/src/avm/av2/common/reconinter.h

Source
/*
 * Copyright (c) 2021, Alliance for Open Media. All rights reserved
 *
 * This source code is subject to the terms of the BSD 3-Clause Clear License
 * and the Alliance for Open Media Patent License 1.0. If the BSD 3-Clause Clear
 * License was not distributed with this source code in the LICENSE file, you
 * can obtain it at aomedia.org/license/software-license/bsd-3-c-c/.  If the
 * Alliance for Open Media Patent License 1.0 was not distributed with this
 * source code in the PATENTS file, you can obtain it at
 * aomedia.org/license/patent-license/.
 */

#ifndef AVM_AV2_COMMON_RECONINTER_H_
#define AVM_AV2_COMMON_RECONINTER_H_

#include "av2/common/av2_common_int.h"
#include "av2/common/convolve.h"
#include "av2/common/filter.h"
#include "av2/common/warped_motion.h"
#include "avm/avm_integer.h"
#include "av2/encoder/block.h"

// Work out how many pixels off the edge of a reference frame we're allowed
// to go when forming an inter prediction.
// The outermost row/col of each referernce frame is extended by
// (AVM_BORDER_IN_PIXELS >> subsampling) pixels, but we need to keep
// at least AVM_INTERP_EXTEND pixels within that to account for filtering.
//
// We have to break this up into two macros to keep both clang-format and
// tools/lint-hunks.py happy.
#define AVM_LEFT_TOP_MARGIN_PX(subsampling) \
  ((AVM_BORDER_IN_PIXELS >> subsampling) - AVM_INTERP_EXTEND)
#define AVM_LEFT_TOP_MARGIN_SCALED(subsampling) \
  (AVM_LEFT_TOP_MARGIN_PX(subsampling) << SCALE_SUBPEL_BITS)

#ifdef __cplusplus
extern "C" {
#endif

#define MAX_WEDGE_SIZE_LOG2 6  // 64x64
#define MAX_WEDGE_SIZE (1 << MAX_WEDGE_SIZE_LOG2)
#define MAX_WEDGE_SQUARE (MAX_WEDGE_SIZE * MAX_WEDGE_SIZE)

#define MAX_WEDGE_BOUNDARY_TYPES 2

#define WEDGE_WEIGHT_BITS 6

static const int wedge_angle_dist_2_index[WEDGE_ANGLES][NUM_WEDGE_DIST] = {
  { -1, 0, 1, 2 },     // WEDGE_0
  { 3, 4, 5, 6 },      // WEDGE_14
  { 7, 8, 9, 10 },     // WEDGE_27
  { 11, 12, 13, 14 },  // WEDGE_45
  { 15, 16, 17, 18 },  // WEDGE_63
  { -1, 19, 20, 21 },  // WEDGE_90
  { 22, 23, 24, 25 },  // WEDGE_117
  { 26, 27, 28, 29 },  // WEDGE_135
  { 30, 31, 32, 33 },  // WEDGE_153
  { 34, 35, 36, 37 },  // WEDGE_166
  { -1, 38, 39, 40 },  // WEDGE_180
  { -1, 41, 42, 43 },  // WEDGE_194
  { -1, 44, 45, 46 },  // WEDGE_207
  { -1, 47, 48, 49 },  // WEDGE_225
  { -1, 50, 51, 52 },  // WEDGE_243
  { -1, 53, 54, 55 },  // WEDGE_270
  { -1, 56, 57, 58 },  // WEDGE_297
  { -1, 59, 60, 61 },  // WEDGE_315
  { -1, 62, 63, 64 },  // WEDGE_333
  { -1, 65, 66, 67 },  // WEDGE_346
};

static const int wedge_index_2_angle[MAX_WEDGE_TYPES] = {
  WEDGE_0,   WEDGE_0,   WEDGE_0,               // WEDGE_0
  WEDGE_14,  WEDGE_14,  WEDGE_14,  WEDGE_14,   // WEDGE_14
  WEDGE_27,  WEDGE_27,  WEDGE_27,  WEDGE_27,   // WEDGE_27
  WEDGE_45,  WEDGE_45,  WEDGE_45,  WEDGE_45,   // WEDGE_45
  WEDGE_63,  WEDGE_63,  WEDGE_63,  WEDGE_63,   // WEDGE_63
  WEDGE_90,  WEDGE_90,  WEDGE_90,              // WEDGE_90
  WEDGE_117, WEDGE_117, WEDGE_117, WEDGE_117,  // WEDGE_117
  WEDGE_135, WEDGE_135, WEDGE_135, WEDGE_135,  // WEDGE_135
  WEDGE_153, WEDGE_153, WEDGE_153, WEDGE_153,  // WEDGE_153
  WEDGE_166, WEDGE_166, WEDGE_166, WEDGE_166,  // WEDGE_166
  WEDGE_180, WEDGE_180, WEDGE_180,             // WEDGE_180
  WEDGE_194, WEDGE_194, WEDGE_194,             // WEDGE_194
  WEDGE_207, WEDGE_207, WEDGE_207,             // WEDGE_207
  WEDGE_225, WEDGE_225, WEDGE_225,             // WEDGE_225
  WEDGE_243, WEDGE_243, WEDGE_243,             // WEDGE_243
  WEDGE_270, WEDGE_270, WEDGE_270,             // WEDGE_270
  WEDGE_297, WEDGE_297, WEDGE_297,             // WEDGE_297
  WEDGE_315, WEDGE_315, WEDGE_315,             // WEDGE_315
  WEDGE_333, WEDGE_333, WEDGE_333,             // WEDGE_333
  WEDGE_346, WEDGE_346, WEDGE_346              // WEDGE_346
};

static const int wedge_index_2_dist[MAX_WEDGE_TYPES] = {
  1, 2, 3,     // WEDGE_0
  0, 1, 2, 3,  // WEDGE_14
  0, 1, 2, 3,  // WEDGE_27
  0, 1, 2, 3,  // WEDGE_45
  0, 1, 2, 3,  // WEDGE_63
  1, 2, 3,     // WEDGE_90
  0, 1, 2, 3,  // WEDGE_117
  0, 1, 2, 3,  // WEDGE_135
  0, 1, 2, 3,  // WEDGE_153
  0, 1, 2, 3,  // WEDGE_166
  1, 2, 3,     // WEDGE_180
  1, 2, 3,     // WEDGE_194
  1, 2, 3,     // WEDGE_207
  1, 2, 3,     // WEDGE_225
  1, 2, 3,     // WEDGE_243
  1, 2, 3,     // WEDGE_270
  1, 2, 3,     // WEDGE_297
  1, 2, 3,     // WEDGE_315
  1, 2, 3,     // WEDGE_333
  1, 2, 3,     // WEDGE_346
};

// 3-tuple: {direction, x_offset, y_offset}
typedef struct {
  WedgeDirectionType direction;
  int x_offset;
  int y_offset;
} wedge_code_type;
typedef uint8_t
    *all_wedge_masks_type[MAX_WEDGE_TYPES][MAX_WEDGE_BOUNDARY_TYPES];
typedef uint8_t
    *wedge_decisions_type[MAX_WEDGE_TYPES][MAX_WEDGE_BOUNDARY_TYPES];

typedef struct {
  int wedge_types;
  const wedge_code_type *codebook;
  uint8_t *signflip;
  all_wedge_masks_type *all_masks;
  wedge_decisions_type *tmvp_mv_decisions;
} wedge_params_type;

extern const wedge_params_type av2_wedge_params_lookup[BLOCK_SIZES_ALL];

typedef struct SubpelParams {
  int xs;
  int ys;
  int subpel_x;
  int subpel_y;
  int x0;  // top left sample horizontal cood.
  int y0;  // top left sample vertical cood.
  int x1;  // x0 + bw
  int y1;  // y0 + bh
} SubpelParams;

#define REFINE_MV_MAX_OFFSET 0

#define EIGHT_TAPS_REF_TOP_BORDER (AVM_INTERP_EXTEND - 1 + REFINE_MV_MAX_OFFSET)
#define EIGHT_TAPS_REF_LEFT_BORDER \
  (AVM_INTERP_EXTEND - 1 + REFINE_MV_MAX_OFFSET)
#define EIGHT_TAPS_REF_RIGHT_BORDER (AVM_INTERP_EXTEND + REFINE_MV_MAX_OFFSET)
#define EIGHT_TAPS_REF_BOTTOM_BORDER (AVM_INTERP_EXTEND + REFINE_MV_MAX_OFFSET)

