IntraSearch.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) 2010-2019, 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     EncSearch.cpp
 *  \brief    encoder intra search class
 */

#include "IntraSearch.h"

#include "EncModeCtrl.h"

#include "CommonLib/CommonDef.h"
#include "CommonLib/Rom.h"
#include "CommonLib/Picture.h"
#include "CommonLib/UnitTools.h"

#include "CommonLib/dtrace_next.h"
#include "CommonLib/dtrace_buffer.h"

#include <math.h>
#include <limits>
 //! \ingroup EncoderLib
 //! \{
#define PLTCtx(c) SubCtx( Ctx::Palette, c )
IntraSearch::IntraSearch()
  : m_pSplitCS      (nullptr)
  , m_pFullCS       (nullptr)
  , m_pBestCS       (nullptr)
  , m_pcEncCfg      (nullptr)
  , m_pcTrQuant     (nullptr)
  , m_pcRdCost      (nullptr)
  , m_pcReshape     (nullptr)
  , m_CABACEstimator(nullptr)
  , m_CtxCache      (nullptr)
  , m_isInitialized (false)
{
  for( uint32_t ch = 0; ch < MAX_NUM_TBLOCKS; ch++ )
  {
    m_pSharedPredTransformSkip[ch] = nullptr;
  }
}


void IntraSearch::destroy()
{
  CHECK( !m_isInitialized, "Not initialized" );

  if( m_pcEncCfg )
  {
    const uint32_t uiNumLayersToAllocateSplit = 1;
    const uint32_t uiNumLayersToAllocateFull  = 1;
    const int uiNumSaveLayersToAllocate = 2;

    for( uint32_t layer = 0; layer < uiNumSaveLayersToAllocate; layer++ )
    {
      m_pSaveCS[layer]->destroy();
      delete m_pSaveCS[layer];
    }

    uint32_t numWidths  = gp_sizeIdxInfo->numWidths();
    uint32_t numHeights = gp_sizeIdxInfo->numHeights();

    for( uint32_t width = 0; width < numWidths; width++ )
    {
      for( uint32_t height = 0; height < numHeights; height++ )
      {
        if( gp_sizeIdxInfo->isCuSize( gp_sizeIdxInfo->sizeFrom( width ) ) && gp_sizeIdxInfo->isCuSize( gp_sizeIdxInfo->sizeFrom( height ) ) )
        {
          for( uint32_t layer = 0; layer < uiNumLayersToAllocateSplit; layer++ )
          {
            m_pSplitCS[width][height][layer]->destroy();

            delete m_pSplitCS[width][height][layer];
          }

          for( uint32_t layer = 0; layer < uiNumLayersToAllocateFull; layer++ )
          {
            m_pFullCS[width][height][layer]->destroy();

            delete m_pFullCS[width][height][layer];
          }

          delete[] m_pSplitCS[width][height];
          delete[] m_pFullCS [width][height];

          m_pBestCS[width][height]->destroy();
          m_pTempCS[width][height]->destroy();

          delete m_pTempCS[width][height];
          delete m_pBestCS[width][height];
        }
      }

      delete[] m_pSplitCS[width];
      delete[] m_pFullCS [width];

      delete[] m_pTempCS[width];
      delete[] m_pBestCS[width];
    }

    delete[] m_pSplitCS;
    delete[] m_pFullCS;

    delete[] m_pBestCS;
    delete[] m_pTempCS;

    delete[] m_pSaveCS;
  }

  m_pSplitCS = m_pFullCS = nullptr;

  m_pBestCS = m_pTempCS = nullptr;

  m_pSaveCS = nullptr;

  for( uint32_t ch = 0; ch < MAX_NUM_TBLOCKS; ch++ )
  {
    delete[] m_pSharedPredTransformSkip[ch];
    m_pSharedPredTransformSkip[ch] = nullptr;
  }

  m_tmpStorageLCU.destroy();
  m_isInitialized = false;
}

IntraSearch::~IntraSearch()
{
  if( m_isInitialized )
  {
    destroy();
  }
}

void IntraSearch::init( EncCfg*        pcEncCfg,
                        TrQuant*       pcTrQuant,
                        RdCost*        pcRdCost,
                        CABACWriter*   CABACEstimator,
                        CtxCache*      ctxCache,
                        const uint32_t     maxCUWidth,
                        const uint32_t     maxCUHeight,
                        const uint32_t     maxTotalCUDepth
                       , EncReshape*   pcReshape
)
{
  CHECK(m_isInitialized, "Already initialized");
  m_pcEncCfg                     = pcEncCfg;
  m_pcTrQuant                    = pcTrQuant;
  m_pcRdCost                     = pcRdCost;
  m_CABACEstimator               = CABACEstimator;
  m_CtxCache                     = ctxCache;
  m_pcReshape                    = pcReshape;

  const ChromaFormat cform = pcEncCfg->getChromaFormatIdc();

  IntraPrediction::init( cform, pcEncCfg->getBitDepth( CHANNEL_TYPE_LUMA ) );
  m_tmpStorageLCU.create(UnitArea(cform, Area(0, 0, MAX_CU_SIZE, MAX_CU_SIZE)));

  for( uint32_t ch = 0; ch < MAX_NUM_TBLOCKS; ch++ )
  {
    m_pSharedPredTransformSkip[ch] = new Pel[MAX_CU_SIZE * MAX_CU_SIZE];
  }

  uint32_t numWidths  = gp_sizeIdxInfo->numWidths();
  uint32_t numHeights = gp_sizeIdxInfo->numHeights();

  const uint32_t uiNumLayersToAllocateSplit = 1;
  const uint32_t uiNumLayersToAllocateFull  = 1;

  m_pBestCS = new CodingStructure**[numWidths];
  m_pTempCS = new CodingStructure**[numWidths];

