cavsdsp.c 23.3 KB
Newer Older
1 2 3 4 5 6 7
/*
 * Chinese AVS video (AVS1-P2, JiZhun profile) decoder.
 *
 * DSP functions
 *
 * Copyright (c) 2006  Stefan Gehrer <stefan.gehrer@gmx.de>
 *
8 9 10
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
11 12
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
13
 * version 2.1 of the License, or (at your option) any later version.
14
 *
15
 * FFmpeg is distributed in the hope that it will be useful,
16 17 18 19 20
 * 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
21
 * License along with FFmpeg; if not, write to the Free Software
22
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23 24 25 26
 */

#include <stdio.h>
#include "dsputil.h"
27
#include "cavsdsp.h"
28
#include "libavutil/common.h"
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

/*****************************************************************************
 *
 * in-loop deblocking filter
 *
 ****************************************************************************/

#define P2 p0_p[-3*stride]
#define P1 p0_p[-2*stride]
#define P0 p0_p[-1*stride]
#define Q0 p0_p[ 0*stride]
#define Q1 p0_p[ 1*stride]
#define Q2 p0_p[ 2*stride]

static inline void loop_filter_l2(uint8_t *p0_p,int stride,int alpha, int beta) {
    int p0 = P0;
    int q0 = Q0;

    if(abs(p0-q0)<alpha && abs(P1-p0)<beta && abs(Q1-q0)<beta) {
        int s = p0 + q0 + 2;
        alpha = (alpha>>2) + 2;
        if(abs(P2-p0) < beta && abs(p0-q0) < alpha) {
            P0 = (P1 + p0 + s) >> 2;
            P1 = (2*P1 + s) >> 2;
        } else
            P0 = (2*P1 + s) >> 2;
        if(abs(Q2-q0) < beta && abs(q0-p0) < alpha) {
            Q0 = (Q1 + q0 + s) >> 2;
            Q1 = (2*Q1 + s) >> 2;
        } else
            Q0 = (2*Q1 + s) >> 2;
    }
}

static inline void loop_filter_l1(uint8_t *p0_p, int stride, int alpha, int beta, int tc) {
    int p0 = P0;
    int q0 = Q0;

    if(abs(p0-q0)<alpha && abs(P1-p0)<beta && abs(Q1-q0)<beta) {
68 69 70
        int delta = av_clip(((q0-p0)*3+P1-Q1+4)>>3,-tc, tc);
        P0 = av_clip_uint8(p0+delta);
        Q0 = av_clip_uint8(q0-delta);
71
        if(abs(P2-p0)<beta) {
72 73
            delta = av_clip(((P0-P1)*3+P2-Q0+4)>>3, -tc, tc);
            P1 = av_clip_uint8(P1+delta);
74 75
        }
        if(abs(Q2-q0)<beta) {
76 77
            delta = av_clip(((Q1-Q0)*3+P0-Q2+4)>>3, -tc, tc);
            Q1 = av_clip_uint8(Q1-delta);
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
        }
    }
}

static inline void loop_filter_c2(uint8_t *p0_p,int stride,int alpha, int beta) {
    int p0 = P0;
    int q0 = Q0;

    if(abs(p0-q0)<alpha && abs(P1-p0)<beta && abs(Q1-q0)<beta) {
        int s = p0 + q0 + 2;
        alpha = (alpha>>2) + 2;
        if(abs(P2-p0) < beta && abs(p0-q0) < alpha) {
            P0 = (P1 + p0 + s) >> 2;
        } else
            P0 = (2*P1 + s) >> 2;
        if(abs(Q2-q0) < beta && abs(q0-p0) < alpha) {
            Q0 = (Q1 + q0 + s) >> 2;
        } else
            Q0 = (2*Q1 + s) >> 2;
    }
}

static inline void loop_filter_c1(uint8_t *p0_p,int stride,int alpha, int beta,
                                  int tc) {
    if(abs(P0-Q0)<alpha && abs(P1-P0)<beta && abs(Q1-Q0)<beta) {
103 104 105
        int delta = av_clip(((Q0-P0)*3+P1-Q1+4)>>3, -tc, tc);
        P0 = av_clip_uint8(P0+delta);
        Q0 = av_clip_uint8(Q0-delta);
106 107 108 109 110 111 112 113 114 115
    }
}

