Mercurial > sdl-ios-xcode

view src/video/SDL_blit_A.c @ 563:04dcaf3da918

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Massive Quartz input enhancements from Darrell Walisser. His email: Enclosed is a patch that addresses the following: --Various minor cleanups. Removed dead/obsolete code, made some style cleanups --Mouse Events Now keep track of what button(s) were pressed so we know when to send the mouse up event. This fixes the case where the mouse is dragged outside of the game window and released (in which case we want to send the mouse up event even though the mouse is outside the game window). --Input Grabbing Here is my take on the grabbing situation, which is the basis for the new implementation. There are 3 grab states, ungrabbed (UG), visible (VG), and invisible (IG). Both VG and IG keep the mouse constrained to the window and produce relative motion events. In VG the cursor is visible (duh), in IG it is not. In VG, absolute motion events also work. There are 6 actions that can affect grabbing: 1. Set Fullscreen/Window (F/W). In fullscreen, a visible grab should do nothing. However, a fullscreen visible grab can be treated just like a windowed visible grab, which is what I have done to help simplify things. 2. Cursor hide/show (H/S). If the cursor is hidden when grabbing, the grab is an invisible grab. If the cursor is visible, the grab should just constrain the mouse to the window. 3. Input grab/ungrab(G/U). If grabbed, the cursor should be confined to the window as should the keyboard input. On Mac OS X, the keyboard input is implicitly grabbed by confining the cursor, except for command-tab which can switch away from the application. Should the window come to the foreground if the application is deactivated and grab input is called? This isn't necessary in this implementation because the grab state will be asserted upon activation. Using my notation, these are all the cases that need to be handled (state + action = new state). UG+U = UG UG+G = VG or IG, if cursor is visible or not UG+H = UG UG+S = UG VG+U = UG VG+G = VG VG+H = IG VG+S = VG IG+U = UG IG+G = IG IG+H = IG IG+S = VG The cases that result in the same state can be ignored in the code, which cuts it down to just 5 cases. Another issue is what happens when the app loses/gains input focus from deactivate/activate or iconify/deiconify. I think that if input focus is ever lost (outside of SDL's control), the grab state should be suspended and the cursor should become visible and active again. When regained, the cursor should reappear in its original location and/or grab state. This way, when reactivating the cursor is still in the same position as before so apps shouldn't get confused when the next motion event comes in. This is what I've done in this patch.
author Ryan C. Gordon <icculus@icculus.org>
date Fri, 27 Dec 2002 20:52:41 +0000
parents f6ffac90895c
children 5bb080d35049
line wrap: on
line source

/*
    SDL - Simple DirectMedia Layer
    Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002  Sam Lantinga

    This library is free software; you can redistribute it and/or
    modify it under the terms of the GNU Library General Public
    License as published by the Free Software Foundation; either
    version 2 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
    Library General Public License for more details.

    You should have received a copy of the GNU Library General Public
    License along with this library; if not, write to the Free
    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

    Sam Lantinga
    slouken@libsdl.org
*/

#ifdef SAVE_RCSID
static char rcsid =
 "@(#) $Id$";
#endif

#include <stdio.h>

#include "SDL_types.h"
#include "SDL_video.h"
#include "SDL_blit.h"

/* Functions to perform alpha blended blitting */

/* N->1 blending with per-surface alpha */
static void BlitNto1SurfaceAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint8 *src = info->s_pixels;
	int srcskip = info->s_skip;
	Uint8 *dst = info->d_pixels;
	int dstskip = info->d_skip;
	Uint8 *palmap = info->table;
	SDL_PixelFormat *srcfmt = info->src;
	SDL_PixelFormat *dstfmt = info->dst;
	int srcbpp = srcfmt->BytesPerPixel;

	const unsigned A = srcfmt->alpha;

