dsputil_template.c 54.9 KB
Newer Older
1 2 3 4 5 6 7
/*
 * DSP utils
 * Copyright (c) 2000, 2001 Fabrice Bellard
 * Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
 *
 * gmc & q-pel & 32/64 bit based MC by Michael Niedermayer <michaelni@gmx.at>
 *
8
 * This file is part of FFmpeg.
9
 *
10
 * FFmpeg is free software; you can redistribute it and/or
11 12 13 14
 * 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.
 *
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 23 24 25 26 27 28 29
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * DSP utils
 */

30
#include "bit_depth_template.c"
31

32 33 34 35 36 37 38 39 40 41
static inline void FUNC(copy_block2)(uint8_t *dst, const uint8_t *src, int dstStride, int srcStride, int h)
{
    int i;
    for(i=0; i<h; i++)
    {
        AV_WN2P(dst   , AV_RN2P(src   ));
        dst+=dstStride;
        src+=srcStride;
    }
}
42

43 44 45 46 47 48 49 50 51 52
static inline void FUNC(copy_block4)(uint8_t *dst, const uint8_t *src, int dstStride, int srcStride, int h)
{
    int i;
    for(i=0; i<h; i++)
    {
        AV_WN4P(dst   , AV_RN4P(src   ));
        dst+=dstStride;
        src+=srcStride;
    }
}
53

54 55 56 57 58 59 60 61 62 63 64
static inline void FUNC(copy_block8)(uint8_t *dst, const uint8_t *src, int dstStride, int srcStride, int h)
{
    int i;
    for(i=0; i<h; i++)
    {
        AV_WN4P(dst                , AV_RN4P(src                ));
        AV_WN4P(dst+4*sizeof(pixel), AV_RN4P(src+4*sizeof(pixel)));
        dst+=dstStride;
        src+=srcStride;
    }
}
65

66 67 68 69 70 71 72 73 74 75 76 77 78
static inline void FUNC(copy_block16)(uint8_t *dst, const uint8_t *src, int dstStride, int srcStride, int h)
{
    int i;
    for(i=0; i<h; i++)
    {
        AV_WN4P(dst                 , AV_RN4P(src                 ));
        AV_WN4P(dst+ 4*sizeof(pixel), AV_RN4P(src+ 4*sizeof(pixel)));
        AV_WN4P(dst+ 8*sizeof(pixel), AV_RN4P(src+ 8*sizeof(pixel)));
        AV_WN4P(dst+12*sizeof(pixel), AV_RN4P(src+12*sizeof(pixel)));
        dst+=dstStride;
        src+=srcStride;
    }
}
79

80 81
/* draw the edges of width 'w' of an image of size width, height */
//FIXME check that this is ok for mpeg4 interlaced
82
static void FUNCC(draw_edges)(uint8_t *p_buf, int p_wrap, int width, int height, int w, int h, int sides)
83
{
84 85
    pixel *buf = (pixel*)p_buf;
    int wrap = p_wrap / sizeof(pixel);
86
    pixel *ptr, *last_line;
87 88 89 90 91
    int i;

    /* left and right */
    ptr = buf;
    for(i=0;i<height;i++) {
92 93 94 95 96 97 98
#if BIT_DEPTH > 8
        int j;
        for (j = 0; j < w; j++) {
            ptr[j-w] = ptr[0];
            ptr[j+width] = ptr[width-1];
        }
#else
99 100
        memset(ptr - w, ptr[0], w);
        memset(ptr + width, ptr[width-1], w);
101
#endif
102 103 104 105 106 107 108
        ptr += wrap;
    }

    /* top and bottom + corners */
    buf -= w;
    last_line = buf + (height - 1) * wrap;
    if (sides & EDGE_TOP)
109
        for(i = 0; i < h; i++)
110
            memcpy(buf - (i + 1) * wrap, buf, (width + w + w) * sizeof(pixel)); // top
111
    if (sides & EDGE_BOTTOM)
112
        for (i = 0; i < h; i++)
113
            memcpy(last_line + (i + 1) * wrap, last_line, (width + w + w) * sizeof(pixel)); // bottom
114 115 116 117 118 119 120 121 122 123 124 125 126 127
}

/**
 * Copy a rectangular area of samples to a temporary buffer and replicate the border samples.
 * @param buf destination buffer
 * @param src source buffer
 * @param linesize number of bytes between 2 vertically adjacent samples in both the source and destination buffers
 * @param block_w width of block
 * @param block_h height of block
 * @param src_x x coordinate of the top left sample of the block in the source buffer
 * @param src_y y coordinate of the top left sample of the block in the source buffer
 * @param w width of the source buffer
 * @param h height of the source buffer
 */
128
void FUNC(ff_emulated_edge_mc)(uint8_t *buf, const uint8_t *src, int linesize, int block_w, int block_h,
129 130 131 132 133 134 135 136 137 138 139 140
                                    int src_x, int src_y, int w, int h){
    int x, y;
    int start_y, start_x, end_y, end_x;

    if(src_y>= h){
        src+= (h-1-src_y)*linesize;
        src_y=h-1;
    }else if(src_y<=-block_h){
        src+= (1-block_h-src_y)*linesize;
        src_y=1-block_h;
    }
    if(src_x>= w){
141
        src+= (w-1-src_x)*sizeof(pixel);
142 143
        src_x=w-1;
    }else if(src_x<=-block_w){
144
        src+= (1-block_w-src_x)*sizeof(pixel);
145 146 147 148 149 150 151 152 153 154 155
        src_x=1-block_w;
    }

    start_y= FFMAX(0, -src_y);
    start_x= FFMAX(0, -src_x);
    end_y= FFMIN(block_h, h-src_y);
    end_x= FFMIN(block_w, w-src_x);
    assert(start_y < end_y && block_h);
    assert(start_x < end_x && block_w);

    w    = end_x - start_x;
156 157
    src += start_y*linesize + start_x*sizeof(pixel);
    buf += start_x*sizeof(pixel);
158 159 160

    //top
    for(y=0; y<start_y; y++){
161
        memcpy(buf, src, w*sizeof(pixel));
162 163 164 165 166
        buf += linesize;
    }

    // copy existing part
    for(; y<end_y; y++){
167
        memcpy(buf, src, w*sizeof(pixel));
168 169 170 171 172 173 174
        src += linesize;
        buf += linesize;
    }

    //bottom
    src -= linesize;
    for(; y<block_h; y++){
175
        memcpy(buf, src, w*sizeof(pixel));
176 177 178
        buf += linesize;
    }

179
    buf -= block_h * linesize + start_x*sizeof(pixel);
180
    while (block_h--){
181
        pixel *bufp = (pixel*)buf;
182 183
       //left
        for(x=0; x<start_x; x++){
184
            bufp[x] = bufp[start_x];
185 186 187 188
        }

       //right
        for(x=end_x; x<block_w; x++){
189
            bufp[x] = bufp[end_x - 1];
190 191 192 193 194
        }
        buf += linesize;
    }
}