#undef P0
#undef P1
#undef P2
#undef Q0
#undef Q1
#undef Q2

116
static void cavs_filter_lv_c(uint8_t *d, int stride, int alpha, int beta, int tc,
117 118 119 120 121 122 123 124 125 126 127 128 129 130 131
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<16;i++)
            loop_filter_l2(d + i*stride,1,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<8;i++)
                loop_filter_l1(d + i*stride,1,alpha,beta,tc);
        if (bs2)
            for(i=8;i<16;i++)
                loop_filter_l1(d + i*stride,1,alpha,beta,tc);
    }
}

132
static void cavs_filter_lh_c(uint8_t *d, int stride, int alpha, int beta, int tc,
133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<16;i++)
            loop_filter_l2(d + i,stride,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<8;i++)
                loop_filter_l1(d + i,stride,alpha,beta,tc);
        if (bs2)
            for(i=8;i<16;i++)
                loop_filter_l1(d + i,stride,alpha,beta,tc);
    }
}

148
static void cavs_filter_cv_c(uint8_t *d, int stride, int alpha, int beta, int tc,
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<8;i++)
            loop_filter_c2(d + i*stride,1,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<4;i++)
                loop_filter_c1(d + i*stride,1,alpha,beta,tc);
        if (bs2)
            for(i=4;i<8;i++)
                loop_filter_c1(d + i*stride,1,alpha,beta,tc);
    }
}

164
static void cavs_filter_ch_c(uint8_t *d, int stride, int alpha, int beta, int tc,
165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<8;i++)
            loop_filter_c2(d + i,stride,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<4;i++)
                loop_filter_c1(d + i,stride,alpha,beta,tc);
        if (bs2)
            for(i=4;i<8;i++)
                loop_filter_c1(d + i,stride,alpha,beta,tc);
    }
}

/*****************************************************************************
 *
 * inverse transform
 *
 ****************************************************************************/

186
static void cavs_idct8_add_c(uint8_t *dst, DCTELEM *block, int stride) {
187 188
    int i;
    DCTELEM (*src)[8] = (DCTELEM(*)[8])block;
189
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
190

191 192
    src[0][0] += 8;

193 194 195 196 197 198 199 200 201 202 203 204 205
    for( i = 0; i < 8; i++ ) {
        const int a0 =  3*src[i][1] - (src[i][7]<<1);
        const int a1 =  3*src[i][3] + (src[i][5]<<1);
        const int a2 =  (src[i][3]<<1) - 3*src[i][5];
        const int a3 =  (src[i][1]<<1) + 3*src[i][7];

        const int b4 = ((a0 + a1 + a3)<<1) + a1;
        const int b5 = ((a0 - a1 + a2)<<1) + a0;
        const int b6 = ((a3 - a2 - a1)<<1) + a3;
        const int b7 = ((a0 - a2 - a3)<<1) - a2;

        const int a7 = (src[i][2]<<2) - 10*src[i][6];
        const int a6 = (src[i][6]<<2) + 10*src[i][2];
206 207
        const int a5 = ((src[i][0] - src[i][4]) << 3) + 4;
        const int a4 = ((src[i][0] + src[i][4]) << 3) + 4;
208 209 210 211 212 213

        const int b0 = a4 + a6;
        const int b1 = a5 + a7;
        const int b2 = a5 - a7;
        const int b3 = a4 - a6;

