FFmpeg
h264qpel_init_aarch64.c
Go to the documentation of this file.
1 /*
2  * ARM NEON optimised DSP functions
3  * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
4  *
5  * This file is part of FFmpeg.
6  *
7  * FFmpeg is free software; you can redistribute it and/or
8  * modify it under the terms of the GNU Lesser General Public
9  * License as published by the Free Software Foundation; either
10  * version 2.1 of the License, or (at your option) any later version.
11  *
12  * FFmpeg 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 GNU
15  * Lesser General Public License for more details.
16  *
17  * You should have received a copy of the GNU Lesser General Public
18  * License along with FFmpeg; if not, write to the Free Software
19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20  */
21 
22 #include <stdint.h>
23 
24 #include "config.h"
25 #include "libavutil/attributes.h"
26 #include "libavutil/cpu.h"
27 #include "libavutil/aarch64/cpu.h"
28 #include "libavcodec/h264qpel.h"
29 
30 void ff_put_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
31 void ff_put_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
32 void ff_put_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
33 void ff_put_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
34 void ff_put_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
35 void ff_put_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
36 void ff_put_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
37 void ff_put_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
38 void ff_put_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
39 void ff_put_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
40 void ff_put_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
41 void ff_put_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
42 void ff_put_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
43 void ff_put_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
44 void ff_put_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
45 void ff_put_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
46 
47 void ff_put_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
48 void ff_put_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
49 void ff_put_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
50 void ff_put_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
51 void ff_put_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
52 void ff_put_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
53 void ff_put_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
54 void ff_put_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
55 void ff_put_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
56 void ff_put_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
57 void ff_put_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
58 void ff_put_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
59 void ff_put_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
60 void ff_put_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
61 void ff_put_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
62 void ff_put_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
63 
64 void ff_avg_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
65 void ff_avg_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
66 void ff_avg_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
67 void ff_avg_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
68 void ff_avg_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
69 void ff_avg_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
70 void ff_avg_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
71 void ff_avg_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
72 void ff_avg_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
73 void ff_avg_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
74 void ff_avg_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
75 void ff_avg_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
76 void ff_avg_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
77 void ff_avg_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
78 void ff_avg_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
79 void ff_avg_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
80 