195
#define DCTELEM_FUNCS(dctcoef, suffix)                                  \
196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218
static void FUNCC(get_pixels ## suffix)(DCTELEM *restrict _block,       \
                                        const uint8_t *_pixels,         \
                                        int line_size)                  \
{                                                                       \
    const pixel *pixels = (const pixel *) _pixels;                      \
    dctcoef *restrict block = (dctcoef *) _block;                       \
    int i;                                                              \
                                                                        \
    /* read the pixels */                                               \
    for(i=0;i<8;i++) {                                                  \
        block[0] = pixels[0];                                           \
        block[1] = pixels[1];                                           \
        block[2] = pixels[2];                                           \
        block[3] = pixels[3];                                           \
        block[4] = pixels[4];                                           \
        block[5] = pixels[5];                                           \
        block[6] = pixels[6];                                           \
        block[7] = pixels[7];                                           \
        pixels += line_size / sizeof(pixel);                            \
        block += 8;                                                     \
    }                                                                   \
}                                                                       \
                                                                        \
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
static void FUNCC(add_pixels8 ## suffix)(uint8_t *restrict _pixels,     \
                                         DCTELEM *_block,               \
                                         int line_size)                 \
{                                                                       \
    int i;                                                              \
    pixel *restrict pixels = (pixel *restrict)_pixels;                  \
    dctcoef *block = (dctcoef*)_block;                                  \
    line_size /= sizeof(pixel);                                         \
                                                                        \
    for(i=0;i<8;i++) {                                                  \
        pixels[0] += block[0];                                          \
        pixels[1] += block[1];                                          \
        pixels[2] += block[2];                                          \
        pixels[3] += block[3];                                          \
        pixels[4] += block[4];                                          \
        pixels[5] += block[5];                                          \
        pixels[6] += block[6];                                          \
        pixels[7] += block[7];                                          \
        pixels += line_size;                                            \
        block += 8;                                                     \
    }                                                                   \
}                                                                       \
                                                                        \
static void FUNCC(add_pixels4 ## suffix)(uint8_t *restrict _pixels,     \
                                         DCTELEM *_block,               \
                                         int line_size)                 \
{                                                                       \
    int i;                                                              \
    pixel *restrict pixels = (pixel *restrict)_pixels;                  \
    dctcoef *block = (dctcoef*)_block;                                  \
    line_size /= sizeof(pixel);                                         \
                                                                        \
    for(i=0;i<4;i++) {                                                  \
        pixels[0] += block[0];                                          \
        pixels[1] += block[1];                                          \
        pixels[2] += block[2];                                          \
        pixels[3] += block[3];                                          \
        pixels += line_size;                                            \
        block += 4;                                                     \
    }                                                                   \
}                                                                       \
                                                                        \
static void FUNCC(clear_block ## suffix)(DCTELEM *block)                \
{                                                                       \
    memset(block, 0, sizeof(dctcoef)*64);                               \
}                                                                       \
                                                                        \
/**                                                                     \
 * memset(blocks, 0, sizeof(DCTELEM)*6*64)                              \
 */                                                                     \
static void FUNCC(clear_blocks ## suffix)(DCTELEM *blocks)              \
{                                                                       \
    memset(blocks, 0, sizeof(dctcoef)*6*64);                            \
272 273
}

274 275 276 277
DCTELEM_FUNCS(DCTELEM, _16)
#if BIT_DEPTH > 8
DCTELEM_FUNCS(dctcoef, _32)
#endif
278 279

