summaryrefslogtreecommitdiffstats
path: root/media/ffvpx/libavcodec/simple_idct_template.c
blob: 5ddd0b45a2b717628dfa0505bae4c7b6e8d647ba (plain)
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
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
/*
 * Simple IDCT
 *
 * Copyright (c) 2001 Michael Niedermayer <michaelni@gmx.at>
 *
 * 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
 */

/**
 * @file
 * simpleidct in C.
 */

/* Based upon some commented-out C code from mpeg2dec (idct_mmx.c
 * written by Aaron Holtzman <aholtzma@ess.engr.uvic.ca>). */

#include "simple_idct.h"

#include "bit_depth_template.c"

#undef W1
#undef W2
#undef W3
#undef W4
#undef W5
#undef W6
#undef W7
#undef ROW_SHIFT
#undef COL_SHIFT
#undef DC_SHIFT
#undef MUL
#undef MAC

#if BIT_DEPTH == 8

#define W1  22725  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
#define W2  21407  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
#define W3  19266  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
#define W4  16383  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
#define W5  12873  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
#define W6  8867   //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
#define W7  4520   //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5

#define ROW_SHIFT 11
#define COL_SHIFT 20
#define DC_SHIFT 3

#define MUL(a, b)    MUL16(a, b)
#define MAC(a, b, c) MAC16(a, b, c)

#elif BIT_DEPTH == 10 || BIT_DEPTH == 12

# if BIT_DEPTH == 10
#define W1 22725 // 90901
#define W2 21407 //  85627
#define W3 19265 //  77062
#define W4 16384 //  65535
#define W5 12873 //  51491
#define W6  8867 //  35468
#define W7  4520 //  18081

#   ifdef EXTRA_SHIFT
#define ROW_SHIFT 13
#define COL_SHIFT 18
#define DC_SHIFT  1
#   elif IN_IDCT_DEPTH == 32
#define ROW_SHIFT 13
#define COL_SHIFT 21
#define DC_SHIFT  2
#   else
#define ROW_SHIFT 12
#define COL_SHIFT 19
#define DC_SHIFT  2
#   endif

# else
#define W1 45451
#define W2 42813
#define W3 38531
#define W4 32767
#define W5 25746
#define W6 17734
#define W7 9041

#define ROW_SHIFT 16
#define COL_SHIFT 17
#define DC_SHIFT -1
# endif

#define MUL(a, b)    ((int)((SUINT)(a) * (b)))
#define MAC(a, b, c) ((a) += (SUINT)(b) * (c))

#else

#error "Unsupported bitdepth"

#endif

#ifdef EXTRA_SHIFT
static inline void FUNC(idctRowCondDC_extrashift)(int16_t *row, int extra_shift)
#else
static inline void FUNC6(idctRowCondDC)(idctin *row, int extra_shift)
#endif
{
    SUINT a0, a1, a2, a3, b0, b1, b2, b3;

// TODO: Add DC-only support for int32_t input
#if IN_IDCT_DEPTH == 16
#if HAVE_FAST_64BIT
#define ROW0_MASK (0xffffULL << 48 * HAVE_BIGENDIAN)
    if (((AV_RN64A(row) & ~ROW0_MASK) | AV_RN64A(row+4)) == 0) {
        uint64_t temp;
        if (DC_SHIFT - extra_shift >= 0) {
            temp = (row[0] * (1 << (DC_SHIFT - extra_shift))) & 0xffff;
        } else {
            temp = ((row[0] + (1<<(extra_shift - DC_SHIFT-1))) >> (extra_shift - DC_SHIFT)) & 0xffff;
        }
        temp += temp * (1 << 16);
        temp += temp * ((uint64_t) 1 << 32);
        AV_WN64A(row, temp);
        AV_WN64A(row + 4, temp);
        return;
    }
#else
    if (!(AV_RN32A(row+2) |
          AV_RN32A(row+4) |
          AV_RN32A(row+6) |
          row[1])) {
        uint32_t temp;
        if (DC_SHIFT - extra_shift >= 0) {
            temp = (row[0] * (1 << (DC_SHIFT - extra_shift))) & 0xffff;
        } else {
            temp = ((row[0] + (1<<(extra_shift - DC_SHIFT-1))) >> (extra_shift - DC_SHIFT)) & 0xffff;
        }
        temp += temp * (1 << 16);
        AV_WN32A(row, temp);
        AV_WN32A(row+2, temp);
        AV_WN32A(row+4, temp);
        AV_WN32A(row+6, temp);
        return;
    }
#endif
#endif