81 void ff_avg_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
82 void ff_avg_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
83 void ff_avg_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
84 void ff_avg_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
85 void ff_avg_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
86 void ff_avg_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
87 void ff_avg_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
88 void ff_avg_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
89 void ff_avg_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
90 void ff_avg_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
91 void ff_avg_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
92 void ff_avg_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
93 void ff_avg_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
94 void ff_avg_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
95 void ff_avg_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
96 void ff_avg_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
97 
98 void ff_put_h264_qpel16_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
99 void ff_put_h264_qpel16_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
100 void ff_put_h264_qpel16_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
101 void ff_put_h264_qpel16_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
102 void ff_put_h264_qpel16_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
103 void ff_put_h264_qpel16_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
104 void ff_put_h264_qpel16_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
105 void ff_put_h264_qpel16_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
106 void ff_put_h264_qpel16_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
107 void ff_put_h264_qpel16_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
108 
109 void ff_put_h264_qpel8_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
110 void ff_put_h264_qpel8_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
111 void ff_put_h264_qpel8_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
112 void ff_put_h264_qpel8_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
113 void ff_put_h264_qpel8_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
114 void ff_put_h264_qpel8_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
115 void ff_put_h264_qpel8_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
116 void ff_put_h264_qpel8_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
117 void ff_put_h264_qpel8_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
118 void ff_put_h264_qpel8_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
119 
120 void ff_avg_h264_qpel16_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
121 void ff_avg_h264_qpel16_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
122 void ff_avg_h264_qpel16_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
123 void ff_avg_h264_qpel16_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
124 void ff_avg_h264_qpel16_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
125 void ff_avg_h264_qpel16_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
126 void ff_avg_h264_qpel16_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
127 void ff_avg_h264_qpel16_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
128 void ff_avg_h264_qpel16_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
129 void ff_avg_h264_qpel16_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
130 
131 void ff_avg_h264_qpel8_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
132 void ff_avg_h264_qpel8_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
133 void ff_avg_h264_qpel8_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
134 void ff_avg_h264_qpel8_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
135 void ff_avg_h264_qpel8_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
136 void ff_avg_h264_qpel8_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
137 void ff_avg_h264_qpel8_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
138 void ff_avg_h264_qpel8_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
139 void ff_avg_h264_qpel8_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
140 void ff_avg_h264_qpel8_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