#define AVM_4TAPS_INTERP_EXTEND 2
#define FOUR_TAPS_REF_TOP_BORDER \
  (AVM_4TAPS_INTERP_EXTEND - 1 + REFINE_MV_MAX_OFFSET)
#define FOUR_TAPS_REF_LEFT_BORDER \
  (AVM_4TAPS_INTERP_EXTEND - 1 + REFINE_MV_MAX_OFFSET)
#define FOUR_TAPS_REF_RIGHT_BORDER \
  (AVM_4TAPS_INTERP_EXTEND + REFINE_MV_MAX_OFFSET)
#define FOUR_TAPS_REF_BOTTOM_BORDER \
  (AVM_4TAPS_INTERP_EXTEND + REFINE_MV_MAX_OFFSET)

#define REF_TOP_BORDER_WARP (AVM_INTERP_EXTEND - 1)
#define REF_LEFT_BORDER_WARP (AVM_INTERP_EXTEND - 1)
#define REF_RIGHT_BORDER_WARP (AVM_INTERP_EXTEND)
#define REF_BOTTOM_BORDER_WARP (AVM_INTERP_EXTEND)

typedef enum InterPredMode {
  TRANSLATION_PRED,
  WARP_PRED,
} InterPredMode;

typedef enum InterCompMode {
  UNIFORM_SINGLE,
  UNIFORM_COMP,
  MASK_COMP,
} InterCompMode;

typedef struct InterPredParams {
  InterPredMode mode;
  InterCompMode comp_mode;
  WarpedMotionParams warp_params;
  ConvolveParams conv_params;
  const InterpFilterParams *interp_filter_params[2];
  int block_width;
  int block_height;
  // In optical flow refinement, block_width and block_height will pass the
  // subblock size into av2_make_inter_predictor, while orig_block_width and
  // orig_block_height keep the original block size that is needed by
  // calc_subpel_params_func
  int orig_block_width;
  int orig_block_height;

  // In refinemV, the prediction is generated maximum 16x16 sub-block basis
  // original_pu_width and  original_pu_height represents the width and height
  // of the original block.
  int original_pu_width;
  int original_pu_height;

  int pix_row;
  int pix_col;
  struct buf_2d ref_frame_buf;
  int subsampling_x;
  int subsampling_y;
  const struct scale_factors *scale_factors;
  int bit_depth;
  INTERINTER_COMPOUND_DATA mask_comp;
  BLOCK_SIZE sb_type;
  int is_intrabc;
  int use_ref_padding;
  ReferenceArea *ref_area;

  int use_warp_bd_box;
  PadBlock *warp_bd_box;

  INTERINTER_COMPOUND_BORDER_DATA border_data;
} InterPredParams;

// Apply bilinear and bicubic interpolation for subpel gradient to avoid
// calls of build_one_inter_predictor function. Bicubic interpolation
// brings better quality but the speed results are neutral. As such, bilinear
// interpolation is used by default for a better trade-off between quality
// and complexity.
#define OPFL_BILINEAR_GRAD 0
#define OPFL_BICUBIC_GRAD 1

// Delta to use for computing gradients in bits, with 0 referring to
// integer-pel. The actual delta value used from the 1/8-pel original MVs
// is 2^(3 - SUBPEL_GRAD_DELTA_BITS). The max value of this macro is 3.
#define SUBPEL_GRAD_DELTA_BITS 2

// Bilinear and bicubic coefficients. Note that, at boundary, we apply
// coefficients that are doubled because spatial distance between the two
// interpolated pixels is halved. In other words, instead of computing
//   coeff * (v[delta] - v[-delta]) / (2 * delta),
// we are practically computing
//   coeff * (v[delta] - v[0]) / (2 * delta).
// Thus, coeff is doubled to get a better gradient quality.
#if OPFL_BILINEAR_GRAD
static const int bilinear_bits = 3;
static const int32_t coeffs_bilinear[4][2] = {
  { 8, 16 },  // delta = 1 (SUBPEL_GRAD_DELTA_BITS = 0)
  { 4, 8 },   // delta = 0.5 (SUBPEL_GRAD_DELTA_BITS = 1)
  { 2, 4 },   // delta = 0.25 (SUBPEL_GRAD_DELTA_BITS = 2)
  { 1, 2 },   // delta = 0.125 (SUBPEL_GRAD_DELTA_BITS = 3)
};
#endif

#if OPFL_BICUBIC_GRAD
static const int bicubic_bits = 7;
static const int32_t coeffs_bicubic[4][2][2] = {
  { { 128, 256 }, { 0, 0 } },    // delta = 1 (SUBPEL_GRAD_DELTA_BITS = 0)
  { { 80, 160 }, { -8, -16 } },  // delta = 0.5 (SUBPEL_GRAD_DELTA_BITS = 1)
  { { 42, 84 }, { -5, -10 } },   // delta = 0.25 (SUBPEL_GRAD_DELTA_BITS = 2)
  { { 21, 42 }, { -3, -6 } },    // delta = 0.125 (SUBPEL_GRAD_DELTA_BITS = 3)
};
#endif

void av2_init_inter_params(InterPredParams *inter_pred_params, int block_width,
                           int block_height, int pix_row, int pix_col,
                           int subsampling_x, int subsampling_y, int bit_depth,
                           int is_intrabc, const struct scale_factors *sf,
                           const struct buf_2d *ref_buf,
                           InterpFilter interp_filter);

// Get the step size and maximum coded index of the warp block
static INLINE void get_warp_model_steps(const MB_MODE_INFO *mbmi,
                                        int *step_size, int *max_coded_index) {
  int step_size_log2 = mbmi->warp_precision_idx == 0 ? 11 : 10;
  *step_size = 1 << step_size_log2;
  *max_coded_index = mbmi->warp_precision_idx == 0 ? 7 : 14;
}

// Get the default value of the six_param_flag
static INLINE int get_default_six_param_flag(const AV2_COMMON *const cm,
                                             const MB_MODE_INFO *mbmi) {
  return (cm->seq_params.enable_six_param_warp_delta &&

          mbmi->warp_ref_idx == 1)
             ? 1
             : 0;
}

static INLINE int allow_warp_inter_intra(const MB_MODE_INFO *mbmi) {
  return mbmi->mode == WARPMV &&
         is_interintra_allowed_bsize(mbmi->sb_type[PLANE_TYPE_Y]) &&
         is_interintra_allowed_ref(mbmi->ref_frame) && mbmi->bawp_flag[0] == 0;
}

// Check if the signaling of the warp delta parameters are allowed
static INLINE int allow_warp_parameter_signaling(const AV2_COMMON *const cm,
                                                 const MB_MODE_INFO *mbmi) {
  // Warp delta parameters are signalled in following two cases
  // Case0: sequence level enable_six_param_warp_delta is enabled AND
  // (mbmi->warp_ref_idx == 0), in this case 6-parameter delta is signaled.
  // Case1: (mbmi->warp_ref_idx == 1), in this case 4-parameter delta is
  // signaled.

  int allow_delta_for_this_warp_ref_idx =
      get_default_six_param_flag(cm, mbmi) ||  // 6-parameter
      (mbmi->warp_ref_idx == 0);               // 4-parameter

  return (mbmi->mode == WARP_NEWMV && cm->features.allow_warpmv_mode &&
          mbmi->motion_mode == WARP_DELTA && allow_delta_for_this_warp_ref_idx);
}

