rdopt.c 153 KB
Newer Older
Jingning Han's avatar
Jingning Han committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
/*
 *  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */

#include <assert.h>
#include <math.h>

#include "./vp10_rtcd.h"
#include "./vpx_dsp_rtcd.h"

17
#include "vpx_dsp/vpx_dsp_common.h"
Jingning Han's avatar
Jingning Han committed
18
19
#include "vpx_mem/vpx_mem.h"
#include "vpx_ports/mem.h"
20
#include "vpx_ports/system_state.h"
Jingning Han's avatar
Jingning Han committed
21

22
23
24
25
26
27
28
29
30
31
32
#include "vp10/common/common.h"
#include "vp10/common/entropy.h"
#include "vp10/common/entropymode.h"
#include "vp10/common/idct.h"
#include "vp10/common/mvref_common.h"
#include "vp10/common/pred_common.h"
#include "vp10/common/quant_common.h"
#include "vp10/common/reconinter.h"
#include "vp10/common/reconintra.h"
#include "vp10/common/scan.h"
#include "vp10/common/seg_common.h"
Jingning Han's avatar
Jingning Han committed
33

34
35
36
37
38
39
40
41
42
43
#include "vp10/encoder/cost.h"
#include "vp10/encoder/encodemb.h"
#include "vp10/encoder/encodemv.h"
#include "vp10/encoder/encoder.h"
#include "vp10/encoder/mcomp.h"
#include "vp10/encoder/quantize.h"
#include "vp10/encoder/ratectrl.h"
#include "vp10/encoder/rd.h"
#include "vp10/encoder/rdopt.h"
#include "vp10/encoder/aq_variance.h"
Jingning Han's avatar
Jingning Han committed
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157

#define LAST_FRAME_MODE_MASK    ((1 << GOLDEN_FRAME) | (1 << ALTREF_FRAME) | \
                                 (1 << INTRA_FRAME))
#define GOLDEN_FRAME_MODE_MASK  ((1 << LAST_FRAME) | (1 << ALTREF_FRAME) | \
                                 (1 << INTRA_FRAME))
#define ALT_REF_MODE_MASK       ((1 << LAST_FRAME) | (1 << GOLDEN_FRAME) | \
                                 (1 << INTRA_FRAME))

#define SECOND_REF_FRAME_MASK   ((1 << ALTREF_FRAME) | 0x01)

#define MIN_EARLY_TERM_INDEX    3
#define NEW_MV_DISCOUNT_FACTOR  8

typedef struct {
  PREDICTION_MODE mode;
  MV_REFERENCE_FRAME ref_frame[2];
} MODE_DEFINITION;

typedef struct {
  MV_REFERENCE_FRAME ref_frame[2];
} REF_DEFINITION;

struct rdcost_block_args {
  MACROBLOCK *x;
  ENTROPY_CONTEXT t_above[16];
  ENTROPY_CONTEXT t_left[16];
  int this_rate;
  int64_t this_dist;
  int64_t this_sse;
  int64_t this_rd;
  int64_t best_rd;
  int exit_early;
  int use_fast_coef_costing;
  const scan_order *so;
  uint8_t skippable;
};

#define LAST_NEW_MV_INDEX 6
static const MODE_DEFINITION vp10_mode_order[MAX_MODES] = {
  {NEARESTMV, {LAST_FRAME,   NONE}},
  {NEARESTMV, {ALTREF_FRAME, NONE}},
  {NEARESTMV, {GOLDEN_FRAME, NONE}},

  {DC_PRED,   {INTRA_FRAME,  NONE}},

  {NEWMV,     {LAST_FRAME,   NONE}},
  {NEWMV,     {ALTREF_FRAME, NONE}},
  {NEWMV,     {GOLDEN_FRAME, NONE}},

  {NEARMV,    {LAST_FRAME,   NONE}},
  {NEARMV,    {ALTREF_FRAME, NONE}},
  {NEARMV,    {GOLDEN_FRAME, NONE}},

  {ZEROMV,    {LAST_FRAME,   NONE}},
  {ZEROMV,    {GOLDEN_FRAME, NONE}},
  {ZEROMV,    {ALTREF_FRAME, NONE}},

  {NEARESTMV, {LAST_FRAME,   ALTREF_FRAME}},
  {NEARESTMV, {GOLDEN_FRAME, ALTREF_FRAME}},

  {TM_PRED,   {INTRA_FRAME,  NONE}},

  {NEARMV,    {LAST_FRAME,   ALTREF_FRAME}},
  {NEWMV,     {LAST_FRAME,   ALTREF_FRAME}},
  {NEARMV,    {GOLDEN_FRAME, ALTREF_FRAME}},
  {NEWMV,     {GOLDEN_FRAME, ALTREF_FRAME}},

  {ZEROMV,    {LAST_FRAME,   ALTREF_FRAME}},
  {ZEROMV,    {GOLDEN_FRAME, ALTREF_FRAME}},

  {H_PRED,    {INTRA_FRAME,  NONE}},
  {V_PRED,    {INTRA_FRAME,  NONE}},
  {D135_PRED, {INTRA_FRAME,  NONE}},
  {D207_PRED, {INTRA_FRAME,  NONE}},
  {D153_PRED, {INTRA_FRAME,  NONE}},
  {D63_PRED,  {INTRA_FRAME,  NONE}},
  {D117_PRED, {INTRA_FRAME,  NONE}},
  {D45_PRED,  {INTRA_FRAME,  NONE}},
};

static const REF_DEFINITION vp10_ref_order[MAX_REFS] = {
  {{LAST_FRAME,   NONE}},
  {{GOLDEN_FRAME, NONE}},
  {{ALTREF_FRAME, NONE}},
  {{LAST_FRAME,   ALTREF_FRAME}},
  {{GOLDEN_FRAME, ALTREF_FRAME}},
  {{INTRA_FRAME,  NONE}},
};

static void swap_block_ptr(MACROBLOCK *x, PICK_MODE_CONTEXT *ctx,
                           int m, int n, int min_plane, int max_plane) {
  int i;