214 215 216 217 218 219 220 221
        src[i][0] = (b0 + b4) >> 3;
        src[i][1] = (b1 + b5) >> 3;
        src[i][2] = (b2 + b6) >> 3;
        src[i][3] = (b3 + b7) >> 3;
        src[i][4] = (b3 - b7) >> 3;
        src[i][5] = (b2 - b6) >> 3;
        src[i][6] = (b1 - b5) >> 3;
        src[i][7] = (b0 - b4) >> 3;
222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
    }
    for( i = 0; i < 8; i++ ) {
        const int a0 =  3*src[1][i] - (src[7][i]<<1);
        const int a1 =  3*src[3][i] + (src[5][i]<<1);
        const int a2 =  (src[3][i]<<1) - 3*src[5][i];
        const int a3 =  (src[1][i]<<1) + 3*src[7][i];

        const int b4 = ((a0 + a1 + a3)<<1) + a1;
        const int b5 = ((a0 - a1 + a2)<<1) + a0;
        const int b6 = ((a3 - a2 - a1)<<1) + a3;
        const int b7 = ((a0 - a2 - a3)<<1) - a2;

        const int a7 = (src[2][i]<<2) - 10*src[6][i];
        const int a6 = (src[6][i]<<2) + 10*src[2][i];
        const int a5 = (src[0][i] - src[4][i]) << 3;
        const int a4 = (src[0][i] + src[4][i]) << 3;

        const int b0 = a4 + a6;
        const int b1 = a5 + a7;
        const int b2 = a5 - a7;
        const int b3 = a4 - a6;

244 245 246 247 248 249 250 251
        dst[i + 0*stride] = cm[ dst[i + 0*stride] + ((b0 + b4) >> 7)];
        dst[i + 1*stride] = cm[ dst[i + 1*stride] + ((b1 + b5) >> 7)];
        dst[i + 2*stride] = cm[ dst[i + 2*stride] + ((b2 + b6) >> 7)];
        dst[i + 3*stride] = cm[ dst[i + 3*stride] + ((b3 + b7) >> 7)];
        dst[i + 4*stride] = cm[ dst[i + 4*stride] + ((b3 - b7) >> 7)];
        dst[i + 5*stride] = cm[ dst[i + 5*stride] + ((b2 - b6) >> 7)];
        dst[i + 6*stride] = cm[ dst[i + 6*stride] + ((b1 - b5) >> 7)];
        dst[i + 7*stride] = cm[ dst[i + 7*stride] + ((b0 - b4) >> 7)];
252 253 254 255 256 257 258 259 260 261 262 263
    }
}

/*****************************************************************************
 *
 * motion compensation
 *
 ****************************************************************************/

