UnitTools.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) 2010-2018, 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     UnitTool.cpp
 *  \brief    defines operations for basic units
 */

#include "UnitTools.h"

#include "dtrace_next.h"

#include "Unit.h"
#include "Slice.h"
#include "Picture.h"

#include <utility>
#include <algorithm>

// CS tools


uint64_t CS::getEstBits(const CodingStructure &cs)
{
  return cs.fracBits >> SCALE_BITS;
}



bool CS::isDualITree( const CodingStructure &cs )
{
  return cs.slice->isIRAP() && !cs.pcv->ISingleTree;
}

UnitArea CS::getArea( const CodingStructure &cs, const UnitArea &area, const ChannelType chType )
{
  return isDualITree( cs ) ? area.singleChan( chType ) : area;
}

#if DMVR_JVET_LOW_LATENCY_K0217
void CS::setRefinedMotionField(CodingStructure &cs)
{
  for (CodingUnit *cu : cs.cus)
  {
    for (auto &pu : CU::traversePUs(*cu))
    {
      if (pu.cs->sps->getSpsNext().getUseDMVR()
        && pu.mergeFlag
        && pu.mergeType == MRG_TYPE_DEFAULT_N
        && !pu.frucMrgMode
        && !pu.cu->LICFlag
        && !pu.cu->affine
        && PU::isBiPredFromDifferentDir(pu))
      {
        pu.mv[REF_PIC_LIST_0] += pu.mvd[REF_PIC_LIST_0];
        pu.mv[REF_PIC_LIST_1] -= pu.mvd[REF_PIC_LIST_0];
        pu.mvd[REF_PIC_LIST_0].setZero();
        PU::spanMotionInfo(pu);
      }
    }
  }
}
#endif
// CU tools

bool CU::isIntra(const CodingUnit &cu)
{
  return cu.predMode == MODE_INTRA;
}

bool CU::isInter(const CodingUnit &cu)
{
  return cu.predMode == MODE_INTER;
}

bool CU::isRDPCMEnabled(const CodingUnit& cu)
{
  return cu.cs->sps->getSpsRangeExtension().getRdpcmEnabledFlag(cu.predMode == MODE_INTRA ? RDPCM_SIGNAL_IMPLICIT : RDPCM_SIGNAL_EXPLICIT);
}

bool CU::isLosslessCoded(const CodingUnit &cu)
{
  return cu.cs->pps->getTransquantBypassEnabledFlag() && cu.transQuantBypass;
}

bool CU::isSameSlice(const CodingUnit& cu, const CodingUnit& cu2)
{
  return cu.slice->getIndependentSliceIdx() == cu2.slice->getIndependentSliceIdx();
}

#if HEVC_TILES_WPP
bool CU::isSameTile(const CodingUnit& cu, const CodingUnit& cu2)
{
  return cu.tileIdx == cu2.tileIdx;
}

bool CU::isSameSliceAndTile(const CodingUnit& cu, const CodingUnit& cu2)
{
  return ( cu.slice->getIndependentSliceIdx() == cu2.slice->getIndependentSliceIdx() ) && ( cu.tileIdx == cu2.tileIdx );
}
#endif

bool CU::isSameCtu(const CodingUnit& cu, const CodingUnit& cu2)
{
  uint32_t ctuSizeBit = g_aucLog2[cu.cs->sps->getMaxCUWidth()];

  Position pos1Ctu(cu.lumaPos().x  >> ctuSizeBit, cu.lumaPos().y  >> ctuSizeBit);
  Position pos2Ctu(cu2.lumaPos().x >> ctuSizeBit, cu2.lumaPos().y >> ctuSizeBit);

  return pos1Ctu.x == pos2Ctu.x && pos1Ctu.y == pos2Ctu.y;
}

uint32_t CU::getIntraSizeIdx(const CodingUnit &cu)
{
  uint8_t uiWidth = cu.lumaSize().width;

  uint32_t  uiCnt   = 0;

  while (uiWidth)
  {
    uiCnt++;
    uiWidth >>= 1;
  }

  uiCnt -= 2;

  return uiCnt > 6 ? 6 : uiCnt;
}

bool CU::isLastSubCUOfCtu( const CodingUnit &cu )
{
  const SPS &sps      = *cu.cs->sps;
  const Area cuAreaY = CS::isDualITree( *cu.cs ) ? Area( recalcPosition( cu.chromaFormat, cu.chType, CHANNEL_TYPE_LUMA, cu.blocks[cu.chType].pos() ), recalcSize( cu.chromaFormat, cu.chType, CHANNEL_TYPE_LUMA, cu.blocks[cu.chType].size() ) ) : ( const Area& ) cu.Y();

  return ( ( ( ( cuAreaY.x + cuAreaY.width  ) & cu.cs->pcv->maxCUWidthMask  ) == 0 || cuAreaY.x + cuAreaY.width  == sps.getPicWidthInLumaSamples()  ) &&
           ( ( ( cuAreaY.y + cuAreaY.height ) & cu.cs->pcv->maxCUHeightMask ) == 0 || cuAreaY.y + cuAreaY.height == sps.getPicHeightInLumaSamples() ) );
}

uint32_t CU::getCtuAddr( const CodingUnit &cu )
{
  return getCtuAddr( cu.blocks[cu.chType].lumaPos(), *cu.cs->pcv );
}

int CU::predictQP( const CodingUnit& cu, const int prevQP )
{
  const CodingStructure &cs = *cu.cs;

