aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/vc1dsp.c
blob: 5773ab1e3c7dd64a4287e90cd5b362fe1320225f (plain) (blame)
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
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
/*
 * VC-1 and WMV3 decoder - DSP functions
 * Copyright (c) 2006 Konstantin Shishkov
 *
 * 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 libavcodec/vc1dsp.c
 * VC-1 and WMV3 decoder
 *
 */

#include "dsputil.h"


/** Apply overlap transform to horizontal edge
*/
static void vc1_v_overlap_c(uint8_t* src, int stride)
{
    int i;
    int a, b, c, d;
    int d1, d2;
    int rnd = 1;
    for(i = 0; i < 8; i++) {
        a = src[-2*stride];
        b = src[-stride];
        c = src[0];
        d = src[stride];
        d1 = (a - d + 3 + rnd) >> 3;
        d2 = (a - d + b - c + 4 - rnd) >> 3;

        src[-2*stride] = a - d1;
        src[-stride] = av_clip_uint8(b - d2);
        src[0] = av_clip_uint8(c + d2);
        src[stride] = d + d1;
        src++;
        rnd = !rnd;
    }
}

/** Apply overlap transform to vertical edge
*/
static void vc1_h_overlap_c(uint8_t* src, int stride)
{
    int i;
    int a, b, c, d;
    int d1, d2;
    int rnd = 1;
    for(i = 0; i < 8; i++) {
        a = src[-2];
        b = src[-1];
        c = src[0];
        d = src[1];
        d1 = (a - d + 3 + rnd) >> 3;
        d2 = (a - d + b - c + 4 - rnd) >> 3;

        src[-2] = a - d1;
        src[-1] = av_clip_uint8(b - d2);
        src[0] = av_clip_uint8(c + d2);
        src[1] = d + d1;
        src += stride;
        rnd = !rnd;
    }
}

/**
 * VC-1 in-loop deblocking filter for one line
 * @param src source block type
 * @param stride block stride
 * @param pq block quantizer
 * @return whether other 3 pairs should be filtered or not
 * @see 8.6
 */
static av_always_inline int vc1_filter_line(uint8_t* src, int stride, int pq){
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;

    int a0 = (2*(src[-2*stride] - src[ 1*stride]) - 5*(src[-1*stride] - src[ 0*stride]) + 4) >> 3;
    int a0_sign = a0 >> 31;        /* Store sign */
    a0 = (a0 ^ a0_sign) - a0_sign; /* a0 = FFABS(a0); */
    if(a0 < pq){
        int a1 = FFABS((2*(src[-4*stride] - src[-1*stride]) - 5*(src[-3*stride] - src[-2*stride]) + 4) >> 3);
        int a2 = FFABS((2*(src[ 0*stride] - src[ 3*stride]) - 5*(src[ 1*stride] - src[ 2*stride]) + 4) >> 3);
        if(a1 < a0 || a2 < a0){
            int clip = src[-1*stride] - src[ 0*stride];
            int clip_sign = clip >> 31;
            clip = ((clip ^ clip_sign) - clip_sign)>>1;
            if(clip){
                int a3 = FFMIN(a1, a2);
                int d = 5 * (a3 - a0);
                int d_sign = (d >> 31);
                d = ((d ^ d_sign) - d_sign) >> 3;
                d_sign ^= a0_sign;

                if( d_sign ^ clip_sign )
                    d = 0;
                else{
                    d = FFMIN(d, clip);
                    d = (d ^ d_sign) - d_sign;          /* Restore sign */
                    src[-1*stride] = cm[src[-1*stride] - d];
                    src[ 0*stride] = cm[src[ 0*stride] + d];
                }
                return 1;
            }
        }
    }
    return 0;
}