  for (i = min_plane; i < max_plane; ++i) {
    struct macroblock_plane *const p = &x->plane[i];
    struct macroblockd_plane *const pd = &x->e_mbd.plane[i];

    p->coeff    = ctx->coeff_pbuf[i][m];
    p->qcoeff   = ctx->qcoeff_pbuf[i][m];
    pd->dqcoeff = ctx->dqcoeff_pbuf[i][m];
    p->eobs     = ctx->eobs_pbuf[i][m];

    ctx->coeff_pbuf[i][m]   = ctx->coeff_pbuf[i][n];
    ctx->qcoeff_pbuf[i][m]  = ctx->qcoeff_pbuf[i][n];
    ctx->dqcoeff_pbuf[i][m] = ctx->dqcoeff_pbuf[i][n];
    ctx->eobs_pbuf[i][m]    = ctx->eobs_pbuf[i][n];

    ctx->coeff_pbuf[i][n]   = p->coeff;
    ctx->qcoeff_pbuf[i][n]  = p->qcoeff;
    ctx->dqcoeff_pbuf[i][n] = pd->dqcoeff;
    ctx->eobs_pbuf[i][n]    = p->eobs;
  }
}

Yaowu Xu's avatar
Yaowu Xu committed
158
static void model_rd_for_sb(VP10_COMP *cpi, BLOCK_SIZE bsize,
Jingning Han's avatar
Jingning Han committed
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
                            MACROBLOCK *x, MACROBLOCKD *xd,
                            int *out_rate_sum, int64_t *out_dist_sum,
                            int *skip_txfm_sb, int64_t *skip_sse_sb) {
  // Note our transform coeffs are 8 times an orthogonal transform.
  // Hence quantizer step is also 8 times. To get effective quantizer
  // we need to divide by 8 before sending to modeling function.
  int i;
  int64_t rate_sum = 0;
  int64_t dist_sum = 0;
  const int ref = xd->mi[0]->mbmi.ref_frame[0];
  unsigned int sse;
  unsigned int var = 0;
  unsigned int sum_sse = 0;
  int64_t total_sse = 0;
  int skip_flag = 1;
  const int shift = 6;
  int rate;
  int64_t dist;
  const int dequant_shift =
#if CONFIG_VP9_HIGHBITDEPTH
      (xd->cur_buf->flags & YV12_FLAG_HIGHBITDEPTH) ?
          xd->bd - 5 :
#endif  // CONFIG_VP9_HIGHBITDEPTH
          3;

  x->pred_sse[ref] = 0;

  for (i = 0; i < MAX_MB_PLANE; ++i) {
    struct macroblock_plane *const p = &x->plane[i];
    struct macroblockd_plane *const pd = &xd->plane[i];
    const BLOCK_SIZE bs = get_plane_block_size(bsize, pd);
    const TX_SIZE max_tx_size = max_txsize_lookup[bs];
    const BLOCK_SIZE unit_size = txsize_to_bsize[max_tx_size];
    const int64_t dc_thr = p->quant_thred[0] >> shift;
    const int64_t ac_thr = p->quant_thred[1] >> shift;
    // The low thresholds are used to measure if the prediction errors are
    // low enough so that we can skip the mode search.
196
197
    const int64_t low_dc_thr = VPXMIN(50, dc_thr >> 2);
    const int64_t low_ac_thr = VPXMIN(80, ac_thr >> 2);
Jingning Han's avatar
Jingning Han committed
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
    int bw = 1 << (b_width_log2_lookup[bs] - b_width_log2_lookup[unit_size]);
    int bh = 1 << (b_height_log2_lookup[bs] - b_width_log2_lookup[unit_size]);
    int idx, idy;
    int lw = b_width_log2_lookup[unit_size] + 2;
    int lh = b_height_log2_lookup[unit_size] + 2;

    sum_sse = 0;

    for (idy = 0; idy < bh; ++idy) {
      for (idx = 0; idx < bw; ++idx) {
        uint8_t *src = p->src.buf + (idy * p->src.stride << lh) + (idx << lw);
        uint8_t *dst = pd->dst.buf + (idy * pd->dst.stride << lh) + (idx << lh);
        int block_idx = (idy << 1) + idx;
        int low_err_skip = 0;

        var = cpi->fn_ptr[unit_size].vf(src, p->src.stride,
                                        dst, pd->dst.stride, &sse);
        x->bsse[(i << 2) + block_idx] = sse;
        sum_sse += sse;

        x->skip_txfm[(i << 2) + block_idx] = SKIP_TXFM_NONE;
        if (!x->select_tx_size) {
          // Check if all ac coefficients can be quantized to zero.
          if (var < ac_thr || var == 0) {
            x->skip_txfm[(i << 2) + block_idx] = SKIP_TXFM_AC_ONLY;

            // Check if dc coefficient can be quantized to zero.
            if (sse - var < dc_thr || sse == var) {
              x->skip_txfm[(i << 2) + block_idx] = SKIP_TXFM_AC_DC;

              if (!sse || (var < low_ac_thr && sse - var < low_dc_thr))
                low_err_skip = 1;
            }
          }
        }

        if (skip_flag && !low_err_skip)
          skip_flag = 0;

        if (i == 0)
          x->pred_sse[ref] += sse;
      }
    }

    total_sse += sum_sse;

    // Fast approximate the modelling function.
    if (cpi->sf.simple_model_rd_from_var) {
      int64_t rate;
      const int64_t square_error = sum_sse;
      int quantizer = (pd->dequant[1] >> dequant_shift);

      if (quantizer < 120)
        rate = (square_error * (280 - quantizer)) >> 8;
      else
        rate = 0;
      dist = (square_error * quantizer) >> 8;
      rate_sum += rate;
      dist_sum += dist;
    } else {
      vp10_model_rd_from_var_lapndz(sum_sse, num_pels_log2_lookup[bs],
                                   pd->dequant[1] >> dequant_shift,
                                   &rate, &dist);
      rate_sum += rate;
      dist_sum += dist;
    }
  }

