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 <stddef.h>
23 #include <stdint.h>
24
25 #include "config.h"
26
27 #include "libavutil/attributes.h"
28 #include "libavutil/cpu.h"
29 #include "libavutil/aarch64/cpu.h"
30 #include "libavcodec/hpeldsp.h"
31
32 void ff_put_pixels16_neon(uint8_t *block, const uint8_t *pixels,
33 ptrdiff_t line_size, int h);
34 void ff_put_pixels16_x2_neon(uint8_t *block, const uint8_t *pixels,
35 ptrdiff_t line_size, int h);
36 void ff_put_pixels16_y2_neon(uint8_t *block, const uint8_t *pixels,
37 ptrdiff_t line_size, int h);
38 void ff_put_pixels16_xy2_neon(uint8_t *block, const uint8_t *pixels,
39 ptrdiff_t line_size, int h);
40 void ff_put_pixels8_neon(uint8_t *block, const uint8_t *pixels,
41 ptrdiff_t line_size, int h);
42 void ff_put_pixels8_x2_neon(uint8_t *block, const uint8_t *pixels,
43 ptrdiff_t line_size, int h);
44 void ff_put_pixels8_y2_neon(uint8_t *block, const uint8_t *pixels,
45 ptrdiff_t line_size, int h);
46 void ff_put_pixels8_xy2_neon(uint8_t *block, const uint8_t *pixels,
47 ptrdiff_t line_size, int h);
48
49 void ff_put_pixels16_x2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
50 ptrdiff_t line_size, int h);
51 void ff_put_pixels16_y2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
52 ptrdiff_t line_size, int h);
53 void ff_put_pixels16_xy2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
54 ptrdiff_t line_size, int h);
55 void ff_put_pixels8_x2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
56 ptrdiff_t line_size, int h);
57 void ff_put_pixels8_y2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
58 ptrdiff_t line_size, int h);
59 void ff_put_pixels8_xy2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
60 ptrdiff_t line_size, int h);
61
62 void ff_avg_pixels16_neon(uint8_t *block, const uint8_t *pixels,
63 ptrdiff_t line_size, int h);
64 void ff_avg_pixels16_x2_neon(uint8_t *block, const uint8_t *pixels,
65 ptrdiff_t line_size, int h);
66 void ff_avg_pixels16_y2_neon(uint8_t *block, const uint8_t *pixels,
67 ptrdiff_t line_size, int h);
68 void ff_avg_pixels16_xy2_neon(uint8_t *block, const uint8_t *pixels,
69 ptrdiff_t line_size, int h);
70 void ff_avg_pixels8_neon(uint8_t *block, const uint8_t *pixels,
71 ptrdiff_t line_size, int h);
72 void ff_avg_pixels8_x2_neon(uint8_t *block, const uint8_t *pixels,
73 ptrdiff_t line_size, int h);
74 void ff_avg_pixels8_y2_neon(uint8_t *block, const uint8_t *pixels,
75 ptrdiff_t line_size, int h);
76 void ff_avg_pixels8_xy2_neon(uint8_t *block, const uint8_t *pixels,
77 ptrdiff_t line_size, int h);
78
79 void ff_avg_pixels16_x2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
80 ptrdiff_t line_size, int h);
81 void ff_avg_pixels16_y2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
82 ptrdiff_t line_size, int h);
83 void ff_avg_pixels16_xy2_no_rnd_neon(uint8_t *block, const uint8_t *pixels,
84 ptrdiff_t line_size, int h);
85
ff_hpeldsp_init_aarch64(HpelDSPContext * c,int flags)86 av_cold void ff_hpeldsp_init_aarch64(HpelDSPContext *c, int flags)
87 {
88 int cpu_flags = av_get_cpu_flags();
89
90 if (have_neon(cpu_flags)) {
91 c->put_pixels_tab[0][0] = ff_put_pixels16_neon;
92 c->put_pixels_tab[0][1] = ff_put_pixels16_x2_neon;
93 c->put_pixels_tab[0][2] = ff_put_pixels16_y2_neon;
94 c->put_pixels_tab[0][3] = ff_put_pixels16_xy2_neon;
95 c->put_pixels_tab[1][0] = ff_put_pixels8_neon;
96 c->put_pixels_tab[1][1] = ff_put_pixels8_x2_neon;
97 c->put_pixels_tab[1][2] = ff_put_pixels8_y2_neon;
98 c->put_pixels_tab[1][3] = ff_put_pixels8_xy2_neon;
99
100 c->put_no_rnd_pixels_tab[0][0] = ff_put_pixels16_neon;
101 c->put_no_rnd_pixels_tab[0][1] = ff_put_pixels16_x2_no_rnd_neon;
102 c->put_no_rnd_pixels_tab[0][2] = ff_put_pixels16_y2_no_rnd_neon;
103 c->put_no_rnd_pixels_tab[0][3] = ff_put_pixels16_xy2_no_rnd_neon;
104 c->put_no_rnd_pixels_tab[1][0] = ff_put_pixels8_neon;
105 c->put_no_rnd_pixels_tab[1][1] = ff_put_pixels8_x2_no_rnd_neon;
106 c->put_no_rnd_pixels_tab[1][2] = ff_put_pixels8_y2_no_rnd_neon;
107 c->put_no_rnd_pixels_tab[1][3] = ff_put_pixels8_xy2_no_rnd_neon;
108
109 c->avg_pixels_tab[0][0] = ff_avg_pixels16_neon;
110 c->avg_pixels_tab[0][1] = ff_avg_pixels16_x2_neon;
111 c->avg_pixels_tab[0][2] = ff_avg_pixels16_y2_neon;
112 c->avg_pixels_tab[0][3] = ff_avg_pixels16_xy2_neon;
113 c->avg_pixels_tab[1][0] = ff_avg_pixels8_neon;
114 c->avg_pixels_tab[1][1] = ff_avg_pixels8_x2_neon;
115 c->avg_pixels_tab[1][2] = ff_avg_pixels8_y2_neon;
116 c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_neon;
117
118 c->avg_no_rnd_pixels_tab[0] = ff_avg_pixels16_neon;
119 c->avg_no_rnd_pixels_tab[1] = ff_avg_pixels16_x2_no_rnd_neon;
120 c->avg_no_rnd_pixels_tab[2] = ff_avg_pixels16_y2_no_rnd_neon;
121 c->avg_no_rnd_pixels_tab[3] = ff_avg_pixels16_xy2_no_rnd_neon;
122 }
123 }
124