#define CAVS_SUBPIX(OPNAME, OP, NAME, A, B, C, D, E, F) \
static void OPNAME ## cavs_filt8_h_ ## NAME(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    const int h=8;\
264
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
    int i;\
    for(i=0; i<h; i++)\
    {\
        OP(dst[0], A*src[-2] + B*src[-1] + C*src[0] + D*src[1] + E*src[2] + F*src[3]);\
        OP(dst[1], A*src[-1] + B*src[ 0] + C*src[1] + D*src[2] + E*src[3] + F*src[4]);\
        OP(dst[2], A*src[ 0] + B*src[ 1] + C*src[2] + D*src[3] + E*src[4] + F*src[5]);\
        OP(dst[3], A*src[ 1] + B*src[ 2] + C*src[3] + D*src[4] + E*src[5] + F*src[6]);\
        OP(dst[4], A*src[ 2] + B*src[ 3] + C*src[4] + D*src[5] + E*src[6] + F*src[7]);\
        OP(dst[5], A*src[ 3] + B*src[ 4] + C*src[5] + D*src[6] + E*src[7] + F*src[8]);\
        OP(dst[6], A*src[ 4] + B*src[ 5] + C*src[6] + D*src[7] + E*src[8] + F*src[9]);\
        OP(dst[7], A*src[ 5] + B*src[ 6] + C*src[7] + D*src[8] + E*src[9] + F*src[10]);\
        dst+=dstStride;\
        src+=srcStride;\
    }\
}\
\
static void OPNAME ## cavs_filt8_v_  ## NAME(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    const int w=8;\
283
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
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
    int i;\
    for(i=0; i<w; i++)\
    {\
        const int srcB= src[-2*srcStride];\
        const int srcA= src[-1*srcStride];\
        const int src0= src[0 *srcStride];\
        const int src1= src[1 *srcStride];\
        const int src2= src[2 *srcStride];\
        const int src3= src[3 *srcStride];\
        const int src4= src[4 *srcStride];\
        const int src5= src[5 *srcStride];\
        const int src6= src[6 *srcStride];\
        const int src7= src[7 *srcStride];\
        const int src8= src[8 *srcStride];\
        const int src9= src[9 *srcStride];\
        const int src10= src[10 *srcStride];\
        OP(dst[0*dstStride], A*srcB + B*srcA + C*src0 + D*src1 + E*src2 + F*src3);\
        OP(dst[1*dstStride], A*srcA + B*src0 + C*src1 + D*src2 + E*src3 + F*src4);\
        OP(dst[2*dstStride], A*src0 + B*src1 + C*src2 + D*src3 + E*src4 + F*src5);\
        OP(dst[3*dstStride], A*src1 + B*src2 + C*src3 + D*src4 + E*src5 + F*src6);\
        OP(dst[4*dstStride], A*src2 + B*src3 + C*src4 + D*src5 + E*src6 + F*src7);\
        OP(dst[5*dstStride], A*src3 + B*src4 + C*src5 + D*src6 + E*src7 + F*src8);\
        OP(dst[6*dstStride], A*src4 + B*src5 + C*src6 + D*src7 + E*src8 + F*src9);\
        OP(dst[7*dstStride], A*src5 + B*src6 + C*src7 + D*src8 + E*src9 + F*src10);\
        dst++;\
        src++;\
    }\
}\
\
static void OPNAME ## cavs_filt16_v_ ## NAME(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    OPNAME ## cavs_filt8_v_ ## NAME(dst  , src  , dstStride, srcStride);\
    OPNAME ## cavs_filt8_v_ ## NAME(dst+8, src+8, dstStride, srcStride);\
    src += 8*srcStride;\
    dst += 8*dstStride;\
    OPNAME ## cavs_filt8_v_ ## NAME(dst  , src  , dstStride, srcStride);\
    OPNAME ## cavs_filt8_v_ ## NAME(dst+8, src+8, dstStride, srcStride);\
}\
\
static void OPNAME ## cavs_filt16_h_ ## NAME(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    OPNAME ## cavs_filt8_h_ ## NAME(dst  , src  , dstStride, srcStride);\
    OPNAME ## cavs_filt8_h_ ## NAME(dst+8, src+8, dstStride, srcStride);\
    src += 8*srcStride;\
    dst += 8*dstStride;\
    OPNAME ## cavs_filt8_h_ ## NAME(dst  , src  , dstStride, srcStride);\
    OPNAME ## cavs_filt8_h_ ## NAME(dst+8, src+8, dstStride, srcStride);\
}\