  m_pFullCS  = new CodingStructure***[numWidths];
  m_pSplitCS = new CodingStructure***[numWidths];

  for( uint32_t width = 0; width < numWidths; width++ )
  {
    m_pBestCS[width] = new CodingStructure*[numHeights];
    m_pTempCS[width] = new CodingStructure*[numHeights];

    m_pFullCS [width] = new CodingStructure**[numHeights];
    m_pSplitCS[width] = new CodingStructure**[numHeights];

    for( uint32_t height = 0; height < numHeights; height++ )
    {
      if(  gp_sizeIdxInfo->isCuSize( gp_sizeIdxInfo->sizeFrom( width ) ) && gp_sizeIdxInfo->isCuSize( gp_sizeIdxInfo->sizeFrom( height ) ) )
      {
        m_pBestCS[width][height] = new CodingStructure( m_unitCache.cuCache, m_unitCache.puCache, m_unitCache.tuCache );
        m_pTempCS[width][height] = new CodingStructure( m_unitCache.cuCache, m_unitCache.puCache, m_unitCache.tuCache );

        m_pBestCS[width][height]->create( m_pcEncCfg->getChromaFormatIdc(), Area( 0, 0, gp_sizeIdxInfo->sizeFrom( width ), gp_sizeIdxInfo->sizeFrom( height ) ), false );
        m_pTempCS[width][height]->create( m_pcEncCfg->getChromaFormatIdc(), Area( 0, 0, gp_sizeIdxInfo->sizeFrom( width ), gp_sizeIdxInfo->sizeFrom( height ) ), false );
        m_pFullCS [width][height] = new CodingStructure*[uiNumLayersToAllocateFull];
        m_pSplitCS[width][height] = new CodingStructure*[uiNumLayersToAllocateSplit];

        for( uint32_t layer = 0; layer < uiNumLayersToAllocateFull; layer++ )
        {
          m_pFullCS [width][height][layer] = new CodingStructure( m_unitCache.cuCache, m_unitCache.puCache, m_unitCache.tuCache );

          m_pFullCS [width][height][layer]->create( m_pcEncCfg->getChromaFormatIdc(), Area( 0, 0, gp_sizeIdxInfo->sizeFrom( width ), gp_sizeIdxInfo->sizeFrom( height ) ), false );
        }

        for( uint32_t layer = 0; layer < uiNumLayersToAllocateSplit; layer++ )
        {
          m_pSplitCS[width][height][layer] = new CodingStructure( m_unitCache.cuCache, m_unitCache.puCache, m_unitCache.tuCache );

          m_pSplitCS[width][height][layer]->create( m_pcEncCfg->getChromaFormatIdc(), Area( 0, 0, gp_sizeIdxInfo->sizeFrom( width ), gp_sizeIdxInfo->sizeFrom( height ) ), false );
        }
      }
      else
      {
        m_pBestCS[width][height] = nullptr;
        m_pTempCS[width][height] = nullptr;

        m_pFullCS [width][height] = nullptr;
        m_pSplitCS[width][height] = nullptr;
      }
    }
  }

  const int uiNumSaveLayersToAllocate = 2;

  m_pSaveCS = new CodingStructure*[uiNumSaveLayersToAllocate];

  for( uint32_t depth = 0; depth < uiNumSaveLayersToAllocate; depth++ )
  {
    m_pSaveCS[depth] = new CodingStructure( m_unitCache.cuCache, m_unitCache.puCache, m_unitCache.tuCache );
    m_pSaveCS[depth]->create( UnitArea( cform, Area( 0, 0, maxCUWidth, maxCUHeight ) ), false );
  }

  m_isInitialized = true;
}


//////////////////////////////////////////////////////////////////////////
// INTRA PREDICTION
//////////////////////////////////////////////////////////////////////////
static constexpr double COST_UNKNOWN = -65536.0;

double IntraSearch::findInterCUCost( CodingUnit &cu )
{
  if( cu.isConsIntra() && !cu.slice->isIntra() )
  {
    //search corresponding inter CU cost
    for( int i = 0; i < m_numCuInSCIPU; i++ )
    {
      if( cu.lumaPos() == m_cuAreaInSCIPU[i].pos() && cu.lumaSize() == m_cuAreaInSCIPU[i].size() )
      {
        return m_cuCostInSCIPU[i];
      }
    }
  }
  return COST_UNKNOWN;
}

bool IntraSearch::estIntraPredLumaQT( CodingUnit &cu, Partitioner &partitioner, const double bestCostSoFar, bool mtsCheckRangeFlag, int mtsFirstCheckId, int mtsLastCheckId, bool moreProbMTSIdxFirst )
{
  CodingStructure       &cs            = *cu.cs;
  const SPS             &sps           = *cs.sps;
  const uint32_t             uiWidthBit    = floorLog2(partitioner.currArea().lwidth() );
  const uint32_t             uiHeightBit   =                   floorLog2(partitioner.currArea().lheight());

  // Lambda calculation at equivalent Qp of 4 is recommended because at that Qp, the quantization divisor is 1.
  const double sqrtLambdaForFirstPass = m_pcRdCost->getMotionLambda(cu.transQuantBypass) * FRAC_BITS_SCALE;

  //===== loop over partitions =====

  const TempCtx ctxStart          ( m_CtxCache, m_CABACEstimator->getCtx() );
  const TempCtx ctxStartMipFlag    ( m_CtxCache, SubCtx( Ctx::MipFlag,          m_CABACEstimator->getCtx() ) );
  const TempCtx ctxStartIspMode    ( m_CtxCache, SubCtx( Ctx::ISPMode,          m_CABACEstimator->getCtx() ) );
  const TempCtx ctxStartPlanarFlag ( m_CtxCache, SubCtx( Ctx::IntraLumaPlanarFlag, m_CABACEstimator->getCtx() ) );
  const TempCtx ctxStartIntraMode(m_CtxCache, SubCtx(Ctx::IntraLumaMpmFlag, m_CABACEstimator->getCtx()));
  const TempCtx ctxStartMrlIdx      ( m_CtxCache, SubCtx( Ctx::MultiRefLineIdx,        m_CABACEstimator->getCtx() ) );

