view src/video/SDL_blit.c @ 3880:74e828c64315 SDL-1.2

Fixed bug #292 I might be on crack here. It looks like SDL_ConvertMono() in src/audio/SDL_audiocvt.c adds the left and right channels of a stereo stream together, and clamps the new mono channel if it would overflow. Shouldn't it be dividing by 2 to average the two sample points instead of clamping? Otherwise the mono sample point's volume doubles in the conversion. This would also make the conversion faster, as it replaces two branches per sample frame with a bitwise shift. --ryan.
author Sam Lantinga <slouken@libsdl.org>
date Sun, 24 Sep 2006 15:45:37 +0000
parents d910939febfa
children e49147870aac c121d94672cb 6832b00d3594
line wrap: on
line source

/*
    SDL - Simple DirectMedia Layer
    Copyright (C) 1997-2006 Sam Lantinga

    This library is free software; you can redistribute it and/or
    modify it under the terms of the GNU Lesser General Public
    License as published by the Free Software Foundation; either
    version 2.1 of the License, or (at your option) any later version.

    This library 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
    Lesser General Public License for more details.

    You should have received a copy of the GNU Lesser General Public
    License along with this library; if not, write to the Free Software
    Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA

    Sam Lantinga
    slouken@libsdl.org
*/
#include "SDL_config.h"

#include "SDL_video.h"
#include "SDL_sysvideo.h"
#include "SDL_blit.h"
#include "SDL_RLEaccel_c.h"
#include "SDL_pixels_c.h"

#if defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__)) && SDL_ASSEMBLY_ROUTINES
#define MMX_ASMBLIT
#endif

#if defined(MMX_ASMBLIT)
#include "SDL_cpuinfo.h"
#include "mmx.h"
#endif

/* The general purpose software blit routine */
static int SDL_SoftBlit(SDL_Surface *src, SDL_Rect *srcrect,
			SDL_Surface *dst, SDL_Rect *dstrect)
{
	int okay;
	int src_locked;
	int dst_locked;

	/* Everything is okay at the beginning...  */
	okay = 1;

	/* Lock the destination if it's in hardware */
	dst_locked = 0;
	if ( SDL_MUSTLOCK(dst) ) {
		if ( SDL_LockSurface(dst) < 0 ) {
			okay = 0;
		} else {
			dst_locked = 1;
		}
	}
	/* Lock the source if it's in hardware */
	src_locked = 0;
	if ( SDL_MUSTLOCK(src) ) {
		if ( SDL_LockSurface(src) < 0 ) {
			okay = 0;
		} else {
			src_locked = 1;
		}
	}

	/* Set up source and destination buffer pointers, and BLIT! */
	if ( okay  && srcrect->w && srcrect->h ) {
		SDL_BlitInfo info;
		SDL_loblit RunBlit;

		/* Set up the blit information */
		info.s_pixels = (Uint8 *)src->pixels +
				(Uint16)srcrect->y*src->pitch +
				(Uint16)srcrect->x*src->format->BytesPerPixel;
		info.s_width = srcrect->w;
		info.s_height = srcrect->h;
		info.s_skip=src->pitch-info.s_width*src->format->BytesPerPixel;
		info.d_pixels = (Uint8 *)dst->pixels +
				(Uint16)dstrect->y*dst->pitch +
				(Uint16)dstrect->x*dst->format->BytesPerPixel;
		info.d_width = dstrect->w;
		info.d_height = dstrect->h;
		info.d_skip=dst->pitch-info.d_width*dst->format->BytesPerPixel;
		info.aux_data = src->map->sw_data->aux_data;
		info.src = src->format;
		info.table = src->map->table;
		info.dst = dst->format;
		RunBlit = src->map->sw_data->blit;

		/* Run the actual software blit */
		RunBlit(&info);
	}

	/* We need to unlock the surfaces if they're locked */
	if ( dst_locked ) {
		SDL_UnlockSurface(dst);
	}
	if ( src_locked ) {
		SDL_UnlockSurface(src);
	}
	/* Blit is done! */
	return(okay ? 0 : -1);
}