// Map the index to weighting factor for compound weighted prediction
static INLINE int get_cwp_coding_idx(int val, int encode,
                                     const AV2_COMMON *const cm,
                                     const MB_MODE_INFO *const mbmi) {
  int is_same_side = 0;
  int cur_ref_side = 0;
  int other_ref_side = 0;
  if (has_second_ref(mbmi)) {
    cur_ref_side = cm->ref_frame_side[mbmi->ref_frame[0]];
    other_ref_side = cm->ref_frame_side[mbmi->ref_frame[1]];

    is_same_side = (cur_ref_side > 0 && other_ref_side > 0) ||
                   (cur_ref_side == 0 && other_ref_side == 0);
  }

  if (encode) {
    for (int i = 0; i < MAX_CWP_NUM; i++) {
      if (cwp_weighting_factor[is_same_side][i] == val) return i;
    }
    return 0;
  } else {
    return cwp_weighting_factor[is_same_side][val];
  }
}

static INLINE int enable_adaptive_mvd_resolution(const AV2_COMMON *const cm,
                                                 const MB_MODE_INFO *mbmi) {
  const int mode = mbmi->mode;
  if (allow_amvd_mode(mode) == 0) return 0;
  if (cm->seq_params.enable_adaptive_mvd == 0) return 0;

  return mbmi->use_amvd;
}

// get the base reference frame list for joint MVD coding, the MVD for base
// reference frame is the same as the joint MVD, the MVD for the other reference
// frame is scaled from the joint MVD.
static INLINE int get_joint_mvd_base_ref_list(const AV2_COMMON *const cm,
                                              const MB_MODE_INFO *mbmi) {
  int base_ref_list = 0;
  int first_ref_dist = 0;
  int sec_ref_dist = 0;
  if (has_second_ref(mbmi)) {
    first_ref_dist = cm->ref_frame_relative_dist[mbmi->ref_frame[0]];
    sec_ref_dist = cm->ref_frame_relative_dist[mbmi->ref_frame[1]];

    if (first_ref_dist >= sec_ref_dist) {
      base_ref_list = 0;
    } else {
      base_ref_list = 1;
    }
  }
  return base_ref_list;
}
// check whether the direction of two reference frames are from same side
static INLINE int is_ref_frame_same_side(const AV2_COMMON *const cm,
                                         const MB_MODE_INFO *mbmi) {
  int is_same_side = 0;
  int cur_ref_side = 0;
  int other_ref_side = 0;
  if (has_second_ref(mbmi)) {
    cur_ref_side = cm->ref_frame_side[mbmi->ref_frame[0]];
    other_ref_side = cm->ref_frame_side[mbmi->ref_frame[1]];

    is_same_side = (cur_ref_side > 0 && other_ref_side > 0) ||
                   (cur_ref_side == 0 && other_ref_side == 0);
  }
  return is_same_side;
}

// check whether the distance of to the two reference frames are the same
static INLINE int is_ref_frame_same_dist(const AV2_COMMON *const cm,
                                         const MB_MODE_INFO *mbmi) {
  int is_same_dist = 0;

  if (has_second_ref(mbmi)) {
    const int first_ref_dist = cm->ref_frame_relative_dist[mbmi->ref_frame[0]];
    const int sec_ref_dist = cm->ref_frame_relative_dist[mbmi->ref_frame[1]];

    is_same_dist = (first_ref_dist == sec_ref_dist);
  }

  return is_same_dist;
}

// obtain context for inter_compound_mode_is_joint
static INLINE int get_inter_compound_mode_is_joint_context(
    const AV2_COMMON *const cm, const MB_MODE_INFO *mbmi) {
  return (is_ref_frame_same_side(cm, mbmi) ||
          (!is_ref_frame_same_dist(cm, mbmi)));
}

void av2_init_comp_mode(InterPredParams *inter_pred_params);

void av2_init_warp_params(InterPredParams *inter_pred_params,
                          const WarpTypesAllowed *warp_types, int ref,
                          const MACROBLOCKD *xd, const MB_MODE_INFO *mi);

static INLINE int has_scale(int xs, int ys) {
  return xs != SCALE_SUBPEL_SHIFTS || ys != SCALE_SUBPEL_SHIFTS;
}

static INLINE void revert_scale_extra_bits(SubpelParams *sp) {
  sp->subpel_x >>= SCALE_EXTRA_BITS;
  sp->subpel_y >>= SCALE_EXTRA_BITS;
  sp->xs >>= SCALE_EXTRA_BITS;
  sp->ys >>= SCALE_EXTRA_BITS;
  assert(sp->subpel_x < SUBPEL_SHIFTS);
  assert(sp->subpel_y < SUBPEL_SHIFTS);
  assert(sp->xs <= SUBPEL_SHIFTS);
  assert(sp->ys <= SUBPEL_SHIFTS);
}

static INLINE void highbd_inter_predictor(
    const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride,
    const SubpelParams *subpel_params, int w, int h,
    ConvolveParams *conv_params, const InterpFilterParams *interp_filters[2],
    int bd, int is_intrabc) {
  assert(conv_params->do_average == 0 || conv_params->do_average == 1);
  const int is_scaled = has_scale(subpel_params->xs, subpel_params->ys);
  if (is_scaled) {
    assert(is_intrabc == 0);
    av2_highbd_convolve_2d_facade(
        src, src_stride, dst, dst_stride, w, h, interp_filters,
        subpel_params->subpel_x, subpel_params->xs, subpel_params->subpel_y,
        subpel_params->ys, 1, conv_params, bd, is_intrabc);
  } else {
    SubpelParams sp = *subpel_params;
    revert_scale_extra_bits(&sp);
    av2_highbd_convolve_2d_facade(
        src, src_stride, dst, dst_stride, w, h, interp_filters, sp.subpel_x,
        sp.xs, sp.subpel_y, sp.ys, 0, conv_params, bd, is_intrabc);
  }
}

static INLINE int is_interinter_compound_used(COMPOUND_TYPE type,
                                              BLOCK_SIZE sb_type) {
  const int comp_allowed = is_comp_ref_allowed(sb_type);
  switch (type) {
    case COMPOUND_AVERAGE: return comp_allowed;
    case COMPOUND_DIFFWTD:
      return comp_allowed && !is_thin_4xn_nx4_block(sb_type);
    case COMPOUND_WEDGE:
      return comp_allowed && av2_wedge_params_lookup[sb_type].wedge_types > 0;
    default: assert(0); return 0;
  }
}

static INLINE int is_any_masked_compound_used(BLOCK_SIZE sb_type) {
  COMPOUND_TYPE comp_type;
  int i;
  if (!is_comp_ref_allowed(sb_type)) return 0;
  for (i = 0; i < COMPOUND_TYPES; i++) {
    comp_type = (COMPOUND_TYPE)i;
    if (is_masked_compound_type(comp_type) &&
        is_interinter_compound_used(comp_type, sb_type))
      return 1;
  }
  return 0;
}

static INLINE int get_wedge_types_lookup(BLOCK_SIZE sb_type) {
  return av2_wedge_params_lookup[sb_type].wedge_types;
}

static INLINE int av2_is_wedge_used(BLOCK_SIZE sb_type) {
  return av2_wedge_params_lookup[sb_type].wedge_types > 0;
}

void av2_make_inter_predictor(const uint16_t *src, int src_stride,
                              uint16_t *dst, int dst_stride,
                              InterPredParams *inter_pred_params,
                              const SubpelParams *subpel_params);

typedef void (*CalcSubpelParamsFunc)(const MV *const src_mv,
                                     InterPredParams *const inter_pred_params,
                                     MACROBLOCKD *xd, int mi_x, int mi_y,
                                     int ref, int use_optflow_refinement,
                                     uint16_t **mc_buf, uint16_t **pre,
                                     SubpelParams *subpel_params,
                                     int *src_stride);

void av2_build_one_inter_predictor(
    uint16_t *dst, int dst_stride, const MV *const src_mv,
    InterPredParams *inter_pred_params, MACROBLOCKD *xd, int mi_x, int mi_y,
    int ref, uint16_t **mc_buf, CalcSubpelParamsFunc calc_subpel_params_func);

void av2_build_inter_predictors(const AV2_COMMON *cm, MACROBLOCKD *xd,
                                int plane, MB_MODE_INFO *mi,
                                const BUFFER_SET *dst_orig,
                                int build_for_refine_mv_only,
                                int build_for_decode, int bw, int bh, int mi_x,
                                int mi_y, uint16_t **mc_buf,
                                CalcSubpelParamsFunc calc_subpel_params_func);

// Precision of refined MV returned, 0 being integer pel. For now, only 1/8 or
// 1/16-pel can be used.
#define MV_REFINE_PREC_BITS 4  // (1/16-pel)

#define OPFL_PRED_MAX ((1 << 10) - 1)

// Apply regularized least squares (RLS). The RLS parameter is bw * bh * 2^(b-4)
// where b = OPFL_RLS_PARAM.
#define OPFL_REGULARIZED_LS 1
#define OPFL_RLS_PARAM 16