141 
143 {
144  int cpu_flags = av_get_cpu_flags();
145 
146  if (have_neon(cpu_flags) && bit_depth <= 8) {
147  c->put_h264_qpel_pixels_tab[0][ 0] = ff_put_h264_qpel16_mc00_neon;
148  c->put_h264_qpel_pixels_tab[0][ 1] = ff_put_h264_qpel16_mc10_neon;
149  c->put_h264_qpel_pixels_tab[0][ 2] = ff_put_h264_qpel16_mc20_neon;
150  c->put_h264_qpel_pixels_tab[0][ 3] = ff_put_h264_qpel16_mc30_neon;
151  c->put_h264_qpel_pixels_tab[0][ 4] = ff_put_h264_qpel16_mc01_neon;
152  c->put_h264_qpel_pixels_tab[0][ 5] = ff_put_h264_qpel16_mc11_neon;
153  c->put_h264_qpel_pixels_tab[0][ 6] = ff_put_h264_qpel16_mc21_neon;
154  c->put_h264_qpel_pixels_tab[0][ 7] = ff_put_h264_qpel16_mc31_neon;
155  c->put_h264_qpel_pixels_tab[0][ 8] = ff_put_h264_qpel16_mc02_neon;
156  c->put_h264_qpel_pixels_tab[0][ 9] = ff_put_h264_qpel16_mc12_neon;
157  c->put_h264_qpel_pixels_tab[0][10] = ff_put_h264_qpel16_mc22_neon;
158  c->put_h264_qpel_pixels_tab[0][11] = ff_put_h264_qpel16_mc32_neon;
159  c->put_h264_qpel_pixels_tab[0][12] = ff_put_h264_qpel16_mc03_neon;
160  c->put_h264_qpel_pixels_tab[0][13] = ff_put_h264_qpel16_mc13_neon;
161  c->put_h264_qpel_pixels_tab[0][14] = ff_put_h264_qpel16_mc23_neon;
162  c->put_h264_qpel_pixels_tab[0][15] = ff_put_h264_qpel16_mc33_neon;
163 
164  c->put_h264_qpel_pixels_tab[1][ 0] = ff_put_h264_qpel8_mc00_neon;
165  c->put_h264_qpel_pixels_tab[1][ 1] = ff_put_h264_qpel8_mc10_neon;
166  c->put_h264_qpel_pixels_tab[1][ 2] = ff_put_h264_qpel8_mc20_neon;
167  c->put_h264_qpel_pixels_tab[1][ 3] = ff_put_h264_qpel8_mc30_neon;
168  c->put_h264_qpel_pixels_tab[1][ 4] = ff_put_h264_qpel8_mc01_neon;
169  c->put_h264_qpel_pixels_tab[1][ 5] = ff_put_h264_qpel8_mc11_neon;
170  c->put_h264_qpel_pixels_tab[1][ 6] = ff_put_h264_qpel8_mc21_neon;
171  c->put_h264_qpel_pixels_tab[1][ 7] = ff_put_h264_qpel8_mc31_neon;
172  c->put_h264_qpel_pixels_tab[1][ 8] = ff_put_h264_qpel8_mc02_neon;
173  c->put_h264_qpel_pixels_tab[1][ 9] = ff_put_h264_qpel8_mc12_neon;
174  c->put_h264_qpel_pixels_tab[1][10] = ff_put_h264_qpel8_mc22_neon;
175  c->put_h264_qpel_pixels_tab[1][11] = ff_put_h264_qpel8_mc32_neon;
176  c->put_h264_qpel_pixels_tab[1][12] = ff_put_h264_qpel8_mc03_neon;
177  c->put_h264_qpel_pixels_tab[1][13] = ff_put_h264_qpel8_mc13_neon;
178  c->put_h264_qpel_pixels_tab[1][14] = ff_put_h264_qpel8_mc23_neon;
179  c->put_h264_qpel_pixels_tab[1][15] = ff_put_h264_qpel8_mc33_neon;
180 
181  c->avg_h264_qpel_pixels_tab[0][ 0] = ff_avg_h264_qpel16_mc00_neon;
182  c->avg_h264_qpel_pixels_tab[0][ 1] = ff_avg_h264_qpel16_mc10_neon;
183  c->avg_h264_qpel_pixels_tab[0][ 2] = ff_avg_h264_qpel16_mc20_neon;
184  c->avg_h264_qpel_pixels_tab[0][ 3] = ff_avg_h264_qpel16_mc30_neon;
185  c->avg_h264_qpel_pixels_tab[0][ 4] = ff_avg_h264_qpel16_mc01_neon;
186  c->avg_h264_qpel_pixels_tab[0][ 5] = ff_avg_h264_qpel16_mc11_neon;
187  c->avg_h264_qpel_pixels_tab[0][ 6] = ff_avg_h264_qpel16_mc21_neon;
188  c->avg_h264_qpel_pixels_tab[0][ 7] = ff_avg_h264_qpel16_mc31_neon;
189  c->avg_h264_qpel_pixels_tab[0][ 8] = ff_avg_h264_qpel16_mc02_neon;
190  c->avg_h264_qpel_pixels_tab[0][ 9] = ff_avg_h264_qpel16_mc12_neon;
191  c->avg_h264_qpel_pixels_tab[0][10] = ff_avg_h264_qpel16_mc22_neon;
192  c->avg_h264_qpel_pixels_tab[0][11] = ff_avg_h264_qpel16_mc32_neon;
193  c->avg_h264_qpel_pixels_tab[0][12] = ff_avg_h264_qpel16_mc03_neon;
194  c->avg_h264_qpel_pixels_tab[0][13] = ff_avg_h264_qpel16_mc13_neon;
195  c->avg_h264_qpel_pixels_tab[0][14] = ff_avg_h264_qpel16_mc23_neon;
196  c->avg_h264_qpel_pixels_tab[0][15] = ff_avg_h264_qpel16_mc33_neon;
197 
198  c->avg_h264_qpel_pixels_tab[1][ 0] = ff_avg_h264_qpel8_mc00_neon;
199  c->avg_h264_qpel_pixels_tab[1][ 1] = ff_avg_h264_qpel8_mc10_neon;
200  c->avg_h264_qpel_pixels_tab[1][ 2] = ff_avg_h264_qpel8_mc20_neon;
201  c->avg_h264_qpel_pixels_tab[1][ 3] = ff_avg_h264_qpel8_mc30_neon;
202  c->avg_h264_qpel_pixels_tab[1][ 4] = ff_avg_h264_qpel8_mc01_neon;
203  c->avg_h264_qpel_pixels_tab[1][ 5] = ff_avg_h264_qpel8_mc11_neon;
204  c->avg_h264_qpel_pixels_tab[1][ 6] = ff_avg_h264_qpel8_mc21_neon;
205  c->avg_h264_qpel_pixels_tab[1][ 7] = ff_avg_h264_qpel8_mc31_neon;
206  c->avg_h264_qpel_pixels_tab[1][ 8] = ff_avg_h264_qpel8_mc02_neon;
207  c->avg_h264_qpel_pixels_tab[1][ 9] = ff_avg_h264_qpel8_mc12_neon;
208  c->avg_h264_qpel_pixels_tab[1][10] = ff_avg_h264_qpel8_mc22_neon;
