Libav
motion_est_template.c
Go to the documentation of this file.
1 /*
2  * Motion estimation
3  * Copyright (c) 2002-2004 Michael Niedermayer
4  *
5  * This file is part of Libav.
6  *
7  * Libav is free software; you can redistribute it and/or
8  * modify it under the terms of the GNU Lesser General Public
9  * License as published by the Free Software Foundation; either
10  * version 2.1 of the License, or (at your option) any later version.
11  *
12  * Libav is distributed in the hope that it will be useful,
13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15  * Lesser General Public License for more details.
16  *
17  * You should have received a copy of the GNU Lesser General Public
18  * License along with Libav; if not, write to the Free Software
19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20  */
21 
27 #include "mpegvideo.h"
28 
29 //Let us hope gcc will remove the unused vars ...(gcc 3.2.2 seems to do it ...)
30 #define LOAD_COMMON\
31  uint32_t av_unused * const score_map= c->score_map;\
32  const int av_unused xmin= c->xmin;\
33  const int av_unused ymin= c->ymin;\
34  const int av_unused xmax= c->xmax;\
35  const int av_unused ymax= c->ymax;\
36  uint8_t *mv_penalty= c->current_mv_penalty;\
37  const int pred_x= c->pred_x;\
38  const int pred_y= c->pred_y;\
39 
40 #define CHECK_HALF_MV(dx, dy, x, y)\
41 {\
42  const int hx= 2*(x)+(dx);\
43  const int hy= 2*(y)+(dy);\
44  d= cmp_hpel(s, x, y, dx, dy, size, h, ref_index, src_index, cmp_sub, chroma_cmp_sub, flags);\
45  d += (mv_penalty[hx - pred_x] + mv_penalty[hy - pred_y])*penalty_factor;\
46  COPY3_IF_LT(dmin, d, bx, hx, by, hy)\
47 }
48 
50  int *mx_ptr, int *my_ptr, int dmin,
51  int src_index, int ref_index,
52  int size, int h)
53 {
54  MotionEstContext * const c= &s->me;
55  const int mx = *mx_ptr;
56  const int my = *my_ptr;
57  const int penalty_factor= c->sub_penalty_factor;
58  me_cmp_func cmp_sub, chroma_cmp_sub;
59  int bx=2*mx, by=2*my;
60 
62  int flags= c->sub_flags;
63 
64  //FIXME factorize
65 
66  cmp_sub = s->mecc.me_sub_cmp[size];
67  chroma_cmp_sub = s->mecc.me_sub_cmp[size + 1];
68 
69  if(c->skip){ //FIXME move out of hpel?
70  *mx_ptr = 0;
71  *my_ptr = 0;
72  return dmin;
73  }
74 
75  if(c->avctx->me_cmp != c->avctx->me_sub_cmp){
76  dmin= cmp(s, mx, my, 0, 0, size, h, ref_index, src_index, cmp_sub, chroma_cmp_sub, flags);
77  if(mx || my || size>0)
78  dmin += (mv_penalty[2*mx - pred_x] + mv_penalty[2*my - pred_y])*penalty_factor;
79  }
80 
81  if (mx > xmin && mx < xmax &&
82  my > ymin && my < ymax) {
83  int d= dmin;
84  const int index= (my<<ME_MAP_SHIFT) + mx;
85  const int t= score_map[(index-(1<<ME_MAP_SHIFT))&(ME_MAP_SIZE-1)]
86  + (mv_penalty[bx - pred_x] + mv_penalty[by-2 - pred_y])*c->penalty_factor;
87  const int l= score_map[(index- 1 )&(ME_MAP_SIZE-1)]
88  + (mv_penalty[bx-2 - pred_x] + mv_penalty[by - pred_y])*c->penalty_factor;
89  const int r= score_map[(index+ 1 )&(ME_MAP_SIZE-1)]
90  + (mv_penalty[bx+2 - pred_x] + mv_penalty[by - pred_y])*c->penalty_factor;
91  const int b= score_map[(index+(1<<ME_MAP_SHIFT))&(ME_MAP_SIZE-1)]
92  + (mv_penalty[bx - pred_x] + mv_penalty[by+2 - pred_y])*c->penalty_factor;
93 
94  unsigned key;
95  unsigned map_generation= c->map_generation;
96 #ifndef NDEBUG
97  uint32_t *map= c->map;
98 #endif
99  key= ((my-1)<<ME_MAP_MV_BITS) + (mx) + map_generation;
100  assert(map[(index-(1<<ME_MAP_SHIFT))&(ME_MAP_SIZE-1)] == key);
101  key= ((my+1)<<ME_MAP_MV_BITS) + (mx) + map_generation;
102  assert(map[(index+(1<<ME_MAP_SHIFT))&(ME_MAP_SIZE-1)] == key);
103  key= ((my)<<ME_MAP_MV_BITS) + (mx+1) + map_generation;
104  assert(map[(index+1)&(ME_MAP_SIZE-1)] == key);
105  key= ((my)<<ME_MAP_MV_BITS) + (mx-1) + map_generation;
106  assert(map[(index-1)&(ME_MAP_SIZE-1)] == key);
107  if(t<=b){
108  CHECK_HALF_MV(0, 1, mx ,my-1)
109  if(l<=r){
110  CHECK_HALF_MV(1, 1, mx-1, my-1)
111  if(t+r<=b+l){
112  CHECK_HALF_MV(1, 1, mx , my-1)
113  }else{
114  CHECK_HALF_MV(1, 1, mx-1, my )
115  }
116  CHECK_HALF_MV(1, 0, mx-1, my )
117  }else{
118  CHECK_HALF_MV(1, 1, mx , my-1)
119  if(t+l<=b+r){
120  CHECK_HALF_MV(1, 1, mx-1, my-1)
121  }else{
122  CHECK_HALF_MV(1, 1, mx , my )
123  }
124  CHECK_HALF_MV(1, 0, mx , my )
125  }
126  }else{
127  if(l<=r){
128  if(t+l<=b+r){
129  CHECK_HALF_MV(1, 1, mx-1, my-1)
130  }else{
131  CHECK_HALF_MV(1, 1, mx , my )
132  }
133  CHECK_HALF_MV(1, 0, mx-1, my)
134  CHECK_HALF_MV(1, 1, mx-1, my)
135  }else{
136  if(t+r<=b+l){
137  CHECK_HALF_MV(1, 1, mx , my-1)
138  }else{
139  CHECK_HALF_MV(1, 1, mx-1, my)
140  }
141  CHECK_HALF_MV(1, 0, mx , my)
142  CHECK_HALF_MV(1, 1, mx , my)
143  }
144  CHECK_HALF_MV(0, 1, mx , my)
145  }
146  assert(bx >= xmin*2 && bx <= xmax*2 && by >= ymin*2 && by <= ymax*2);
147  }
148 
149  *mx_ptr = bx;
150  *my_ptr = by;
151 
152  return dmin;
153 }
154 
156  int *mx_ptr, int *my_ptr, int dmin,
157  int src_index, int ref_index,
158  int size, int h)
159 {
160  (*mx_ptr)<<=1;
161  (*my_ptr)<<=1;
162  return dmin;
163 }
164 
165 static inline int get_mb_score(MpegEncContext *s, int mx, int my,
166  int src_index, int ref_index, int size,
167  int h, int add_rate)
168 {
169  MotionEstContext * const c= &s->me;
170  const int penalty_factor= c->mb_penalty_factor;
171  const int flags= c->mb_flags;
172  const int qpel= flags & FLAG_QPEL;
173  const int mask= 1+2*qpel;
174  me_cmp_func cmp_sub, chroma_cmp_sub;
175  int d;
176 
178 
179  //FIXME factorize
180 
181  cmp_sub = s->mecc.mb_cmp[size];