/**
 * VC-1 in-loop deblocking filter
 * @param src source block type
 * @param step distance between horizontally adjacent elements
 * @param stride distance between vertically adjacent elements
 * @param len edge length to filter (4 or 8 pixels)
 * @param pq block quantizer
 * @see 8.6
 */
static inline void vc1_loop_filter(uint8_t* src, int step, int stride, int len, int pq)
{
    int i;
    int filt3;

    for(i = 0; i < len; i += 4){
        filt3 = vc1_filter_line(src + 2*step, stride, pq);
        if(filt3){
            vc1_filter_line(src + 0*step, stride, pq);
            vc1_filter_line(src + 1*step, stride, pq);
            vc1_filter_line(src + 3*step, stride, pq);
        }
        src += step * 4;
    }
}

static void vc1_v_loop_filter4_c(uint8_t *src, int stride, int pq)
{
    vc1_loop_filter(src, 1, stride, 4, pq);
}

static void vc1_h_loop_filter4_c(uint8_t *src, int stride, int pq)
{
    vc1_loop_filter(src, stride, 1, 4, pq);
}

static void vc1_v_loop_filter8_c(uint8_t *src, int stride, int pq)
{
    vc1_loop_filter(src, 1, stride, 8, pq);
}

static void vc1_h_loop_filter8_c(uint8_t *src, int stride, int pq)
{
    vc1_loop_filter(src, stride, 1, 8, pq);
}

static void vc1_v_loop_filter16_c(uint8_t *src, int stride, int pq)
{
    vc1_loop_filter(src, 1, stride, 16, pq);
}

static void vc1_h_loop_filter16_c(uint8_t *src, int stride, int pq)
{
    vc1_loop_filter(src, stride, 1, 16, pq);
}

/** Do inverse transform on 8x8 block
*/
static void vc1_inv_trans_8x8_c(DCTELEM block[64])
{
    int i;
    register int t1,t2,t3,t4,t5,t6,t7,t8;
    DCTELEM *src, *dst;

    src = block;
    dst = block;
    for(i = 0; i < 8; i++){
        t1 = 12 * (src[0] + src[4]) + 4;
        t2 = 12 * (src[0] - src[4]) + 4;
        t3 = 16 * src[2] +  6 * src[6];
        t4 =  6 * src[2] - 16 * src[6];

        t5 = t1 + t3;
        t6 = t2 + t4;
        t7 = t2 - t4;
        t8 = t1 - t3;

        t1 = 16 * src[1] + 15 * src[3] +  9 * src[5] +  4 * src[7];
        t2 = 15 * src[1] -  4 * src[3] - 16 * src[5] -  9 * src[7];
        t3 =  9 * src[1] - 16 * src[3] +  4 * src[5] + 15 * src[7];
        t4 =  4 * src[1] -  9 * src[3] + 15 * src[5] - 16 * src[7];

        dst[0] = (t5 + t1) >> 3;
        dst[1] = (t6 + t2) >> 3;
        dst[2] = (t7 + t3) >> 3;
        dst[3] = (t8 + t4) >> 3;
        dst[4] = (t8 - t4) >> 3;
        dst[5] = (t7 - t3) >> 3;
        dst[6] = (t6 - t2) >> 3;
        dst[7] = (t5 - t1) >> 3;

        src += 8;
        dst += 8;
    }

    src = block;
    dst = block;
    for(i = 0; i < 8; i++){
        t1 = 12 * (src[ 0] + src[32]) + 64;
        t2 = 12 * (src[ 0] - src[32]) + 64;
        t3 = 16 * src[16] +  6 * src[48];
        t4 =  6 * src[16] - 16 * src[48];

        t5 = t1 + t3;
        t6 = t2 + t4;
        t7 = t2 - t4;
        t8 = t1 - t3;

        t1 = 16 * src[ 8] + 15 * src[24] +  9 * src[40] +  4 * src[56];
        t2 = 15 * src[ 8] -  4 * src[24] - 16 * src[40] -  9 * src[56];
        t3 =  9 * src[ 8] - 16 * src[24] +  4 * src[40] + 15 * src[56];
        t4 =  4 * src[ 8] -  9 * src[24] + 15 * src[40] - 16 * src[56];

        dst[ 0] = (t5 + t1) >> 7;
        dst[ 8] = (t6 + t2) >> 7;
        dst[16] = (t7 + t3) >> 7;
        dst[24] = (t8 + t4) >> 7;
        dst[32] = (t8 - t4 + 1) >> 7;
        dst[40] = (t7 - t3 + 1) >> 7;
        dst[48] = (t6 - t2 + 1) >> 7;
        dst[56] = (t5 - t1 + 1) >> 7;

