jccoefct.c

基于Linux的ffmepg decoder

/*
 * jccoefct.c
 *
 * Copyright (C) 1994-1997, Thomas G. Lane.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
 * This file contains the coefficient buffer controller for compression.
 * This controller is the top level of the JPEG compressor proper.
 * The coefficient buffer lies between forward-DCT and entropy encoding steps.
 */

#define JPEG_INTERNALS
#include "jinclude.h"
#include "jpeglib.h"
#include "vpe_m.h"
#include "local_mem.h"


/* We use a full-image coefficient buffer when doing Huffman optimization,
 * and also for writing multiple-scan JPEG files.  In all cases, the DCT
 * step is run during the first pass, and subsequent passes need only read
 * the buffered coefficients.
 */
#ifdef ENTROPY_OPT_SUPPORTED
#define FULL_COEF_BUFFER_SUPPORTED
#else
#ifdef C_MULTISCAN_FILES_SUPPORTED
#define FULL_COEF_BUFFER_SUPPORTED
#endif
#endif


/* Private buffer controller object */

typedef struct {
  struct jpeg_c_coef_controller pub; /* public fields */

  JDIMENSION iMCU_row_num;	/* iMCU row # within image */
  JDIMENSION mcu_ctr;		/* counts MCUs processed in current row */
  int MCU_vert_offset;		/* counts MCU rows within iMCU row */
  int MCU_rows_per_iMCU_row;	/* number of such rows needed */

  /* For single-pass compression, it's sufficient to buffer just one MCU
   * (although this may prove a bit slow in practice).  We allocate a
   * workspace of C_MAX_BLOCKS_IN_MCU coefficient blocks, and reuse it for each
   * MCU constructed and sent.  (On 80x86, the workspace is FAR even though
   * it's not really very big; this is to keep the module interfaces unchanged
   * when a large coefficient buffer is necessary.)
   * In multi-pass modes, this array points to the current MCU's blocks
   * within the virtual arrays.
   */
  JBLOCKROW MCU_buffer[C_MAX_BLOCKS_IN_MCU];

  /* In multi-pass modes, we need a virtual block array for each component. */
  jvirt_barray_ptr whole_image[MAX_COMPONENTS];
} my_coef_controller;

typedef my_coef_controller * my_coef_ptr;



LOCAL(void)
start_iMCU_row (j_compress_ptr cinfo)
/* Reset within-iMCU-row counters for a new row */
{
  my_coef_ptr coef = (my_coef_ptr) cinfo->coef;

  /* In an interleaved scan, an MCU row is the same as an iMCU row.
   * In a noninterleaved scan, an iMCU row has v_samp_factor MCU rows.
   * But at the bottom of the image, process only what's left.
   */
  if (cinfo->comps_in_scan > 1) {
    coef->MCU_rows_per_iMCU_row = 1;
  } else {
    if (coef->iMCU_row_num < (cinfo->total_iMCU_rows-1))
      coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->v_samp_factor;
    else
      coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->last_row_height;
  }

  coef->mcu_ctr = 0;
  coef->MCU_vert_offset = 0;
}


/*
 * Initialize for a processing pass.
 */

METHODDEF(void)
start_pass_coef (j_compress_ptr cinfo, J_BUF_MODE pass_mode)
{
  my_coef_ptr coef = (my_coef_ptr) cinfo->coef;

  coef->iMCU_row_num = 0;
  start_iMCU_row(cinfo);

}


/*
 * Process some data in the single-pass case.
 * We process the equivalent of one fully interleaved MCU row ("iMCU" row)
 * per call, ie, v_samp_factor block rows for each component in the image.
 * Returns TRUE if the iMCU row is completed, FALSE if suspended.
 *
 * NB: input_buf contains a plane for each component in image,
 * which we index according to the component's SOF position.
 */

boolean compress_data1 (j_compress_ptr cinfo)
{
  my_coef_ptr coef = (my_coef_ptr) cinfo->coef;
  JDIMENSION MCU_col_num;	/* index of current MCU within row */
  JDIMENSION last_MCU_col = cinfo->MCUs_per_row - 1; 
  //unsigned int DCT_col_num; 
  volatile MDMA *pmdma = MDMA1;
  int i;


     for (MCU_col_num = coef->mcu_ctr; MCU_col_num <= last_MCU_col;
	 MCU_col_num++) {		//pwhsu:20031017	For this case:0~14