#if ENABLE_WPP_PARALLELISM
  if( cs.sps->getSpsNext().getUseNextDQP() )
  {
    // Inter-CTU 2D "planar"   c(orner)  a(bove)
    // predictor arrangement:  b(efore)  p(rediction)

    // restrict the lookup, as it might cross CTU/slice/tile boundaries
    const CodingUnit *cuA = cs.getCURestricted( cu.blocks[cu.chType].pos().offset(  0, -1 ), cu, cu.chType );
    const CodingUnit *cuB = cs.getCURestricted( cu.blocks[cu.chType].pos().offset( -1,  0 ), cu, cu.chType );
    const CodingUnit *cuC = cs.getCURestricted( cu.blocks[cu.chType].pos().offset( -1, -1 ), cu, cu.chType );

    const int a = cuA ? cuA->qp : cs.slice->getSliceQpBase();
    const int b = cuB ? cuB->qp : cs.slice->getSliceQpBase();
    const int c = cuC ? cuC->qp : cs.slice->getSliceQpBase();

    return Clip3( ( a < b ? a : b ), ( a > b ? a : b ), a + b - c ); // derived from Martucci's Median Adaptive Prediction, 1990
  }

#endif
  // only predict within the same CTU, use HEVC's above+left prediction
  const int a = ( cu.blocks[cu.chType].y & ( cs.pcv->maxCUHeightMask >> getChannelTypeScaleY( cu.chType, cu.chromaFormat ) ) ) ? ( cs.getCU( cu.blocks[cu.chType].pos().offset( 0, -1 ), cu.chType ) )->qp : prevQP;
  const int b = ( cu.blocks[cu.chType].x & ( cs.pcv->maxCUWidthMask  >> getChannelTypeScaleX( cu.chType, cu.chromaFormat ) ) ) ? ( cs.getCU( cu.blocks[cu.chType].pos().offset( -1, 0 ), cu.chType ) )->qp : prevQP;

  return ( a + b + 1 ) >> 1;
}

bool CU::isQGStart( const CodingUnit& cu )
{
  const SPS &sps = *cu.cs->sps;
  const PPS &pps = *cu.cs->pps;

  return ( cu.blocks[cu.chType].x % ( ( 1 << ( g_aucLog2[sps.getMaxCUWidth()]  - pps.getMaxCuDQPDepth() ) ) >> getChannelTypeScaleX( cu.chType, cu.chromaFormat ) ) ) == 0 &&
         ( cu.blocks[cu.chType].y % ( ( 1 << ( g_aucLog2[sps.getMaxCUHeight()] - pps.getMaxCuDQPDepth() ) ) >> getChannelTypeScaleY( cu.chType, cu.chromaFormat ) ) ) == 0;
}

uint32_t CU::getNumPUs( const CodingUnit& cu )
{
  uint32_t cnt = 0;
  PredictionUnit *pu = cu.firstPU;

  do
  {
    cnt++;
  } while( ( pu != cu.lastPU ) && ( pu = pu->next ) );

  return cnt;
}

void CU::addPUs( CodingUnit& cu )
{
  cu.cs->addPU( CS::getArea( *cu.cs, cu, cu.chType ), cu.chType );
}


PartSplit CU::getSplitAtDepth( const CodingUnit& cu, const unsigned depth )
{
  if( depth >= cu.depth ) return CU_DONT_SPLIT;

  const PartSplit cuSplitType = PartSplit( ( cu.splitSeries >> ( depth * SPLIT_DMULT ) ) & SPLIT_MASK );

  if     ( cuSplitType == CU_QUAD_SPLIT    ) return CU_QUAD_SPLIT;

  else if( cuSplitType == CU_HORZ_SPLIT    ) return CU_HORZ_SPLIT;

  else if( cuSplitType == CU_VERT_SPLIT    ) return CU_VERT_SPLIT;

  else if( cuSplitType == CU_TRIH_SPLIT    ) return CU_TRIH_SPLIT;
  else if( cuSplitType == CU_TRIV_SPLIT    ) return CU_TRIV_SPLIT;
  else   { THROW( "Unknown split mode"    ); return CU_QUAD_SPLIT; }
}