        src++;
        dst++;
    }
}

/** Do inverse transform on 8x4 part of block
*/
static void vc1_inv_trans_8x4_c(uint8_t *dest, int linesize, DCTELEM *block)
{
    int i;
    register int t1,t2,t3,t4,t5,t6,t7,t8;
    DCTELEM *src, *dst;
    const uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;

    src = block;
    dst = block;
    for(i = 0; i < 4; i++){
        t1 = 12 * (src[0] + src[4]) + 4;
        t2 = 12 * (src[0] - src[4]) + 4;
        t3 = 16 * src[2] +  6 * src[6];
        t4 =  6 * src[2] - 16 * src[6];

        t5 = t1 + t3;
        t6 = t2 + t4;
        t7 = t2 - t4;
        t8 = t1 - t3;

        t1 = 16 * src[1] + 15 * src[3] +  9 * src[5] +  4 * src[7];
        t2 = 15 * src[1] -  4 * src[3] - 16 * src[5] -  9 * src[7];
        t3 =  9 * src[1] - 16 * src[3] +  4 * src[5] + 15 * src[7];
        t4 =  4 * src[1] -  9 * src[3] + 15 * src[5] - 16 * src[7];

        dst[0] = (t5 + t1) >> 3;
        dst[1] = (t6 + t2) >> 3;
        dst[2] = (t7 + t3) >> 3;
        dst[3] = (t8 + t4) >> 3;
        dst[4] = (t8 - t4) >> 3;
        dst[5] = (t7 - t3) >> 3;
        dst[6] = (t6 - t2) >> 3;
        dst[7] = (t5 - t1) >> 3;

        src += 8;
        dst += 8;
    }

    src = block;
    for(i = 0; i < 8; i++){
        t1 = 17 * (src[ 0] + src[16]) + 64;
        t2 = 17 * (src[ 0] - src[16]) + 64;
        t3 = 22 * src[ 8] + 10 * src[24];
        t4 = 22 * src[24] - 10 * src[ 8];

        dest[0*linesize] = cm[dest[0*linesize] + ((t1 + t3) >> 7)];
        dest[1*linesize] = cm[dest[1*linesize] + ((t2 - t4) >> 7)];
        dest[2*linesize] = cm[dest[2*linesize] + ((t2 + t4) >> 7)];
        dest[3*linesize] = cm[dest[3*linesize] + ((t1 - t3) >> 7)];

        src ++;
        dest++;
    }
}

/** Do inverse transform on 4x8 parts of block
*/
static void vc1_inv_trans_4x8_c(uint8_t *dest, int linesize, DCTELEM *block)
{
    int i;
    register int t1,t2,t3,t4,t5,t6,t7,t8;
    DCTELEM *src, *dst;
    const uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;

    src = block;
    dst = block;
    for(i = 0; i < 8; i++){
        t1 = 17 * (src[0] + src[2]) + 4;
        t2 = 17 * (src[0] - src[2]) + 4;
        t3 = 22 * src[1] + 10 * src[3];
        t4 = 22 * src[3] - 10 * src[1];

        dst[0] = (t1 + t3) >> 3;
        dst[1] = (t2 - t4) >> 3;
        dst[2] = (t2 + t4) >> 3;
        dst[3] = (t1 - t3) >> 3;

        src += 8;
        dst += 8;
    }

    src = block;
    for(i = 0; i < 4; i++){
        t1 = 12 * (src[ 0] + src[32]) + 64;
        t2 = 12 * (src[ 0] - src[32]) + 64;
        t3 = 16 * src[16] +  6 * src[48];
        t4 =  6 * src[16] - 16 * src[48];

        t5 = t1 + t3;
        t6 = t2 + t4;
        t7 = t2 - t4;
        t8 = t1 - t3;

        t1 = 16 * src[ 8] + 15 * src[24] +  9 * src[40] +  4 * src[56];
        t2 = 15 * src[ 8] -  4 * src[24] - 16 * src[40] -  9 * src[56];
        t3 =  9 * src[ 8] - 16 * src[24] +  4 * src[40] + 15 * src[56];
        t4 =  4 * src[ 8] -  9 * src[24] + 15 * src[40] - 16 * src[56];