	while ( height-- ) {
	    DUFFS_LOOP4(
	    {
		Uint32 pixel;
		unsigned sR;
		unsigned sG;
		unsigned sB;
		unsigned dR;
		unsigned dG;
		unsigned dB;
		DISEMBLE_RGB(src, srcbpp, srcfmt, pixel, sR, sG, sB);
		dR = dstfmt->palette->colors[*dst].r;
		dG = dstfmt->palette->colors[*dst].g;
		dB = dstfmt->palette->colors[*dst].b;
		ALPHA_BLEND(sR, sG, sB, A, dR, dG, dB);
		dR &= 0xff;
		dG &= 0xff;
		dB &= 0xff;
		/* Pack RGB into 8bit pixel */
		if ( palmap == NULL ) {
		    *dst =((dR>>5)<<(3+2))|
			  ((dG>>5)<<(2))|
			  ((dB>>6)<<(0));
		} else {
		    *dst = palmap[((dR>>5)<<(3+2))|
				  ((dG>>5)<<(2))  |
				  ((dB>>6)<<(0))];
		}
		dst++;
		src += srcbpp;
	    },
	    width);
	    src += srcskip;
	    dst += dstskip;
	}
}

/* N->1 blending with pixel alpha */
static void BlitNto1PixelAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint8 *src = info->s_pixels;
	int srcskip = info->s_skip;
	Uint8 *dst = info->d_pixels;
	int dstskip = info->d_skip;
	Uint8 *palmap = info->table;
	SDL_PixelFormat *srcfmt = info->src;
	SDL_PixelFormat *dstfmt = info->dst;
	int srcbpp = srcfmt->BytesPerPixel;

	/* FIXME: fix alpha bit field expansion here too? */
	while ( height-- ) {
	    DUFFS_LOOP4(
	    {
		Uint32 pixel;
		unsigned sR;
		unsigned sG;
		unsigned sB;
		unsigned sA;
		unsigned dR;
		unsigned dG;
		unsigned dB;
		DISEMBLE_RGBA(src,srcbpp,srcfmt,pixel,sR,sG,sB,sA);
		dR = dstfmt->palette->colors[*dst].r;
		dG = dstfmt->palette->colors[*dst].g;
		dB = dstfmt->palette->colors[*dst].b;
		ALPHA_BLEND(sR, sG, sB, sA, dR, dG, dB);
		dR &= 0xff;
		dG &= 0xff;
		dB &= 0xff;
		/* Pack RGB into 8bit pixel */
		if ( palmap == NULL ) {
		    *dst =((dR>>5)<<(3+2))|
			  ((dG>>5)<<(2))|
			  ((dB>>6)<<(0));
		} else {
		    *dst = palmap[((dR>>5)<<(3+2))|
				  ((dG>>5)<<(2))  |
				  ((dB>>6)<<(0))  ];
		}
		dst++;
		src += srcbpp;
	    },
	    width);
	    src += srcskip;
	    dst += dstskip;
	}
}

/* colorkeyed N->1 blending with per-surface alpha */
static void BlitNto1SurfaceAlphaKey(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint8 *src = info->s_pixels;
	int srcskip = info->s_skip;
	Uint8 *dst = info->d_pixels;
	int dstskip = info->d_skip;
	Uint8 *palmap = info->table;
	SDL_PixelFormat *srcfmt = info->src;
	SDL_PixelFormat *dstfmt = info->dst;
	int srcbpp = srcfmt->BytesPerPixel;
	Uint32 ckey = srcfmt->colorkey;

	const int A = srcfmt->alpha;

	while ( height-- ) {
	    DUFFS_LOOP(
	    {
		Uint32 pixel;
		unsigned sR;
		unsigned sG;
		unsigned sB;
		unsigned dR;
		unsigned dG;
		unsigned dB;
		DISEMBLE_RGB(src, srcbpp, srcfmt, pixel, sR, sG, sB);
		if ( pixel != ckey ) {
		    dR = dstfmt->palette->colors[*dst].r;
		    dG = dstfmt->palette->colors[*dst].g;
		    dB = dstfmt->palette->colors[*dst].b;
		    ALPHA_BLEND(sR, sG, sB, A, dR, dG, dB);
		    dR &= 0xff;
		    dG &= 0xff;
		    dB &= 0xff;
		    /* Pack RGB into 8bit pixel */
		    if ( palmap == NULL ) {
			*dst =((dR>>5)<<(3+2))|
			      ((dG>>5)<<(2)) |
			      ((dB>>6)<<(0));
		    } else {
			*dst = palmap[((dR>>5)<<(3+2))|
				      ((dG>>5)<<(2))  |
				      ((dB>>6)<<(0))  ];
		    }
		}
		dst++;
		src += srcbpp;
	    },
	    width);
	    src += srcskip;
	    dst += dstskip;
	}
}

