EncCu.cpp

/* The copyright in this software is being made available under the BSD
 * License, included below. This software may be subject to other third party
 * and contributor rights, including patent rights, and no such rights are
 * granted under this license.
 *
 * Copyright (c) 2010-2025, ITU/ISO/IEC
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 *  * Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 *  * Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 *  * Neither the name of the ITU/ISO/IEC nor the names of its contributors may
 *    be used to endorse or promote products derived from this software without
 *    specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS
 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
 * THE POSSIBILITY OF SUCH DAMAGE.
 */

/** \file     EncCu.cpp
    \brief    Coding Unit (CU) encoder class
*/

#include "EncCu.h"

#include "EncLib.h"
#include "Analyze.h"
#include "AQp.h"

#include "CommonLib/dtrace_codingstruct.h"
#include "CommonLib/Picture.h"
#include "CommonLib/UnitTools.h"
#include "MCTS.h"


#include "CommonLib/dtrace_buffer.h"

#include <stdio.h>
#include <cmath>
#include <algorithm>

//! \ingroup EncoderLib
//! \{

// ====================================================================================================================

const MergeIdxPair EncCu::m_geoModeTest[GEO_MAX_NUM_CANDS] = {
  MergeIdxPair{ 0, 1 }, MergeIdxPair{ 1, 0 }, MergeIdxPair{ 0, 2 }, MergeIdxPair{ 1, 2 }, MergeIdxPair{ 2, 0 },
  MergeIdxPair{ 2, 1 }, MergeIdxPair{ 0, 3 }, MergeIdxPair{ 1, 3 }, MergeIdxPair{ 2, 3 }, MergeIdxPair{ 3, 0 },
  MergeIdxPair{ 3, 1 }, MergeIdxPair{ 3, 2 }, MergeIdxPair{ 0, 4 }, MergeIdxPair{ 1, 4 }, MergeIdxPair{ 2, 4 },
  MergeIdxPair{ 3, 4 }, MergeIdxPair{ 4, 0 }, MergeIdxPair{ 4, 1 }, MergeIdxPair{ 4, 2 }, MergeIdxPair{ 4, 3 },
  MergeIdxPair{ 0, 5 }, MergeIdxPair{ 1, 5 }, MergeIdxPair{ 2, 5 }, MergeIdxPair{ 3, 5 }, MergeIdxPair{ 4, 5 },
  MergeIdxPair{ 5, 0 }, MergeIdxPair{ 5, 1 }, MergeIdxPair{ 5, 2 }, MergeIdxPair{ 5, 3 }, MergeIdxPair{ 5, 4 }
};

EncCu::EncCu() {}

void EncCu::create( EncCfg* encCfg )
{
  unsigned      uiMaxWidth    = encCfg->getMaxCUWidth();
  unsigned      uiMaxHeight   = encCfg->getMaxCUHeight();
  ChromaFormat  chromaFormat  = encCfg->getChromaFormatIdc();

  unsigned      numWidths     = gp_sizeIdxInfo->numWidths();
  unsigned      numHeights    = gp_sizeIdxInfo->numHeights();
  m_pTempCS = new CodingStructure**  [numWidths];
  m_pBestCS = new CodingStructure**  [numWidths];
  m_pTempCS2 = new CodingStructure** [numWidths];
  m_pBestCS2 = new CodingStructure** [numWidths];

  m_pelUnitBufPool.initPelUnitBufPool(chromaFormat, uiMaxWidth, uiMaxHeight);
  m_mergeItemList.init(encCfg->getMaxMergeRdCandNumTotal(), chromaFormat, uiMaxWidth, uiMaxHeight);

  for( unsigned w = 0; w < numWidths; w++ )
  {
    m_pTempCS[w] = new CodingStructure*  [numHeights];
    m_pBestCS[w] = new CodingStructure*  [numHeights];
    m_pTempCS2[w] = new CodingStructure* [numHeights];
    m_pBestCS2[w] = new CodingStructure* [numHeights];

    for( unsigned h = 0; h < numHeights; h++ )
    {
      unsigned width  = gp_sizeIdxInfo->sizeFrom( w );
      unsigned height = gp_sizeIdxInfo->sizeFrom( h );

      if( gp_sizeIdxInfo->isCuSize( width ) && gp_sizeIdxInfo->isCuSize( height ) 
        && width <= uiMaxWidth && height <= uiMaxHeight)
      {
        m_pTempCS[w][h] = new CodingStructure(m_unitPool);
        m_pBestCS[w][h] = new CodingStructure(m_unitPool);

        m_pTempCS[w][h]->create(chromaFormat, Area(0, 0, width, height), false, (bool)encCfg->getPLTMode());
        m_pBestCS[w][h]->create(chromaFormat, Area(0, 0, width, height), false, (bool)encCfg->getPLTMode());

        m_pTempCS2[w][h] = new CodingStructure(m_unitPool);
        m_pBestCS2[w][h] = new CodingStructure(m_unitPool);

        m_pTempCS2[w][h]->create(chromaFormat, Area(0, 0, width, height), false, (bool)encCfg->getPLTMode());
        m_pBestCS2[w][h]->create(chromaFormat, Area(0, 0, width, height), false, (bool)encCfg->getPLTMode());
      }
      else
      {
        m_pTempCS[w][h] = nullptr;
        m_pBestCS[w][h] = nullptr;
        m_pTempCS2[w][h] = nullptr;
        m_pBestCS2[w][h] = nullptr;
      }
    }
  }

  m_cuChromaQpOffsetIdxPlus1 = 0;

  unsigned maxDepth = numWidths + numHeights;