  *skip_txfm_sb = skip_flag;
  *skip_sse_sb = total_sse << 4;
  *out_rate_sum = (int)rate_sum;
  *out_dist_sum = dist_sum << 4;
}

int64_t vp10_block_error_c(const tran_low_t *coeff, const tran_low_t *dqcoeff,
                          intptr_t block_size, int64_t *ssz) {
  int i;
  int64_t error = 0, sqcoeff = 0;

  for (i = 0; i < block_size; i++) {
    const int diff = coeff[i] - dqcoeff[i];
    error +=  diff * diff;
    sqcoeff += coeff[i] * coeff[i];
  }

  *ssz = sqcoeff;
  return error;
}

int64_t vp10_block_error_fp_c(const int16_t *coeff, const int16_t *dqcoeff,
                             int block_size) {
  int i;
  int64_t error = 0;

  for (i = 0; i < block_size; i++) {
    const int diff = coeff[i] - dqcoeff[i];
    error +=  diff * diff;
  }

  return error;
}

#if CONFIG_VP9_HIGHBITDEPTH
int64_t vp10_highbd_block_error_c(const tran_low_t *coeff,
                                 const tran_low_t *dqcoeff,
                                 intptr_t block_size,
                                 int64_t *ssz, int bd) {
  int i;
  int64_t error = 0, sqcoeff = 0;
  int shift = 2 * (bd - 8);
  int rounding = shift > 0 ? 1 << (shift - 1) : 0;

  for (i = 0; i < block_size; i++) {
    const int64_t diff = coeff[i] - dqcoeff[i];
    error +=  diff * diff;
    sqcoeff += (int64_t)coeff[i] * (int64_t)coeff[i];
  }
  assert(error >= 0 && sqcoeff >= 0);
  error = (error + rounding) >> shift;
  sqcoeff = (sqcoeff + rounding) >> shift;

  *ssz = sqcoeff;
  return error;
}
#endif  // CONFIG_VP9_HIGHBITDEPTH

/* The trailing '0' is a terminator which is used inside cost_coeffs() to
 * decide whether to include cost of a trailing EOB node or not (i.e. we
 * can skip this if the last coefficient in this transform block, e.g. the
 * 16th coefficient in a 4x4 block or the 64th coefficient in a 8x8 block,
 * were non-zero). */
static const int16_t band_counts[TX_SIZES][8] = {
  { 1, 2, 3, 4,  3,   16 - 13, 0 },
  { 1, 2, 3, 4, 11,   64 - 21, 0 },
  { 1, 2, 3, 4, 11,  256 - 21, 0 },
  { 1, 2, 3, 4, 11, 1024 - 21, 0 },
};
static int cost_coeffs(MACROBLOCK *x,
                       int plane, int block,
                       ENTROPY_CONTEXT *A, ENTROPY_CONTEXT *L,
                       TX_SIZE tx_size,
                       const int16_t *scan, const int16_t *nb,
                       int use_fast_coef_costing) {
  MACROBLOCKD *const xd = &x->e_mbd;
  MB_MODE_INFO *mbmi = &xd->mi[0]->mbmi;
  const struct macroblock_plane *p = &x->plane[plane];
  const struct macroblockd_plane *pd = &xd->plane[plane];
  const PLANE_TYPE type = pd->plane_type;
  const int16_t *band_count = &band_counts[tx_size][1];
  const int eob = p->eobs[block];
  const tran_low_t *const qcoeff = BLOCK_OFFSET(p->qcoeff, block);
  unsigned int (*token_costs)[2][COEFF_CONTEXTS][ENTROPY_TOKENS] =
                   x->token_costs[tx_size][type][is_inter_block(mbmi)];
  uint8_t token_cache[32 * 32];
  int pt = combine_entropy_contexts(*A, *L);
  int c, cost;
#if CONFIG_VP9_HIGHBITDEPTH
  const int16_t *cat6_high_cost = vp10_get_high_cost_table(xd->bd);
#else
  const int16_t *cat6_high_cost = vp10_get_high_cost_table(8);
#endif

  // Check for consistency of tx_size with mode info
  assert(type == PLANE_TYPE_Y ? mbmi->tx_size == tx_size
                              : get_uv_tx_size(mbmi, pd) == tx_size);

  if (eob == 0) {
    // single eob token
    cost = token_costs[0][0][pt][EOB_TOKEN];
    c = 0;
  } else {
    int band_left = *band_count++;

    // dc token
    int v = qcoeff[0];
    int16_t prev_t;
    EXTRABIT e;
    vp10_get_token_extra(v, &prev_t, &e);
    cost = (*token_costs)[0][pt][prev_t] +
        vp10_get_cost(prev_t, e, cat6_high_cost);

    token_cache[0] = vp10_pt_energy_class[prev_t];
    ++token_costs;

    // ac tokens
    for (c = 1; c < eob; c++) {
      const int rc = scan[c];
      int16_t t;

      v = qcoeff[rc];
      vp10_get_token_extra(v, &t, &e);
      if (use_fast_coef_costing) {
        cost += (*token_costs)[!prev_t][!prev_t][t] +
            vp10_get_cost(t, e, cat6_high_cost);
      } else {
        pt = get_coef_context(nb, token_cache, c);
        cost += (*token_costs)[!prev_t][pt][t] +
            vp10_get_cost(t, e, cat6_high_cost);
        token_cache[rc] = vp10_pt_energy_class[t];
      }
      prev_t = t;
      if (!--band_left) {
        band_left = *band_count++;
        ++token_costs;
      }
    }

    // eob token
    if (band_left) {
      if (use_fast_coef_costing) {
        cost += (*token_costs)[0][!prev_t][EOB_TOKEN];
      } else {
        pt = get_coef_context(nb, token_cache, c);
        cost += (*token_costs)[0][pt][EOB_TOKEN];
      }
    }
  }