182  chroma_cmp_sub = s->mecc.mb_cmp[size + 1];
183 
184 // assert(!c->skip);
185 // assert(c->avctx->me_sub_cmp != c->avctx->mb_cmp);
186 
187  d= cmp(s, mx>>(qpel+1), my>>(qpel+1), mx&mask, my&mask, size, h, ref_index, src_index, cmp_sub, chroma_cmp_sub, flags);
188  //FIXME check cbp before adding penalty for (0,0) vector
189  if(add_rate && (mx || my || size>0))
190  d += (mv_penalty[mx - pred_x] + mv_penalty[my - pred_y])*penalty_factor;
191 
192  return d;
193 }
194 
195 int ff_get_mb_score(MpegEncContext *s, int mx, int my, int src_index,
196  int ref_index, int size, int h, int add_rate)
197 {
198  return get_mb_score(s, mx, my, src_index, ref_index, size, h, add_rate);
199 }
200 
201 #define CHECK_QUARTER_MV(dx, dy, x, y)\
202 {\
203  const int hx= 4*(x)+(dx);\
204  const int hy= 4*(y)+(dy);\
205  d= cmp_qpel(s, x, y, dx, dy, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);\
206  d += (mv_penalty[hx - pred_x] + mv_penalty[hy - pred_y])*penalty_factor;\
207  COPY3_IF_LT(dmin, d, bx, hx, by, hy)\
208 }
209 
211  int *mx_ptr, int *my_ptr, int dmin,
212  int src_index, int ref_index,
213  int size, int h)
214 {
215  MotionEstContext * const c= &s->me;
216  const int mx = *mx_ptr;
217  const int my = *my_ptr;
218  const int penalty_factor= c->sub_penalty_factor;
219  const unsigned map_generation = c->map_generation;
220  const int subpel_quality= c->avctx->me_subpel_quality;
221  uint32_t *map= c->map;
222  me_cmp_func cmpf, chroma_cmpf;
223  me_cmp_func cmp_sub, chroma_cmp_sub;
224 
226  int flags= c->sub_flags;
227 
228  cmpf = s->mecc.me_cmp[size];
229  chroma_cmpf = s->mecc.me_cmp[size + 1]; // FIXME: factorize
230  //FIXME factorize
231 
232  cmp_sub = s->mecc.me_sub_cmp[size];
233  chroma_cmp_sub = s->mecc.me_sub_cmp[size + 1];
234 
235  if(c->skip){ //FIXME somehow move up (benchmark)
236  *mx_ptr = 0;
237  *my_ptr = 0;
238  return dmin;
239  }
240 
241  if(c->avctx->me_cmp != c->avctx->me_sub_cmp){
242  dmin= cmp(s, mx, my, 0, 0, size, h, ref_index, src_index, cmp_sub, chroma_cmp_sub, flags);
243  if(mx || my || size>0)
244  dmin += (mv_penalty[4*mx - pred_x] + mv_penalty[4*my - pred_y])*penalty_factor;
245  }
246 
247  if (mx > xmin && mx < xmax &&
248  my > ymin && my < ymax) {
249  int bx=4*mx, by=4*my;
250  int d= dmin;
251  int i, nx, ny;
252  const int index= (my<<ME_MAP_SHIFT) + mx;
253  const int t= score_map[(index-(1<<ME_MAP_SHIFT) )&(ME_MAP_SIZE-1)];
254  const int l= score_map[(index- 1 )&(ME_MAP_SIZE-1)];
255  const int r= score_map[(index+ 1 )&(ME_MAP_SIZE-1)];
256  const int b= score_map[(index+(1<<ME_MAP_SHIFT) )&(ME_MAP_SIZE-1)];
257  const int c= score_map[(index )&(ME_MAP_SIZE-1)];
258  int best[8];
259  int best_pos[8][2];
260 
261  memset(best, 64, sizeof(int)*8);
262  if(s->me.dia_size>=2){
263  const int tl= score_map[(index-(1<<ME_MAP_SHIFT)-1)&(ME_MAP_SIZE-1)];
264  const int bl= score_map[(index+(1<<ME_MAP_SHIFT)-1)&(ME_MAP_SIZE-1)];
265  const int tr= score_map[(index-(1<<ME_MAP_SHIFT)+1)&(ME_MAP_SIZE-1)];
266  const int br= score_map[(index+(1<<ME_MAP_SHIFT)+1)&(ME_MAP_SIZE-1)];
267 
268  for(ny= -3; ny <= 3; ny++){
269  for(nx= -3; nx <= 3; nx++){
270  //FIXME this could overflow (unlikely though)
271  const int64_t t2= nx*nx*(tr + tl - 2*t) + 4*nx*(tr-tl) + 32*t;
272  const int64_t c2= nx*nx*( r + l - 2*c) + 4*nx*( r- l) + 32*c;
273  const int64_t b2= nx*nx*(br + bl - 2*b) + 4*nx*(br-bl) + 32*b;
274  int score= (ny*ny*(b2 + t2 - 2*c2) + 4*ny*(b2 - t2) + 32*c2 + 512)>>10;
275  int i;
276 
277  if((nx&3)==0 && (ny&3)==0) continue;
278 
279  score += (mv_penalty[4*mx + nx - pred_x] + mv_penalty[4*my + ny - pred_y])*penalty_factor;
280 
281 // if(nx&1) score-=1024*c->penalty_factor;
282 // if(ny&1) score-=1024*c->penalty_factor;
283 
284  for(i=0; i<8; i++){
285  if(score < best[i]){
286  memmove(&best[i+1], &best[i], sizeof(int)*(7-i));
287  memmove(&best_pos[i+1][0], &best_pos[i][0], sizeof(int)*2*(7-i));
288  best[i]= score;
289  best_pos[i][0]= nx + 4*mx;
290  best_pos[i][1]= ny + 4*my;
291  break;
292  }
293  }
294  }
295  }
296  }else{
297  int tl;
298  //FIXME this could overflow (unlikely though)
299  const int cx = 4*(r - l);
300  const int cx2= r + l - 2*c;
301  const int cy = 4*(b - t);
302  const int cy2= b + t - 2*c;
303  int cxy;
304 
305  if(map[(index-(1<<ME_MAP_SHIFT)-1)&(ME_MAP_SIZE-1)] == (my<<ME_MAP_MV_BITS) + mx + map_generation && 0){ //FIXME
306  tl= score_map[(index-(1<<ME_MAP_SHIFT)-1)&(ME_MAP_SIZE-1)];
307  }else{
308  tl= cmp(s, mx-1, my-1, 0, 0, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);//FIXME wrong if chroma me is different
309  }
310 
311  cxy= 2*tl + (cx + cy)/4 - (cx2 + cy2) - 2*c;
312 
313  assert(16*cx2 + 4*cx + 32*c == 32*r);
314  assert(16*cx2 - 4*cx + 32*c == 32*l);
315  assert(16*cy2 + 4*cy + 32*c == 32*b);
316  assert(16*cy2 - 4*cy + 32*c == 32*t);
317  assert(16*cxy + 16*cy2 + 16*cx2 - 4*cy - 4*cx + 32*c == 32*tl);
318 
319  for(ny= -3; ny <= 3; ny++){
320  for(nx= -3; nx <= 3; nx++){
321  //FIXME this could overflow (unlikely though)
322  int score= ny*nx*cxy + nx*nx*cx2 + ny*ny*cy2 + nx*cx + ny*cy + 32*c; //FIXME factor
323  int i;
324 
325  if((nx&3)==0 && (ny&3)==0) continue;
326 
327  score += 32*(mv_penalty[4*mx + nx - pred_x] + mv_penalty[4*my + ny - pred_y])*penalty_factor;
328 // if(nx&1) score-=32*c->penalty_factor;
329  // if(ny&1) score-=32*c->penalty_factor;
330 
331  for(i=0; i<8; i++){
332  if(score < best[i]){
333  memmove(&best[i+1], &best[i], sizeof(int)*(7-i));