#define CAVS_SUBPIX_HV(OPNAME, OP, NAME, AH, BH, CH, DH, EH, FH, AV, BV, CV, DV, EV, FV, FULL) \
static void OPNAME ## cavs_filt8_hv_ ## NAME(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int srcStride){\
    int16_t temp[8*(8+5)];\
    int16_t *tmp = temp;\
    const int h=8;\
    const int w=8;\
337
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
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
    int i;\
    src1 -= 2*srcStride;\
    for(i=0; i<h+5; i++)\
    {\
        tmp[0]= AH*src1[-2] + BH*src1[-1] + CH*src1[0] + DH*src1[1] + EH*src1[2] + FH*src1[3];\
        tmp[1]= AH*src1[-1] + BH*src1[ 0] + CH*src1[1] + DH*src1[2] + EH*src1[3] + FH*src1[4];\
        tmp[2]= AH*src1[ 0] + BH*src1[ 1] + CH*src1[2] + DH*src1[3] + EH*src1[4] + FH*src1[5];\
        tmp[3]= AH*src1[ 1] + BH*src1[ 2] + CH*src1[3] + DH*src1[4] + EH*src1[5] + FH*src1[6];\
        tmp[4]= AH*src1[ 2] + BH*src1[ 3] + CH*src1[4] + DH*src1[5] + EH*src1[6] + FH*src1[7];\
        tmp[5]= AH*src1[ 3] + BH*src1[ 4] + CH*src1[5] + DH*src1[6] + EH*src1[7] + FH*src1[8];\
        tmp[6]= AH*src1[ 4] + BH*src1[ 5] + CH*src1[6] + DH*src1[7] + EH*src1[8] + FH*src1[9];\
        tmp[7]= AH*src1[ 5] + BH*src1[ 6] + CH*src1[7] + DH*src1[8] + EH*src1[9] + FH*src1[10];\
        tmp+=8;\
        src1+=srcStride;\
    }\
    if(FULL) {\
      tmp = temp+8*2;                           \
      for(i=0; i<w; i++)                        \
        {                                       \
          const int tmpB= tmp[-2*8];    \
          const int tmpA= tmp[-1*8];    \
          const int tmp0= tmp[0 *8];    \
          const int tmp1= tmp[1 *8];    \
          const int tmp2= tmp[2 *8];    \
          const int tmp3= tmp[3 *8];    \
          const int tmp4= tmp[4 *8];    \
          const int tmp5= tmp[5 *8];    \
          const int tmp6= tmp[6 *8];    \
          const int tmp7= tmp[7 *8];    \
          const int tmp8= tmp[8 *8];    \
          const int tmp9= tmp[9 *8];    \
          const int tmp10=tmp[10*8];                            \
          OP(dst[0*dstStride], AV*tmpB + BV*tmpA + CV*tmp0 + DV*tmp1 + EV*tmp2 + FV*tmp3 + 64*src2[0*srcStride]); \
          OP(dst[1*dstStride], AV*tmpA + BV*tmp0 + CV*tmp1 + DV*tmp2 + EV*tmp3 + FV*tmp4 + 64*src2[1*srcStride]); \
          OP(dst[2*dstStride], AV*tmp0 + BV*tmp1 + CV*tmp2 + DV*tmp3 + EV*tmp4 + FV*tmp5 + 64*src2[2*srcStride]); \
          OP(dst[3*dstStride], AV*tmp1 + BV*tmp2 + CV*tmp3 + DV*tmp4 + EV*tmp5 + FV*tmp6 + 64*src2[3*srcStride]); \
          OP(dst[4*dstStride], AV*tmp2 + BV*tmp3 + CV*tmp4 + DV*tmp5 + EV*tmp6 + FV*tmp7 + 64*src2[4*srcStride]); \
          OP(dst[5*dstStride], AV*tmp3 + BV*tmp4 + CV*tmp5 + DV*tmp6 + EV*tmp7 + FV*tmp8 + 64*src2[5*srcStride]); \
          OP(dst[6*dstStride], AV*tmp4 + BV*tmp5 + CV*tmp6 + DV*tmp7 + EV*tmp8 + FV*tmp9 + 64*src2[6*srcStride]); \
          OP(dst[7*dstStride], AV*tmp5 + BV*tmp6 + CV*tmp7 + DV*tmp8 + EV*tmp9 + FV*tmp10 + 64*src2[7*srcStride]); \
          dst++;                                                        \
          tmp++;                                                        \
          src2++;                                                       \
        }                                                               \
    } else {\
      tmp = temp+8*2;                           \
      for(i=0; i<w; i++)                        \
        {                                       \
          const int tmpB= tmp[-2*8];    \
          const int tmpA= tmp[-1*8];    \
          const int tmp0= tmp[0 *8];    \
          const int tmp1= tmp[1 *8];    \
          const int tmp2= tmp[2 *8];    \
          const int tmp3= tmp[3 *8];    \
          const int tmp4= tmp[4 *8];    \
          const int tmp5= tmp[5 *8];    \
          const int tmp6= tmp[6 *8];    \
          const int tmp7= tmp[7 *8];    \
          const int tmp8= tmp[8 *8];    \
          const int tmp9= tmp[9 *8];    \
          const int tmp10=tmp[10*8];                            \
          OP(dst[0*dstStride], AV*tmpB + BV*tmpA + CV*tmp0 + DV*tmp1 + EV*tmp2 + FV*tmp3); \
          OP(dst[1*dstStride], AV*tmpA + BV*tmp0 + CV*tmp1 + DV*tmp2 + EV*tmp3 + FV*tmp4); \
          OP(dst[2*dstStride], AV*tmp0 + BV*tmp1 + CV*tmp2 + DV*tmp3 + EV*tmp4 + FV*tmp5); \
          OP(dst[3*dstStride], AV*tmp1 + BV*tmp2 + CV*tmp3 + DV*tmp4 + EV*tmp5 + FV*tmp6); \
          OP(dst[4*dstStride], AV*tmp2 + BV*tmp3 + CV*tmp4 + DV*tmp5 + EV*tmp6 + FV*tmp7); \
          OP(dst[5*dstStride], AV*tmp3 + BV*tmp4 + CV*tmp5 + DV*tmp6 + EV*tmp7 + FV*tmp8); \
          OP(dst[6*dstStride], AV*tmp4 + BV*tmp5 + CV*tmp6 + DV*tmp7 + EV*tmp8 + FV*tmp9); \
          OP(dst[7*dstStride], AV*tmp5 + BV*tmp6 + CV*tmp7 + DV*tmp8 + EV*tmp9 + FV*tmp10); \
          dst++;                                                        \
          tmp++;                                                        \
        }                                                               \
    }\
}\
\
static void OPNAME ## cavs_filt16_hv_ ## NAME(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int srcStride){ \
    OPNAME ## cavs_filt8_hv_ ## NAME(dst  , src1,   src2  , dstStride, srcStride); \
    OPNAME ## cavs_filt8_hv_ ## NAME(dst+8, src1+8, src2+8, dstStride, srcStride); \
    src1 += 8*srcStride;\
    src2 += 8*srcStride;\
    dst += 8*dstStride;\
    OPNAME ## cavs_filt8_hv_ ## NAME(dst  , src1,   src2  , dstStride, srcStride); \
    OPNAME ## cavs_filt8_hv_ ## NAME(dst+8, src1+8, src2+8, dstStride, srcStride); \
}\

