mem_transfer.c

用MPEG-4对YUV视频文件编码压缩成divx视频文件

/*****************************************************************************
 *
 *  XVID MPEG-4 VIDEO CODEC
 *  - 8bit<->16bit transfer  -
 *
 *  This program is an implementation of a part of one or more MPEG-4
 *  Video tools as specified in ISO/IEC 14496-2 standard.  Those intending
 *  to use this software module in hardware or software products are
 *  advised that its use may infringe existing patents or copyrights, and
 *  any such use would be at such party's own risk.  The original
 *  developer of this software module and his/her company, and subsequent
 *  editors and their companies, will have no liability for use of this
 *  software or modifications or derivatives thereof.
 *
 *  This program is free software ; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation ; either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY ; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program ; if not, write to the Free Software
 *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
 *
 ****************************************************************************/
/*****************************************************************************
 *
 *  History
 *
 *  - 14.06.2002 Changed legal header with the new FSF address
 *	- 14.04.2002 added transfer_8to16sub2
 *	- 07.01.2002 merge functions from compensate; rename functions
 *	- 22.12.2001 transfer_8to8add16 limit fix
 *	- 07.11.2001 initial version; (c)2001 peter ross <[e-mail]>
 *
 *  $Id: mem_transfer.c,v 1.4 2002/06/15 22:15:57 edgomez Exp $
 *
 ****************************************************************************/

#include "../global.h"
#include "mem_transfer.h"
#ifdef _TRIMEDIA 
#include <custom_defs.h>
#endif
/* Function pointers - Initialized in the xvid.c module */
/*========8 to 16 copy,for transform====================*/
TRANSFER_8TO16COPY_PTR transfer_8to16copy;
/*========16 to 8 copy,after transform==================*/
TRANSFER_16TO8COPY_PTR transfer_16to8copy;
/*========8 to 16 sub,do in compensate,for transform====*/
TRANSFER_8TO16SUB_PTR  transfer_8to16sub;
/*========16 to 8 add,after transform===================*/
TRANSFER_16TO8ADD_PTR  transfer_16to8add;
/*========8 to 8 copy,for interpolate8x8_switch=========*/
TRANSFER8X8_COPY_PTR transfer8x8_copy;


/*****************************************************************************
 *
 * All these functions are used to transfer data from a 8 bit data array
 * to a 16 bit data array.
 *
 * This is typically used during motion compensation, that's why some
 * functions also do the addition/substraction of another buffer during the
 * so called  transfer.
 *
 ****************************************************************************/

/*!
 ************************************************************************   
 * \brief
 *    the function does the 8->16 bit transfer and this serie of operations 
 *
 ************************************************************************
 */

/*#ifdef _TRIMEDIA*/
void
transfer_8to16copy_c(int16_t * const dst,		/*<--> the destination buffer*/
					 const uint8_t * const src,	/*<--  the source buffer*/
					 uint32_t stride			/*<--  stride*/
					 )
{
	uint32_t i, j;

	for (j = 0; j < 8; j++) {
		for (i = 0; i < 8; i++) {
			dst[j * 8 + i] = (int16_t) src[j * stride + i];
		}
	}
}
/*#endif*/
/*just loop unroll*/
/*#define optimize_8to16copy3*/
#ifdef optimize_8to16copy1
void
transfer_8to16copy_c(int16_t * const dst,
					 const uint8_t * const src,
					 uint32_t stride)
{
	uint32_t i, j,k;

	for ( j = 0; j < 8; j++) {

		
		dst[j * 8 + 0]   = (int16_t) src[j * stride + 0];
		dst[j * 8 + 1]   = (int16_t) src[j * stride + 1];
		dst[j * 8 + 2]   = (int16_t) src[j * stride + 2];
		dst[j * 8 + 3]   = (int16_t) src[j * stride + 3];
		dst[j * 8 + 4]   = (int16_t) src[j * stride + 4];
		dst[j * 8 + 5]   = (int16_t) src[j * stride + 5];
		dst[j * 8 + 6]   = (int16_t) src[j * stride + 6];
		dst[j * 8 + 7]   = (int16_t) src[j * stride + 7];
					
