1.1 --- a/configure.ac	Sun Mar 04 18:30:06 2012 -0600
     1.2 +++ b/configure.ac	Sun Mar 04 20:32:31 2012 -0600
     1.3 @@ -18,7 +18,7 @@
     1.4  AC_PROG_RANLIB
     1.5  AC_PROG_CPP
     1.6  AC_PROG_MKDIR_P
     1.7 -
     1.8 +AC_PATH_PROG(NASM, nasm)
     1.9  
    1.10  # Checks for libraries.
    1.11  AC_CHECK_LIB([SDL], [SDL_Init])
    1.12 @@ -61,7 +61,8 @@
    1.13  		 src/gb/Makefile
    1.14  		 src/gba/Makefile
    1.15  		 src/common/Makefile
    1.16 -		 src/SFMT/Makefile])
    1.17 +		 src/SFMT/Makefile
    1.18 +		 src/filters/Makefile])
    1.19  
    1.20  
    1.21  

     2.1 --- a/src/Makefile.am	Sun Mar 04 18:30:06 2012 -0600
     2.2 +++ b/src/Makefile.am	Sun Mar 04 20:32:31 2012 -0600
     2.3 @@ -1,3 +1,3 @@
     2.4 -SUBDIRS = SFMT lua gb gba common
     2.5 +SUBDIRS = SFMT lua gb gba common filters
     2.6  
     2.7  noinst_HEADERS = Port.h NLS.h

     3.1 --- /dev/null	Thu Jan 01 00:00:00 1970 +0000
     3.2 +++ b/src/filters/2xSaI.cpp	Sun Mar 04 20:32:31 2012 -0600
     3.3 @@ -0,0 +1,1434 @@
     3.4 +#include "../common/System.h"
     3.5 +
     3.6 +extern "C"
     3.7 +{
     3.8 +#ifdef MMX
     3.9 +	void _2xSaILine(u8 *srcPtr, u8 *deltaPtr, u32 srcPitch,
    3.10 +	                u32 width, u8 *dstPtr, u32 dstPitch);
    3.11 +	void _2xSaISuperEagleLine(u8 *srcPtr, u8 *deltaPtr,
    3.12 +	                          u32 srcPitch, u32 width,
    3.13 +	                          u8 *dstPtr, u32 dstPitch);
    3.14 +	void _2xSaISuper2xSaILine(u8 *srcPtr, u8 *deltaPtr,
    3.15 +	                          u32 srcPitch, u32 width,
    3.16 +	                          u8 *dstPtr, u32 dstPitch);
    3.17 +	void Init_2xSaIMMX(u32 BitFormat);
    3.18 +	void BilinearMMX(u16 *A, u16 *B, u16 *C, u16 *D,
    3.19 +	                 u16 *dx, u16 *dy, u8 *dP);
    3.20 +	void BilinearMMXGrid0(u16 *A, u16 *B, u16 *C, u16 *D,
    3.21 +	                      u16 *dx, u16 *dy, u8 *dP);
    3.22 +	void BilinearMMXGrid1(u16 *A, u16 *B, u16 *C, u16 *D,
    3.23 +	                      u16 *dx, u16 *dy, u8 *dP);
    3.24 +	void EndMMX();
    3.25 +
    3.26 +	bool cpu_mmx = 1;
    3.27 +#endif
    3.28 +}
    3.29 +static u32 colorMask	 = 0xF7DEF7DE;
    3.30 +static u32 lowPixelMask	 = 0x08210821;
    3.31 +static u32 qcolorMask	 = 0xE79CE79C;
    3.32 +static u32 qlowpixelMask = 0x18631863;
    3.33 +static u32 redblueMask	 = 0xF81F;
    3.34 +static u32 greenMask	 = 0x7E0;
    3.35 +
    3.36 +u32 qRGB_COLOR_MASK[2] = { 0xF7DEF7DE, 0xF7DEF7DE };
    3.37 +
    3.38 +extern void hq2x_init(unsigned);
    3.39 +
    3.40 +int Init_2xSaI(u32 BitFormat)
    3.41 +{
    3.42 +	if (systemColorDepth == 16)
    3.43 +	{
    3.44 +		if (BitFormat == 565)
    3.45 +		{
    3.46 +			colorMask		   = 0xF7DEF7DE;
    3.47 +			lowPixelMask	   = 0x08210821;
    3.48 +			qcolorMask		   = 0xE79CE79C;
    3.49 +			qlowpixelMask	   = 0x18631863;
    3.50 +			redblueMask		   = 0xF81F;
    3.51 +			greenMask		   = 0x7E0;
    3.52 +			qRGB_COLOR_MASK[0] = qRGB_COLOR_MASK[1] = 0xF7DEF7DE;
    3.53 +			hq2x_init(16);
    3.54 +		}
    3.55 +		else if (BitFormat == 555)
    3.56 +		{
    3.57 +			colorMask		   = 0x7BDE7BDE;
    3.58 +			lowPixelMask	   = 0x04210421;
    3.59 +			qcolorMask		   = 0x739C739C;
    3.60 +			qlowpixelMask	   = 0x0C630C63;
    3.61 +			redblueMask		   = 0x7C1F;
    3.62 +			greenMask		   = 0x3E0;
    3.63 +			qRGB_COLOR_MASK[0] = qRGB_COLOR_MASK[1] = 0x7BDE7BDE;
    3.64 +			hq2x_init(15);
    3.65 +		}
    3.66 +		else
    3.67 +		{
    3.68 +			return 0;
    3.69 +		}
    3.70 +	}
    3.71 +	else if (systemColorDepth == 32)
    3.72 +	{
    3.73 +		colorMask		   = 0xfefefe;
    3.74 +		lowPixelMask	   = 0x010101;
    3.75 +		qcolorMask		   = 0xfcfcfc;
    3.76 +		qlowpixelMask	   = 0x030303;
    3.77 +		qRGB_COLOR_MASK[0] = qRGB_COLOR_MASK[1] = 0xfefefe;
    3.78 +		hq2x_init(32);
    3.79 +	}
    3.80 +	else
    3.81 +		return 0;
    3.82 +
    3.83 +#ifdef MMX
    3.84 +	Init_2xSaIMMX(BitFormat);
    3.85 +#endif
    3.86 +
    3.87 +	return 1;
    3.88 +}
    3.89 +
    3.90 +static inline int GetResult1(u32 A, u32 B, u32 C, u32 D,
    3.91 +                             u32 /* E */)
    3.92 +{
    3.93 +	int x = 0;
    3.94 +	int y = 0;
    3.95 +	int r = 0;
    3.96 +
    3.97 +	if (A == C)
    3.98 +		x += 1;
    3.99 +	else if (B == C)
   3.100 +		y += 1;
   3.101 +	if (A == D)
   3.102 +		x += 1;
   3.103 +	else if (B == D)
   3.104 +		y += 1;
   3.105 +	if (x <= 1)
   3.106 +		r += 1;
   3.107 +	if (y <= 1)
   3.108 +		r -= 1;
   3.109 +	return r;
   3.110 +}
   3.111 +
   3.112 +static inline int GetResult2(u32 A, u32 B, u32 C, u32 D,
   3.113 +                             u32 /* E */)
   3.114 +{
   3.115 +	int x = 0;
   3.116 +	int y = 0;
   3.117 +	int r = 0;
   3.118 +
   3.119 +	if (A == C)
   3.120 +		x += 1;
   3.121 +	else if (B == C)
   3.122 +		y += 1;
   3.123 +	if (A == D)
   3.124 +		x += 1;
   3.125 +	else if (B == D)
   3.126 +		y += 1;
   3.127 +	if (x <= 1)
   3.128 +		r -= 1;
   3.129 +	if (y <= 1)
   3.130 +		r += 1;
   3.131 +	return r;
   3.132 +}
   3.133 +
   3.134 +static inline int GetResult(u32 A, u32 B, u32 C, u32 D)
   3.135 +{
   3.136 +	int x = 0;
   3.137 +	int y = 0;
   3.138 +	int r = 0;
   3.139 +
   3.140 +	if (A == C)
   3.141 +		x += 1;
   3.142 +	else if (B == C)
   3.143 +		y += 1;
   3.144 +	if (A == D)
   3.145 +		x += 1;
   3.146 +	else if (B == D)
   3.147 +		y += 1;
   3.148 +	if (x <= 1)
   3.149 +		r += 1;
   3.150 +	if (y <= 1)
   3.151 +		r -= 1;
   3.152 +	return r;
   3.153 +}
   3.154 +
   3.155 +static inline u32 INTERPOLATE(u32 A, u32 B)
   3.156 +{
   3.157 +	if (A != B)
   3.158 +	{
   3.159 +		return (((A & colorMask) >> 1) + ((B & colorMask) >> 1) +
   3.160 +		        (A & B & lowPixelMask));
   3.161 +	}
   3.162 +	else
   3.163 +		return A;
   3.164 +}
   3.165 +
   3.166 +static inline u32 Q_INTERPOLATE(u32 A, u32 B, u32 C, u32 D)
   3.167 +{
   3.168 +	register u32 x = ((A & qcolorMask) >> 2) +
   3.169 +	                 ((B & qcolorMask) >> 2) +
   3.170 +	                 ((C & qcolorMask) >> 2) + ((D & qcolorMask) >> 2);
   3.171 +	register u32 y = (A & qlowpixelMask) +
   3.172 +	                 (B & qlowpixelMask) + (C & qlowpixelMask) + (D & qlowpixelMask);
   3.173 +
   3.174 +	y = (y >> 2) & qlowpixelMask;
   3.175 +	return x + y;
   3.176 +}
   3.177 +
   3.178 +static inline int GetResult1_32(u32 A, u32 B, u32 C, u32 D,
   3.179 +                                u32 /* E */)
   3.180 +{
   3.181 +	int x = 0;
   3.182 +	int y = 0;
   3.183 +	int r = 0;
   3.184 +
   3.185 +	if (A == C)
   3.186 +		x += 1;
   3.187 +	else if (B == C)
   3.188 +		y += 1;
   3.189 +	if (A == D)
   3.190 +		x += 1;
   3.191 +	else if (B == D)
   3.192 +		y += 1;
   3.193 +	if (x <= 1)
   3.194 +		r += 1;
   3.195 +	if (y <= 1)
   3.196 +		r -= 1;
   3.197 +	return r;
   3.198 +}
   3.199 +
   3.200 +static inline int GetResult2_32(u32 A, u32 B, u32 C, u32 D,
   3.201 +                                u32 /* E */)
   3.202 +{
   3.203 +	int x = 0;
   3.204 +	int y = 0;
   3.205 +	int r = 0;
   3.206 +
   3.207 +	if (A == C)
   3.208 +		x += 1;
   3.209 +	else if (B == C)
   3.210 +		y += 1;
   3.211 +	if (A == D)
   3.212 +		x += 1;
   3.213 +	else if (B == D)
   3.214 +		y += 1;
   3.215 +	if (x <= 1)
   3.216 +		r -= 1;
   3.217 +	if (y <= 1)
   3.218 +		r += 1;
   3.219 +	return r;
   3.220 +}
   3.221 +
   3.222 +#define BLUE_MASK565 0x001F001F
   3.223 +#define RED_MASK565 0xF800F800
   3.224 +#define GREEN_MASK565 0x07E007E0
   3.225 +
   3.226 +#define BLUE_MASK555 0x001F001F
   3.227 +#define RED_MASK555 0x7C007C00
   3.228 +#define GREEN_MASK555 0x03E003E0
   3.229 +
   3.230 +void Super2xSaI(u8 *srcPtr, u32 srcPitch,
   3.231 +                u8 *deltaPtr, u8 *dstPtr, u32 dstPitch,
   3.232 +                int width, int height)
   3.233 +{
   3.234 +	u16 *bP;
   3.235 +	u8 * dP;
   3.236 +	u32	 inc_bP;
   3.237 +	u32	 Nextline = srcPitch >> 1;
   3.238 +#ifdef MMX
   3.239 +	if (cpu_mmx)
   3.240 +	{
   3.241 +		for (; height; height--)
   3.242 +		{
   3.243 +			_2xSaISuper2xSaILine(srcPtr, deltaPtr, srcPitch, width,
   3.244 +			                     dstPtr, dstPitch);
   3.245 +			srcPtr	 += srcPitch;
   3.246 +			dstPtr	 += dstPitch * 2;
   3.247 +			deltaPtr += srcPitch;
   3.248 +		}
   3.249 +	}
   3.250 +	else
   3.251 +#endif
   3.252 +	{
   3.253 +		inc_bP = 1;
   3.254 +
   3.255 +		for (; height; height--)
   3.256 +		{
   3.257 +			bP = (u16 *) srcPtr;
   3.258 +			dP = (u8 *) dstPtr;
   3.259 +
   3.260 +			for (u32 finish = width; finish; finish -= inc_bP)
   3.261 +			{
   3.262 +				u32 color4, color5, color6;
   3.263 +				u32 color1, color2, color3;
   3.264 +				u32 colorA0, colorA1, colorA2, colorA3,
   3.265 +				    colorB0, colorB1, colorB2, colorB3, colorS1, colorS2;
   3.266 +				u32 product1a, product1b, product2a, product2b;
   3.267 +
   3.268 +				//---------------------------------------    B1 B2
   3.269 +				//                                         4  5  6 S2
   3.270 +				//                                         1  2  3 S1
   3.271 +				//                                           A1 A2
   3.272 +
   3.273 +				colorB0 = *(bP - Nextline - 1);
   3.274 +				colorB1 = *(bP - Nextline);
   3.275 +				colorB2 = *(bP - Nextline + 1);
   3.276 +				colorB3 = *(bP - Nextline + 2);
   3.277 +
   3.278 +				color4	= *(bP - 1);
   3.279 +				color5	= *(bP);
   3.280 +				color6	= *(bP + 1);
   3.281 +				colorS2 = *(bP + 2);
   3.282 +
   3.283 +				color1	= *(bP + Nextline - 1);
   3.284 +				color2	= *(bP + Nextline);
   3.285 +				color3	= *(bP + Nextline + 1);
   3.286 +				colorS1 = *(bP + Nextline + 2);
   3.287 +
   3.288 +				colorA0 = *(bP + Nextline + Nextline - 1);
   3.289 +				colorA1 = *(bP + Nextline + Nextline);
   3.290 +				colorA2 = *(bP + Nextline + Nextline + 1);
   3.291 +				colorA3 = *(bP + Nextline + Nextline + 2);
   3.292 +
   3.293 +				//--------------------------------------
   3.294 +				if (color2 == color6 && color5 != color3)
   3.295 +				{
   3.296 +					product2b = product1b = color2;
   3.297 +				}
   3.298 +				else if (color5 == color3 && color2 != color6)
   3.299 +				{
   3.300 +					product2b = product1b = color5;
   3.301 +				}
   3.302 +				else if (color5 == color3 && color2 == color6)
   3.303 +				{
   3.304 +					register int r = 0;
   3.305 +
   3.306 +					r += GetResult(color6, color5, color1, colorA1);
   3.307 +					r += GetResult(color6, color5, color4, colorB1);
   3.308 +					r += GetResult(color6, color5, colorA2, colorS1);
   3.309 +					r += GetResult(color6, color5, colorB2, colorS2);
   3.310 +
   3.311 +					if (r > 0)
   3.312 +						product2b = product1b = color6;
   3.313 +					else if (r < 0)
   3.314 +						product2b = product1b = color5;
   3.315 +					else
   3.316 +					{
   3.317 +						product2b = product1b = INTERPOLATE(color5, color6);
   3.318 +					}
   3.319 +				}
   3.320 +				else
   3.321 +				{
   3.322 +					if (color6 == color3 && color3 == colorA1
   3.323 +					    && color2 != colorA2 && color3 != colorA0)
   3.324 +						product2b =
   3.325 +						    Q_INTERPOLATE(color3, color3, color3, color2);
   3.326 +					else if (color5 == color2 && color2 == colorA2
   3.327 +					         && colorA1 != color3 && color2 != colorA3)
   3.328 +						product2b =
   3.329 +						    Q_INTERPOLATE(color2, color2, color2, color3);
   3.330 +					else
   3.331 +						product2b = INTERPOLATE(color2, color3);
   3.332 +
   3.333 +					if (color6 == color3 && color6 == colorB1
   3.334 +					    && color5 != colorB2 && color6 != colorB0)
   3.335 +						product1b =
   3.336 +						    Q_INTERPOLATE(color6, color6, color6, color5);
   3.337 +					else if (color5 == color2 && color5 == colorB2
   3.338 +					         && colorB1 != color6 && color5 != colorB3)
   3.339 +						product1b =
   3.340 +						    Q_INTERPOLATE(color6, color5, color5, color5);
   3.341 +					else
   3.342 +						product1b = INTERPOLATE(color5, color6);
   3.343 +				}
   3.344 +
   3.345 +				if (color5 == color3 && color2 != color6 && color4 == color5
   3.346 +				    && color5 != colorA2)
   3.347 +					product2a = INTERPOLATE(color2, color5);
   3.348 +				else
   3.349 +					if (color5 == color1 && color6 == color5
   3.350 +					    && color4 != color2 && color5 != colorA0)
   3.351 +						product2a = INTERPOLATE(color2, color5);
   3.352 +					else
   3.353 +						product2a = color2;
   3.354 +
   3.355 +				if (color2 == color6 && color5 != color3 && color1 == color2
   3.356 +				    && color2 != colorB2)
   3.357 +					product1a = INTERPOLATE(color2, color5);
   3.358 +				else
   3.359 +					if (color4 == color2 && color3 == color2
   3.360 +					    && color1 != color5 && color2 != colorB0)
   3.361 +						product1a = INTERPOLATE(color2, color5);
   3.362 +					else
   3.363 +						product1a = color5;
   3.364 +
   3.365 +#ifdef WORDS_BIGENDIAN
   3.366 +				product1a = (product1a << 16) | product1b;
   3.367 +				product2a = (product2a << 16) | product2b;
   3.368 +#else
   3.369 +				product1a = product1a | (product1b << 16);
   3.370 +				product2a = product2a | (product2b << 16);
   3.371 +#endif
   3.372 +
   3.373 +				*((u32 *) dP) = product1a;
   3.374 +				*((u32 *) (dP + dstPitch)) = product2a;
   3.375 +
   3.376 +				bP += inc_bP;
   3.377 +				dP += sizeof(u32);
   3.378 +			}                   // end of for ( finish= width etc..)