334  memmove(&best_pos[i+1][0], &best_pos[i][0], sizeof(int)*2*(7-i));
335  best[i]= score;
336  best_pos[i][0]= nx + 4*mx;
337  best_pos[i][1]= ny + 4*my;
338  break;
339  }
340  }
341  }
342  }
343  }
344  for(i=0; i<subpel_quality; i++){
345  nx= best_pos[i][0];
346  ny= best_pos[i][1];
347  CHECK_QUARTER_MV(nx&3, ny&3, nx>>2, ny>>2)
348  }
349 
350  assert(bx >= xmin*4 && bx <= xmax*4 && by >= ymin*4 && by <= ymax*4);
351 
352  *mx_ptr = bx;
353  *my_ptr = by;
354  }else{
355  *mx_ptr =4*mx;
356  *my_ptr =4*my;
357  }
358 
359  return dmin;
360 }
361 
362 
363 #define CHECK_MV(x,y)\
364 {\
365  const unsigned key = ((y)<<ME_MAP_MV_BITS) + (x) + map_generation;\
366  const int index= (((y)<<ME_MAP_SHIFT) + (x))&(ME_MAP_SIZE-1);\
367  assert((x) >= xmin);\
368  assert((x) <= xmax);\
369  assert((y) >= ymin);\
370  assert((y) <= ymax);\
371  if(map[index]!=key){\
372  d= cmp(s, x, y, 0, 0, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);\
373  map[index]= key;\
374  score_map[index]= d;\
375  d += (mv_penalty[((x)<<shift)-pred_x] + mv_penalty[((y)<<shift)-pred_y])*penalty_factor;\
376  COPY3_IF_LT(dmin, d, best[0], x, best[1], y)\
377  }\
378 }
379 
380 #define CHECK_CLIPPED_MV(ax,ay)\
381 {\
382  const int Lx= ax;\
383  const int Ly= ay;\
384  const int Lx2= FFMAX(xmin, FFMIN(Lx, xmax));\
385  const int Ly2= FFMAX(ymin, FFMIN(Ly, ymax));\
386  CHECK_MV(Lx2, Ly2)\
387 }
388 
389 #define CHECK_MV_DIR(x,y,new_dir)\
390 {\
391  const unsigned key = ((y)<<ME_MAP_MV_BITS) + (x) + map_generation;\
392  const int index= (((y)<<ME_MAP_SHIFT) + (x))&(ME_MAP_SIZE-1);\
393  if(map[index]!=key){\
394  d= cmp(s, x, y, 0, 0, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);\
395  map[index]= key;\
396  score_map[index]= d;\
397  d += (mv_penalty[((x)<<shift)-pred_x] + mv_penalty[((y)<<shift)-pred_y])*penalty_factor;\
398  if(d<dmin){\
399  best[0]=x;\
400  best[1]=y;\
401  dmin=d;\
402  next_dir= new_dir;\
403  }\
404  }\
405 }
406 
407 #define check(x,y,S,v)\
408 if( (x)<(xmin<<(S)) ) printf("%d %d %d %d %d xmin" #v, xmin, (x), (y), s->mb_x, s->mb_y);\
409 if( (x)>(xmax<<(S)) ) printf("%d %d %d %d %d xmax" #v, xmax, (x), (y), s->mb_x, s->mb_y);\
410 if( (y)<(ymin<<(S)) ) printf("%d %d %d %d %d ymin" #v, ymin, (x), (y), s->mb_x, s->mb_y);\
411 if( (y)>(ymax<<(S)) ) printf("%d %d %d %d %d ymax" #v, ymax, (x), (y), s->mb_x, s->mb_y);\
412 
413 #define LOAD_COMMON2\
414  uint32_t *map= c->map;\
415  const int qpel= flags&FLAG_QPEL;\
416  const int shift= 1+qpel;\
417 
418 static av_always_inline int small_diamond_search(MpegEncContext * s, int *best, int dmin,
419  int src_index, int ref_index, const int penalty_factor,
420  int size, int h, int flags)
421 {
422  MotionEstContext * const c= &s->me;
423  me_cmp_func cmpf, chroma_cmpf;
424  int next_dir=-1;
427  unsigned map_generation = c->map_generation;
428 
429  cmpf = s->mecc.me_cmp[size];
430  chroma_cmpf = s->mecc.me_cmp[size + 1];
431 
432  { /* ensure that the best point is in the MAP as h/qpel refinement needs it */
433  const unsigned key = (best[1]<<ME_MAP_MV_BITS) + best[0] + map_generation;
434  const int index= ((best[1]<<ME_MAP_SHIFT) + best[0])&(ME_MAP_SIZE-1);
435  if (map[index] != key) { // this will be executed only very rarely
436  score_map[index]= cmp(s, best[0], best[1], 0, 0, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);
437  map[index]= key;
438  }
439  }
440 
441  for(;;){
442  int d;
443  const int dir= next_dir;
444  const int x= best[0];
445  const int y= best[1];
446  next_dir=-1;
447 
448  if(dir!=2 && x>xmin) CHECK_MV_DIR(x-1, y , 0)
449  if(dir!=3 && y>ymin) CHECK_MV_DIR(x , y-1, 1)
450  if(dir!=0 && x<xmax) CHECK_MV_DIR(x+1, y , 2)
451  if(dir!=1 && y<ymax) CHECK_MV_DIR(x , y+1, 3)
452 
453  if(next_dir==-1){
454  return dmin;
455  }
456  }
457 }
458 
459 static int funny_diamond_search(MpegEncContext * s, int *best, int dmin,
460  int src_index, int ref_index, const int penalty_factor,
461  int size, int h, int flags)
462 {
463  MotionEstContext * const c= &s->me;
464  me_cmp_func cmpf, chroma_cmpf;
465  int dia_size;
468  unsigned map_generation = c->map_generation;
469 
470  cmpf = s->mecc.me_cmp[size];
471  chroma_cmpf = s->mecc.me_cmp[size + 1];
472 
473  for(dia_size=1; dia_size<=4; dia_size++){
474  int dir;
475  const int x= best[0];
476  const int y= best[1];
477 
478  if(dia_size&(dia_size-1)) continue;
479 
480  if( x + dia_size > xmax
481  || x - dia_size < xmin
482  || y + dia_size > ymax
483  || y - dia_size < ymin)
484  continue;
485 
486  for(dir= 0; dir<dia_size; dir+=2){
487  int d;
488 
489  CHECK_MV(x + dir , y + dia_size - dir);
490  CHECK_MV(x + dia_size - dir, y - dir );
491  CHECK_MV(x - dir , y - dia_size + dir);
492  CHECK_MV(x - dia_size + dir, y + dir );
493  }
494 
495  if(x!=best[0] || y!=best[1])
496  dia_size=0;
497  }
498  return dmin;
499 }
500 
501 static int hex_search(MpegEncContext * s, int *best, int dmin,
502  int src_index, int ref_index, const int penalty_factor,
503  int size, int h, int flags, int dia_size)
504 {
505  MotionEstContext * const c= &s->me;
506  me_cmp_func cmpf, chroma_cmpf;
509  unsigned map_generation = c->map_generation;
510  int x,y,d;
511  const int dec= dia_size & (dia_size-1);
512 
513  cmpf = s->mecc.me_cmp[size];
514  chroma_cmpf = s->mecc.me_cmp[size + 1];
515 
516  for(;dia_size; dia_size= dec ? dia_size-1 : dia_size>>1){
517  do{
518  x= best[0];
519  y= best[1];
520 
521  CHECK_CLIPPED_MV(x -dia_size , y);