  CHECK( !cu.firstPU, "CU has no PUs" );
  const bool keepResi   = cs.pps->getPpsRangeExtension().getCrossComponentPredictionEnabledFlag() || KEEP_PRED_AND_RESI_SIGNALS;

  // variables for saving fast intra modes scan results across multiple LFNST passes
  bool LFNSTLoadFlag = sps.getUseLFNST() && cu.lfnstIdx != 0;
  bool LFNSTSaveFlag = sps.getUseLFNST() && cu.lfnstIdx == 0;

  LFNSTSaveFlag &= sps.getUseIntraMTS() ? cu.mtsFlag == 0 : true;

  const uint32_t lfnstIdx = cu.lfnstIdx;
  double costInterCU = findInterCUCost( cu );


  const int width  = partitioner.currArea().lwidth();
  const int height = partitioner.currArea().lheight();

  // Marking MTS usage for faster MTS
  // 0: MTS is either not applicable for current CU (cuWidth > MTS_INTRA_MAX_CU_SIZE or cuHeight > MTS_INTRA_MAX_CU_SIZE), not active in the config file or the fast decision algorithm is not used in this case
  // 1: MTS fast algorithm can be applied for the current CU, and the DCT2 is being checked
  // 2: MTS is being checked for current CU. Stored results of DCT2 can be utilized for speedup
  uint8_t mtsUsageFlag = 0;
  const int maxSizeEMT = MTS_INTRA_MAX_CU_SIZE;
  if( width <= maxSizeEMT && height <= maxSizeEMT && sps.getUseIntraMTS() )
  {
    mtsUsageFlag = ( sps.getUseLFNST() && cu.mtsFlag == 1 ) ? 2 : 1;
  }

  if( width * height < 64 && !m_pcEncCfg->getUseFastLFNST() )
  {
    mtsUsageFlag = 0;
  }

  double bestCurrentCost = bestCostSoFar;
  bool testISP = sps.getUseISP() && cu.mtsFlag == 0 && cu.lfnstIdx == 0 && CU::canUseISP( width, height, cu.cs->sps->getMaxTbSize() );
  if( testISP )
  {
    //reset the variables used for the tests
    m_ispCandListHor.clear();
    m_ispCandListVer.clear();
    m_regIntraRDListWithCosts.clear();
    m_ispTestedModes.clear();
    //save the number of subpartitions
    m_ispTestedModes.numTotalParts[0] = (int)height >> floorLog2(CU::getISPSplitDim(width, height, TU_1D_HORZ_SPLIT));
    m_ispTestedModes.numTotalParts[1] = (int)width >> floorLog2(CU::getISPSplitDim(width, height, TU_1D_VERT_SPLIT));
  }

  const bool testBDPCM = sps.getBDPCMEnabledFlag() && CU::bdpcmAllowed( cu, ComponentID( partitioner.chType ) ) && cu.mtsFlag == 0 && cu.lfnstIdx == 0;
  static_vector<ModeInfo, FAST_UDI_MAX_RDMODE_NUM> uiHadModeList;
  static_vector<double, FAST_UDI_MAX_RDMODE_NUM> CandCostList;
  static_vector<double, FAST_UDI_MAX_RDMODE_NUM> CandHadList;


  auto &pu = *cu.firstPU;
  bool validReturn = false;
  {
    CandHadList.clear();
    CandCostList.clear();
    uiHadModeList.clear();

    CHECK(pu.cu != &cu, "PU is not contained in the CU");

    //===== determine set of modes to be tested (using prediction signal only) =====
    int numModesAvailable = NUM_LUMA_MODE; // total number of Intra modes
    const bool fastMip    = sps.getUseMIP() && m_pcEncCfg->getUseFastMIP();
    const bool mipAllowed = sps.getUseMIP() && isLuma(partitioner.chType) && pu.lwidth() <= cu.cs->sps->getMaxTbSize() && pu.lheight() <= cu.cs->sps->getMaxTbSize() && ((cu.lfnstIdx == 0) || allowLfnstWithMip(cu.firstPU->lumaSize()));
    const bool testMip    = mipAllowed && mipModesAvailable(pu.Y());

    static_vector<ModeInfo, FAST_UDI_MAX_RDMODE_NUM> uiRdModeList;

    int numModesForFullRD = 3;
    numModesForFullRD = g_aucIntraModeNumFast_UseMPM_2D[uiWidthBit - MIN_CU_LOG2][uiHeightBit - MIN_CU_LOG2];

#if INTRA_FULL_SEARCH
    numModesForFullRD = numModesAvailable;
#endif

    if( mtsUsageFlag != 2 )
    {
      // this should always be true
      CHECK( !pu.Y().valid(), "PU is not valid" );
      bool isFirstLineOfCtu = (((pu.block(COMPONENT_Y).y)&((pu.cs->sps)->getMaxCUWidth() - 1)) == 0);
      int numOfPassesExtendRef = (isFirstLineOfCtu ? 1 : MRL_NUM_REF_LINES);
      pu.multiRefIdx = 0;

      if( numModesForFullRD != numModesAvailable )
      {
        CHECK( numModesForFullRD >= numModesAvailable, "Too many modes for full RD search" );

        const CompArea &area = pu.Y();

        PelBuf piOrg         = cs.getOrgBuf(area);
        PelBuf piPred        = cs.getPredBuf(area);