#define PIXOP2(OPNAME, OP) \
280
static void FUNCC(OPNAME ## _pixels2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
281 282
    int i;\
    for(i=0; i<h; i++){\
283
        OP(*((pixel2*)(block  )), AV_RN2P(pixels  ));\
284 285 286 287
        pixels+=line_size;\
        block +=line_size;\
    }\
}\
288
static void FUNCC(OPNAME ## _pixels4)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
289 290
    int i;\
    for(i=0; i<h; i++){\
291
        OP(*((pixel4*)(block  )), AV_RN4P(pixels  ));\
292 293 294 295
        pixels+=line_size;\
        block +=line_size;\
    }\
}\
296
static void FUNCC(OPNAME ## _pixels8)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
297 298
    int i;\
    for(i=0; i<h; i++){\
299 300
        OP(*((pixel4*)(block                )), AV_RN4P(pixels                ));\
        OP(*((pixel4*)(block+4*sizeof(pixel))), AV_RN4P(pixels+4*sizeof(pixel)));\
301 302 303 304
        pixels+=line_size;\
        block +=line_size;\
    }\
}\
305 306
static inline void FUNCC(OPNAME ## _no_rnd_pixels8)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNCC(OPNAME ## _pixels8)(block, pixels, line_size, h);\
307 308
}\
\
309
static inline void FUNC(OPNAME ## _no_rnd_pixels8_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
310 311 312
                                                int src_stride1, int src_stride2, int h){\
    int i;\
    for(i=0; i<h; i++){\
313 314 315 316 317 318 319
        pixel4 a,b;\
        a= AV_RN4P(&src1[i*src_stride1  ]);\
        b= AV_RN4P(&src2[i*src_stride2  ]);\
        OP(*((pixel4*)&dst[i*dst_stride  ]), no_rnd_avg_pixel4(a, b));\
        a= AV_RN4P(&src1[i*src_stride1+4*sizeof(pixel)]);\
        b= AV_RN4P(&src2[i*src_stride2+4*sizeof(pixel)]);\
        OP(*((pixel4*)&dst[i*dst_stride+4*sizeof(pixel)]), no_rnd_avg_pixel4(a, b));\
320 321 322
    }\
}\
\
323
static inline void FUNC(OPNAME ## _pixels8_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
324 325 326
                                                int src_stride1, int src_stride2, int h){\
    int i;\
    for(i=0; i<h; i++){\
327 328 329 330 331 332 333
        pixel4 a,b;\
        a= AV_RN4P(&src1[i*src_stride1  ]);\
        b= AV_RN4P(&src2[i*src_stride2  ]);\
        OP(*((pixel4*)&dst[i*dst_stride  ]), rnd_avg_pixel4(a, b));\
        a= AV_RN4P(&src1[i*src_stride1+4*sizeof(pixel)]);\
        b= AV_RN4P(&src2[i*src_stride2+4*sizeof(pixel)]);\
        OP(*((pixel4*)&dst[i*dst_stride+4*sizeof(pixel)]), rnd_avg_pixel4(a, b));\
334 335 336
    }\
}\
\
337
static inline void FUNC(OPNAME ## _pixels4_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
338 339 340
                                                int src_stride1, int src_stride2, int h){\
    int i;\
    for(i=0; i<h; i++){\
341 342 343 344
        pixel4 a,b;\
        a= AV_RN4P(&src1[i*src_stride1  ]);\
        b= AV_RN4P(&src2[i*src_stride2  ]);\
        OP(*((pixel4*)&dst[i*dst_stride  ]), rnd_avg_pixel4(a, b));\
345 346 347
    }\
}\
\
348
static inline void FUNC(OPNAME ## _pixels2_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
349 350 351
                                                int src_stride1, int src_stride2, int h){\
    int i;\
    for(i=0; i<h; i++){\
352 353 354 355
        pixel4 a,b;\
        a= AV_RN2P(&src1[i*src_stride1  ]);\
        b= AV_RN2P(&src2[i*src_stride2  ]);\
        OP(*((pixel2*)&dst[i*dst_stride  ]), rnd_avg_pixel4(a, b));\
356 357 358
    }\
}\
\
359
static inline void FUNC(OPNAME ## _pixels16_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
360
                                                int src_stride1, int src_stride2, int h){\
361 362
    FUNC(OPNAME ## _pixels8_l2)(dst  , src1  , src2  , dst_stride, src_stride1, src_stride2, h);\
    FUNC(OPNAME ## _pixels8_l2)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, h);\
363 364
}\
\
365
static inline void FUNC(OPNAME ## _no_rnd_pixels16_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
366
                                                int src_stride1, int src_stride2, int h){\
367 368
    FUNC(OPNAME ## _no_rnd_pixels8_l2)(dst  , src1  , src2  , dst_stride, src_stride1, src_stride2, h);\
    FUNC(OPNAME ## _no_rnd_pixels8_l2)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, h);\
369 370
}\
\
371 372
static inline void FUNCC(OPNAME ## _no_rnd_pixels8_x2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _no_rnd_pixels8_l2)(block, pixels, pixels+sizeof(pixel), line_size, line_size, line_size, h);\
373 374
}\
\
375 376
static inline void FUNCC(OPNAME ## _pixels8_x2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _pixels8_l2)(block, pixels, pixels+sizeof(pixel), line_size, line_size, line_size, h);\
377 378
}\
\
379 380
static inline void FUNCC(OPNAME ## _no_rnd_pixels8_y2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _no_rnd_pixels8_l2)(block, pixels, pixels+line_size, line_size, line_size, line_size, h);\
381 382
}\
\
383 384
static inline void FUNCC(OPNAME ## _pixels8_y2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _pixels8_l2)(block, pixels, pixels+line_size, line_size, line_size, line_size, h);\
385 386
}\
\
387
static inline void FUNC(OPNAME ## _pixels8_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\
388
                 int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
389
    /* FIXME HIGH BIT DEPTH */\
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;\
    for(i=0; i<h; i++){\
        uint32_t a, b, c, d, l0, l1, h0, h1;\
        a= AV_RN32(&src1[i*src_stride1]);\
        b= AV_RN32(&src2[i*src_stride2]);\
        c= AV_RN32(&src3[i*src_stride3]);\
        d= AV_RN32(&src4[i*src_stride4]);\
        l0=  (a&0x03030303UL)\
           + (b&0x03030303UL)\
           + 0x02020202UL;\
        h0= ((a&0xFCFCFCFCUL)>>2)\
          + ((b&0xFCFCFCFCUL)>>2);\
        l1=  (c&0x03030303UL)\
           + (d&0x03030303UL);\
        h1= ((c&0xFCFCFCFCUL)>>2)\
          + ((d&0xFCFCFCFCUL)>>2);\
        OP(*((uint32_t*)&dst[i*dst_stride]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
        a= AV_RN32(&src1[i*src_stride1+4]);\
        b= AV_RN32(&src2[i*src_stride2+4]);\
        c= AV_RN32(&src3[i*src_stride3+4]);\
        d= AV_RN32(&src4[i*src_stride4+4]);\
        l0=  (a&0x03030303UL)\
           + (b&0x03030303UL)\
           + 0x02020202UL;\
        h0= ((a&0xFCFCFCFCUL)>>2)\
          + ((b&0xFCFCFCFCUL)>>2);\
        l1=  (c&0x03030303UL)\
           + (d&0x03030303UL);\
        h1= ((c&0xFCFCFCFCUL)>>2)\
          + ((d&0xFCFCFCFCUL)>>2);\
        OP(*((uint32_t*)&dst[i*dst_stride+4]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
    }\
}\
\
424 425
static inline void FUNCC(OPNAME ## _pixels4_x2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _pixels4_l2)(block, pixels, pixels+sizeof(pixel), line_size, line_size, line_size, h);\
426 427
}\
\
428 429
static inline void FUNCC(OPNAME ## _pixels4_y2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _pixels4_l2)(block, pixels, pixels+line_size, line_size, line_size, line_size, h);\
430 431
}\
\
432 433
static inline void FUNCC(OPNAME ## _pixels2_x2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _pixels2_l2)(block, pixels, pixels+sizeof(pixel), line_size, line_size, line_size, h);\
434 435
}\
\
436 437
static inline void FUNCC(OPNAME ## _pixels2_y2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
    FUNC(OPNAME ## _pixels2_l2)(block, pixels, pixels+line_size, line_size, line_size, line_size, h);\
438 439
}\
\
440
static inline void FUNC(OPNAME ## _no_rnd_pixels8_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\
441
                 int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
442
    /* FIXME HIGH BIT DEPTH*/\
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
    int i;\
    for(i=0; i<h; i++){\
        uint32_t a, b, c, d, l0, l1, h0, h1;\
        a= AV_RN32(&src1[i*src_stride1]);\
        b= AV_RN32(&src2[i*src_stride2]);\
        c= AV_RN32(&src3[i*src_stride3]);\
        d= AV_RN32(&src4[i*src_stride4]);\
        l0=  (a&0x03030303UL)\
           + (b&0x03030303UL)\
           + 0x01010101UL;\
        h0= ((a&0xFCFCFCFCUL)>>2)\