522  CHECK_CLIPPED_MV(x+ dia_size , y);
523  CHECK_CLIPPED_MV(x+( dia_size>>1), y+dia_size);
524  CHECK_CLIPPED_MV(x+( dia_size>>1), y-dia_size);
525  if(dia_size>1){
526  CHECK_CLIPPED_MV(x+(-dia_size>>1), y+dia_size);
527  CHECK_CLIPPED_MV(x+(-dia_size>>1), y-dia_size);
528  }
529  }while(best[0] != x || best[1] != y);
530  }
531 
532  return dmin;
533 }
534 
535 static int l2s_dia_search(MpegEncContext * s, int *best, int dmin,
536  int src_index, int ref_index, const int penalty_factor,
537  int size, int h, int flags)
538 {
539  MotionEstContext * const c= &s->me;
540  me_cmp_func cmpf, chroma_cmpf;
543  unsigned map_generation = c->map_generation;
544  int x,y,i,d;
545  int dia_size= c->dia_size&0xFF;
546  const int dec= dia_size & (dia_size-1);
547  static const int hex[8][2]={{-2, 0}, {-1,-1}, { 0,-2}, { 1,-1},
548  { 2, 0}, { 1, 1}, { 0, 2}, {-1, 1}};
549 
550  cmpf = s->mecc.me_cmp[size];
551  chroma_cmpf = s->mecc.me_cmp[size + 1];
552 
553  for(; dia_size; dia_size= dec ? dia_size-1 : dia_size>>1){
554  do{
555  x= best[0];
556  y= best[1];
557  for(i=0; i<8; i++){
558  CHECK_CLIPPED_MV(x+hex[i][0]*dia_size, y+hex[i][1]*dia_size);
559  }
560  }while(best[0] != x || best[1] != y);
561  }
562 
563  x= best[0];
564  y= best[1];
565  CHECK_CLIPPED_MV(x+1, y);
566  CHECK_CLIPPED_MV(x, y+1);
567  CHECK_CLIPPED_MV(x-1, y);
568  CHECK_CLIPPED_MV(x, y-1);
569 
570  return dmin;
571 }
572 
573 static int umh_search(MpegEncContext * s, int *best, int dmin,
574  int src_index, int ref_index, const int penalty_factor,
575  int size, int h, int flags)
576 {
577  MotionEstContext * const c= &s->me;
578  me_cmp_func cmpf, chroma_cmpf;
581  unsigned map_generation = c->map_generation;
582  int x,y,x2,y2, i, j, d;
583  const int dia_size= c->dia_size&0xFE;
584  static const int hex[16][2]={{-4,-2}, {-4,-1}, {-4, 0}, {-4, 1}, {-4, 2},
585  { 4,-2}, { 4,-1}, { 4, 0}, { 4, 1}, { 4, 2},
586  {-2, 3}, { 0, 4}, { 2, 3},
587  {-2,-3}, { 0,-4}, { 2,-3},};
588 
589  cmpf = s->mecc.me_cmp[size];
590  chroma_cmpf = s->mecc.me_cmp[size + 1];
591 
592  x= best[0];
593  y= best[1];
594  for(x2=FFMAX(x-dia_size+1, xmin); x2<=FFMIN(x+dia_size-1,xmax); x2+=2){
595  CHECK_MV(x2, y);
596  }
597  for(y2=FFMAX(y-dia_size/2+1, ymin); y2<=FFMIN(y+dia_size/2-1,ymax); y2+=2){
598  CHECK_MV(x, y2);
599  }
600 
601  x= best[0];
602  y= best[1];
603  for(y2=FFMAX(y-2, ymin); y2<=FFMIN(y+2,ymax); y2++){
604  for(x2=FFMAX(x-2, xmin); x2<=FFMIN(x+2,xmax); x2++){
605  CHECK_MV(x2, y2);
606  }
607  }
608 
609 //FIXME prevent the CLIP stuff
610 
611  for(j=1; j<=dia_size/4; j++){
612  for(i=0; i<16; i++){
613  CHECK_CLIPPED_MV(x+hex[i][0]*j, y+hex[i][1]*j);
614  }
615  }
616 
617  return hex_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags, 2);
618 }
619 
620 static int full_search(MpegEncContext * s, int *best, int dmin,
621  int src_index, int ref_index, const int penalty_factor,
622  int size, int h, int flags)
623 {
624  MotionEstContext * const c= &s->me;
625  me_cmp_func cmpf, chroma_cmpf;
628  unsigned map_generation = c->map_generation;
629  int x,y, d;
630  const int dia_size= c->dia_size&0xFF;
631 
632  cmpf = s->mecc.me_cmp[size];
633  chroma_cmpf = s->mecc.me_cmp[size + 1];
634 
635  for(y=FFMAX(-dia_size, ymin); y<=FFMIN(dia_size,ymax); y++){
636  for(x=FFMAX(-dia_size, xmin); x<=FFMIN(dia_size,xmax); x++){
637  CHECK_MV(x, y);
638  }
639  }
640 
641  x= best[0];
642  y= best[1];
643  d= dmin;
644  CHECK_CLIPPED_MV(x , y);
645  CHECK_CLIPPED_MV(x+1, y);
646  CHECK_CLIPPED_MV(x, y+1);
647  CHECK_CLIPPED_MV(x-1, y);
648  CHECK_CLIPPED_MV(x, y-1);
649  best[0]= x;
650  best[1]= y;
651 
652  return d;
653 }
654 
655 #define SAB_CHECK_MV(ax,ay)\
656 {\
657  const unsigned key = ((ay)<<ME_MAP_MV_BITS) + (ax) + map_generation;\
658  const int index= (((ay)<<ME_MAP_SHIFT) + (ax))&(ME_MAP_SIZE-1);\
659  if(map[index]!=key){\
660  d= cmp(s, ax, ay, 0, 0, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);\
661  map[index]= key;\
662  score_map[index]= d;\
663  d += (mv_penalty[((ax)<<shift)-pred_x] + mv_penalty[((ay)<<shift)-pred_y])*penalty_factor;\
664  if(d < minima[minima_count-1].height){\
665  int j=0;\
666  \
667  while(d >= minima[j].height) j++;\
668 \
669  memmove(&minima [j+1], &minima [j], (minima_count - j - 1)*sizeof(Minima));\
670 \
671  minima[j].checked= 0;\
672  minima[j].height= d;\
673  minima[j].x= ax;\
674  minima[j].y= ay;\
675  \
676  i=-1;\
677  continue;\
678  }\
679  }\
680 }
681 
682 #define MAX_SAB_SIZE ME_MAP_SIZE
683 static int sab_diamond_search(MpegEncContext * s, int *best, int dmin,
684  int src_index, int ref_index, const int penalty_factor,
685  int size, int h, int flags)
686 {
687  MotionEstContext * const c= &s->me;
688  me_cmp_func cmpf, chroma_cmpf;
689  Minima minima[MAX_SAB_SIZE];
690  const int minima_count= FFABS(c->dia_size);
691  int i, j;
694  unsigned map_generation = c->map_generation;
695 
696  cmpf = s->mecc.me_cmp[size];
697  chroma_cmpf = s->mecc.me_cmp[size + 1];
698 
699  /*Note j<MAX_SAB_SIZE is needed if MAX_SAB_SIZE < ME_MAP_SIZE as j can
700  become larger due to MVs overflowing their ME_MAP_MV_BITS bits space in map
701  */
702  for(j=i=0; i<ME_MAP_SIZE && j<MAX_SAB_SIZE; i++){
703  uint32_t key= map[i];
704 
705  key += (1<<(ME_MAP_MV_BITS-1)) + (1<<(2*ME_MAP_MV_BITS-1));
706 
707  if ((key & (-(1 << (2 * ME_MAP_MV_BITS)))) != map_generation)
708  continue;
709 
710  minima[j].height= score_map[i];