  // is eob first coefficient;
  *A = *L = (c > 0);

  return cost;
}

static void dist_block(MACROBLOCK *x, int plane, int block, TX_SIZE tx_size,
                       int64_t *out_dist, int64_t *out_sse) {
  const int ss_txfrm_size = tx_size << 1;
  MACROBLOCKD* const xd = &x->e_mbd;
  const struct macroblock_plane *const p = &x->plane[plane];
  const struct macroblockd_plane *const pd = &xd->plane[plane];
  int64_t this_sse;
  int shift = tx_size == TX_32X32 ? 0 : 2;
  tran_low_t *const coeff = BLOCK_OFFSET(p->coeff, block);
  tran_low_t *const dqcoeff = BLOCK_OFFSET(pd->dqcoeff, block);
#if CONFIG_VP9_HIGHBITDEPTH
  const int bd = (xd->cur_buf->flags & YV12_FLAG_HIGHBITDEPTH) ? xd->bd : 8;
  *out_dist = vp10_highbd_block_error(coeff, dqcoeff, 16 << ss_txfrm_size,
                                     &this_sse, bd) >> shift;
#else
  *out_dist = vp10_block_error(coeff, dqcoeff, 16 << ss_txfrm_size,
                              &this_sse) >> shift;
#endif  // CONFIG_VP9_HIGHBITDEPTH
  *out_sse = this_sse >> shift;
}

static int rate_block(int plane, int block, BLOCK_SIZE plane_bsize,
                      TX_SIZE tx_size, struct rdcost_block_args* args) {
  int x_idx, y_idx;
  txfrm_block_to_raster_xy(plane_bsize, tx_size, block, &x_idx, &y_idx);

  return cost_coeffs(args->x, plane, block, args->t_above + x_idx,
                     args->t_left + y_idx, tx_size,
                     args->so->scan, args->so->neighbors,
                     args->use_fast_coef_costing);
}

static void block_rd_txfm(int plane, int block, BLOCK_SIZE plane_bsize,
                          TX_SIZE tx_size, void *arg) {
  struct rdcost_block_args *args = arg;
  MACROBLOCK *const x = args->x;
  MACROBLOCKD *const xd = &x->e_mbd;
  MB_MODE_INFO *const mbmi = &xd->mi[0]->mbmi;
  int64_t rd1, rd2, rd;
  int rate;
  int64_t dist;
  int64_t sse;

  if (args->exit_early)
    return;

  if (!is_inter_block(mbmi)) {
    struct encode_b_args arg = {x, NULL, &mbmi->skip};
    vp10_encode_block_intra(plane, block, plane_bsize, tx_size, &arg);
    dist_block(x, plane, block, tx_size, &dist, &sse);
  } else if (max_txsize_lookup[plane_bsize] == tx_size) {
    if (x->skip_txfm[(plane << 2) + (block >> (tx_size << 1))] ==
        SKIP_TXFM_NONE) {
      // full forward transform and quantization
      vp10_xform_quant(x, plane, block, plane_bsize, tx_size);
      dist_block(x, plane, block, tx_size, &dist, &sse);
    } else if (x->skip_txfm[(plane << 2) + (block >> (tx_size << 1))] ==
               SKIP_TXFM_AC_ONLY) {
      // compute DC coefficient
      tran_low_t *const coeff   = BLOCK_OFFSET(x->plane[plane].coeff, block);
      tran_low_t *const dqcoeff = BLOCK_OFFSET(xd->plane[plane].dqcoeff, block);
      vp10_xform_quant_dc(x, plane, block, plane_bsize, tx_size);
      sse  = x->bsse[(plane << 2) + (block >> (tx_size << 1))] << 4;
      dist = sse;
      if (x->plane[plane].eobs[block]) {
        const int64_t orig_sse = (int64_t)coeff[0] * coeff[0];
        const int64_t resd_sse = coeff[0] - dqcoeff[0];
        int64_t dc_correct = orig_sse - resd_sse * resd_sse;
#if CONFIG_VP9_HIGHBITDEPTH
        dc_correct >>= ((xd->bd - 8) * 2);
#endif
        if (tx_size != TX_32X32)
          dc_correct >>= 2;

496
        dist = VPXMAX(0, sse - dc_correct);
Jingning Han's avatar
Jingning Han committed
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
      }
    } else {
      // SKIP_TXFM_AC_DC
      // skip forward transform
      x->plane[plane].eobs[block] = 0;
      sse  = x->bsse[(plane << 2) + (block >> (tx_size << 1))] << 4;
      dist = sse;
    }
  } else {
    // full forward transform and quantization
    vp10_xform_quant(x, plane, block, plane_bsize, tx_size);
    dist_block(x, plane, block, tx_size, &dist, &sse);
  }

  rd = RDCOST(x->rdmult, x->rddiv, 0, dist);
  if (args->this_rd + rd > args->best_rd) {
    args->exit_early = 1;
    return;
  }

  rate = rate_block(plane, block, plane_bsize, tx_size, args);
  rd1 = RDCOST(x->rdmult, x->rddiv, rate, dist);
  rd2 = RDCOST(x->rdmult, x->rddiv, 0, sse);

  // TODO(jingning): temporarily enabled only for luma component
522
  rd = VPXMIN(rd1, rd2);
Jingning Han's avatar
Jingning Han committed
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
  if (plane == 0)
    x->zcoeff_blk[tx_size][block] = !x->plane[plane].eobs[block] ||
                                    (rd1 > rd2 && !xd->lossless);

  args->this_rate += rate;
  args->this_dist += dist;
  args->this_sse += sse;
  args->this_rd += rd;

  if (args->this_rd > args->best_rd) {
    args->exit_early = 1;
    return;
  }

  args->skippable &= !x->plane[plane].eobs[block];
}

static void txfm_rd_in_plane(MACROBLOCK *x,
                             int *rate, int64_t *distortion,
                             int *skippable, int64_t *sse,
                             int64_t ref_best_rd, int plane,
                             BLOCK_SIZE bsize, TX_SIZE tx_size,
                             int use_fast_coef_casting) {
  MACROBLOCKD *const xd = &x->e_mbd;
  const struct macroblockd_plane *const pd = &xd->plane[plane];
hui su's avatar
hui su committed
548
  TX_TYPE tx_type;
Jingning Han's avatar
Jingning Han committed
549
550
551
552
553
554
555
556
557
558
559
560
  struct rdcost_block_args args;
  vp10_zero(args);
  args.x = x;
  args.best_rd = ref_best_rd;
  args.use_fast_coef_costing = use_fast_coef_casting;
  args.skippable = 1;