          + ((b&0xFCFCFCFCUL)>>2);\
        l1=  (c&0x03030303UL)\
           + (d&0x03030303UL);\
        h1= ((c&0xFCFCFCFCUL)>>2)\
          + ((d&0xFCFCFCFCUL)>>2);\
        OP(*((uint32_t*)&dst[i*dst_stride]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
        a= AV_RN32(&src1[i*src_stride1+4]);\
        b= AV_RN32(&src2[i*src_stride2+4]);\
        c= AV_RN32(&src3[i*src_stride3+4]);\
        d= AV_RN32(&src4[i*src_stride4+4]);\
        l0=  (a&0x03030303UL)\
           + (b&0x03030303UL)\
           + 0x01010101UL;\
        h0= ((a&0xFCFCFCFCUL)>>2)\
          + ((b&0xFCFCFCFCUL)>>2);\
        l1=  (c&0x03030303UL)\
           + (d&0x03030303UL);\
        h1= ((c&0xFCFCFCFCUL)>>2)\
          + ((d&0xFCFCFCFCUL)>>2);\
        OP(*((uint32_t*)&dst[i*dst_stride+4]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
    }\
}\
476
static inline void FUNC(OPNAME ## _pixels16_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\
477
                 int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
478 479
    FUNC(OPNAME ## _pixels8_l4)(dst  , src1  , src2  , src3  , src4  , dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
    FUNC(OPNAME ## _pixels8_l4)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), src3+8*sizeof(pixel), src4+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
480
}\
481
static inline void FUNC(OPNAME ## _no_rnd_pixels16_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\
482
                 int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
483 484
    FUNC(OPNAME ## _no_rnd_pixels8_l4)(dst  , src1  , src2  , src3  , src4  , dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
    FUNC(OPNAME ## _no_rnd_pixels8_l4)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), src3+8*sizeof(pixel), src4+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
485 486
}\
\
487
static inline void FUNCC(OPNAME ## _pixels2_xy2)(uint8_t *p_block, const uint8_t *p_pixels, int line_size, int h)\
488 489
{\
        int i, a0, b0, a1, b1;\
490 491 492
        pixel *block = (pixel*)p_block;\
        const pixel *pixels = (const pixel*)p_pixels;\
        line_size >>= sizeof(pixel)-1;\
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
        a0= pixels[0];\
        b0= pixels[1] + 2;\
        a0 += b0;\
        b0 += pixels[2];\
\
        pixels+=line_size;\
        for(i=0; i<h; i+=2){\
            a1= pixels[0];\
            b1= pixels[1];\
            a1 += b1;\
            b1 += pixels[2];\
\
            block[0]= (a1+a0)>>2; /* FIXME non put */\
            block[1]= (b1+b0)>>2;\
\
            pixels+=line_size;\
            block +=line_size;\
\
            a0= pixels[0];\
            b0= pixels[1] + 2;\
            a0 += b0;\
            b0 += pixels[2];\
\
            block[0]= (a1+a0)>>2;\
            block[1]= (b1+b0)>>2;\
            pixels+=line_size;\
            block +=line_size;\
        }\
}\
\
523
static inline void FUNCC(OPNAME ## _pixels4_xy2)(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
524
{\
525
        /* FIXME HIGH BIT DEPTH */\
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
        int i;\
        const uint32_t a= AV_RN32(pixels  );\
        const uint32_t b= AV_RN32(pixels+1);\
        uint32_t l0=  (a&0x03030303UL)\
                    + (b&0x03030303UL)\
                    + 0x02020202UL;\
        uint32_t h0= ((a&0xFCFCFCFCUL)>>2)\
                   + ((b&0xFCFCFCFCUL)>>2);\
        uint32_t l1,h1;\
\
        pixels+=line_size;\
        for(i=0; i<h; i+=2){\
            uint32_t a= AV_RN32(pixels  );\
            uint32_t b= AV_RN32(pixels+1);\
            l1=  (a&0x03030303UL)\
               + (b&0x03030303UL);\
            h1= ((a&0xFCFCFCFCUL)>>2)\
              + ((b&0xFCFCFCFCUL)>>2);\
            OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
            pixels+=line_size;\
            block +=line_size;\
            a= AV_RN32(pixels  );\
            b= AV_RN32(pixels+1);\
            l0=  (a&0x03030303UL)\
               + (b&0x03030303UL)\
               + 0x02020202UL;\
            h0= ((a&0xFCFCFCFCUL)>>2)\
              + ((b&0xFCFCFCFCUL)>>2);\
            OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
            pixels+=line_size;\
            block +=line_size;\
        }\
}\
\
560
static inline void FUNCC(OPNAME ## _pixels8_xy2)(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
561
{\
562
    /* FIXME HIGH BIT DEPTH */\
563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
    int j;\
    for(j=0; j<2; j++){\
        int i;\
        const uint32_t a= AV_RN32(pixels  );\
        const uint32_t b= AV_RN32(pixels+1);\
        uint32_t l0=  (a&0x03030303UL)\
                    + (b&0x03030303UL)\
                    + 0x02020202UL;\
        uint32_t h0= ((a&0xFCFCFCFCUL)>>2)\
                   + ((b&0xFCFCFCFCUL)>>2);\
        uint32_t l1,h1;\
\
        pixels+=line_size;\
        for(i=0; i<h; i+=2){\
            uint32_t a= AV_RN32(pixels  );\
            uint32_t b= AV_RN32(pixels+1);\
            l1=  (a&0x03030303UL)\
               + (b&0x03030303UL);\
            h1= ((a&0xFCFCFCFCUL)>>2)\
              + ((b&0xFCFCFCFCUL)>>2);\
            OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
            pixels+=line_size;\
            block +=line_size;\
            a= AV_RN32(pixels  );\
            b= AV_RN32(pixels+1);\
            l0=  (a&0x03030303UL)\
               + (b&0x03030303UL)\
               + 0x02020202UL;\
            h0= ((a&0xFCFCFCFCUL)>>2)\
              + ((b&0xFCFCFCFCUL)>>2);\
            OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
            pixels+=line_size;\
            block +=line_size;\
        }\
        pixels+=4-line_size*(h+1);\
        block +=4-line_size*h;\
    }\
}\
\
602
static inline void FUNCC(OPNAME ## _no_rnd_pixels8_xy2)(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
603
{\
604
    /* FIXME HIGH BIT DEPTH */\
605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643
    int j;\
    for(j=0; j<2; j++){\
        int i;\
        const uint32_t a= AV_RN32(pixels  );\
        const uint32_t b= AV_RN32(pixels+1);\
        uint32_t l0=  (a&0x03030303UL)\
                    + (b&0x03030303UL)\
                    + 0x01010101UL;\
        uint32_t h0= ((a&0xFCFCFCFCUL)>>2)\
                   + ((b&0xFCFCFCFCUL)>>2);\
        uint32_t l1,h1;\
\
        pixels+=line_size;\
        for(i=0; i<h; i+=2){\
            uint32_t a= AV_RN32(pixels  );\
            uint32_t b= AV_RN32(pixels+1);\
            l1=  (a&0x03030303UL)\
               + (b&0x03030303UL);\
            h1= ((a&0xFCFCFCFCUL)>>2)\
              + ((b&0xFCFCFCFCUL)>>2);\
            OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
            pixels+=line_size;\
            block +=line_size;\
            a= AV_RN32(pixels  );\
            b= AV_RN32(pixels+1);\
            l0=  (a&0x03030303UL)\
               + (b&0x03030303UL)\
               + 0x01010101UL;\
            h0= ((a&0xFCFCFCFCUL)>>2)\
              + ((b&0xFCFCFCFCUL)>>2);\
            OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
            pixels+=line_size;\
            block +=line_size;\
        }\
        pixels+=4-line_size*(h+1);\
        block +=4-line_size*h;\
    }\
}\
\
644 645 646 647 648 649 650 651
CALL_2X_PIXELS(FUNCC(OPNAME ## _pixels16)    , FUNCC(OPNAME ## _pixels8)    , 8*sizeof(pixel))\
CALL_2X_PIXELS(FUNCC(OPNAME ## _pixels16_x2) , FUNCC(OPNAME ## _pixels8_x2) , 8*sizeof(pixel))\
CALL_2X_PIXELS(FUNCC(OPNAME ## _pixels16_y2) , FUNCC(OPNAME ## _pixels8_y2) , 8*sizeof(pixel))\
CALL_2X_PIXELS(FUNCC(OPNAME ## _pixels16_xy2), FUNCC(OPNAME ## _pixels8_xy2), 8*sizeof(pixel))\
av_unused CALL_2X_PIXELS(FUNCC(OPNAME ## _no_rnd_pixels16)    , FUNCC(OPNAME ## _pixels8) , 8*sizeof(pixel))\
CALL_2X_PIXELS(FUNCC(OPNAME ## _no_rnd_pixels16_x2) , FUNCC(OPNAME ## _no_rnd_pixels8_x2) , 8*sizeof(pixel))\
CALL_2X_PIXELS(FUNCC(OPNAME ## _no_rnd_pixels16_y2) , FUNCC(OPNAME ## _no_rnd_pixels8_y2) , 8*sizeof(pixel))\
CALL_2X_PIXELS(FUNCC(OPNAME ## _no_rnd_pixels16_xy2), FUNCC(OPNAME ## _no_rnd_pixels8_xy2), 8*sizeof(pixel))\
652