711  minima[j].x= key & ((1<<ME_MAP_MV_BITS)-1); key>>=ME_MAP_MV_BITS;
712  minima[j].y= key & ((1<<ME_MAP_MV_BITS)-1);
713  minima[j].x-= (1<<(ME_MAP_MV_BITS-1));
714  minima[j].y-= (1<<(ME_MAP_MV_BITS-1));
715 
716  // all entries in map should be in range except if the mv overflows their ME_MAP_MV_BITS bits space
717  if( minima[j].x > xmax || minima[j].x < xmin
718  || minima[j].y > ymax || minima[j].y < ymin)
719  continue;
720 
721  minima[j].checked=0;
722  if(minima[j].x || minima[j].y)
723  minima[j].height+= (mv_penalty[((minima[j].x)<<shift)-pred_x] + mv_penalty[((minima[j].y)<<shift)-pred_y])*penalty_factor;
724 
725  j++;
726  }
727 
728  qsort(minima, j, sizeof(Minima), minima_cmp);
729 
730  for(; j<minima_count; j++){
731  minima[j].height=256*256*256*64;
732  minima[j].checked=0;
733  minima[j].x= minima[j].y=0;
734  }
735 
736  for(i=0; i<minima_count; i++){
737  const int x= minima[i].x;
738  const int y= minima[i].y;
739  int d;
740 
741  if(minima[i].checked) continue;
742 
743  if( x >= xmax || x <= xmin
744  || y >= ymax || y <= ymin)
745  continue;
746 
747  SAB_CHECK_MV(x-1, y)
748  SAB_CHECK_MV(x+1, y)
749  SAB_CHECK_MV(x , y-1)
750  SAB_CHECK_MV(x , y+1)
751 
752  minima[i].checked= 1;
753  }
754 
755  best[0]= minima[0].x;
756  best[1]= minima[0].y;
757  dmin= minima[0].height;
758 
759  if( best[0] < xmax && best[0] > xmin
760  && best[1] < ymax && best[1] > ymin){
761  int d;
762  // ensure that the reference samples for hpel refinement are in the map
763  CHECK_MV(best[0]-1, best[1])
764  CHECK_MV(best[0]+1, best[1])
765  CHECK_MV(best[0], best[1]-1)
766  CHECK_MV(best[0], best[1]+1)
767  }
768  return dmin;
769 }
770 
771 static int var_diamond_search(MpegEncContext * s, int *best, int dmin,
772  int src_index, int ref_index, const int penalty_factor,
773  int size, int h, int flags)
774 {
775  MotionEstContext * const c= &s->me;
776  me_cmp_func cmpf, chroma_cmpf;
777  int dia_size;
780  unsigned map_generation = c->map_generation;
781 
782  cmpf = s->mecc.me_cmp[size];
783  chroma_cmpf = s->mecc.me_cmp[size + 1];
784 
785  for(dia_size=1; dia_size<=c->dia_size; dia_size++){
786  int dir, start, end;
787  const int x= best[0];
788  const int y= best[1];
789 
790  start= FFMAX(0, y + dia_size - ymax);
791  end = FFMIN(dia_size, xmax - x + 1);
792  for(dir= start; dir<end; dir++){
793  int d;
794 
795 //check(x + dir,y + dia_size - dir,0, a0)
796  CHECK_MV(x + dir , y + dia_size - dir);
797  }
798 
799  start= FFMAX(0, x + dia_size - xmax);
800  end = FFMIN(dia_size, y - ymin + 1);
801  for(dir= start; dir<end; dir++){
802  int d;
803 
804 //check(x + dia_size - dir, y - dir,0, a1)
805  CHECK_MV(x + dia_size - dir, y - dir );
806  }
807 
808  start= FFMAX(0, -y + dia_size + ymin );
809  end = FFMIN(dia_size, x - xmin + 1);
810  for(dir= start; dir<end; dir++){
811  int d;
812 
813 //check(x - dir,y - dia_size + dir,0, a2)
814  CHECK_MV(x - dir , y - dia_size + dir);
815  }
816 
817  start= FFMAX(0, -x + dia_size + xmin );
818  end = FFMIN(dia_size, ymax - y + 1);
819  for(dir= start; dir<end; dir++){
820  int d;
821 
822 //check(x - dia_size + dir, y + dir,0, a3)
823  CHECK_MV(x - dia_size + dir, y + dir );
824  }
825 
826  if(x!=best[0] || y!=best[1])
827  dia_size=0;
828  }
829  return dmin;
830 }
831 
832 static av_always_inline int diamond_search(MpegEncContext * s, int *best, int dmin,
833  int src_index, int ref_index, const int penalty_factor,
834  int size, int h, int flags){
835  MotionEstContext * const c= &s->me;
836  if(c->dia_size==-1)
837  return funny_diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
838  else if(c->dia_size<-1)
839  return sab_diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
840  else if(c->dia_size<2)
841  return small_diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
842  else if(c->dia_size>1024)
843  return full_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
844  else if(c->dia_size>768)
845  return umh_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
846  else if(c->dia_size>512)
847  return hex_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags, c->dia_size&0xFF);
848  else if(c->dia_size>256)
849  return l2s_dia_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
850  else
851  return var_diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
852 }
853 
860 static av_always_inline int epzs_motion_search_internal(MpegEncContext * s, int *mx_ptr, int *my_ptr,
861  int P[10][2], int src_index, int ref_index, int16_t (*last_mv)[2],
862  int ref_mv_scale, int flags, int size, int h)
863 {
864  MotionEstContext * const c= &s->me;
865  int best[2]={0, 0};
869  int d;
870  int dmin;
872  unsigned map_generation;
873  int penalty_factor;
874  const int ref_mv_stride= s->mb_stride; //pass as arg FIXME
875  const int ref_mv_xy = s->mb_x + s->mb_y * ref_mv_stride; // add to last_mv before passing FIXME
876  me_cmp_func cmpf, chroma_cmpf;
877 
880 
881  if(c->pre_pass){
882  penalty_factor= c->pre_penalty_factor;
883  cmpf = s->mecc.me_pre_cmp[size];
884  chroma_cmpf = s->mecc.me_pre_cmp[size + 1];
885  }else{
886  penalty_factor= c->penalty_factor;
887  cmpf = s->mecc.me_cmp[size];
888  chroma_cmpf = s->mecc.me_cmp[size + 1];
889  }
890 
891  map_generation= update_map_generation(c);
892 
893  assert(cmpf);
894  dmin= cmp(s, 0, 0, 0, 0, size, h, ref_index, src_index, cmpf, chroma_cmpf, flags);
895  map[0]= map_generation;
896  score_map[0]= dmin;
897 
898  //FIXME precalc first term below?