  if (plane == 0)
    xd->mi[0]->mbmi.tx_size = tx_size;

  vp10_get_entropy_contexts(bsize, tx_size, pd, args.t_above, args.t_left);

hui su's avatar
hui su committed
561
562
  tx_type = get_tx_type(pd->plane_type, xd, 0);
  args.so = get_scan(tx_size, tx_type);
Jingning Han's avatar
Jingning Han committed
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578

  vp10_foreach_transformed_block_in_plane(xd, bsize, plane,
                                         block_rd_txfm, &args);
  if (args.exit_early) {
    *rate       = INT_MAX;
    *distortion = INT64_MAX;
    *sse        = INT64_MAX;
    *skippable  = 0;
  } else {
    *distortion = args.this_dist;
    *rate       = args.this_rate;
    *sse        = args.this_sse;
    *skippable  = args.skippable;
  }
}

Yaowu Xu's avatar
Yaowu Xu committed
579
static void choose_largest_tx_size(VP10_COMP *cpi, MACROBLOCK *x,
Jingning Han's avatar
Jingning Han committed
580
581
582
583
584
                                   int *rate, int64_t *distortion,
                                   int *skip, int64_t *sse,
                                   int64_t ref_best_rd,
                                   BLOCK_SIZE bs) {
  const TX_SIZE max_tx_size = max_txsize_lookup[bs];
Yaowu Xu's avatar
Yaowu Xu committed
585
  VP10_COMMON *const cm = &cpi->common;
Jingning Han's avatar
Jingning Han committed
586
587
588
589
  const TX_SIZE largest_tx_size = tx_mode_to_biggest_tx_size[cm->tx_mode];
  MACROBLOCKD *const xd = &x->e_mbd;
  MB_MODE_INFO *const mbmi = &xd->mi[0]->mbmi;

590
  mbmi->tx_size = VPXMIN(max_tx_size, largest_tx_size);
Jingning Han's avatar
Jingning Han committed
591
592
593
594
595
596

  txfm_rd_in_plane(x, rate, distortion, skip,
                   sse, ref_best_rd, 0, bs,
                   mbmi->tx_size, cpi->sf.use_fast_coef_costing);
}

Yaowu Xu's avatar
Yaowu Xu committed
597
static void choose_tx_size_from_rd(VP10_COMP *cpi, MACROBLOCK *x,
Jingning Han's avatar
Jingning Han committed
598
599
600
601
602
603
604
                                   int *rate,
                                   int64_t *distortion,
                                   int *skip,
                                   int64_t *psse,
                                   int64_t ref_best_rd,
                                   BLOCK_SIZE bs) {
  const TX_SIZE max_tx_size = max_txsize_lookup[bs];
Yaowu Xu's avatar
Yaowu Xu committed
605
  VP10_COMMON *const cm = &cpi->common;
Jingning Han's avatar
Jingning Han committed
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
  MACROBLOCKD *const xd = &x->e_mbd;
  MB_MODE_INFO *const mbmi = &xd->mi[0]->mbmi;
  vpx_prob skip_prob = vp10_get_skip_prob(cm, xd);
  int r[TX_SIZES][2], s[TX_SIZES];
  int64_t d[TX_SIZES], sse[TX_SIZES];
  int64_t rd[TX_SIZES][2] = {{INT64_MAX, INT64_MAX},
                             {INT64_MAX, INT64_MAX},
                             {INT64_MAX, INT64_MAX},
                             {INT64_MAX, INT64_MAX}};
  int n, m;
  int s0, s1;
  int64_t best_rd = INT64_MAX;
  TX_SIZE best_tx = max_tx_size;
  int start_tx, end_tx;

  const vpx_prob *tx_probs = get_tx_probs2(max_tx_size, xd, &cm->fc->tx_probs);
  assert(skip_prob > 0);
  s0 = vp10_cost_bit(skip_prob, 0);
  s1 = vp10_cost_bit(skip_prob, 1);

  if (cm->tx_mode == TX_MODE_SELECT) {
    start_tx = max_tx_size;
    end_tx = 0;
  } else {
630
631
    TX_SIZE chosen_tx_size = VPXMIN(max_tx_size,
                                    tx_mode_to_biggest_tx_size[cm->tx_mode]);
Jingning Han's avatar
Jingning Han committed
632
633
634
635
636
    start_tx = chosen_tx_size;
    end_tx = chosen_tx_size;
  }

  for (n = start_tx; n >= end_tx; n--) {
637
638
639
640
641
642
643
    int r_tx_size = 0;
    for (m = 0; m <= n - (n == (int) max_tx_size); m++) {
      if (m == n)
        r_tx_size += vp10_cost_zero(tx_probs[m]);
      else
        r_tx_size += vp10_cost_one(tx_probs[m]);
    }
Jingning Han's avatar
Jingning Han committed
644
645
646
647
648
    txfm_rd_in_plane(x, &r[n][0], &d[n], &s[n],
                     &sse[n], ref_best_rd, 0, bs, n,
                     cpi->sf.use_fast_coef_costing);
    r[n][1] = r[n][0];
    if (r[n][0] < INT_MAX) {
649
      r[n][1] += r_tx_size;
Jingning Han's avatar
Jingning Han committed
650
    }
651
    if (d[n] == INT64_MAX || r[n][0] == INT_MAX) {
Jingning Han's avatar
Jingning Han committed
652
653
      rd[n][0] = rd[n][1] = INT64_MAX;
    } else if (s[n]) {
654
655
      if (is_inter_block(mbmi)) {
        rd[n][0] = rd[n][1] = RDCOST(x->rdmult, x->rddiv, s1, sse[n]);
656
        r[n][1] -= r_tx_size;
657
658
659
660
      } else {
        rd[n][0] = RDCOST(x->rdmult, x->rddiv, s1, sse[n]);
        rd[n][1] = RDCOST(x->rdmult, x->rddiv, s1 + r_tx_size, sse[n]);
      }
Jingning Han's avatar
Jingning Han committed
661
662
663
664
665
    } else {
      rd[n][0] = RDCOST(x->rdmult, x->rddiv, r[n][0] + s0, d[n]);
      rd[n][1] = RDCOST(x->rdmult, x->rddiv, r[n][1] + s0, d[n]);
    }