#define CAVS_MC(OPNAME, SIZE) \
424
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, int stride){\
425 426 427
    OPNAME ## cavs_filt ## SIZE ## _h_qpel_l(dst, src, stride, stride);\
}\
\
428
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc20_c(uint8_t *dst, uint8_t *src, int stride){\
429 430 431
    OPNAME ## cavs_filt ## SIZE ## _h_hpel(dst, src, stride, stride);\
}\
\
432
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc30_c(uint8_t *dst, uint8_t *src, int stride){\
433 434 435
    OPNAME ## cavs_filt ## SIZE ## _h_qpel_r(dst, src, stride, stride);\
}\
\
436
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, int stride){\
437 438 439
    OPNAME ## cavs_filt ## SIZE ## _v_qpel_l(dst, src, stride, stride);\
}\
\
440
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc02_c(uint8_t *dst, uint8_t *src, int stride){\
441 442 443
    OPNAME ## cavs_filt ## SIZE ## _v_hpel(dst, src, stride, stride);\
}\
\
444
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc03_c(uint8_t *dst, uint8_t *src, int stride){\
445 446 447
    OPNAME ## cavs_filt ## SIZE ## _v_qpel_r(dst, src, stride, stride);\
}\
\
448
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, int stride){\
449 450 451
  OPNAME ## cavs_filt ## SIZE ## _hv_jj(dst, src, NULL, stride, stride); \
}\
\
452
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, int stride){\
453 454 455
  OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src, stride, stride); \
}\
\
456
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc13_c(uint8_t *dst, uint8_t *src, int stride){\
457 458 459
  OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src+stride, stride, stride); \
}\
\
460
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc31_c(uint8_t *dst, uint8_t *src, int stride){\
461 462 463
  OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src+1, stride, stride); \
}\
\
464
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc33_c(uint8_t *dst, uint8_t *src, int stride){\
465 466 467
  OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src+stride+1,stride, stride); \
}\
\
468
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, int stride){\
469 470 471
  OPNAME ## cavs_filt ## SIZE ## _hv_ff(dst, src, src+stride+1,stride, stride); \
}\
\
472
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, int stride){\
473 474 475
  OPNAME ## cavs_filt ## SIZE ## _hv_ii(dst, src, src+stride+1,stride, stride); \
}\
\
476
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc32_c(uint8_t *dst, uint8_t *src, int stride){\
477 478 479
  OPNAME ## cavs_filt ## SIZE ## _hv_kk(dst, src, src+stride+1,stride, stride); \
}\
\
480
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc23_c(uint8_t *dst, uint8_t *src, int stride){\
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
  OPNAME ## cavs_filt ## SIZE ## _hv_qq(dst, src, src+stride+1,stride, stride); \
}\