        dest[0*linesize] = cm[dest[0*linesize] + ((t5 + t1) >> 7)];
        dest[1*linesize] = cm[dest[1*linesize] + ((t6 + t2) >> 7)];
        dest[2*linesize] = cm[dest[2*linesize] + ((t7 + t3) >> 7)];
        dest[3*linesize] = cm[dest[3*linesize] + ((t8 + t4) >> 7)];
        dest[4*linesize] = cm[dest[4*linesize] + ((t8 - t4 + 1) >> 7)];
        dest[5*linesize] = cm[dest[5*linesize] + ((t7 - t3 + 1) >> 7)];
        dest[6*linesize] = cm[dest[6*linesize] + ((t6 - t2 + 1) >> 7)];
        dest[7*linesize] = cm[dest[7*linesize] + ((t5 - t1 + 1) >> 7)];

        src ++;
        dest++;
    }
}

/** Do inverse transform on 4x4 part of block
*/
static void vc1_inv_trans_4x4_c(uint8_t *dest, int linesize, DCTELEM *block)
{
    int i;
    register int t1,t2,t3,t4;
    DCTELEM *src, *dst;
    const uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;

    src = block;
    dst = block;
    for(i = 0; i < 4; i++){
        t1 = 17 * (src[0] + src[2]) + 4;
        t2 = 17 * (src[0] - src[2]) + 4;
        t3 = 22 * src[1] + 10 * src[3];
        t4 = 22 * src[3] - 10 * src[1];

        dst[0] = (t1 + t3) >> 3;
        dst[1] = (t2 - t4) >> 3;
        dst[2] = (t2 + t4) >> 3;
        dst[3] = (t1 - t3) >> 3;

        src += 8;
        dst += 8;
    }

    src = block;
    for(i = 0; i < 4; i++){
        t1 = 17 * (src[ 0] + src[16]) + 64;
        t2 = 17 * (src[ 0] - src[16]) + 64;
        t3 = 22 * src[ 8] + 10 * src[24];
        t4 = 22 * src[24] - 10 * src[ 8];

        dest[0*linesize] = cm[dest[0*linesize] + ((t1 + t3) >> 7)];
        dest[1*linesize] = cm[dest[1*linesize] + ((t2 - t4) >> 7)];
        dest[2*linesize] = cm[dest[2*linesize] + ((t2 + t4) >> 7)];
        dest[3*linesize] = cm[dest[3*linesize] + ((t1 - t3) >> 7)];

        src ++;
        dest++;
    }
}

/* motion compensation functions */
/** Filter in case of 2 filters */
#define VC1_MSPEL_FILTER_16B(DIR, TYPE)                                 \
static av_always_inline int vc1_mspel_ ## DIR ## _filter_16bits(const TYPE *src, int stride, int mode) \
{                                                                       \
    switch(mode){                                                       \
    case 0: /* no shift - should not occur */                           \
        return 0;                                                       \
    case 1: /* 1/4 shift */                                             \
        return -4*src[-stride] + 53*src[0] + 18*src[stride] - 3*src[stride*2]; \
    case 2: /* 1/2 shift */                                             \
        return -src[-stride] + 9*src[0] + 9*src[stride] - src[stride*2]; \
    case 3: /* 3/4 shift */                                             \
        return -3*src[-stride] + 18*src[0] + 53*src[stride] - 4*src[stride*2]; \
    }                                                                   \
    return 0; /* should not occur */                                    \
}

VC1_MSPEL_FILTER_16B(ver, uint8_t);
VC1_MSPEL_FILTER_16B(hor, int16_t);


/** Filter used to interpolate fractional pel values
 */
static av_always_inline int vc1_mspel_filter(const uint8_t *src, int stride, int mode, int r)
{
    switch(mode){
    case 0: //no shift
        return src[0];
    case 1: // 1/4 shift
        return (-4*src[-stride] + 53*src[0] + 18*src[stride] - 3*src[stride*2] + 32 - r) >> 6;
    case 2: // 1/2 shift
        return (-src[-stride] + 9*src[0] + 9*src[stride] - src[stride*2] + 8 - r) >> 4;
    case 3: // 3/4 shift
        return (-3*src[-stride] + 18*src[0] + 53*src[stride] - 4*src[stride*2] + 32 - r) >> 6;
    }
    return 0; //should not occur
}