// Number of bits allowed for all intermediate results of covariance matrix
// filling
#define MAX_OPFL_AUTOCORR_BITS 24
// Clamp range for u/v/w. If it uses h unsigned bits, then u2/v2 uses 2h
// unsigned bits. Every sum of 8 u2/v2 use at most 2h+3 unsigned bits, and
// must not exceed max bd of su2/sv2 minus 2. Thus, 2h+3 <= H-2
#define OPFL_GRAD_CLAMP_VAL ((1 << ((MAX_OPFL_AUTOCORR_BITS - 6) >> 1)) - 1)

void reduce_temporal_dist(int *d0, int *d1);

void av2_opfl_build_inter_predictor(
    const AV2_COMMON *cm, MACROBLOCKD *xd, int plane, const MB_MODE_INFO *mi,
    int bw, int bh, int mi_x, int mi_y, uint16_t **mc_buf,
    InterPredParams *inter_pred_params,
    CalcSubpelParamsFunc calc_subpel_params_func, int ref, uint16_t *pred_dst,
    const MV *const src_mv, int pu_width, int pu_height);

// Generate refined MVs using optflow refinement
void av2_get_optflow_based_mv(const AV2_COMMON *cm, MACROBLOCKD *xd, int plane,
                              const MB_MODE_INFO *mbmi, int_mv *mv_refined,
                              int bw, int bh, int mi_x, int mi_y,
                              int build_for_decode, uint16_t **mc_buf,
                              CalcSubpelParamsFunc calc_subpel_params_func,
                              int16_t *gx0, int16_t *gy0, int16_t *gx1,
                              int16_t *gy1, int *vx0, int *vy0, int *vx1,
                              int *vy1, uint16_t *dst0, uint16_t *dst1,
                              int dst_stride, int do_pred, int use_4x4,
                              MV *best_mv_ref, int pu_width, int pu_height);

// With the refined MVs, generate the inter prediction for the block.
void av2_opfl_rebuild_inter_predictor(
    uint16_t *dst, int dst_stride, int plane, int_mv *const mv_refined,
    InterPredParams *inter_pred_params, MACROBLOCKD *xd, int mi_x, int mi_y,
    int build_for_decode, const AV2_COMMON *cm, int pu_width, int ref,
    uint16_t **mc_buf, CalcSubpelParamsFunc calc_subpel_params_func,
    int use_4x4, MB_MODE_INFO *mi, int pu_height, const MV mi_mv[2],
    int use_sub_pad);

// We consider this tunable number K=MAX_LS_BITS-1 (sign bit excluded)
// as the target maximum bit depth of all intermediate results for LS problem.
#define MAX_LS_BITS 26

// Divide all elements of a vector by a common factor, and apply shifts.
// The integer division is based on lookup table.
// sol: numerator (will be updated to the solution)
// den: denominator
// out: output result (sol / den)
static INLINE void divide_and_round_array(int32_t *sol, int32_t den,
                                          const int dim, int *shifts) {
  assert(den != 0);
  if (den < 0) {
    for (int i = 0; i < dim; i++) sol[i] = -sol[i];
    divide_and_round_array(sol, -den, dim, shifts);
    return;
  }
  int16_t den_shift = 0;
  int16_t inv_den =
      (den == 1) ? 1 : resolve_divisor_32((uint32_t)den, &den_shift);
  int inv_den_msb = get_msb_signed(inv_den);

  // Apply shifts to sol[i] and den to keep both bit depths within K.
  for (int i = 0; i < dim; i++) {
    if (sol[i] == 0) continue;
    int sign = sol[i] > 0;
    sol[i] = sign ? sol[i] : -sol[i];
    int num_red_bits =
        AVMMAX(0, get_msb_signed(sol[i]) + inv_den_msb + 4 - MAX_LS_BITS);
    if (num_red_bits > 0) sol[i] = ROUND_POWER_OF_TWO(sol[i], num_red_bits);

    int inc_bits = shifts[i] + num_red_bits - den_shift;
    if (inc_bits >= 0)
      sol[i] = sol[i] * inv_den * (1 << inc_bits);
    else if (-inc_bits >= 31)  // ROUND_POWER_OF_TWO can only handle n<=30
      sol[i] = ROUND_POWER_OF_TWO(
          ROUND_POWER_OF_TWO(sol[i], -inc_bits - 30) * inv_den, 30);
    else
      sol[i] = ROUND_POWER_OF_TWO(sol[i] * inv_den, -inc_bits);
    sol[i] = sign ? sol[i] : -sol[i];
  }
}

static INLINE int is_translational_refinement_allowed(const AV2_COMMON *cm,
                                                      BLOCK_SIZE bsize,
                                                      const MACROBLOCKD *xd,
                                                      const int mode) {
  assert(cm->seq_params.enable_opfl_refine);
  (void)cm;
  if (mode < NEAR_NEARMV_OPTFLOW) return 0;
  if (is_thin_4xn_nx4_block(bsize)) return 0;
  (void)xd;
  return 1;
}

void highbd_build_mc_border(const uint16_t *src, int src_stride, uint16_t *dst,
                            int dst_stride, int x, int y, int b_w, int b_h,
                            int w, int h);

// Compute the SAD between the two predictors when refinemv is ON
int get_refinemv_sad(uint16_t *src1, uint16_t *src2, int stride, int width,
                     int height, int bd);
// Generate two prediction signals of a given mv0 and mv1
void av2_refinemv_build_predictors(MACROBLOCKD *xd, int mi_x, int mi_y,
                                   uint16_t **mc_buf,
                                   CalcSubpelParamsFunc calc_subpel_params_func,
                                   uint16_t *dst_ref0, uint16_t *dst_ref1,
                                   int dst_stride, MV mv0, MV mv1,
                                   InterPredParams *inter_pred_params);

// Get the context index to code refinemv flag
int av2_get_refinemv_context(const AV2_COMMON *cm, const MACROBLOCKD *xd,
                             BLOCK_SIZE bsize);

// Full blocks refine MVs are stored in 4x4 grid so that the MVs can be reused
// for chroma
void fill_subblock_refine_mv(REFINEMV_SUBMB_INFO *refinemv_subinfo, int bw,
                             int bh, MV mv0, MV mv1);

void av2_get_reference_area_with_padding_single(
    const AV2_COMMON *cm, MACROBLOCKD *xd, int plane, const MB_MODE_INFO *mi,
    const MV mv, int bw, int bh, int mi_x, int mi_y, PadBlock *ref_area,
    int pu_width, int pu_height, int ref, int is_warp);

// Generate the reference area (bounding box) based on the signaled MV
void av2_get_reference_area_with_padding(const AV2_COMMON *cm, MACROBLOCKD *xd,
                                         int plane, MB_MODE_INFO *mi,
                                         const MV mv[2], int bw, int bh,
                                         int mi_x, int mi_y,
                                         ReferenceArea ref_area[2],
                                         int pu_width, int pu_height);

// Derive the sub-pixel related parameters of non-TIP blocks
// Sub-pel related parameters are stored in the structures pointed by
// "subpel_params" and "block"
void dec_calc_subpel_params(const MV *const src_mv,
                            InterPredParams *const inter_pred_params,
                            const MACROBLOCKD *const xd, int mi_x, int mi_y,
                            uint16_t **pre, SubpelParams *subpel_params,
                            int *src_stride, PadBlock *block,
                            int use_optflow_refinement, MV32 *scaled_mv,
                            int *subpel_x_mv, int *subpel_y_mv);

