8334
|
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 "libavcodec/avcodec.h"
|
|
25 #include "libavcodec/dsputil.h"
|
|
26
|
|
27 void ff_put_pixels16_neon(uint8_t *, const uint8_t *, int, int);
|
|
28 void ff_put_pixels16_x2_neon(uint8_t *, const uint8_t *, int, int);
|
|
29 void ff_put_pixels16_y2_neon(uint8_t *, const uint8_t *, int, int);
|
|
30 void ff_put_pixels16_xy2_neon(uint8_t *, const uint8_t *, int, int);
|
|
31 void ff_put_pixels8_neon(uint8_t *, const uint8_t *, int, int);
|
|
32 void ff_put_pixels8_x2_neon(uint8_t *, const uint8_t *, int, int);
|
|
33 void ff_put_pixels8_y2_neon(uint8_t *, const uint8_t *, int, int);
|
|
34 void ff_put_pixels8_xy2_neon(uint8_t *, const uint8_t *, int, int);
|
|
35 void ff_put_pixels16_x2_no_rnd_neon(uint8_t *, const uint8_t *, int, int);
|
|
36 void ff_put_pixels16_y2_no_rnd_neon(uint8_t *, const uint8_t *, int, int);
|
|
37 void ff_put_pixels16_xy2_no_rnd_neon(uint8_t *, const uint8_t *, int, int);
|
|
38 void ff_put_pixels8_x2_no_rnd_neon(uint8_t *, const uint8_t *, int, int);
|
|
39 void ff_put_pixels8_y2_no_rnd_neon(uint8_t *, const uint8_t *, int, int);
|
|
40 void ff_put_pixels8_xy2_no_rnd_neon(uint8_t *, const uint8_t *, int, int);
|
|
41
|
|
42 void ff_avg_pixels16_neon(uint8_t *, const uint8_t *, int, int);
|
|
43
|
|
44 void ff_put_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
|
8338
|
45 void ff_put_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
|
|
46 void ff_put_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
|
|
47 void ff_put_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
|
|
48 void ff_put_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
|
|
49 void ff_put_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
|
|
50 void ff_put_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
|
|
51 void ff_put_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
|
|
52 void ff_put_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
|
|
53 void ff_put_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
|
|
54 void ff_put_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
|
|
55 void ff_put_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
|
|
56 void ff_put_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
|
|
57 void ff_put_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
|
|
58 void ff_put_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
|
|
59 void ff_put_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
|
|
60
|
8334
|
61 void ff_put_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
|
8338
|
62 void ff_put_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
|
|
63 void ff_put_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
|
|
64 void ff_put_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
|
|
65 void ff_put_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
|
|
66 void ff_put_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
|
|
67 void ff_put_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
|
|
68 void ff_put_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