899  if ((s->pict_type == AV_PICTURE_TYPE_B && !(c->flags & FLAG_DIRECT)) ||
901  dmin += (mv_penalty[pred_x] + mv_penalty[pred_y])*penalty_factor;
902 
903  /* first line */
904  if (s->first_slice_line) {
905  CHECK_MV(P_LEFT[0]>>shift, P_LEFT[1]>>shift)
906  CHECK_CLIPPED_MV((last_mv[ref_mv_xy][0]*ref_mv_scale + (1<<15))>>16,
907  (last_mv[ref_mv_xy][1]*ref_mv_scale + (1<<15))>>16)
908  }else{
909  if(dmin<((h*h*s->avctx->mv0_threshold)>>8)
910  && ( P_LEFT[0] |P_LEFT[1]
911  |P_TOP[0] |P_TOP[1]
912  |P_TOPRIGHT[0]|P_TOPRIGHT[1])==0){
913  *mx_ptr= 0;
914  *my_ptr= 0;
915  c->skip=1;
916  return dmin;
917  }
918  CHECK_MV( P_MEDIAN[0] >>shift , P_MEDIAN[1] >>shift)
919  CHECK_CLIPPED_MV((P_MEDIAN[0]>>shift) , (P_MEDIAN[1]>>shift)-1)
920  CHECK_CLIPPED_MV((P_MEDIAN[0]>>shift) , (P_MEDIAN[1]>>shift)+1)
921  CHECK_CLIPPED_MV((P_MEDIAN[0]>>shift)-1, (P_MEDIAN[1]>>shift) )
922  CHECK_CLIPPED_MV((P_MEDIAN[0]>>shift)+1, (P_MEDIAN[1]>>shift) )
923  CHECK_CLIPPED_MV((last_mv[ref_mv_xy][0]*ref_mv_scale + (1<<15))>>16,
924  (last_mv[ref_mv_xy][1]*ref_mv_scale + (1<<15))>>16)
925  CHECK_MV(P_LEFT[0] >>shift, P_LEFT[1] >>shift)
926  CHECK_MV(P_TOP[0] >>shift, P_TOP[1] >>shift)
927  CHECK_MV(P_TOPRIGHT[0]>>shift, P_TOPRIGHT[1]>>shift)
928  }
929  if(dmin>h*h*4){
930  if(c->pre_pass){
931  CHECK_CLIPPED_MV((last_mv[ref_mv_xy-1][0]*ref_mv_scale + (1<<15))>>16,
932  (last_mv[ref_mv_xy-1][1]*ref_mv_scale + (1<<15))>>16)
933  if(!s->first_slice_line)
934  CHECK_CLIPPED_MV((last_mv[ref_mv_xy-ref_mv_stride][0]*ref_mv_scale + (1<<15))>>16,
935  (last_mv[ref_mv_xy-ref_mv_stride][1]*ref_mv_scale + (1<<15))>>16)
936  }else{
937  CHECK_CLIPPED_MV((last_mv[ref_mv_xy+1][0]*ref_mv_scale + (1<<15))>>16,
938  (last_mv[ref_mv_xy+1][1]*ref_mv_scale + (1<<15))>>16)
939  if(s->mb_y+1<s->end_mb_y) //FIXME replace at least with last_slice_line
940  CHECK_CLIPPED_MV((last_mv[ref_mv_xy+ref_mv_stride][0]*ref_mv_scale + (1<<15))>>16,
941  (last_mv[ref_mv_xy+ref_mv_stride][1]*ref_mv_scale + (1<<15))>>16)
942  }
943  }
944 
945  if(c->avctx->last_predictor_count){
946  const int count= c->avctx->last_predictor_count;
947  const int xstart= FFMAX(0, s->mb_x - count);
948  const int ystart= FFMAX(0, s->mb_y - count);
949  const int xend= FFMIN(s->mb_width , s->mb_x + count + 1);
950  const int yend= FFMIN(s->mb_height, s->mb_y + count + 1);
951  int mb_y;
952 
953  for(mb_y=ystart; mb_y<yend; mb_y++){
954  int mb_x;
955  for(mb_x=xstart; mb_x<xend; mb_x++){
956  const int xy= mb_x + 1 + (mb_y + 1)*ref_mv_stride;
957  int mx= (last_mv[xy][0]*ref_mv_scale + (1<<15))>>16;
958  int my= (last_mv[xy][1]*ref_mv_scale + (1<<15))>>16;
959 
960  if(mx>xmax || mx<xmin || my>ymax || my<ymin) continue;
961  CHECK_MV(mx,my)
962  }
963  }
964  }
965 
966 //check(best[0],best[1],0, b0)
967  dmin= diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
968 
969 //check(best[0],best[1],0, b1)
970  *mx_ptr= best[0];
971  *my_ptr= best[1];
972 
973  return dmin;
974 }
975 
976 //this function is dedicated to the brain damaged gcc
977 int ff_epzs_motion_search(MpegEncContext *s, int *mx_ptr, int *my_ptr,
978  int P[10][2], int src_index, int ref_index,
979  int16_t (*last_mv)[2], int ref_mv_scale,
980  int size, int h)
981 {
982  MotionEstContext * const c= &s->me;
983 //FIXME convert other functions in the same way if faster
984  if(c->flags==0 && h==16 && size==0){
985  return epzs_motion_search_internal(s, mx_ptr, my_ptr, P, src_index, ref_index, last_mv, ref_mv_scale, 0, 0, 16);
986 // case FLAG_QPEL:
987 // return epzs_motion_search_internal(s, mx_ptr, my_ptr, P, src_index, ref_index, last_mv, ref_mv_scale, FLAG_QPEL);
988  }else{
989  return epzs_motion_search_internal(s, mx_ptr, my_ptr, P, src_index, ref_index, last_mv, ref_mv_scale, c->flags, size, h);
990  }
991 }
992 
994  int *mx_ptr, int *my_ptr, int P[10][2],
995  int src_index, int ref_index, int16_t (*last_mv)[2],
996  int ref_mv_scale)
997 {
998  MotionEstContext * const c= &s->me;
999  int best[2]={0, 0};
1000  int d, dmin;
1001  unsigned map_generation;
1002  const int penalty_factor= c->penalty_factor;
1003  const int size=1;
1004  const int h=8;
1005  const int ref_mv_stride= s->mb_stride;
1006  const int ref_mv_xy= s->mb_x + s->mb_y *ref_mv_stride;
1007  me_cmp_func cmpf, chroma_cmpf;
1008  LOAD_COMMON
1009  int flags= c->flags;
1010  LOAD_COMMON2
1011 
1012  cmpf = s->mecc.me_cmp[size];