  m_modeCtrl = new EncModeCtrlMTnoRQT();

  m_modeCtrl->create( *encCfg );

  m_ctxBuffer.resize(maxDepth);
  m_CurrCtx = 0;
}


void EncCu::destroy()
{
  unsigned numWidths  = gp_sizeIdxInfo->numWidths();
  unsigned numHeights = gp_sizeIdxInfo->numHeights();

  for( unsigned w = 0; w < numWidths; w++ )
  {
    for( unsigned h = 0; h < numHeights; h++ )
    {
      if (m_pBestCS[w][h])
      {
        m_pBestCS[w][h]->destroy();
      }
      if (m_pTempCS[w][h])
      {
        m_pTempCS[w][h]->destroy();
      }

      delete m_pBestCS[w][h];
      delete m_pTempCS[w][h];

      if (m_pBestCS2[w][h])
      {
        m_pBestCS2[w][h]->destroy();
      }
      if (m_pTempCS2[w][h])
      {
        m_pTempCS2[w][h]->destroy();
      }

      delete m_pBestCS2[w][h];
      delete m_pTempCS2[w][h];
    }

    delete[] m_pTempCS[w];
    delete[] m_pBestCS[w];
    delete[] m_pTempCS2[w];
    delete[] m_pBestCS2[w];
  }

  delete[] m_pBestCS; m_pBestCS = nullptr;
  delete[] m_pTempCS; m_pTempCS = nullptr;
  delete[] m_pBestCS2; m_pBestCS2 = nullptr;
  delete[] m_pTempCS2; m_pTempCS2 = nullptr;

#if REUSE_CU_RESULTS
  if (m_tmpStorageCtu)
  {
    m_tmpStorageCtu->destroy();
    delete m_tmpStorageCtu;
    m_tmpStorageCtu = nullptr;
  }
#endif

#if REUSE_CU_RESULTS
  m_modeCtrl->destroy();

#endif
  delete m_modeCtrl;
  m_modeCtrl = nullptr;

}

EncCu::~EncCu()
{
}

/** \param    pcEncLib      pointer of encoder class
 */
void EncCu::init( EncLib* pcEncLib, const SPS& sps )
{
  m_pcEncCfg           = pcEncLib;
  m_pcIntraSearch      = pcEncLib->getIntraSearch();
  m_pcInterSearch      = pcEncLib->getInterSearch();
  m_pcTrQuant          = pcEncLib->getTrQuant();
  m_pcRdCost           = pcEncLib->getRdCost ();
  m_CABACEstimator     = pcEncLib->getCABACEncoder()->getCABACEstimator( &sps );
  m_CABACEstimator->setEncCu(this);
  m_ctxPool            = pcEncLib->getCtxCache();
  m_pcRateCtrl         = pcEncLib->getRateCtrl();
  m_pcSliceEncoder     = pcEncLib->getSliceEncoder();
  m_deblockingFilter   = pcEncLib->getDeblockingFilter();
  m_geoCostList.init(m_pcEncCfg->getMaxNumGeoCand());
  m_AFFBestSATDCost = MAX_DOUBLE;

  DecCu::init( m_pcTrQuant, m_pcIntraSearch, m_pcInterSearch );

  m_modeCtrl->init( m_pcEncCfg, m_pcRateCtrl, m_pcRdCost );
  m_modeCtrl->setBIMQPMap( m_pcEncCfg->getAdaptQPmap() );

  m_pcInterSearch->setModeCtrl( m_modeCtrl );
  m_modeCtrl->setInterSearch(m_pcInterSearch);
  m_pcIntraSearch->setModeCtrl( m_modeCtrl );

  m_pcGOPEncoder = pcEncLib->getGOPEncoder();
  m_pcGOPEncoder->setModeCtrl( m_modeCtrl );
}

// ====================================================================================================================
// Public member functions
// ====================================================================================================================

void EncCu::compressCtu(CodingStructure &cs, const UnitArea &area, const unsigned ctuRsAddr,
                        const EnumArray<int, ChannelType> &prevQP, const EnumArray<int, ChannelType> &currQP)
{
  m_modeCtrl->initCTUEncoding( *cs.slice );
  cs.treeType = TREE_D;

  cs.slice->m_mapPltCost[0].clear();
  cs.slice->m_mapPltCost[1].clear();
  // init the partitioning manager
  QTBTPartitioner partitioner;
  partitioner.initCtu(area, ChannelType::LUMA, *cs.slice);
  if (m_pcEncCfg->getIBCMode())
  {
    if (area.lx() == 0 && area.ly() == 0)
    {
      m_pcInterSearch->resetIbcSearch();
    }
    m_pcInterSearch->resetCtuRecord();
    m_ctuIbcSearchRangeX = m_pcEncCfg->getIBCLocalSearchRangeX();
    m_ctuIbcSearchRangeY = m_pcEncCfg->getIBCLocalSearchRangeY();
  }
  if (m_pcEncCfg->getIBCMode() && m_pcEncCfg->getIBCHashSearch() && (m_pcEncCfg->getIBCFastMethod() & IBC_FAST_METHOD_ADAPTIVE_SEARCHRANGE))
  {
    const int hashHitRatio = m_ibcHashMap.getHashHitRatio(area.Y()); // in percent
    if (hashHitRatio < 5) // 5%
    {
      m_ctuIbcSearchRangeX >>= 1;
      m_ctuIbcSearchRangeY >>= 1;
    }
    if (cs.slice->getNumRefIdx(REF_PIC_LIST_0) > 0)
    {
      m_ctuIbcSearchRangeX >>= 1;
      m_ctuIbcSearchRangeY >>= 1;
    }
  }
  // init current context pointer
  m_CurrCtx = m_ctxBuffer.data();