|
|
69 void ff_put_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
|
|
70 void ff_put_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
|
|
71 void ff_put_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
|
|
72 void ff_put_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
|
|
73 void ff_put_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
|
|
74 void ff_put_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
|
|
75 void ff_put_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
|
|
76 void ff_put_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
|
8334
|
77
|
|
78 void ff_avg_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
|
|
79
|
8336
|
80 void ff_put_h264_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int);
|
|
81 void ff_put_h264_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int);
|
|
82
|
|
83 void ff_avg_h264_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int);
|
|
84 void ff_avg_h264_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int);
|
|
85
|
8337
|
86 void ff_h264_v_loop_filter_luma_neon(uint8_t *pix, int stride, int alpha,
|
|
87 int beta, int8_t *tc0);
|
|
88 void ff_h264_h_loop_filter_luma_neon(uint8_t *pix, int stride, int alpha,
|
|
89 int beta, int8_t *tc0);
|
|
90 void ff_h264_v_loop_filter_chroma_neon(uint8_t *pix, int stride, int alpha,
|
|
91 int beta, int8_t *tc0);
|
|
92 void ff_h264_h_loop_filter_chroma_neon(uint8_t *pix, int stride, int alpha,
|
|
93 int beta, int8_t *tc0);
|
|
94
|
8339
|
95 void ff_h264_idct_add_neon(uint8_t *dst, DCTELEM *block, int stride);
|
8340
|
96 void ff_h264_idct_dc_add_neon(uint8_t *dst, DCTELEM *block, int stride);
|
8339
|
97
|
8334
|
98 void ff_dsputil_init_neon(DSPContext *c, AVCodecContext *avctx)
|
|
99 {
|
|
100 c->put_pixels_tab[0][0] = ff_put_pixels16_neon;
|
|
101 c->put_pixels_tab[0][1] = ff_put_pixels16_x2_neon;
|
|
102 c->put_pixels_tab[0][2] = ff_put_pixels16_y2_neon;
|
|
103 c->put_pixels_tab[0][3] = ff_put_pixels16_xy2_neon;
|
|
104 c->put_pixels_tab[1][0] = ff_put_pixels8_neon;
|
|
105 c->put_pixels_tab[1][1] = ff_put_pixels8_x2_neon;
|
|
106 c->put_pixels_tab[1][2] = ff_put_pixels8_y2_neon;
|
|
107 c->put_pixels_tab[1][3] = ff_put_pixels8_xy2_neon;
|
|
108
|
|
109 c->put_no_rnd_pixels_tab[0][0] = ff_put_pixels16_neon;
|
|
110 c->put_no_rnd_pixels_tab[0][1] = ff_put_pixels16_x2_no_rnd_neon;
|
|
111 c->put_no_rnd_pixels_tab[0][2] = ff_put_pixels16_y2_no_rnd_neon;
|
|
112 c->put_no_rnd_pixels_tab[0][3] = ff_put_pixels16_xy2_no_rnd_neon;
|
|
113 c->put_no_rnd_pixels_tab[1][0] = ff_put_pixels8_neon;
|
|
114 c->put_no_rnd_pixels_tab[1][1] = ff_put_pixels8_x2_no_rnd_neon;
|
|
115 c->put_no_rnd_pixels_tab[1][2] = ff_put_pixels8_y2_no_rnd_neon;
|
|
116 c->put_no_rnd_pixels_tab[1][3] = ff_put_pixels8_xy2_no_rnd_neon;
|
|
117
|
|
118 c->avg_pixels_tab[0][0] = ff_avg_pixels16_neon;
|
|
119
|
8336
|
120 c->put_h264_chroma_pixels_tab[0] = ff_put_h264_chroma_mc8_neon;
|
|
121 c->put_h264_chroma_pixels_tab[1] = ff_put_h264_chroma_mc4_neon;
|
|
122
|
|
