openh264/codec/encoder/core/inc/mb_cache.h

/*!
 * \copy
 *     Copyright (c)  2013, Cisco Systems
 *     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.
 *
 *     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.
 *
 */

//mb_cache.h
#ifndef WELS_MACROBLOCK_CACHE_H__
#define WELS_MACROBLOCK_CACHE_H__

#include "typedefs.h"
#include "wels_common_basis.h"
#include "wels_const.h"
#include "macros.h"

namespace WelsSVCEnc {

/*
 *	MB Cache information, such one cache should be defined within a slice
 */
/*
 * Cache for Luma				Cache for Chroma(Cb, Cr)
 *
 *	TL T T T T					TL T T
 *	 L - - - -					 L - -
 *	 L - - - -					 L - - TR
 *	 L - - - -
 *   L - - - - TR
 *
 */

////////////////////////mapping scan index////////////////////////

extern const uint8_t g_kuiSmb4AddrIn256[16];
extern const uint8_t g_kuiMbCountScan4Idx[24];
extern const uint8_t g_kuiCache30ScanIdx[16];
extern const uint8_t g_kuiCache12_8x8RefIdx[4];
extern const uint8_t g_kuiCache48CountScan4Idx[24];

typedef	struct TagDCTCoeff {
//ALIGNED_DECLARE( int16_t, residual_ac[16], 16 ); //I_16x16
  int16_t iLumaBlock[16][16]; //based on block4x4 luma DC/AC
//ALIGNED_DECLARE( int16_t, iLumaI16x16Dc[16], 16 ); //I_16x16 DC
  int16_t iLumaI16x16Dc[16];
//ALIGNED_DECLARE( int16_t, iChromaDc[2][4], 16 ); //chroma DC
  int16_t iChromaBlock[8][16]; //based on block4x4  chroma DC/AC
  int16_t iChromaDc[2][4];
} SDCTCoeff ;

typedef struct TagMbCache {
//the followed pData now is promised aligned to 16 bytes
  ALIGNED_DECLARE (SMVComponentUnit, sMvComponents, 16);

  ALIGNED_DECLARE_MATRIX_1D (iNonZeroCoeffCount, 48, int8_t, 16);	// Cache line size
// 	int8_t		iNonZeroCoeffCount[6 * 8];	// Right luma, Chroma(Left Top Cb, Left btm Cr); must follow by iIntraPredMode!
  ALIGNED_DECLARE_MATRIX_1D (iIntraPredMode, 48, int8_t, 16);
//	must follow with iNonZeroCoeffCount!

  int32_t     iSadCost[4];			//avail 1; unavail 0
  SMVUnitXY  sMbMvp[MB_BLOCK8x8_NUM];// for write bs

//for residual decoding (recovery) at the side of Encoder
  int16_t* pCoeffLevel;		// tmep
//malloc memory for prediction
  uint8_t* pSkipMb;

//ALIGNED_DECLARE(uint8_t, pMemPredMb[2][256],  16);//One: Best I_16x16 Luma and refine frac_pixel pBuffer; another: PingPong I_8x8&&Inter Cb + Cr
  uint8_t* pMemPredMb;
  uint8_t* pMemPredLuma;// inter && intra share same pointer;
//ALIGNED_DECLARE(uint8_t, pMemPredChroma[2][64*2], 16); //another PingPong pBuffer: Best Cb + Cr;
  uint8_t* pMemPredChroma;// inter && intra share same pointer;
  uint8_t* pBestPredIntraChroma; //Cb:0~63;   Cr:64~127

//ALIGNED_DECLARE(uint8_t, pMemPredBlk4[2][16], 16); //I_4x4
  uint8_t* pMemPredBlk4;

  uint8_t* pBestPredI4x4Blk4;//I_4x4

//ALIGNED_DECLARE(uint8_t, pBufferInterPredMe[4][400], 16);//inter type pBuffer for ME h & v & hv
  uint8_t* pBufferInterPredMe;    // [4][400] is enough because only h&v or v&hv or h&hv. but if both h&v&hv is needed when 8 quart pixel, future we have to use [5][400].

//no scan4[] order, just as memory order to store
//ALIGNED_DECLARE(bool, pPrevIntra4x4PredModeFlag[16], 16);//if 1, means no rem_intra4x4_pred_mode; if 0, means rem_intra4x4_pred_mode != 0
  bool* pPrevIntra4x4PredModeFlag;
//ALIGNED_DECLARE(int8_t, pRemIntra4x4PredModeFlag[16], 16);//-1 as default; if pPrevIntra4x4PredModeFlag==0,
//pRemIntra4x4PredModeFlag or added by 1 is the best pred_mode
  int8_t* pRemIntra4x4PredModeFlag;

  int32_t     iSadCostSkip[4];	     //avail 1; unavail 0
  bool      bMbTypeSkip[4];         //1: skip; 0: non-skip
  int32_t*     pEncSad;

//for residual encoding at the side of Encoder
  SDCTCoeff* pDct;

  uint8_t      uiNeighborIntra; // LEFT_MB_POS:0x01, TOP_MB_POS:0x02, TOPLEFT_MB_POS = 0x04 ,TOPRIGHT_MB_POS = 0x08;
  uint8_t uiLumaI16x16Mode;
  uint8_t uiChmaI8x8Mode;

  bool		bCollocatedPredFlag;//denote if current MB is collocated predicted (MV==0).
  uint32_t	uiRefMbType;

  struct {
    /* pointer of current mb location in original frame */
    uint8_t* pEncMb[3];
    /* pointer of current mb location in recovery frame */
    uint8_t* pDecMb[3];
    /* pointer of co-located mb location in reference frame */
    uint8_t* pRefMb[3];
    //for SVC
    uint8_t*	pCsMb[3];//locating current mb's CS in whole frame
//		int16_t *p_rs[3];//locating current mb's RS	in whole frame

  } SPicData;
} SMbCache;

}//end of namespace

#endif//WELS_MACROBLOCK_CACHE_H__