	}
}
#endif
/*use pointer add,but cannot parallel */
#ifdef optimize_8to16copy2
void
transfer_8to16copy_c(int16_t * const dst,
					 const uint8_t * const src,
					 uint32_t stride)
{
	uint32_t i, j;
	int16_t  * ptr_dst = dst;
	uint8_t const * ptr_src = src;

	/*
	for (k = i = j = 0; j < 8; j++) {
		
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		*(ptr_dst + i++)   =  *(ptr_src + k++);
		ptr_dst += 8;
		ptr_src += stride;
					
	}
	*/
	uint32_t stride1 = stride-8;
	
	for (j = 0; j < 8; j++) {
		
		
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		*(ptr_dst++)   =  *(ptr_src++);
		
		ptr_src += stride1;
					
	}
	
}
#endif
#ifdef optimize_8to16copy3
void
transfer_8to16copy_c(int16_t * const dst,
					 const uint8_t * const src,
					 uint32_t stride)
{
	uint32_t i, j;
	int16_t  * ptr_dst = dst;
	uint8_t const * ptr_src = src;

	
	
	
	for (j = 0; j < 8; j++) {
		
		
		*(ptr_dst+0)   =  *(ptr_src+0);
		*(ptr_dst+1)   =  *(ptr_src+1);
		*(ptr_dst+2)   =  *(ptr_src+2);
		*(ptr_dst+3)   =  *(ptr_src+3);
		*(ptr_dst+4)   =  *(ptr_src+4);
		*(ptr_dst+5)   =  *(ptr_src+5);
		*(ptr_dst+6)   =  *(ptr_src+6);
		*(ptr_dst+7)   =  *(ptr_src+7);
		
		ptr_src += stride;
		ptr_dst +=8;
					
	}
	
}
#endif


/*!
 ************************************************************************   
 * \brief
 *    the function does the 16->8 bit transfer and this serie of operations 
 *	  SRC (16bit) = SRC
 *	  DST (8bit)  = max(min(SRC, 255), 0)
 ************************************************************************
 */
#ifdef _TRIMEDIA
#define optimize_16to8copy_3         /*useful*/
#endif
#ifndef _TRIMEDIA
void
transfer_16to8copy_c(uint8_t * const dst,		/*<--> the destination buffer*/
					 const int16_t * const src,	/*<--  the source buffer*/
					 uint32_t stride)			/*<--  stride*/
{
	uint32_t i, j;
	int16_t pixel ;
	

	for (j = 0; j < 8; j++) {
		for (i = 0; i < 8; i++) {
			pixel = src[j * 8 + i];

			if (pixel < 0) {
				pixel = 0;
			} 
			else if (pixel > 255) {
				pixel = 255;
			}
			dst[j * stride + i] = (uint8_t) pixel;
		}
	}
}
#endif

#ifdef optimize_16to8copy_2

void
transfer_16to8copy_c(uint8_t * const dst,		/*<--> the destination buffer*/
					 const int16_t * const src,	/*<--  the source buffer*/
					 uint32_t stride)			/*<--  stride*/
{
	uint32_t i, j;
	int16_t pixel ;
	
	int16_t src0,src1,src2,src3;
	int16_t src4,src5,src6,src7;
	
	
	uint8_t *ptr_dst = dst;
	int16_t *ptr_src = src;
	
	uint32_t clip_src0,clip_src1,clip_src2,clip_src3;
	uint32_t clip_src4,clip_src5,clip_src6,clip_src7;
	
	uint32_t stride1;
	stride1 = stride - 8;

	for (j = 0; j < 8; j++) {
		
		src0 = *(ptr_src++);
		src1 = *(ptr_src++);
		src2 = *(ptr_src++);
		src3 = *(ptr_src++);
		src4 = *(ptr_src++);
		src5 = *(ptr_src++);
		src6 = *(ptr_src++);
		src7 = *(ptr_src++);
		
		
		clip_src0 = UCLIPI(src0,255);
		clip_src1 = UCLIPI(src1,255);
		clip_src2 = UCLIPI(src2,255);
		clip_src3 = UCLIPI(src3,255);
		clip_src4 = UCLIPI(src4,255);
		clip_src5 = UCLIPI(src5,255);
		clip_src6 = UCLIPI(src6,255);
		clip_src7 = UCLIPI(src7,255);
		
