• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /*
2  * Copyright (c) 2010 Mans Rullgard <mans@mansr.com>
3  *
4  * This file is part of FFmpeg.
5  *
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.
10  *
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.
15  *
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
19  */
20 
21 #include <stdint.h>
22 
23 #include "libavutil/attributes.h"
24 #include "libavutil/cpu.h"
25 #include "libavutil/aarch64/cpu.h"
26 #include "libavcodec/h264dsp.h"
27 
28 void ff_h264_v_loop_filter_luma_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
29                                      int beta, int8_t *tc0);
30 void ff_h264_h_loop_filter_luma_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
31                                      int beta, int8_t *tc0);
32 void ff_h264_v_loop_filter_luma_intra_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
33                                            int beta);
34 void ff_h264_h_loop_filter_luma_intra_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
35                                            int beta);
36 void ff_h264_v_loop_filter_chroma_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
37                                        int beta, int8_t *tc0);
38 void ff_h264_h_loop_filter_chroma_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
39                                        int beta, int8_t *tc0);
40 void ff_h264_h_loop_filter_chroma422_neon(uint8_t *pix, ptrdiff_t stride, int alpha,
41                                           int beta, int8_t *tc0);
42 void ff_h264_v_loop_filter_chroma_intra_neon(uint8_t *pix, ptrdiff_t stride,
43                                              int alpha, int beta);
44 void ff_h264_h_loop_filter_chroma_intra_neon(uint8_t *pix, ptrdiff_t stride,
45                                              int alpha, int beta);
46 void ff_h264_h_loop_filter_chroma422_intra_neon(uint8_t *pix, ptrdiff_t stride,
47                                                 int alpha, int beta);
48 void ff_h264_h_loop_filter_chroma_mbaff_intra_neon(uint8_t *pix, ptrdiff_t stride,
49                                                    int alpha, int beta);
50 
51 void ff_weight_h264_pixels_16_neon(uint8_t *dst, ptrdiff_t stride, int height,
52                                    int log2_den, int weight, int offset);
53 void ff_weight_h264_pixels_8_neon(uint8_t *dst, ptrdiff_t stride, int height,
54                                   int log2_den, int weight, int offset);
55 void ff_weight_h264_pixels_4_neon(uint8_t *dst, ptrdiff_t stride, int height,
56                                   int log2_den, int weight, int offset);
57 
58 void ff_biweight_h264_pixels_16_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
59                                      int height, int log2_den, int weightd,
60                                      int weights, int offset);
61 void ff_biweight_h264_pixels_8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
62                                     int height, int log2_den, int weightd,
63                                     int weights, int offset);
64 void ff_biweight_h264_pixels_4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
65                                     int height, int log2_den, int weightd,
66                                     int weights, int offset);
67 
68 void ff_h264_idct_add_neon(uint8_t *dst, int16_t *block, int stride);
69 void ff_h264_idct_dc_add_neon(uint8_t *dst, int16_t *block, int stride);
70 void ff_h264_idct_add16_neon(uint8_t *dst, const int *block_offset,
71                              int16_t *block, int stride,
72                              const uint8_t nnzc[6*8]);
73 void ff_h264_idct_add16intra_neon(uint8_t *dst, const int *block_offset,
74                                   int16_t *block, int stride,
75                                   const uint8_t nnzc[6*8]);
76 void ff_h264_idct_add8_neon(uint8_t **dest, const int *block_offset,
77                             int16_t *block, int stride,
78                             const uint8_t nnzc[6*8]);
79 
80 void ff_h264_idct8_add_neon(uint8_t *dst, int16_t *block, int stride);
81 void ff_h264_idct8_dc_add_neon(uint8_t *dst, int16_t *block, int stride);
82 void ff_h264_idct8_add4_neon(uint8_t *dst, const int *block_offset,
83                              int16_t *block, int stride,
84                              const uint8_t nnzc[6*8]);
85 
ff_h264dsp_init_aarch64(H264DSPContext * c,const int bit_depth,const int chroma_format_idc)86 av_cold void ff_h264dsp_init_aarch64(H264DSPContext *c, const int bit_depth,
87                                      const int chroma_format_idc)
88 {
89     int cpu_flags = av_get_cpu_flags();
90 
91     if (have_neon(cpu_flags) && bit_depth == 8) {
92         c->h264_v_loop_filter_luma   = ff_h264_v_loop_filter_luma_neon;
93         c->h264_h_loop_filter_luma   = ff_h264_h_loop_filter_luma_neon;
94         c->h264_v_loop_filter_luma_intra= ff_h264_v_loop_filter_luma_intra_neon;
95         c->h264_h_loop_filter_luma_intra= ff_h264_h_loop_filter_luma_intra_neon;
96 
97         c->h264_v_loop_filter_chroma = ff_h264_v_loop_filter_chroma_neon;
98         c->h264_v_loop_filter_chroma_intra = ff_h264_v_loop_filter_chroma_intra_neon;
99 
100         if (chroma_format_idc <= 1) {
101             c->h264_h_loop_filter_chroma = ff_h264_h_loop_filter_chroma_neon;
102             c->h264_h_loop_filter_chroma_intra = ff_h264_h_loop_filter_chroma_intra_neon;
103             c->h264_h_loop_filter_chroma_mbaff_intra = ff_h264_h_loop_filter_chroma_mbaff_intra_neon;
104         } else {
105             c->h264_h_loop_filter_chroma = ff_h264_h_loop_filter_chroma422_neon;
106             c->h264_h_loop_filter_chroma_mbaff = ff_h264_h_loop_filter_chroma_neon;
107             c->h264_h_loop_filter_chroma_intra = ff_h264_h_loop_filter_chroma422_intra_neon;
108             c->h264_h_loop_filter_chroma_mbaff_intra = ff_h264_h_loop_filter_chroma_intra_neon;
109         }
110 
111         c->weight_h264_pixels_tab[0] = ff_weight_h264_pixels_16_neon;
112         c->weight_h264_pixels_tab[1] = ff_weight_h264_pixels_8_neon;
113         c->weight_h264_pixels_tab[2] = ff_weight_h264_pixels_4_neon;
114 
115         c->biweight_h264_pixels_tab[0] = ff_biweight_h264_pixels_16_neon;
116         c->biweight_h264_pixels_tab[1] = ff_biweight_h264_pixels_8_neon;
117         c->biweight_h264_pixels_tab[2] = ff_biweight_h264_pixels_4_neon;
118 
119         c->h264_idct_add        = ff_h264_idct_add_neon;
120         c->h264_idct_dc_add     = ff_h264_idct_dc_add_neon;
121         c->h264_idct_add16      = ff_h264_idct_add16_neon;
122         c->h264_idct_add16intra = ff_h264_idct_add16intra_neon;
123         if (chroma_format_idc <= 1)
124             c->h264_idct_add8   = ff_h264_idct_add8_neon;
125         c->h264_idct8_add       = ff_h264_idct8_add_neon;
126         c->h264_idct8_dc_add    = ff_h264_idct8_dc_add_neon;
127         c->h264_idct8_add4      = ff_h264_idct8_add4_neon;
128     }
129 }
130