/** Function used to do motion compensation with bicubic interpolation
 */
#define VC1_MSPEL_MC(OP, OPNAME)\
static void OPNAME ## vc1_mspel_mc(uint8_t *dst, const uint8_t *src, int stride, int hmode, int vmode, int rnd)\
{\
    int     i, j;\
\
    if (vmode) { /* Horizontal filter to apply */\
        int r;\
\
        if (hmode) { /* Vertical filter to apply, output to tmp */\
            static const int shift_value[] = { 0, 5, 1, 5 };\
            int              shift = (shift_value[hmode]+shift_value[vmode])>>1;\
            int16_t          tmp[11*8], *tptr = tmp;\
\
            r = (1<<(shift-1)) + rnd-1;\
\
            src -= 1;\
            for(j = 0; j < 8; j++) {\
                for(i = 0; i < 11; i++)\
                    tptr[i] = (vc1_mspel_ver_filter_16bits(src + i, stride, vmode)+r)>>shift;\
                src += stride;\
                tptr += 11;\
            }\
\
            r = 64-rnd;\
            tptr = tmp+1;\
            for(j = 0; j < 8; j++) {\
                for(i = 0; i < 8; i++)\
                    OP(dst[i], (vc1_mspel_hor_filter_16bits(tptr + i, 1, hmode)+r)>>7);\
                dst += stride;\
                tptr += 11;\
            }\
\
            return;\
        }\
        else { /* No horizontal filter, output 8 lines to dst */\
            r = 1-rnd;\
\
            for(j = 0; j < 8; j++) {\
                for(i = 0; i < 8; i++)\
                    OP(dst[i], vc1_mspel_filter(src + i, stride, vmode, r));\
                src += stride;\
                dst += stride;\
            }\
            return;\
        }\
    }\
\
    /* Horizontal mode with no vertical mode */\
    for(j = 0; j < 8; j++) {\
        for(i = 0; i < 8; i++)\
            OP(dst[i], vc1_mspel_filter(src + i, 1, hmode, rnd));\
        dst += stride;\
        src += stride;\
    }\
}

#define op_put(a, b) a = av_clip_uint8(b)
#define op_avg(a, b) a = (a + av_clip_uint8(b) + 1) >> 1

VC1_MSPEL_MC(op_put, put_)
VC1_MSPEL_MC(op_avg, avg_)

/* pixel functions - really are entry points to vc1_mspel_mc */

/* this one is defined in dsputil.c */
void ff_put_vc1_mspel_mc00_c(uint8_t *dst, const uint8_t *src, int stride, int rnd);
void ff_avg_vc1_mspel_mc00_c(uint8_t *dst, const uint8_t *src, int stride, int rnd);

#define PUT_VC1_MSPEL(a, b)\
static void put_vc1_mspel_mc ## a ## b ##_c(uint8_t *dst, const uint8_t *src, int stride, int rnd) { \
     put_vc1_mspel_mc(dst, src, stride, a, b, rnd);                         \
}\
static void avg_vc1_mspel_mc ## a ## b ##_c(uint8_t *dst, const uint8_t *src, int stride, int rnd) { \
     avg_vc1_mspel_mc(dst, src, stride, a, b, rnd);                         \
}

PUT_VC1_MSPEL(1, 0)
PUT_VC1_MSPEL(2, 0)
PUT_VC1_MSPEL(3, 0)

PUT_VC1_MSPEL(0, 1)
PUT_VC1_MSPEL(1, 1)
PUT_VC1_MSPEL(2, 1)
PUT_VC1_MSPEL(3, 1)

PUT_VC1_MSPEL(0, 2)
PUT_VC1_MSPEL(1, 2)
PUT_VC1_MSPEL(2, 2)
PUT_VC1_MSPEL(3, 2)

PUT_VC1_MSPEL(0, 3)
PUT_VC1_MSPEL(1, 3)
PUT_VC1_MSPEL(2, 3)
PUT_VC1_MSPEL(3, 3)