bool CU::hasNonTsCodedBlock( const CodingUnit& cu )
{
  bool hasAnyNonTSCoded = false;

  for( auto &currTU : traverseTUs( cu ) )
  {
    for( uint32_t i = 0; i < ::getNumberValidTBlocks( *cu.cs->pcv ); i++ )
    {
      hasAnyNonTSCoded |= ( currTU.blocks[i].valid() && !currTU.transformSkip[i] && TU::getCbf( currTU, ComponentID( i ) ) );
    }
  }

  return hasAnyNonTSCoded;
}

uint32_t CU::getNumNonZeroCoeffNonTs( const CodingUnit& cu )
{
  uint32_t count = 0;
  for( auto &currTU : traverseTUs( cu ) )
  {
    count += TU::getNumNonZeroCoeffsNonTS( currTU );
  }

  return count;
}




PUTraverser CU::traversePUs( CodingUnit& cu )
{
  return PUTraverser( cu.firstPU, cu.lastPU->next );
}

TUTraverser CU::traverseTUs( CodingUnit& cu )
{
  return TUTraverser( cu.firstTU, cu.lastTU->next );
}

cPUTraverser CU::traversePUs( const CodingUnit& cu )
{
  return cPUTraverser( cu.firstPU, cu.lastPU->next );
}

cTUTraverser CU::traverseTUs( const CodingUnit& cu )
{
  return cTUTraverser( cu.firstTU, cu.lastTU->next );
}

// PU tools

int PU::getIntraMPMs( const PredictionUnit &pu, unsigned* mpm, const ChannelType &channelType /*= CHANNEL_TYPE_LUMA*/ )
{
  const unsigned numMPMs = pu.cs->pcv->numMPMs;
  {
    int numCand      = -1;
    int leftIntraDir = DC_IDX, aboveIntraDir = DC_IDX;

    const CompArea &area = pu.block(getFirstComponentOfChannel(channelType));
    const Position &pos  = area.pos();

    // Get intra direction of left PU
    const PredictionUnit *puLeft = pu.cs->getPURestricted(pos.offset(-1, 0), pu, channelType);

    if (puLeft && CU::isIntra(*puLeft->cu))
    {
      leftIntraDir = puLeft->intraDir[channelType];

      if (isChroma(channelType) && leftIntraDir == DM_CHROMA_IDX)
      {
        leftIntraDir = puLeft->intraDir[0];
      }
    }

    // Get intra direction of above PU
    const PredictionUnit *puAbove = pu.cs->getPURestricted(pos.offset(0, -1), pu, channelType);

    if (puAbove && CU::isIntra(*puAbove->cu) && CU::isSameCtu(*pu.cu, *puAbove->cu))
    {
      aboveIntraDir = puAbove->intraDir[channelType];

      if (isChroma(channelType) && aboveIntraDir == DM_CHROMA_IDX)
      {
        aboveIntraDir = puAbove->intraDir[0];
      }
    }

    CHECK(2 >= numMPMs, "Invalid number of most probable modes");

    const int offset = 61;
    const int mod    = 64;

    if (leftIntraDir == aboveIntraDir)
    {
      numCand = 1;

      if (leftIntraDir > DC_IDX)   // angular modes
      {
        mpm[0] = leftIntraDir;
        mpm[1] = ((leftIntraDir + offset) % mod) + 2;
        mpm[2] = ((leftIntraDir - 1) % mod) + 2;
      }
      else   // non-angular
      {
        mpm[0] = PLANAR_IDX;
        mpm[1] = DC_IDX;
        mpm[2] = VER_IDX;
      }
    }
    else
    {
      numCand = 2;

      mpm[0] = leftIntraDir;
      mpm[1] = aboveIntraDir;

      if (leftIntraDir && aboveIntraDir)   // both modes are non-planar
      {
        mpm[2] = PLANAR_IDX;
      }
      else
      {
        mpm[2] = (leftIntraDir + aboveIntraDir) < 2 ? VER_IDX : DC_IDX;
      }
    }
    for (int i = 0; i < numMPMs; i++)
    {
      CHECK(mpm[i] >= NUM_LUMA_MODE, "Invalid MPM");
    }
    CHECK(numCand == 0, "No candidates found");
    return numCand;
  }
}


void PU::getIntraChromaCandModes( const PredictionUnit &pu, unsigned modeList[NUM_CHROMA_MODE] )
{
  {
    modeList[  0 ] = PLANAR_IDX;
    modeList[  1 ] = VER_IDX;
    modeList[  2 ] = HOR_IDX;
    modeList[  3 ] = DC_IDX;
    modeList[4] = LM_CHROMA_IDX;
    modeList[5] = DM_CHROMA_IDX;