    a0 = ((SUINT)W4 * row[0]) + (1 << (ROW_SHIFT + extra_shift - 1));
    a1 = a0;
    a2 = a0;
    a3 = a0;

    a0 += (SUINT)W2 * row[2];
    a1 += (SUINT)W6 * row[2];
    a2 -= (SUINT)W6 * row[2];
    a3 -= (SUINT)W2 * row[2];

    b0 = MUL(W1, row[1]);
    MAC(b0, W3, row[3]);
    b1 = MUL(W3, row[1]);
    MAC(b1, -W7, row[3]);
    b2 = MUL(W5, row[1]);
    MAC(b2, -W1, row[3]);
    b3 = MUL(W7, row[1]);
    MAC(b3, -W5, row[3]);

#if IN_IDCT_DEPTH == 32
    if (AV_RN64A(row + 4) | AV_RN64A(row + 6)) {
#else
    if (AV_RN64A(row + 4)) {
#endif
        a0 += (SUINT)  W4*row[4] + (SUINT)W6*row[6];
        a1 += (SUINT)- W4*row[4] - (SUINT)W2*row[6];
        a2 += (SUINT)- W4*row[4] + (SUINT)W2*row[6];
        a3 += (SUINT)  W4*row[4] - (SUINT)W6*row[6];

        MAC(b0,  W5, row[5]);
        MAC(b0,  W7, row[7]);

        MAC(b1, -W1, row[5]);
        MAC(b1, -W5, row[7]);

        MAC(b2,  W7, row[5]);
        MAC(b2,  W3, row[7]);

        MAC(b3,  W3, row[5]);
        MAC(b3, -W1, row[7]);
    }

    row[0] = (int)(a0 + b0) >> (ROW_SHIFT + extra_shift);
    row[7] = (int)(a0 - b0) >> (ROW_SHIFT + extra_shift);
    row[1] = (int)(a1 + b1) >> (ROW_SHIFT + extra_shift);
    row[6] = (int)(a1 - b1) >> (ROW_SHIFT + extra_shift);
    row[2] = (int)(a2 + b2) >> (ROW_SHIFT + extra_shift);
    row[5] = (int)(a2 - b2) >> (ROW_SHIFT + extra_shift);
    row[3] = (int)(a3 + b3) >> (ROW_SHIFT + extra_shift);
    row[4] = (int)(a3 - b3) >> (ROW_SHIFT + extra_shift);
}

#define IDCT_COLS do {                                  \
        a0 = (SUINT)W4 * (col[8*0] + ((1<<(COL_SHIFT-1))/W4)); \
        a1 = a0;                                        \
        a2 = a0;                                        \
        a3 = a0;                                        \
                                                        \
        a0 += (SUINT) W2*col[8*2];                             \
        a1 += (SUINT) W6*col[8*2];                             \
        a2 += (SUINT)-W6*col[8*2];                             \
        a3 += (SUINT)-W2*col[8*2];                             \
                                                        \
        b0 = MUL(W1, col[8*1]);                         \
        b1 = MUL(W3, col[8*1]);                         \
        b2 = MUL(W5, col[8*1]);                         \
        b3 = MUL(W7, col[8*1]);                         \
                                                        \
        MAC(b0,  W3, col[8*3]);                         \
        MAC(b1, -W7, col[8*3]);                         \
        MAC(b2, -W1, col[8*3]);                         \
        MAC(b3, -W5, col[8*3]);                         \
                                                        \
        if (col[8*4]) {                                 \
            a0 += (SUINT) W4*col[8*4];                         \
            a1 += (SUINT)-W4*col[8*4];                         \
            a2 += (SUINT)-W4*col[8*4];                         \
            a3 += (SUINT) W4*col[8*4];                         \
        }                                               \
                                                        \
        if (col[8*5]) {                                 \
            MAC(b0,  W5, col[8*5]);                     \
            MAC(b1, -W1, col[8*5]);                     \
            MAC(b2,  W7, col[8*5]);                     \
            MAC(b3,  W3, col[8*5]);                     \
        }                                               \
                                                        \
        if (col[8*6]) {                                 \
            a0 += (SUINT) W6*col[8*6];                         \
            a1 += (SUINT)-W2*col[8*6];                         \
            a2 += (SUINT) W2*col[8*6];                         \
            a3 += (SUINT)-W6*col[8*6];                         \
        }                                               \
                                                        \
        if (col[8*7]) {                                 \
            MAC(b0,  W7, col[8*7]);                     \
            MAC(b1, -W5, col[8*7]);                     \
            MAC(b2,  W3, col[8*7]);                     \
            MAC(b3, -W1, col[8*7]);                     \
        }                                               \
    } while (0)