/* fast RGB888->(A)RGB888 blending with surface alpha=128 special case */
static void BlitRGBtoRGBSurfaceAlpha128(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint32 *srcp = (Uint32 *)info->s_pixels;
	int srcskip = info->s_skip >> 2;
	Uint32 *dstp = (Uint32 *)info->d_pixels;
	int dstskip = info->d_skip >> 2;

	while(height--) {
	    DUFFS_LOOP4({
		    Uint32 s = *srcp++;
		    Uint32 d = *dstp;
		    *dstp++ = ((((s & 0x00fefefe) + (d & 0x00fefefe)) >> 1)
			       + (s & d & 0x00010101)) | 0xff000000;
	    }, width);
	    srcp += srcskip;
	    dstp += dstskip;
	}
}

/* fast RGB888->(A)RGB888 blending with surface alpha */
static void BlitRGBtoRGBSurfaceAlpha(SDL_BlitInfo *info)
{
	unsigned alpha = info->src->alpha;
	if(alpha == 128) {
		BlitRGBtoRGBSurfaceAlpha128(info);
	} else {
		int width = info->d_width;
		int height = info->d_height;
		Uint32 *srcp = (Uint32 *)info->s_pixels;
		int srcskip = info->s_skip >> 2;
		Uint32 *dstp = (Uint32 *)info->d_pixels;
		int dstskip = info->d_skip >> 2;

		while(height--) {
			DUFFS_LOOP4({
				Uint32 s;
				Uint32 d;
				Uint32 s1;
				Uint32 d1;
				s = *srcp;
				d = *dstp;
				s1 = s & 0xff00ff;
				d1 = d & 0xff00ff;
				d1 = (d1 + ((s1 - d1) * alpha >> 8))
				     & 0xff00ff;
				s &= 0xff00;
				d &= 0xff00;
				d = (d + ((s - d) * alpha >> 8)) & 0xff00;
				*dstp = d1 | d | 0xff000000;
				++srcp;
				++dstp;
			}, width);
			srcp += srcskip;
			dstp += dstskip;
		}
	}
}

/* fast ARGB888->(A)RGB888 blending with pixel alpha */
static void BlitRGBtoRGBPixelAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint32 *srcp = (Uint32 *)info->s_pixels;
	int srcskip = info->s_skip >> 2;
	Uint32 *dstp = (Uint32 *)info->d_pixels;
	int dstskip = info->d_skip >> 2;

	while(height--) {
	    DUFFS_LOOP4({
		Uint32 dalpha;
		Uint32 d;
		Uint32 s1;
		Uint32 d1;
		Uint32 s = *srcp;
		Uint32 alpha = s >> 24;
		/* FIXME: Here we special-case opaque alpha since the
		   compositioning used (>>8 instead of /255) doesn't handle
		   it correctly. Also special-case alpha=0 for speed?
		   Benchmark this! */
		if(alpha == SDL_ALPHA_OPAQUE) {
		    *dstp = (s & 0x00ffffff) | (*dstp & 0xff000000);
		} else {
		    /*
		     * take out the middle component (green), and process
		     * the other two in parallel. One multiply less.
		     */
		    d = *dstp;
		    dalpha = d & 0xff000000;
		    s1 = s & 0xff00ff;
		    d1 = d & 0xff00ff;
		    d1 = (d1 + ((s1 - d1) * alpha >> 8)) & 0xff00ff;
		    s &= 0xff00;
		    d &= 0xff00;
		    d = (d + ((s - d) * alpha >> 8)) & 0xff00;
		    *dstp = d1 | d | dalpha;
		}
		++srcp;
		++dstp;
	    }, width);
	    srcp += srcskip;
	    dstp += dstskip;
	}
}