    const PredictionUnit *lumaPU = CS::isDualITree( *pu.cs ) ? pu.cs->picture->cs->getPU( pu.blocks[pu.chType].lumaPos(), CHANNEL_TYPE_LUMA ) : &pu;
    const uint32_t lumaMode = lumaPU->intraDir[CHANNEL_TYPE_LUMA];
    for( int i = 0; i < 4; i++ )
    {
      if( lumaMode == modeList[i] )
      {
        modeList[i] = VDIA_IDX;
        break;
      }
    }
  }
}


bool PU::isLMCMode(unsigned mode)
{
  return (mode == LM_CHROMA_IDX);
}
bool PU::isLMCModeEnabled(const PredictionUnit &pu, unsigned mode)
{
  if ( pu.cs->sps->getSpsNext().getUseLMChroma() )
  {
    return true;
  }
  return false;
}

int PU::getLMSymbolList(const PredictionUnit &pu, int *pModeList)
{
  const int iNeighbors = 5;
  const PredictionUnit* neighboringPUs[ iNeighbors ];

  const CompArea& area = pu.Cb();
  const Position posLT = area.topLeft();
  const Position posRT = area.topRight();
  const Position posLB = area.bottomLeft();

  neighboringPUs[ 0 ] = pu.cs->getPURestricted( posLB.offset(-1,  0), pu, CHANNEL_TYPE_CHROMA ); //left
  neighboringPUs[ 1 ] = pu.cs->getPURestricted( posRT.offset( 0, -1), pu, CHANNEL_TYPE_CHROMA ); //above
  neighboringPUs[ 2 ] = pu.cs->getPURestricted( posRT.offset( 1, -1), pu, CHANNEL_TYPE_CHROMA ); //aboveRight
  neighboringPUs[ 3 ] = pu.cs->getPURestricted( posLB.offset(-1,  1), pu, CHANNEL_TYPE_CHROMA ); //BelowLeft
  neighboringPUs[ 4 ] = pu.cs->getPURestricted( posLT.offset(-1, -1), pu, CHANNEL_TYPE_CHROMA ); //AboveLeft

  int iCount = 0;
  for ( int i = 0; i < iNeighbors; i++ )
  {
    if ( neighboringPUs[i] && CU::isIntra( *(neighboringPUs[i]->cu) ) )
    {
      int iMode = neighboringPUs[i]->intraDir[CHANNEL_TYPE_CHROMA];
      if ( ! PU::isLMCMode( iMode ) )
      {
        iCount++;
      }
    }
  }

  bool bNonLMInsert = false;
  int iIdx = 0;

  pModeList[ iIdx++ ] = LM_CHROMA_IDX;

  if ( iCount >= g_aiNonLMPosThrs[0] && ! bNonLMInsert )
  {
    pModeList[ iIdx++ ] = -1;
    bNonLMInsert = true;
  }

  if ( iCount >= g_aiNonLMPosThrs[1] && ! bNonLMInsert )
  {
    pModeList[ iIdx++ ] = -1;
    bNonLMInsert = true;
  }
  if ( ! bNonLMInsert )
  {
    pModeList[ iIdx++ ] = -1;
    bNonLMInsert = true;
  }

  return iIdx;
}



bool PU::isChromaIntraModeCrossCheckMode( const PredictionUnit &pu )
{
  return pu.intraDir[CHANNEL_TYPE_CHROMA] == DM_CHROMA_IDX;
}

uint32_t PU::getFinalIntraMode( const PredictionUnit &pu, const ChannelType &chType )
{
  uint32_t uiIntraMode = pu.intraDir[chType];

  if( uiIntraMode == DM_CHROMA_IDX && !isLuma( chType ) )
  {
    const PredictionUnit &lumaPU = CS::isDualITree( *pu.cs ) ? *pu.cs->picture->cs->getPU( pu.blocks[chType].lumaPos(), CHANNEL_TYPE_LUMA ) : *pu.cs->getPU( pu.blocks[chType].lumaPos(), CHANNEL_TYPE_LUMA );
    uiIntraMode = lumaPU.intraDir[0];
  }
  if( pu.chromaFormat == CHROMA_422 && !isLuma( chType ) )
  {
    uiIntraMode = g_chroma422IntraAngleMappingTable[uiIntraMode];
  }
  return uiIntraMode;
}


void PU::getInterMergeCandidates( const PredictionUnit &pu, MergeCtx& mrgCtx, const int& mrgCandIdx )
{
  const CodingStructure &cs  = *pu.cs;
  const Slice &slice         = *pu.cs->slice;
  const uint32_t maxNumMergeCand = slice.getMaxNumMergeCand();
  const bool canFastExit     = pu.cs->pps->getLog2ParallelMergeLevelMinus2() == 0;

  bool isCandInter[MRG_MAX_NUM_CANDS];

  for (uint32_t ui = 0; ui < maxNumMergeCand; ++ui)
  {
    isCandInter[ui] = false;
    mrgCtx.interDirNeighbours[ui] = 0;
    mrgCtx.mrgTypeNeighbours [ui] = MRG_TYPE_DEFAULT_N;
    mrgCtx.mvFieldNeighbours[(ui << 1)    ].refIdx = NOT_VALID;
    mrgCtx.mvFieldNeighbours[(ui << 1) + 1].refIdx = NOT_VALID;
  }

  mrgCtx.numValidMergeCand = maxNumMergeCand;
  // compute the location of the current PU

  int cnt = 0;
  const Position posLT = pu.Y().topLeft();
  const Position posRT = pu.Y().topRight();
  const Position posLB = pu.Y().bottomLeft();