1013  chroma_cmpf = s->mecc.me_cmp[size + 1];
1014 
1015  map_generation= update_map_generation(c);
1016 
1017  dmin = 1000000;
1018 
1019  /* first line */
1020  if (s->first_slice_line) {
1021  CHECK_MV(P_LEFT[0]>>shift, P_LEFT[1]>>shift)
1022  CHECK_CLIPPED_MV((last_mv[ref_mv_xy][0]*ref_mv_scale + (1<<15))>>16,
1023  (last_mv[ref_mv_xy][1]*ref_mv_scale + (1<<15))>>16)
1024  CHECK_MV(P_MV1[0]>>shift, P_MV1[1]>>shift)
1025  }else{
1026  CHECK_MV(P_MV1[0]>>shift, P_MV1[1]>>shift)
1027  //FIXME try some early stop
1028  CHECK_MV(P_MEDIAN[0]>>shift, P_MEDIAN[1]>>shift)
1029  CHECK_MV(P_LEFT[0]>>shift, P_LEFT[1]>>shift)
1030  CHECK_MV(P_TOP[0]>>shift, P_TOP[1]>>shift)
1031  CHECK_MV(P_TOPRIGHT[0]>>shift, P_TOPRIGHT[1]>>shift)
1032  CHECK_CLIPPED_MV((last_mv[ref_mv_xy][0]*ref_mv_scale + (1<<15))>>16,
1033  (last_mv[ref_mv_xy][1]*ref_mv_scale + (1<<15))>>16)
1034  }
1035  if(dmin>64*4){
1036  CHECK_CLIPPED_MV((last_mv[ref_mv_xy+1][0]*ref_mv_scale + (1<<15))>>16,
1037  (last_mv[ref_mv_xy+1][1]*ref_mv_scale + (1<<15))>>16)
1038  if(s->mb_y+1<s->end_mb_y) //FIXME replace at least with last_slice_line
1039  CHECK_CLIPPED_MV((last_mv[ref_mv_xy+ref_mv_stride][0]*ref_mv_scale + (1<<15))>>16,
1040  (last_mv[ref_mv_xy+ref_mv_stride][1]*ref_mv_scale + (1<<15))>>16)
1041  }
1042 
1043  dmin= diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
1044 
1045  *mx_ptr= best[0];
1046  *my_ptr= best[1];
1047 
1048  return dmin;
1049 }
1050 
1051 //try to merge with above FIXME (needs PSNR test)
1053  int *mx_ptr, int *my_ptr, int P[10][2],
1054  int src_index, int ref_index, int16_t (*last_mv)[2],
1055  int ref_mv_scale)
1056 {
1057  MotionEstContext * const c= &s->me;
1058  int best[2]={0, 0};
1059  int d, dmin;
1060  unsigned map_generation;
1061  const int penalty_factor= c->penalty_factor;
1062  const int size=0; //FIXME pass as arg
1063  const int h=8;
1064  const int ref_mv_stride= s->mb_stride;
1065  const int ref_mv_xy= s->mb_x + s->mb_y *ref_mv_stride;
1066  me_cmp_func cmpf, chroma_cmpf;
1067  LOAD_COMMON
1068  int flags= c->flags;
1069  LOAD_COMMON2
1070 
1071  cmpf = s->mecc.me_cmp[size];
1072  chroma_cmpf = s->mecc.me_cmp[size + 1];
1073 
1074  map_generation= update_map_generation(c);
1075 
1076  dmin = 1000000;
1077 
1078  /* first line */
1079  if (s->first_slice_line) {
1080  CHECK_MV(P_LEFT[0]>>shift, P_LEFT[1]>>shift)
1081  CHECK_CLIPPED_MV((last_mv[ref_mv_xy][0]*ref_mv_scale + (1<<15))>>16,
1082  (last_mv[ref_mv_xy][1]*ref_mv_scale + (1<<15))>>16)
1083  CHECK_MV(P_MV1[0]>>shift, P_MV1[1]>>shift)
1084  }else{
1085  CHECK_MV(P_MV1[0]>>shift, P_MV1[1]>>shift)
1086  //FIXME try some early stop
1087  CHECK_MV(P_MEDIAN[0]>>shift, P_MEDIAN[1]>>shift)
1088  CHECK_MV(P_LEFT[0]>>shift, P_LEFT[1]>>shift)
1089  CHECK_MV(P_TOP[0]>>shift, P_TOP[1]>>shift)
1090  CHECK_MV(P_TOPRIGHT[0]>>shift, P_TOPRIGHT[1]>>shift)
1091  CHECK_CLIPPED_MV((last_mv[ref_mv_xy][0]*ref_mv_scale + (1<<15))>>16,
1092  (last_mv[ref_mv_xy][1]*ref_mv_scale + (1<<15))>>16)
1093  }
1094  if(dmin>64*4){
1095  CHECK_CLIPPED_MV((last_mv[ref_mv_xy+1][0]*ref_mv_scale + (1<<15))>>16,
1096  (last_mv[ref_mv_xy+1][1]*ref_mv_scale + (1<<15))>>16)
1097  if(s->mb_y+1<s->end_mb_y) //FIXME replace at least with last_slice_line
1098  CHECK_CLIPPED_MV((last_mv[ref_mv_xy+ref_mv_stride][0]*ref_mv_scale + (1<<15))>>16,
1099  (last_mv[ref_mv_xy+ref_mv_stride][1]*ref_mv_scale + (1<<15))>>16)
1100  }
1101 
1102  dmin= diamond_search(s, best, dmin, src_index, ref_index, penalty_factor, size, h, flags);
1103 
1104  *mx_ptr= best[0];
1105  *my_ptr= best[1];
1106 
1107  return dmin;
1108 }
static int minima_cmp(const void *a, const void *b)
Definition: motion_est.c:75
static unsigned update_map_generation(MotionEstContext *c)
Definition: motion_est.c:58
static int epzs_motion_search2(MpegEncContext *s, int *mx_ptr, int *my_ptr, int P[10][2], int src_index, int ref_index, int16_t(*last_mv)[2], int ref_mv_scale)
int size
#define P_TOPRIGHT
Definition: motion_est.c:46
int skip
set if ME is skipped for the current MB
Definition: motion_est.h:44
int last_mv[2][2][2]
last MV, used for MV prediction in MPEG-1 & B-frame MPEG-4
Definition: mpegvideo.h:272
int pre_pass
= 1 for the pre pass
Definition: motion_est.h:67
#define SAB_CHECK_MV(ax, ay)
#define P_LEFT
Definition: motion_est.c:44
int end_mb_y
end mb_y of this thread (so current thread should process start_mb_y <= row < end_mb_y) ...