   3.379 +
   3.380 +			srcPtr	 += srcPitch;
   3.381 +			dstPtr	 += dstPitch << 1;
   3.382 +			deltaPtr += srcPitch;
   3.383 +		}               // endof: for (; height; height--)
   3.384 +	}
   3.385 +}
   3.386 +
   3.387 +void Super2xSaI32(u8 *srcPtr, u32 srcPitch,
   3.388 +                  u8 * /* deltaPtr */, u8 *dstPtr, u32 dstPitch,
   3.389 +                  int width, int height)
   3.390 +{
   3.391 +	u32 *bP;
   3.392 +	u32 *dP;
   3.393 +	u32	 inc_bP;
   3.394 +	u32	 Nextline = srcPitch >> 2;
   3.395 +	inc_bP = 1;
   3.396 +
   3.397 +	for (; height; height--)
   3.398 +	{
   3.399 +		bP = (u32 *) srcPtr;
   3.400 +		dP = (u32 *) dstPtr;
   3.401 +
   3.402 +		for (u32 finish = width; finish; finish -= inc_bP)
   3.403 +		{
   3.404 +			u32 color4, color5, color6;
   3.405 +			u32 color1, color2, color3;
   3.406 +			u32 colorA0, colorA1, colorA2, colorA3,
   3.407 +			    colorB0, colorB1, colorB2, colorB3, colorS1, colorS2;
   3.408 +			u32 product1a, product1b, product2a, product2b;
   3.409 +
   3.410 +			//---------------------------------------    B1 B2
   3.411 +			//                                         4  5  6 S2
   3.412 +			//                                         1  2  3 S1
   3.413 +			//                                           A1 A2
   3.414 +
   3.415 +			colorB0 = *(bP - Nextline - 1);
   3.416 +			colorB1 = *(bP - Nextline);
   3.417 +			colorB2 = *(bP - Nextline + 1);
   3.418 +			colorB3 = *(bP - Nextline + 2);
   3.419 +
   3.420 +			color4	= *(bP - 1);
   3.421 +			color5	= *(bP);
   3.422 +			color6	= *(bP + 1);
   3.423 +			colorS2 = *(bP + 2);
   3.424 +
   3.425 +			color1	= *(bP + Nextline - 1);
   3.426 +			color2	= *(bP + Nextline);
   3.427 +			color3	= *(bP + Nextline + 1);
   3.428 +			colorS1 = *(bP + Nextline + 2);
   3.429 +
   3.430 +			colorA0 = *(bP + Nextline + Nextline - 1);
   3.431 +			colorA1 = *(bP + Nextline + Nextline);
   3.432 +			colorA2 = *(bP + Nextline + Nextline + 1);
   3.433 +			colorA3 = *(bP + Nextline + Nextline + 2);
   3.434 +
   3.435 +			//--------------------------------------
   3.436 +			if (color2 == color6 && color5 != color3)
   3.437 +			{
   3.438 +				product2b = product1b = color2;
   3.439 +			}
   3.440 +			else if (color5 == color3 && color2 != color6)
   3.441 +			{
   3.442 +				product2b = product1b = color5;
   3.443 +			}
   3.444 +			else if (color5 == color3 && color2 == color6)
   3.445 +			{
   3.446 +				register int r = 0;
   3.447 +
   3.448 +				r += GetResult(color6, color5, color1, colorA1);
   3.449 +				r += GetResult(color6, color5, color4, colorB1);
   3.450 +				r += GetResult(color6, color5, colorA2, colorS1);
   3.451 +				r += GetResult(color6, color5, colorB2, colorS2);
   3.452 +
   3.453 +				if (r > 0)
   3.454 +					product2b = product1b = color6;
   3.455 +				else if (r < 0)
   3.456 +					product2b = product1b = color5;
   3.457 +				else
   3.458 +				{
   3.459 +					product2b = product1b = INTERPOLATE(color5, color6);
   3.460 +				}
   3.461 +			}
   3.462 +			else
   3.463 +			{
   3.464 +				if (color6 == color3 && color3 == colorA1
   3.465 +				    && color2 != colorA2 && color3 != colorA0)
   3.466 +					product2b =
   3.467 +					    Q_INTERPOLATE(color3, color3, color3, color2);
   3.468 +				else if (color5 == color2 && color2 == colorA2
   3.469 +				         && colorA1 != color3 && color2 != colorA3)
   3.470 +					product2b =
   3.471 +					    Q_INTERPOLATE(color2, color2, color2, color3);
   3.472 +				else
   3.473 +					product2b = INTERPOLATE(color2, color3);
   3.474 +
   3.475 +				if (color6 == color3 && color6 == colorB1
   3.476 +				    && color5 != colorB2 && color6 != colorB0)
   3.477 +					product1b =
   3.478 +					    Q_INTERPOLATE(color6, color6, color6, color5);
   3.479 +				else if (color5 == color2 && color5 == colorB2
   3.480 +				         && colorB1 != color6 && color5 != colorB3)
   3.481 +					product1b =
   3.482 +					    Q_INTERPOLATE(color6, color5, color5, color5);
   3.483 +				else
   3.484 +					product1b = INTERPOLATE(color5, color6);
   3.485 +			}
   3.486 +
   3.487 +			if (color5 == color3 && color2 != color6 && color4 == color5
   3.488 +			    && color5 != colorA2)
   3.489 +				product2a = INTERPOLATE(color2, color5);
   3.490 +			else
   3.491 +				if (color5 == color1 && color6 == color5
   3.492 +				    && color4 != color2 && color5 != colorA0)
   3.493 +					product2a = INTERPOLATE(color2, color5);
   3.494 +				else
   3.495 +					product2a = color2;
   3.496 +
   3.497 +			if (color2 == color6 && color5 != color3 && color1 == color2
   3.498 +			    && color2 != colorB2)
   3.499 +				product1a = INTERPOLATE(color2, color5);
   3.500 +			else
   3.501 +				if (color4 == color2 && color3 == color2
   3.502 +				    && color1 != color5 && color2 != colorB0)
   3.503 +					product1a = INTERPOLATE(color2, color5);
   3.504 +				else
   3.505 +					product1a = color5;
   3.506 +			*(dP)	  = product1a;
   3.507 +			*(dP + 1) = product1b;
   3.508 +			*(dP + (dstPitch >> 2))		= product2a;
   3.509 +			*(dP + (dstPitch >> 2) + 1) = product2b;
   3.510 +
   3.511 +			bP += inc_bP;
   3.512 +			dP += 2;
   3.513 +		}                   // end of for ( finish= width etc..)
   3.514 +
   3.515 +		srcPtr += srcPitch;
   3.516 +		dstPtr += dstPitch << 1;
   3.517 +		//        deltaPtr += srcPitch;
   3.518 +	}               // endof: for (; height; height--)
   3.519 +}
   3.520 +
   3.521 +void SuperEagle(u8 *srcPtr, u32 srcPitch, u8 *deltaPtr,
   3.522 +                u8 *dstPtr, u32 dstPitch, int width, int height)
   3.523 +{
   3.524 +	u8 * dP;
   3.525 +	u16 *bP;
   3.526 +	u16 *xP;
   3.527 +	u32	 inc_bP;
   3.528 +
   3.529 +#ifdef MMX
   3.530 +	if (cpu_mmx)
   3.531 +	{
   3.532 +		for (; height; height--)
   3.533 +		{
   3.534 +			_2xSaISuperEagleLine(srcPtr, deltaPtr, srcPitch, width,
   3.535 +			                     dstPtr, dstPitch);
   3.536 +			srcPtr	 += srcPitch;
   3.537 +			dstPtr	 += dstPitch * 2;
   3.538 +			deltaPtr += srcPitch;
   3.539 +		}
   3.540 +	}
   3.541 +	else
   3.542 +#endif
   3.543 +	{
   3.544 +		inc_bP = 1;
   3.545 +
   3.546 +		u32 Nextline = srcPitch >> 1;
   3.547 +
   3.548 +		for (; height; height--)
   3.549 +		{
   3.550 +			bP = (u16 *) srcPtr;
   3.551 +			xP = (u16 *) deltaPtr;
   3.552 +			dP = dstPtr;
   3.553 +			for (u32 finish = width; finish; finish -= inc_bP)
   3.554 +			{
   3.555 +				u32 color4, color5, color6;
   3.556 +				u32 color1, color2, color3;
   3.557 +				u32 colorA1, colorA2, colorB1, colorB2, colorS1, colorS2;
   3.558 +				u32 product1a, product1b, product2a, product2b;
   3.559 +
   3.560 +				colorB1 = *(bP - Nextline);
   3.561 +				colorB2 = *(bP - Nextline + 1);
   3.562 +
   3.563 +				color4	= *(bP - 1);
   3.564 +				color5	= *(bP);
   3.565 +				color6	= *(bP + 1);
   3.566 +				colorS2 = *(bP + 2);
   3.567 +
   3.568 +				color1	= *(bP + Nextline - 1);
   3.569 +				color2	= *(bP + Nextline);
   3.570 +				color3	= *(bP + Nextline + 1);
   3.571 +				colorS1 = *(bP + Nextline + 2);
   3.572 +
   3.573 +				colorA1 = *(bP + Nextline + Nextline);
   3.574 +				colorA2 = *(bP + Nextline + Nextline + 1);
   3.575 +
   3.576 +				// --------------------------------------
   3.577 +				if (color2 == color6 && color5 != color3)
   3.578 +				{
   3.579 +					product1b = product2a = color2;
   3.580 +					if ((color1 == color2) || (color6 == colorB2))
   3.581 +					{
   3.582 +						product1a = INTERPOLATE(color2, color5);
   3.583 +						product1a = INTERPOLATE(color2, product1a);
   3.584 +						//                       product1a = color2;
   3.585 +					}
   3.586 +					else
   3.587 +					{
   3.588 +						product1a = INTERPOLATE(color5, color6);
   3.589 +					}
   3.590 +
   3.591 +					if ((color6 == colorS2) || (color2 == colorA1))
   3.592 +					{
   3.593 +						product2b = INTERPOLATE(color2, color3);
   3.594 +						product2b = INTERPOLATE(color2, product2b);
   3.595 +						//                       product2b = color2;
   3.596 +					}
   3.597 +					else
   3.598 +					{
   3.599 +						product2b = INTERPOLATE(color2, color3);
   3.600 +					}
   3.601 +				}
   3.602 +				else if (color5 == color3 && color2 != color6)
   3.603 +				{
   3.604 +					product2b = product1a = color5;
   3.605 +
   3.606 +					if ((colorB1 == color5) || (color3 == colorS1))
   3.607 +					{
   3.608 +						product1b = INTERPOLATE(color5, color6);
   3.609 +						product1b = INTERPOLATE(color5, product1b);
   3.610 +						//                       product1b = color5;
   3.611 +					}
   3.612 +					else
   3.613 +					{
   3.614 +						product1b = INTERPOLATE(color5, color6);
   3.615 +					}
   3.616 +
   3.617 +					if ((color3 == colorA2) || (color4 == color5))
   3.618 +					{
   3.619 +						product2a = INTERPOLATE(color5, color2);
   3.620 +						product2a = INTERPOLATE(color5, product2a);
   3.621 +						//                       product2a = color5;
   3.622 +					}
   3.623 +					else
   3.624 +					{
   3.625 +						product2a = INTERPOLATE(color2, color3);
   3.626 +					}
   3.627 +				}
   3.628 +				else if (color5 == color3 && color2 == color6)
   3.629 +				{
   3.630 +					register int r = 0;
   3.631 +
   3.632 +					r += GetResult(color6, color5, color1, colorA1);
   3.633 +					r += GetResult(color6, color5, color4, colorB1);
   3.634 +					r += GetResult(color6, color5, colorA2, colorS1);
   3.635 +					r += GetResult(color6, color5, colorB2, colorS2);
   3.636 +
   3.637 +					if (r > 0)
   3.638 +					{
   3.639 +						product1b = product2a = color2;
   3.640 +						product1a = product2b = INTERPOLATE(color5, color6);
   3.641 +					}
   3.642 +					else if (r < 0)
   3.643 +					{
   3.644 +						product2b = product1a = color5;
   3.645 +						product1b = product2a = INTERPOLATE(color5, color6);
   3.646 +					}
   3.647 +					else
   3.648 +					{
   3.649 +						product2b = product1a = color5;
   3.650 +						product1b = product2a = color2;
   3.651 +					}
   3.652 +				}
   3.653 +				else
   3.654 +				{
   3.655 +					product2b = product1a = INTERPOLATE(color2, color6);
   3.656 +					product2b =
   3.657 +					    Q_INTERPOLATE(color3, color3, color3, product2b);
   3.658 +					product1a =
   3.659 +					    Q_INTERPOLATE(color5, color5, color5, product1a);
   3.660 +
   3.661 +					product2a = product1b = INTERPOLATE(color5, color3);
   3.662 +					product2a =
   3.663 +					    Q_INTERPOLATE(color2, color2, color2, product2a);
   3.664 +					product1b =
   3.665 +					    Q_INTERPOLATE(color6, color6, color6, product1b);
   3.666 +
   3.667 +					//                    product1a = color5;
   3.668 +					//                    product1b = color6;
   3.669 +					//                    product2a = color2;
   3.670 +					//                    product2b = color3;
   3.671 +				}
   3.672 +#ifdef WORDS_BIGENDIAN
   3.673 +				product1a = (product1a << 16) | product1b;
   3.674 +				product2a = (product2a << 16) | product2b;
   3.675 +#else
   3.676 +				product1a = product1a | (product1b << 16);
   3.677 +				product2a = product2a | (product2b << 16);
   3.678 +#endif
   3.679 +
   3.680 +				*((u32 *) dP) = product1a;
   3.681 +				*((u32 *) (dP + dstPitch)) = product2a;
   3.682 +				*xP = color5;
   3.683 +
   3.684 +				bP += inc_bP;
   3.685 +				xP += inc_bP;
   3.686 +				dP += sizeof(u32);
   3.687 +			}           // end of for ( finish= width etc..)