  MotionInfo miAbove, miLeft, miAboveLeft, miAboveRight, miBelowLeft;

  //left
  const PredictionUnit* puLeft = cs.getPURestricted( posLB.offset( -1, 0 ), pu, pu.chType );

  const bool isAvailableA1 = puLeft && isDiffMER( pu, *puLeft ) && pu.cu != puLeft->cu && CU::isInter( *puLeft->cu );

  if( isAvailableA1 )
  {
    miLeft = puLeft->getMotionInfo( posLB.offset(-1, 0) );

    isCandInter[cnt] = true;

    // get Inter Dir
    mrgCtx.interDirNeighbours[cnt] = miLeft.interDir;
    // get Mv from Left
    mrgCtx.mvFieldNeighbours[cnt << 1].setMvField(miLeft.mv[0], miLeft.refIdx[0]);

    if (slice.isInterB())
    {
      mrgCtx.mvFieldNeighbours[(cnt << 1) + 1].setMvField(miLeft.mv[1], miLeft.refIdx[1]);
    }

    if( mrgCandIdx == cnt && canFastExit )
    {
      return;
    }

    cnt++;
  }

  // early termination
  if (cnt == maxNumMergeCand)
  {
    return;
  }


  // above
  const PredictionUnit *puAbove = cs.getPURestricted( posRT.offset( 0, -1 ), pu, pu.chType );

  bool isAvailableB1 = puAbove && isDiffMER( pu, *puAbove ) && pu.cu != puAbove->cu && CU::isInter( *puAbove->cu );

  if( isAvailableB1 )
  {
    miAbove = puAbove->getMotionInfo( posRT.offset( 0, -1 ) );

    if( !isAvailableA1 || ( miAbove != miLeft ) )
    {
      isCandInter[cnt] = true;

      // get Inter Dir
      mrgCtx.interDirNeighbours[cnt] = miAbove.interDir;
      // get Mv from Above
      mrgCtx.mvFieldNeighbours[cnt << 1].setMvField( miAbove.mv[0], miAbove.refIdx[0] );

      if( slice.isInterB() )
      {
        mrgCtx.mvFieldNeighbours[( cnt << 1 ) + 1].setMvField( miAbove.mv[1], miAbove.refIdx[1] );
      }

      if( mrgCandIdx == cnt && canFastExit )
      {
        return;
      }

      cnt++;
    }
  }

  // early termination
  if( cnt == maxNumMergeCand )
  {
    return;
  }

  // above right
  const PredictionUnit *puAboveRight = cs.getPURestricted( posRT.offset( 1, -1 ), pu, pu.chType );

  bool isAvailableB0 = puAboveRight && isDiffMER( pu, *puAboveRight ) && CU::isInter( *puAboveRight->cu );

  if( isAvailableB0 )
  {
    miAboveRight = puAboveRight->getMotionInfo( posRT.offset( 1, -1 ) );

#if HM_JEM_MERGE_CANDS
    if( ( !isAvailableB1 || ( miAbove != miAboveRight ) ) && ( !isAvailableA1 || ( miLeft != miAboveRight ) ) )
#else
    if( !isAvailableB1 || ( miAbove != miAboveRight ) )
#endif
    {
      isCandInter[cnt] = true;

      // get Inter Dir
      mrgCtx.interDirNeighbours[cnt] = miAboveRight.interDir;
      // get Mv from Above-right
      mrgCtx.mvFieldNeighbours[cnt << 1].setMvField( miAboveRight.mv[0], miAboveRight.refIdx[0] );

      if( slice.isInterB() )
      {
        mrgCtx.mvFieldNeighbours[( cnt << 1 ) + 1].setMvField( miAboveRight.mv[1], miAboveRight.refIdx[1] );
      }

      if( mrgCandIdx == cnt && canFastExit )
      {
        return;
      }

      cnt++;
    }
  }
  // early termination
  if( cnt == maxNumMergeCand )
  {
    return;
  }

  //left bottom
  const PredictionUnit *puLeftBottom = cs.getPURestricted( posLB.offset( -1, 1 ), pu, pu.chType );

  bool isAvailableA0 = puLeftBottom && isDiffMER( pu, *puLeftBottom ) && CU::isInter( *puLeftBottom->cu );

  if( isAvailableA0 )
  {
    miBelowLeft = puLeftBottom->getMotionInfo( posLB.offset( -1, 1 ) );