// check if the refinemv mode is allwed for a given blocksize
static INLINE int is_refinemv_allowed_bsize(BLOCK_SIZE bsize) {
  assert(bsize < BLOCK_SIZES_ALL);
  if (AVMMIN(block_size_wide[bsize], block_size_high[bsize]) < 8) return 0;

  return (block_size_wide[bsize] >= 16 || block_size_high[bsize] >= 16);
}

// check if the refinemv mode is allwed for a given mode and precision
static INLINE int is_refinemv_allowed_mode_precision(
    PREDICTION_MODE mode, MvSubpelPrecision precision,
    const AV2_COMMON *const cm) {
  (void)precision;
  if (mode == GLOBAL_GLOBALMV) return 0;

  if (cm->features.opfl_refine_type == REFINE_SWITCHABLE &&
      (mode == JOINT_NEWMV || mode == NEAR_NEWMV || mode == NEW_NEARMV ||
       mode == NEW_NEWMV))
    return 0;

  return (mode >= NEAR_NEARMV);
}

// check if the prediction mode infered to refimemv to always 1.
static INLINE int default_refinemv_modes(const MB_MODE_INFO *mbmi) {
  return (mbmi->mode == NEAR_NEARMV || mbmi->mode == NEAR_NEARMV_OPTFLOW ||
          mbmi->mode == JOINT_NEWMV_OPTFLOW);
}
// Check if the compound and equal distance references
static INLINE int is_refinemv_allowed_reference(const AV2_COMMON *cm,
                                                const MB_MODE_INFO *mbmi) {
  if (!cm->seq_params.enable_refinemv) return 0;
  const unsigned int cur_index = cm->cur_frame->display_order_hint;
  int d0, d1;
  int is_tip = (mbmi->ref_frame[0] == TIP_FRAME);
  // In error resilient mode, whether a frame is scaled may be unknown, and the
  // following logic for parsing is not reliable, so SMVR is disabled.
  if (frame_is_sframe(cm)) return 0;

  if ((get_ref_frame_buf(cm, mbmi->ref_frame[0]) != NULL &&
       get_ref_frame_buf(cm, mbmi->ref_frame[0])->is_restricted) ||
      (get_ref_frame_buf(cm, mbmi->ref_frame[1]) != NULL &&
       get_ref_frame_buf(cm, mbmi->ref_frame[1])->is_restricted))
    return 0;

  // If one of the reference frame is different resolution than the current
  // frame, refinemv is disabled.
  const struct scale_factors *const sf0 =
      is_tip ? cm->tip_ref.ref_scale_factor[0]
             : get_ref_scale_factors_const(cm, mbmi->ref_frame[0]);
  const struct scale_factors *const sf1 =
      is_tip ? cm->tip_ref.ref_scale_factor[1]
             : get_ref_scale_factors_const(cm, mbmi->ref_frame[1]);
  if (av2_is_scaled(sf0) || av2_is_scaled(sf1)) {
    return 0;
  }

  if (is_tip) {
    d0 = cm->tip_ref.ref_offset[0];
    d1 = cm->tip_ref.ref_offset[1];
  } else {
    if (!has_second_ref(mbmi)) return 0;
    const RefCntBuffer *const ref0 = get_ref_frame_buf(cm, mbmi->ref_frame[0]);
    const RefCntBuffer *const ref1 = get_ref_frame_buf(cm, mbmi->ref_frame[1]);
    d0 = get_relative_dist(&cm->seq_params.order_hint_info, cur_index,
                           ref0->display_order_hint);
    d1 = get_relative_dist(&cm->seq_params.order_hint_info, cur_index,
                           ref1->display_order_hint);
  }

  // reference frame has to be both sides to apply refinemv
  if (!((d0 <= 0) ^ (d1 <= 0))) return 0;

  // Current implementation only supports when both has the same distance
  if (abs(d0) != abs(d1)) return 0;

  return 1;
}

// check if the refinemv mode is allowed for a given block
static INLINE int is_refinemv_allowed(const AV2_COMMON *const cm,
                                      const MB_MODE_INFO *mbmi,
                                      BLOCK_SIZE bsize) {
  if (!cm->seq_params.enable_refinemv) return 0;
  int is_tip = is_tip_ref_frame(mbmi->ref_frame[0]);
  if (is_tip) return 0;
  assert(!mbmi->skip_mode);
  int is_compound = has_second_ref(mbmi);
  return mbmi->motion_mode == SIMPLE_TRANSLATION && is_compound &&
         is_refinemv_allowed_bsize(bsize) &&
         is_refinemv_allowed_mode_precision(mbmi->mode, mbmi->pb_mv_precision,
                                            cm) &&
         is_refinemv_allowed_reference(cm, mbmi);
}

// check if any mv refinement mode is allowed in TIP at frame level
static INLINE int is_any_mv_refinement_allowed_in_tip(
    const AV2_COMMON *const cm) {
  if (!cm->has_both_sides_refs) return 0;

  if (cm->features.opfl_refine_type == REFINE_NONE &&
      !cm->seq_params.enable_refinemv)
    return 0;

  if (!cm->seq_params.enable_tip_refinemv) return 0;

  const int tip_wtd_index = cm->tip_global_wtd_index;
  const int8_t tip_weight = tip_weighting_factors[tip_wtd_index];
  if (tip_weight != TIP_EQUAL_WTD) return 0;

  return 1;
}

// check if any refinement algorithm is applied in TIP at block level
static INLINE int is_tip_block_with_mv_refinement(const AV2_COMMON *const cm,
                                                  const MACROBLOCKD *xd) {
  if (!is_any_mv_refinement_allowed_in_tip(cm)) return 0;

  if (cm->features.tip_frame_mode == TIP_FRAME_AS_OUTPUT) {
    // No OPFL and subblock MV refinement on TIP direct output mode when
    // the interpolation filter is not MULTITAP_SHARP
    if (cm->tip_interp_filter != MULTITAP_SHARP) return 0;

    // No subblock MV refinement on TIP direct output mode when the
    // interpolation filter is MULTITAP_SHARP
    if (cm->seq_params.enable_opfl_refine == AVM_OPFL_REFINE_NONE) return 0;
  }

  if (cm->features.tip_frame_mode == TIP_FRAME_AS_REF) {
    // No OPFL MV refinement on TIP reference mode when the coding block
    // is 256x256
    const MB_MODE_INFO *mi = xd->mi[0];

    const int bw = block_size_wide[mi->sb_type[xd->tree_type == CHROMA_PART]];
    const int bh = block_size_high[mi->sb_type[xd->tree_type == CHROMA_PART]];
    if (bw >= 256 && bh >= 256 && !cm->seq_params.enable_refinemv) return 0;
  }

  return 1;
}

// Is the coding block a TIP 16x16 coding block.
static AVM_INLINE bool is_tip_coded_as_16x16_block(const AV2_COMMON *cm,
                                                   const MB_MODE_INFO *mi) {
  if (!is_tip_ref_frame(mi->ref_frame[0])) return false;

  const int bw = block_size_wide[mi->sb_type[0]];
  const int bh = block_size_high[mi->sb_type[0]];

  bool is_tip_16_16 = disable_opfl_for_16x16_tip_ref(
      cm->features.tip_frame_mode, bw, bh, cm->seq_params.enable_tip_refinemv);
  is_tip_16_16 |= disable_opfl_for_tip_direct(
      cm->features.tip_frame_mode, cm->tip_interp_filter,
      cm->seq_params.enable_tip_refinemv);
  return is_tip_16_16;
}

// check if unequal weight is allowed for TIP at frame level
static INLINE int is_unequal_weighted_tip_allowed(const AV2_COMMON *const cm) {
  if (!cm->has_both_sides_refs) return 1;

  if (!cm->seq_params.enable_tip_refinemv) return 1;

  if (cm->features.opfl_refine_type == REFINE_NONE &&
      !cm->seq_params.enable_refinemv)
    return 1;

  return 0;
}

// set the weighting factor of TIP frame
static INLINE void set_tip_interp_weight_factor(
    const AV2_COMMON *const cm, const int ref_index,
    InterPredParams *const inter_pred_params) {
  if (ref_index == 1 && inter_pred_params->conv_params.do_average == 1 &&
      !is_any_mv_refinement_allowed_in_tip(cm)) {
    const int tip_wtd_index = cm->tip_global_wtd_index;
    const int8_t tip_weight = tip_weighting_factors[tip_wtd_index];
    assert(tip_wtd_index >= 0 && tip_wtd_index < MAX_TIP_WTD_NUM);
    inter_pred_params->conv_params.fwd_offset = tip_weight;
    inter_pred_params->conv_params.bck_offset =
        (1 << DIST_PRECISION_BITS) - tip_weight;
  }
}

// check if the refinemv mode is allowed for a given block for TIP mode
static INLINE int is_refinemv_allowed_tip_blocks(const AV2_COMMON *const cm,
                                                 const MB_MODE_INFO *mbmi) {
  assert(is_tip_ref_frame(mbmi->ref_frame[0]));
  return cm->seq_params.enable_refinemv && cm->seq_params.enable_tip_refinemv &&
         is_refinemv_allowed_reference(cm, mbmi) &&
         (cm->features.tip_frame_mode != TIP_FRAME_AS_OUTPUT);
}