   3.688 +
   3.689 +			srcPtr	 += srcPitch;
   3.690 +			dstPtr	 += dstPitch << 1;
   3.691 +			deltaPtr += srcPitch;
   3.692 +		}               // endof: for (height; height; height--)
   3.693 +	}
   3.694 +}
   3.695 +
   3.696 +void SuperEagle32(u8 *srcPtr, u32 srcPitch, u8 *deltaPtr,
   3.697 +                  u8 *dstPtr, u32 dstPitch, int width, int height)
   3.698 +{
   3.699 +	u32 *dP;
   3.700 +	u32 *bP;
   3.701 +	u32 *xP;
   3.702 +	u32	 inc_bP;
   3.703 +
   3.704 +	inc_bP = 1;
   3.705 +
   3.706 +	u32 Nextline = srcPitch >> 2;
   3.707 +
   3.708 +	for (; height; height--)
   3.709 +	{
   3.710 +		bP = (u32 *) srcPtr;
   3.711 +		xP = (u32 *) deltaPtr;
   3.712 +		dP = (u32 *)dstPtr;
   3.713 +		for (u32 finish = width; finish; finish -= inc_bP)
   3.714 +		{
   3.715 +			u32 color4, color5, color6;
   3.716 +			u32 color1, color2, color3;
   3.717 +			u32 colorA1, colorA2, colorB1, colorB2, colorS1, colorS2;
   3.718 +			u32 product1a, product1b, product2a, product2b;
   3.719 +
   3.720 +			colorB1 = *(bP - Nextline);
   3.721 +			colorB2 = *(bP - Nextline + 1);
   3.722 +
   3.723 +			color4	= *(bP - 1);
   3.724 +			color5	= *(bP);
   3.725 +			color6	= *(bP + 1);
   3.726 +			colorS2 = *(bP + 2);
   3.727 +
   3.728 +			color1	= *(bP + Nextline - 1);
   3.729 +			color2	= *(bP + Nextline);
   3.730 +			color3	= *(bP + Nextline + 1);
   3.731 +			colorS1 = *(bP + Nextline + 2);
   3.732 +
   3.733 +			colorA1 = *(bP + Nextline + Nextline);
   3.734 +			colorA2 = *(bP + Nextline + Nextline + 1);
   3.735 +
   3.736 +			// --------------------------------------
   3.737 +			if (color2 == color6 && color5 != color3)
   3.738 +			{
   3.739 +				product1b = product2a = color2;
   3.740 +				if ((color1 == color2) || (color6 == colorB2))
   3.741 +				{
   3.742 +					product1a = INTERPOLATE(color2, color5);
   3.743 +					product1a = INTERPOLATE(color2, product1a);
   3.744 +					//                       product1a = color2;
   3.745 +				}
   3.746 +				else
   3.747 +				{
   3.748 +					product1a = INTERPOLATE(color5, color6);
   3.749 +				}
   3.750 +
   3.751 +				if ((color6 == colorS2) || (color2 == colorA1))
   3.752 +				{
   3.753 +					product2b = INTERPOLATE(color2, color3);
   3.754 +					product2b = INTERPOLATE(color2, product2b);
   3.755 +					//                       product2b = color2;
   3.756 +				}
   3.757 +				else
   3.758 +				{
   3.759 +					product2b = INTERPOLATE(color2, color3);
   3.760 +				}
   3.761 +			}
   3.762 +			else if (color5 == color3 && color2 != color6)
   3.763 +			{
   3.764 +				product2b = product1a = color5;
   3.765 +
   3.766 +				if ((colorB1 == color5) || (color3 == colorS1))
   3.767 +				{
   3.768 +					product1b = INTERPOLATE(color5, color6);
   3.769 +					product1b = INTERPOLATE(color5, product1b);
   3.770 +					//                       product1b = color5;
   3.771 +				}
   3.772 +				else
   3.773 +				{
   3.774 +					product1b = INTERPOLATE(color5, color6);
   3.775 +				}
   3.776 +
   3.777 +				if ((color3 == colorA2) || (color4 == color5))
   3.778 +				{
   3.779 +					product2a = INTERPOLATE(color5, color2);
   3.780 +					product2a = INTERPOLATE(color5, product2a);
   3.781 +					//                       product2a = color5;
   3.782 +				}
   3.783 +				else
   3.784 +				{
   3.785 +					product2a = INTERPOLATE(color2, color3);
   3.786 +				}
   3.787 +			}
   3.788 +			else if (color5 == color3 && color2 == color6)
   3.789 +			{
   3.790 +				register int r = 0;
   3.791 +
   3.792 +				r += GetResult(color6, color5, color1, colorA1);
   3.793 +				r += GetResult(color6, color5, color4, colorB1);
   3.794 +				r += GetResult(color6, color5, colorA2, colorS1);
   3.795 +				r += GetResult(color6, color5, colorB2, colorS2);
   3.796 +
   3.797 +				if (r > 0)
   3.798 +				{
   3.799 +					product1b = product2a = color2;
   3.800 +					product1a = product2b = INTERPOLATE(color5, color6);
   3.801 +				}
   3.802 +				else if (r < 0)
   3.803 +				{
   3.804 +					product2b = product1a = color5;
   3.805 +					product1b = product2a = INTERPOLATE(color5, color6);
   3.806 +				}
   3.807 +				else
   3.808 +				{
   3.809 +					product2b = product1a = color5;
   3.810 +					product1b = product2a = color2;
   3.811 +				}
   3.812 +			}
   3.813 +			else
   3.814 +			{
   3.815 +				product2b = product1a = INTERPOLATE(color2, color6);
   3.816 +				product2b =
   3.817 +				    Q_INTERPOLATE(color3, color3, color3, product2b);
   3.818 +				product1a =
   3.819 +				    Q_INTERPOLATE(color5, color5, color5, product1a);
   3.820 +
   3.821 +				product2a = product1b = INTERPOLATE(color5, color3);
   3.822 +				product2a =
   3.823 +				    Q_INTERPOLATE(color2, color2, color2, product2a);
   3.824 +				product1b =
   3.825 +				    Q_INTERPOLATE(color6, color6, color6, product1b);
   3.826 +
   3.827 +				//                    product1a = color5;
   3.828 +				//                    product1b = color6;
   3.829 +				//                    product2a = color2;
   3.830 +				//                    product2b = color3;
   3.831 +			}
   3.832 +			*(dP)	  = product1a;
   3.833 +			*(dP + 1) = product1b;
   3.834 +			*(dP + (dstPitch >> 2))		= product2a;
   3.835 +			*(dP + (dstPitch >> 2) + 1) = product2b;
   3.836 +			*xP = color5;
   3.837 +
   3.838 +			bP += inc_bP;
   3.839 +			xP += inc_bP;
   3.840 +			dP += 2;
   3.841 +		}             // end of for ( finish= width etc..)
   3.842 +
   3.843 +		srcPtr	 += srcPitch;
   3.844 +		dstPtr	 += dstPitch << 1;
   3.845 +		deltaPtr += srcPitch;
   3.846 +	}                 // endof: for (height; height; height--)
   3.847 +}
   3.848 +
   3.849 +void _2xSaI(u8 *srcPtr, u32 srcPitch, u8 *deltaPtr,
   3.850 +            u8 *dstPtr, u32 dstPitch, int width, int height)
   3.851 +{
   3.852 +	u8 * dP;
   3.853 +	u16 *bP;
   3.854 +	u32	 inc_bP;
   3.855 +
   3.856 +#ifdef MMX
   3.857 +	if (cpu_mmx)
   3.858 +	{
   3.859 +		for (; height; height -= 1)
   3.860 +		{
   3.861 +			_2xSaILine(srcPtr, deltaPtr, srcPitch, width, dstPtr, dstPitch);
   3.862 +			srcPtr	 += srcPitch;
   3.863 +			dstPtr	 += dstPitch * 2;
   3.864 +			deltaPtr += srcPitch;
   3.865 +		}
   3.866 +	}
   3.867 +	else
   3.868 +#endif
   3.869 +	{
   3.870 +		inc_bP = 1;
   3.871 +
   3.872 +		u32 Nextline = srcPitch >> 1;
   3.873 +
   3.874 +		for (; height; height--)
   3.875 +		{
   3.876 +			bP = (u16 *) srcPtr;
   3.877 +			dP = dstPtr;
   3.878 +
   3.879 +			for (u32 finish = width; finish; finish -= inc_bP)
   3.880 +			{
   3.881 +				register u32 colorA, colorB;
   3.882 +				u32 colorC, colorD,
   3.883 +				    colorE, colorF, colorG, colorH,
   3.884 +				    colorI, colorJ, colorK, colorL,
   3.885 +
   3.886 +				    colorM, colorN, colorO, colorP;
   3.887 +				u32 product, product1, product2;
   3.888 +
   3.889 +				//---------------------------------------
   3.890 +				// Map of the pixels:                    I|E F|J
   3.891 +				//                                       G|A B|K
   3.892 +				//                                       H|C D|L
   3.893 +				//                                       M|N O|P
   3.894 +				colorI = *(bP - Nextline - 1);
   3.895 +				colorE = *(bP - Nextline);
   3.896 +				colorF = *(bP - Nextline + 1);
   3.897 +				colorJ = *(bP - Nextline + 2);
   3.898 +
   3.899 +				colorG = *(bP - 1);
   3.900 +				colorA = *(bP);
   3.901 +				colorB = *(bP + 1);
   3.902 +				colorK = *(bP + 2);
   3.903 +
   3.904 +				colorH = *(bP + Nextline - 1);
   3.905 +				colorC = *(bP + Nextline);
   3.906 +				colorD = *(bP + Nextline + 1);
   3.907 +				colorL = *(bP + Nextline + 2);
   3.908 +
   3.909 +				colorM = *(bP + Nextline + Nextline - 1);
   3.910 +				colorN = *(bP + Nextline + Nextline);
   3.911 +				colorO = *(bP + Nextline + Nextline + 1);
   3.912 +				colorP = *(bP + Nextline + Nextline + 2);
   3.913 +
   3.914 +				if ((colorA == colorD) && (colorB != colorC))
   3.915 +				{
   3.916 +					if (((colorA == colorE) && (colorB == colorL)) ||
   3.917 +					    ((colorA == colorC) && (colorA == colorF)
   3.918 +					     && (colorB != colorE) && (colorB == colorJ)))
   3.919 +					{
   3.920 +						product = colorA;
   3.921 +					}
   3.922 +					else
   3.923 +					{
   3.924 +						product = INTERPOLATE(colorA, colorB);
   3.925 +					}
   3.926 +
   3.927 +					if (((colorA == colorG) && (colorC == colorO)) ||
   3.928 +					    ((colorA == colorB) && (colorA == colorH)
   3.929 +					     && (colorG != colorC) && (colorC == colorM)))
   3.930 +					{
   3.931 +						product1 = colorA;
   3.932 +					}
   3.933 +					else
   3.934 +					{
   3.935 +						product1 = INTERPOLATE(colorA, colorC);
   3.936 +					}
   3.937 +					product2 = colorA;
   3.938 +				}
   3.939 +				else if ((colorB == colorC) && (colorA != colorD))
   3.940 +				{
   3.941 +					if (((colorB == colorF) && (colorA == colorH)) ||
   3.942 +					    ((colorB == colorE) && (colorB == colorD)
   3.943 +					     && (colorA != colorF) && (colorA == colorI)))
   3.944 +					{
   3.945 +						product = colorB;
   3.946 +					}
   3.947 +					else
   3.948 +					{
   3.949 +						product = INTERPOLATE(colorA, colorB);
   3.950 +					}
   3.951 +
   3.952 +					if (((colorC == colorH) && (colorA == colorF)) ||
   3.953 +					    ((colorC == colorG) && (colorC == colorD)
   3.954 +					     && (colorA != colorH) && (colorA == colorI)))
   3.955 +					{
   3.956 +						product1 = colorC;
   3.957 +					}
   3.958 +					else
   3.959 +					{
   3.960 +						product1 = INTERPOLATE(colorA, colorC);
   3.961 +					}
   3.962 +					product2 = colorB;
   3.963 +				}
   3.964 +				else if ((colorA == colorD) && (colorB == colorC))
   3.965 +				{
   3.966 +					if (colorA == colorB)
   3.967 +					{
   3.968 +						product	 = colorA;
   3.969 +						product1 = colorA;
   3.970 +						product2 = colorA;
   3.971 +					}
   3.972 +					else
   3.973 +					{
   3.974 +						register int r = 0;
   3.975 +
   3.976 +						product1 = INTERPOLATE(colorA, colorC);
   3.977 +						product	 = INTERPOLATE(colorA, colorB);
   3.978 +
   3.979 +						r +=
   3.980 +						    GetResult1(colorA, colorB, colorG, colorE,
   3.981 +						               colorI);
   3.982 +						r +=
   3.983 +						    GetResult2(colorB, colorA, colorK, colorF,
   3.984 +						               colorJ);
   3.985 +						r +=
   3.986 +						    GetResult2(colorB, colorA, colorH, colorN,
   3.987 +						               colorM);
   3.988 +						r +=
   3.989 +						    GetResult1(colorA, colorB, colorL, colorO,
   3.990 +						               colorP);
   3.991 +
   3.992 +						if (r > 0)
   3.993 +							product2 = colorA;
   3.994 +						else if (r < 0)
   3.995 +							product2 = colorB;
   3.996 +						else
   3.997 +						{
   3.998 +							product2 =
   3.999 +							    Q_INTERPOLATE(colorA, colorB, colorC,
  3.1000 +							                  colorD);
  3.1001 +						}
  3.1002 +					}
  3.1003 +				}
  3.1004 +				else
  3.1005 +				{
  3.1006 +					product2 = Q_INTERPOLATE(colorA, colorB, colorC, colorD);
  3.1007 +
  3.1008 +					if ((colorA == colorC) && (colorA == colorF)
  3.1009 +					    && (colorB != colorE) && (colorB == colorJ))
  3.1010 +					{
  3.1011 +						product = colorA;
  3.1012 +					}
  3.1013 +					else if ((colorB == colorE) && (colorB == colorD)
  3.1014 +					         && (colorA != colorF) && (colorA == colorI))
  3.1015 +					{
  3.1016 +						product = colorB;
  3.1017 +					}
  3.1018 +					else
  3.1019 +					{
  3.1020 +						product = INTERPOLATE(colorA, colorB);
  3.1021 +					}
  3.1022 +
  3.1023 +					if ((colorA == colorB) && (colorA == colorH)
  3.1024 +					    && (colorG != colorC) && (colorC == colorM))
  3.1025 +					{
  3.1026 +						product1 = colorA;
  3.1027 +					}
  3.1028 +					else if ((colorC == colorG) && (colorC == colorD)
  3.1029 +					         && (colorA != colorH) && (colorA == colorI))
  3.1030 +					{
  3.1031 +						product1 = colorC;
  3.1032 +					}
  3.1033 +					else
  3.1034 +					{
  3.1035 +						product1 = INTERPOLATE(colorA, colorC);
  3.1036 +					}
  3.1037 +				}
  3.1038 +
  3.1039 +#ifdef WORDS_BIGENDIAN
  3.1040 +				product	 = (colorA << 16) | product;
  3.1041 +				product1 = (product1 << 16) | product2;
  3.1042 +#else
  3.1043 +				product	 = colorA | (product << 16);
  3.1044 +				product1 = product1 | (product2 << 16);
  3.1045 +#endif
  3.1046 +				*((s32 *) dP) = product;
  3.1047 +				*((u32 *) (dP + dstPitch)) = product1;
  3.1048 +
  3.1049 +				bP += inc_bP;
  3.1050 +				dP += sizeof(u32);
  3.1051 +			}           // end of for ( finish= width etc..)