653
#define op_avg(a, b) a = rnd_avg_pixel4(a, b)
654 655 656 657 658 659 660 661 662 663
#define op_put(a, b) a = b

PIXOP2(avg, op_avg)
PIXOP2(put, op_put)
#undef op_avg
#undef op_put

#define put_no_rnd_pixels8_c  put_pixels8_c
#define put_no_rnd_pixels16_c put_pixels16_c

664 665
static void FUNCC(put_no_rnd_pixels16_l2)(uint8_t *dst, const uint8_t *a, const uint8_t *b, int stride, int h){
    FUNC(put_no_rnd_pixels16_l2)(dst, a, b, stride, stride, stride, h);
666 667
}

668 669
static void FUNCC(put_no_rnd_pixels8_l2)(uint8_t *dst, const uint8_t *a, const uint8_t *b, int stride, int h){
    FUNC(put_no_rnd_pixels8_l2)(dst, a, b, stride, stride, stride, h);
670 671 672
}

#define H264_CHROMA_MC(OPNAME, OP)\
673 674 675
static void FUNCC(OPNAME ## h264_chroma_mc2)(uint8_t *p_dst/*align 8*/, uint8_t *p_src/*align 1*/, int stride, int h, int x, int y){\
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
676 677 678 679 680
    const int A=(8-x)*(8-y);\
    const int B=(  x)*(8-y);\
    const int C=(8-x)*(  y);\
    const int D=(  x)*(  y);\
    int i;\
681
    stride >>= sizeof(pixel)-1;\
682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703
    \
    assert(x<8 && y<8 && x>=0 && y>=0);\
\
    if(D){\
        for(i=0; i<h; i++){\
            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
            OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
            dst+= stride;\
            src+= stride;\
        }\
    }else{\
        const int E= B+C;\
        const int step= C ? stride : 1;\
        for(i=0; i<h; i++){\
            OP(dst[0], (A*src[0] + E*src[step+0]));\
            OP(dst[1], (A*src[1] + E*src[step+1]));\
            dst+= stride;\
            src+= stride;\
        }\
    }\
}\
\
704 705 706
static void FUNCC(OPNAME ## h264_chroma_mc4)(uint8_t *p_dst/*align 8*/, uint8_t *p_src/*align 1*/, int stride, int h, int x, int y){\
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
707 708 709 710 711
    const int A=(8-x)*(8-y);\
    const int B=(  x)*(8-y);\
    const int C=(8-x)*(  y);\
    const int D=(  x)*(  y);\
    int i;\
712
    stride >>= sizeof(pixel)-1;\
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
    \
    assert(x<8 && y<8 && x>=0 && y>=0);\
\
    if(D){\
        for(i=0; i<h; i++){\
            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
            OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
            OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\
            OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\
            dst+= stride;\
            src+= stride;\
        }\
    }else{\
        const int E= B+C;\
        const int step= C ? stride : 1;\
        for(i=0; i<h; i++){\
            OP(dst[0], (A*src[0] + E*src[step+0]));\
            OP(dst[1], (A*src[1] + E*src[step+1]));\
            OP(dst[2], (A*src[2] + E*src[step+2]));\
            OP(dst[3], (A*src[3] + E*src[step+3]));\
            dst+= stride;\
            src+= stride;\
        }\
    }\
}\
\
739 740 741
static void FUNCC(OPNAME ## h264_chroma_mc8)(uint8_t *p_dst/*align 8*/, uint8_t *p_src/*align 1*/, int stride, int h, int x, int y){\
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
742 743 744 745 746
    const int A=(8-x)*(8-y);\
    const int B=(  x)*(8-y);\
    const int C=(8-x)*(  y);\
    const int D=(  x)*(  y);\
    int i;\
747
    stride >>= sizeof(pixel)-1;\
748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790
    \
    assert(x<8 && y<8 && x>=0 && y>=0);\
\
    if(D){\
        for(i=0; i<h; i++){\
            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
            OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
            OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\
            OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\
            OP(dst[4], (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5]));\
            OP(dst[5], (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6]));\
            OP(dst[6], (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7]));\
            OP(dst[7], (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8]));\
            dst+= stride;\
            src+= stride;\
        }\
    }else{\
        const int E= B+C;\
        const int step= C ? stride : 1;\
        for(i=0; i<h; i++){\
            OP(dst[0], (A*src[0] + E*src[step+0]));\
            OP(dst[1], (A*src[1] + E*src[step+1]));\
            OP(dst[2], (A*src[2] + E*src[step+2]));\
            OP(dst[3], (A*src[3] + E*src[step+3]));\
            OP(dst[4], (A*src[4] + E*src[step+4]));\
            OP(dst[5], (A*src[5] + E*src[step+5]));\
            OP(dst[6], (A*src[6] + E*src[step+6]));\
            OP(dst[7], (A*src[7] + E*src[step+7]));\
            dst+= stride;\
            src+= stride;\
        }\
    }\
}

#define op_avg(a, b) a = (((a)+(((b) + 32)>>6)+1)>>1)
#define op_put(a, b) a = (((b) + 32)>>6)

H264_CHROMA_MC(put_       , op_put)
H264_CHROMA_MC(avg_       , op_avg)
#undef op_avg
#undef op_put