// check if the refinemv mode is allowed for a given block for skip mode
static INLINE int is_refinemv_allowed_skip_mode(const AV2_COMMON *const cm,
                                                const MB_MODE_INFO *mbmi) {
  (void)cm;
  (void)mbmi;
  return 0;
}
static INLINE int get_default_refinemv_flag(const AV2_COMMON *const cm,
                                            const MB_MODE_INFO *mbmi) {
  if (!cm->seq_params.enable_refinemv) return 0;
  int is_refinemv =
      (mbmi->skip_mode
           ? is_refinemv_allowed_skip_mode(cm, mbmi)
           : is_refinemv_allowed(cm, mbmi, mbmi->sb_type[PLANE_TYPE_Y]));
  if (is_refinemv) {
    if (default_refinemv_modes(mbmi)) return 1;
  }
  return 0;
}

// check if the refinemv mode is switchable for a given block
static INLINE int switchable_refinemv_flag(const AV2_COMMON *const cm,
                                           const MB_MODE_INFO *mbmi) {
  if (!cm->seq_params.enable_refinemv) return 0;
  int is_refinemv =
      (mbmi->skip_mode
           ? is_refinemv_allowed_skip_mode(cm, mbmi)
           : is_refinemv_allowed(cm, mbmi, mbmi->sb_type[PLANE_TYPE_Y]));
  if (is_refinemv && !is_tip_ref_frame(mbmi->ref_frame[0])) {
    if (default_refinemv_modes(mbmi)) return 0;
    return 1;
  }

  return 0;
}

// Check if refined MV needs to be stored in the TMVP list.
static INLINE int enable_refined_mvs_in_tmvp(const AV2_COMMON *cm,
                                             const MACROBLOCKD *xd,
                                             const MB_MODE_INFO *mbmi) {
  return (
      opfl_allowed_cur_pred_mode(cm, xd, mbmi) ||
      (mbmi->refinemv_flag && mbmi->interinter_comp.type == COMPOUND_AVERAGE) ||
      (is_tip_ref_frame(mbmi->ref_frame[0]) &&
       is_tip_block_with_mv_refinement(cm, xd)));
}

// This function conduct the SAD search between two predictors and find the best
// MVs
void apply_mv_refinement(const AV2_COMMON *cm, MACROBLOCKD *xd, int plane,
                         MB_MODE_INFO *mi, int bw, int bh, int mi_x, int mi_y,
                         uint16_t **mc_buf, const MV mv[2],
                         CalcSubpelParamsFunc calc_subpel_params_func,
                         int pre_x, int pre_y, uint16_t *dst_ref0,
                         uint16_t *dst_ref1, uint16_t **dst_ref0_ptr,
                         uint16_t **dst_ref1_ptr, MV *best_mv_ref, int pu_width,
                         int pu_height, ReferenceArea ref_area[2]);

// check if padding is required during motion compensation
// return 1 means reference pixel is outside of the reference range and padding
// is required return 0 means no padding.
int update_extend_mc_border_params(const struct scale_factors *const sf,
                                   struct buf_2d *const pre_buf, MV32 scaled_mv,
                                   PadBlock *block, int subpel_x_mv,
                                   int subpel_y_mv, int do_warp, int is_intrabc,
                                   int *x_pad, int *y_pad,
                                   const ReferenceArea *ref_area);

// Derive the sub-pixel related parameters of refinemv non-TIP blocks
// Sub-pel related parameters are stored in the structures pointed by
// "subpel_params" Also do padding if required This function is used for
// both encoder and decoder
void common_calc_subpel_params_and_extend(
    const MV *const src_mv, InterPredParams *const inter_pred_params,
    MACROBLOCKD *const xd, int mi_x, int mi_y, int ref,
    int use_optflow_refinement, uint16_t **mc_buf, uint16_t **pre,
    SubpelParams *subpel_params, int *src_stride);

#define MAKE_BFP_SAD_WRAPPER_COMMON(fnname)                                   \
  static unsigned int fnname##_8(const uint16_t *src_ptr, int source_stride,  \
                                 const uint16_t *ref_ptr, int ref_stride) {   \
    return fnname(src_ptr, source_stride, ref_ptr, ref_stride);               \
  }                                                                           \
  static unsigned int fnname##_10(const uint16_t *src_ptr, int source_stride, \
                                  const uint16_t *ref_ptr, int ref_stride) {  \
    return fnname(src_ptr, source_stride, ref_ptr, ref_stride) >> 2;          \
  }                                                                           \
  static unsigned int fnname##_12(const uint16_t *src_ptr, int source_stride, \
                                  const uint16_t *ref_ptr, int ref_stride) {  \
    return fnname(src_ptr, source_stride, ref_ptr, ref_stride) >> 4;          \
  }

unsigned int get_highbd_sad(const uint16_t *src_ptr, int source_stride,
                            const uint16_t *ref_ptr, int ref_stride, int bd,
                            int bw, int bh);

unsigned int get_highbd_sad_ds(const uint16_t *src_ptr, int source_stride,
                               const uint16_t *ref_ptr, int ref_stride, int bd,
                               int bw, int bh);

void calc_mv_process(int32_t su2, int32_t sv2, int32_t suv, int32_t suw,
                     int32_t svw, const int d0, const int d1, const int bits,
                     const int rls_alpha, int *vx0, int *vy0, int *vx1,
                     int *vy1);
void av2_opfl_mv_refinement(const int16_t *pdiff, int pstride0,
                            const int16_t *gx, const int16_t *gy, int gstride,
                            int bw, int bh, int d0, int d1, int grad_prec_bits,
                            int mv_prec_bits, int *vx0, int *vy0, int *vx1,
                            int *vy1);
void av2_compute_subpel_gradients_interp(int16_t *pred_dst, int bw, int bh,
                                         int *grad_prec_bits, int16_t *x_grad,
                                         int16_t *y_grad);

// TODO(jkoleszar): yet another mv clamping function :-(
static INLINE MV clamp_mv_to_umv_border_sb(const MACROBLOCKD *xd,
                                           const MV *src_mv, int bw, int bh,
                                           int use_optflow_refinement, int ss_x,
                                           int ss_y) {
  // If the MV points so far into the UMV border that no visible pixels
  // are used for reconstruction, the subpel part of the MV can be
  // discarded and the MV limited to 16 pixels with equivalent results.
  const int spel_left = (AVM_INTERP_EXTEND + bw) << SUBPEL_BITS;
  const int spel_right = spel_left - SUBPEL_SHIFTS;
  const int spel_top = (AVM_INTERP_EXTEND + bh) << SUBPEL_BITS;
  const int spel_bottom = spel_top - SUBPEL_SHIFTS;
  MV clamped_mv;
  if (use_optflow_refinement) {
    // optflow refinement always returns MVs with 1/16 precision so it is not
    // necessary to shift the MV before clamping
    clamped_mv.row = (MV_COMP_DATA_TYPE)ROUND_POWER_OF_TWO_SIGNED(
        src_mv->row * (1 << SUBPEL_BITS), MV_REFINE_PREC_BITS + ss_y);
    clamped_mv.col = (MV_COMP_DATA_TYPE)ROUND_POWER_OF_TWO_SIGNED(
        src_mv->col * (1 << SUBPEL_BITS), MV_REFINE_PREC_BITS + ss_x);
  } else {
    clamped_mv.row = (MV_COMP_DATA_TYPE)(src_mv->row * (1 << (1 - ss_y)));
    clamped_mv.col = (MV_COMP_DATA_TYPE)(src_mv->col * (1 << (1 - ss_x)));
  }
  assert(ss_x <= 1);
  assert(ss_y <= 1);
  const SubpelMvLimits mv_limits = {
    xd->mb_to_left_edge * (1 << (1 - ss_x)) - spel_left,
    xd->mb_to_right_edge * (1 << (1 - ss_x)) + spel_right,
    xd->mb_to_top_edge * (1 << (1 - ss_y)) - spel_top,
    xd->mb_to_bottom_edge * (1 << (1 - ss_y)) + spel_bottom
  };

  clamp_mv(&clamped_mv, &mv_limits);

  return clamped_mv;
}

void make_masked_inter_predictor(const uint16_t *pre, int pre_stride,
                                 uint16_t *dst, int dst_stride,
                                 InterPredParams *inter_pred_params,
                                 const SubpelParams *subpel_params,
                                 int use_bacp, int sub_block_id);

static INLINE int use_border_aware_compound(const AV2_COMMON *cm,
                                            MACROBLOCKD *xd,
                                            const MB_MODE_INFO *mbmi) {
  if (is_masked_compound_type(mbmi->interinter_comp.type) ||
      mbmi->mode == GLOBAL_GLOBALMV)
    return 0;

  if (mbmi->mode == NEW_NEWMV && mbmi->motion_mode == WARP_CAUSAL) return 0;