  3.1052 +
  3.1053 +			srcPtr	 += srcPitch;
  3.1054 +			dstPtr	 += dstPitch << 1;
  3.1055 +			deltaPtr += srcPitch;
  3.1056 +		}               // endof: for (height; height; height--)
  3.1057 +	}
  3.1058 +}
  3.1059 +
  3.1060 +void _2xSaI32(u8 *srcPtr, u32 srcPitch, u8 * /* deltaPtr */,
  3.1061 +              u8 *dstPtr, u32 dstPitch, int width, int height)
  3.1062 +{
  3.1063 +	u32 *dP;
  3.1064 +	u32 *bP;
  3.1065 +	u32	 inc_bP = 1;
  3.1066 +
  3.1067 +	u32 Nextline = srcPitch >> 2;
  3.1068 +
  3.1069 +	for (; height; height--)
  3.1070 +	{
  3.1071 +		bP = (u32 *) srcPtr;
  3.1072 +		dP = (u32 *) dstPtr;
  3.1073 +
  3.1074 +		for (u32 finish = width; finish; finish -= inc_bP)
  3.1075 +		{
  3.1076 +			register u32 colorA, colorB;
  3.1077 +			u32 colorC, colorD,
  3.1078 +			    colorE, colorF, colorG, colorH,
  3.1079 +			    colorI, colorJ, colorK, colorL,
  3.1080 +
  3.1081 +			    colorM, colorN, colorO, colorP;
  3.1082 +			u32 product, product1, product2;
  3.1083 +
  3.1084 +			//---------------------------------------
  3.1085 +			// Map of the pixels:                    I|E F|J
  3.1086 +			//                                       G|A B|K
  3.1087 +			//                                       H|C D|L
  3.1088 +			//                                       M|N O|P
  3.1089 +			colorI = *(bP - Nextline - 1);
  3.1090 +			colorE = *(bP - Nextline);
  3.1091 +			colorF = *(bP - Nextline + 1);
  3.1092 +			colorJ = *(bP - Nextline + 2);
  3.1093 +
  3.1094 +			colorG = *(bP - 1);
  3.1095 +			colorA = *(bP);
  3.1096 +			colorB = *(bP + 1);
  3.1097 +			colorK = *(bP + 2);
  3.1098 +
  3.1099 +			colorH = *(bP + Nextline - 1);
  3.1100 +			colorC = *(bP + Nextline);
  3.1101 +			colorD = *(bP + Nextline + 1);
  3.1102 +			colorL = *(bP + Nextline + 2);
  3.1103 +
  3.1104 +			colorM = *(bP + Nextline + Nextline - 1);
  3.1105 +			colorN = *(bP + Nextline + Nextline);
  3.1106 +			colorO = *(bP + Nextline + Nextline + 1);
  3.1107 +			colorP = *(bP + Nextline + Nextline + 2);
  3.1108 +
  3.1109 +			if ((colorA == colorD) && (colorB != colorC))
  3.1110 +			{
  3.1111 +				if (((colorA == colorE) && (colorB == colorL)) ||
  3.1112 +				    ((colorA == colorC) && (colorA == colorF)
  3.1113 +				     && (colorB != colorE) && (colorB == colorJ)))
  3.1114 +				{
  3.1115 +					product = colorA;
  3.1116 +				}
  3.1117 +				else
  3.1118 +				{
  3.1119 +					product = INTERPOLATE(colorA, colorB);
  3.1120 +				}
  3.1121 +
  3.1122 +				if (((colorA == colorG) && (colorC == colorO)) ||
  3.1123 +				    ((colorA == colorB) && (colorA == colorH)
  3.1124 +				     && (colorG != colorC) && (colorC == colorM)))
  3.1125 +				{
  3.1126 +					product1 = colorA;
  3.1127 +				}
  3.1128 +				else
  3.1129 +				{
  3.1130 +					product1 = INTERPOLATE(colorA, colorC);
  3.1131 +				}
  3.1132 +				product2 = colorA;
  3.1133 +			}
  3.1134 +			else if ((colorB == colorC) && (colorA != colorD))
  3.1135 +			{
  3.1136 +				if (((colorB == colorF) && (colorA == colorH)) ||
  3.1137 +				    ((colorB == colorE) && (colorB == colorD)
  3.1138 +				     && (colorA != colorF) && (colorA == colorI)))
  3.1139 +				{
  3.1140 +					product = colorB;
  3.1141 +				}
  3.1142 +				else
  3.1143 +				{
  3.1144 +					product = INTERPOLATE(colorA, colorB);
  3.1145 +				}
  3.1146 +
  3.1147 +				if (((colorC == colorH) && (colorA == colorF)) ||
  3.1148 +				    ((colorC == colorG) && (colorC == colorD)
  3.1149 +				     && (colorA != colorH) && (colorA == colorI)))
  3.1150 +				{
  3.1151 +					product1 = colorC;
  3.1152 +				}
  3.1153 +				else
  3.1154 +				{
  3.1155 +					product1 = INTERPOLATE(colorA, colorC);
  3.1156 +				}
  3.1157 +				product2 = colorB;
  3.1158 +			}
  3.1159 +			else if ((colorA == colorD) && (colorB == colorC))
  3.1160 +			{
  3.1161 +				if (colorA == colorB)
  3.1162 +				{
  3.1163 +					product	 = colorA;
  3.1164 +					product1 = colorA;
  3.1165 +					product2 = colorA;
  3.1166 +				}
  3.1167 +				else
  3.1168 +				{
  3.1169 +					register int r = 0;
  3.1170 +
  3.1171 +					product1 = INTERPOLATE(colorA, colorC);
  3.1172 +					product	 = INTERPOLATE(colorA, colorB);
  3.1173 +
  3.1174 +					r +=
  3.1175 +					    GetResult1(colorA, colorB, colorG, colorE,
  3.1176 +					               colorI);
  3.1177 +					r +=
  3.1178 +					    GetResult2(colorB, colorA, colorK, colorF,
  3.1179 +					               colorJ);
  3.1180 +					r +=
  3.1181 +					    GetResult2(colorB, colorA, colorH, colorN,
  3.1182 +					               colorM);
  3.1183 +					r +=
  3.1184 +					    GetResult1(colorA, colorB, colorL, colorO,
  3.1185 +					               colorP);
  3.1186 +
  3.1187 +					if (r > 0)
  3.1188 +						product2 = colorA;
  3.1189 +					else if (r < 0)
  3.1190 +						product2 = colorB;
  3.1191 +					else
  3.1192 +					{
  3.1193 +						product2 =
  3.1194 +						    Q_INTERPOLATE(colorA, colorB, colorC,
  3.1195 +						                  colorD);
  3.1196 +					}
  3.1197 +				}
  3.1198 +			}
  3.1199 +			else
  3.1200 +			{
  3.1201 +				product2 = Q_INTERPOLATE(colorA, colorB, colorC, colorD);
  3.1202 +
  3.1203 +				if ((colorA == colorC) && (colorA == colorF)
  3.1204 +				    && (colorB != colorE) && (colorB == colorJ))
  3.1205 +				{
  3.1206 +					product = colorA;
  3.1207 +				}
  3.1208 +				else if ((colorB == colorE) && (colorB == colorD)
  3.1209 +				         && (colorA != colorF) && (colorA == colorI))
  3.1210 +				{
  3.1211 +					product = colorB;
  3.1212 +				}
  3.1213 +				else
  3.1214 +				{
  3.1215 +					product = INTERPOLATE(colorA, colorB);
  3.1216 +				}
  3.1217 +
  3.1218 +				if ((colorA == colorB) && (colorA == colorH)
  3.1219 +				    && (colorG != colorC) && (colorC == colorM))
  3.1220 +				{
  3.1221 +					product1 = colorA;
  3.1222 +				}
  3.1223 +				else if ((colorC == colorG) && (colorC == colorD)
  3.1224 +				         && (colorA != colorH) && (colorA == colorI))
  3.1225 +				{
  3.1226 +					product1 = colorC;
  3.1227 +				}
  3.1228 +				else
  3.1229 +				{
  3.1230 +					product1 = INTERPOLATE(colorA, colorC);
  3.1231 +				}
  3.1232 +			}
  3.1233 +			*(dP)	  = colorA;
  3.1234 +			*(dP + 1) = product;
  3.1235 +			*(dP + (dstPitch >> 2))		= product1;
  3.1236 +			*(dP + (dstPitch >> 2) + 1) = product2;
  3.1237 +
  3.1238 +			bP += inc_bP;
  3.1239 +			dP += 2;
  3.1240 +		}             // end of for ( finish= width etc..)
  3.1241 +
  3.1242 +		srcPtr += srcPitch;
  3.1243 +		dstPtr += dstPitch << 1;
  3.1244 +		//    deltaPtr += srcPitch;
  3.1245 +	}                 // endof: for (height; height; height--)
  3.1246 +}
  3.1247 +
  3.1248 +static u32 Bilinear(u32 A, u32 B, u32 x)
  3.1249 +{
  3.1250 +	unsigned long areaA, areaB;
  3.1251 +	unsigned long result;
  3.1252 +
  3.1253 +	if (A == B)
  3.1254 +		return A;
  3.1255 +
  3.1256 +	areaB = (x >> 11) & 0x1f;   // reduce 16 bit fraction to 5 bits
  3.1257 +	areaA = 0x20 - areaB;
  3.1258 +
  3.1259 +	A = (A & redblueMask) | ((A & greenMask) << 16);
  3.1260 +	B = (B & redblueMask) | ((B & greenMask) << 16);
  3.1261 +
  3.1262 +	result = ((areaA * A) + (areaB * B)) >> 5;
  3.1263 +
  3.1264 +	return (result & redblueMask) | ((result >> 16) & greenMask);
  3.1265 +}
  3.1266 +
  3.1267 +static u32 Bilinear4(u32 A, u32 B, u32 C, u32 D, u32 x,
  3.1268 +                     u32 y)
  3.1269 +{
  3.1270 +	unsigned long areaA, areaB, areaC, areaD;
  3.1271 +	unsigned long result, xy;
  3.1272 +
  3.1273 +	x  = (x >> 11) & 0x1f;
  3.1274 +	y  = (y >> 11) & 0x1f;
  3.1275 +	xy = (x * y) >> 5;
  3.1276 +
  3.1277 +	A = (A & redblueMask) | ((A & greenMask) << 16);
  3.1278 +	B = (B & redblueMask) | ((B & greenMask) << 16);
  3.1279 +	C = (C & redblueMask) | ((C & greenMask) << 16);
  3.1280 +	D = (D & redblueMask) | ((D & greenMask) << 16);
  3.1281 +
  3.1282 +	areaA = 0x20 + xy - x - y;
  3.1283 +	areaB = x - xy;
  3.1284 +	areaC = y - xy;
  3.1285 +	areaD = xy;
  3.1286 +
  3.1287 +	result = ((areaA * A) + (areaB * B) + (areaC * C) + (areaD * D)) >> 5;
  3.1288 +
  3.1289 +	return (result & redblueMask) | ((result >> 16) & greenMask);
  3.1290 +}
  3.1291 +
  3.1292 +void Scale_2xSaI(u8 *srcPtr, u32 srcPitch, u8 * /* deltaPtr */,
  3.1293 +                 u8 *dstPtr, u32 dstPitch,
  3.1294 +                 u32 dstWidth, u32 dstHeight, int width, int height)
  3.1295 +{
  3.1296 +	u8 * dP;
  3.1297 +	u16 *bP;
  3.1298 +
  3.1299 +	u32 w;
  3.1300 +	u32 h;
  3.1301 +	u32 dw;
  3.1302 +	u32 dh;
  3.1303 +	u32 hfinish;
  3.1304 +	u32 wfinish;
  3.1305 +
  3.1306 +	u32 Nextline = srcPitch >> 1;
  3.1307 +
  3.1308 +	wfinish = (width - 1) << 16; // convert to fixed point
  3.1309 +	dw		= wfinish / (dstWidth - 1);
  3.1310 +	hfinish = (height - 1) << 16; // convert to fixed point
  3.1311 +	dh		= hfinish / (dstHeight - 1);
  3.1312 +
  3.1313 +	for (h = 0; h < hfinish; h += dh)
  3.1314 +	{
  3.1315 +		u32 y1, y2;
  3.1316 +
  3.1317 +		y1 = h & 0xffff; // fraction part of fixed point
  3.1318 +		bP = (u16 *) (srcPtr + ((h >> 16) * srcPitch));
  3.1319 +		dP = dstPtr;
  3.1320 +		y2 = 0x10000 - y1;
  3.1321 +
  3.1322 +		w = 0;
  3.1323 +
  3.1324 +		for (; w < wfinish; )
  3.1325 +		{
  3.1326 +			u32 A, B, C, D;
  3.1327 +			u32 E, F, G, H;
  3.1328 +			u32 I, J, K, L;
  3.1329 +			u32 x1, x2, a1, f1, f2;
  3.1330 +			u32 position, product1;
  3.1331 +
  3.1332 +			position = w >> 16;
  3.1333 +			A		 = bP[position]; // current pixel
  3.1334 +			B		 = bP[position + 1]; // next pixel
  3.1335 +			C		 = bP[position + Nextline];
  3.1336 +			D		 = bP[position + Nextline + 1];
  3.1337 +			E		 = bP[position - Nextline];
  3.1338 +			F		 = bP[position - Nextline + 1];
  3.1339 +			G		 = bP[position - 1];
  3.1340 +			H		 = bP[position + Nextline - 1];
  3.1341 +			I		 = bP[position + 2];
  3.1342 +			J		 = bP[position + Nextline + 2];
  3.1343 +			K		 = bP[position + Nextline + Nextline];
  3.1344 +			L		 = bP[position + Nextline + Nextline + 1];
  3.1345 +
  3.1346 +			x1 = w & 0xffff; // fraction part of fixed point
  3.1347 +			x2 = 0x10000 - x1;
  3.1348 +
  3.1349 +			/*0*/
  3.1350 +			if (A == B && C == D && A == C)
  3.1351 +				product1 = A;
  3.1352 +			else /*1*/ if (A == D && B != C)
  3.1353 +			{
  3.1354 +				f1 = (x1 >> 1) + (0x10000 >> 2);
  3.1355 +				f2 = (y1 >> 1) + (0x10000 >> 2);
  3.1356 +				if (y1 <= f1 && A == J && A != E) // close to B
  3.1357 +				{
  3.1358 +					a1		 = f1 - y1;
  3.1359 +					product1 = Bilinear(A, B, a1);
  3.1360 +				}
  3.1361 +				else if (y1 >= f1 && A == G && A != L)  // close to C
  3.1362 +				{
  3.1363 +					a1		 = y1 - f1;
  3.1364 +					product1 = Bilinear(A, C, a1);
  3.1365 +				}
  3.1366 +				else if (x1 >= f2 && A == E && A != J) // close to B
  3.1367 +				{
  3.1368 +					a1		 = x1 - f2;
  3.1369 +					product1 = Bilinear(A, B, a1);
  3.1370 +				}
  3.1371 +				else if (x1 <= f2 && A == L && A != G) // close to C
  3.1372 +				{
  3.1373 +					a1		 = f2 - x1;
  3.1374 +					product1 = Bilinear(A, C, a1);
  3.1375 +				}
  3.1376 +				else if (y1 >= x1) // close to C
  3.1377 +				{
  3.1378 +					a1		 = y1 - x1;
  3.1379 +					product1 = Bilinear(A, C, a1);
  3.1380 +				}
  3.1381 +				else if (y1 <= x1) // close to B
  3.1382 +				{
  3.1383 +					a1		 = x1 - y1;
  3.1384 +					product1 = Bilinear(A, B, a1);
  3.1385 +				}
  3.1386 +			}
  3.1387 +			else
  3.1388 +				/*2*/
  3.1389 +				if (B == C && A != D)
  3.1390 +				{
  3.1391 +					f1 = (x1 >> 1) + (0x10000 >> 2);
  3.1392 +					f2 = (y1 >> 1) + (0x10000 >> 2);
  3.1393 +					if (y2 >= f1 && B == H && B != F) // close to A
  3.1394 +					{
  3.1395 +						a1		 = y2 - f1;
  3.1396 +						product1 = Bilinear(B, A, a1);
  3.1397 +					}
  3.1398 +					else if (y2 <= f1 && B == I && B != K) // close to D
  3.1399 +					{
  3.1400 +						a1		 = f1 - y2;
  3.1401 +						product1 = Bilinear(B, D, a1);
  3.1402 +					}
  3.1403 +					else if (x2 >= f2 && B == F && B != H) // close to A
  3.1404 +					{
  3.1405 +						a1		 = x2 - f2;
  3.1406 +						product1 = Bilinear(B, A, a1);
  3.1407 +					}
  3.1408 +					else if (x2 <= f2 && B == K && B != I) // close to D
  3.1409 +					{
  3.1410 +						a1		 = f2 - x2;
  3.1411 +						product1 = Bilinear(B, D, a1);
  3.1412 +					}
  3.1413 +					else if (y2 >= x1) // close to A
  3.1414 +					{
  3.1415 +						a1		 = y2 - x1;
  3.1416 +						product1 = Bilinear(B, A, a1);
  3.1417 +					}
  3.1418 +					else if (y2 <= x1) // close to D
  3.1419 +					{
  3.1420 +						a1		 = x1 - y2;
  3.1421 +						product1 = Bilinear(B, D, a1);
  3.1422 +					}
  3.1423 +				}
  3.1424 +				/*3*/
  3.1425 +				else
  3.1426 +				{
  3.1427 +					product1 = Bilinear4(A, B, C, D, x1, y1);
  3.1428 +				}
  3.1429 +
  3.1430 +			//end First Pixel
  3.1431 +			*(u32 *) dP = product1;
  3.1432 +			dP += 2;
  3.1433 +			w  += dw;
  3.1434 +		}
  3.1435 +		dstPtr += dstPitch;
  3.1436 +	}
  3.1437 +}

     4.1 --- /dev/null	Thu Jan 01 00:00:00 1970 +0000
     4.2 +++ b/src/filters/2xSaImmx.asm	Sun Mar 04 20:32:31 2012 -0600
     4.3 @@ -0,0 +1,2109 @@
     4.4 +;/*---------------------------------------------------------------------*
     4.5 +; * The following (piece of) code, (part of) the 2xSaI engine,          *
     4.6 +; * copyright (c) 1999 - 2001 by Derek Liauw Kie Fa.                    *
     4.7 +; * Non-Commercial use of this software is allowed and is encouraged,   *
     4.8 +; * provided that appropriate credit be given.                          *
     4.9 +; * You may freely modify this code, but I request                      *
    4.10 +; * that any improvements to the engine be submitted to me, so          *
    4.11 +; * that I can implement these improvements in newer versions of        *
    4.12 +; * the software.                                                       *
    4.13 +; * If you need more information, have any comments or suggestions,     *
    4.14 +; * you can e-mail me. My e-mail: derek-liauw@usa.net.                  *
    4.15 +; *---------------------------------------------------------------------*/
    4.16 +
    4.17 +;----------------------
    4.18 +; 2xSaI version 0.59 WIP, soon to become version 0.60