        DistParam distParamSad;
        DistParam distParamHad;
        if (cu.slice->getLmcsEnabledFlag() && m_pcReshape->getCTUFlag())
        {
          CompArea      tmpArea(COMPONENT_Y, area.chromaFormat, Position(0, 0), area.size());
          PelBuf tmpOrg = m_tmpStorageLCU.getBuf(tmpArea);
          tmpOrg.copyFrom(piOrg);
          tmpOrg.rspSignal(m_pcReshape->getFwdLUT());
          m_pcRdCost->setDistParam(distParamSad, tmpOrg, piPred, sps.getBitDepth(CHANNEL_TYPE_LUMA), COMPONENT_Y, false); // Use SAD cost
          m_pcRdCost->setDistParam(distParamHad, tmpOrg, piPred, sps.getBitDepth(CHANNEL_TYPE_LUMA), COMPONENT_Y,  true); // Use HAD (SATD) cost
        }
        else
        {
          m_pcRdCost->setDistParam(distParamSad, piOrg, piPred, sps.getBitDepth(CHANNEL_TYPE_LUMA), COMPONENT_Y, false); // Use SAD cost
          m_pcRdCost->setDistParam(distParamHad, piOrg, piPred, sps.getBitDepth(CHANNEL_TYPE_LUMA), COMPONENT_Y,  true); // Use HAD (SATD) cost
        }

        distParamSad.applyWeight = false;
        distParamHad.applyWeight = false;

        if( testMip)
        {
          numModesForFullRD += fastMip? std::max(numModesForFullRD, floorLog2(std::min(pu.lwidth(), pu.lheight())) - 1) : numModesForFullRD;
        }
        const int numHadCand = (testMip ? 2 : 1) * 3;

        //*** Derive (regular) candidates using Hadamard
        cu.mipFlag = false;

        //===== init pattern for luma prediction =====
        initIntraPatternChType(cu, pu.Y(), true);
        bool bSatdChecked[NUM_INTRA_MODE];
        memset( bSatdChecked, 0, sizeof( bSatdChecked ) );

        if( !LFNSTLoadFlag )
        {
          for( int modeIdx = 0; modeIdx < numModesAvailable; modeIdx++ )
          {
            uint32_t       uiMode = modeIdx;
            Distortion minSadHad = 0;

            // Skip checking extended Angular modes in the first round of SATD
            if( uiMode > DC_IDX && ( uiMode & 1 ) )
            {
              continue;
            }

            bSatdChecked[uiMode] = true;

            pu.intraDir[0] = modeIdx;

            initPredIntraParams(pu, pu.Y(), sps);
            if( useDPCMForFirstPassIntraEstimation( pu, uiMode ) )
            {
              encPredIntraDPCM( COMPONENT_Y, piOrg, piPred, uiMode );
            }
            else
            {
              predIntraAng( COMPONENT_Y, piPred, pu);
            }
            // Use the min between SAD and HAD as the cost criterion
            // SAD is scaled by 2 to align with the scaling of HAD
            minSadHad += std::min(distParamSad.distFunc(distParamSad)*2, distParamHad.distFunc(distParamHad));

            // NB xFracModeBitsIntra will not affect the mode for chroma that may have already been pre-estimated.
            m_CABACEstimator->getCtx() = SubCtx( Ctx::MipFlag, ctxStartMipFlag );
            m_CABACEstimator->getCtx() = SubCtx( Ctx::ISPMode, ctxStartIspMode );
            m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaPlanarFlag, ctxStartPlanarFlag);
            m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaMpmFlag, ctxStartIntraMode);
            m_CABACEstimator->getCtx() = SubCtx( Ctx::MultiRefLineIdx, ctxStartMrlIdx );

            uint64_t fracModeBits = xFracModeBitsIntra(pu, uiMode, CHANNEL_TYPE_LUMA);

            double cost = ( double ) minSadHad + (double)fracModeBits * sqrtLambdaForFirstPass;

            DTRACE(g_trace_ctx, D_INTRA_COST, "IntraHAD: %u, %llu, %f (%d)\n", minSadHad, fracModeBits, cost, uiMode);

            updateCandList( ModeInfo(false, 0, NOT_INTRA_SUBPARTITIONS, uiMode), cost,          uiRdModeList,  CandCostList, numModesForFullRD );
            updateCandList( ModeInfo(false, 0, NOT_INTRA_SUBPARTITIONS, uiMode), (double)minSadHad, uiHadModeList, CandHadList,  numHadCand );
          }
          if( !sps.getUseMIP() && LFNSTSaveFlag )
          {
            // save found best modes
            m_uiSavedNumRdModesLFNST   = numModesForFullRD;
            m_uiSavedRdModeListLFNST   = uiRdModeList;
            m_dSavedModeCostLFNST      = CandCostList;
            // PBINTRA fast
            m_uiSavedHadModeListLFNST  = uiHadModeList;
            m_dSavedHadListLFNST       = CandHadList;
            LFNSTSaveFlag              = false;
          }
        } // NSSTFlag
        if( !sps.getUseMIP() && LFNSTLoadFlag )
        {
          // restore saved modes
          numModesForFullRD = m_uiSavedNumRdModesLFNST;
          uiRdModeList      = m_uiSavedRdModeListLFNST;
          CandCostList      = m_dSavedModeCostLFNST;
          // PBINTRA fast
          uiHadModeList     = m_uiSavedHadModeListLFNST;
          CandHadList       = m_dSavedHadListLFNST;
        } // !LFNSTFlag

        if (!(sps.getUseMIP() && LFNSTLoadFlag))
        {
        static_vector<ModeInfo, FAST_UDI_MAX_RDMODE_NUM> parentCandList = uiRdModeList;