#if HM_JEM_MERGE_CANDS
    if( ( !isAvailableA1 || ( miBelowLeft != miLeft ) ) && ( !isAvailableB1 || ( miBelowLeft != miAbove ) ) && ( !isAvailableB0 || ( miBelowLeft != miAboveRight ) ) )
#else
    if( !isAvailableA1 || ( miBelowLeft != miLeft ) )
#endif
    {
      isCandInter[cnt] = true;

      // get Inter Dir
      mrgCtx.interDirNeighbours[cnt] = miBelowLeft.interDir;
      // get Mv from Bottom-Left
      mrgCtx.mvFieldNeighbours[cnt << 1].setMvField( miBelowLeft.mv[0], miBelowLeft.refIdx[0] );

      if( slice.isInterB() )
      {
        mrgCtx.mvFieldNeighbours[( cnt << 1 ) + 1].setMvField( miBelowLeft.mv[1], miBelowLeft.refIdx[1] );
      }

      if( mrgCandIdx == cnt && canFastExit )
      {
        return;
      }

      cnt++;
    }
  }
  // early termination
  if( cnt == maxNumMergeCand )
  {
    return;
  }

  bool enableSubPuMvp = slice.getSPS()->getSpsNext().getUseSubPuMvp();
  bool isAvailableSubPu = false;
  unsigned subPuMvpPos = 0;

  if( enableSubPuMvp )
  {
    CHECK( mrgCtx.subPuMvpMiBuf   .area() == 0 || !mrgCtx.subPuMvpMiBuf   .buf, "Buffer not initialized" );

    mrgCtx.subPuMvpMiBuf   .fill( MotionInfo() );
  }

  if( enableSubPuMvp && slice.getEnableTMVPFlag() )
  {
    bool bMrgIdxMatchATMVPCan = ( mrgCandIdx == cnt );
    bool tmpLICFlag           = false;

    isAvailableSubPu = cs.sps->getSpsNext().getUseATMVP() && getInterMergeSubPuMvpCand( pu, mrgCtx, tmpLICFlag, cnt 
    );

    if( isAvailableSubPu )
    {
      isCandInter[cnt] = true;

      mrgCtx.mrgTypeNeighbours[cnt] = MRG_TYPE_SUBPU_ATMVP;

      if( bMrgIdxMatchATMVPCan )
      {
        return;
      }
      subPuMvpPos = cnt;
      cnt++;

      if( cnt == maxNumMergeCand )
      {
        return;
      }
    }

  }

  // above left
  if( cnt < ( enableSubPuMvp ? 6 : 4 ) )
  {
    const PredictionUnit *puAboveLeft = cs.getPURestricted( posLT.offset( -1, -1 ), pu, pu.chType );

    bool isAvailableB2 = puAboveLeft && isDiffMER( pu, *puAboveLeft ) && CU::isInter( *puAboveLeft->cu );

    if( isAvailableB2 )
    {
      miAboveLeft = puAboveLeft->getMotionInfo( posLT.offset( -1, -1 ) );

#if HM_JEM_MERGE_CANDS
      if( ( !isAvailableA1 || ( miLeft != miAboveLeft ) ) && ( !isAvailableB1 || ( miAbove != miAboveLeft ) ) && ( !isAvailableA0 || ( miBelowLeft != miAboveLeft ) ) && ( !isAvailableB0 || ( miAboveRight != miAboveLeft ) ) )
#else
      if( ( !isAvailableA1 || ( miLeft != miAboveLeft ) ) && ( !isAvailableB1 || ( miAbove != miAboveLeft ) ) )
#endif
      {
        isCandInter[cnt] = true;

        // get Inter Dir
        mrgCtx.interDirNeighbours[cnt] = miAboveLeft.interDir;
        // get Mv from Above-Left
        mrgCtx.mvFieldNeighbours[cnt << 1].setMvField( miAboveLeft.mv[0], miAboveLeft.refIdx[0] );

        if( slice.isInterB() )
        {
          mrgCtx.mvFieldNeighbours[( cnt << 1 ) + 1].setMvField( miAboveLeft.mv[1], miAboveLeft.refIdx[1] );
        }

        if( mrgCandIdx == cnt && canFastExit )
        {
          return;
        }

        cnt++;
      }
    }
  }
  // early termination
  if (cnt == maxNumMergeCand)
  {
    return;
  }

  if (slice.getEnableTMVPFlag())
  {
    //>> MTK colocated-RightBottom
    // offset the pos to be sure to "point" to the same position the uiAbsPartIdx would've pointed to
    Position posRB = pu.Y().bottomRight().offset(-3, -3);

    const PreCalcValues& pcv = *cs.pcv;