#define H264_LOWPASS(OPNAME, OP, OP2) \
791
static av_unused void FUNC(OPNAME ## h264_qpel2_h_lowpass)(uint8_t *p_dst, uint8_t *p_src, int dstStride, int srcStride){\
792
    const int h=2;\
793
    INIT_CLIP\
794
    int i;\
795 796 797 798
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
799 800 801 802 803 804 805 806 807
    for(i=0; i<h; i++)\
    {\
        OP(dst[0], (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3]));\
        OP(dst[1], (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4]));\
        dst+=dstStride;\
        src+=srcStride;\
    }\
}\
\
808
static av_unused void FUNC(OPNAME ## h264_qpel2_v_lowpass)(uint8_t *p_dst, uint8_t *p_src, int dstStride, int srcStride){\
809
    const int w=2;\
810
    INIT_CLIP\
811
    int i;\
812 813 814 815
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831
    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];\
        OP(dst[0*dstStride], (src0+src1)*20 - (srcA+src2)*5 + (srcB+src3));\
        OP(dst[1*dstStride], (src1+src2)*20 - (src0+src3)*5 + (srcA+src4));\
        dst++;\
        src++;\
    }\
}\
\
832
static av_unused void FUNC(OPNAME ## h264_qpel2_hv_lowpass)(uint8_t *p_dst, int16_t *tmp, uint8_t *p_src, int dstStride, int tmpStride, int srcStride){\
833 834
    const int h=2;\
    const int w=2;\
835 836
    const int pad = (BIT_DEPTH > 9) ? (-10 * ((1<<BIT_DEPTH)-1)) : 0;\
    INIT_CLIP\
837
    int i;\
838 839 840 841
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
842 843 844
    src -= 2*srcStride;\
    for(i=0; i<h+5; i++)\
    {\
845 846
        tmp[0]= (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3]) + pad;\
        tmp[1]= (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4]) + pad;\
847 848 849 850 851 852
        tmp+=tmpStride;\
        src+=srcStride;\
    }\
    tmp -= tmpStride*(h+5-2);\
    for(i=0; i<w; i++)\
    {\
853 854 855 856 857 858 859
        const int tmpB= tmp[-2*tmpStride] - pad;\
        const int tmpA= tmp[-1*tmpStride] - pad;\
        const int tmp0= tmp[0 *tmpStride] - pad;\
        const int tmp1= tmp[1 *tmpStride] - pad;\
        const int tmp2= tmp[2 *tmpStride] - pad;\
        const int tmp3= tmp[3 *tmpStride] - pad;\
        const int tmp4= tmp[4 *tmpStride] - pad;\
860 861 862 863 864 865
        OP2(dst[0*dstStride], (tmp0+tmp1)*20 - (tmpA+tmp2)*5 + (tmpB+tmp3));\
        OP2(dst[1*dstStride], (tmp1+tmp2)*20 - (tmp0+tmp3)*5 + (tmpA+tmp4));\
        dst++;\
        tmp++;\
    }\
}\
866
static void FUNC(OPNAME ## h264_qpel4_h_lowpass)(uint8_t *p_dst, uint8_t *p_src, int dstStride, int srcStride){\
867
    const int h=4;\
868
    INIT_CLIP\
869
    int i;\
870 871 872 873
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
874 875 876 877 878 879 880 881 882 883 884
    for(i=0; i<h; i++)\
    {\
        OP(dst[0], (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3]));\
        OP(dst[1], (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4]));\
        OP(dst[2], (src[2]+src[3])*20 - (src[1 ]+src[4])*5 + (src[0 ]+src[5]));\
        OP(dst[3], (src[3]+src[4])*20 - (src[2 ]+src[5])*5 + (src[1 ]+src[6]));\
        dst+=dstStride;\
        src+=srcStride;\
    }\
}\
\
885
static void FUNC(OPNAME ## h264_qpel4_v_lowpass)(uint8_t *p_dst, uint8_t *p_src, int dstStride, int srcStride){\
886
    const int w=4;\
887
    INIT_CLIP\
888
    int i;\
889 890 891 892
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
    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];\
        OP(dst[0*dstStride], (src0+src1)*20 - (srcA+src2)*5 + (srcB+src3));\
        OP(dst[1*dstStride], (src1+src2)*20 - (src0+src3)*5 + (srcA+src4));\
        OP(dst[2*dstStride], (src2+src3)*20 - (src1+src4)*5 + (src0+src5));\
        OP(dst[3*dstStride], (src3+src4)*20 - (src2+src5)*5 + (src1+src6));\
        dst++;\
        src++;\
    }\
}\
\
913
static void FUNC(OPNAME ## h264_qpel4_hv_lowpass)(uint8_t *p_dst, int16_t *tmp, uint8_t *p_src, int dstStride, int tmpStride, int srcStride){\
914 915
    const int h=4;\
    const int w=4;\
916 917
    const int pad = (BIT_DEPTH > 9) ? (-10 * ((1<<BIT_DEPTH)-1)) : 0;\
    INIT_CLIP\
918
    int i;\
919 920 921 922
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
923 924 925
    src -= 2*srcStride;\
    for(i=0; i<h+5; i++)\
    {\
926 927 928 929
        tmp[0]= (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3]) + pad;\
        tmp[1]= (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4]) + pad;\
        tmp[2]= (src[2]+src[3])*20 - (src[1 ]+src[4])*5 + (src[0 ]+src[5]) + pad;\
        tmp[3]= (src[3]+src[4])*20 - (src[2 ]+src[5])*5 + (src[1 ]+src[6]) + pad;\
930 931 932 933 934 935
        tmp+=tmpStride;\
        src+=srcStride;\
    }\
    tmp -= tmpStride*(h+5-2);\
    for(i=0; i<w; i++)\
    {\
936 937 938 939 940 941 942 943 944
        const int tmpB= tmp[-2*tmpStride] - pad;\
        const int tmpA= tmp[-1*tmpStride] - pad;\
        const int tmp0= tmp[0 *tmpStride] - pad;\
        const int tmp1= tmp[1 *tmpStride] - pad;\
        const int tmp2= tmp[2 *tmpStride] - pad;\
        const int tmp3= tmp[3 *tmpStride] - pad;\
        const int tmp4= tmp[4 *tmpStride] - pad;\
        const int tmp5= tmp[5 *tmpStride] - pad;\
        const int tmp6= tmp[6 *tmpStride] - pad;\
945 946 947 948 949 950 951 952 953
        OP2(dst[0*dstStride], (tmp0+tmp1)*20 - (tmpA+tmp2)*5 + (tmpB+tmp3));\
        OP2(dst[1*dstStride], (tmp1+tmp2)*20 - (tmp0+tmp3)*5 + (tmpA+tmp4));\
        OP2(dst[2*dstStride], (tmp2+tmp3)*20 - (tmp1+tmp4)*5 + (tmp0+tmp5));\
        OP2(dst[3*dstStride], (tmp3+tmp4)*20 - (tmp2+tmp5)*5 + (tmp1+tmp6));\
        dst++;\
        tmp++;\
    }\
}\
\
954
static void FUNC(OPNAME ## h264_qpel8_h_lowpass)(uint8_t *p_dst, uint8_t *p_src, int dstStride, int srcStride){\
955
    const int h=8;\
956
    INIT_CLIP\
957
    int i;\
958 959 960 961
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
962 963 964 965 966 967 968 969 970 971 972 973 974 975 976
    for(i=0; i<h; i++)\
    {\
        OP(dst[0], (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3 ]));\
        OP(dst[1], (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4 ]));\
        OP(dst[2], (src[2]+src[3])*20 - (src[1 ]+src[4])*5 + (src[0 ]+src[5 ]));\
        OP(dst[3], (src[3]+src[4])*20 - (src[2 ]+src[5])*5 + (src[1 ]+src[6 ]));\
        OP(dst[4], (src[4]+src[5])*20 - (src[3 ]+src[6])*5 + (src[2 ]+src[7 ]));\
        OP(dst[5], (src[5]+src[6])*20 - (src[4 ]+src[7])*5 + (src[3 ]+src[8 ]));\
        OP(dst[6], (src[6]+src[7])*20 - (src[5 ]+src[8])*5 + (src[4 ]+src[9 ]));\
        OP(dst[7], (src[7]+src[8])*20 - (src[6 ]+src[9])*5 + (src[5 ]+src[10]));\
        dst+=dstStride;\
        src+=srcStride;\
    }\
}\
\
977
static void FUNC(OPNAME ## h264_qpel8_v_lowpass)(uint8_t *p_dst, uint8_t *p_src, int dstStride, int srcStride){\
978
    const int w=8;\
979
    INIT_CLIP\
980
    int i;\
981 982 983 984
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012
    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], (src0+src1)*20 - (srcA+src2)*5 + (srcB+src3));\
        OP(dst[1*dstStride], (src1+src2)*20 - (src0+src3)*5 + (srcA+src4));\
        OP(dst[2*dstStride], (src2+src3)*20 - (src1+src4)*5 + (src0+src5));\
        OP(dst[3*dstStride], (src3+src4)*20 - (src2+src5)*5 + (src1+src6));\
        OP(dst[4*dstStride], (src4+src5)*20 - (src3+src6)*5 + (src2+src7));\
        OP(dst[5*dstStride], (src5+src6)*20 - (src4+src7)*5 + (src3+src8));\
        OP(dst[6*dstStride], (src6+src7)*20 - (src5+src8)*5 + (src4+src9));\
        OP(dst[7*dstStride], (src7+src8)*20 - (src6+src9)*5 + (src5+src10));\
        dst++;\
        src++;\
    }\
}\
\
1013
static void FUNC(OPNAME ## h264_qpel8_hv_lowpass)(uint8_t *p_dst, int16_t *tmp, uint8_t *p_src, int dstStride, int tmpStride, int srcStride){\
1014 1015
    const int h=8;\
    const int w=8;\
1016 1017
    const int pad = (BIT_DEPTH > 9) ? (-10 * ((1<<BIT_DEPTH)-1)) : 0;\
    INIT_CLIP\
1018
    int i;\
1019 1020 1021 1022
    pixel *dst = (pixel*)p_dst;\
    pixel *src = (pixel*)p_src;\
    dstStride >>= sizeof(pixel)-1;\
    srcStride >>= sizeof(pixel)-1;\
1023 1024 1025
    src -= 2*srcStride;\
    for(i=0; i<h+5; i++)\
    {\
1026 1027 1028 1029 1030 1031 1032 1033
        tmp[0]= (src[0]+src[1])*20 - (src[-1]+src[2])*5 + (src[-2]+src[3 ]) + pad;\
        tmp[1]= (src[1]+src[2])*20 - (src[0 ]+src[3])*5 + (src[-1]+src[4 ]) + pad;\
        tmp[2]= (src[2]+src[3])*20 - (src[1 ]+src[4])*5 + (src[0 ]+src[5 ]) + pad;\
        tmp[3]= (src[3]+src[4])*20 - (src[2 ]+src[5])*5 + (src[1 ]+src[6 ]) + pad;\
        tmp[4]= (src[4]+src[5])*20 - (src[3 ]+src[6])*5 + (src[2 ]+src[7 ]) + pad;\
        tmp[5]= (src[5]+src[6])*20 - (src[4 ]+src[7])*5 + (src[3 ]+src[8 ]) + pad;\
        tmp[6]= (src[6]+src[7])*20 - (src[5 ]+src[8])*5 + (src[4 ]+src[9 ]) + pad;\
        tmp[7]= (src[7]+src[8])*20 - (src[6 ]+src[9])*5 + (src[5 ]+src[10]) + pad;\
1034 1035 1036 1037 1038 1039
        tmp+=tmpStride;\
        src+=srcStride;\
    }\
    tmp -= tmpStride*(h+5-2);\
    for(i=0; i<w; i++)\
    {\
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052
        const int tmpB= tmp[-2*tmpStride] - pad;\
        const int tmpA= tmp[-1*tmpStride] - pad;\
        const int tmp0= tmp[0 *tmpStride] - pad;\
        const int tmp1= tmp[1 *tmpStride] - pad;\
        const int tmp2= tmp[2 *tmpStride] - pad;\
        const int tmp3= tmp[3 *tmpStride] - pad;\
        const int tmp4= tmp[4 *tmpStride] - pad;\
        const int tmp5= tmp[5 *tmpStride] - pad;\
        const int tmp6= tmp[6 *tmpStride] - pad;\
        const int tmp7= tmp[7 *tmpStride] - pad;\
        const int tmp8= tmp[8 *tmpStride] - pad;\
        const int tmp9= tmp[9 *tmpStride] - pad;\
        const int tmp10=tmp[10*tmpStride] - pad;\
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065
        OP2(dst[0*dstStride], (tmp0+tmp1)*20 - (tmpA+tmp2)*5 + (tmpB+tmp3));\
        OP2(dst[1*dstStride], (tmp1+tmp2)*20 - (tmp0+tmp3)*5 + (tmpA+tmp4));\
        OP2(dst[2*dstStride], (tmp2+tmp3)*20 - (tmp1+tmp4)*5 + (tmp0+tmp5));\
        OP2(dst[3*dstStride], (tmp3+tmp4)*20 - (tmp2+tmp5)*5 + (tmp1+tmp6));\
        OP2(dst[4*dstStride], (tmp4+tmp5)*20 - (tmp3+tmp6)*5 + (tmp2+tmp7));\
        OP2(dst[5*dstStride], (tmp5+tmp6)*20 - (tmp4+tmp7)*5 + (tmp3+tmp8));\
        OP2(dst[6*dstStride], (tmp6+tmp7)*20 - (tmp5+tmp8)*5 + (tmp4+tmp9));\
        OP2(dst[7*dstStride], (tmp7+tmp8)*20 - (tmp6+tmp9)*5 + (tmp5+tmp10));\
        dst++;\
        tmp++;\
    }\
}\
\
1066 1067 1068
static void FUNC(OPNAME ## h264_qpel16_v_lowpass)(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    FUNC(OPNAME ## h264_qpel8_v_lowpass)(dst                , src                , dstStride, srcStride);\
    FUNC(OPNAME ## h264_qpel8_v_lowpass)(dst+8*sizeof(pixel), src+8*sizeof(pixel), dstStride, srcStride);\
1069 1070
    src += 8*srcStride;\
    dst += 8*dstStride;\
1071 1072
    FUNC(OPNAME ## h264_qpel8_v_lowpass)(dst                , src                , dstStride, srcStride);\
    FUNC(OPNAME ## h264_qpel8_v_lowpass)(dst+8*sizeof(pixel), src+8*sizeof(pixel), dstStride, srcStride);\
1073 1074
}\
\
1075 1076 1077
static void FUNC(OPNAME ## h264_qpel16_h_lowpass)(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    FUNC(OPNAME ## h264_qpel8_h_lowpass)(dst                , src                , dstStride, srcStride);\
    FUNC(OPNAME ## h264_qpel8_h_lowpass)(dst+8*sizeof(pixel), src+8*sizeof(pixel), dstStride, srcStride);\
1078 1079
    src += 8*srcStride;\
    dst += 8*dstStride;\
1080 1081
    FUNC(OPNAME ## h264_qpel8_h_lowpass)(dst                , src                , dstStride, srcStride);\
    FUNC(OPNAME ## h264_qpel8_h_lowpass)(dst+8*sizeof(pixel), src+8*sizeof(pixel), dstStride, srcStride);\
1082 1083
}\
\
1084 1085 1086
static void FUNC(OPNAME ## h264_qpel16_hv_lowpass)(uint8_t *dst, int16_t *tmp, uint8_t *src, int dstStride, int tmpStride, int srcStride){\
    FUNC(OPNAME ## h264_qpel8_hv_lowpass)(dst                , tmp  , src                , dstStride, tmpStride, srcStride);\
    FUNC(OPNAME ## h264_qpel8_hv_lowpass)(dst+8*sizeof(pixel), tmp+8, src+8*sizeof(pixel), dstStride, tmpStride, srcStride);\
1087 1088
    src += 8*srcStride;\
    dst += 8*dstStride;\
1089 1090
    FUNC(OPNAME ## h264_qpel8_hv_lowpass)(dst                , tmp  , src                , dstStride, tmpStride, srcStride);\
    FUNC(OPNAME ## h264_qpel8_hv_lowpass)(dst+8*sizeof(pixel), tmp+8, src+8*sizeof(pixel), dstStride, tmpStride, srcStride);\
1091 1092 1093
}\

