1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
|
/*
* VVC filters DSP
*
* Copyright (C) 2024 Zhao Zhili
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "libavutil/cpu.h"
#include "libavutil/aarch64/cpu.h"
#include "libavcodec/aarch64/h26x/dsp.h"
#include "libavcodec/vvc/dsp.h"
#include "libavcodec/vvc/dec.h"
#include "libavcodec/vvc/ctu.h"
#define BIT_DEPTH 8
#include "alf_template.c"
#undef BIT_DEPTH
#define BIT_DEPTH 10
#include "alf_template.c"
#undef BIT_DEPTH
#define BIT_DEPTH 12
#include "alf_template.c"
#undef BIT_DEPTH
int ff_vvc_sad_neon(const int16_t *src0, const int16_t *src1, int dx, int dy,
const int block_w, const int block_h);
void ff_vvc_avg_8_neon(uint8_t *dst, ptrdiff_t dst_stride,
const int16_t *src0, const int16_t *src1, int width,
int height);
void ff_vvc_avg_10_neon(uint8_t *dst, ptrdiff_t dst_stride,
const int16_t *src0, const int16_t *src1, int width,
int height);
void ff_vvc_avg_12_neon(uint8_t *dst, ptrdiff_t dst_stride,
const int16_t *src0, const int16_t *src1, int width,
int height);
void ff_vvc_w_avg_8_neon(uint8_t *_dst, ptrdiff_t _dst_stride,
const int16_t *src0, const int16_t *src1,
int width, int height,
uintptr_t w0_w1, uintptr_t offset_shift);
void ff_vvc_w_avg_10_neon(uint8_t *_dst, ptrdiff_t _dst_stride,
const int16_t *src0, const int16_t *src1,
int width, int height,
uintptr_t w0_w1, uintptr_t offset_shift);
void ff_vvc_w_avg_12_neon(uint8_t *_dst, ptrdiff_t _dst_stride,
const int16_t *src0, const int16_t *src1,
int width, int height,
uintptr_t w0_w1, uintptr_t offset_shift);
/* When passing arguments to functions, Apple platforms diverge from the ARM64
* standard ABI for functions that require passing arguments on the stack. To
* simplify portability in the assembly function interface, use a different
* function signature that doesn't require passing arguments on the stack.
*/
#define W_AVG_FUN(bit_depth) \
static void vvc_w_avg_ ## bit_depth(uint8_t *dst, ptrdiff_t dst_stride, \
const int16_t *src0, const int16_t *src1, int width, int height, \
int denom, int w0, int w1, int o0, int o1) \
{ \
int shift = denom + FFMAX(3, 15 - bit_depth); \
int offset = ((o0 + o1) * (1 << (bit_depth - 8)) + 1) * (1 << (shift - 1)); \
uintptr_t w0_w1 = ((uintptr_t)w0 << 32) | (uint32_t)w1; \
uintptr_t offset_shift = ((uintptr_t)offset << 32) | (uint32_t)shift; \
ff_vvc_w_avg_ ## bit_depth ## _neon(dst, dst_stride, src0, src1, width, height, w0_w1, offset_shift); \
}
W_AVG_FUN(8)
W_AVG_FUN(10)
W_AVG_FUN(12)
#define DMVR_FUN(fn, bd) \
void ff_vvc_dmvr_ ## fn ## bd ## _neon(int16_t *dst, \
const uint8_t *_src, ptrdiff_t _src_stride, int height, \
intptr_t mx, intptr_t my, int width);
DMVR_FUN(, 8)
DMVR_FUN(, 12)
DMVR_FUN(hv_, 8)
DMVR_FUN(hv_, 10)
DMVR_FUN(hv_, 12)
void ff_vvc_dsp_init_aarch64(VVCDSPContext *const c, const int bd)