    4.19 +;----------------------
    4.20 +
    4.21 +;%define FAR_POINTER
    4.22 +
    4.23 +
    4.24 +
    4.25 +          BITS 32
    4.26 +%ifdef __DJGPP__
    4.27 +          GLOBAL __2xSaILine
    4.28 +          GLOBAL __2xSaISuperEagleLine
    4.29 +                  GLOBAL __2xSaISuper2xSaILine
    4.30 +          GLOBAL _Init_2xSaIMMX
    4.31 +%else
    4.32 +          GLOBAL _2xSaILine
    4.33 +          GLOBAL _2xSaISuperEagleLine
    4.34 +                  GLOBAL _2xSaISuper2xSaILine
    4.35 +          GLOBAL Init_2xSaIMMX
    4.36 +%endif
    4.37 +          SECTION .text ALIGN = 32
    4.38 +
    4.39 +%ifdef FAR_POINTER
    4.40 +;EXTERN_C void _2xSaILine (uint8 *srcPtr, uint32 srcPitch, uint32 width,
    4.41 +;                        uint8 *dstPtr, uint32 dstPitch, uint16 dstSegment);
    4.42 +%else
    4.43 +;EXTERN_C void _2xSaILine (uint8 *srcPtr, uint32 srcPitch, uint32 width,
    4.44 +;                        uint8 *dstPtr, uint32 dstPitch);
    4.45 +%endif
    4.46 +
    4.47 +srcPtr        equ 8
    4.48 +deltaPtr      equ 12
    4.49 +srcPitch      equ 16
    4.50 +width         equ 20
    4.51 +dstOffset     equ 24
    4.52 +dstPitch      equ 28
    4.53 +dstSegment    equ 32
    4.54 +
    4.55 +
    4.56 +
    4.57 +
    4.58 +colorB0   equ -2
    4.59 +colorB1   equ 0
    4.60 +colorB2   equ 2
    4.61 +colorB3   equ 4
    4.62 +
    4.63 +color7   equ -2
    4.64 +color8   equ 0
    4.65 +color9   equ 2
    4.66 +
    4.67 +color4   equ -2
    4.68 +color5   equ 0
    4.69 +color6   equ 2
    4.70 +colorS2   equ 4
    4.71 +
    4.72 +color1   equ -2
    4.73 +color2   equ 0
    4.74 +color3   equ 2
    4.75 +colorS1   equ 4
    4.76 +
    4.77 +colorA0   equ -2
    4.78 +colorA1   equ 0
    4.79 +colorA2   equ 2
    4.80 +colorA3   equ 4
    4.81 +
    4.82 +
    4.83 +
    4.84 +
    4.85 +%ifdef __DJGPP__
    4.86 +__2xSaISuper2xSaILine:
    4.87 +%else
    4.88 +_2xSaISuper2xSaILine:
    4.89 +%endif
    4.90 +; Store some stuff
    4.91 +         push ebp
    4.92 +         mov ebp, esp
    4.93 +         pushad
    4.94 +
    4.95 +; Prepare the destination
    4.96 +%ifdef FAR_POINTER
    4.97 +         ; Set the selector
    4.98 +         mov eax, [ebp+dstSegment]
    4.99 +         mov fs, ax
   4.100 +%endif
   4.101 +         mov edx, [ebp+dstOffset]         ; edx points to the screen
   4.102 +; Prepare the source
   4.103 +         ; eax points to colorA
   4.104 +         mov eax, [ebp+srcPtr]                          ;eax points to colorA
   4.105 +         mov ebx, [ebp+srcPitch]                        ;ebx contains the source pitch
   4.106 +         mov ecx, [ebp+width]                           ;ecx contains the number of pixels to process
   4.107 +         ; eax now points to colorB1
   4.108 +         sub eax, ebx                                           ;eax points to B1 which is the base 
   4.109 +
   4.110 +; Main Loop
   4.111 +.Loop:   push ecx
   4.112 +
   4.113 +         ;-----Check Delta------------------
   4.114 +         mov ecx, [ebp+deltaPtr]
   4.115 +
   4.116 +
   4.117 +                ;load source img
   4.118 +         movq mm0, [eax+colorB0]
   4.119 +         movq mm1, [eax+colorB3]
   4.120 +         movq mm2, [eax+ebx+color4]
   4.121 +         movq mm3, [eax+ebx+colorS2]
   4.122 +         movq mm4, [eax+ebx+ebx+color1]
   4.123 +         movq mm5, [eax+ebx+ebx+colorS1]
   4.124 +         push eax
   4.125 +         add eax, ebx
   4.126 +         movq mm6, [eax+ebx+ebx+colorA0]
   4.127 +         movq mm7, [eax+ebx+ebx+colorA3]
   4.128 +         pop eax
   4.129 +
   4.130 +                ;compare to delta
   4.131 +         pcmpeqw mm0, [ecx+2+colorB0]
   4.132 +         pcmpeqw mm1, [ecx+2+colorB3]
   4.133 +         pcmpeqw mm2, [ecx+ebx+2+color4]
   4.134 +         pcmpeqw mm3, [ecx+ebx+2+colorS2]
   4.135 +         pcmpeqw mm4, [ecx+ebx+ebx+2+color1]
   4.136 +         pcmpeqw mm5, [ecx+ebx+ebx+2+colorS1]
   4.137 +         add ecx, ebx
   4.138 +         pcmpeqw mm6, [ecx+ebx+ebx+2+colorA0]
   4.139 +         pcmpeqw mm7, [ecx+ebx+ebx+2+colorA3]
   4.140 +         sub ecx, ebx
   4.141 +
   4.142 +
   4.143 +                ;compose results
   4.144 +         pand mm0, mm1
   4.145 +         pand mm2, mm3
   4.146 +         pand mm4, mm5
   4.147 +         pand mm6, mm7
   4.148 +         pand mm0, mm2
   4.149 +         pand mm4, mm6
   4.150 +         pxor mm7, mm7
   4.151 +         pand mm0, mm4
   4.152 +         movq mm6, [eax+colorB0]
   4.153 +         pcmpeqw mm7, mm0                       ;did any compare give us a zero ?
   4.154 +
   4.155 +         movq [ecx+2+colorB0], mm6
   4.156 +
   4.157 +         packsswb mm7, mm7
   4.158 +         movd ecx, mm7
   4.159 +         test ecx, ecx                          
   4.160 +         jz near .SKIP_PROCESS          ;no, so we can skip
   4.161 +
   4.162 +         ;End Delta
   4.163 +
   4.164 +         ;---------------------------------
   4.165 +         movq mm0, [eax+ebx+color5]
   4.166 +         movq mm1, [eax+ebx+color6]
   4.167 +         movq mm2, mm0
   4.168 +         movq mm3, mm1
   4.169 +         movq mm4, mm0
   4.170 +         movq mm5, mm1
   4.171 +
   4.172 +         pand mm0, [colorMask]
   4.173 +         pand mm1, [colorMask]
   4.174 +
   4.175 +         psrlw mm0, 1
   4.176 +         psrlw mm1, 1
   4.177 +
   4.178 +         pand mm3, [lowPixelMask]
   4.179 +         paddw mm0, mm1
   4.180 +
   4.181 +         pand mm3, mm2
   4.182 +         paddw mm0, mm3                ;mm0 contains the interpolated values
   4.183 +         movq [I56Pixel], mm0
   4.184 +         movq mm7, mm0
   4.185 +
   4.186 +         ;-------------------
   4.187 +         movq mm0, mm7
   4.188 +         movq mm1, mm4  ;5,5,5,6
   4.189 +         movq mm2, mm0
   4.190 +         movq mm3, mm1
   4.191 +
   4.192 +         pand mm0, [colorMask]
   4.193 +         pand mm1, [colorMask]
   4.194 +
   4.195 +         psrlw mm0, 1
   4.196 +         psrlw mm1, 1
   4.197 +
   4.198 +         pand mm3, [lowPixelMask]
   4.199 +         paddw mm0, mm1
   4.200 +
   4.201 +         pand mm3, mm2
   4.202 +         paddw mm0, mm3                ;mm0 contains the interpolated values
   4.203 +         movq [I5556Pixel], mm0
   4.204 +         ;--------------------
   4.205 +
   4.206 +         movq mm0, mm7
   4.207 +         movq mm1, mm5  ;6,6,6,5
   4.208 +         movq mm2, mm0
   4.209 +         movq mm3, mm1
   4.210 +
   4.211 +         pand mm0, [colorMask]
   4.212 +         pand mm1, [colorMask]
   4.213 +
   4.214 +         psrlw mm0, 1
   4.215 +         psrlw mm1, 1
   4.216 +
   4.217 +         pand mm3, [lowPixelMask]
   4.218 +         paddw mm0, mm1
   4.219 +
   4.220 +         pand mm3, mm2
   4.221 +         paddw mm0, mm3
   4.222 +         movq [I5666Pixel], mm0
   4.223 +
   4.224 +         ;-------------------------
   4.225 +         ;-------------------------
   4.226 +         movq mm0, [eax+ebx+ebx+color2]
   4.227 +         movq mm1, [eax+ebx+ebx+color3]
   4.228 +         movq mm2, mm0
   4.229 +         movq mm3, mm1
   4.230 +         movq mm4, mm0
   4.231 +         movq mm5, mm1
   4.232 +
   4.233 +         pand mm0, [colorMask]
   4.234 +         pand mm1, [colorMask]
   4.235 +
   4.236 +         psrlw mm0, 1
   4.237 +         psrlw mm1, 1
   4.238 +
   4.239 +         pand mm3, [lowPixelMask]
   4.240 +         paddw mm0, mm1
   4.241 +
   4.242 +         pand mm3, mm2
   4.243 +         paddw mm0, mm3
   4.244 +         movq [I23Pixel], mm0
   4.245 +         movq mm7, mm0
   4.246 +
   4.247 +         ;---------------------
   4.248 +         movq mm0, mm7
   4.249 +         movq mm1, mm4  ;2,2,2,3
   4.250 +         movq mm2, mm0
   4.251 +         movq mm3, mm1
   4.252 +
   4.253 +         pand mm0, [colorMask]
   4.254 +         pand mm1, [colorMask]
   4.255 +
   4.256 +         psrlw mm0, 1
   4.257 +         psrlw mm1, 1
   4.258 +
   4.259 +         pand mm3, [lowPixelMask]
   4.260 +         paddw mm0, mm1
   4.261 +
   4.262 +         pand mm3, mm2
   4.263 +         paddw mm0, mm3
   4.264 +         movq [I2223Pixel], mm0
   4.265 +
   4.266 +         ;----------------------
   4.267 +         movq mm0, mm7
   4.268 +         movq mm1, mm5  ;3,3,3,2
   4.269 +         movq mm2, mm0
   4.270 +         movq mm3, mm1
   4.271 +
   4.272 +         pand mm0, [colorMask]
   4.273 +         pand mm1, [colorMask]
   4.274 +
   4.275 +         psrlw mm0, 1
   4.276 +         psrlw mm1, 1
   4.277 +
   4.278 +         pand mm3, [lowPixelMask]
   4.279 +         paddw mm0, mm1
   4.280 +
   4.281 +         pand mm3, mm2
   4.282 +         paddw mm0, mm3
   4.283 +         movq [I2333Pixel], mm0
   4.284 +
   4.285 +
   4.286 +         ;--------------------
   4.287 +;////////////////////////////////
   4.288 +; Decide which "branch" to take
   4.289 +;--------------------------------
   4.290 +         movq mm0, [eax+ebx+color5]
   4.291 +         movq mm1, [eax+ebx+color6]
   4.292 +         movq mm6, mm0
   4.293 +         movq mm7, mm1
   4.294 +         pcmpeqw mm0, [eax+ebx+ebx+color3]
   4.295 +         pcmpeqw mm1, [eax+ebx+ebx+color2]
   4.296 +         pcmpeqw mm6, mm7
   4.297 +
   4.298 +         movq mm2, mm0
   4.299 +         movq mm3, mm0
   4.300 +
   4.301 +         pand mm0, mm1       ;colorA == colorD && colorB == colorC
   4.302 +         pxor mm7, mm7
   4.303 +
   4.304 +         pcmpeqw mm2, mm7
   4.305 +         pand mm6, mm0
   4.306 +         pand mm2, mm1       ;colorA != colorD && colorB == colorC
   4.307 +
   4.308 +         pcmpeqw mm1, mm7
   4.309 +
   4.310 +         pand mm1, mm3       ;colorA == colorD && colorB != colorC
   4.311 +         pxor mm0, mm6
   4.312 +         por mm1, mm6
   4.313 +         movq mm7, mm0
   4.314 +         movq [Mask26], mm2
   4.315 +         packsswb mm7, mm7
   4.316 +         movq [Mask35], mm1
   4.317 +
   4.318 +         movd ecx, mm7
   4.319 +         test ecx, ecx
   4.320 +         jz near .SKIP_GUESS
   4.321 +
   4.322 +;---------------------------------------------
   4.323 +         movq mm6, mm0
   4.324 +         movq mm4, [eax+ebx+colorA]
   4.325 +         movq mm5, [eax+ebx+colorB]
   4.326 +         pxor mm7, mm7
   4.327 +         pand mm6, [ONE]
   4.328 +
   4.329 +         movq mm0, [eax+colorE]
   4.330 +         movq mm1, [eax+ebx+colorG]
   4.331 +         movq mm2, mm0
   4.332 +         movq mm3, mm1
   4.333 +         pcmpeqw mm0, mm4
   4.334 +         pcmpeqw mm1, mm4
   4.335 +         pcmpeqw mm2, mm5
   4.336 +         pcmpeqw mm3, mm5
   4.337 +         pand mm0, mm6
   4.338 +         pand mm1, mm6
   4.339 +         pand mm2, mm6
   4.340 +         pand mm3, mm6
   4.341 +         paddw mm0, mm1
   4.342 +         paddw mm2, mm3
   4.343 +
   4.344 +         pxor mm3, mm3
   4.345 +         pcmpgtw mm0, mm6
   4.346 +         pcmpgtw mm2, mm6
   4.347 +         pcmpeqw mm0, mm3
   4.348 +         pcmpeqw mm2, mm3
   4.349 +         pand mm0, mm6
   4.350 +         pand mm2, mm6
   4.351 +         paddw mm7, mm0
   4.352 +         psubw mm7, mm2
   4.353 +
   4.354 +         movq mm0, [eax+colorF]
   4.355 +         movq mm1, [eax+ebx+colorK]
   4.356 +         movq mm2, mm0
   4.357 +         movq mm3, mm1
   4.358 +         pcmpeqw mm0, mm4
   4.359 +         pcmpeqw mm1, mm4
   4.360 +         pcmpeqw mm2, mm5
   4.361 +         pcmpeqw mm3, mm5
   4.362 +         pand mm0, mm6
   4.363 +         pand mm1, mm6
   4.364 +         pand mm2, mm6
   4.365 +         pand mm3, mm6
   4.366 +         paddw mm0, mm1
   4.367 +         paddw mm2, mm3
   4.368 +
   4.369 +         pxor mm3, mm3
   4.370 +         pcmpgtw mm0, mm6
   4.371 +         pcmpgtw mm2, mm6
   4.372 +         pcmpeqw mm0, mm3
   4.373 +         pcmpeqw mm2, mm3
   4.374 +         pand mm0, mm6
   4.375 +         pand mm2, mm6
   4.376 +         paddw mm7, mm0
   4.377 +         psubw mm7, mm2
   4.378 +
   4.379 +         push eax
   4.380 +         add eax, ebx
   4.381 +         movq mm0, [eax+ebx+colorH]
   4.382 +         movq mm1, [eax+ebx+ebx+colorN]
   4.383 +         movq mm2, mm0
   4.384 +         movq mm3, mm1
   4.385 +         pcmpeqw mm0, mm4
   4.386 +         pcmpeqw mm1, mm4
   4.387 +         pcmpeqw mm2, mm5
   4.388 +         pcmpeqw mm3, mm5
   4.389 +         pand mm0, mm6
   4.390 +         pand mm1, mm6
   4.391 +         pand mm2, mm6
   4.392 +         pand mm3, mm6
   4.393 +         paddw mm0, mm1
   4.394 +         paddw mm2, mm3
   4.395 +
   4.396 +         pxor mm3, mm3
   4.397 +         pcmpgtw mm0, mm6
   4.398 +         pcmpgtw mm2, mm6
   4.399 +         pcmpeqw mm0, mm3
   4.400 +         pcmpeqw mm2, mm3
   4.401 +         pand mm0, mm6
   4.402 +         pand mm2, mm6
   4.403 +         paddw mm7, mm0
   4.404 +         psubw mm7, mm2
   4.405 +
   4.406 +         movq mm0, [eax+ebx+colorL]
   4.407 +         movq mm1, [eax+ebx+ebx+colorO]
   4.408 +         movq mm2, mm0
   4.409 +         movq mm3, mm1
   4.410 +         pcmpeqw mm0, mm4
   4.411 +         pcmpeqw mm1, mm4
   4.412 +         pcmpeqw mm2, mm5
   4.413 +         pcmpeqw mm3, mm5
   4.414 +         pand mm0, mm6
   4.415 +         pand mm1, mm6
   4.416 +         pand mm2, mm6
   4.417 +         pand mm3, mm6
   4.418 +         paddw mm0, mm1
   4.419 +         paddw mm2, mm3
   4.420 +
   4.421 +         pxor mm3, mm3
   4.422 +         pcmpgtw mm0, mm6
   4.423 +         pcmpgtw mm2, mm6
   4.424 +         pcmpeqw mm0, mm3
   4.425 +         pcmpeqw mm2, mm3
   4.426 +         pand mm0, mm6
   4.427 +         pand mm2, mm6
   4.428 +         paddw mm7, mm0
   4.429 +         psubw mm7, mm2
   4.430 +
   4.431 +         pop eax
   4.432 +         movq mm1, mm7
   4.433 +         pxor mm0, mm0
   4.434 +         pcmpgtw mm7, mm0
   4.435 +         pcmpgtw mm0, mm1
   4.436 +
   4.437 +         por mm7, [Mask35]
   4.438 +         por mm0, [Mask26] 
   4.439 +         movq [Mask35], mm7
   4.440 +         movq [Mask26], mm0
   4.441 +
   4.442 +.SKIP_GUESS:
   4.443 +
   4.444 +         ;Start the ASSEMBLY !!!        eh... compose all the results together to form the final image...