void ff_vc1dsp_init(DSPContext* dsp, AVCodecContext *avctx) {
    dsp->vc1_inv_trans_8x8 = vc1_inv_trans_8x8_c;
    dsp->vc1_inv_trans_4x8 = vc1_inv_trans_4x8_c;
    dsp->vc1_inv_trans_8x4 = vc1_inv_trans_8x4_c;
    dsp->vc1_inv_trans_4x4 = vc1_inv_trans_4x4_c;
    dsp->vc1_h_overlap = vc1_h_overlap_c;
    dsp->vc1_v_overlap = vc1_v_overlap_c;
    dsp->vc1_v_loop_filter4 = vc1_v_loop_filter4_c;
    dsp->vc1_h_loop_filter4 = vc1_h_loop_filter4_c;
    dsp->vc1_v_loop_filter8 = vc1_v_loop_filter8_c;
    dsp->vc1_h_loop_filter8 = vc1_h_loop_filter8_c;
    dsp->vc1_v_loop_filter16 = vc1_v_loop_filter16_c;
    dsp->vc1_h_loop_filter16 = vc1_h_loop_filter16_c;

    dsp->put_vc1_mspel_pixels_tab[ 0] = ff_put_vc1_mspel_mc00_c;
    dsp->put_vc1_mspel_pixels_tab[ 1] = put_vc1_mspel_mc10_c;
    dsp->put_vc1_mspel_pixels_tab[ 2] = put_vc1_mspel_mc20_c;
    dsp->put_vc1_mspel_pixels_tab[ 3] = put_vc1_mspel_mc30_c;
    dsp->put_vc1_mspel_pixels_tab[ 4] = put_vc1_mspel_mc01_c;
    dsp->put_vc1_mspel_pixels_tab[ 5] = put_vc1_mspel_mc11_c;
    dsp->put_vc1_mspel_pixels_tab[ 6] = put_vc1_mspel_mc21_c;
    dsp->put_vc1_mspel_pixels_tab[ 7] = put_vc1_mspel_mc31_c;
    dsp->put_vc1_mspel_pixels_tab[ 8] = put_vc1_mspel_mc02_c;
    dsp->put_vc1_mspel_pixels_tab[ 9] = put_vc1_mspel_mc12_c;
    dsp->put_vc1_mspel_pixels_tab[10] = put_vc1_mspel_mc22_c;
    dsp->put_vc1_mspel_pixels_tab[11] = put_vc1_mspel_mc32_c;
    dsp->put_vc1_mspel_pixels_tab[12] = put_vc1_mspel_mc03_c;
    dsp->put_vc1_mspel_pixels_tab[13] = put_vc1_mspel_mc13_c;
    dsp->put_vc1_mspel_pixels_tab[14] = put_vc1_mspel_mc23_c;
    dsp->put_vc1_mspel_pixels_tab[15] = put_vc1_mspel_mc33_c;

    dsp->avg_vc1_mspel_pixels_tab[ 0] = ff_avg_vc1_mspel_mc00_c;
    dsp->avg_vc1_mspel_pixels_tab[ 1] = avg_vc1_mspel_mc10_c;
    dsp->avg_vc1_mspel_pixels_tab[ 2] = avg_vc1_mspel_mc20_c;
    dsp->avg_vc1_mspel_pixels_tab[ 3] = avg_vc1_mspel_mc30_c;
    dsp->avg_vc1_mspel_pixels_tab[ 4] = avg_vc1_mspel_mc01_c;
    dsp->avg_vc1_mspel_pixels_tab[ 5] = avg_vc1_mspel_mc11_c;
    dsp->avg_vc1_mspel_pixels_tab[ 6] = avg_vc1_mspel_mc21_c;
    dsp->avg_vc1_mspel_pixels_tab[ 7] = avg_vc1_mspel_mc31_c;
    dsp->avg_vc1_mspel_pixels_tab[ 8] = avg_vc1_mspel_mc02_c;
    dsp->avg_vc1_mspel_pixels_tab[ 9] = avg_vc1_mspel_mc12_c;
    dsp->avg_vc1_mspel_pixels_tab[10] = avg_vc1_mspel_mc22_c;
    dsp->avg_vc1_mspel_pixels_tab[11] = avg_vc1_mspel_mc32_c;
    dsp->avg_vc1_mspel_pixels_tab[12] = avg_vc1_mspel_mc03_c;
    dsp->avg_vc1_mspel_pixels_tab[13] = avg_vc1_mspel_mc13_c;
    dsp->avg_vc1_mspel_pixels_tab[14] = avg_vc1_mspel_mc23_c;
    dsp->avg_vc1_mspel_pixels_tab[15] = avg_vc1_mspel_mc33_c;
}