of_2(frame_size);
146 
147  sanitize_bytes(u, elements);
148  clean_up_intermediate_values(llrs, u, frame_size, elements);
149  generate_error_free_input_vector(llrs + frame_exp * frame_size, u, frame_size);
150 
151  unsigned int u_num = 0;
152  for (; u_num < frame_size; u_num++) {
153  volk_32f_8u_polarbutterfly_32f_u_avx2(llrs, u, frame_exp, 0, u_num, u_num);
154  u[u_num] = llrs[u_num] > 0 ? 0 : 1;
155  }
156 
157  clean_up_intermediate_values(llrs, u, frame_size, elements);
158 }
159 #endif /* LV_HAVE_AVX2 */
160 
161 
162 #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:47
static void print_llr_tree(const float *llrs, const int frame_size, const int frame_exp)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:71
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:93
static int maximum_frame_size(const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:85
static void generate_error_free_input_vector(float *llrs, unsigned char *u, const int frame_size)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:57
static void sanitize_bytes(unsigned char *u, const int elements)
Definition: volk_32f_8u_polarbutterflypuppet_32f.h:37
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:116
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:63
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