#ifdef MMX_ASMBLIT
static __inline__ void SDL_memcpyMMX(Uint8 *to, const Uint8 *from, int len)
{
	int i;

	for(i=0; i<len/8; i++) {
		__asm__ __volatile__ (
		"	movq (%0), %%mm0\n"
		"	movq %%mm0, (%1)\n"
		: : "r" (from), "r" (to) : "memory");
		from+=8;
		to+=8;
	}
	if (len&7)
		SDL_memcpy(to, from, len&7);
}

static __inline__ void SDL_memcpySSE(Uint8 *to, const Uint8 *from, int len)
{
	int i;

	__asm__ __volatile__ (
	"	prefetchnta (%0)\n"
	"	prefetchnta 64(%0)\n"
	"	prefetchnta 128(%0)\n"
	"	prefetchnta 192(%0)\n"
	: : "r" (from) );

	for(i=0; i<len/8; i++) {
		__asm__ __volatile__ (
		"	prefetchnta 256(%0)\n"
		"	movq (%0), %%mm0\n"
		"	movntq %%mm0, (%1)\n"
		: : "r" (from), "r" (to) : "memory");
		from+=8;
		to+=8;
	}
	if (len&7)
		SDL_memcpy(to, from, len&7);
}
#endif

static void SDL_BlitCopy(SDL_BlitInfo *info)
{
	Uint8 *src, *dst;
	int w, h;
	int srcskip, dstskip;

	w = info->d_width*info->dst->BytesPerPixel;
	h = info->d_height;
	src = info->s_pixels;
	dst = info->d_pixels;
	srcskip = w+info->s_skip;
	dstskip = w+info->d_skip;
#ifdef MMX_ASMBLIT
	if(SDL_HasSSE())
	{
		while ( h-- ) {
			SDL_memcpySSE(dst, src, w);
			src += srcskip;
			dst += dstskip;
		}
		__asm__ __volatile__ (
		"	emms\n"
		::);
	}
	else
	if(SDL_HasMMX())
	{
		while ( h-- ) {
			SDL_memcpyMMX(dst, src, w);
			src += srcskip;
			dst += dstskip;
		}
		__asm__ __volatile__ (
		"	emms\n"
		::);
	}
	else
#endif
	while ( h-- ) {
		SDL_memcpy(dst, src, w);
		src += srcskip;
		dst += dstskip;
	}
}

static void SDL_BlitCopyOverlap(SDL_BlitInfo *info)
{
	Uint8 *src, *dst;
	int w, h;
	int srcskip, dstskip;

	w = info->d_width*info->dst->BytesPerPixel;
	h = info->d_height;
	src = info->s_pixels;
	dst = info->d_pixels;
	srcskip = w+info->s_skip;
	dstskip = w+info->d_skip;
	if ( dst < src ) {
		while ( h-- ) {
			SDL_memcpy(dst, src, w);
			src += srcskip;
			dst += dstskip;
		}
	} else {
		src += ((h-1) * srcskip);
		dst += ((h-1) * dstskip);
		while ( h-- ) {
			SDL_revcpy(dst, src, w);
			src -= srcskip;
			dst -= dstskip;
		}
	}
}

/* Figure out which of many blit routines to set up on a surface */
int SDL_CalculateBlit(SDL_Surface *surface)
{
	int blit_index;

	/* Clean everything out to start */
	if ( (surface->flags & SDL_RLEACCEL) == SDL_RLEACCEL ) {
		SDL_UnRLESurface(surface, 1);
	}
	surface->map->sw_blit = NULL;