        // Second round of SATD for extended Angular modes
        for (int modeIdx = 0; modeIdx < numModesForFullRD; modeIdx++)
        {
          unsigned parentMode = parentCandList[modeIdx].modeId;
          if (parentMode > (DC_IDX + 1) && parentMode < (NUM_LUMA_MODE - 1))
          {
            for (int subModeIdx = -1; subModeIdx <= 1; subModeIdx += 2)
            {
              unsigned mode = parentMode + subModeIdx;


              if (!bSatdChecked[mode])
              {
                pu.intraDir[0] = mode;

                initPredIntraParams(pu, pu.Y(), sps);
                if (useDPCMForFirstPassIntraEstimation(pu, mode))
                {
                  encPredIntraDPCM(COMPONENT_Y, piOrg, piPred, mode);
                }
                else
                {
                  predIntraAng(COMPONENT_Y, piPred, pu );
                }

                // Use the min between SAD and SATD as the cost criterion
                // SAD is scaled by 2 to align with the scaling of HAD
                Distortion minSadHad = std::min(distParamSad.distFunc(distParamSad)*2, distParamHad.distFunc(distParamHad));

                // NB xFracModeBitsIntra will not affect the mode for chroma that may have already been pre-estimated.
                m_CABACEstimator->getCtx() = SubCtx( Ctx::MipFlag, ctxStartMipFlag );
                m_CABACEstimator->getCtx() = SubCtx( Ctx::ISPMode, ctxStartIspMode );
                m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaPlanarFlag, ctxStartPlanarFlag);
                m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaMpmFlag, ctxStartIntraMode);
                m_CABACEstimator->getCtx() = SubCtx( Ctx::MultiRefLineIdx, ctxStartMrlIdx );

                uint64_t fracModeBits = xFracModeBitsIntra(pu, mode, CHANNEL_TYPE_LUMA);

                double cost = (double) minSadHad + (double) fracModeBits * sqrtLambdaForFirstPass;

                updateCandList( ModeInfo( false, 0, NOT_INTRA_SUBPARTITIONS, mode ), cost,        uiRdModeList,  CandCostList, numModesForFullRD );
                updateCandList( ModeInfo( false, 0, NOT_INTRA_SUBPARTITIONS, mode ), (double)minSadHad, uiHadModeList, CandHadList,  numHadCand );

                bSatdChecked[mode] = true;
              }
            }
          }
        }
        if ( testISP )
        {
          // we save the regular intra modes list
          m_ispCandListHor = uiRdModeList;
        }
        pu.multiRefIdx = 1;
        const int  numMPMs = NUM_MOST_PROBABLE_MODES;
        unsigned  multiRefMPM [numMPMs];
        PU::getIntraMPMs(pu, multiRefMPM);
        for (int mRefNum = 1; mRefNum < numOfPassesExtendRef; mRefNum++)
        {
          int multiRefIdx = MULTI_REF_LINE_IDX[mRefNum];

          pu.multiRefIdx = multiRefIdx;
          {
            initIntraPatternChType(cu, pu.Y(), true);
          }
          for (int x = 1; x < numMPMs; x++)
          {
            uint32_t mode = multiRefMPM[x];
            {
              pu.intraDir[0] = mode;
              initPredIntraParams(pu, pu.Y(), sps);

              if (useDPCMForFirstPassIntraEstimation(pu, mode))
              {
                encPredIntraDPCM(COMPONENT_Y, piOrg, piPred, mode);
              }
              else
              {
                predIntraAng(COMPONENT_Y, piPred, pu);
              }

              // Use the min between SAD and SATD as the cost criterion
              // SAD is scaled by 2 to align with the scaling of HAD
              Distortion minSadHad = std::min(distParamSad.distFunc(distParamSad)*2, distParamHad.distFunc(distParamHad));

              // NB xFracModeBitsIntra will not affect the mode for chroma that may have already been pre-estimated.
              m_CABACEstimator->getCtx() = SubCtx( Ctx::MipFlag, ctxStartMipFlag );
              m_CABACEstimator->getCtx() = SubCtx( Ctx::ISPMode, ctxStartIspMode );
              m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaPlanarFlag, ctxStartPlanarFlag);
              m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaMpmFlag, ctxStartIntraMode);
              m_CABACEstimator->getCtx() = SubCtx( Ctx::MultiRefLineIdx, ctxStartMrlIdx );

              uint64_t fracModeBits = xFracModeBitsIntra(pu, mode, CHANNEL_TYPE_LUMA);

              double cost = (double)minSadHad + (double)fracModeBits * sqrtLambdaForFirstPass;
              updateCandList( ModeInfo( false, multiRefIdx, NOT_INTRA_SUBPARTITIONS, mode ), cost,        uiRdModeList,  CandCostList, numModesForFullRD );
              updateCandList( ModeInfo( false, multiRefIdx, NOT_INTRA_SUBPARTITIONS, mode ), (double)minSadHad, uiHadModeList, CandHadList,  numHadCand );
            }
          }
        }
        CHECKD( uiRdModeList.size() != numModesForFullRD, "Error: RD mode list size" );

        if (LFNSTSaveFlag && testMip && !allowLfnstWithMip(cu.firstPU->lumaSize())) // save a different set for the next run
        {
          // save found best modes
          m_uiSavedRdModeListLFNST = uiRdModeList;
          m_dSavedModeCostLFNST = CandCostList;
          // PBINTRA fast
          m_uiSavedHadModeListLFNST = uiHadModeList;
          m_dSavedHadListLFNST = CandHadList;
          m_uiSavedNumRdModesLFNST = g_aucIntraModeNumFast_UseMPM_2D[uiWidthBit - MIN_CU_LOG2][uiHeightBit - MIN_CU_LOG2];
          m_uiSavedRdModeListLFNST.resize(m_uiSavedNumRdModesLFNST);
          m_dSavedModeCostLFNST.resize(m_uiSavedNumRdModesLFNST);
          // PBINTRA fast
          m_uiSavedHadModeListLFNST.resize(3);
          m_dSavedHadListLFNST.resize(3);
          LFNSTSaveFlag = false;
        }
          //*** Derive MIP candidates using Hadamard
          if (testMip)
          {
            cu.mipFlag = true;
            pu.multiRefIdx = 0;

            double mipHadCost[MAX_NUM_MIP_MODE] = { MAX_DOUBLE };

            initIntraPatternChType(cu, pu.Y());
            initIntraMip( pu );

            for (uint32_t uiMode = 0; uiMode < getNumModesMip(pu.Y()); uiMode++)
            {
              pu.intraDir[CHANNEL_TYPE_LUMA] = uiMode;
              predIntraMip(COMPONENT_Y, piPred, pu);