209  c->avg_h264_qpel_pixels_tab[1][11] = ff_avg_h264_qpel8_mc32_neon;
210  c->avg_h264_qpel_pixels_tab[1][12] = ff_avg_h264_qpel8_mc03_neon;
211  c->avg_h264_qpel_pixels_tab[1][13] = ff_avg_h264_qpel8_mc13_neon;
212  c->avg_h264_qpel_pixels_tab[1][14] = ff_avg_h264_qpel8_mc23_neon;
213  c->avg_h264_qpel_pixels_tab[1][15] = ff_avg_h264_qpel8_mc33_neon;
214  } else if (have_neon(cpu_flags) && bit_depth == 10) {
215  c->put_h264_qpel_pixels_tab[0][ 1] = ff_put_h264_qpel16_mc10_neon_10;
216  c->put_h264_qpel_pixels_tab[0][ 2] = ff_put_h264_qpel16_mc20_neon_10;
217  c->put_h264_qpel_pixels_tab[0][ 3] = ff_put_h264_qpel16_mc30_neon_10;
218  c->put_h264_qpel_pixels_tab[0][ 4] = ff_put_h264_qpel16_mc01_neon_10;
219  c->put_h264_qpel_pixels_tab[0][ 5] = ff_put_h264_qpel16_mc11_neon_10;
220  c->put_h264_qpel_pixels_tab[0][ 7] = ff_put_h264_qpel16_mc31_neon_10;
221  c->put_h264_qpel_pixels_tab[0][ 8] = ff_put_h264_qpel16_mc02_neon_10;
222  c->put_h264_qpel_pixels_tab[0][12] = ff_put_h264_qpel16_mc03_neon_10;
223  c->put_h264_qpel_pixels_tab[0][13] = ff_put_h264_qpel16_mc13_neon_10;
224  c->put_h264_qpel_pixels_tab[0][15] = ff_put_h264_qpel16_mc33_neon_10;
225 
226  c->put_h264_qpel_pixels_tab[1][ 1] = ff_put_h264_qpel8_mc10_neon_10;
227  c->put_h264_qpel_pixels_tab[1][ 2] = ff_put_h264_qpel8_mc20_neon_10;
228  c->put_h264_qpel_pixels_tab[1][ 3] = ff_put_h264_qpel8_mc30_neon_10;
229  c->put_h264_qpel_pixels_tab[1][ 4] = ff_put_h264_qpel8_mc01_neon_10;
230  c->put_h264_qpel_pixels_tab[1][ 5] = ff_put_h264_qpel8_mc11_neon_10;
231  c->put_h264_qpel_pixels_tab[1][ 7] = ff_put_h264_qpel8_mc31_neon_10;
232  c->put_h264_qpel_pixels_tab[1][ 8] = ff_put_h264_qpel8_mc02_neon_10;
233  c->put_h264_qpel_pixels_tab[1][12] = ff_put_h264_qpel8_mc03_neon_10;
234  c->put_h264_qpel_pixels_tab[1][13] = ff_put_h264_qpel8_mc13_neon_10;
235  c->put_h264_qpel_pixels_tab[1][15] = ff_put_h264_qpel8_mc33_neon_10;
236 
237  c->avg_h264_qpel_pixels_tab[0][ 1] = ff_avg_h264_qpel16_mc10_neon_10;
238  c->avg_h264_qpel_pixels_tab[0][ 2] = ff_avg_h264_qpel16_mc20_neon_10;
239  c->avg_h264_qpel_pixels_tab[0][ 3] = ff_avg_h264_qpel16_mc30_neon_10;
240  c->avg_h264_qpel_pixels_tab[0][ 4] = ff_avg_h264_qpel16_mc01_neon_10;
241  c->avg_h264_qpel_pixels_tab[0][ 5] = ff_avg_h264_qpel16_mc11_neon_10;
242  c->avg_h264_qpel_pixels_tab[0][ 7] = ff_avg_h264_qpel16_mc31_neon_10;
243  c->avg_h264_qpel_pixels_tab[0][ 8] = ff_avg_h264_qpel16_mc02_neon_10;
244  c->avg_h264_qpel_pixels_tab[0][12] = ff_avg_h264_qpel16_mc03_neon_10;
245  c->avg_h264_qpel_pixels_tab[0][13] = ff_avg_h264_qpel16_mc13_neon_10;
246  c->avg_h264_qpel_pixels_tab[0][15] = ff_avg_h264_qpel16_mc33_neon_10;
247 
248  c->avg_h264_qpel_pixels_tab[1][ 1] = ff_avg_h264_qpel8_mc10_neon_10;
249  c->avg_h264_qpel_pixels_tab[1][ 2] = ff_avg_h264_qpel8_mc20_neon_10;
250  c->avg_h264_qpel_pixels_tab[1][ 3] = ff_avg_h264_qpel8_mc30_neon_10;
251  c->avg_h264_qpel_pixels_tab[1][ 4] = ff_avg_h264_qpel8_mc01_neon_10;
252  c->avg_h264_qpel_pixels_tab[1][ 5] = ff_avg_h264_qpel8_mc11_neon_10;
253  c->avg_h264_qpel_pixels_tab[1][ 7] = ff_avg_h264_qpel8_mc31_neon_10;
254  c->avg_h264_qpel_pixels_tab[1][ 8] = ff_avg_h264_qpel8_mc02_neon_10;
255  c->avg_h264_qpel_pixels_tab[1][12] = ff_avg_h264_qpel8_mc03_neon_10;
256  c->avg_h264_qpel_pixels_tab[1][13] = ff_avg_h264_qpel8_mc13_neon_10;
257  c->avg_h264_qpel_pixels_tab[1][15] = ff_avg_h264_qpel8_mc33_neon_10;
258  }
259 }
ff_avg_h264_qpel8_mc33_neon_10
void ff_avg_h264_qpel8_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc03_neon_10
void ff_put_h264_qpel16_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc13_neon
void ff_avg_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc03_neon
void ff_put_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc23_neon
void ff_put_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc30_neon
void ff_avg_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc13_neon
void ff_put_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc31_neon
void ff_avg_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc33_neon_10
void ff_put_h264_qpel16_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc22_neon
void ff_put_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc33_neon_10
void ff_put_h264_qpel8_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
