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
230
231
232
233
|
/*
* Copyright (c) 2014 Seppo Tomperi <seppo.tomperi@vtt.fi>
*
* 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/attributes.h"
#include "libavutil/arm/cpu.h"
#include "libavcodec/hevcdsp.h"
void ff_hevc_v_loop_filter_luma_neon(uint8_t *_pix, ptrdiff_t _stride, int _beta, int *_tc, uint8_t *_no_p, uint8_t *_no_q);
void ff_hevc_h_loop_filter_luma_neon(uint8_t *_pix, ptrdiff_t _stride, int _beta, int *_tc, uint8_t *_no_p, uint8_t *_no_q);
void ff_hevc_v_loop_filter_chroma_neon(uint8_t *_pix, ptrdiff_t _stride, int *_tc, uint8_t *_no_p, uint8_t *_no_q);
void ff_hevc_h_loop_filter_chroma_neon(uint8_t *_pix, ptrdiff_t _stride, int *_tc, uint8_t *_no_p, uint8_t *_no_q);
void ff_hevc_transform_4x4_neon_8(int16_t *coeffs, int col_limit);
void ff_hevc_transform_8x8_neon_8(int16_t *coeffs, int col_limit);
void ff_hevc_idct_4x4_dc_neon_8(int16_t *coeffs);
void ff_hevc_idct_8x8_dc_neon_8(int16_t *coeffs);
void ff_hevc_idct_16x16_dc_neon_8(int16_t *coeffs);
void ff_hevc_idct_32x32_dc_neon_8(int16_t *coeffs);
void ff_hevc_transform_luma_4x4_neon_8(int16_t *coeffs);
void ff_hevc_transform_add_4x4_neon_8(uint8_t *_dst, int16_t *coeffs,
ptrdiff_t stride);
void ff_hevc_transform_add_8x8_neon_8(uint8_t *_dst, int16_t *coeffs,
ptrdiff_t stride);
void ff_hevc_transform_add_16x16_neon_8(uint8_t *_dst, int16_t *coeffs,
ptrdiff_t stride);
void ff_hevc_transform_add_32x32_neon_8(uint8_t *_dst, int16_t *coeffs,
ptrdiff_t stride);
#define PUT_PIXELS(name) \
void name(int16_t *dst, uint8_t *src, \
ptrdiff_t srcstride, int height, \
intptr_t mx, intptr_t my, int width)
PUT_PIXELS(ff_hevc_put_pixels_w2_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w4_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w6_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w8_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w12_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w16_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w24_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w32_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w48_neon_8);
PUT_PIXELS(ff_hevc_put_pixels_w64_neon_8);
#undef PUT_PIXELS
static void (*put_hevc_qpel_neon[4][4])(int16_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride,
int height, int width);
static void (*put_hevc_qpel_uw_neon[4][4])(uint8_t *dst, ptrdiff_t dststride, uint8_t *_src, ptrdiff_t _srcstride,
int width, int height, int16_t* src2, ptrdiff_t src2stride);
void ff_hevc_put_qpel_neon_wrapper(int16_t *dst, uint8_t *src, ptrdiff_t srcstride,
int height, intptr_t mx, intptr_t my, int width);
void ff_hevc_put_qpel_uni_neon_wrapper(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride,
int height, intptr_t mx, intptr_t my, int width);
void ff_hevc_put_qpel_bi_neon_wrapper(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride,
int16_t *src2,
int height, intptr_t mx, intptr_t my, int width);
#define QPEL_FUNC(name) \
void name(int16_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, \
int height, int width)
QPEL_FUNC(ff_hevc_put_qpel_v1_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_v2_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_v3_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h1_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h2_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h3_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h1v1_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h1v2_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h1v3_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h2v1_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h2v2_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h2v3_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h3v1_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h3v2_neon_8);