#define op_put1(a, b)  a = cm[((b)+4)>>3]
#define op_put2(a, b)  a = cm[((b)+64)>>7]
#define op_put3(a, b)  a = cm[((b)+32)>>6]
#define op_put4(a, b)  a = cm[((b)+512)>>10]
#define op_avg1(a, b)  a = ((a)+cm[((b)+4)>>3]   +1)>>1
#define op_avg2(a, b)  a = ((a)+cm[((b)+64)>>7]  +1)>>1
#define op_avg3(a, b)  a = ((a)+cm[((b)+32)>>6]  +1)>>1
#define op_avg4(a, b)  a = ((a)+cm[((b)+512)>>10]+1)>>1
CAVS_SUBPIX(put_   , op_put1, hpel,    0, -1,  5,  5, -1,  0)
CAVS_SUBPIX(put_   , op_put2, qpel_l, -1, -2, 96, 42, -7,  0)
CAVS_SUBPIX(put_   , op_put2, qpel_r,  0, -7, 42, 96, -2, -1)
CAVS_SUBPIX_HV(put_, op_put3, jj,      0, -1,  5,  5, -1,  0,  0, -1,  5,  5, -1, 0, 0)
CAVS_SUBPIX_HV(put_, op_put4, ff,      0, -1,  5,  5, -1,  0, -1, -2, 96, 42, -7, 0, 0)
CAVS_SUBPIX_HV(put_, op_put4, ii,     -1, -2, 96, 42, -7,  0,  0, -1,  5,  5, -1, 0, 0)
CAVS_SUBPIX_HV(put_, op_put4, kk,      0, -7, 42, 96, -2, -1,  0, -1,  5,  5, -1, 0, 0)
CAVS_SUBPIX_HV(put_, op_put4, qq,      0, -1,  5,  5, -1,  0,  0, -7, 42, 96, -2,-1, 0)
CAVS_SUBPIX_HV(put_, op_put2, egpr,    0, -1,  5,  5, -1,  0,  0, -1,  5,  5, -1, 0, 1)
CAVS_SUBPIX(avg_   , op_avg1, hpel,    0, -1,  5,  5, -1,  0)
CAVS_SUBPIX(avg_   , op_avg2, qpel_l, -1, -2, 96, 42, -7,  0)
CAVS_SUBPIX(avg_   , op_avg2, qpel_r,  0, -7, 42, 96, -2, -1)
CAVS_SUBPIX_HV(avg_, op_avg3, jj,      0, -1,  5,  5, -1,  0,  0, -1,  5,  5, -1, 0, 0)
CAVS_SUBPIX_HV(avg_, op_avg4, ff,      0, -1,  5,  5, -1,  0, -1, -2, 96, 42, -7, 0, 0)
CAVS_SUBPIX_HV(avg_, op_avg4, ii,     -1, -2, 96, 42, -7,  0,  0, -1,  5,  5, -1, 0, 0)
CAVS_SUBPIX_HV(avg_, op_avg4, kk,      0, -7, 42, 96, -2, -1,  0, -1,  5,  5, -1, 0, 0)
CAVS_SUBPIX_HV(avg_, op_avg4, qq,      0, -1,  5,  5, -1,  0,  0, -7, 42, 96, -2,-1, 0)
CAVS_SUBPIX_HV(avg_, op_avg2, egpr,    0, -1,  5,  5, -1,  0,  0, -1,  5,  5, -1, 0, 1)
CAVS_MC(put_, 8)
CAVS_MC(put_, 16)
CAVS_MC(avg_, 8)
CAVS_MC(avg_, 16)
514