#define H264_MC(OPNAME, SIZE) \
1094 1095
static av_unused void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc00)(uint8_t *dst, uint8_t *src, int stride){\
    FUNCC(OPNAME ## pixels ## SIZE)(dst, src, stride, SIZE);\
1096 1097
}\
\
1098 1099 1100 1101
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc10)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t half[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(half, src, SIZE*sizeof(pixel), stride);\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, src, half, stride, stride, SIZE*sizeof(pixel), SIZE);\
1102 1103
}\
\
1104 1105
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc20)(uint8_t *dst, uint8_t *src, int stride){\
    FUNC(OPNAME ## h264_qpel ## SIZE ## _h_lowpass)(dst, src, stride, stride);\
1106 1107
}\
\
1108 1109 1110 1111
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc30)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t half[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(half, src, SIZE*sizeof(pixel), stride);\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, src+sizeof(pixel), half, stride, stride, SIZE*sizeof(pixel), SIZE);\
1112 1113
}\
\
1114 1115 1116 1117 1118 1119 1120
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc01)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    uint8_t half[SIZE*SIZE*sizeof(pixel)];\
    FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(half, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, full_mid, half, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1121 1122
}\
\
1123 1124 1125 1126 1127
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc02)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(OPNAME ## h264_qpel ## SIZE ## _v_lowpass)(dst, full_mid, stride, SIZE*sizeof(pixel));\
1128 1129
}\
\
1130 1131 1132 1133 1134 1135 1136
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc03)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    uint8_t half[SIZE*SIZE*sizeof(pixel)];\
    FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(half, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, full_mid+SIZE*sizeof(pixel), half, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1137 1138
}\
\
1139 1140 1141 1142 1143 1144 1145 1146 1147
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc11)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(halfH, src, SIZE*sizeof(pixel), stride);\
    FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(halfV, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1148 1149
}\
\
1150 1151 1152 1153 1154 1155 1156 1157 1158
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc31)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(halfH, src, SIZE*sizeof(pixel), stride);\
    FUNC(copy_block ## SIZE )(full, src - stride*2 + sizeof(pixel), SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(halfV, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1159 1160
}\
\
1161 1162 1163 1164 1165 1166 1167 1168 1169
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc13)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(halfH, src + stride, SIZE*sizeof(pixel), stride);\
    FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(halfV, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1170 1171
}\
\
1172 1173 1174 1175 1176 1177 1178 1179 1180
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc33)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(halfH, src + stride, SIZE*sizeof(pixel), stride);\
    FUNC(copy_block ## SIZE )(full, src - stride*2 + sizeof(pixel), SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(halfV, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1181 1182
}\
\
1183 1184 1185
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc22)(uint8_t *dst, uint8_t *src, int stride){\
    int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
    FUNC(OPNAME ## h264_qpel ## SIZE ## _hv_lowpass)(dst, tmp, src, stride, SIZE*sizeof(pixel), stride);\
1186 1187
}\
\
1188 1189 1190 1191 1192 1193 1194
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc21)(uint8_t *dst, uint8_t *src, int stride){\
    int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfHV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(halfH, src, SIZE*sizeof(pixel), stride);\
    FUNC(put_h264_qpel ## SIZE ## _hv_lowpass)(halfHV, tmp, src, SIZE*sizeof(pixel), SIZE*sizeof(pixel), stride);\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1195 1196
}\
\
1197 1198 1199 1200 1201 1202 1203
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc23)(uint8_t *dst, uint8_t *src, int stride){\
    int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfHV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(halfH, src + stride, SIZE*sizeof(pixel), stride);\
    FUNC(put_h264_qpel ## SIZE ## _hv_lowpass)(halfHV, tmp, src, SIZE*sizeof(pixel), SIZE*sizeof(pixel), stride);\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1204 1205
}\
\
1206 1207 1208 1209 1210 1211 1212 1213 1214 1215
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc12)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t halfV[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfHV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(halfV, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(put_h264_qpel ## SIZE ## _hv_lowpass)(halfHV, tmp, src, SIZE*sizeof(pixel), SIZE*sizeof(pixel), stride);\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfV, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1216 1217
}\
\
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc32)(uint8_t *dst, uint8_t *src, int stride){\
    uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
    int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
    uint8_t halfV[SIZE*SIZE*sizeof(pixel)];\
    uint8_t halfHV[SIZE*SIZE*sizeof(pixel)];\
    FUNC(copy_block ## SIZE )(full, src - stride*2 + sizeof(pixel), SIZE*sizeof(pixel),  stride, SIZE + 5);\
    FUNC(put_h264_qpel ## SIZE ## _v_lowpass)(halfV, full_mid, SIZE*sizeof(pixel), SIZE*sizeof(pixel));\
    FUNC(put_h264_qpel ## SIZE ## _hv_lowpass)(halfHV, tmp, src, SIZE*sizeof(pixel), SIZE*sizeof(pixel), stride);\
    FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfV, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
1228 1229
}\