		 if(rinfo.nCount==0){
			
			DMA_mcu(cinfo, MCU_col_num, last_MCU_col, rinfo.Dma_bnum);
			while((pmdma->Status & 0x1) == 0){}
			pmdma->SMaddr =
			pmdma->CCA = (unsigned int)DMA_COMMAND_system_phy;
//			pmdma->LMaddr = DMA_COMMAND_local+rinfo.Dma_bnum*40;
			pmdma->LMaddr = DMA_COMMAND_local_OFF+rinfo.Dma_bnum*40 * 4;
			pmdma->BlkWidth = 0;
			pmdma->Control =  dmactrl;//0x4B00000 | (rinfo.mcublkn)*4;		//start DMA	

			rinfo.Dma_bnum =  (rinfo.Dma_bnum+1)&1;
			mVpe_Indicator(0x90000000 | cMCUrow<<4 | MCU_col_num);	//pwhsu++:20040128
			
			 if(rinfo.nCount==rinfo.MCUnum-1) {
				mFa526DrainWrBuf ();

				while((pmdma->Status & 0x1) == 0){}
				mVpe_Indicator (0x90000000 | cMCUrow<<4 | MCU_col_num);	//pwhsu++:20040128
				pbufaddr = (unsigned int *)(cur_b0 + rinfo.Coef_bnum * stride_MCU);			 
				rinfo.Coef_bnum =  (rinfo.Coef_bnum+1)&1;  //switch the pixel buffer number
				SET_MCCADDR(pbufaddr);
				SET_MCCTL(mcctrl);

				mFa526DrainWrBuf ();

				do {
					READ_VLDSTS(vldreg);
				} while ((vldreg & 32) == 0);


			}

		 }else if (rinfo.nCount==1){

			DMA_mcu(cinfo, MCU_col_num, last_MCU_col, rinfo.Dma_bnum);	//gen DMA command queue

			while((pmdma->Status & 0x1) == 0)		//DMA process run
			{}
			pmdma->SMaddr =
			pmdma->CCA = (unsigned int)DMA_COMMAND_system_phy;
//			pmdma->LMaddr = DMA_COMMAND_local+rinfo.Dma_bnum*40;
			pmdma->LMaddr = DMA_COMMAND_local_OFF+rinfo.Dma_bnum*40 * 4;
			pmdma->BlkWidth = 0;
			pmdma->Control =  dmactrl;//0x4B00000 | (rinfo.mcublkn)*4;		//start DMA	

			mVpe_Indicator (0x90000000 | cMCUrow<<4 | MCU_col_num);	

			rinfo.Dma_bnum =  (rinfo.Dma_bnum+1)&1;

  			pbufaddr = (unsigned int *)(cur_b0 + rinfo.Coef_bnum * stride_MCU);		//MC_go
			rinfo.Coef_bnum =  (rinfo.Coef_bnum+1)&1;

			SET_MCCADDR(pbufaddr);
			SET_MCCTL(mcctrl);			

  			if(rinfo.nCount==rinfo.MCUnum-1){
				mFa526DrainWrBuf ();
				while((pmdma->Status & 0x1) == 0){}

				mVpe_Indicator ( 0x90000000 | cMCUrow<<4 | MCU_col_num);	//pwhsu++:20040128

				//Check VLC is done
				do {
					READ_CPSTS(sreg);
				} while ((sreg & 32) == 0);


				//pwhsu++:2040607	VAD found bug
				Encode_mcu (cinfo);       
				mFa526DrainWrBuf ();
				//pwhsu++:2040607	VAD found bug

				pbufaddr = (unsigned int *)(cur_b0 + rinfo.Coef_bnum * stride_MCU);			 
				rinfo.Coef_bnum =  (rinfo.Coef_bnum+1)&1;  //switch the pixel buffer number
				SET_MCCADDR(pbufaddr);
				SET_MCCTL(mcctrl);

				mFa526DrainWrBuf ();
				do {
					READ_CPSTS(sreg);
				} while ((sreg & 32) == 0);

			}


		}else {

			DMA_mcu(cinfo, MCU_col_num, last_MCU_col, rinfo.Dma_bnum);	//gen DMA command queue
			mFa526DrainWrBuf ();
			do {
				READ_CPSTS(sreg);
			} while ((sreg & 32) == 0);