		*(ptr_dst++) = (uint8_t)clip_src0;
		*(ptr_dst++) = (uint8_t)clip_src1;
		*(ptr_dst++) = (uint8_t)clip_src2;
		*(ptr_dst++) = (uint8_t)clip_src3;
		*(ptr_dst++) = (uint8_t)clip_src4;
		*(ptr_dst++) = (uint8_t)clip_src5;
		*(ptr_dst++) = (uint8_t)clip_src6; 
		*(ptr_dst++) = (uint8_t)clip_src7;
		ptr_dst += stride1;
	
		/*for (i = 0; i < 8; i++) {
			pixel = src[j * 8 + i];

			if (pixel < 0) {
				pixel = 0;
			} 
			else if (pixel > 255) {
				pixel = 255;
			}
			dst[j * stride + i] = (uint8_t) pixel;
		}*/
		
	}
}
#endif
#ifdef optimize_16to8copy_3

void
transfer_16to8copy_c(uint8_t * const dst,		/*<--> the destination buffer*/
					 const int16_t * const src,	/*<--  the source buffer*/
					 uint32_t stride)			/*<--  stride*/
{
	uint32_t i, j;
	int16_t pixel ;
	
	int16_t src0,src1,src2,src3;
	int16_t src4,src5,src6,src7;
	
	
/*	uint8_t * restrict ptr_dst = dst;
	const int16_t * restrict ptr_src = src;*/
	uint8_t *ptr_dst = dst;
	const int16_t *ptr_src = src;
	
	uint32_t clip_src0,clip_src1,clip_src2,clip_src3;
	uint32_t clip_src4,clip_src5,clip_src6,clip_src7;

	for (j = 0; j < 8; j++) {
		
		src0 = *(ptr_src+0);
		src1 = *(ptr_src+1);
		src2 = *(ptr_src+2);
		src3 = *(ptr_src+3);
		src4 = *(ptr_src+4);
		src5 = *(ptr_src+5);
		src6 = *(ptr_src+6);
		src7 = *(ptr_src+7);
		
		
		clip_src0 = UCLIPI(src0,255);
		clip_src1 = UCLIPI(src1,255);
		clip_src2 = UCLIPI(src2,255);
		clip_src3 = UCLIPI(src3,255);
		clip_src4 = UCLIPI(src4,255);
		clip_src5 = UCLIPI(src5,255);
		clip_src6 = UCLIPI(src6,255);
		clip_src7 = UCLIPI(src7,255);
		
		*(ptr_dst+0) = (uint8_t)clip_src0;
		*(ptr_dst+1) = (uint8_t)clip_src1;
		*(ptr_dst+2) = (uint8_t)clip_src2;
		*(ptr_dst+3) = (uint8_t)clip_src3;
		*(ptr_dst+4) = (uint8_t)clip_src4;
		*(ptr_dst+5) = (uint8_t)clip_src5;
		*(ptr_dst+6) = (uint8_t)clip_src6; 
		*(ptr_dst+7) = (uint8_t)clip_src7;
		ptr_dst += stride;
		ptr_src += 8;
	
		/*for (i = 0; i < 8; i++) {
			pixel = src[j * 8 + i];

			if (pixel < 0) {
				pixel = 0;
			} 
			else if (pixel > 255) {
				pixel = 255;
			}
			dst[j * stride + i] = (uint8_t) pixel;
		}*/
		
	}
}
#endif
#ifdef optimize_16to8copy_1
void
transfer_16to8copy_c(uint8_t * const dst,		/*<--> the destination buffer*/
					 const int16_t * const src,	/*<--  the source buffer*/
					 uint32_t stride)			/*<--  stride*/
{
	uint32_t i, j;
	int16_t pixel ;