	/* Figure out if an accelerated hardware blit is possible */
	surface->flags &= ~SDL_HWACCEL;
	if ( surface->map->identity ) {
		int hw_blit_ok;

		if ( (surface->flags & SDL_HWSURFACE) == SDL_HWSURFACE ) {
			/* We only support accelerated blitting to hardware */
			if ( surface->map->dst->flags & SDL_HWSURFACE ) {
				hw_blit_ok = current_video->info.blit_hw;
			} else {
				hw_blit_ok = 0;
			}
			if (hw_blit_ok && (surface->flags & SDL_SRCCOLORKEY)) {
				hw_blit_ok = current_video->info.blit_hw_CC;
			}
			if ( hw_blit_ok && (surface->flags & SDL_SRCALPHA) ) {
				hw_blit_ok = current_video->info.blit_hw_A;
			}
		} else {
			/* We only support accelerated blitting to hardware */
			if ( surface->map->dst->flags & SDL_HWSURFACE ) {
				hw_blit_ok = current_video->info.blit_sw;
			} else {
				hw_blit_ok = 0;
			}
			if (hw_blit_ok && (surface->flags & SDL_SRCCOLORKEY)) {
				hw_blit_ok = current_video->info.blit_sw_CC;
			}
			if ( hw_blit_ok && (surface->flags & SDL_SRCALPHA) ) {
				hw_blit_ok = current_video->info.blit_sw_A;
			}
		}
		if ( hw_blit_ok ) {
			SDL_VideoDevice *video = current_video;
			SDL_VideoDevice *this  = current_video;
			video->CheckHWBlit(this, surface, surface->map->dst);
		}
	}
	
	/* if an alpha pixel format is specified, we can accelerate alpha blits */
	if (((surface->flags & SDL_HWSURFACE) == SDL_HWSURFACE )&&(current_video->displayformatalphapixel)) 
	{
		if ( (surface->flags & SDL_SRCALPHA) ) 
			if ( current_video->info.blit_hw_A ) {
				SDL_VideoDevice *video = current_video;
				SDL_VideoDevice *this  = current_video;
				video->CheckHWBlit(this, surface, surface->map->dst);
			}
	}

	/* Get the blit function index, based on surface mode */
	/* { 0 = nothing, 1 = colorkey, 2 = alpha, 3 = colorkey+alpha } */
	blit_index = 0;
	blit_index |= (!!(surface->flags & SDL_SRCCOLORKEY))      << 0;
	if ( surface->flags & SDL_SRCALPHA
	     && (surface->format->alpha != SDL_ALPHA_OPAQUE
		 || surface->format->Amask) ) {
	        blit_index |= 2;
	}

	/* Check for special "identity" case -- copy blit */
	if ( surface->map->identity && blit_index == 0 ) {
	        surface->map->sw_data->blit = SDL_BlitCopy;

		/* Handle overlapping blits on the same surface */
		if ( surface == surface->map->dst ) {
		        surface->map->sw_data->blit = SDL_BlitCopyOverlap;
		}
	} else {
		if ( surface->format->BitsPerPixel < 8 ) {
			surface->map->sw_data->blit =
			    SDL_CalculateBlit0(surface, blit_index);
		} else {
			switch ( surface->format->BytesPerPixel ) {
			    case 1:
				surface->map->sw_data->blit =
				    SDL_CalculateBlit1(surface, blit_index);
				break;
			    case 2:
			    case 3:
			    case 4:
				surface->map->sw_data->blit =
				    SDL_CalculateBlitN(surface, blit_index);
				break;
			    default:
				surface->map->sw_data->blit = NULL;
				break;
			}
		}
	}
	/* Make sure we have a blit function */
	if ( surface->map->sw_data->blit == NULL ) {
		SDL_InvalidateMap(surface->map);
		SDL_SetError("Blit combination not supported");
		return(-1);
	}

	/* Choose software blitting function */
	if(surface->flags & SDL_RLEACCELOK
	   && (surface->flags & SDL_HWACCEL) != SDL_HWACCEL) {

	        if(surface->map->identity
		   && (blit_index == 1
		       || (blit_index == 3 && !surface->format->Amask))) {
		        if ( SDL_RLESurface(surface) == 0 )
			        surface->map->sw_blit = SDL_RLEBlit;
		} else if(blit_index == 2 && surface->format->Amask) {
		        if ( SDL_RLESurface(surface) == 0 )
			        surface->map->sw_blit = SDL_RLEAlphaBlit;
		}
	}
	
	if ( surface->map->sw_blit == NULL ) {
		surface->map->sw_blit = SDL_SoftBlit;
	}
	return(0);
}