  (void)cm;
  const struct scale_factors *const sf0 = xd->block_ref_scale_factors[0];
  const struct scale_factors *const sf1 = xd->block_ref_scale_factors[1];
  return has_second_ref(mbmi) &&
         (mbmi->mode >= COMP_INTER_MODE_START &&
          mbmi->mode < COMP_INTER_MODE_END) &&
         !av2_is_scaled(sf0) && !av2_is_scaled(sf1) &&
         (mbmi->interinter_comp.type == COMPOUND_DIFFWTD ||
          mbmi->interinter_comp.type == COMPOUND_AVERAGE);
}
int is_out_of_frame_block(InterPredParams const *inter_pred_params,
                          int frame_width, int frame_height, int sub_block_id);

static INLINE int64_t scaled_buffer_offset(int x_offset, int y_offset,
                                           int stride,
                                           const struct scale_factors *sf) {
  const int x =
      sf ? sf->scale_value_x(x_offset, sf) >> SCALE_EXTRA_BITS : x_offset;
  const int y =
      sf ? sf->scale_value_y(y_offset, sf) >> SCALE_EXTRA_BITS : y_offset;
  return (int64_t)y * stride + x;
}

static INLINE void setup_pred_plane(struct buf_2d *dst, uint16_t *src,
                                    int width, int height, int crop_width,
                                    int crop_height, int stride, int mi_row,
                                    int mi_col,
                                    const struct scale_factors *scale,
                                    int subsampling_x, int subsampling_y,
                                    const CHROMA_REF_INFO *chroma_ref_info) {
  // Offset the buffer pointer
  if (chroma_ref_info && (subsampling_x || subsampling_y)) {
    mi_row = chroma_ref_info->mi_row_chroma_base;
    mi_col = chroma_ref_info->mi_col_chroma_base;
  }

  const int x = (MI_SIZE * mi_col) >> subsampling_x;
  const int y = (MI_SIZE * mi_row) >> subsampling_y;
  dst->buf = src + scaled_buffer_offset(x, y, stride, scale);
  dst->buf0 = src;
  dst->width = width;
  dst->height = height;
  dst->crop_width = crop_width;
  dst->crop_height = crop_height;
  dst->stride = stride;
}

void av2_setup_dst_planes(struct macroblockd_plane *planes,
                          const YV12_BUFFER_CONFIG *src, int mi_row, int mi_col,
                          const int plane_start, const int plane_end,
                          const CHROMA_REF_INFO *chroma_ref_info);

void av2_setup_pre_planes(MACROBLOCKD *xd, int idx,
                          const YV12_BUFFER_CONFIG *src, int mi_row, int mi_col,
                          const struct scale_factors *sf, const int num_planes,
                          const CHROMA_REF_INFO *chroma_ref_info);

static AVM_INLINE void setup_pred_planes_for_tip(const TIP *tip_ref,
                                                 MACROBLOCKD *xd,
                                                 int plane_start, int plane_end,
                                                 int mi_col, int mi_row) {
  for (int plane = plane_start; plane < AVMMIN(plane_end, MAX_MB_PLANE);
       ++plane) {
    struct macroblockd_plane *const pd = &xd->plane[plane];
    int is_uv = plane > 0;

    for (int ref = 0; ref < 2; ++ref) {
      const YV12_BUFFER_CONFIG *ref_buf = &tip_ref->ref_frame_buffer[ref]->buf;
      setup_pred_plane(&pd->pre[ref], ref_buf->buffers[plane],
                       ref_buf->widths[is_uv], ref_buf->heights[is_uv],
                       ref_buf->crop_widths[is_uv],
                       ref_buf->crop_heights[is_uv], ref_buf->strides[is_uv],
                       mi_row, mi_col, tip_ref->ref_scale_factor[ref],
                       pd->subsampling_x, pd->subsampling_y, NULL);
    }
  }
}

static INLINE void set_default_interp_filters(
    MB_MODE_INFO *const mbmi, const AV2_COMMON *cm, const MACROBLOCKD *xd,
    InterpFilter frame_interp_filter) {
  if (mbmi->skip_mode) {
    mbmi->interp_fltr = MULTITAP_SHARP;
    return;
  }
  mbmi->interp_fltr =
      (opfl_allowed_cur_pred_mode(cm, xd, mbmi) || mbmi->refinemv_flag ||
       is_tip_ref_frame(mbmi->ref_frame[0]))
          ? MULTITAP_SHARP
          : av2_unswitchable_filter(frame_interp_filter);
}

static INLINE int av2_is_interp_needed(const AV2_COMMON *const cm,
                                       const MACROBLOCKD *const xd) {
  (void)cm;
  const MB_MODE_INFO *const mbmi = xd->mi[0];
  if (mbmi->skip_mode) return 0;

  if (mbmi->mode == WARPMV) return 0;
  // No interpolation filter search when optical flow MV refinement is used.
  if (opfl_allowed_cur_pred_mode(cm, xd, mbmi)) return 0;

  // No interpolation filter search when MV refinement is used.
  if (mbmi->refinemv_flag) return 0;
  if (is_warp_mode(mbmi->motion_mode)) return 0;
  if (is_nontrans_global_motion(xd, xd->mi[0])) return 0;
  if (is_tip_ref_frame(mbmi->ref_frame[0])) return 0;
  return 1;
}

#define MASK_MASTER_SIZE ((MAX_WEDGE_SIZE) << 1)
#define MASK_MASTER_STRIDE (MASK_MASTER_SIZE)

void av2_init_wedge_masks();

static INLINE const uint8_t *av2_get_all_contiguous_soft_mask(
    int8_t wedge_index, int8_t wedge_sign, BLOCK_SIZE sb_type,
    int boundary_index) {
  return av2_wedge_params_lookup[sb_type]
      .all_masks[wedge_sign][wedge_index][boundary_index];
}

static INLINE int8_t get_wedge_boundary_type(BLOCK_SIZE bsize) {
  int8_t boundary_type = 1;  // smooth mask
  if (bsize <= BLOCK_16X16) {
    boundary_type = 0;  // sharp mask
  }
  return boundary_type;
}

static INLINE const uint8_t *av2_get_contiguous_soft_mask_decision(
    int8_t wedge_index, int8_t wedge_sign, int boundary_index,
    BLOCK_SIZE sb_type) {
  return av2_wedge_params_lookup[sb_type]
      .tmvp_mv_decisions[wedge_sign][wedge_index][boundary_index];
}

const uint8_t *av2_get_compound_type_mask(
    const INTERINTER_COMPOUND_DATA *const comp_data, BLOCK_SIZE sb_type);

// Init the masks for compound weighted prediction
void init_cwp_masks();
// Get the mask for compound weighted prediction
const int8_t *av2_get_cwp_mask(int list_idx, int idx);