              // Use the min between SAD and HAD as the cost criterion
              // SAD is scaled by 2 to align with the scaling of HAD
              Distortion minSadHad = std::min(distParamSad.distFunc(distParamSad)*2, distParamHad.distFunc(distParamHad));

              m_CABACEstimator->getCtx() = SubCtx( Ctx::MipFlag, ctxStartMipFlag );

              uint64_t fracModeBits = xFracModeBitsIntra(pu, uiMode, CHANNEL_TYPE_LUMA);

              double cost = double(minSadHad) + double(fracModeBits) * sqrtLambdaForFirstPass;
              mipHadCost[uiMode] = cost;
              DTRACE(g_trace_ctx, D_INTRA_COST, "IntraMIP: %u, %llu, %f (%d)\n", minSadHad, fracModeBits, cost, uiMode);

              updateCandList(ModeInfo(true, 0, NOT_INTRA_SUBPARTITIONS, uiMode), cost, uiRdModeList, CandCostList, numModesForFullRD + 1);
              updateCandList(ModeInfo(true, 0, NOT_INTRA_SUBPARTITIONS, uiMode), 0.8*double(minSadHad), uiHadModeList, CandHadList, numHadCand);
            }

            const double thresholdHadCost = 1.0 + 1.4 / sqrt((double)(pu.lwidth()*pu.lheight()));
            reduceHadCandList(uiRdModeList, CandCostList, numModesForFullRD, thresholdHadCost, mipHadCost, pu, fastMip);
          }
          if ( sps.getUseMIP() && LFNSTSaveFlag)
          {
            // save found best modes
            m_uiSavedNumRdModesLFNST = numModesForFullRD;
            m_uiSavedRdModeListLFNST = uiRdModeList;
            m_dSavedModeCostLFNST = CandCostList;
            // PBINTRA fast
            m_uiSavedHadModeListLFNST = uiHadModeList;
            m_dSavedHadListLFNST = CandHadList;
            LFNSTSaveFlag = false;
          }
        }
        else //if( sps.getUseMIP() && LFNSTLoadFlag)
        {
          // restore saved modes
          numModesForFullRD = m_uiSavedNumRdModesLFNST;
          uiRdModeList = m_uiSavedRdModeListLFNST;
          CandCostList = m_dSavedModeCostLFNST;
          // PBINTRA fast
          uiHadModeList = m_uiSavedHadModeListLFNST;
          CandHadList = m_dSavedHadListLFNST;
        }

        if( m_pcEncCfg->getFastUDIUseMPMEnabled() )
        {
          const int numMPMs = NUM_MOST_PROBABLE_MODES;
          unsigned  uiPreds[numMPMs];

          pu.multiRefIdx = 0;

          const int numCand = PU::getIntraMPMs( pu, uiPreds );

          for( int j = 0; j < numCand; j++ )
          {
            bool mostProbableModeIncluded = false;
            ModeInfo mostProbableMode( false, 0, NOT_INTRA_SUBPARTITIONS, uiPreds[j] );


            for( int i = 0; i < numModesForFullRD; i++ )
            {
              mostProbableModeIncluded |= ( mostProbableMode == uiRdModeList[i] );
            }
            if( !mostProbableModeIncluded )
            {
              numModesForFullRD++;
              uiRdModeList.push_back( mostProbableMode );
              CandCostList.push_back(0);
            }
          }
          if ( testISP )
          {
            // we add the MPMs to the list that contains only regular intra modes
            for (int j = 0; j < numCand; j++)
            {
              bool     mostProbableModeIncluded = false;
              ModeInfo mostProbableMode(false, 0, NOT_INTRA_SUBPARTITIONS, uiPreds[j]);

              for (int i = 0; i < m_ispCandListHor.size(); i++)
              {
                mostProbableModeIncluded |= (mostProbableMode == m_ispCandListHor[i]);
              }
              if (!mostProbableModeIncluded)
              {
                m_ispCandListHor.push_back(mostProbableMode);
              }
            }
          }
        }
      }
      else
      {
        THROW( "Full search not supported for MIP" );
      }
      if( sps.getUseLFNST() && mtsUsageFlag == 1 )
      {
        // Store the modes to be checked with RD
        m_savedNumRdModes[ lfnstIdx ]     = numModesForFullRD;
        std::copy_n( uiRdModeList.begin(),  numModesForFullRD, m_savedRdModeList[ lfnstIdx ] );
      }
    }
    else //mtsUsage = 2 (here we potentially reduce the number of modes that will be full-RD checked)
    {
      if( ( m_pcEncCfg->getUseFastLFNST() || !cu.slice->isIntra() ) && m_bestModeCostValid[ lfnstIdx ] )
      {
        numModesForFullRD = 0;

        double thresholdSkipMode = 1.0 + ( ( cu.lfnstIdx > 0 ) ? 0.1 : 1.0 ) * ( 1.4 / sqrt( ( double ) ( width*height ) ) );