  CodingStructure *tempCS = m_pTempCS[gp_sizeIdxInfo->idxFrom( area.lumaSize().width )][gp_sizeIdxInfo->idxFrom( area.lumaSize().height )];
  CodingStructure *bestCS = m_pBestCS[gp_sizeIdxInfo->idxFrom( area.lumaSize().width )][gp_sizeIdxInfo->idxFrom( area.lumaSize().height )];

  cs.initSubStructure(*tempCS, partitioner.chType, partitioner.currArea(), false);
  cs.initSubStructure(*bestCS, partitioner.chType, partitioner.currArea(), false);
  tempCS->currQP[ChannelType::LUMA] = bestCS->currQP[ChannelType::LUMA] = tempCS->baseQP = bestCS->baseQP =
    currQP[ChannelType::LUMA];
  tempCS->prevQP[ChannelType::LUMA] = bestCS->prevQP[ChannelType::LUMA] = prevQP[ChannelType::LUMA];

  xCompressCU(tempCS, bestCS, partitioner);
  cs.slice->m_mapPltCost[0].clear();
  cs.slice->m_mapPltCost[1].clear();
  // all signals were already copied during compression if the CTU was split - at this point only the structures are copied to the top level CS
  const bool copyUnsplitCTUSignals = bestCS->cus.size() == 1;
  cs.useSubStructure(*bestCS, partitioner.chType, CS::getArea(*bestCS, area, partitioner.chType), copyUnsplitCTUSignals,
                     false, false, copyUnsplitCTUSignals, true);

  if (CS::isDualITree (cs) && isChromaEnabled (cs.pcv->chrFormat))
  {
    m_CABACEstimator->getCtx() = m_CurrCtx->start;

    partitioner.initCtu(area, ChannelType::CHROMA, *cs.slice);

    cs.initSubStructure(*tempCS, partitioner.chType, partitioner.currArea(), false);
    cs.initSubStructure(*bestCS, partitioner.chType, partitioner.currArea(), false);
    tempCS->currQP[ChannelType::CHROMA] = bestCS->currQP[ChannelType::CHROMA] = tempCS->baseQP = bestCS->baseQP =
      currQP[ChannelType::CHROMA];
    tempCS->prevQP[ChannelType::CHROMA] = bestCS->prevQP[ChannelType::CHROMA] = prevQP[ChannelType::CHROMA];

    xCompressCU(tempCS, bestCS, partitioner);

    const bool copyUnsplitCTUSignals = bestCS->cus.size() == 1;
    cs.useSubStructure(*bestCS, partitioner.chType, CS::getArea(*bestCS, area, partitioner.chType),
                       copyUnsplitCTUSignals, false, false, copyUnsplitCTUSignals, true);
  }

  if (m_pcEncCfg->getUseRateCtrl())
  {
    (m_pcRateCtrl->getRCPic()->getLCU(ctuRsAddr)).m_actualMSE = (double)bestCS->dist / (double)m_pcRateCtrl->getRCPic()->getLCU(ctuRsAddr).m_numberOfPixel;
  }
  // reset context states and uninit context pointer
  m_CABACEstimator->getCtx() = m_CurrCtx->start;
  m_CurrCtx                  = 0;


  // Ensure that a coding was found
  // Selected mode's RD-cost must be not MAX_DOUBLE.
  CHECK( bestCS->cus.empty()                                   , "No possible encoding found" );
  CHECK( bestCS->cus[0]->predMode == NUMBER_OF_PREDICTION_MODES, "No possible encoding found" );
  CHECK( bestCS->cost             == MAX_DOUBLE                , "No possible encoding found" );
}

// ====================================================================================================================
// Protected member functions
// ====================================================================================================================

static int xCalcHADs8x8_ISlice(const Pel *piOrg, const ptrdiff_t strideOrg)
{
  int k, i, j, jj;
  int diff[64], m1[8][8], m2[8][8], m3[8][8], iSumHad = 0;

  for (k = 0; k < 64; k += 8)
  {
    diff[k + 0] = piOrg[0];
    diff[k + 1] = piOrg[1];
    diff[k + 2] = piOrg[2];
    diff[k + 3] = piOrg[3];
    diff[k + 4] = piOrg[4];
    diff[k + 5] = piOrg[5];
    diff[k + 6] = piOrg[6];
    diff[k + 7] = piOrg[7];

    piOrg += strideOrg;
  }

  //horizontal
  for (j = 0; j < 8; j++)
  {
    jj = j << 3;
    m2[j][0] = diff[jj    ] + diff[jj + 4];
    m2[j][1] = diff[jj + 1] + diff[jj + 5];
    m2[j][2] = diff[jj + 2] + diff[jj + 6];
    m2[j][3] = diff[jj + 3] + diff[jj + 7];
    m2[j][4] = diff[jj    ] - diff[jj + 4];
    m2[j][5] = diff[jj + 1] - diff[jj + 5];
    m2[j][6] = diff[jj + 2] - diff[jj + 6];
    m2[j][7] = diff[jj + 3] - diff[jj + 7];

    m1[j][0] = m2[j][0] + m2[j][2];
    m1[j][1] = m2[j][1] + m2[j][3];
    m1[j][2] = m2[j][0] - m2[j][2];
    m1[j][3] = m2[j][1] - m2[j][3];
    m1[j][4] = m2[j][4] + m2[j][6];
    m1[j][5] = m2[j][5] + m2[j][7];
    m1[j][6] = m2[j][4] - m2[j][6];
    m1[j][7] = m2[j][5] - m2[j][7];

    m2[j][0] = m1[j][0] + m1[j][1];
    m2[j][1] = m1[j][0] - m1[j][1];
    m2[j][2] = m1[j][2] + m1[j][3];
    m2[j][3] = m1[j][2] - m1[j][3];
    m2[j][4] = m1[j][4] + m1[j][5];
    m2[j][5] = m1[j][4] - m1[j][5];
    m2[j][6] = m1[j][6] + m1[j][7];
    m2[j][7] = m1[j][6] - m1[j][7];
  }