// build interintra_predictors for one plane
void av2_build_interintra_predictor(const AV2_COMMON *cm, MACROBLOCKD *xd,
                                    uint16_t *pred, int stride,
                                    const BUFFER_SET *ctx, int plane,
                                    BLOCK_SIZE bsize);

void av2_build_intra_predictors_for_interintra(const AV2_COMMON *cm,
                                               MACROBLOCKD *xd, int plane,
                                               const BUFFER_SET *ctx,
                                               uint16_t *dst, int dst_stride);

void av2_combine_interintra(MACROBLOCKD *xd, BLOCK_SIZE bsize, int plane,
                            const uint16_t *inter_pred, int inter_stride,
                            const uint16_t *intra_pred, int intra_stride);

int av2_allow_warp(const MB_MODE_INFO *const mbmi,
                   const WarpTypesAllowed *const warp_types,
                   const WarpedMotionParams *const gm_params, int ref,
                   const struct scale_factors *const sf,
                   WarpedMotionParams *final_warp_params);

static INLINE int av2_allow_bawp(const AV2_COMMON *const cm,
                                 const MB_MODE_INFO *mbmi, int mi_row,
                                 int mi_col) {
  if (mbmi->mode == WARPMV) return 0;
  if (mbmi->mode == GLOBALMV) return 0;
  // In error resilient mode, whether a frame is scaled may be unknown, and the
  // following logic for parsing is not reliable, so we disable BAWP.
  if (frame_is_sframe(cm)) return 0;

  if ((get_ref_frame_buf(cm, mbmi->ref_frame[0]) != NULL &&
       get_ref_frame_buf(cm, mbmi->ref_frame[0])->is_restricted) ||
      (get_ref_frame_buf(cm, mbmi->ref_frame[1]) != NULL &&
       get_ref_frame_buf(cm, mbmi->ref_frame[1])->is_restricted))
    return 0;

  // If one of the reference frame is different resolution than the current
  // frame, bawp is disabled.
  const struct scale_factors *const sf0 =
      get_ref_scale_factors_const(cm, mbmi->ref_frame[0]);
  const struct scale_factors *const sf1 =
      get_ref_scale_factors_const(cm, mbmi->ref_frame[1]);
  if ((sf0 != NULL && av2_is_scaled(sf0)) ||
      (sf1 != NULL && av2_is_scaled(sf1))) {
    return 0;
  }
  if (mbmi->mode == WARP_NEWMV) return 0;
  if (is_tip_ref_frame(mbmi->ref_frame[0])) return 0;
  if (is_motion_variation_allowed_bsize(mbmi->sb_type[PLANE_TYPE_Y], mi_row,
                                        mi_col) &&
      is_inter_singleref_mode(mbmi->mode))
    return 1;
  else
    return 0;
}

static INLINE int av2_allow_explicit_bawp(const MB_MODE_INFO *mbmi) {
  return mbmi->mode == NEWMV || mbmi->mode == NEARMV;
}

// derive the context of the mpp_flag
int av2_get_mpp_flag_context(const AV2_COMMON *cm, const MACROBLOCKD *xd);

// derive the context of the precision signaling
int av2_get_pb_mv_precision_down_context(const AV2_COMMON *cm,
                                         const MACROBLOCKD *xd);

// set the precision of a block to the precision
void set_mv_precision(MB_MODE_INFO *mbmi, MvSubpelPrecision precision);
void set_amvd_mv_precision(MB_MODE_INFO *mbmi, MvSubpelPrecision precision);

// Function to check if precision need to be signaled or not
int is_intraBC_bv_precision_active(const AV2_COMMON *const cm,
                                   const int intrabc_mode);
// Set max value as default precision
void set_default_intraBC_bv_precision(const AV2_COMMON *const cm,
                                      MB_MODE_INFO *mbmi);

// set the most probable mv precision of the block
// Currently, the most probable MV precision is same as the maximum precision of
// the block.
void set_most_probable_mv_precision(const AV2_COMMON *const cm,
                                    MB_MODE_INFO *mbmi, const BLOCK_SIZE bsize);

// Set the default value fo the precision set. Currently the value is always 0.
void set_default_precision_set(const AV2_COMMON *const cm, MB_MODE_INFO *mbmi,
                               const BLOCK_SIZE bsize);

// Set the precision set of the block. Currently, the value is 0.
void set_precision_set(const AV2_COMMON *const cm, MACROBLOCKD *const xd,
                       MB_MODE_INFO *mbmi, const BLOCK_SIZE bsize,
                       int *ref_mv_idx);
// Get the index of the precision
// this index is signalled when precision is not same as the most probable
// precision
int av2_get_pb_mv_precision_index(const MB_MODE_INFO *mbmi);

// get the actual precision value from the signalled index
MvSubpelPrecision av2_get_precision_from_index(MB_MODE_INFO *mbmi,
                                               int precision_idx_coded_value);

// Set the maximum precision to the default value
void set_default_max_mv_precision(MB_MODE_INFO *mbmi,
                                  MvSubpelPrecision precision);

// get the maximum allowed precision value of the block
MvSubpelPrecision av2_get_mbmi_max_mv_precision(const AV2_COMMON *const cm,
                                                const SB_INFO *sbi,
                                                const MB_MODE_INFO *mbmi);

// check if pb_mv_precision is allowed or not
int is_pb_mv_precision_active(const AV2_COMMON *const cm,
                              const MB_MODE_INFO *mbmi, const BLOCK_SIZE bsize);

// check if the WARPMV mode is allwed for a given blocksize
static INLINE int is_warpmv_allowed_bsize(BLOCK_SIZE bsize) {
  assert(bsize < BLOCK_SIZES_ALL);
  return AVMMIN(block_size_wide[bsize], block_size_high[bsize]) >= 8;
}

// check if WARPMV mode is allowed
static INLINE int is_warpmv_mode_allowed(const AV2_COMMON *const cm,
                                         const MB_MODE_INFO *mbmi,
                                         BLOCK_SIZE bsize) {
  int frame_warp_delta_allowed =
      (cm->features.enabled_motion_modes & (1 << WARP_DELTA)) != 0;

  if (has_second_ref(mbmi) || !frame_warp_delta_allowed ||
      is_tip_ref_frame(mbmi->ref_frame[0]) || !cm->features.allow_warpmv_mode)
    return 0;

  return frame_warp_delta_allowed && is_warpmv_allowed_bsize(bsize);
}

// check if warpmv with mvd is allowed or not
static INLINE int allow_warpmv_with_mvd_coding(const AV2_COMMON *const cm,
                                               const MB_MODE_INFO *mbmi) {
  if (!cm->features.allow_warpmv_mode) return 0;
  return (mbmi->mode == WARPMV && mbmi->warp_ref_idx < 2);
}

// Return the threshold value for number of non-zero componenets for sign
// derivation.
static INLINE int get_derive_sign_nzero_th(const MB_MODE_INFO *mbmi) {
  return (mbmi->mode == NEW_NEWMV || mbmi->mode == NEW_NEWMV_OPTFLOW) ? 4 : 1;
}
// Check if the sign derivation method is allowed or not for current block
static INLINE int is_mvd_sign_derive_allowed(const AV2_COMMON *const cm,
                                             const MACROBLOCKD *const xd,
                                             const MB_MODE_INFO *mbmi) {
  if (!cm->seq_params.enable_mvd_sign_derive ||
      mbmi->motion_mode != SIMPLE_TRANSLATION ||
      is_intrabc_block(mbmi, xd->tree_type) ||
      enable_adaptive_mvd_resolution(cm, mbmi) || mbmi->skip_mode ||
      cm->features.allow_screen_content_tools ||
      cm->features.fr_mv_precision > MV_PRECISION_QTR_PEL ||
      mbmi->pb_mv_precision >= MV_PRECISION_QTR_PEL)
    return 0;

  if (has_second_ref(mbmi)) {
    int drl_idx = mbmi->ref_mv_idx[0];
    if (has_second_drl(mbmi)) {
      drl_idx = AVMMAX(mbmi->ref_mv_idx[0], mbmi->ref_mv_idx[1]);
    }
    if (drl_idx > 0) return 0;
  }
  return (mbmi->mode == NEWMV || mbmi->mode == WARP_NEWMV ||
          mbmi->mode == JOINT_NEWMV || mbmi->mode == JOINT_NEWMV_OPTFLOW ||
          mbmi->mode == NEW_NEWMV || mbmi->mode == NEW_NEWMV_OPTFLOW);
}

bool av2_build_morph_pred(const AV2_COMMON *const cm, MACROBLOCKD *const xd,
                          const BLOCK_SIZE bsize, const int mi_row,
                          const int mi_col);

static AVM_INLINE bool is_subblock_outside(int x, int y, int mi_cols,
                                           int mi_rows, int build_for_decode) {
  if (!build_for_decode) return 0;
  return (x >= mi_cols * MI_SIZE || y >= mi_rows * MI_SIZE);
}

#ifdef __cplusplus
}  // extern "C"
#endif

#endif  // AVM_AV2_COMMON_RECONINTER_H_

Coverage Report

Created: 2026-09-14 06:49