av_get_cpu_flags
int av_get_cpu_flags(void)
Return the flags which specify extensions supported by the CPU.
Definition: cpu.c:109
ff_avg_h264_qpel8_mc00_neon
void ff_avg_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
bit_depth
static void bit_depth(AudioStatsContext *s, const uint64_t *const mask, uint8_t *depth)
Definition: af_astats.c:246
cpu_flags
static atomic_int cpu_flags
Definition: cpu.c:56
ff_avg_h264_qpel16_mc03_neon_10
void ff_avg_h264_qpel16_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc10_neon_10
void ff_avg_h264_qpel8_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc21_neon
void ff_avg_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc33_neon
void ff_avg_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
h264qpel.h
ff_avg_h264_qpel8_mc30_neon
void ff_avg_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc02_neon_10
void ff_put_h264_qpel8_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc11_neon
void ff_put_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc21_neon
void ff_put_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc02_neon_10
void ff_put_h264_qpel16_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc21_neon
void ff_avg_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc11_neon
void ff_avg_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc31_neon
void ff_put_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc20_neon
void ff_avg_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc30_neon
void ff_put_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc10_neon_10
void ff_put_h264_qpel16_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
av_cold
#define av_cold
Definition: attributes.h:90
ff_avg_h264_qpel8_mc02_neon_10
void ff_avg_h264_qpel8_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc01_neon_10
void ff_put_h264_qpel8_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc20_neon
void ff_avg_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc31_neon_10
void ff_put_h264_qpel8_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc32_neon
void ff_put_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc12_neon
void ff_avg_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc30_neon_10
void ff_avg_h264_qpel16_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc03_neon
void ff_avg_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc20_neon_10
void ff_put_h264_qpel16_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc10_neon
void ff_avg_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc01_neon
void ff_put_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc11_neon
void ff_put_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc33_neon
void ff_avg_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc11_neon_10
void ff_put_h264_qpel16_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc31_neon_10
void ff_avg_h264_qpel16_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc11_neon_10
void ff_put_h264_qpel8_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc01_neon_10
void ff_avg_h264_qpel8_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc13_neon_10
void ff_avg_h264_qpel8_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc01_neon_10
void ff_avg_h264_qpel16_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc11_neon_10
void ff_avg_h264_qpel8_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc20_neon_10
void ff_avg_h264_qpel8_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc13_neon
void ff_put_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc02_neon
void ff_avg_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc13_neon
void ff_avg_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc30_neon_10
void ff_avg_h264_qpel8_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
c
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
Definition: undefined.txt:32
ff_put_h264_qpel8_mc00_neon
void ff_put_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc31_neon
void ff_put_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc11_neon
void ff_avg_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc31_neon_10
void ff_avg_h264_qpel8_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc10_neon
void ff_put_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc12_neon
void ff_put_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc10_neon_10