  //vertical
  for (i = 0; i < 8; i++)
  {
    m3[0][i] = m2[0][i] + m2[4][i];
    m3[1][i] = m2[1][i] + m2[5][i];
    m3[2][i] = m2[2][i] + m2[6][i];
    m3[3][i] = m2[3][i] + m2[7][i];
    m3[4][i] = m2[0][i] - m2[4][i];
    m3[5][i] = m2[1][i] - m2[5][i];
    m3[6][i] = m2[2][i] - m2[6][i];
    m3[7][i] = m2[3][i] - m2[7][i];

    m1[0][i] = m3[0][i] + m3[2][i];
    m1[1][i] = m3[1][i] + m3[3][i];
    m1[2][i] = m3[0][i] - m3[2][i];
    m1[3][i] = m3[1][i] - m3[3][i];
    m1[4][i] = m3[4][i] + m3[6][i];
    m1[5][i] = m3[5][i] + m3[7][i];
    m1[6][i] = m3[4][i] - m3[6][i];
    m1[7][i] = m3[5][i] - m3[7][i];

    m2[0][i] = m1[0][i] + m1[1][i];
    m2[1][i] = m1[0][i] - m1[1][i];
    m2[2][i] = m1[2][i] + m1[3][i];
    m2[3][i] = m1[2][i] - m1[3][i];
    m2[4][i] = m1[4][i] + m1[5][i];
    m2[5][i] = m1[4][i] - m1[5][i];
    m2[6][i] = m1[6][i] + m1[7][i];
    m2[7][i] = m1[6][i] - m1[7][i];
  }

  for (i = 0; i < 8; i++)
  {
    for (j = 0; j < 8; j++)
    {
      iSumHad += abs(m2[i][j]);
    }
  }
  iSumHad -= abs(m2[0][0]);
  iSumHad = (iSumHad + 2) >> 2;
  return(iSumHad);
}

int  EncCu::updateCtuDataISlice(const CPelBuf buf)
{
  int  xBl, yBl;
  const int iBlkSize = 8;
  const Pel* pOrgInit = buf.buf;
  ptrdiff_t  iStrideOrig = buf.stride;

  int iSumHad = 0;
  for( yBl = 0; ( yBl + iBlkSize ) <= buf.height; yBl += iBlkSize )
  {
    for( xBl = 0; ( xBl + iBlkSize ) <= buf.width; xBl += iBlkSize )
    {
      const Pel* pOrg = pOrgInit + iStrideOrig*yBl + xBl;
      iSumHad += xCalcHADs8x8_ISlice( pOrg, iStrideOrig );
    }
  }
  return( iSumHad );
}

bool EncCu::xCheckBestMode( CodingStructure *&tempCS, CodingStructure *&bestCS, Partitioner &partitioner, const EncTestMode& encTestMode )
{
  bool bestCSUpdated = false;

  if( !tempCS->cus.empty() )
  {
    if( tempCS->cus.size() == 1 )
    {
      const CodingUnit& cu = *tempCS->cus.front();
      CHECK( cu.skip && !cu.firstPU->mergeFlag, "Skip flag without a merge flag is not allowed!" );
    }

#if WCG_EXT
    DTRACE_BEST_MODE( tempCS, bestCS, m_pcRdCost->getLambda( true ) );
#else
    DTRACE_BEST_MODE( tempCS, bestCS, m_pcRdCost->getLambda() );
#endif

    if( m_modeCtrl->useModeResult( encTestMode, tempCS, partitioner ) )
    {
      std::swap( tempCS, bestCS );
      // store temp best CI for next CU coding
      m_CurrCtx->best = m_CABACEstimator->getCtx();
      m_bestModeUpdated = true;
      bestCSUpdated = true;
    }
  }

  // reset context states
  m_CABACEstimator->getCtx() = m_CurrCtx->start;
  return bestCSUpdated;
}

void EncCu::xCompressCU( CodingStructure*& tempCS, CodingStructure*& bestCS, Partitioner& partitioner, double maxCostAllowed )
{
  CHECK(maxCostAllowed < 0, "Wrong value of maxCostAllowed!");

  uint32_t compBegin;
  uint32_t numComp;
  bool jointPLT = false;
  if (partitioner.isSepTree( *tempCS ))
  {
    if( !CS::isDualITree(*tempCS) && partitioner.treeType != TREE_D )
    {
      compBegin = COMPONENT_Y;
      numComp   = getNumberValidComponents(tempCS->area.chromaFormat);
      jointPLT = true;
    }
    else
    {
      if (isLuma(partitioner.chType))
      {
        compBegin = COMPONENT_Y;
        numComp   = 1;
      }
      else
      {
        compBegin = COMPONENT_Cb;
        numComp   = 2;
      }
    }
  }
  else
  {
    compBegin = COMPONENT_Y;
    numComp   = getNumberValidComponents(tempCS->area.chromaFormat);
    jointPLT = true;
  }
  SplitSeries splitmode = -1;
  uint8_t   bestLastPLTSize[MAX_NUM_CHANNEL_TYPE];
  Pel       bestLastPLT[MAX_NUM_COMPONENT][MAXPLTPREDSIZE]; // store LastPLT for
  uint8_t   curLastPLTSize[MAX_NUM_CHANNEL_TYPE];
  Pel       curLastPLT[MAX_NUM_COMPONENT][MAXPLTPREDSIZE]; // store LastPLT if no partition
  for (int i = compBegin; i < (compBegin + numComp); i++)
  {
    ComponentID comID = jointPLT ? (ComponentID)compBegin : ((i > 0) ? COMPONENT_Cb : COMPONENT_Y);
    bestLastPLTSize[comID] = 0;
    curLastPLTSize[comID] = tempCS->prevPLT.curPLTSize[comID];
    memcpy(curLastPLT[i], tempCS->prevPLT.curPLT[i], tempCS->prevPLT.curPLTSize[comID] * sizeof(Pel));
  }