/* 16bpp special case for per-surface alpha=50%: blend 2 pixels in parallel */

/* blend a single 16 bit pixel at 50% */
#define BLEND16_50(d, s, mask)						\
	((((s & mask) + (d & mask)) >> 1) + (s & d & (~mask & 0xffff)))

/* blend two 16 bit pixels at 50% */
#define BLEND2x16_50(d, s, mask)					     \
	(((s & (mask | mask << 16)) >> 1) + ((d & (mask | mask << 16)) >> 1) \
	 + (s & d & (~(mask | mask << 16))))

static void Blit16to16SurfaceAlpha128(SDL_BlitInfo *info, Uint16 mask)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint16 *srcp = (Uint16 *)info->s_pixels;
	int srcskip = info->s_skip >> 1;
	Uint16 *dstp = (Uint16 *)info->d_pixels;
	int dstskip = info->d_skip >> 1;

	while(height--) {
		if(((unsigned long)srcp ^ (unsigned long)dstp) & 2) {
			/*
			 * Source and destination not aligned, pipeline it.
			 * This is mostly a win for big blits but no loss for
			 * small ones
			 */
			Uint32 prev_sw;
			int w = width;

			/* handle odd destination */
			if((unsigned long)dstp & 2) {
				Uint16 d = *dstp, s = *srcp;
				*dstp = BLEND16_50(d, s, mask);
				dstp++;
				srcp++;
				w--;
			}
			srcp++;	/* srcp is now 32-bit aligned */

			/* bootstrap pipeline with first halfword */
			prev_sw = ((Uint32 *)srcp)[-1];

			while(w > 1) {
				Uint32 sw, dw, s;
				sw = *(Uint32 *)srcp;
				dw = *(Uint32 *)dstp;
				if(SDL_BYTEORDER == SDL_BIG_ENDIAN)
					s = (prev_sw << 16) + (sw >> 16);
				else
					s = (prev_sw >> 16) + (sw << 16);
				prev_sw = sw;
				*(Uint32 *)dstp = BLEND2x16_50(dw, s, mask);
				dstp += 2;
				srcp += 2;
				w -= 2;
			}

			/* final pixel if any */
			if(w) {
				Uint16 d = *dstp, s;
				if(SDL_BYTEORDER == SDL_BIG_ENDIAN)
					s = prev_sw;
				else
					s = prev_sw >> 16;
				*dstp = BLEND16_50(d, s, mask);
				srcp++;
				dstp++;
			}
			srcp += srcskip - 1;
			dstp += dstskip;
		} else {
			/* source and destination are aligned */
			int w = width;

			/* first odd pixel? */
			if((unsigned long)srcp & 2) {
				Uint16 d = *dstp, s = *srcp;
				*dstp = BLEND16_50(d, s, mask);
				srcp++;
				dstp++;
				w--;
			}
			/* srcp and dstp are now 32-bit aligned */

			while(w > 1) {
				Uint32 sw = *(Uint32 *)srcp;
				Uint32 dw = *(Uint32 *)dstp;
				*(Uint32 *)dstp = BLEND2x16_50(dw, sw, mask);
				srcp += 2;
				dstp += 2;
				w -= 2;
			}

			/* last odd pixel? */
			if(w) {
				Uint16 d = *dstp, s = *srcp;
				*dstp = BLEND16_50(d, s, mask);
				srcp++;
				dstp++;
			}
			srcp += srcskip;
			dstp += dstskip;
		}
	}
}

/* fast RGB565->RGB565 blending with surface alpha */
static void Blit565to565SurfaceAlpha(SDL_BlitInfo *info)
{
	unsigned alpha = info->src->alpha;
	if(alpha == 128) {
		Blit16to16SurfaceAlpha128(info, 0xf7de);
	} else {
		int width = info->d_width;
		int height = info->d_height;
		Uint16 *srcp = (Uint16 *)info->s_pixels;
		int srcskip = info->s_skip >> 1;
		Uint16 *dstp = (Uint16 *)info->d_pixels;
		int dstskip = info->d_skip >> 1;
		alpha >>= 3;	/* downscale alpha to 5 bits */

		while(height--) {
			DUFFS_LOOP4({
				Uint32 s = *srcp++;
				Uint32 d = *dstp;
				/*
				 * shift out the middle component (green) to
				 * the high 16 bits, and process all three RGB
				 * components at the same time.
				 */
				s = (s | s << 16) & 0x07e0f81f;
				d = (d | d << 16) & 0x07e0f81f;
				d += (s - d) * alpha >> 5;
				d &= 0x07e0f81f;
				*dstp++ = d | d >> 16;
			}, width);
			srcp += srcskip;
			dstp += dstskip;
		}
	}
}