515 516 517 518 519
#define ff_put_cavs_qpel8_mc00_c  ff_put_pixels8x8_c
#define ff_avg_cavs_qpel8_mc00_c  ff_avg_pixels8x8_c
#define ff_put_cavs_qpel16_mc00_c ff_put_pixels16x16_c
#define ff_avg_cavs_qpel16_mc00_c ff_avg_pixels16x16_c

520
av_cold void ff_cavsdsp_init(CAVSDSPContext* c, AVCodecContext *avctx) {
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
#define dspfunc(PFX, IDX, NUM) \
    c->PFX ## _pixels_tab[IDX][ 0] = ff_ ## PFX ## NUM ## _mc00_c; \
    c->PFX ## _pixels_tab[IDX][ 1] = ff_ ## PFX ## NUM ## _mc10_c; \
    c->PFX ## _pixels_tab[IDX][ 2] = ff_ ## PFX ## NUM ## _mc20_c; \
    c->PFX ## _pixels_tab[IDX][ 3] = ff_ ## PFX ## NUM ## _mc30_c; \
    c->PFX ## _pixels_tab[IDX][ 4] = ff_ ## PFX ## NUM ## _mc01_c; \
    c->PFX ## _pixels_tab[IDX][ 5] = ff_ ## PFX ## NUM ## _mc11_c; \
    c->PFX ## _pixels_tab[IDX][ 6] = ff_ ## PFX ## NUM ## _mc21_c; \
    c->PFX ## _pixels_tab[IDX][ 7] = ff_ ## PFX ## NUM ## _mc31_c; \
    c->PFX ## _pixels_tab[IDX][ 8] = ff_ ## PFX ## NUM ## _mc02_c; \
    c->PFX ## _pixels_tab[IDX][ 9] = ff_ ## PFX ## NUM ## _mc12_c; \
    c->PFX ## _pixels_tab[IDX][10] = ff_ ## PFX ## NUM ## _mc22_c; \
    c->PFX ## _pixels_tab[IDX][11] = ff_ ## PFX ## NUM ## _mc32_c; \
    c->PFX ## _pixels_tab[IDX][12] = ff_ ## PFX ## NUM ## _mc03_c; \
    c->PFX ## _pixels_tab[IDX][13] = ff_ ## PFX ## NUM ## _mc13_c; \
    c->PFX ## _pixels_tab[IDX][14] = ff_ ## PFX ## NUM ## _mc23_c; \
    c->PFX ## _pixels_tab[IDX][15] = ff_ ## PFX ## NUM ## _mc33_c
    dspfunc(put_cavs_qpel, 0, 16);
    dspfunc(put_cavs_qpel, 1, 8);
    dspfunc(avg_cavs_qpel, 0, 16);
    dspfunc(avg_cavs_qpel, 1, 8);
    c->cavs_filter_lv = cavs_filter_lv_c;
    c->cavs_filter_lh = cavs_filter_lh_c;
    c->cavs_filter_cv = cavs_filter_cv_c;
    c->cavs_filter_ch = cavs_filter_ch_c;
    c->cavs_idct8_add = cavs_idct8_add_c;
547
    c->idct_perm = FF_NO_IDCT_PERM;
548 549

    if (HAVE_MMX) ff_cavsdsp_init_mmx(c, avctx);
550
}