{
int cpu_flags = av_get_cpu_flags();
if (!have_neon(cpu_flags))
return;
if (bd == 8) {
c->inter.put[0][1][0][0] = ff_vvc_put_pel_pixels4_8_neon;
c->inter.put[0][2][0][0] = ff_vvc_put_pel_pixels8_8_neon;
c->inter.put[0][3][0][0] = ff_vvc_put_pel_pixels16_8_neon;
c->inter.put[0][4][0][0] = ff_vvc_put_pel_pixels32_8_neon;
c->inter.put[0][5][0][0] = ff_vvc_put_pel_pixels64_8_neon;
c->inter.put[0][6][0][0] = ff_vvc_put_pel_pixels128_8_neon;
c->inter.put[0][1][0][1] = ff_vvc_put_qpel_h4_8_neon;
c->inter.put[0][2][0][1] = ff_vvc_put_qpel_h8_8_neon;
c->inter.put[0][3][0][1] = ff_vvc_put_qpel_h16_8_neon;
c->inter.put[0][4][0][1] =
c->inter.put[0][5][0][1] =
c->inter.put[0][6][0][1] = ff_vvc_put_qpel_h32_8_neon;
c->inter.put[0][1][1][0] = ff_vvc_put_qpel_v4_8_neon;
c->inter.put[0][2][1][0] =
c->inter.put[0][3][1][0] =
c->inter.put[0][4][1][0] =
c->inter.put[0][5][1][0] =
c->inter.put[0][6][1][0] = ff_vvc_put_qpel_v8_8_neon;
c->inter.put[0][1][1][1] = ff_vvc_put_qpel_hv4_8_neon;
c->inter.put[0][2][1][1] = ff_vvc_put_qpel_hv8_8_neon;
c->inter.put[0][3][1][1] = ff_vvc_put_qpel_hv16_8_neon;
c->inter.put[0][4][1][1] = ff_vvc_put_qpel_hv32_8_neon;
c->inter.put[0][5][1][1] = ff_vvc_put_qpel_hv64_8_neon;
c->inter.put[0][6][1][1] = ff_vvc_put_qpel_hv128_8_neon;
c->inter.put[1][1][0][1] = ff_vvc_put_epel_h4_8_neon;
c->inter.put[1][2][0][1] = ff_vvc_put_epel_h8_8_neon;
c->inter.put[1][3][0][1] = ff_vvc_put_epel_h16_8_neon;
c->inter.put[1][4][0][1] =
c->inter.put[1][5][0][1] =
c->inter.put[1][6][0][1] = ff_vvc_put_epel_h32_8_neon;
c->inter.put[1][1][1][1] = ff_vvc_put_epel_hv4_8_neon;
c->inter.put[1][2][1][1] = ff_vvc_put_epel_hv8_8_neon;
c->inter.put[1][3][1][1] = ff_vvc_put_epel_hv16_8_neon;
c->inter.put[1][4][1][1] = ff_vvc_put_epel_hv32_8_neon;
c->inter.put[1][5][1][1] = ff_vvc_put_epel_hv64_8_neon;
c->inter.put[1][6][1][1] = ff_vvc_put_epel_hv128_8_neon;
c->inter.put_uni[0][1][0][0] = ff_vvc_put_pel_uni_pixels4_8_neon;
c->inter.put_uni[0][2][0][0] = ff_vvc_put_pel_uni_pixels8_8_neon;
c->inter.put_uni[0][3][0][0] = ff_vvc_put_pel_uni_pixels16_8_neon;
c->inter.put_uni[0][4][0][0] = ff_vvc_put_pel_uni_pixels32_8_neon;
c->inter.put_uni[0][5][0][0] = ff_vvc_put_pel_uni_pixels64_8_neon;
c->inter.put_uni[0][6][0][0] = ff_vvc_put_pel_uni_pixels128_8_neon;
c->inter.put_uni[0][1][0][1] = ff_vvc_put_qpel_uni_h4_8_neon;
c->inter.put_uni[0][2][0][1] = ff_vvc_put_qpel_uni_h8_8_neon;
c->inter.put_uni[0][3][0][1] = ff_vvc_put_qpel_uni_h16_8_neon;
c->inter.put_uni[0][4][0][1] =
c->inter.put_uni[0][5][0][1] =
c->inter.put_uni[0][6][0][1] = ff_vvc_put_qpel_uni_h32_8_neon;
c->inter.put_uni_w[0][1][0][0] = ff_vvc_put_pel_uni_w_pixels4_8_neon;
c->inter.put_uni_w[0][2][0][0] = ff_vvc_put_pel_uni_w_pixels8_8_neon;
c->inter.put_uni_w[0][3][0][0] = ff_vvc_put_pel_uni_w_pixels16_8_neon;