			Encode_mcu (cinfo);
			while((pmdma->Status & 0x1) == 0)			//DMA process run
				{}
			pmdma->SMaddr =
			pmdma->CCA = (unsigned int)DMA_COMMAND_system_phy;
//			pmdma->LMaddr = DMA_COMMAND_local+rinfo.Dma_bnum*40;
			pmdma->LMaddr = DMA_COMMAND_local_OFF+rinfo.Dma_bnum*40 * 4;
			pmdma->BlkWidth = 0;
			pmdma->Control =  dmactrl;//0x4B00000 | (rinfo.mcublkn)*4;		//start DMA	

			rinfo.Dma_bnum =  (rinfo.Dma_bnum+1)&1;

			mVpe_Indicator ( 0x90000000 | cMCUrow<<4 | MCU_col_num);

  			pbufaddr = (unsigned int *)(cur_b0 + rinfo.Coef_bnum * stride_MCU);		//MC_go
			rinfo.Coef_bnum =  (rinfo.Coef_bnum+1)&1;

			SET_MCCADDR(pbufaddr);
			SET_MCCTL(mcctrl)	;			
			

  			if(rinfo.nCount==rinfo.MCUnum-1){

				mFa526DrainWrBuf ();

				//while((pmdma->Status & 0x1) == 0){}
				while((pmdma->Status & 0x1) == 0)			//DMA process run
					{}

				mVpe_Indicator (0x90000000 | cMCUrow<<4 | MCU_col_num);

				do {
					READ_CPSTS(sreg);
				} while ((sreg & 32) == 0);

				//pwhsu++:2040607	VAD found bug
				Encode_mcu (cinfo);       
				mFa526DrainWrBuf ();
				//pwhsu++:2040607	VAD found bug

				pbufaddr = (unsigned int *)(cur_b0 + rinfo.Coef_bnum * stride_MCU);			 
				rinfo.Coef_bnum =  (rinfo.Coef_bnum+1)&1;  //switch the pixel buffer number
				SET_MCCADDR(pbufaddr);
				SET_MCCTL(mcctrl);

				mFa526DrainWrBuf ();
				do {
					READ_CPSTS(sreg);
				} while ((sreg & 32) == 0);
			}
		}
		//pwhsu++:20031031
		rinfo.nCount+=1;//pwhsu++:20031030
	     }//end of for MCU_col_num
  return TRUE;
}

/*
 * Initialize coefficient buffer controller.
 */

GLOBAL(void)
jinit_c_coef_controller (j_compress_ptr cinfo, boolean need_full_buffer)
{
  my_coef_ptr coef;

  coef = (my_coef_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
				SIZEOF(my_coef_controller));
  cinfo->coef = (struct jpeg_c_coef_controller *) coef;
  coef->pub.start_pass = start_pass_coef;

}



void
DMA_mcu(j_compress_ptr cinfo,unsigned int MCU_col_num, 
			unsigned int last_MCU_col, int dbnum)
{
	int blkn, xpos, ypos, bi, ci, blockcnt, yindex;
	jpeg_component_info *compptr;
	int cmdidx;
	unsigned char* puchexin, *pucdataptr;
	unsigned int uintpos;
	int sysdataidx;
	blkn = 0;	
	sysdataidx = dbnum*40;

	for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
		puchexin = curdata[ci];
		compptr = cinfo->cur_comp_info[ci];
		blockcnt = compptr->MCU_width;
		xpos = MCU_col_num * compptr->MCU_sample_width;
		ypos = 0;
		for (yindex = 0; yindex < compptr->MCU_height; yindex++) {

			uintpos = puchexin + (ypos+cMCUrow*mcu_height[ci])*comp_width[ci];

			for (bi = 0; bi < blockcnt; bi++, xpos += DCTSIZE) {  
				
				//DMA_access(ypos ,xpos, blkn, bi, dbnum, ci/*, cmdidx*/);
				
				pucdataptr = uintpos + xpos;
				//DMA_COMMAND_local[0+(blkn+bi)*4+dbnum*40] = (unsigned int) pucdataptr;
				DMA_COMMAND_local[sysdataidx] = (unsigned int) pucdataptr;

				//puchexin += (start_row + (cMCUrow*v_sampf[ci])*8) * (comp_width[ci]) + start_col;
				//DMA_COMMAND_local[0+(blkn+bi)*4+dbnum*40] = (unsigned int) puchexin;

				//move one block per comment  //BlkWidth
				//DMA_COMMAND_local[2+(blkn+bi)*4+dbnum*40] = ((( cinfo->MCUs_per_row * blockcnt)*2-1)<<8) | 0x02;

				sysdataidx += 4;
		
			} //end of block num
			xpos = MCU_col_num * compptr->MCU_sample_width;

			blkn += compptr->MCU_width;
			ypos += DCTSIZE;
		}// end of yindex
	}// end of ci

}