   4.445 +
   4.446 +                 
   4.447 +         movq mm0, [eax+ebx+color5]
   4.448 +         movq mm1, [eax+ebx+ebx+color2]
   4.449 +         movq mm2, mm0
   4.450 +         movq mm3, mm1
   4.451 +         movq mm4, mm0
   4.452 +         movq mm5, mm1
   4.453 +
   4.454 +         pand mm0, [colorMask]
   4.455 +         pand mm1, [colorMask]
   4.456 +
   4.457 +         psrlw mm0, 1
   4.458 +         psrlw mm1, 1
   4.459 +
   4.460 +         pand mm3, [lowPixelMask]
   4.461 +         paddw mm0, mm1
   4.462 +
   4.463 +         pand mm3, mm2
   4.464 +         paddw mm0, mm3                ;mm0 contains the interpolated values
   4.465 +                 ;---------------------------
   4.466 +
   4.467 +
   4.468 +
   4.469 +%ifdef dfhsdfhsdahdsfhdsfh
   4.470 +
   4.471 +                if (color5 == color3 && color2 != color6 && color4 == color5 && color5 != colorA2)
   4.472 +                   product2a = INTERPOLATE (color2, color5);
   4.473 +                else
   4.474 +                if (color5 == color1 && color6 == color5 && color4 != color2 && color5 != colorA0)
   4.475 +                   product2a = INTERPOLATE(color2, color5);
   4.476 +                else
   4.477 +                   product2a = color2;
   4.478 +
   4.479 +                if (color2 == color6 && color5 != color3 && color1 == color2 && color2 != colorB2)
   4.480 +                   product1a = INTERPOLATE (color2, color5);
   4.481 +                else
   4.482 +                if (color4 == color2 && color3 == color2 && color1 != color5 && color2 != colorB0)
   4.483 +                   product1a = INTERPOLATE(color2, color5);
   4.484 +                else
   4.485 +                   product1a = color5;
   4.486 +
   4.487 +%endif
   4.488 +
   4.489 +
   4.490 +                 movq mm7, [Mask26]
   4.491 +                 movq mm6, [eax+colorB2]
   4.492 +                 movq mm5, [eax+ebx+ebx+color2]
   4.493 +                 movq mm4, [eax+ebx+ebx+color1]
   4.494 +                 pcmpeqw mm4, mm5
   4.495 +                 pcmpeqw mm6, mm5
   4.496 +                 pxor mm5, mm5
   4.497 +                 pand mm7, mm4
   4.498 +                 pcmpeqw mm6, mm5
   4.499 +                 pand mm7, mm6
   4.500 +
   4.501 +
   4.502 +
   4.503 +                 movq mm6, [eax+ebx+ebx+color3]
   4.504 +                 movq mm5, [eax+ebx+ebx+color2]
   4.505 +                 movq mm4, [eax+ebx+ebx+color1]
   4.506 +                 movq mm2, [eax+ebx+color5]
   4.507 +                 movq mm1, [eax+ebx+color4]
   4.508 +                 movq mm3, [eax+colorB0]
   4.509 +
   4.510 +                 pcmpeqw mm2, mm4
   4.511 +                 pcmpeqw mm6, mm5
   4.512 +                 pcmpeqw mm1, mm5
   4.513 +                 pcmpeqw mm3, mm5
   4.514 +                 pxor mm5, mm5
   4.515 +                 pcmpeqw mm2, mm5
   4.516 +                 pcmpeqw mm3, mm5
   4.517 +                 pand mm6, mm1
   4.518 +                 pand mm2, mm3
   4.519 +                 pand mm6, mm2
   4.520 +                 por mm7, mm6
   4.521 +
   4.522 +                 
   4.523 +                 movq mm6, mm7
   4.524 +                 pcmpeqw mm6, mm5
   4.525 +                 pand mm7, mm0
   4.526 +
   4.527 +                 movq mm1, [eax+ebx+color5]
   4.528 +                 pand mm6, mm1
   4.529 +                 por mm7, mm6
   4.530 +                 movq [final1a], mm7                    ;finished  1a
   4.531 +
   4.532 +
   4.533 +         
   4.534 +             ;--------------------------------           
   4.535 +
   4.536 +                 movq mm7, [Mask35]
   4.537 +                 push eax
   4.538 +                 add eax, ebx
   4.539 +                 movq mm6, [eax+ebx+ebx+colorA2]
   4.540 +                 pop eax
   4.541 +                 movq mm5, [eax+ebx+color5]
   4.542 +                 movq mm4, [eax+ebx+color4]
   4.543 +                 pcmpeqw mm4, mm5
   4.544 +                 pcmpeqw mm6, mm5
   4.545 +                 pxor mm5, mm5
   4.546 +                 pand mm7, mm4
   4.547 +                 pcmpeqw mm6, mm5
   4.548 +                 pand mm7, mm6
   4.549 +
   4.550 +
   4.551 +
   4.552 +                 movq mm6, [eax+ebx+color6]
   4.553 +                 movq mm5, [eax+ebx+color5]
   4.554 +                 movq mm4, [eax+ebx+color4]
   4.555 +                 movq mm2, [eax+ebx+ebx+color2]
   4.556 +                 movq mm1, [eax+ebx+ebx+color1]
   4.557 +                 push eax
   4.558 +                 add eax, ebx
   4.559 +                 movq mm3, [eax+ebx+ebx+colorA0]
   4.560 +                 pop eax
   4.561 +
   4.562 +                 pcmpeqw mm2, mm4
   4.563 +                 pcmpeqw mm6, mm5
   4.564 +                 pcmpeqw mm1, mm5
   4.565 +                 pcmpeqw mm3, mm5
   4.566 +                 pxor mm5, mm5
   4.567 +                 pcmpeqw mm2, mm5
   4.568 +                 pcmpeqw mm3, mm5
   4.569 +                 pand mm6, mm1
   4.570 +                 pand mm2, mm3
   4.571 +                 pand mm6, mm2
   4.572 +                 por mm7, mm6
   4.573 +
   4.574 +                 
   4.575 +                 movq mm6, mm7
   4.576 +                 pcmpeqw mm6, mm5
   4.577 +                 pand mm7, mm0
   4.578 +
   4.579 +                 movq mm1, [eax+ebx+ebx+color2]
   4.580 +                 pand mm6, mm1
   4.581 +                 por mm7, mm6
   4.582 +                 movq [final2a], mm7                    ;finished  2a
   4.583 +
   4.584 +
   4.585 +                 ;--------------------------------------------
   4.586 + 
   4.587 +
   4.588 +%ifdef dfhsdfhsdahdsfhdsfh
   4.589 +                   if (color6 == color3 && color3 == colorA1 && color2 != colorA2 && color3 != colorA0)
   4.590 +                      product2b = Q_INTERPOLATE (color3, color3, color3, color2);
   4.591 +                   else
   4.592 +                   if (color5 == color2 && color2 == colorA2 && colorA1 != color3 && color2 != colorA3)
   4.593 +                      product2b = Q_INTERPOLATE (color2, color2, color2, color3);
   4.594 +                   else
   4.595 +                      product2b = INTERPOLATE (color2, color3);
   4.596 +
   4.597 +                   if (color6 == color3 && color6 == colorB1 && color5 != colorB2 && color6 != colorB0)
   4.598 +                      product1b = Q_INTERPOLATE (color6, color6, color6, color5);
   4.599 +                   else
   4.600 +                   if (color5 == color2 && color5 == colorB2 && colorB1 != color6 && color5 != colorB3)
   4.601 +                      product1b = Q_INTERPOLATE (color6, color5, color5, color5);
   4.602 +                   else
   4.603 +                      product1b = INTERPOLATE (color5, color6);
   4.604 +%endif
   4.605 +
   4.606 +                 push eax
   4.607 +                 add eax, ebx
   4.608 +                 pxor mm7, mm7
   4.609 +                 movq mm0, [eax+ebx+ebx+colorA0]
   4.610 +                 movq mm1, [eax+ebx+ebx+colorA1]
   4.611 +                 movq mm2, [eax+ebx+ebx+colorA2]
   4.612 +                 movq mm3, [eax+ebx+ebx+colorA3]
   4.613 +                 pop eax
   4.614 +                 movq mm4, [eax+ebx+ebx+color2]
   4.615 +                 movq mm5, [eax+ebx+ebx+color3]
   4.616 +                 movq mm6, [eax+ebx+color6]
   4.617 +
   4.618 +                 pcmpeqw mm6, mm5
   4.619 +                 pcmpeqw mm1, mm5
   4.620 +                 pcmpeqw mm4, mm2
   4.621 +                 pcmpeqw mm0, mm5
   4.622 +                 pcmpeqw mm4, mm7
   4.623 +                 pcmpeqw mm0, mm7
   4.624 +                 pand mm0, mm4
   4.625 +                 pand mm6, mm1
   4.626 +                 pand mm0, mm6
   4.627 +
   4.628 +
   4.629 +                 push eax
   4.630 +                 add eax, ebx
   4.631 +                 movq mm1, [eax+ebx+ebx+colorA1]
   4.632 +                 pop eax
   4.633 +                 movq mm4, [eax+ebx+ebx+color2]
   4.634 +                 movq mm5, [eax+ebx+color5]
   4.635 +                 movq mm6, [eax+ebx+ebx+color3]
   4.636 +
   4.637 +                 pcmpeqw mm5, mm4
   4.638 +                 pcmpeqw mm2, mm4
   4.639 +                 pcmpeqw mm1, mm6
   4.640 +                 pcmpeqw mm3, mm4
   4.641 +                 pcmpeqw mm1, mm7
   4.642 +                 pcmpeqw mm3, mm7
   4.643 +                 pand mm2, mm5
   4.644 +                 pand mm1, mm3
   4.645 +                 pand mm1, mm2
   4.646 +
   4.647 +
   4.648 +                 movq mm7, mm0
   4.649 +                 por mm7, mm1
   4.650 +
   4.651 +                 movq mm4, [Mask35]
   4.652 +                 movq mm3, [Mask26]
   4.653 +                 
   4.654 +                 movq mm6, mm4
   4.655 +                 pand mm6, mm7
   4.656 +                 pxor mm4, mm6
   4.657 +
   4.658 +                 movq mm6, mm3
   4.659 +                 pand mm6, mm7
   4.660 +                 pxor mm3, mm6
   4.661 +
   4.662 +                 movq mm2, mm0
   4.663 +                 movq mm7, [I2333Pixel]
   4.664 +                 movq mm6, [I2223Pixel]
   4.665 +                 movq mm5, [I23Pixel]
   4.666 +
   4.667 +
   4.668 +                 por mm2, mm4
   4.669 +                 pand mm4, [eax+ebx+ebx+color3]
   4.670 +                 por mm2, mm3
   4.671 +                 pand mm3, [eax+ebx+ebx+color2]
   4.672 +                 por mm2, mm1
   4.673 +                 pand mm0, mm7
   4.674 +                 pand mm1, mm6
   4.675 +                 pxor mm7, mm7
   4.676 +                 pcmpeqw mm2, mm7
   4.677 +                 por mm0, mm1
   4.678 +                 por mm3, mm4
   4.679 +                 pand mm2, mm5
   4.680 +                 por mm0, mm3
   4.681 +                 por mm0, mm2
   4.682 +                 movq [final2b], mm0
   4.683 +
   4.684 +                 ;-----------------------------------
   4.685 +                 
   4.686 +
   4.687 +                 pxor mm7, mm7
   4.688 +                 movq mm0, [eax+colorB0]
   4.689 +                 movq mm1, [eax+colorB1]
   4.690 +                 movq mm2, [eax+colorB2]
   4.691 +                 movq mm3, [eax+colorB3]
   4.692 +                 movq mm4, [eax+ebx+color5]
   4.693 +                 movq mm5, [eax+ebx+color6]
   4.694 +                 movq mm6, [eax+ebx+ebx+color3]
   4.695 +
   4.696 +                 pcmpeqw mm6, mm5
   4.697 +                 pcmpeqw mm1, mm5
   4.698 +                 pcmpeqw mm4, mm2
   4.699 +                 pcmpeqw mm0, mm5
   4.700 +                 pcmpeqw mm4, mm7
   4.701 +                 pcmpeqw mm0, mm7
   4.702 +                 pand mm0, mm4
   4.703 +                 pand mm6, mm1
   4.704 +                 pand mm0, mm6
   4.705 +
   4.706 +                 movq mm1, [eax+colorB1]
   4.707 +                 movq mm4, [eax+ebx+color5]
   4.708 +                 movq mm5, [eax+ebx+ebx+color2]
   4.709 +                 movq mm6, [eax+ebx+color6]
   4.710 +
   4.711 +                 pcmpeqw mm5, mm4
   4.712 +                 pcmpeqw mm2, mm4
   4.713 +                 pcmpeqw mm1, mm6
   4.714 +                 pcmpeqw mm3, mm4
   4.715 +                 pcmpeqw mm1, mm7
   4.716 +                 pcmpeqw mm3, mm7
   4.717 +                 pand mm2, mm5
   4.718 +                 pand mm1, mm3
   4.719 +                 pand mm1, mm2
   4.720 +
   4.721 +
   4.722 +                 movq mm7, mm0
   4.723 +                 por mm7, mm1
   4.724 +
   4.725 +                 movq mm4, [Mask35]
   4.726 +                 movq mm3, [Mask26]
   4.727 +                 
   4.728 +                 movq mm6, mm4
   4.729 +                 pand mm6, mm7
   4.730 +                 pxor mm4, mm6
   4.731 +
   4.732 +                 movq mm6, mm3
   4.733 +                 pand mm6, mm7
   4.734 +                 pxor mm3, mm6
   4.735 +
   4.736 +                 movq mm2, mm0
   4.737 +                 movq mm7, [I5666Pixel]
   4.738 +                 movq mm6, [I5556Pixel]
   4.739 +                 movq mm5, [I56Pixel]
   4.740 +
   4.741 +
   4.742 +                 por mm2, mm4
   4.743 +                 pand mm4, [eax+ebx+color5]
   4.744 +                 por mm2, mm3
   4.745 +                 pand mm3, [eax+ebx+color6]
   4.746 +                 por mm2, mm1
   4.747 +                 pand mm0, mm7
   4.748 +                 pand mm1, mm6
   4.749 +                 pxor mm7, mm7
   4.750 +                 pcmpeqw mm2, mm7
   4.751 +                 por mm0, mm1
   4.752 +                 por mm3, mm4
   4.753 +                 pand mm2, mm5
   4.754 +                 por mm0, mm3
   4.755 +                 por mm0, mm2
   4.756 +                 movq [final1b], mm0
   4.757 +                 
   4.758 +          ;---------
   4.759 +
   4.760 +                 movq mm0, [final1a]
   4.761 +                 movq mm4, [final2a]
   4.762 +                 movq mm2, [final1b]
   4.763 +                 movq mm6, [final2b]
   4.764 +
   4.765 +
   4.766 +                 movq mm1, mm0
   4.767 +                 movq mm5, mm4
   4.768 +
   4.769 +
   4.770 +         punpcklwd mm0, mm2
   4.771 +         punpckhwd mm1, mm2
   4.772 +
   4.773 +         punpcklwd mm4, mm6
   4.774 +         punpckhwd mm5, mm6
   4.775 +
   4.776 +
   4.777 +%ifdef FAR_POINTER
   4.778 +         movq [fs:edx], mm0
   4.779 +         movq [fs:edx+8], mm1
   4.780 +         push edx
   4.781 +         add edx, [ebp+dstPitch]
   4.782 +         movq [fs:edx], mm4
   4.783 +         movq [fs:edx+8], mm5
   4.784 +         pop edx
   4.785 +%else
   4.786 +         movq [edx], mm0
   4.787 +         movq [edx+8], mm1
   4.788 +         push edx
   4.789 +         add edx, [ebp+dstPitch]
   4.790 +         movq [edx], mm4
   4.791 +         movq [edx+8], mm5
   4.792 +         pop edx
   4.793 +%endif
   4.794 +.SKIP_PROCESS:
   4.795 +         mov ecx, [ebp+deltaPtr]
   4.796 +         add ecx, 8
   4.797 +         mov [ebp+deltaPtr], ecx
   4.798 +         add edx, 16
   4.799 +         add eax, 8
   4.800 +
   4.801 +         pop ecx
   4.802 +         sub ecx, 4
   4.803 +         cmp ecx, 0
   4.804 +         jg  near .Loop
   4.805 +