  Slice&   slice      = *tempCS->slice;
  const PPS &pps      = *tempCS->pps;
  const SPS &sps      = *tempCS->sps;
  const uint32_t uiLPelX  = tempCS->area.Y().lumaPos().x;
  const uint32_t uiTPelY  = tempCS->area.Y().lumaPos().y;

  const ModeType modeTypeParent  = partitioner.modeType;
  const TreeType treeTypeParent  = partitioner.treeType;
  const ChannelType chTypeParent = partitioner.chType;
  const UnitArea currCsArea = clipArea( CS::getArea( *bestCS, bestCS->area, partitioner.chType ), *tempCS->picture );

  tempCS->splitRdCostBest = nullptr;
  if (m_pcEncCfg->getDPF())
  {
    m_modeCtrl->setCurrCsArea(currCsArea);
    m_modeCtrl->setQpCtu(m_pcSliceEncoder->getQpCtu());
  }
  m_modeCtrl->initCULevel( partitioner, *tempCS );
#if GDR_ENABLED
  if (m_pcEncCfg->getGdrEnabled())
  {
    bool isInGdrInterval = slice.getPic()->gdrParam.inGdrInterval;

    // 1.0 applicable to inter picture only
    if (isInGdrInterval)
    {
      int gdrPocStart = m_pcEncCfg->getGdrPocStart();
      int gdrInterval = m_pcEncCfg->getGdrInterval();
      int gdrPeriod = m_pcEncCfg->getGdrPeriod();

      int picWidth = slice.getPPS()->getPicWidthInLumaSamples();
      int m1, m2, n1;

      int curPoc = slice.getPOC();
      int gdrPoc = (curPoc - gdrPocStart) % gdrPeriod;

      int begGdrX = 0;
      int endGdrX = 0;

      double dd = (picWidth / (double)gdrInterval);
      int mm = (int)((picWidth / (double)gdrInterval) + 0.49999);
      m1 = ((mm + 7) >> 3) << 3;
      m2 = ((mm + 0) >> 3) << 3;

      if (dd > mm && m1 == m2)
      {
        m1 = m1 + 8;
      }

      n1 = (picWidth - m2 * gdrInterval) / 8;

      if (gdrPoc < n1)
      {
        begGdrX = m1 * gdrPoc;
        endGdrX = begGdrX + m1;
      }
      else
      {
        begGdrX = m1 * n1 + m2 * (gdrPoc - n1);
        endGdrX = begGdrX + m2;
        if (picWidth <= endGdrX)
        {
          begGdrX = picWidth;
          endGdrX = picWidth;
        }
      }

      bool isInRefreshArea = tempCS->withinRefresh(begGdrX, endGdrX);

      if (isInRefreshArea)
      {
        m_modeCtrl->forceIntraMode();
      }
      else if (tempCS->containRefresh(begGdrX, endGdrX) || tempCS->overlapRefresh(begGdrX, endGdrX))
      {
        // 1.3.1 enable only vertical splits (QT, BT_V, TT_V)
        m_modeCtrl->forceVerSplitOnly();

        // 1.3.2 remove TT_V if it does not satisfy the condition
        if (tempCS->refreshCrossTTV(begGdrX, endGdrX))
        {
          m_modeCtrl->forceRemoveTTV();
        }
      }

      if (tempCS->area.lwidth() != tempCS->area.lheight())
      {
        m_modeCtrl->forceRemoveQT();
      }

      if (!m_modeCtrl->anyPredModeLeft())
      {
        m_modeCtrl->forceRemoveDontSplit();
      }

      if (isInRefreshArea && !m_modeCtrl->anyIntraIBCMode() && (tempCS->area.lwidth() == 4 || tempCS->area.lheight() == 4))
      {
        m_modeCtrl->finishCULevel(partitioner);
        return;
      }
    }
  }
#endif

  if (partitioner.currQtDepth == 0 && partitioner.currMtDepth == 0 && !tempCS->slice->isIntra()
      && (sps.getUseSBT() || sps.getExplicitMtsInterEnabled()))
  {
    auto slsSbt = dynamic_cast<SaveLoadEncInfoSbt*>( m_modeCtrl );
    int maxSLSize = sps.getUseSBT() ? tempCS->slice->getSPS()->getMaxTbSize() : MTS_INTER_MAX_CU_SIZE;
    slsSbt->resetSaveloadSbt( maxSLSize );
  }
  m_sbtCostSave[0] = m_sbtCostSave[1] = MAX_DOUBLE;

  m_CurrCtx->start = m_CABACEstimator->getCtx();

  if( slice.getUseChromaQpAdj() )
  {
    // TODO M0133 : double check encoder decisions with respect to chroma QG detection and actual encode
    int lgMinCuSize = sps.getLog2MinCodingBlockSize() +
      std::max<int>(0, floorLog2(sps.getCTUSize()) - sps.getLog2MinCodingBlockSize() - int((slice.getCuChromaQpOffsetSubdiv()+1) / 2));
    if( partitioner.currQgChromaEnable() )
    {
      m_cuChromaQpOffsetIdxPlus1 = ( ( uiLPelX >> lgMinCuSize ) + ( uiTPelY >> lgMinCuSize ) ) % ( pps.getChromaQpOffsetListLen() + 1 );
    }
  }
  else
  {
    m_cuChromaQpOffsetIdxPlus1 = 0;
  }

  if( !m_modeCtrl->anyMode() )
  {
    m_modeCtrl->finishCULevel( partitioner );
    return;
  }

