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
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
|
/*
* FFv1 codec
*
* Copyright (c) 2024 Lynne <dev@lynne.ee>
*
* 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
*/
#ifndef GOLOMB
#ifdef CACHED_SYMBOL_READER
shared uint8_t state[CONTEXT_SIZE];
#define READ(c, off) get_rac_direct(c, state[off])
#else
#define READ(c, off) get_rac(c, uint64_t(slice_state) + (state_off + off))
#endif
int get_isymbol(inout RangeCoder c, uint state_off)
{
if (READ(c, 0))
return 0;
uint e = 1;
for (; e < 33; e++)
if (!READ(c, min(e, 10)))
break;
if (expectEXT(e == 1, false)) {
return READ(c, 11) ? -1 : 1;
} else if (expectEXT(e == 33, false)) {
corrupt = true;
return 0;
}
int a = 1;
for (uint i = e + 20; i >= 22; i--) {
a <<= 1;
a |= int(READ(c, min(i, 31)));
}
return READ(c, min(e + 10, 21)) ? -a : a;
}
void decode_line_pcm(inout SliceContext sc, ivec2 sp, int w, int y, int p, int bits)
{
#ifdef CACHED_SYMBOL_READER
if (gl_LocalInvocationID.x > 0)
return;
#endif
#ifndef RGB
if (p > 0 && p < 3) {
w = ceil_rshift(w, chroma_shift.x);
sp >>= chroma_shift;
}
#endif
for (int x = 0; x < w; x++) {
uint v = 0;
for (int i = (bits - 1); i >= 0; i--)
v |= uint(get_rac_equi(sc.c)) << i;
imageStore(dec[p], sp + LADDR(ivec2(x, y)), uvec4(v));
}
}
void decode_line(inout SliceContext sc, ivec2 sp, int w,
int y, int p, int bits, uint state_off,
uint8_t quant_table_idx, const int run_index)
{
#ifndef RGB
if (p > 0 && p < 3) {
w = ceil_rshift(w, chroma_shift.x);
sp >>= chroma_shift;
}
#endif
for (int x = 0; x < w; x++) {
ivec2 pr = get_pred(dec[p], sp, ivec2(x, y), 0, w,
quant_table_idx, extend_lookup[quant_table_idx] > 0);
uint context_off = state_off + CONTEXT_SIZE*abs(pr[0]);
#ifdef CACHED_SYMBOL_READER
u8buf sb = u8buf(uint64_t(slice_state) + context_off + gl_LocalInvocationID.x);
state[gl_LocalInvocationID.x] = sb.v;
barrier();
if (gl_LocalInvocationID.x == 0) {
#endif
int diff = get_isymbol(sc.c, context_off);
if (pr[0] < 0)
diff = -diff;
uint v = zero_extend(pr[1] + diff, bits);
imageStore(dec[p], sp + LADDR(ivec2(x, y)), uvec4(v));
#ifdef CACHED_SYMBOL_READER
}
barrier();
sb.v = state[gl_LocalInvocationID.x];
#endif
}
}
#else /* GOLOMB */
void decode_line(inout SliceContext sc, ivec2 sp, int w,
int y, int p, int bits, uint state_off,
uint8_t quant_table_idx, inout int run_index)
{
#ifndef RGB
if (p > 0 && p < 3) {
w = ceil_rshift(w, chroma_shift.x);
sp >>= chroma_shift;
}
#endif
int run_count = 0;
int run_mode = 0;
for (int x = 0; x < w; x++) {
ivec2 pos = sp + ivec2(x, y);
int diff;
ivec2 pr = get_pred(dec[p], sp, ivec2(x, y), 0, w,
quant_table_idx, extend_lookup[quant_table_idx] > 0);
uint context_off = state_off + VLC_STATE_SIZE*abs(pr[0]);
VlcState sb = VlcState(uint64_t(slice_state) + context_off);
if (pr[0] == 0 && run_mode == 0)
run_mode = 1;
if (run_mode != 0) {
if (run_count == 0 && run_mode == 1) {
int tmp_idx = int(log2_run[run_index]);
if (get_bit(sc.gb)) {
run_count = 1 << tmp_idx;
if (x + run_count <= w)
run_index++;
} else {
if (tmp_idx != 0) {
run_count = int(get_bits(sc.gb, tmp_idx));
} else
run_count = 0;