   4.806 +; Restore some stuff
   4.807 +         popad
   4.808 +         mov esp, ebp
   4.809 +         pop ebp
   4.810 +         emms
   4.811 +         ret
   4.812 +
   4.813 +
   4.814 +;-------------------------------------------------------------------------
   4.815 +;-------------------------------------------------------------------------
   4.816 +;-------------------------------------------------------------------------
   4.817 +;-------------------------------------------------------------------------
   4.818 +;-------------------------------------------------------------------------
   4.819 +;-------------------------------------------------------------------------
   4.820 +;-------------------------------------------------------------------------
   4.821 +
   4.822 +
   4.823 +
   4.824 +%ifdef __DJGPP__
   4.825 +__2xSaISuperEagleLine:
   4.826 +%else
   4.827 +_2xSaISuperEagleLine:
   4.828 +%endif
   4.829 +; Store some stuff
   4.830 +         push ebp
   4.831 +         mov ebp, esp
   4.832 +         pushad
   4.833 +
   4.834 +; Prepare the destination
   4.835 +%ifdef FAR_POINTER
   4.836 +         ; Set the selector
   4.837 +         mov eax, [ebp+dstSegment]
   4.838 +         mov fs, ax
   4.839 +%endif
   4.840 +         mov edx, [ebp+dstOffset]         ; edx points to the screen
   4.841 +; Prepare the source
   4.842 +         ; eax points to colorA
   4.843 +         mov eax, [ebp+srcPtr]
   4.844 +         mov ebx, [ebp+srcPitch]
   4.845 +         mov ecx, [ebp+width]
   4.846 +         ; eax now points to colorB1
   4.847 +         sub eax, ebx
   4.848 +
   4.849 +; Main Loop
   4.850 +.Loop:   push ecx
   4.851 +
   4.852 +         ;-----Check Delta------------------
   4.853 +         mov ecx, [ebp+deltaPtr]
   4.854 +
   4.855 +         movq mm0, [eax+colorB0]
   4.856 +         movq mm1, [eax+colorB3]
   4.857 +         movq mm2, [eax+ebx+color4]
   4.858 +         movq mm3, [eax+ebx+colorS2]
   4.859 +         movq mm4, [eax+ebx+ebx+color1]
   4.860 +         movq mm5, [eax+ebx+ebx+colorS1]
   4.861 +         push eax
   4.862 +         add eax, ebx
   4.863 +         movq mm6, [eax+ebx+ebx+colorA0]
   4.864 +         movq mm7, [eax+ebx+ebx+colorA3]
   4.865 +         pop eax
   4.866 +
   4.867 +         pcmpeqw mm0, [ecx+2+colorB0]
   4.868 +         pcmpeqw mm1, [ecx+2+colorB3]
   4.869 +         pcmpeqw mm2, [ecx+ebx+2+color4]
   4.870 +         pcmpeqw mm3, [ecx+ebx+2+colorS2]
   4.871 +         pcmpeqw mm4, [ecx+ebx+ebx+2+color1]
   4.872 +         pcmpeqw mm5, [ecx+ebx+ebx+2+colorS1]
   4.873 +         add ecx, ebx
   4.874 +         pcmpeqw mm6, [ecx+ebx+ebx+2+colorA0]
   4.875 +         pcmpeqw mm7, [ecx+ebx+ebx+2+colorA3]
   4.876 +         sub ecx, ebx
   4.877 +
   4.878 +
   4.879 +         pand mm0, mm1
   4.880 +         pand mm2, mm3
   4.881 +         pand mm4, mm5
   4.882 +         pand mm6, mm7
   4.883 +         pand mm0, mm2
   4.884 +         pand mm4, mm6
   4.885 +         pxor mm7, mm7
   4.886 +         pand mm0, mm4
   4.887 +         movq mm6, [eax+colorB0]
   4.888 +         pcmpeqw mm7, mm0
   4.889 +
   4.890 +         movq [ecx+2+colorB0], mm6
   4.891 +
   4.892 +         packsswb mm7, mm7
   4.893 +         movd ecx, mm7
   4.894 +         test ecx, ecx
   4.895 +         jz near .SKIP_PROCESS
   4.896 +
   4.897 +         ;End Delta
   4.898 +
   4.899 +         ;---------------------------------
   4.900 +         movq mm0, [eax+ebx+color5]
   4.901 +         movq mm1, [eax+ebx+color6]
   4.902 +         movq mm2, mm0
   4.903 +         movq mm3, mm1
   4.904 +         movq mm4, mm0
   4.905 +         movq mm5, mm1
   4.906 +
   4.907 +         pand mm0, [colorMask]
   4.908 +         pand mm1, [colorMask]
   4.909 +
   4.910 +         psrlw mm0, 1
   4.911 +         psrlw mm1, 1
   4.912 +
   4.913 +         pand mm3, [lowPixelMask]
   4.914 +         paddw mm0, mm1
   4.915 +
   4.916 +         pand mm3, mm2
   4.917 +         paddw mm0, mm3                ;mm0 contains the interpolated values
   4.918 +         movq [I56Pixel], mm0
   4.919 +         movq mm7, mm0
   4.920 +
   4.921 +         ;-------------------
   4.922 +         movq mm0, mm7
   4.923 +         movq mm1, mm4  ;5,5,5,6
   4.924 +         movq mm2, mm0
   4.925 +         movq mm3, mm1
   4.926 +
   4.927 +         pand mm0, [colorMask]
   4.928 +         pand mm1, [colorMask]
   4.929 +
   4.930 +         psrlw mm0, 1
   4.931 +         psrlw mm1, 1
   4.932 +
   4.933 +         pand mm3, [lowPixelMask]
   4.934 +         paddw mm0, mm1
   4.935 +
   4.936 +         pand mm3, mm2
   4.937 +         paddw mm0, mm3                ;mm0 contains the interpolated values
   4.938 +         movq [product1a], mm0
   4.939 +         ;--------------------
   4.940 +
   4.941 +         movq mm0, mm7
   4.942 +         movq mm1, mm5  ;6,6,6,5
   4.943 +         movq mm2, mm0
   4.944 +         movq mm3, mm1
   4.945 +
   4.946 +         pand mm0, [colorMask]
   4.947 +         pand mm1, [colorMask]
   4.948 +
   4.949 +         psrlw mm0, 1
   4.950 +         psrlw mm1, 1
   4.951 +
   4.952 +         pand mm3, [lowPixelMask]
   4.953 +         paddw mm0, mm1
   4.954 +
   4.955 +         pand mm3, mm2
   4.956 +         paddw mm0, mm3
   4.957 +         movq [product1b], mm0
   4.958 +
   4.959 +         ;-------------------------
   4.960 +         ;-------------------------
   4.961 +         movq mm0, [eax+ebx+ebx+color2]
   4.962 +         movq mm1, [eax+ebx+ebx+color3]
   4.963 +         movq mm2, mm0
   4.964 +         movq mm3, mm1
   4.965 +         movq mm4, mm0
   4.966 +         movq mm5, mm1
   4.967 +
   4.968 +         pand mm0, [colorMask]
   4.969 +         pand mm1, [colorMask]
   4.970 +
   4.971 +         psrlw mm0, 1
   4.972 +         psrlw mm1, 1
   4.973 +
   4.974 +         pand mm3, [lowPixelMask]
   4.975 +         paddw mm0, mm1
   4.976 +
   4.977 +         pand mm3, mm2
   4.978 +         paddw mm0, mm3
   4.979 +         movq [I23Pixel], mm0
   4.980 +         movq mm7, mm0
   4.981 +
   4.982 +         ;---------------------
   4.983 +         movq mm0, mm7
   4.984 +         movq mm1, mm4  ;2,2,2,3
   4.985 +         movq mm2, mm0
   4.986 +         movq mm3, mm1
   4.987 +
   4.988 +         pand mm0, [colorMask]
   4.989 +         pand mm1, [colorMask]
   4.990 +
   4.991 +         psrlw mm0, 1
   4.992 +         psrlw mm1, 1
   4.993 +
   4.994 +         pand mm3, [lowPixelMask]
   4.995 +         paddw mm0, mm1
   4.996 +
   4.997 +         pand mm3, mm2
   4.998 +         paddw mm0, mm3
   4.999 +         movq [product2a], mm0
  4.1000 +
  4.1001 +         ;----------------------
  4.1002 +         movq mm0, mm7
  4.1003 +         movq mm1, mm5  ;3,3,3,2
  4.1004 +         movq mm2, mm0
  4.1005 +         movq mm3, mm1
  4.1006 +
  4.1007 +         pand mm0, [colorMask]
  4.1008 +         pand mm1, [colorMask]
  4.1009 +
  4.1010 +         psrlw mm0, 1
  4.1011 +         psrlw mm1, 1
  4.1012 +
  4.1013 +         pand mm3, [lowPixelMask]
  4.1014 +         paddw mm0, mm1
  4.1015 +
  4.1016 +         pand mm3, mm2
  4.1017 +         paddw mm0, mm3
  4.1018 +         movq [product2b], mm0
  4.1019 +
  4.1020 +
  4.1021 +         ;////////////////////////////////
  4.1022 +         ; Decide which "branch" to take
  4.1023 +         ;--------------------------------
  4.1024 +         movq mm4, [eax+ebx+color5]
  4.1025 +         movq mm5, [eax+ebx+color6]
  4.1026 +         movq mm6, [eax+ebx+ebx+color3]
  4.1027 +         movq mm7, [eax+ebx+ebx+color2]
  4.1028 +
  4.1029 +         pxor mm3, mm3
  4.1030 +         movq mm0, mm4
  4.1031 +         movq mm1, mm5
  4.1032 +
  4.1033 +         pcmpeqw mm0, mm6
  4.1034 +         pcmpeqw mm1, mm7
  4.1035 +         pcmpeqw mm1, mm3
  4.1036 +         pand mm0, mm1
  4.1037 +         movq [Mask35], mm0
  4.1038 +
  4.1039 +         movq mm0, [eax+ebx+ebx+colorS1]
  4.1040 +         movq mm1, [eax+ebx+color4]
  4.1041 +         push eax
  4.1042 +         add eax, ebx
  4.1043 +         movq mm2, [eax+ebx+ebx+colorA2]
  4.1044 +         pop eax
  4.1045 +         movq mm3, [eax+colorB1]
  4.1046 +         pcmpeqw mm0, mm4
  4.1047 +         pcmpeqw mm1, mm4
  4.1048 +         pcmpeqw mm2, mm4
  4.1049 +         pcmpeqw mm3, mm4
  4.1050 +         pand mm0, mm1
  4.1051 +         pand mm2, mm3
  4.1052 +         por mm0, mm2
  4.1053 +         pand mm0, [Mask35]
  4.1054 +         movq [Mask35b], mm0
  4.1055 +
  4.1056 +         ;-----------
  4.1057 +         pxor mm3, mm3
  4.1058 +         movq mm0, mm4
  4.1059 +         movq mm1, mm5
  4.1060 +
  4.1061 +         pcmpeqw mm0, mm6
  4.1062 +         pcmpeqw mm1, mm7
  4.1063 +         pcmpeqw mm0, mm3
  4.1064 +         pand mm0, mm1
  4.1065 +         movq [Mask26], mm0
  4.1066 +
  4.1067 +         movq mm0, [eax+ebx+ebx+color1]
  4.1068 +         movq mm1, [eax+ebx+colorS2]
  4.1069 +         push eax
  4.1070 +         add eax, ebx
  4.1071 +         movq mm2, [eax+ebx+ebx+colorA1]
  4.1072 +         pop eax
  4.1073 +         movq mm3, [eax+colorB2]
  4.1074 +         pcmpeqw mm0, mm5
  4.1075 +         pcmpeqw mm1, mm5
  4.1076 +         pcmpeqw mm2, mm5
  4.1077 +         pcmpeqw mm3, mm5
  4.1078 +         pand mm0, mm1
  4.1079 +         pand mm2, mm3
  4.1080 +         por mm0, mm2
  4.1081 +         pand mm0, [Mask26]
  4.1082 +         movq [Mask26b], mm0
  4.1083 +
  4.1084 +         ;--------------------
  4.1085 +         movq mm0, mm4
  4.1086 +         movq mm1, mm5
  4.1087 +         movq mm2, mm0
  4.1088 +
  4.1089 +         pcmpeqw mm2, mm1
  4.1090 +         pcmpeqw mm0, mm6
  4.1091 +         pcmpeqw mm1, mm7
  4.1092 +         pand mm0, mm1
  4.1093 +         pand mm2, mm0
  4.1094 +         pxor mm0, mm2
  4.1095 +         movq mm7, mm0
  4.1096 +
  4.1097 +         ;------------------
  4.1098 +         packsswb mm7, mm7
  4.1099 +         movd ecx, mm7
  4.1100 +         test ecx, ecx
  4.1101 +         jz near .SKIP_GUESS
  4.1102 +
  4.1103 +;---------------------------------------------
  4.1104 +; Map of the pixels:                    I|E F|J
  4.1105 +;                                       G|A B|K
  4.1106 +;                                       H|C D|L
  4.1107 +;                                       M|N O|P
  4.1108 +         movq mm6, mm0
  4.1109 +         movq mm4, [eax+ebx+color5]
  4.1110 +         movq mm5, [eax+ebx+color6]
  4.1111 +         pxor mm7, mm7
  4.1112 +         pand mm6, [ONE]
  4.1113 +
  4.1114 +         movq mm0, [eax+colorB1]
  4.1115 +         movq mm1, [eax+ebx+color4]
  4.1116 +         movq mm2, mm0
  4.1117 +         movq mm3, mm1
  4.1118 +         pcmpeqw mm0, mm4
  4.1119 +         pcmpeqw mm1, mm4
  4.1120 +         pcmpeqw mm2, mm5
  4.1121 +         pcmpeqw mm3, mm5
  4.1122 +         pand mm0, mm6
  4.1123 +         pand mm1, mm6
  4.1124 +         pand mm2, mm6
  4.1125 +         pand mm3, mm6
  4.1126 +         paddw mm0, mm1
  4.1127 +         paddw mm2, mm3
  4.1128 +
  4.1129 +         pxor mm3, mm3
  4.1130 +         pcmpgtw mm0, mm6
  4.1131 +         pcmpgtw mm2, mm6
  4.1132 +         pcmpeqw mm0, mm3
  4.1133 +         pcmpeqw mm2, mm3
  4.1134 +         pand mm0, mm6
  4.1135 +         pand mm2, mm6
  4.1136 +         paddw mm7, mm0
  4.1137 +         psubw mm7, mm2
  4.1138 +
  4.1139 +         movq mm0, [eax+colorB2]
  4.1140 +         movq mm1, [eax+ebx+colorS2]
  4.1141 +         movq mm2, mm0
  4.1142 +         movq mm3, mm1
  4.1143 +         pcmpeqw mm0, mm4
  4.1144 +         pcmpeqw mm1, mm4
  4.1145 +         pcmpeqw mm2, mm5
  4.1146 +         pcmpeqw mm3, mm5
  4.1147 +         pand mm0, mm6
  4.1148 +         pand mm1, mm6
  4.1149 +         pand mm2, mm6
  4.1150 +         pand mm3, mm6
  4.1151 +         paddw mm0, mm1
  4.1152 +         paddw mm2, mm3
  4.1153 +
  4.1154 +         pxor mm3, mm3
  4.1155 +         pcmpgtw mm0, mm6
  4.1156 +         pcmpgtw mm2, mm6
  4.1157 +         pcmpeqw mm0, mm3
  4.1158 +         pcmpeqw mm2, mm3
  4.1159 +         pand mm0, mm6
  4.1160 +         pand mm2, mm6
  4.1161 +         paddw mm7, mm0
  4.1162 +         psubw mm7, mm2
  4.1163 +
  4.1164 +         push eax
  4.1165 +         add eax, ebx
  4.1166 +         movq mm0, [eax+ebx+color1]
  4.1167 +         movq mm1, [eax+ebx+ebx+colorA1]
  4.1168 +         movq mm2, mm0
  4.1169 +         movq mm3, mm1
  4.1170 +         pcmpeqw mm0, mm4
  4.1171 +         pcmpeqw mm1, mm4
  4.1172 +         pcmpeqw mm2, mm5
  4.1173 +         pcmpeqw mm3, mm5
  4.1174 +         pand mm0, mm6
  4.1175 +         pand mm1, mm6
  4.1176 +         pand mm2, mm6
  4.1177 +         pand mm3, mm6
  4.1178 +         paddw mm0, mm1
  4.1179 +         paddw mm2, mm3
  4.1180 +
  4.1181 +         pxor mm3, mm3
  4.1182 +         pcmpgtw mm0, mm6
  4.1183 +         pcmpgtw mm2, mm6
  4.1184 +         pcmpeqw mm0, mm3
  4.1185 +         pcmpeqw mm2, mm3
  4.1186 +         pand mm0, mm6
  4.1187 +         pand mm2, mm6
  4.1188 +         paddw mm7, mm0
  4.1189 +         psubw mm7, mm2
  4.1190 +
  4.1191 +         movq mm0, [eax+ebx+colorS1]
  4.1192 +         movq mm1, [eax+ebx+ebx+colorA2]
  4.1193 +         movq mm2, mm0
  4.1194 +         movq mm3, mm1
  4.1195 +         pcmpeqw mm0, mm4
  4.1196 +         pcmpeqw mm1, mm4
  4.1197 +         pcmpeqw mm2, mm5
  4.1198 +         pcmpeqw mm3, mm5
  4.1199 +         pand mm0, mm6
  4.1200 +         pand mm1, mm6
  4.1201 +         pand mm2, mm6
  4.1202 +         pand mm3, mm6
  4.1203 +         paddw mm0, mm1
  4.1204 +         paddw mm2, mm3
  4.1205 +
  4.1206 +         pxor mm3, mm3
  4.1207 +         pcmpgtw mm0, mm6
  4.1208 +         pcmpgtw mm2, mm6
  4.1209 +         pcmpeqw mm0, mm3
  4.1210 +         pcmpeqw mm2, mm3
  4.1211 +         pand mm0, mm6
  4.1212 +         pand mm2, mm6
  4.1213 +         paddw mm7, mm0
  4.1214 +         psubw mm7, mm2
  4.1215 +
  4.1216 +         pop eax
  4.1217 +         movq mm1, mm7
  4.1218 +         pxor mm0, mm0
  4.1219 +         pcmpgtw mm7, mm0
  4.1220 +         pcmpgtw mm0, mm1
  4.1221 +
  4.1222 +         por mm7, [Mask35]
  4.1223 +         por mm0, [Mask26]
  4.1224 +         movq [Mask35], mm7
  4.1225 +         movq [Mask26], mm0
  4.1226 +
  4.1227 +.SKIP_GUESS:
  4.1228 +         ;Start the ASSEMBLY !!!