1230
#define op_avg(a, b)  a = (((a)+CLIP(((b) + 16)>>5)+1)>>1)
1231
//#define op_avg2(a, b) a = (((a)*w1+cm[((b) + 16)>>5]*w2 + o + 64)>>7)
1232 1233 1234
#define op_put(a, b)  a = CLIP(((b) + 16)>>5)
#define op2_avg(a, b)  a = (((a)+CLIP(((b) + 512)>>10)+1)>>1)
#define op2_put(a, b)  a = CLIP(((b) + 512)>>10)
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250

H264_LOWPASS(put_       , op_put, op2_put)
H264_LOWPASS(avg_       , op_avg, op2_avg)
H264_MC(put_, 2)
H264_MC(put_, 4)
H264_MC(put_, 8)
H264_MC(put_, 16)
H264_MC(avg_, 4)
H264_MC(avg_, 8)
H264_MC(avg_, 16)

#undef op_avg
#undef op_put
#undef op2_avg
#undef op2_put

1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
#if BIT_DEPTH == 8
#   define put_h264_qpel8_mc00_8_c  ff_put_pixels8x8_8_c
#   define avg_h264_qpel8_mc00_8_c  ff_avg_pixels8x8_8_c
#   define put_h264_qpel16_mc00_8_c ff_put_pixels16x16_8_c
#   define avg_h264_qpel16_mc00_8_c ff_avg_pixels16x16_8_c
#elif BIT_DEPTH == 9
#   define put_h264_qpel8_mc00_9_c  ff_put_pixels8x8_9_c
#   define avg_h264_qpel8_mc00_9_c  ff_avg_pixels8x8_9_c
#   define put_h264_qpel16_mc00_9_c ff_put_pixels16x16_9_c
#   define avg_h264_qpel16_mc00_9_c ff_avg_pixels16x16_9_c
#elif BIT_DEPTH == 10
#   define put_h264_qpel8_mc00_10_c  ff_put_pixels8x8_10_c
#   define avg_h264_qpel8_mc00_10_c  ff_avg_pixels8x8_10_c
#   define put_h264_qpel16_mc00_10_c ff_put_pixels16x16_10_c
#   define avg_h264_qpel16_mc00_10_c ff_avg_pixels16x16_10_c
#endif
1267

1268 1269
void FUNCC(ff_put_pixels8x8)(uint8_t *dst, uint8_t *src, int stride) {
    FUNCC(put_pixels8)(dst, src, stride, 8);
1270
}
1271 1272
void FUNCC(ff_avg_pixels8x8)(uint8_t *dst, uint8_t *src, int stride) {
    FUNCC(avg_pixels8)(dst, src, stride, 8);
1273
}
1274 1275
void FUNCC(ff_put_pixels16x16)(uint8_t *dst, uint8_t *src, int stride) {
    FUNCC(put_pixels16)(dst, src, stride, 16);
1276
}
1277 1278
void FUNCC(ff_avg_pixels16x16)(uint8_t *dst, uint8_t *src, int stride) {
    FUNCC(avg_pixels16)(dst, src, stride, 16);
1279 1280
}