        // Skip checking the modes with much larger R-D cost than the best mode
        for( int i = 0; i < m_savedNumRdModes[ lfnstIdx ]; i++ )
        {
          if( m_modeCostStore[ lfnstIdx ][ i ] <= thresholdSkipMode * m_bestModeCostStore[ lfnstIdx ] )
          {
            uiRdModeList.push_back( m_savedRdModeList[ lfnstIdx ][ i ] );
            numModesForFullRD++;
          }
        }
      }
      else //this is necessary because we skip the candidates list calculation, since it was already obtained for the DCT-II. Now we load it
      {
        // Restore the modes to be checked with RD
        numModesForFullRD = m_savedNumRdModes[ lfnstIdx ];
        uiRdModeList.resize( numModesForFullRD );
        std::copy_n( m_savedRdModeList[ lfnstIdx ], m_savedNumRdModes[ lfnstIdx ], uiRdModeList.begin() );
        CandCostList.resize( numModesForFullRD );
      }
    }



    CHECK( numModesForFullRD != uiRdModeList.size(), "Inconsistent state!" );

    // after this point, don't use numModesForFullRD

    // PBINTRA fast
    if( m_pcEncCfg->getUsePbIntraFast() && !cs.slice->isIntra() && uiRdModeList.size() < numModesAvailable && !cs.slice->getDisableSATDForRD() && ( mtsUsageFlag != 2 || lfnstIdx > 0 ) )
    {
      double pbintraRatio = (lfnstIdx > 0) ? 1.25 : PBINTRA_RATIO;
      int maxSize = -1;
      ModeInfo bestMipMode;
      int bestMipIdx = -1;
      for( int idx = 0; idx < uiRdModeList.size(); idx++ )
      {
        if( uiRdModeList[idx].mipFlg )
        {
          bestMipMode = uiRdModeList[idx];
          bestMipIdx = idx;
          break;
        }
      }
      const int numHadCand = 3;
      for (int k = numHadCand - 1; k >= 0; k--)
      {
        if (CandHadList.size() < (k + 1) || CandHadList[k] > cs.interHad * pbintraRatio) { maxSize = k; }
      }
      if (maxSize > 0)
      {
        uiRdModeList.resize(std::min<size_t>(uiRdModeList.size(), maxSize));
        if( bestMipIdx >= 0 )
        {
          if( uiRdModeList.size() <= bestMipIdx )
          {
            uiRdModeList.push_back(bestMipMode);
          }
        }
        if ( testISP )
        {
          m_ispCandListHor.resize(std::min<size_t>(m_ispCandListHor.size(), maxSize));
        }
      }
      if (maxSize == 0)
      {
        cs.dist = std::numeric_limits<Distortion>::max();
        cs.interHad = 0;

        //===== reset context models =====
        m_CABACEstimator->getCtx() = SubCtx(Ctx::MipFlag, ctxStartMipFlag);
        m_CABACEstimator->getCtx() = SubCtx(Ctx::ISPMode, ctxStartIspMode);
        m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaPlanarFlag, ctxStartPlanarFlag);
        m_CABACEstimator->getCtx() = SubCtx(Ctx::IntraLumaMpmFlag, ctxStartIntraMode);
        m_CABACEstimator->getCtx() = SubCtx(Ctx::MultiRefLineIdx, ctxStartMrlIdx);

        return false;
      }
    }

    int numNonISPModes = (int)uiRdModeList.size();

    if ( testISP )
    {
      // we reserve positions for ISP in the common full RD list
      const int maxNumRDModesISP = 16;
      for (int i = 0; i < maxNumRDModesISP; i++)
        uiRdModeList.push_back(ModeInfo(false, 0, INTRA_SUBPARTITIONS_RESERVED, 0));
    }

    //===== check modes (using r-d costs) =====
    ModeInfo       uiBestPUMode;
    int            bestBDPCMMode = 0;
    double         bestCostNonBDPCM = MAX_DOUBLE;

    CodingStructure *csTemp = m_pTempCS[gp_sizeIdxInfo->idxFrom( cu.lwidth() )][gp_sizeIdxInfo->idxFrom( cu.lheight() )];
    CodingStructure *csBest = m_pBestCS[gp_sizeIdxInfo->idxFrom( cu.lwidth() )][gp_sizeIdxInfo->idxFrom( cu.lheight() )];

    csTemp->slice = cs.slice;
    csBest->slice = cs.slice;
    csTemp->initStructData();
    csBest->initStructData();
    csTemp->picture = cs.picture;
    csBest->picture = cs.picture;

    static_vector<int, FAST_UDI_MAX_RDMODE_NUM> rdModeIdxList;
    if (testMip)
    {
    static_vector<ModeInfo, FAST_UDI_MAX_RDMODE_NUM> rdModeListTemp;
    for( int i = 0; i < uiRdModeList.size(); i++)
    {
      if( !uiRdModeList[i].mipFlg && uiRdModeList[i].ispMod==NOT_INTRA_SUBPARTITIONS )
      {
        rdModeListTemp.push_back( uiRdModeList[i] );
        rdModeIdxList.push_back( i );
      }
    }
    for( int i = 0; i < uiRdModeList.size(); i++)
    {
      if( uiRdModeList[i].mipFlg || uiRdModeList[i].ispMod!=NOT_INTRA_SUBPARTITIONS )
      {
        rdModeListTemp.push_back( uiRdModeList[i] );
        rdModeIdxList.push_back( i );
      }
    }
      uiRdModeList.resize(rdModeListTemp.size());
    for( int i = 0; i < uiRdModeList.size(); i++)
    {
      uiRdModeList[i] = rdModeListTemp[i];
    }
    }
    else
    {
      static_vector<ModeInfo, FAST_UDI_MAX_RDMODE_NUM> rdModeListTemp;
      for( int i = 0; i < uiRdModeList.size(); i++ )
      {
        if( !uiRdModeList[i].mipFlg  )
        {
          rdModeListTemp.push_back( uiRdModeList[i] );
        }
      }
      uiRdModeList.resize(rdModeListTemp.size());
      for( int i = 0; i < rdModeListTemp.size(); i++ )
      {
        uiRdModeList[i] = rdModeListTemp[i];
      }
    }