QPEL_FUNC(ff_hevc_put_qpel_h3v3_neon_8);
#undef QPEL_FUNC
#define QPEL_FUNC_UW_PIX(name) \
void name(uint8_t *dst, ptrdiff_t dststride, uint8_t *_src, ptrdiff_t _srcstride, \
int height, intptr_t mx, intptr_t my, int width);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w4_neon_8);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w8_neon_8);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w16_neon_8);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w24_neon_8);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w32_neon_8);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w48_neon_8);
QPEL_FUNC_UW_PIX(ff_hevc_put_qpel_uw_pixels_w64_neon_8);
#undef QPEL_FUNC_UW_PIX
#define QPEL_FUNC_UW(name) \
void name(uint8_t *dst, ptrdiff_t dststride, uint8_t *_src, ptrdiff_t _srcstride, \
int width, int height, int16_t* src2, ptrdiff_t src2stride);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_pixels_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_v1_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_v2_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_v3_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h1_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h2_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h3_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h1v1_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h1v2_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h1v3_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h2v1_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h2v2_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h2v3_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h3v1_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h3v2_neon_8);
QPEL_FUNC_UW(ff_hevc_put_qpel_uw_h3v3_neon_8);
#undef QPEL_FUNC_UW
void ff_hevc_put_qpel_neon_wrapper(int16_t *dst, uint8_t *src, ptrdiff_t srcstride,
int height, intptr_t mx, intptr_t my, int width) {
put_hevc_qpel_neon[my][mx](dst, MAX_PB_SIZE, src, srcstride, height, width);
}
void ff_hevc_put_qpel_uni_neon_wrapper(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride,
int height, intptr_t mx, intptr_t my, int width) {
put_hevc_qpel_uw_neon[my][mx](dst, dststride, src, srcstride, width, height, NULL, 0);
}
void ff_hevc_put_qpel_bi_neon_wrapper(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride,
int16_t *src2,
int height, intptr_t mx, intptr_t my, int width) {
put_hevc_qpel_uw_neon[my][mx](dst, dststride, src, srcstride, width, height, src2, MAX_PB_SIZE);
}
static av_cold void hevcdsp_init_neon(HEVCDSPContext *c, const int bit_depth)
{
#if HAVE_NEON
if (bit_depth == 8) {
int x;
c->hevc_v_loop_filter_luma = ff_hevc_v_loop_filter_luma_neon;
c->hevc_h_loop_filter_luma = ff_hevc_h_loop_filter_luma_neon;
c->hevc_v_loop_filter_chroma = ff_hevc_v_loop_filter_chroma_neon;
c->hevc_h_loop_filter_chroma = ff_hevc_h_loop_filter_chroma_neon;
c->idct[0] = ff_hevc_transform_4x4_neon_8;
c->idct[1] = ff_hevc_transform_8x8_neon_8;
c->idct_dc[0] = ff_hevc_idct_4x4_dc_neon_8;
c->idct_dc[1] = ff_hevc_idct_8x8_dc_neon_8;
c->idct_dc[2] = ff_hevc_idct_16x16_dc_neon_8;
c->idct_dc[3] = ff_hevc_idct_32x32_dc_neon_8;
c->transform_add[0] = ff_hevc_transform_add_4x4_neon_8;
c->transform_add[1] = ff_hevc_transform_add_8x8_neon_8;
c->transform_add[2] = ff_hevc_transform_add_16x16_neon_8;
c->transform_add[3] = ff_hevc_transform_add_32x32_neon_8;
c->idct_4x4_luma = ff_hevc_transform_luma_4x4_neon_8;