/* fast RGB555->RGB555 blending with surface alpha */
static void Blit555to555SurfaceAlpha(SDL_BlitInfo *info)
{
	unsigned alpha = info->src->alpha; /* downscale alpha to 5 bits */
	if(alpha == 128) {
		Blit16to16SurfaceAlpha128(info, 0xfbde);
	} else {
		int width = info->d_width;
		int height = info->d_height;
		Uint16 *srcp = (Uint16 *)info->s_pixels;
		int srcskip = info->s_skip >> 1;
		Uint16 *dstp = (Uint16 *)info->d_pixels;
		int dstskip = info->d_skip >> 1;
		alpha >>= 3;		/* downscale alpha to 5 bits */

		while(height--) {
			DUFFS_LOOP4({
				Uint32 s = *srcp++;
				Uint32 d = *dstp;
				/*
				 * shift out the middle component (green) to
				 * the high 16 bits, and process all three RGB
				 * components at the same time.
				 */
				s = (s | s << 16) & 0x03e07c1f;
				d = (d | d << 16) & 0x03e07c1f;
				d += (s - d) * alpha >> 5;
				d &= 0x03e07c1f;
				*dstp++ = d | d >> 16;
			}, width);
			srcp += srcskip;
			dstp += dstskip;
		}
	}
}

/* fast ARGB8888->RGB565 blending with pixel alpha */
static void BlitARGBto565PixelAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint32 *srcp = (Uint32 *)info->s_pixels;
	int srcskip = info->s_skip >> 2;
	Uint16 *dstp = (Uint16 *)info->d_pixels;
	int dstskip = info->d_skip >> 1;

	while(height--) {
	    DUFFS_LOOP4({
		Uint32 s = *srcp;
		unsigned alpha = s >> 27; /* downscale alpha to 5 bits */
		/* FIXME: Here we special-case opaque alpha since the
		   compositioning used (>>8 instead of /255) doesn't handle
		   it correctly. Also special-case alpha=0 for speed?
		   Benchmark this! */
		if(alpha == (SDL_ALPHA_OPAQUE >> 3)) {
		    *dstp = (s >> 8 & 0xf800) + (s >> 5 & 0x7e0)
			  + (s >> 3  & 0x1f);
		} else {
		    Uint32 d = *dstp;
		    /*
		     * convert source and destination to G0RAB65565
		     * and blend all components at the same time
		     */
		    s = ((s & 0xfc00) << 11) + (s >> 8 & 0xf800)
		      + (s >> 3 & 0x1f);
		    d = (d | d << 16) & 0x07e0f81f;
		    d += (s - d) * alpha >> 5;
		    d &= 0x07e0f81f;
		    *dstp = d | d >> 16;
		}
		srcp++;
		dstp++;
	    }, width);
	    srcp += srcskip;
	    dstp += dstskip;
	}
}

/* fast ARGB8888->RGB555 blending with pixel alpha */
static void BlitARGBto555PixelAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint32 *srcp = (Uint32 *)info->s_pixels;
	int srcskip = info->s_skip >> 2;
	Uint16 *dstp = (Uint16 *)info->d_pixels;
	int dstskip = info->d_skip >> 1;