    // just to be sure
    numModesForFullRD = ( int ) uiRdModeList.size();
    TUIntraSubPartitioner subTuPartitioner( partitioner );
    if( !cu.ispMode && !cu.mtsFlag )
    {
      m_modeCtrl->setMtsFirstPassNoIspCost( MAX_DOUBLE );
    }

    for (int mode = -2 * int(testBDPCM); mode < (int)uiRdModeList.size(); mode++)
    {
      // set CU/PU to luma prediction mode
      ModeInfo uiOrgMode;
      if ( mode < 0 )
      {
        cu.bdpcmMode = -mode;

        uiOrgMode = ModeInfo(false, 0, NOT_INTRA_SUBPARTITIONS, cu.bdpcmMode == 2 ? VER_IDX : HOR_IDX);
        cu.mipFlag                     = uiOrgMode.mipFlg;
        cu.ispMode                     = uiOrgMode.ispMod;
        pu.multiRefIdx                 = uiOrgMode.mRefId;
        pu.intraDir[CHANNEL_TYPE_LUMA] = uiOrgMode.modeId;
      }
      else
      {
        cu.bdpcmMode = 0;
        if (uiRdModeList[mode].ispMod == INTRA_SUBPARTITIONS_RESERVED)
        {
          if (mode == numNonISPModes) // the list needs to be sorted only once
          {
            xSortISPCandList(bestCurrentCost, csBest->cost);
          }
          xGetNextISPMode(uiRdModeList[mode], (mode > 0 ? &uiRdModeList[mode - 1] : nullptr), Size(width, height));
          if (uiRdModeList[mode].ispMod == INTRA_SUBPARTITIONS_RESERVED)
            continue;
        }
        uiOrgMode = uiRdModeList[mode];
      cu.mipFlag                     = uiOrgMode.mipFlg;
      cu.ispMode                     = uiOrgMode.ispMod;
      pu.multiRefIdx                 = uiOrgMode.mRefId;
      pu.intraDir[CHANNEL_TYPE_LUMA] = uiOrgMode.modeId;

      CHECK(cu.mipFlag && pu.multiRefIdx, "Error: combination of MIP and MRL not supported");
      CHECK(pu.multiRefIdx && (pu.intraDir[0] == PLANAR_IDX), "Error: combination of MRL and Planar mode not supported");
      CHECK(cu.ispMode && cu.mipFlag, "Error: combination of ISP and MIP not supported");
      CHECK(cu.ispMode && pu.multiRefIdx, "Error: combination of ISP and MRL not supported");
      }

      // set context models
      m_CABACEstimator->getCtx() = ctxStart;

      // determine residual for partition
      cs.initSubStructure( *csTemp, partitioner.chType, cs.area, true );

      bool tmpValidReturn = false;
      if( cu.ispMode )
      {
        tmpValidReturn = xIntraCodingLumaISP(*csTemp, subTuPartitioner, bestCurrentCost);
        if (csTemp->tus.size() == 0)
        {
          // no TUs were coded
          csTemp->cost = MAX_DOUBLE;
          continue;
        }
        if (!cu.mtsFlag && !cu.lfnstIdx)
        {
          // we save the data for future tests
          m_ispTestedModes.setModeResults((ISPType)cu.ispMode, (int)uiOrgMode.modeId, (int)csTemp->tus.size(), csTemp->cus[0]->firstTU->cbf[COMPONENT_Y] ? csTemp->cost : MAX_DOUBLE, csBest->cost);
        }
      }
      else
      {
        tmpValidReturn = xRecurIntraCodingLumaQT( *csTemp, partitioner, uiBestPUMode.ispMod ? bestCurrentCost : MAX_DOUBLE, -1, TU_NO_ISP, uiBestPUMode.ispMod,
                                                  mtsCheckRangeFlag, mtsFirstCheckId, mtsLastCheckId, moreProbMTSIdxFirst );
      }

      if (!cu.ispMode && !cu.mtsFlag && !cu.lfnstIdx && !cu.bdpcmMode && !pu.multiRefIdx && !cu.mipFlag && testISP)
      {
        m_regIntraRDListWithCosts.push_back(ModeInfoWithCost(cu.mipFlag, pu.multiRefIdx, cu.ispMode, uiOrgMode.modeId, csTemp->cost));
      }

      if( cu.ispMode && !csTemp->cus[0]->firstTU->cbf[COMPONENT_Y] )
      {
        csTemp->cost = MAX_DOUBLE;
        csTemp->costDbOffset = 0;
        tmpValidReturn = false;
      }
      validReturn |= tmpValidReturn;

      if( sps.getUseLFNST() && mtsUsageFlag == 1 && !cu.ispMode && mode >= 0 )
      {
        m_modeCostStore[ lfnstIdx ][ testMip ? rdModeIdxList[ mode ] : mode ] = tmpValidReturn ? csTemp->cost : ( MAX_DOUBLE / 2.0 ); //(MAX_DOUBLE / 2.0) ??
      }

      DTRACE(g_trace_ctx, D_INTRA_COST, "IntraCost T [x=%d,y=%d,w=%d,h=%d] %f (%d,%d,%d,%d,%d,%d) \n", cu.blocks[0].x,
        cu.blocks[0].y, (int)width, (int)height, csTemp->cost, uiOrgMode.modeId, uiOrgMode.ispMod,
        pu.multiRefIdx, cu.mipFlag, cu.lfnstIdx, cu.mtsFlag);


      if( tmpValidReturn )
      {
        // check r-d cost
        if( csTemp->cost < csBest->cost )
        {
          std::swap( csTemp, csBest );

          uiBestPUMode  = uiOrgMode;
          bestBDPCMMode = cu.bdpcmMode;
          if( sps.getUseLFNST() && mtsUsageFlag == 1 && !cu.ispMode )
          {
            m_bestModeCostStore[ lfnstIdx ] = csBest->cost; //cs.cost;