  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cux", uiLPelX ) );
  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cuy", uiTPelY ) );
  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cuw", tempCS->area.lwidth() ) );
  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cuh", tempCS->area.lheight() ) );
  DTRACE( g_trace_ctx, D_COMMON, "@(%4d,%4d) [%2dx%2d]\n", tempCS->area.lx(), tempCS->area.ly(), tempCS->area.lwidth(), tempCS->area.lheight() );


  m_pcInterSearch->resetSavedAffineMotion();

  double bestIntPelCost = MAX_DOUBLE;

  if (tempCS->slice->getSPS()->getUseColorTrans())
  {
    tempCS->tmpColorSpaceCost = MAX_DOUBLE;
    bestCS->tmpColorSpaceCost = MAX_DOUBLE;
    tempCS->firstColorSpaceSelected = true;
    bestCS->firstColorSpaceSelected = true;
  }

  if (tempCS->slice->getSPS()->getUseColorTrans() && !CS::isDualITree(*tempCS))
  {
    tempCS->firstColorSpaceTestOnly = false;
    bestCS->firstColorSpaceTestOnly = false;
    tempCS->tmpColorSpaceIntraCost[0] = MAX_DOUBLE;
    tempCS->tmpColorSpaceIntraCost[1] = MAX_DOUBLE;
    bestCS->tmpColorSpaceIntraCost[0] = MAX_DOUBLE;
    bestCS->tmpColorSpaceIntraCost[1] = MAX_DOUBLE;

    if (tempCS->bestParent && tempCS->bestParent->firstColorSpaceTestOnly)
    {
      tempCS->firstColorSpaceTestOnly = bestCS->firstColorSpaceTestOnly = true;
    }
  }

  double splitRdCostBest[NUM_PART_SPLIT];
  std::fill(std::begin(splitRdCostBest), std::end(splitRdCostBest), MAX_DOUBLE);
  if (tempCS->slice->getCheckLDC())
  {
    m_bestBcwCost.fill(std::numeric_limits<double>::max());
    m_bestBcwIdx.fill(BCW_NUM);
  }
  do
  {
    for (int i = compBegin; i < (compBegin + numComp); i++)
    {
      ComponentID comID = jointPLT ? (ComponentID)compBegin : ((i > 0) ? COMPONENT_Cb : COMPONENT_Y);
      tempCS->prevPLT.curPLTSize[comID] = curLastPLTSize[comID];
      memcpy(tempCS->prevPLT.curPLT[i], curLastPLT[i], curLastPLTSize[comID] * sizeof(Pel));
    }
    EncTestMode currTestMode = m_modeCtrl->currTestMode();
    currTestMode.maxCostAllowed = maxCostAllowed;

    if (pps.getUseDQP() && partitioner.isSepTree(*tempCS) && isChroma( partitioner.chType ))
    {
      const Position chromaCentral(tempCS->area.Cb().chromaPos().offset(tempCS->area.Cb().chromaSize().width >> 1, tempCS->area.Cb().chromaSize().height >> 1));
      const Position lumaRefPos(chromaCentral.x << getComponentScaleX(COMPONENT_Cb, tempCS->area.chromaFormat), chromaCentral.y << getComponentScaleY(COMPONENT_Cb, tempCS->area.chromaFormat));
      const CodingStructure* baseCS = bestCS->picture->cs;
      const CodingUnit      *colLumaCu = baseCS->getCU(lumaRefPos, ChannelType::LUMA);

      if (colLumaCu)
      {
        currTestMode.qp = colLumaCu->qp;
      }
    }

#if SHARP_LUMA_DELTA_QP || ENABLE_QPA_SUB_CTU
    if (partitioner.currQgEnable() && (
        (m_pcEncCfg->getBIM()) ||
#if SHARP_LUMA_DELTA_QP
        (m_pcEncCfg->getLumaLevelToDeltaQPMapping().isEnabled()) ||
#endif
        (m_pcEncCfg->getSmoothQPReductionEnable()) ||
#if ENABLE_QPA_SUB_CTU
        (m_pcEncCfg->getUsePerceptQPA() && !m_pcEncCfg->getUseRateCtrl() && pps.getUseDQP())
#else
        false
#endif
      ))
    {
      if (currTestMode.qp >= 0)
      {
        updateLambda (&slice, currTestMode.qp,
 #if WCG_EXT && ER_CHROMA_QP_WCG_PPS
                      m_pcEncCfg->getWCGChromaQPControl().isEnabled(),
 #endif
                      CS::isDualITree (*tempCS) || (partitioner.currDepth == 0));
      }
    }
#endif

    if( currTestMode.type == ETM_INTER_ME )
    {
      if( ( currTestMode.opts & ETO_IMV ) != 0 )
      {
        const bool skipAltHpelIF = (currTestMode.getAmvrSearchMode() == EncTestMode::AmvrSearchMode::HALF_PEL)
                                   && (bestIntPelCost > 1.25 * bestCS->cost);
        if (!skipAltHpelIF)
        {
          tempCS->bestCS = bestCS;
          xCheckRDCostInterAmvr(tempCS, bestCS, partitioner, currTestMode, bestIntPelCost);
          tempCS->bestCS = nullptr;
          splitRdCostBest[CTU_LEVEL] = bestCS->cost;
          tempCS->splitRdCostBest = splitRdCostBest;
        }
      }
      else
      {
        tempCS->bestCS = bestCS;
        xCheckRDCostInter( tempCS, bestCS, partitioner, currTestMode );
        tempCS->bestCS = nullptr;
        splitRdCostBest[CTU_LEVEL] = bestCS->cost;
        tempCS->splitRdCostBest = splitRdCostBest;
      }