#ifdef EXTRA_SHIFT
static inline void FUNC(idctSparseCol_extrashift)(int16_t *col)
#else
static inline void FUNC6(idctSparseColPut)(pixel *dest, ptrdiff_t line_size,
                                          idctin *col)
{
    SUINT a0, a1, a2, a3, b0, b1, b2, b3;

    IDCT_COLS;

    dest[0] = av_clip_pixel((int)(a0 + b0) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a1 + b1) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a2 + b2) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a3 + b3) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a3 - b3) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a2 - b2) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a1 - b1) >> COL_SHIFT);
    dest += line_size;
    dest[0] = av_clip_pixel((int)(a0 - b0) >> COL_SHIFT);
}

static inline void FUNC6(idctSparseColAdd)(pixel *dest, ptrdiff_t line_size,
                                          idctin *col)
{
    unsigned a0, a1, a2, a3, b0, b1, b2, b3;

    IDCT_COLS;

    dest[0] = av_clip_pixel(dest[0] + ((int)(a0 + b0) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a1 + b1) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a2 + b2) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a3 + b3) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a3 - b3) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a2 - b2) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a1 - b1) >> COL_SHIFT));
    dest += line_size;
    dest[0] = av_clip_pixel(dest[0] + ((int)(a0 - b0) >> COL_SHIFT));
}

static inline void FUNC6(idctSparseCol)(idctin *col)
#endif
{
    unsigned a0, a1, a2, a3, b0, b1, b2, b3;

    IDCT_COLS;

    col[0 ] = ((int)(a0 + b0) >> COL_SHIFT);
    col[8 ] = ((int)(a1 + b1) >> COL_SHIFT);
    col[16] = ((int)(a2 + b2) >> COL_SHIFT);
    col[24] = ((int)(a3 + b3) >> COL_SHIFT);
    col[32] = ((int)(a3 - b3) >> COL_SHIFT);
    col[40] = ((int)(a2 - b2) >> COL_SHIFT);
    col[48] = ((int)(a1 - b1) >> COL_SHIFT);
    col[56] = ((int)(a0 - b0) >> COL_SHIFT);
}

#ifndef EXTRA_SHIFT
void FUNC6(ff_simple_idct_put)(uint8_t *dest_, ptrdiff_t line_size, int16_t *block_)
{
    idctin *block = (idctin *)block_;
    pixel *dest = (pixel *)dest_;
    int i;

    line_size /= sizeof(pixel);

    for (i = 0; i < 8; i++)
        FUNC6(idctRowCondDC)(block + i*8, 0);

    for (i = 0; i < 8; i++)
        FUNC6(idctSparseColPut)(dest + i, line_size, block + i);
}

#if IN_IDCT_DEPTH == 16
void FUNC6(ff_simple_idct_add)(uint8_t *dest_, ptrdiff_t line_size, int16_t *block)
{
    pixel *dest = (pixel *)dest_;
    int i;

    line_size /= sizeof(pixel);

    for (i = 0; i < 8; i++)
        FUNC6(idctRowCondDC)(block + i*8, 0);

    for (i = 0; i < 8; i++)
        FUNC6(idctSparseColAdd)(dest + i, line_size, block + i);
}

void FUNC6(ff_simple_idct)(int16_t *block)
{
    int i;

    for (i = 0; i < 8; i++)
        FUNC6(idctRowCondDC)(block + i*8, 0);

    for (i = 0; i < 8; i++)
        FUNC6(idctSparseCol)(block + i);
}
#endif
#endif