    Position posC0;
    Position posC1 = pu.Y().center();
    bool C0Avail = false;

    if (((posRB.x + pcv.minCUWidth) < pcv.lumaWidth) && ((posRB.y + pcv.minCUHeight) < pcv.lumaHeight))
    {
      {
        Position posInCtu( posRB.x & pcv.maxCUWidthMask, posRB.y & pcv.maxCUHeightMask );

        if( ( posInCtu.x + 4 < pcv.maxCUWidth ) &&           // is not at the last column of CTU
            ( posInCtu.y + 4 < pcv.maxCUHeight ) )           // is not at the last row    of CTU
        {
          posC0 = posRB.offset( 4, 4 );
          C0Avail = true;
        }
        else if( posInCtu.x + 4 < pcv.maxCUWidth )           // is not at the last column of CTU But is last row of CTU
        {
          posC0 = posRB.offset( 4, 4 );
          // in the reference the CTU address is not set - thus probably resulting in no using this C0 possibility
        }
        else if( posInCtu.y + 4 < pcv.maxCUHeight )          // is not at the last row of CTU But is last column of CTU
        {
          posC0 = posRB.offset( 4, 4 );
          C0Avail = true;
        }
        else //is the right bottom corner of CTU
        {
          posC0 = posRB.offset( 4, 4 );
          // same as for last column but not last row
        }
      }
    }

    Mv        cColMv;
    int       iRefIdx     = 0;
    int       dir         = 0;
    unsigned  uiArrayAddr = cnt;
    bool      bExistMV    = ( C0Avail && getColocatedMVP(pu, REF_PIC_LIST_0, posC0, cColMv, iRefIdx ) )
                                      || getColocatedMVP(pu, REF_PIC_LIST_0, posC1, cColMv, iRefIdx );

    if (bExistMV)
    {
      dir     |= 1;
      mrgCtx.mvFieldNeighbours[2 * uiArrayAddr].setMvField(cColMv, iRefIdx);
    }

    if (slice.isInterB())
    {
      bExistMV = ( C0Avail && getColocatedMVP(pu, REF_PIC_LIST_1, posC0, cColMv, iRefIdx ) )
                           || getColocatedMVP(pu, REF_PIC_LIST_1, posC1, cColMv, iRefIdx );
      if (bExistMV)
      {
        dir     |= 2;
        mrgCtx.mvFieldNeighbours[2 * uiArrayAddr + 1].setMvField(cColMv, iRefIdx);
      }
    }

    if( dir != 0 )
    {
      bool addTMvp = !( cs.sps->getSpsNext().getUseSubPuMvp() && isAvailableSubPu );
      if( !addTMvp )
      {
        if ( dir != mrgCtx.interDirNeighbours[subPuMvpPos] )
        {
          addTMvp = true;
        }
        else
        {
          for( unsigned refList = 0; refList < NUM_REF_PIC_LIST_01; refList++ )
          {
            if( dir & ( 1 << refList ) )
            {
              if( mrgCtx.mvFieldNeighbours[( cnt << 1 ) + refList] != mrgCtx.mvFieldNeighbours[(subPuMvpPos << 1) + refList] )
              {
                addTMvp = true;
                break;
              }
            }
          }
        }
      }
#if HM_JEM_MERGE_CANDS
      int iSpanCand = isAvailableSubPu ? cnt - 1 : cnt;
      for( int i = 0; i < iSpanCand; i++ )
      {
        if( mrgCtx.interDirNeighbours[  i           ] == dir &&
            mrgCtx.mvFieldNeighbours [  i << 1      ] == mrgCtx.mvFieldNeighbours[  uiArrayAddr << 1      ] &&
            mrgCtx.mvFieldNeighbours [( i << 1 ) + 1] == mrgCtx.mvFieldNeighbours[( uiArrayAddr << 1 ) + 1] )
        {
          addTMvp = false;
        }
      }
#endif
      if( addTMvp )
      {
        mrgCtx.interDirNeighbours[uiArrayAddr] = dir;
        isCandInter              [uiArrayAddr] = true;

        if( mrgCandIdx == cnt && canFastExit )
        {
          return;
        }

        cnt++;
      }
    }
  }

  // early termination
  if (cnt == maxNumMergeCand)
  {
    return;
  }

  uint32_t uiArrayAddr = cnt;
  uint32_t uiCutoff    = std::min( uiArrayAddr, 4u );

  if (slice.isInterB())
  {
    static const uint32_t NUM_PRIORITY_LIST = 12;
    static const uint32_t uiPriorityList0[NUM_PRIORITY_LIST] = { 0 , 1, 0, 2, 1, 2, 0, 3, 1, 3, 2, 3 };
    static const uint32_t uiPriorityList1[NUM_PRIORITY_LIST] = { 1 , 0, 2, 0, 2, 1, 3, 0, 3, 1, 3, 2 };

    for (int idx = 0; idx < uiCutoff * (uiCutoff - 1) && uiArrayAddr != maxNumMergeCand; idx++)
    {
      CHECK( idx >= NUM_PRIORITY_LIST, "Invalid priority list number" );
      int i = uiPriorityList0[idx];
      int j = uiPriorityList1[idx];
      if (isCandInter[i] && isCandInter[j] && (mrgCtx.interDirNeighbours[i] & 0x1) && (mrgCtx.interDirNeighbours[j] & 0x2))
      {
        isCandInter[uiArrayAddr] = true;
        mrgCtx.interDirNeighbours[uiArrayAddr] = 3;