    }
    else if (currTestMode.type == ETM_HASH_INTER)
    {
      xCheckRDCostHashInter( tempCS, bestCS, partitioner, currTestMode );
      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
#if REUSE_CU_RESULTS
    else if( currTestMode.type == ETM_RECO_CACHED )
    {
      xReuseCachedResult( tempCS, bestCS, partitioner );
      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
#endif
    else if( currTestMode.type == ETM_MERGE_SKIP )
    {
      xCheckRDCostUnifiedMerge(tempCS, bestCS, partitioner, currTestMode);
      CodingUnit* cu = bestCS->getCU(partitioner.chType);
      if (cu)
      {
        cu->mmvdSkip = cu->skip == false ? false : cu->mmvdSkip;
      }
      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
    else if( currTestMode.type == ETM_INTRA )
    {
      if (slice.getSPS()->getUseColorTrans() && !CS::isDualITree(*tempCS))
      {
        bool skipSecColorSpace = false;
        skipSecColorSpace = xCheckRDCostIntra(tempCS, bestCS, partitioner, currTestMode, (m_pcEncCfg->getRGBFormatFlag() ? true : false));
        if ((m_pcEncCfg->getCostMode() == COST_LOSSLESS_CODING && slice.isLossless()) && !m_pcEncCfg->getRGBFormatFlag())
        {
          skipSecColorSpace = true;
        }
        if (!skipSecColorSpace && !tempCS->firstColorSpaceTestOnly)
        {
          xCheckRDCostIntra(tempCS, bestCS, partitioner, currTestMode, (m_pcEncCfg->getRGBFormatFlag() ? false : true));
        }

        if (!tempCS->firstColorSpaceTestOnly)
        {
          if (tempCS->tmpColorSpaceIntraCost[0] != MAX_DOUBLE && tempCS->tmpColorSpaceIntraCost[1] != MAX_DOUBLE)
          {
            double skipCostRatio = m_pcEncCfg->getRGBFormatFlag() ? 1.1 : 1.0;
            if (tempCS->tmpColorSpaceIntraCost[1] > (skipCostRatio*tempCS->tmpColorSpaceIntraCost[0]))
            {
              tempCS->firstColorSpaceTestOnly = bestCS->firstColorSpaceTestOnly = true;
            }
          }
        }
        else
        {
          CHECK(tempCS->tmpColorSpaceIntraCost[1] != MAX_DOUBLE, "the RD test of the second color space should be skipped");
        }
      }
      else
      {
        xCheckRDCostIntra(tempCS, bestCS, partitioner, currTestMode, false);
      }
      if (partitioner.currQtDepth == 1 && partitioner.currBtDepth == 0 && partitioner.currArea().lwidth() == 64
          && partitioner.currArea().lheight() == 64)
      {
        if ((partitioner.chType == ChannelType::LUMA)
            && ((partitioner.currArea().Y().x + 63 < bestCS->picture->lwidth())
                && (partitioner.currArea().Y().y + 63 < bestCS->picture->lheight())))
        {      
          m_modeCtrl->setNoSplitIntraCost(bestCS->cost);
        }
      }

      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
    else if (currTestMode.type == ETM_PALETTE)
    {
      xCheckPLT( tempCS, bestCS, partitioner, currTestMode );
      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
    else if (currTestMode.type == ETM_IBC)
    {
      xCheckRDCostIBCMode(tempCS, bestCS, partitioner, currTestMode);
      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
    else if (currTestMode.type == ETM_IBC_MERGE)
    {
      xCheckRDCostIBCModeMerge2Nx2N(tempCS, bestCS, partitioner, currTestMode);
      splitRdCostBest[CTU_LEVEL] = bestCS->cost;
      tempCS->splitRdCostBest = splitRdCostBest;
    }
    else if( isModeSplit( currTestMode ) )
    {
      if (bestCS->cus.size() != 0)
      {
        splitmode = bestCS->cus[0]->splitSeries;
      }
      assert( partitioner.modeType == tempCS->modeType );
      int signalModeConsVal = tempCS->signalModeCons( getPartSplit( currTestMode ), partitioner, modeTypeParent );
      int numRoundRdo = signalModeConsVal == LDT_MODE_TYPE_SIGNAL ? 2 : 1;
      bool skipInterPass = false;
      for( int i = 0; i < numRoundRdo; i++ )
      {
        //change cons modes
        if( signalModeConsVal == LDT_MODE_TYPE_SIGNAL )
        {
          CHECK( numRoundRdo != 2, "numRoundRdo shall be 2 - [LDT_MODE_TYPE_SIGNAL]" );
          tempCS->modeType = partitioner.modeType = (i == 0) ? MODE_TYPE_INTER : MODE_TYPE_INTRA;
        }
        else if( signalModeConsVal == LDT_MODE_TYPE_INFER )
        {
          CHECK( numRoundRdo != 1, "numRoundRdo shall be 1 - [LDT_MODE_TYPE_INFER]" );
          tempCS->modeType = partitioner.modeType = MODE_TYPE_INTRA;
        }
        else if( signalModeConsVal == LDT_MODE_TYPE_INHERIT )
        {
          CHECK( numRoundRdo != 1, "numRoundRdo shall be 1 - [LDT_MODE_TYPE_INHERIT]" );
          tempCS->modeType = partitioner.modeType = modeTypeParent;
        }

