2 * Copyright (c) 2010 Mans Rullgard <mans@mansr.com>
4 * This file is part of FFmpeg.
6 * FFmpeg is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
11 * FFmpeg is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with FFmpeg; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23 #include "libavcodec/dsputil.h"
24 #include "libavcodec/h264dsp.h"
26 void ff_h264_v_loop_filter_luma_neon(uint8_t *pix, int stride, int alpha,
27 int beta, int8_t *tc0);
28 void ff_h264_h_loop_filter_luma_neon(uint8_t *pix, int stride, int alpha,
29 int beta, int8_t *tc0);
30 void ff_h264_v_loop_filter_chroma_neon(uint8_t *pix, int stride, int alpha,
31 int beta, int8_t *tc0);
32 void ff_h264_h_loop_filter_chroma_neon(uint8_t *pix, int stride, int alpha,
33 int beta, int8_t *tc0);
35 void ff_weight_h264_pixels_16x16_neon(uint8_t *ds, int stride, int log2_den,
36 int weight, int offset);
37 void ff_weight_h264_pixels_16x8_neon(uint8_t *ds, int stride, int log2_den,
38 int weight, int offset);
39 void ff_weight_h264_pixels_8x16_neon(uint8_t *ds, int stride, int log2_den,
40 int weight, int offset);
41 void ff_weight_h264_pixels_8x8_neon(uint8_t *ds, int stride, int log2_den,
42 int weight, int offset);
43 void ff_weight_h264_pixels_8x4_neon(uint8_t *ds, int stride, int log2_den,
44 int weight, int offset);
45 void ff_weight_h264_pixels_4x8_neon(uint8_t *ds, int stride, int log2_den,
46 int weight, int offset);
47 void ff_weight_h264_pixels_4x4_neon(uint8_t *ds, int stride, int log2_den,
48 int weight, int offset);
49 void ff_weight_h264_pixels_4x2_neon(uint8_t *ds, int stride, int log2_den,
50 int weight, int offset);
52 void ff_biweight_h264_pixels_16x16_neon(uint8_t *dst, uint8_t *src, int stride,
53 int log2_den, int weightd, int weights,
55 void ff_biweight_h264_pixels_16x8_neon(uint8_t *dst, uint8_t *src, int stride,
56 int log2_den, int weightd, int weights,
58 void ff_biweight_h264_pixels_8x16_neon(uint8_t *dst, uint8_t *src, int stride,
59 int log2_den, int weightd, int weights,
61 void ff_biweight_h264_pixels_8x8_neon(uint8_t *dst, uint8_t *src, int stride,
62 int log2_den, int weightd, int weights,
64 void ff_biweight_h264_pixels_8x4_neon(uint8_t *dst, uint8_t *src, int stride,
65 int log2_den, int weightd, int weights,
67 void ff_biweight_h264_pixels_4x8_neon(uint8_t *dst, uint8_t *src, int stride,
68 int log2_den, int weightd, int weights,
70 void ff_biweight_h264_pixels_4x4_neon(uint8_t *dst, uint8_t *src, int stride,
71 int log2_den, int weightd, int weights,
73 void ff_biweight_h264_pixels_4x2_neon(uint8_t *dst, uint8_t *src, int stride,
74 int log2_den, int weightd, int weights,
77 void ff_h264_idct_add_neon(uint8_t *dst, DCTELEM *block, int stride);
78 void ff_h264_idct_dc_add_neon(uint8_t *dst, DCTELEM *block, int stride);
79 void ff_h264_idct_add16_neon(uint8_t *dst, const int *block_offset,
80 DCTELEM *block, int stride,
81 const uint8_t nnzc[6*8]);
82 void ff_h264_idct_add16intra_neon(uint8_t *dst, const int *block_offset,
83 DCTELEM *block, int stride,
84 const uint8_t nnzc[6*8]);
85 void ff_h264_idct_add8_neon(uint8_t **dest, const int *block_offset,
86 DCTELEM *block, int stride,
87 const uint8_t nnzc[6*8]);
89 void ff_h264_idct8_add_neon(uint8_t *dst, DCTELEM *block, int stride);
90 void ff_h264_idct8_dc_add_neon(uint8_t *dst, DCTELEM *block, int stride);
91 void ff_h264_idct8_add4_neon(uint8_t *dst, const int *block_offset,
92 DCTELEM *block, int stride,
93 const uint8_t nnzc[6*8]);
95 static void ff_h264dsp_init_neon(H264DSPContext *c, const int bit_depth)
98 c->h264_v_loop_filter_luma = ff_h264_v_loop_filter_luma_neon;
99 c->h264_h_loop_filter_luma = ff_h264_h_loop_filter_luma_neon;
100 c->h264_v_loop_filter_chroma = ff_h264_v_loop_filter_chroma_neon;
101 c->h264_h_loop_filter_chroma = ff_h264_h_loop_filter_chroma_neon;
103 c->weight_h264_pixels_tab[0] = ff_weight_h264_pixels_16x16_neon;
104 c->weight_h264_pixels_tab[1] = ff_weight_h264_pixels_16x8_neon;
105 c->weight_h264_pixels_tab[2] = ff_weight_h264_pixels_8x16_neon;
106 c->weight_h264_pixels_tab[3] = ff_weight_h264_pixels_8x8_neon;
107 c->weight_h264_pixels_tab[4] = ff_weight_h264_pixels_8x4_neon;
108 c->weight_h264_pixels_tab[5] = ff_weight_h264_pixels_4x8_neon;
109 c->weight_h264_pixels_tab[6] = ff_weight_h264_pixels_4x4_neon;
110 c->weight_h264_pixels_tab[7] = ff_weight_h264_pixels_4x2_neon;
112 c->biweight_h264_pixels_tab[0] = ff_biweight_h264_pixels_16x16_neon;
113 c->biweight_h264_pixels_tab[1] = ff_biweight_h264_pixels_16x8_neon;
114 c->biweight_h264_pixels_tab[2] = ff_biweight_h264_pixels_8x16_neon;
115 c->biweight_h264_pixels_tab[3] = ff_biweight_h264_pixels_8x8_neon;
116 c->biweight_h264_pixels_tab[4] = ff_biweight_h264_pixels_8x4_neon;
117 c->biweight_h264_pixels_tab[5] = ff_biweight_h264_pixels_4x8_neon;
118 c->biweight_h264_pixels_tab[6] = ff_biweight_h264_pixels_4x4_neon;
119 c->biweight_h264_pixels_tab[7] = ff_biweight_h264_pixels_4x2_neon;
121 c->h264_idct_add = ff_h264_idct_add_neon;
122 c->h264_idct_dc_add = ff_h264_idct_dc_add_neon;
123 c->h264_idct_add16 = ff_h264_idct_add16_neon;
124 c->h264_idct_add16intra = ff_h264_idct_add16intra_neon;
125 c->h264_idct_add8 = ff_h264_idct_add8_neon;
126 c->h264_idct8_add = ff_h264_idct8_add_neon;
127 c->h264_idct8_dc_add = ff_h264_idct8_dc_add_neon;
128 c->h264_idct8_add4 = ff_h264_idct8_add4_neon;
132 void ff_h264dsp_init_arm(H264DSPContext *c, const int bit_depth)
134 if (HAVE_NEON) ff_h264dsp_init_neon(c, bit_depth);