666
667
668
669
670
    if (is_inter_block(mbmi) && !xd->lossless && !s[n] && sse[n] != INT64_MAX) {
      rd[n][0] = VPXMIN(rd[n][0], RDCOST(x->rdmult, x->rddiv, s1, sse[n]));
      rd[n][1] = VPXMIN(rd[n][1], RDCOST(x->rdmult, x->rddiv, s1, sse[n]));
    }

Jingning Han's avatar
Jingning Han committed
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
    // Early termination in transform size search.
    if (cpi->sf.tx_size_search_breakout &&
        (rd[n][1] == INT64_MAX ||
        (n < (int) max_tx_size && rd[n][1] > rd[n + 1][1]) ||
        s[n] == 1))
      break;

    if (rd[n][1] < best_rd) {
      best_tx = n;
      best_rd = rd[n][1];
    }
  }
  mbmi->tx_size = best_tx;

  *distortion = d[mbmi->tx_size];
  *rate       = r[mbmi->tx_size][cm->tx_mode == TX_MODE_SELECT];
  *skip       = s[mbmi->tx_size];
  *psse       = sse[mbmi->tx_size];
}

Yaowu Xu's avatar
Yaowu Xu committed
691
static void super_block_yrd(VP10_COMP *cpi, MACROBLOCK *x, int *rate,
Jingning Han's avatar
Jingning Han committed
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
                            int64_t *distortion, int *skip,
                            int64_t *psse, BLOCK_SIZE bs,
                            int64_t ref_best_rd) {
  MACROBLOCKD *xd = &x->e_mbd;
  int64_t sse;
  int64_t *ret_sse = psse ? psse : &sse;

  assert(bs == xd->mi[0]->mbmi.sb_type);

  if (cpi->sf.tx_size_search_method == USE_LARGESTALL || xd->lossless) {
    choose_largest_tx_size(cpi, x, rate, distortion, skip, ret_sse, ref_best_rd,
                           bs);
  } else {
    choose_tx_size_from_rd(cpi, x, rate, distortion, skip, ret_sse,
                           ref_best_rd, bs);
  }
}

static int conditional_skipintra(PREDICTION_MODE mode,
                                 PREDICTION_MODE best_intra_mode) {
  if (mode == D117_PRED &&
      best_intra_mode != V_PRED &&
      best_intra_mode != D135_PRED)
    return 1;
  if (mode == D63_PRED &&
      best_intra_mode != V_PRED &&
      best_intra_mode != D45_PRED)
    return 1;
  if (mode == D207_PRED &&
      best_intra_mode != H_PRED &&
      best_intra_mode != D45_PRED)
    return 1;
  if (mode == D153_PRED &&
      best_intra_mode != H_PRED &&
      best_intra_mode != D135_PRED)
    return 1;
  return 0;
}

Yaowu Xu's avatar
Yaowu Xu committed
731
static int64_t rd_pick_intra4x4block(VP10_COMP *cpi, MACROBLOCK *x,
Jingning Han's avatar
Jingning Han committed
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
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
791
792
                                     int row, int col,
                                     PREDICTION_MODE *best_mode,
                                     const int *bmode_costs,
                                     ENTROPY_CONTEXT *a, ENTROPY_CONTEXT *l,
                                     int *bestrate, int *bestratey,
                                     int64_t *bestdistortion,
                                     BLOCK_SIZE bsize, int64_t rd_thresh) {
  PREDICTION_MODE mode;
  MACROBLOCKD *const xd = &x->e_mbd;
  int64_t best_rd = rd_thresh;
  struct macroblock_plane *p = &x->plane[0];
  struct macroblockd_plane *pd = &xd->plane[0];
  const int src_stride = p->src.stride;
  const int dst_stride = pd->dst.stride;
  const uint8_t *src_init = &p->src.buf[row * 4 * src_stride + col * 4];
  uint8_t *dst_init = &pd->dst.buf[row * 4 * src_stride + col * 4];
  ENTROPY_CONTEXT ta[2], tempa[2];
  ENTROPY_CONTEXT tl[2], templ[2];
  const int num_4x4_blocks_wide = num_4x4_blocks_wide_lookup[bsize];
  const int num_4x4_blocks_high = num_4x4_blocks_high_lookup[bsize];
  int idx, idy;
  uint8_t best_dst[8 * 8];
#if CONFIG_VP9_HIGHBITDEPTH
  uint16_t best_dst16[8 * 8];
#endif

  memcpy(ta, a, sizeof(ta));
  memcpy(tl, l, sizeof(tl));
  xd->mi[0]->mbmi.tx_size = TX_4X4;

#if CONFIG_VP9_HIGHBITDEPTH
  if (xd->cur_buf->flags & YV12_FLAG_HIGHBITDEPTH) {
    for (mode = DC_PRED; mode <= TM_PRED; ++mode) {
      int64_t this_rd;
      int ratey = 0;
      int64_t distortion = 0;
      int rate = bmode_costs[mode];

      if (!(cpi->sf.intra_y_mode_mask[TX_4X4] & (1 << mode)))
        continue;