if (run_index != 0)
run_index--;
run_mode = 2;
}
}
run_count--;
if (run_count < 0) {
run_mode = 0;
run_count = 0;
diff = read_vlc_symbol(sc.gb, sb, bits);
if (diff >= 0)
diff++;
} else {
diff = 0;
}
} else {
diff = read_vlc_symbol(sc.gb, sb, bits);
}
if (pr[0] < 0)
diff = -diff;
uint v = zero_extend(pr[1] + diff, bits);
imageStore(dec[p], sp + LADDR(ivec2(x, y)), uvec4(v));
}
}
#endif
#ifdef RGB
ivec4 transform_sample(ivec4 pix, ivec2 rct_coef)
{
pix.b -= rct_offset;
pix.r -= rct_offset;
pix.g -= (pix.b*rct_coef.y + pix.r*rct_coef.x) >> 2;
pix.b += pix.g;
pix.r += pix.g;
return ivec4(pix[fmt_lut[0]], pix[fmt_lut[1]],
pix[fmt_lut[2]], pix[fmt_lut[3]]);
}
void writeout_rgb(in SliceContext sc, ivec2 sp, int w, int y, bool apply_rct)
{
for (uint x = gl_LocalInvocationID.x; x < w; x += gl_WorkGroupSize.x) {
ivec2 lpos = sp + LADDR(ivec2(x, y));
ivec2 pos = sc.slice_pos + ivec2(x, y);
ivec4 pix;
pix.r = int(imageLoad(dec[2], lpos)[0]);
pix.g = int(imageLoad(dec[0], lpos)[0]);
pix.b = int(imageLoad(dec[1], lpos)[0]);
if (transparency != 0)
pix.a = int(imageLoad(dec[3], lpos)[0]);
if (expectEXT(apply_rct, true))
pix = transform_sample(pix, sc.slice_rct_coef);
else
pix = ivec4(pix[fmt_lut[0]], pix[fmt_lut[1]],
pix[fmt_lut[2]], pix[fmt_lut[3]]);
imageStore(dst[0], pos, pix);
if (planar_rgb != 0) {
for (int i = 1; i < color_planes; i++)
imageStore(dst[i], pos, ivec4(pix[i]));
}
}
}
#endif
void decode_slice(inout SliceContext sc, const uint slice_idx)
{
int w = sc.slice_dim.x;
ivec2 sp = sc.slice_pos;
#ifndef RGB
int bits = bits_per_raw_sample;
#else
int bits = 9;
if (bits != 8 || sc.slice_coding_mode != 0)
bits = bits_per_raw_sample + int(sc.slice_coding_mode != 1);
sp.y = int(gl_WorkGroupID.y)*RGB_LINECACHE;
#endif
/* PCM coding */
#ifndef GOLOMB
if (sc.slice_coding_mode == 1) {
#ifndef RGB
for (int p = 0; p < planes; p++) {
int h = sc.slice_dim.y;
if (p > 0 && p < 3)
h = ceil_rshift(h, chroma_shift.y);
for (int y = 0; y < h; y++)
decode_line_pcm(sc, sp, w, y, p, bits);
}
#else
for (int y = 0; y < sc.slice_dim.y; y++) {
for (int p = 0; p < color_planes; p++)
decode_line_pcm(sc, sp, w, y, p, bits);
writeout_rgb(sc, sp, w, y, false);
}
#endif
} else
/* Arithmetic coding */
#endif
{
u8vec4 quant_table_idx = sc.quant_table_idx.xyyz;
u32vec4 slice_state_off = (slice_idx*codec_planes + uvec4(0, 1, 1, 2))*plane_state_size;
#ifndef RGB
for (int p = 0; p < planes; p++) {
int h = sc.slice_dim.y;
if (p > 0 && p < 3)
h = ceil_rshift(h, chroma_shift.y);
int run_index = 0;
for (int y = 0; y < h; y++)
decode_line(sc, sp, w, y, p, bits,
slice_state_off[p], quant_table_idx[p], run_index);
}
#else
int run_index = 0;
for (int y = 0; y < sc.slice_dim.y; y++) {
for (int p = 0; p < color_planes; p++)
decode_line(sc, sp, w, y, p, bits,
slice_state_off[p], quant_table_idx[p], run_index);
writeout_rgb(sc, sp, w, y, true);
}
#endif
}
}
void main(void)
{
const uint slice_idx = gl_WorkGroupID.y*gl_NumWorkGroups.x + gl_WorkGroupID.x;
decode_slice(slice_ctx[slice_idx], slice_idx);
uint32_t status = corrupt ? uint32_t(corrupt) : overread;
if (status != 0)
slice_status[2*slice_idx + 1] = status;
}
|