put_hevc_qpel_neon[1][0] = ff_hevc_put_qpel_v1_neon_8;
put_hevc_qpel_neon[2][0] = ff_hevc_put_qpel_v2_neon_8;
put_hevc_qpel_neon[3][0] = ff_hevc_put_qpel_v3_neon_8;
put_hevc_qpel_neon[0][1] = ff_hevc_put_qpel_h1_neon_8;
put_hevc_qpel_neon[0][2] = ff_hevc_put_qpel_h2_neon_8;
put_hevc_qpel_neon[0][3] = ff_hevc_put_qpel_h3_neon_8;
put_hevc_qpel_neon[1][1] = ff_hevc_put_qpel_h1v1_neon_8;
put_hevc_qpel_neon[1][2] = ff_hevc_put_qpel_h2v1_neon_8;
put_hevc_qpel_neon[1][3] = ff_hevc_put_qpel_h3v1_neon_8;
put_hevc_qpel_neon[2][1] = ff_hevc_put_qpel_h1v2_neon_8;
put_hevc_qpel_neon[2][2] = ff_hevc_put_qpel_h2v2_neon_8;
put_hevc_qpel_neon[2][3] = ff_hevc_put_qpel_h3v2_neon_8;
put_hevc_qpel_neon[3][1] = ff_hevc_put_qpel_h1v3_neon_8;
put_hevc_qpel_neon[3][2] = ff_hevc_put_qpel_h2v3_neon_8;
put_hevc_qpel_neon[3][3] = ff_hevc_put_qpel_h3v3_neon_8;
put_hevc_qpel_uw_neon[1][0] = ff_hevc_put_qpel_uw_v1_neon_8;
put_hevc_qpel_uw_neon[2][0] = ff_hevc_put_qpel_uw_v2_neon_8;
put_hevc_qpel_uw_neon[3][0] = ff_hevc_put_qpel_uw_v3_neon_8;
put_hevc_qpel_uw_neon[0][1] = ff_hevc_put_qpel_uw_h1_neon_8;
put_hevc_qpel_uw_neon[0][2] = ff_hevc_put_qpel_uw_h2_neon_8;
put_hevc_qpel_uw_neon[0][3] = ff_hevc_put_qpel_uw_h3_neon_8;
put_hevc_qpel_uw_neon[1][1] = ff_hevc_put_qpel_uw_h1v1_neon_8;
put_hevc_qpel_uw_neon[1][2] = ff_hevc_put_qpel_uw_h2v1_neon_8;
put_hevc_qpel_uw_neon[1][3] = ff_hevc_put_qpel_uw_h3v1_neon_8;
put_hevc_qpel_uw_neon[2][1] = ff_hevc_put_qpel_uw_h1v2_neon_8;
put_hevc_qpel_uw_neon[2][2] = ff_hevc_put_qpel_uw_h2v2_neon_8;
put_hevc_qpel_uw_neon[2][3] = ff_hevc_put_qpel_uw_h3v2_neon_8;
put_hevc_qpel_uw_neon[3][1] = ff_hevc_put_qpel_uw_h1v3_neon_8;
put_hevc_qpel_uw_neon[3][2] = ff_hevc_put_qpel_uw_h2v3_neon_8;
put_hevc_qpel_uw_neon[3][3] = ff_hevc_put_qpel_uw_h3v3_neon_8;
for (x = 0; x < 10; x++) {
c->put_hevc_qpel[x][1][0] = ff_hevc_put_qpel_neon_wrapper;
c->put_hevc_qpel[x][0][1] = ff_hevc_put_qpel_neon_wrapper;
c->put_hevc_qpel[x][1][1] = ff_hevc_put_qpel_neon_wrapper;
c->put_hevc_qpel_uni[x][1][0] = ff_hevc_put_qpel_uni_neon_wrapper;
c->put_hevc_qpel_uni[x][0][1] = ff_hevc_put_qpel_uni_neon_wrapper;
c->put_hevc_qpel_uni[x][1][1] = ff_hevc_put_qpel_uni_neon_wrapper;
c->put_hevc_qpel_bi[x][1][0] = ff_hevc_put_qpel_bi_neon_wrapper;
c->put_hevc_qpel_bi[x][0][1] = ff_hevc_put_qpel_bi_neon_wrapper;
c->put_hevc_qpel_bi[x][1][1] = ff_hevc_put_qpel_bi_neon_wrapper;
}
c->put_hevc_qpel[0][0][0] = ff_hevc_put_pixels_w2_neon_8;
c->put_hevc_qpel[1][0][0] = ff_hevc_put_pixels_w4_neon_8;
c->put_hevc_qpel[2][0][0] = ff_hevc_put_pixels_w6_neon_8;
c->put_hevc_qpel[3][0][0] = ff_hevc_put_pixels_w8_neon_8;
c->put_hevc_qpel[4][0][0] = ff_hevc_put_pixels_w12_neon_8;
c->put_hevc_qpel[5][0][0] = ff_hevc_put_pixels_w16_neon_8;
c->put_hevc_qpel[6][0][0] = ff_hevc_put_pixels_w24_neon_8;
c->put_hevc_qpel[7][0][0] = ff_hevc_put_pixels_w32_neon_8;
c->put_hevc_qpel[8][0][0] = ff_hevc_put_pixels_w48_neon_8;
c->put_hevc_qpel[9][0][0] = ff_hevc_put_pixels_w64_neon_8;
c->put_hevc_qpel_uni[1][0][0] = ff_hevc_put_qpel_uw_pixels_w4_neon_8;
c->put_hevc_qpel_uni[3][0][0] = ff_hevc_put_qpel_uw_pixels_w8_neon_8;
c->put_hevc_qpel_uni[5][0][0] = ff_hevc_put_qpel_uw_pixels_w16_neon_8;
c->put_hevc_qpel_uni[6][0][0] = ff_hevc_put_qpel_uw_pixels_w24_neon_8;
c->put_hevc_qpel_uni[7][0][0] = ff_hevc_put_qpel_uw_pixels_w32_neon_8;
c->put_hevc_qpel_uni[8][0][0] = ff_hevc_put_qpel_uw_pixels_w48_neon_8;
c->put_hevc_qpel_uni[9][0][0] = ff_hevc_put_qpel_uw_pixels_w64_neon_8;
}
#endif // HAVE_NEON
}
void ff_hevcdsp_init_arm(HEVCDSPContext *c, const int bit_depth)
{
int cpu_flags = av_get_cpu_flags();
if (have_neon(cpu_flags))
hevcdsp_init_neon(c, bit_depth);
}
|