      // Only do the oblique modes if the best so far is
      // one of the neighboring directional modes
      if (cpi->sf.mode_search_skip_flags & FLAG_SKIP_INTRA_DIRMISMATCH) {
        if (conditional_skipintra(mode, *best_mode))
            continue;
      }

      memcpy(tempa, ta, sizeof(ta));
      memcpy(templ, tl, sizeof(tl));

      for (idy = 0; idy < num_4x4_blocks_high; ++idy) {
        for (idx = 0; idx < num_4x4_blocks_wide; ++idx) {
          const int block = (row + idy) * 2 + (col + idx);
          const uint8_t *const src = &src_init[idx * 4 + idy * 4 * src_stride];
          uint8_t *const dst = &dst_init[idx * 4 + idy * 4 * dst_stride];
          int16_t *const src_diff = vp10_raster_block_offset_int16(BLOCK_8X8,
                                                                  block,
                                                                  p->src_diff);
          tran_low_t *const coeff = BLOCK_OFFSET(x->plane[0].coeff, block);
          xd->mi[0]->bmi[block].as_mode = mode;
Jingning Han's avatar
Jingning Han committed
793
          vp10_predict_intra_block(xd, 1, TX_4X4, mode, dst, dst_stride,
Jingning Han's avatar
Jingning Han committed
794
795
796
797
798
                                  dst, dst_stride,
                                  col + idx, row + idy, 0);
          vpx_highbd_subtract_block(4, 4, src_diff, 8, src, src_stride,
                                    dst, dst_stride, xd->bd);
          if (xd->lossless) {
hui su's avatar
hui su committed
799
800
            TX_TYPE tx_type = get_tx_type(PLANE_TYPE_Y, xd, block);
            const scan_order *so = get_scan(TX_4X4, tx_type);
801
            vp10_highbd_fwd_txfm_4x4(src_diff, coeff, 8, DCT_DCT, 1);
Jingning Han's avatar
Jingning Han committed
802
803
804
805
806
807
            vp10_regular_quantize_b_4x4(x, 0, block, so->scan, so->iscan);
            ratey += cost_coeffs(x, 0, block, tempa + idx, templ + idy, TX_4X4,
                                 so->scan, so->neighbors,
                                 cpi->sf.use_fast_coef_costing);
            if (RDCOST(x->rdmult, x->rddiv, ratey, distortion) >= best_rd)
              goto next_highbd;
hui su's avatar
hui su committed
808
809
            vp10_highbd_inv_txfm_add_4x4(BLOCK_OFFSET(pd->dqcoeff, block),
                                         dst, dst_stride, p->eobs[block],
810
                                         xd->bd, DCT_DCT, 1);
Jingning Han's avatar
Jingning Han committed
811
812
          } else {
            int64_t unused;
hui su's avatar
hui su committed
813
814
            TX_TYPE tx_type = get_tx_type(PLANE_TYPE_Y, xd, block);
            const scan_order *so = get_scan(TX_4X4, tx_type);
815
            vp10_highbd_fwd_txfm_4x4(src_diff, coeff, 8, tx_type, 0);
Jingning Han's avatar
Jingning Han committed
816
817
818
819
820
821
822
823
824
            vp10_regular_quantize_b_4x4(x, 0, block, so->scan, so->iscan);
            ratey += cost_coeffs(x, 0, block, tempa + idx, templ + idy, TX_4X4,
                                 so->scan, so->neighbors,
                                 cpi->sf.use_fast_coef_costing);
            distortion += vp10_highbd_block_error(
                coeff, BLOCK_OFFSET(pd->dqcoeff, block),
                16, &unused, xd->bd) >> 2;
            if (RDCOST(x->rdmult, x->rddiv, ratey, distortion) >= best_rd)
              goto next_highbd;
hui su's avatar
hui su committed
825
826
            vp10_highbd_inv_txfm_add_4x4(BLOCK_OFFSET(pd->dqcoeff, block),
                                         dst, dst_stride, p->eobs[block],
827
                                         xd->bd, tx_type, 0);
Jingning Han's avatar
Jingning Han committed
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
          }
        }
      }

      rate += ratey;
      this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);

      if (this_rd < best_rd) {
        *bestrate = rate;
        *bestratey = ratey;
        *bestdistortion = distortion;
        best_rd = this_rd;
        *best_mode = mode;
        memcpy(a, tempa, sizeof(tempa));
        memcpy(l, templ, sizeof(templ));
        for (idy = 0; idy < num_4x4_blocks_high * 4; ++idy) {
          memcpy(best_dst16 + idy * 8,
                 CONVERT_TO_SHORTPTR(dst_init + idy * dst_stride),
                 num_4x4_blocks_wide * 4 * sizeof(uint16_t));
        }
      }
    next_highbd:
      {}
    }
Jingning Han's avatar
Jingning Han committed
852
    if (best_rd >= rd_thresh)
Jingning Han's avatar
Jingning Han committed
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
      return best_rd;

    for (idy = 0; idy < num_4x4_blocks_high * 4; ++idy) {
      memcpy(CONVERT_TO_SHORTPTR(dst_init + idy * dst_stride),
             best_dst16 + idy * 8,
             num_4x4_blocks_wide * 4 * sizeof(uint16_t));
    }

    return best_rd;
  }
#endif  // CONFIG_VP9_HIGHBITDEPTH

  for (mode = DC_PRED; mode <= TM_PRED; ++mode) {
    int64_t this_rd;
    int ratey = 0;
    int64_t distortion = 0;
    int rate = bmode_costs[mode];

    if (!(cpi->sf.intra_y_mode_mask[TX_4X4] & (1 << mode)))
      continue;

    // Only do the oblique modes if the best so far is
    // one of the neighboring directional modes
    if (cpi->sf.mode_search_skip_flags & FLAG_SKIP_INTRA_DIRMISMATCH) {
      if (conditional_skipintra(mode, *best_mode))
          continue;
    }

    memcpy(tempa, ta, sizeof(ta));
    memcpy(templ, tl, sizeof(tl));

    for (idy = 0; idy < num_4x4_blocks_high; ++idy) {
      for (idx = 0; idx < num_4x4_blocks_wide; ++idx) {
        const int block = (row + idy) * 2 + (col + idx);
        const uint8_t *const src = &src_init[idx * 4 + idy * 4 * src_stride];
        uint8_t *const dst = &dst_init[idx * 4 + idy * 4 * dst_stride];
        int16_t *const src_diff =
            vp10_raster_block_offset_int16(BLOCK_8X8, block, p->src_diff);
        tran_low_t *const coeff = BLOCK_OFFSET(x->plane[0].coeff, block);
        xd->mi[0]->bmi[block].as_mode = mode;
893
        vp10_predict_intra_block(xd, 1, TX_4X4, mode, dst, dst_stride,
Jingning Han's avatar
Jingning Han committed
894
895
896
897
                                dst, dst_stride, col + idx, row + idy, 0);
        vpx_subtract_block(4, 4, src_diff, 8, src, src_stride, dst, dst_stride);