        // get Mv from cand[i] and cand[j]
        mrgCtx.mvFieldNeighbours[ uiArrayAddr << 1     ].setMvField(mrgCtx.mvFieldNeighbours[ i << 1     ].mv, mrgCtx.mvFieldNeighbours[ i << 1     ].refIdx);
        mrgCtx.mvFieldNeighbours[(uiArrayAddr << 1) + 1].setMvField(mrgCtx.mvFieldNeighbours[(j << 1) + 1].mv, mrgCtx.mvFieldNeighbours[(j << 1) + 1].refIdx);

        int iRefPOCL0 = slice.getRefPOC(REF_PIC_LIST_0, mrgCtx.mvFieldNeighbours[(uiArrayAddr << 1)    ].refIdx);
        int iRefPOCL1 = slice.getRefPOC(REF_PIC_LIST_1, mrgCtx.mvFieldNeighbours[(uiArrayAddr << 1) + 1].refIdx);

        if( iRefPOCL0 == iRefPOCL1 && mrgCtx.mvFieldNeighbours[( uiArrayAddr << 1 )].mv == mrgCtx.mvFieldNeighbours[( uiArrayAddr << 1 ) + 1].mv )
        {
          isCandInter[uiArrayAddr] = false;
        }
        else
        {
          uiArrayAddr++;
        }
      }
    }
  }

  // early termination
  if (uiArrayAddr == maxNumMergeCand)
  {
    return;
  }

  int iNumRefIdx = slice.isInterB() ? std::min(slice.getNumRefIdx(REF_PIC_LIST_0), slice.getNumRefIdx(REF_PIC_LIST_1)) : slice.getNumRefIdx(REF_PIC_LIST_0);

  int r = 0;
  int refcnt = 0;
  while (uiArrayAddr < maxNumMergeCand)
  {
    isCandInter               [uiArrayAddr     ] = true;
    mrgCtx.interDirNeighbours [uiArrayAddr     ] = 1;
    mrgCtx.mvFieldNeighbours  [uiArrayAddr << 1].setMvField(Mv(0, 0), r);

    if (slice.isInterB())
    {
      mrgCtx.interDirNeighbours [ uiArrayAddr          ] = 3;
      mrgCtx.mvFieldNeighbours  [(uiArrayAddr << 1) + 1].setMvField(Mv(0, 0), r);
    }

    uiArrayAddr++;

    if (refcnt == iNumRefIdx - 1)
    {
      r = 0;
    }
    else
    {
      ++r;
      ++refcnt;
    }
  }
  mrgCtx.numValidMergeCand = uiArrayAddr;
}


static int xGetDistScaleFactor(const int &iCurrPOC, const int &iCurrRefPOC, const int &iColPOC, const int &iColRefPOC)
{
  int iDiffPocD = iColPOC - iColRefPOC;
  int iDiffPocB = iCurrPOC - iCurrRefPOC;

  if (iDiffPocD == iDiffPocB)
  {
    return 4096;
  }
  else
  {
    int iTDB = Clip3(-128, 127, iDiffPocB);
    int iTDD = Clip3(-128, 127, iDiffPocD);
    int iX = (0x4000 + abs(iTDD / 2)) / iTDD;
    int iScale = Clip3(-4096, 4095, (iTDB * iX + 32) >> 6);
    return iScale;
  }
}

bool PU::getColocatedMVP(const PredictionUnit &pu, const RefPicList &eRefPicList, const Position &_pos, Mv& rcMv, const int &refIdx )
{
  // don't perform MV compression when generally disabled or subPuMvp is used
  const unsigned scale = ( pu.cs->pcv->noMotComp ? 1 : 4 * std::max<int>(1, 4 * AMVP_DECIMATION_FACTOR / 4) );
  const unsigned mask  = ~( scale - 1 );

  const Position pos = Position{ PosType( _pos.x & mask ), PosType( _pos.y & mask ) };

  const Slice &slice = *pu.cs->slice;

  // use coldir.
  const Picture* const pColPic = slice.getRefPic(RefPicList(slice.isInterB() ? 1 - slice.getColFromL0Flag() : 0), slice.getColRefIdx());

  if( !pColPic )
  {
    return false;
  }