123 c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_neon;
|
|
124 c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_neon;
|
|
125
|
8338
|
126 c->put_h264_qpel_pixels_tab[0][ 0] = ff_put_h264_qpel16_mc00_neon;
|
|
127 c->put_h264_qpel_pixels_tab[0][ 1] = ff_put_h264_qpel16_mc10_neon;
|
|
128 c->put_h264_qpel_pixels_tab[0][ 2] = ff_put_h264_qpel16_mc20_neon;
|
|
129 c->put_h264_qpel_pixels_tab[0][ 3] = ff_put_h264_qpel16_mc30_neon;
|
|
130 c->put_h264_qpel_pixels_tab[0][ 4] = ff_put_h264_qpel16_mc01_neon;
|
|
131 c->put_h264_qpel_pixels_tab[0][ 5] = ff_put_h264_qpel16_mc11_neon;
|
|
132 c->put_h264_qpel_pixels_tab[0][ 6] = ff_put_h264_qpel16_mc21_neon;
|
|
133 c->put_h264_qpel_pixels_tab[0][ 7] = ff_put_h264_qpel16_mc31_neon;
|
|
134 c->put_h264_qpel_pixels_tab[0][ 8] = ff_put_h264_qpel16_mc02_neon;
|
|
135 c->put_h264_qpel_pixels_tab[0][ 9] = ff_put_h264_qpel16_mc12_neon;
|
|
136 c->put_h264_qpel_pixels_tab[0][10] = ff_put_h264_qpel16_mc22_neon;
|
|
137 c->put_h264_qpel_pixels_tab[0][11] = ff_put_h264_qpel16_mc32_neon;
|
|
138 c->put_h264_qpel_pixels_tab[0][12] = ff_put_h264_qpel16_mc03_neon;
|
|
139 c->put_h264_qpel_pixels_tab[0][13] = ff_put_h264_qpel16_mc13_neon;
|
|
140 c->put_h264_qpel_pixels_tab[0][14] = ff_put_h264_qpel16_mc23_neon;
|
|
141 c->put_h264_qpel_pixels_tab[0][15] = ff_put_h264_qpel16_mc33_neon;
|
|
142
|
|
143 c->put_h264_qpel_pixels_tab[1][ 0] = ff_put_h264_qpel8_mc00_neon;
|
|
144 c->put_h264_qpel_pixels_tab[1][ 1] = ff_put_h264_qpel8_mc10_neon;
|
|
145 c->put_h264_qpel_pixels_tab[1][ 2] = ff_put_h264_qpel8_mc20_neon;
|
|
146 c->put_h264_qpel_pixels_tab[1][ 3] = ff_put_h264_qpel8_mc30_neon;
|
|
147 c->put_h264_qpel_pixels_tab[1][ 4] = ff_put_h264_qpel8_mc01_neon;
|
|
148 c->put_h264_qpel_pixels_tab[1][ 5] = ff_put_h264_qpel8_mc11_neon;
|
|
149 c->put_h264_qpel_pixels_tab[1][ 6] = ff_put_h264_qpel8_mc21_neon;
|
|
150 c->put_h264_qpel_pixels_tab[1][ 7] = ff_put_h264_qpel8_mc31_neon;
|
|
151 c->put_h264_qpel_pixels_tab[1][ 8] = ff_put_h264_qpel8_mc02_neon;
|
|
152 c->put_h264_qpel_pixels_tab[1][ 9] = ff_put_h264_qpel8_mc12_neon;
|
|
153 c->put_h264_qpel_pixels_tab[1][10] = ff_put_h264_qpel8_mc22_neon;
|
|
154 c->put_h264_qpel_pixels_tab[1][11] = ff_put_h264_qpel8_mc32_neon;
|
|
155 c->put_h264_qpel_pixels_tab[1][12] = ff_put_h264_qpel8_mc03_neon;
|
|
156 c->put_h264_qpel_pixels_tab[1][13] = ff_put_h264_qpel8_mc13_neon;
|
|
157 c->put_h264_qpel_pixels_tab[1][14] = ff_put_h264_qpel8_mc23_neon;
|
|
158 c->put_h264_qpel_pixels_tab[1][15] = ff_put_h264_qpel8_mc33_neon;
|
8334
|
159
|
|
160 c->avg_h264_qpel_pixels_tab[0][ 0] = ff_avg_h264_qpel16_mc00_neon;
|
8337
|
161
|
|
162 c->h264_v_loop_filter_luma = ff_h264_v_loop_filter_luma_neon;
|
|
163 c->h264_h_loop_filter_luma = ff_h264_h_loop_filter_luma_neon;
|
|
164 c->h264_v_loop_filter_chroma = ff_h264_v_loop_filter_chroma_neon;
|
|
165 c->h264_h_loop_filter_chroma = ff_h264_h_loop_filter_chroma_neon;
|
8339
|
166
|
|
167 c->h264_idct_add = ff_h264_idct_add_neon;
|
8340
|
168 c->h264_idct_dc_add = ff_h264_idct_dc_add_neon;
|
8334
|
169 }
|