Definition: mpegvideo.h:149
int sub_penalty_factor
Definition: motion_est.h:62
mpegvideo header.
int pre_penalty_factor
Definition: motion_est.h:56
int mpv_flags
flags set by private options
Definition: mpegvideo.h:504
me_cmp_func me_pre_cmp[6]
Definition: me_cmp.h:55
#define P_MV1
Definition: motion_est.c:48
#define CHECK_MV(x, y)
static int checked
Definition: cpu.c:45
int y
Definition: motion_est.c:71
#define ME_MAP_SIZE
Definition: motion_est.h:33
me_cmp_func mb_cmp[6]
Definition: me_cmp.h:58
#define FLAG_DIRECT
Definition: motion_est.c:84
#define b
Definition: input.c:52
Motion estimation context.
Definition: motion_est.h:42
int me_cmp
motion estimation comparison function
Definition: avcodec.h:1811
static int flags
Definition: log.c:50
int(* me_cmp_func)(struct MpegEncContext *c, uint8_t *blk1, uint8_t *blk2, ptrdiff_t stride, int h)
Definition: me_cmp.h:34
int ff_epzs_motion_search(MpegEncContext *s, int *mx_ptr, int *my_ptr, int P[10][2], int src_index, int ref_index, int16_t(*last_mv)[2], int ref_mv_scale, int size, int h)
int mb_height
number of MBs horizontally & vertically
Definition: mpegvideo.h:124
#define r
Definition: input.c:51
#define FF_MPV_FLAG_MV0
Definition: mpegvideo.h:553
#define CHECK_CLIPPED_MV(ax, ay)
int ff_get_mb_score(MpegEncContext *s, int mx, int my, int src_index, int ref_index, int size, int h, int add_rate)
int height
Definition: motion_est.c:70
static int hex_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags, int dia_size)
static const uint16_t mask[17]
Definition: lzw.c:38
static int epzs_motion_search4(MpegEncContext *s, int *mx_ptr, int *my_ptr, int P[10][2], int src_index, int ref_index, int16_t(*last_mv)[2], int ref_mv_scale)
static int no_sub_motion_search(MpegEncContext *s, int *mx_ptr, int *my_ptr, int dmin, int src_index, int ref_index, int size, int h)
#define LOAD_COMMON
int me_sub_cmp
subpixel motion estimation comparison function
Definition: avcodec.h:1817
#define P_MEDIAN
Definition: motion_est.c:47
static av_always_inline int diamond_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
#define FFMAX(a, b)
Definition: common.h:64
#define CHECK_MV_DIR(x, y, new_dir)
int checked
Definition: motion_est.c:72
#define CHECK_HALF_MV(dx, dy, x, y)
static av_always_inline int epzs_motion_search_internal(MpegEncContext *s, int *mx_ptr, int *my_ptr, int P[10][2], int src_index, int ref_index, int16_t(*last_mv)[2], int ref_mv_scale, int flags, int size, int h)
#define FLAG_QPEL
Definition: motion_est.c:82
#define FFMIN(a, b)
Definition: common.h:66
static int l2s_dia_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
#define ME_MAP_SHIFT
Definition: motion_est.c:50
static int hpel_motion_search(MpegEncContext *s, int *mx_ptr, int *my_ptr, int dmin, int src_index, int ref_index, int size, int h)
#define P_TOP
Definition: motion_est.c:45
unsigned map_generation
Definition: motion_est.h:55
#define FFABS(a)
Definition: common.h:61
static av_always_inline int cmp(MpegEncContext *s, const int x, const int y, const int subx, const int suby, const int size, const int h, int ref_index, int src_index, me_cmp_func cmp_func, me_cmp_func chroma_cmp_func, const int flags)
compares a block (either a full macroblock or a partition thereof) against a proposed motion-compensa...
Definition: motion_est.c:257
MotionEstContext me
Definition: mpegvideo.h:276
int last_predictor_count
amount of previous MV predictors (2a+1 x 2a+1 square)
Definition: avcodec.h:1857
static int get_mb_score(MpegEncContext *s, int mx, int my, int src_index, int ref_index, int size, int h, int add_rate)
static int qpel_motion_search(MpegEncContext *s, int *mx_ptr, int *my_ptr, int dmin, int src_index, int ref_index, int size, int h)
int penalty_factor
an estimate of the bits required to code a given mv value, e.g.
Definition: motion_est.h:57
int first_slice_line
used in MPEG-4 too to handle resync markers
Definition: mpegvideo.h:419
AVCodecContext * avctx
Definition: motion_est.h:43
static int full_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
int index
Definition: gxfenc.c:72
static int umh_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
#define MAX_SAB_SIZE
const VDPAUPixFmtMap * map
#define CHECK_QUARTER_MV(dx, dy, x, y)
static int funny_diamond_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
#define ME_MAP_MV_BITS
Definition: motion_est.c:51
int pict_type
AV_PICTURE_TYPE_I, AV_PICTURE_TYPE_P, AV_PICTURE_TYPE_B, ...
Definition: mpegvideo.h:206
me_cmp_func me_cmp[6]
Definition: me_cmp.h:56
MpegEncContext.
Definition: mpegvideo.h:76
struct AVCodecContext * avctx
Definition: mpegvideo.h:93
static uint8_t mv_penalty[MAX_FCODE+1][MAX_MV *2+1]
Table of number of bits a motion vector component needs.
Definition: ituh263enc.c:47
MECmpContext mecc
Definition: mpegvideo.h:222
int mb_stride
mb_width+1 used for some arrays to allow simple addressing of left & top MBs without sig11 ...
Definition: mpegvideo.h:125
static int sab_diamond_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
Bi-dir predicted.
Definition: avutil.h:262
me_cmp_func me_sub_cmp[6]
Definition: me_cmp.h:57
uint32_t * map
map to avoid duplicate evaluations
Definition: motion_est.h:53
int mv0_threshold
Note: Value depends upon the compare function used for fullpel ME.
Definition: avcodec.h:2093
static int var_diamond_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)
#define av_always_inline
Definition: attributes.h:40
int me_subpel_quality
subpel ME quality
Definition: avcodec.h:1884
int x
Definition: motion_est.c:71
#define LOAD_COMMON2
static av_always_inline int small_diamond_search(MpegEncContext *s, int *best, int dmin, int src_index, int ref_index, const int penalty_factor, int size, int h, int flags)