c->inter.put_uni_w[0][4][0][0] = ff_vvc_put_pel_uni_w_pixels32_8_neon;
c->inter.put_uni_w[0][5][0][0] = ff_vvc_put_pel_uni_w_pixels64_8_neon;
c->inter.put_uni_w[0][6][0][0] = ff_vvc_put_pel_uni_w_pixels128_8_neon;
c->inter.avg = ff_vvc_avg_8_neon;
c->inter.w_avg = vvc_w_avg_8;
c->inter.dmvr[0][0] = ff_vvc_dmvr_8_neon;
c->inter.dmvr[1][1] = ff_vvc_dmvr_hv_8_neon;
for (int i = 0; i < FF_ARRAY_ELEMS(c->sao.band_filter); i++)
c->sao.band_filter[i] = ff_h26x_sao_band_filter_8x8_8_neon;
c->sao.edge_filter[0] = ff_vvc_sao_edge_filter_8x8_8_neon;
for (int i = 1; i < FF_ARRAY_ELEMS(c->sao.edge_filter); i++)
c->sao.edge_filter[i] = ff_vvc_sao_edge_filter_16x16_8_neon;
c->alf.filter[LUMA] = alf_filter_luma_8_neon;
c->alf.filter[CHROMA] = alf_filter_chroma_8_neon;
if (have_i8mm(cpu_flags)) {
c->inter.put[0][1][0][1] = ff_vvc_put_qpel_h4_8_neon_i8mm;
c->inter.put[0][2][0][1] = ff_vvc_put_qpel_h8_8_neon_i8mm;
c->inter.put[0][3][0][1] = ff_vvc_put_qpel_h16_8_neon_i8mm;
c->inter.put[0][4][0][1] = ff_vvc_put_qpel_h32_8_neon_i8mm;
c->inter.put[0][5][0][1] = ff_vvc_put_qpel_h64_8_neon_i8mm;
c->inter.put[0][6][0][1] = ff_vvc_put_qpel_h128_8_neon_i8mm;
c->inter.put[0][1][1][1] = ff_vvc_put_qpel_hv4_8_neon_i8mm;
c->inter.put[0][2][1][1] = ff_vvc_put_qpel_hv8_8_neon_i8mm;
c->inter.put[0][3][1][1] = ff_vvc_put_qpel_hv16_8_neon_i8mm;
c->inter.put[0][4][1][1] = ff_vvc_put_qpel_hv32_8_neon_i8mm;
c->inter.put[0][5][1][1] = ff_vvc_put_qpel_hv64_8_neon_i8mm;
c->inter.put[0][6][1][1] = ff_vvc_put_qpel_hv128_8_neon_i8mm;
c->inter.put[1][1][0][1] = ff_vvc_put_epel_h4_8_neon_i8mm;
c->inter.put[1][2][0][1] = ff_vvc_put_epel_h8_8_neon_i8mm;
c->inter.put[1][3][0][1] = ff_vvc_put_epel_h16_8_neon_i8mm;
c->inter.put[1][4][0][1] = ff_vvc_put_epel_h32_8_neon_i8mm;
c->inter.put[1][5][0][1] = ff_vvc_put_epel_h64_8_neon_i8mm;
c->inter.put[1][6][0][1] = ff_vvc_put_epel_h128_8_neon_i8mm;
c->inter.put[1][1][1][1] = ff_vvc_put_epel_hv4_8_neon_i8mm;
c->inter.put[1][2][1][1] = ff_vvc_put_epel_hv8_8_neon_i8mm;
c->inter.put[1][3][1][1] = ff_vvc_put_epel_hv16_8_neon_i8mm;
c->inter.put[1][4][1][1] = ff_vvc_put_epel_hv32_8_neon_i8mm;
c->inter.put[1][5][1][1] = ff_vvc_put_epel_hv64_8_neon_i8mm;
c->inter.put[1][6][1][1] = ff_vvc_put_epel_hv128_8_neon_i8mm;
}
} else if (bd == 10) {
c->inter.avg = ff_vvc_avg_10_neon;
c->inter.w_avg = vvc_w_avg_10;
c->inter.dmvr[1][1] = ff_vvc_dmvr_hv_10_neon;
c->alf.filter[LUMA] = alf_filter_luma_10_neon;
c->alf.filter[CHROMA] = alf_filter_chroma_10_neon;
} else if (bd == 12) {
c->inter.avg = ff_vvc_avg_12_neon;
c->inter.w_avg = vvc_w_avg_12;
c->inter.dmvr[0][0] = ff_vvc_dmvr_12_neon;
c->inter.dmvr[1][1] = ff_vvc_dmvr_hv_12_neon;
c->alf.filter[LUMA] = alf_filter_luma_12_neon;
c->alf.filter[CHROMA] = alf_filter_chroma_12_neon;
}
c->inter.sad = ff_vvc_sad_neon;
}
|