	while(height--) {
	    DUFFS_LOOP4({
		unsigned alpha;
		Uint32 s = *srcp;
		alpha = s >> 27; /* downscale alpha to 5 bits */
		/* FIXME: Here we special-case opaque alpha since the
		   compositioning used (>>8 instead of /255) doesn't handle
		   it correctly. Also special-case alpha=0 for speed?
		   Benchmark this! */
		if(alpha == (SDL_ALPHA_OPAQUE >> 3)) {
		    *dstp = (s >> 9 & 0x7c00) + (s >> 6 & 0x3e0)
			  + (s >> 3  & 0x1f);
		} else {
		    Uint32 d = *dstp;
		    /*
		     * convert source and destination to G0RAB65565
		     * and blend all components at the same time
		     */
		    s = ((s & 0xf800) << 10) + (s >> 9 & 0x7c00)
		      + (s >> 3 & 0x1f);
		    d = (d | d << 16) & 0x03e07c1f;
		    d += (s - d) * alpha >> 5;
		    d &= 0x03e07c1f;
		    *dstp = d | d >> 16;
		}
		srcp++;
		dstp++;
	    }, width);
	    srcp += srcskip;
	    dstp += dstskip;
	}
}

/* General (slow) N->N blending with per-surface alpha */
static void BlitNtoNSurfaceAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint8 *src = info->s_pixels;
	int srcskip = info->s_skip;
	Uint8 *dst = info->d_pixels;
	int dstskip = info->d_skip;
	SDL_PixelFormat *srcfmt = info->src;
	SDL_PixelFormat *dstfmt = info->dst;
	int srcbpp = srcfmt->BytesPerPixel;
	int dstbpp = dstfmt->BytesPerPixel;
	unsigned sA = srcfmt->alpha;
	unsigned dA = dstfmt->Amask ? SDL_ALPHA_OPAQUE : 0;

	while ( height-- ) {
	    DUFFS_LOOP4(
	    {
		Uint32 pixel;
		unsigned sR;
		unsigned sG;
		unsigned sB;
		unsigned dR;
		unsigned dG;
		unsigned dB;
		DISEMBLE_RGB(src, srcbpp, srcfmt, pixel, sR, sG, sB);
		DISEMBLE_RGB(dst, dstbpp, dstfmt, pixel, dR, dG, dB);
		ALPHA_BLEND(sR, sG, sB, sA, dR, dG, dB);
		ASSEMBLE_RGBA(dst, dstbpp, dstfmt, dR, dG, dB, dA);
		src += srcbpp;
		dst += dstbpp;
	    },
	    width);
	    src += srcskip;
	    dst += dstskip;
	}
}

/* General (slow) colorkeyed N->N blending with per-surface alpha */
static void BlitNtoNSurfaceAlphaKey(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint8 *src = info->s_pixels;
	int srcskip = info->s_skip;
	Uint8 *dst = info->d_pixels;
	int dstskip = info->d_skip;
	SDL_PixelFormat *srcfmt = info->src;
	SDL_PixelFormat *dstfmt = info->dst;
	Uint32 ckey = srcfmt->colorkey;
	int srcbpp = srcfmt->BytesPerPixel;
	int dstbpp = dstfmt->BytesPerPixel;
	unsigned sA = srcfmt->alpha;
	unsigned dA = dstfmt->Amask ? SDL_ALPHA_OPAQUE : 0;

	while ( height-- ) {
	    DUFFS_LOOP4(
	    {
		Uint32 pixel;
		unsigned sR;
		unsigned sG;
		unsigned sB;
		unsigned dR;
		unsigned dG;
		unsigned dB;
		RETRIEVE_RGB_PIXEL(src, srcbpp, pixel);
		if(pixel != ckey) {
		    RGB_FROM_PIXEL(pixel, srcfmt, sR, sG, sB);
		    DISEMBLE_RGB(dst, dstbpp, dstfmt, pixel, dR, dG, dB);
		    ALPHA_BLEND(sR, sG, sB, sA, dR, dG, dB);
		    ASSEMBLE_RGBA(dst, dstbpp, dstfmt, dR, dG, dB, dA);
		}
		src += srcbpp;
		dst += dstbpp;
	    },
	    width);
	    src += srcskip;
	    dst += dstskip;
	}
}

/* General (slow) N->N blending with pixel alpha */
static void BlitNtoNPixelAlpha(SDL_BlitInfo *info)
{
	int width = info->d_width;
	int height = info->d_height;
	Uint8 *src = info->s_pixels;
	int srcskip = info->s_skip;
	Uint8 *dst = info->d_pixels;
	int dstskip = info->d_skip;
	SDL_PixelFormat *srcfmt = info->src;
	SDL_PixelFormat *dstfmt = info->dst;

	int  srcbpp;
	int  dstbpp;