        if (xd->lossless) {
hui su's avatar
hui su committed
898
899
          TX_TYPE tx_type = get_tx_type(PLANE_TYPE_Y, xd, block);
          const scan_order *so = get_scan(TX_4X4, tx_type);
900
          vp10_fwd_txfm_4x4(src_diff, coeff, 8, DCT_DCT, 1);
Jingning Han's avatar
Jingning Han committed
901
902
903
904
905
906
          vp10_regular_quantize_b_4x4(x, 0, block, so->scan, so->iscan);
          ratey += cost_coeffs(x, 0, block, tempa + idx, templ + idy, TX_4X4,
                               so->scan, so->neighbors,
                               cpi->sf.use_fast_coef_costing);
          if (RDCOST(x->rdmult, x->rddiv, ratey, distortion) >= best_rd)
            goto next;
hui su's avatar
hui su committed
907
          vp10_inv_txfm_add_4x4(BLOCK_OFFSET(pd->dqcoeff, block),
908
                                dst, dst_stride, p->eobs[block], DCT_DCT, 1);
Jingning Han's avatar
Jingning Han committed
909
910
        } else {
          int64_t unused;
hui su's avatar
hui su committed
911
912
          TX_TYPE tx_type = get_tx_type(PLANE_TYPE_Y, xd, block);
          const scan_order *so = get_scan(TX_4X4, tx_type);
913
          vp10_fwd_txfm_4x4(src_diff, coeff, 8, tx_type, 0);
Jingning Han's avatar
Jingning Han committed
914
915
916
917
918
919
920
921
          vp10_regular_quantize_b_4x4(x, 0, block, so->scan, so->iscan);
          ratey += cost_coeffs(x, 0, block, tempa + idx, templ + idy, TX_4X4,
                             so->scan, so->neighbors,
                             cpi->sf.use_fast_coef_costing);
          distortion += vp10_block_error(coeff, BLOCK_OFFSET(pd->dqcoeff, block),
                                        16, &unused) >> 2;
          if (RDCOST(x->rdmult, x->rddiv, ratey, distortion) >= best_rd)
            goto next;
hui su's avatar
hui su committed
922
          vp10_inv_txfm_add_4x4(BLOCK_OFFSET(pd->dqcoeff, block),
923
                                dst, dst_stride, p->eobs[block], tx_type, 0);
Jingning Han's avatar
Jingning Han committed
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
        }
      }
    }

    rate += ratey;
    this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);

    if (this_rd < best_rd) {
      *bestrate = rate;
      *bestratey = ratey;
      *bestdistortion = distortion;
      best_rd = this_rd;
      *best_mode = mode;
      memcpy(a, tempa, sizeof(tempa));
      memcpy(l, templ, sizeof(templ));
      for (idy = 0; idy < num_4x4_blocks_high * 4; ++idy)
        memcpy(best_dst + idy * 8, dst_init + idy * dst_stride,
               num_4x4_blocks_wide * 4);
    }
  next:
    {}
  }

947
  if (best_rd >= rd_thresh)
Jingning Han's avatar
Jingning Han committed
948
949
950
951
952
953
954
955
956
    return best_rd;

  for (idy = 0; idy < num_4x4_blocks_high * 4; ++idy)
    memcpy(dst_init + idy * dst_stride, best_dst + idy * 8,
           num_4x4_blocks_wide * 4);

  return best_rd;
}

Yaowu Xu's avatar
Yaowu Xu committed
957
static int64_t rd_pick_intra_sub_8x8_y_mode(VP10_COMP *cpi, MACROBLOCK *mb,
Jingning Han's avatar
Jingning Han committed
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
                                            int *rate, int *rate_y,
                                            int64_t *distortion,
                                            int64_t best_rd) {
  int i, j;
  const MACROBLOCKD *const xd = &mb->e_mbd;
  MODE_INFO *const mic = xd->mi[0];
  const MODE_INFO *above_mi = xd->above_mi;
  const MODE_INFO *left_mi = xd->left_mi;
  const BLOCK_SIZE bsize = xd->mi[0]->mbmi.sb_type;
  const int num_4x4_blocks_wide = num_4x4_blocks_wide_lookup[bsize];
  const int num_4x4_blocks_high = num_4x4_blocks_high_lookup[bsize];
  int idx, idy;
  int cost = 0;
  int64_t total_distortion = 0;
  int tot_rate_y = 0;
  int64_t total_rd = 0;
  ENTROPY_CONTEXT t_above[4], t_left[4];
  const int *bmode_costs = cpi->mbmode_cost;

  memcpy(t_above, xd->plane[0].above_context, sizeof(t_above));
  memcpy(t_left, xd->plane[0].left_context, sizeof(t_left));

  // Pick modes for each sub-block (of size 4x4, 4x8, or 8x4) in an 8x8 block.
  for (idy = 0; idy < 2; idy += num_4x4_blocks_high) {
    for (idx = 0; idx < 2; idx += num_4x4_blocks_wide) {
      PREDICTION_MODE best_mode = DC_PRED;
      int r = INT_MAX, ry = INT_MAX;
      int64_t d = INT64_MAX, this_rd = INT64_MAX;
      i = idy * 2 + idx;
      if (cpi->common.frame_type == KEY_FRAME) {
        const PREDICTION_MODE A = vp10_above_block_mode(mic, above_mi, i);
        const PREDICTION_MODE L = vp10_left_block_mode(mic, left_mi, i);

        bmode_costs  = cpi->y_mode_costs[A][L];
      }

      this_rd = rd_pick_intra4x4block(cpi, mb, idy, idx, &best_mode,
                                      bmode_costs, t_above + idx, t_left + idy,
                                      &r, &ry, &d, bsize, best_rd - total_rd);
      if (this_rd >= best_rd - total_rd)
        return INT64_MAX;

      total_rd += this_rd;