void ff_put_h264_qpel8_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
dst
uint8_t ptrdiff_t const uint8_t ptrdiff_t int intptr_t intptr_t int int16_t * dst
Definition: dsp.h:83
cpu.h
ff_avg_h264_qpel16_mc32_neon
void ff_avg_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc22_neon
void ff_put_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc00_neon
void ff_avg_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc31_neon
void ff_avg_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc20_neon_10
void ff_avg_h264_qpel16_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
have_neon
#define have_neon(flags)
Definition: cpu.h:26
ff_avg_h264_qpel16_mc10_neon_10
void ff_avg_h264_qpel16_mc10_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc01_neon
void ff_avg_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
attributes.h
ff_put_h264_qpel16_mc01_neon_10
void ff_put_h264_qpel16_mc01_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc12_neon
void ff_avg_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc03_neon_10
void ff_put_h264_qpel8_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc20_neon
void ff_put_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc30_neon
void ff_put_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc20_neon
void ff_put_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc11_neon_10
void ff_avg_h264_qpel16_mc11_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_h264qpel_init_aarch64
av_cold void ff_h264qpel_init_aarch64(H264QpelContext *c, int bit_depth)
Definition: h264qpel_init_aarch64.c:142
ff_put_h264_qpel8_mc23_neon
void ff_put_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
H264QpelContext
Definition: h264qpel.h:27
stride
#define stride
Definition: h264pred_template.c:537
ff_put_h264_qpel8_mc13_neon_10
void ff_put_h264_qpel8_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc02_neon_10
void ff_avg_h264_qpel16_mc02_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc33_neon
void ff_put_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc02_neon
void ff_avg_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc02_neon
void ff_put_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc21_neon
void ff_put_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc00_neon
void ff_put_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc10_neon
void ff_avg_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc33_neon_10
void ff_avg_h264_qpel16_mc33_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc23_neon
void ff_avg_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc03_neon_10
void ff_avg_h264_qpel8_mc03_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc22_neon
void ff_avg_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc10_neon
void ff_put_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc31_neon_10
void ff_put_h264_qpel16_mc31_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc13_neon_10
void ff_put_h264_qpel16_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc01_neon
void ff_put_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc33_neon
void ff_put_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc23_neon
void ff_avg_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc03_neon
void ff_put_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc32_neon
void ff_avg_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel16_mc30_neon_10
void ff_put_h264_qpel16_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc32_neon
void ff_put_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
cpu.h
ff_put_h264_qpel8_mc20_neon_10
void ff_put_h264_qpel8_mc20_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc02_neon
void ff_put_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc03_neon
void ff_avg_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
src
#define src
Definition: vp8dsp.c:248
ff_avg_h264_qpel16_mc13_neon_10
void ff_avg_h264_qpel16_mc13_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel8_mc22_neon
void ff_avg_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc12_neon
void ff_put_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_put_h264_qpel8_mc30_neon_10
void ff_put_h264_qpel8_mc30_neon_10(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
ff_avg_h264_qpel16_mc01_neon
void ff_avg_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)