        //for lite intra encoding fast algorithm, set the status to save inter coding info
        if( modeTypeParent == MODE_TYPE_ALL && tempCS->modeType == MODE_TYPE_INTER )
        {
          m_pcIntraSearch->setSaveCuCostInSCIPU( true );
          m_pcIntraSearch->setNumCuInSCIPU( 0 );
        }
        else if( modeTypeParent == MODE_TYPE_ALL && tempCS->modeType != MODE_TYPE_INTER )
        {
          m_pcIntraSearch->setSaveCuCostInSCIPU( false );
          if( tempCS->modeType == MODE_TYPE_ALL )
          {
            m_pcIntraSearch->setNumCuInSCIPU( 0 );
          }
        }

        xCheckModeSplit( tempCS, bestCS, partitioner, currTestMode, modeTypeParent, skipInterPass, splitRdCostBest );
        tempCS->splitRdCostBest = splitRdCostBest;
        //recover cons modes
        tempCS->modeType = partitioner.modeType = modeTypeParent;
        tempCS->treeType = partitioner.treeType = treeTypeParent;
        partitioner.chType = chTypeParent;
        if( modeTypeParent == MODE_TYPE_ALL )
        {
          m_pcIntraSearch->setSaveCuCostInSCIPU( false );
          if( numRoundRdo == 2 && tempCS->modeType == MODE_TYPE_INTRA )
          {
            m_pcIntraSearch->initCuAreaCostInSCIPU();
          }
        }
        if( skipInterPass )
        {
          break;
        }
      }
#if GDR_ENABLED
      if (bestCS->cus.size() > 0 && splitmode != bestCS->cus[0]->splitSeries)
#else
      if (splitmode != bestCS->cus[0]->splitSeries)
#endif
      {
        splitmode = bestCS->cus[0]->splitSeries;
        const CodingUnit&     cu = *bestCS->cus.front();
        cu.cs->prevPLT = bestCS->prevPLT;
        for (int i = compBegin; i < (compBegin + numComp); i++)
        {
          ComponentID comID = jointPLT ? (ComponentID)compBegin : ((i > 0) ? COMPONENT_Cb : COMPONENT_Y);
          bestLastPLTSize[comID] = bestCS->cus[0]->cs->prevPLT.curPLTSize[comID];
          memcpy(bestLastPLT[i], bestCS->cus[0]->cs->prevPLT.curPLT[i], bestCS->cus[0]->cs->prevPLT.curPLTSize[comID] * sizeof(Pel));
        }
      }
    }
    else
    {
      THROW( "Don't know how to handle mode: type = " << currTestMode.type << ", options = " << currTestMode.opts );
    }
  } while( m_modeCtrl->nextMode( *tempCS, partitioner ) );


  //////////////////////////////////////////////////////////////////////////
  // Finishing CU
  if( tempCS->cost == MAX_DOUBLE && bestCS->cost == MAX_DOUBLE )
  {
    //although some coding modes were planned to be tried in RDO, no coding mode actually finished encoding due to early termination
    //thus tempCS->cost and bestCS->cost are both MAX_DOUBLE; in this case, skip the following process for normal case
    m_modeCtrl->finishCULevel( partitioner );
    return;
  }

  // set context states
  m_CABACEstimator->getCtx() = m_CurrCtx->best;

  // QP from last processed CU for further processing
  //copy the qp of the last non-chroma CU
  int numCUInThisNode = (int)bestCS->cus.size();
  if (numCUInThisNode > 1 && bestCS->cus.back()->chType == ChannelType::CHROMA && !CS::isDualITree(*bestCS))
  {
    CHECK(bestCS->cus[numCUInThisNode - 2]->chType != ChannelType::LUMA, "wrong chType");
    bestCS->prevQP[partitioner.chType] = bestCS->cus[numCUInThisNode - 2]->qp;
  }
  else
  {
    bestCS->prevQP[partitioner.chType] = bestCS->cus.back()->qp;
  }
  if ((!slice.isIntra() || slice.getSPS()->getIBCFlag()) && isLuma(partitioner.chType) && bestCS->cus.size() == 1
      && (CU::isInter(*bestCS->cus.back()) || CU::isIBC(*bestCS->cus.back()))
      && bestCS->area.Y() == (*bestCS->cus.back()).Y())
  {
    const CodingUnit&     cu = *bestCS->cus.front();

    CU::saveMotionForHmvp(cu);
  }
  bestCS->picture->getPredBuf(currCsArea).copyFrom(bestCS->getPredBuf(currCsArea));
  bestCS->picture->getRecoBuf( currCsArea ).copyFrom( bestCS->getRecoBuf( currCsArea ) );
  m_modeCtrl->finishCULevel( partitioner );
  if( m_pcIntraSearch->getSaveCuCostInSCIPU() && bestCS->cus.size() == 1 )
  {
    m_pcIntraSearch->saveCuAreaCostInSCIPU( Area( partitioner.currArea().lumaPos(), partitioner.currArea().lumaSize() ), bestCS->cost );
  }

  if (bestCS->cus.size() == 1) // no partition
  {