	/* Set up some basic variables */
	srcbpp = srcfmt->BytesPerPixel;
	dstbpp = dstfmt->BytesPerPixel;

	/* FIXME: for 8bpp source alpha, this doesn't get opaque values
	   quite right. for <8bpp source alpha, it gets them very wrong
	   (check all macros!)
	   It is unclear whether there is a good general solution that doesn't
	   need a branch (or a divide). */
	while ( height-- ) {
	    DUFFS_LOOP4(
	    {
		Uint32 pixel;
		unsigned sR;
		unsigned sG;
		unsigned sB;
		unsigned dR;
		unsigned dG;
		unsigned dB;
		unsigned sA;
		unsigned dA;
		DISEMBLE_RGBA(src, srcbpp, srcfmt, pixel, sR, sG, sB, sA);
		DISEMBLE_RGBA(dst, dstbpp, dstfmt, pixel, dR, dG, dB, dA);
		ALPHA_BLEND(sR, sG, sB, sA, dR, dG, dB);
		ASSEMBLE_RGBA(dst, dstbpp, dstfmt, dR, dG, dB, dA);
		src += srcbpp;
		dst += dstbpp;
	    },
	    width);
	    src += srcskip;
	    dst += dstskip;
	}
}


SDL_loblit SDL_CalculateAlphaBlit(SDL_Surface *surface, int blit_index)
{
    SDL_PixelFormat *sf = surface->format;
    SDL_PixelFormat *df = surface->map->dst->format;

    if(sf->Amask == 0) {
	if((surface->flags & SDL_SRCCOLORKEY) == SDL_SRCCOLORKEY) {
	    if(df->BytesPerPixel == 1)
		return BlitNto1SurfaceAlphaKey;
	    else
		return BlitNtoNSurfaceAlphaKey;
	} else {
	    /* Per-surface alpha blits */
	    switch(df->BytesPerPixel) {
	    case 1:
		return BlitNto1SurfaceAlpha;

	    case 2:
		if(surface->map->identity) {
		    if(df->Gmask == 0x7e0)
			return Blit565to565SurfaceAlpha;
		    else if(df->Gmask == 0x3e0)
			return Blit555to555SurfaceAlpha;
		}
		return BlitNtoNSurfaceAlpha;

	    case 4:
		if(sf->Rmask == df->Rmask
		   && sf->Gmask == df->Gmask
		   && sf->Bmask == df->Bmask
		   && (sf->Rmask | sf->Gmask | sf->Bmask) == 0xffffff
		   && sf->BytesPerPixel == 4)
		    return BlitRGBtoRGBSurfaceAlpha;
		else
		    return BlitNtoNSurfaceAlpha;

	    case 3:
	    default:
		return BlitNtoNSurfaceAlpha;
	    }
	}
    } else {
	/* Per-pixel alpha blits */
	switch(df->BytesPerPixel) {
	case 1:
	    return BlitNto1PixelAlpha;

	case 2:
	    if(sf->BytesPerPixel == 4 && sf->Amask == 0xff000000
	       && sf->Gmask == 0xff00
	       && ((sf->Rmask == 0xff && df->Rmask == 0x1f)
		   || (sf->Bmask == 0xff && df->Bmask == 0x1f))) {
		if(df->Gmask == 0x7e0)
		    return BlitARGBto565PixelAlpha;
		else if(df->Gmask == 0x3e0)
		    return BlitARGBto555PixelAlpha;
	    }
	    return BlitNtoNPixelAlpha;

	case 4:
	    if(sf->Amask == 0xff000000
	       && sf->Rmask == df->Rmask
	       && sf->Gmask == df->Gmask
	       && sf->Bmask == df->Bmask
	       && sf->BytesPerPixel == 4)
		return BlitRGBtoRGBPixelAlpha;
	    return BlitNtoNPixelAlpha;

	case 3:
	default:
	    return BlitNtoNPixelAlpha;
	}
    }
}