  4.1229 +
  4.1230 +         movq mm4, [Mask35]
  4.1231 +         movq mm5, [Mask26]
  4.1232 +         movq mm6, [Mask35b]
  4.1233 +         movq mm7, [Mask26b]
  4.1234 +
  4.1235 +         movq mm0, [eax+ebx+color5]
  4.1236 +         movq mm1, [eax+ebx+color6]
  4.1237 +         movq mm2, [eax+ebx+ebx+color2]
  4.1238 +         movq mm3, [eax+ebx+ebx+color3]
  4.1239 +         pcmpeqw mm0, mm2
  4.1240 +         pcmpeqw mm1, mm3
  4.1241 +         movq mm2, mm4
  4.1242 +         movq mm3, mm5
  4.1243 +         por mm0, mm1
  4.1244 +         por mm2, mm3
  4.1245 +         pand mm2, mm0
  4.1246 +         pxor mm0, mm2
  4.1247 +         movq mm3, mm0
  4.1248 +
  4.1249 +         movq mm2, mm0
  4.1250 +         pxor mm0, mm0
  4.1251 +         por mm2, mm4
  4.1252 +         pxor mm4, mm6
  4.1253 +         por mm2, mm5
  4.1254 +         pxor mm5, mm7
  4.1255 +         pcmpeqw mm2, mm0
  4.1256 +         ;----------------
  4.1257 +
  4.1258 +         movq mm0, [eax+ebx+color5]
  4.1259 +         movq mm1, mm3
  4.1260 +         por mm1, mm4
  4.1261 +         por mm1, mm6
  4.1262 +         pand mm0, mm1
  4.1263 +         movq mm1, mm5
  4.1264 +         pand mm1, [I56Pixel]
  4.1265 +         por mm0, mm1
  4.1266 +         movq mm1, mm7
  4.1267 +         pand mm1, [product1b]
  4.1268 +         por mm0, mm1
  4.1269 +         movq mm1, mm2
  4.1270 +         pand mm1, [product1a]
  4.1271 +         por mm0, mm1
  4.1272 +         movq [final1a], mm0
  4.1273 +
  4.1274 +         movq mm0, [eax+ebx+color6]
  4.1275 +         movq mm1, mm3
  4.1276 +         por mm1, mm5
  4.1277 +         por mm1, mm7
  4.1278 +         pand mm0, mm1
  4.1279 +         movq mm1, mm4
  4.1280 +         pand mm1, [I56Pixel]
  4.1281 +         por mm0, mm1
  4.1282 +         movq mm1, mm6
  4.1283 +         pand mm1, [product1a]
  4.1284 +         por mm0, mm1
  4.1285 +         movq mm1, mm2
  4.1286 +         pand mm1, [product1b]
  4.1287 +         por mm0, mm1
  4.1288 +         movq [final1b], mm0
  4.1289 +
  4.1290 +         movq mm0, [eax+ebx+ebx+color2]
  4.1291 +         movq mm1, mm3
  4.1292 +         por mm1, mm5
  4.1293 +         por mm1, mm7
  4.1294 +         pand mm0, mm1
  4.1295 +         movq mm1, mm4
  4.1296 +         pand mm1, [I23Pixel]
  4.1297 +         por mm0, mm1
  4.1298 +         movq mm1, mm6
  4.1299 +         pand mm1, [product2b]
  4.1300 +         por mm0, mm1
  4.1301 +         movq mm1, mm2
  4.1302 +         pand mm1, [product2a]
  4.1303 +         por mm0, mm1
  4.1304 +         movq [final2a], mm0
  4.1305 +
  4.1306 +         movq mm0, [eax+ebx+ebx+color3]
  4.1307 +         movq mm1, mm3
  4.1308 +         por mm1, mm4
  4.1309 +         por mm1, mm6
  4.1310 +         pand mm0, mm1
  4.1311 +         movq mm1, mm5
  4.1312 +         pand mm1, [I23Pixel]
  4.1313 +         por mm0, mm1
  4.1314 +         movq mm1, mm7
  4.1315 +         pand mm1, [product2a]
  4.1316 +         por mm0, mm1
  4.1317 +         movq mm1, mm2
  4.1318 +         pand mm1, [product2b]
  4.1319 +         por mm0, mm1
  4.1320 +         movq [final2b], mm0
  4.1321 +
  4.1322 +
  4.1323 +         movq mm0, [final1a]
  4.1324 +         movq mm2, [final1b]
  4.1325 +         movq mm1, mm0
  4.1326 +         movq mm4, [final2a]
  4.1327 +         movq mm6, [final2b]
  4.1328 +         movq mm5, mm4
  4.1329 +         punpcklwd mm0, mm2
  4.1330 +         punpckhwd mm1, mm2
  4.1331 +         punpcklwd mm4, mm6
  4.1332 +         punpckhwd mm5, mm6
  4.1333 +
  4.1334 +
  4.1335 +
  4.1336 +
  4.1337 +%ifdef FAR_POINTER
  4.1338 +         movq [fs:edx], mm0
  4.1339 +         movq [fs:edx+8], mm1
  4.1340 +         push edx
  4.1341 +         add edx, [ebp+dstPitch]
  4.1342 +         movq [fs:edx], mm4
  4.1343 +         movq [fs:edx+8], mm5
  4.1344 +         pop edx
  4.1345 +%else
  4.1346 +         movq [edx], mm0
  4.1347 +         movq [edx+8], mm1
  4.1348 +         push edx
  4.1349 +         add edx, [ebp+dstPitch]
  4.1350 +         movq [edx], mm4
  4.1351 +         movq [edx+8], mm5
  4.1352 +         pop edx
  4.1353 +%endif
  4.1354 +.SKIP_PROCESS:
  4.1355 +         mov ecx, [ebp+deltaPtr]
  4.1356 +         add ecx, 8
  4.1357 +         mov [ebp+deltaPtr], ecx
  4.1358 +         add edx, 16
  4.1359 +         add eax, 8
  4.1360 +
  4.1361 +         pop ecx
  4.1362 +         sub ecx, 4
  4.1363 +         cmp ecx, 0
  4.1364 +         jg  near .Loop
  4.1365 +
  4.1366 +; Restore some stuff
  4.1367 +         popad
  4.1368 +         mov esp, ebp
  4.1369 +         pop ebp
  4.1370 +         emms
  4.1371 +         ret
  4.1372 +
  4.1373 +
  4.1374 +;-------------------------------------------------------------------------
  4.1375 +;-------------------------------------------------------------------------
  4.1376 +;-------------------------------------------------------------------------
  4.1377 +;-------------------------------------------------------------------------
  4.1378 +;-------------------------------------------------------------------------
  4.1379 +;-------------------------------------------------------------------------
  4.1380 +;-------------------------------------------------------------------------
  4.1381 +
  4.1382 +
  4.1383 +;This is version 0.50
  4.1384 +colorI   equ -2
  4.1385 +colorE   equ 0
  4.1386 +colorF   equ 2
  4.1387 +colorJ   equ 4
  4.1388 +
  4.1389 +colorG   equ -2
  4.1390 +colorA   equ 0
  4.1391 +colorB   equ 2
  4.1392 +colorK   equ 4
  4.1393 +
  4.1394 +colorH   equ -2
  4.1395 +colorC   equ 0
  4.1396 +colorD   equ 2
  4.1397 +colorL   equ 4
  4.1398 +
  4.1399 +colorM   equ -2
  4.1400 +colorN   equ 0
  4.1401 +colorO   equ 2
  4.1402 +colorP   equ 4
  4.1403 +
  4.1404 +%ifdef __DJGPP__
  4.1405 +__2xSaILine:
  4.1406 +%else
  4.1407 +_2xSaILine:
  4.1408 +%endif
  4.1409 +; Store some stuff
  4.1410 +         push ebp
  4.1411 +         mov ebp, esp
  4.1412 +         pushad
  4.1413 +
  4.1414 +; Prepare the destination
  4.1415 +%ifdef FAR_POINTER
  4.1416 +         ; Set the selector
  4.1417 +         mov eax, [ebp+dstSegment]
  4.1418 +         mov fs, ax
  4.1419 +%endif
  4.1420 +         mov edx, [ebp+dstOffset]         ; edx points to the screen
  4.1421 +; Prepare the source
  4.1422 +         ; eax points to colorA
  4.1423 +         mov eax, [ebp+srcPtr]
  4.1424 +         mov ebx, [ebp+srcPitch]
  4.1425 +         mov ecx, [ebp+width]
  4.1426 +         ; eax now points to colorE
  4.1427 +         sub eax, ebx
  4.1428 +
  4.1429 +
  4.1430 +; Main Loop
  4.1431 +.Loop:   push ecx
  4.1432 +
  4.1433 +         ;-----Check Delta------------------
  4.1434 +         mov ecx, [ebp+deltaPtr]
  4.1435 +
  4.1436 +         movq mm0, [eax+colorI]
  4.1437 +         movq mm1, [eax+colorJ]
  4.1438 +         movq mm2, [eax+ebx+colorG]
  4.1439 +         movq mm3, [eax+ebx+colorK]
  4.1440 +         movq mm4, [eax+ebx+ebx+colorH]
  4.1441 +         movq mm5, [eax+ebx+ebx+colorL]
  4.1442 +         push eax
  4.1443 +         add eax, ebx
  4.1444 +         movq mm6, [eax+ebx+ebx+colorM]
  4.1445 +         movq mm7, [eax+ebx+ebx+colorP]
  4.1446 +         pop eax
  4.1447 +
  4.1448 +         pcmpeqw mm0, [ecx+2+colorI]
  4.1449 +         pcmpeqw mm1, [ecx+2+colorK]
  4.1450 +         pcmpeqw mm2, [ecx+ebx+2+colorG]
  4.1451 +         pcmpeqw mm3, [ecx+ebx+2+colorK]
  4.1452 +         pcmpeqw mm4, [ecx+ebx+ebx+2+colorH]
  4.1453 +         pcmpeqw mm5, [ecx+ebx+ebx+2+colorL]
  4.1454 +         add ecx, ebx
  4.1455 +         pcmpeqw mm6, [ecx+ebx+ebx+2+colorM]
  4.1456 +         pcmpeqw mm7, [ecx+ebx+ebx+2+colorP]
  4.1457 +         sub ecx, ebx
  4.1458 +
  4.1459 +
  4.1460 +         pand mm0, mm1
  4.1461 +         pand mm2, mm3
  4.1462 +         pand mm4, mm5
  4.1463 +         pand mm6, mm7
  4.1464 +         pand mm0, mm2
  4.1465 +         pand mm4, mm6
  4.1466 +         pxor mm7, mm7
  4.1467 +         pand mm0, mm4
  4.1468 +         movq mm6, [eax+colorI]
  4.1469 +         pcmpeqw mm7, mm0
  4.1470 +
  4.1471 +         movq [ecx+2+colorI], mm6
  4.1472 +
  4.1473 +         packsswb mm7, mm7
  4.1474 +         movd ecx, mm7
  4.1475 +         test ecx, ecx
  4.1476 +         jz near .SKIP_PROCESS
  4.1477 +
  4.1478 +         ;End Delta
  4.1479 +
  4.1480 +         ;---------------------------------
  4.1481 +
  4.1482 +
  4.1483 +;1
  4.1484 +         ;if ((colorA == colorD) && (colorB != colorC) && (colorA == colorE) && (colorB == colorL)
  4.1485 +         movq mm0, [eax+ebx+colorA]        ;mm0 and mm1 contain colorA
  4.1486 +         movq mm2, [eax+ebx+colorB]        ;mm2 and mm3 contain colorB
  4.1487 +
  4.1488 +         movq mm1, mm0
  4.1489 +         movq mm3, mm2
  4.1490 +
  4.1491 +         pcmpeqw mm0, [eax+ebx+ebx+colorD]
  4.1492 +         pcmpeqw mm1, [eax+colorE]
  4.1493 +         pcmpeqw mm2, [eax+ebx+ebx+colorL]
  4.1494 +         pcmpeqw mm3, [eax+ebx+ebx+colorC]
  4.1495 +
  4.1496 +         pand mm0, mm1
  4.1497 +         pxor mm1, mm1
  4.1498 +         pand mm0, mm2
  4.1499 +         pcmpeqw mm3, mm1
  4.1500 +         pand mm0, mm3                 ;result in mm0
  4.1501 +
  4.1502 +         ;if ((colorA == colorC) && (colorB != colorE) && (colorA == colorF) && (colorB == colorJ)
  4.1503 +         movq mm4, [eax+ebx+colorA]        ;mm4 and mm5 contain colorA
  4.1504 +         movq mm6, [eax+ebx+colorB]        ;mm6 and mm7 contain colorB
  4.1505 +         movq mm5, mm4
  4.1506 +         movq mm7, mm6
  4.1507 +
  4.1508 +         pcmpeqw mm4, [eax+ebx+ebx+colorC]
  4.1509 +         pcmpeqw mm5, [eax+colorF]
  4.1510 +         pcmpeqw mm6, [eax+colorJ]
  4.1511 +         pcmpeqw mm7, [eax+colorE]
  4.1512 +
  4.1513 +         pand mm4, mm5
  4.1514 +         pxor mm5, mm5
  4.1515 +         pand mm4, mm6
  4.1516 +         pcmpeqw mm7, mm5
  4.1517 +         pand mm4, mm7                 ;result in mm4
  4.1518 +
  4.1519 +         por mm0, mm4                  ;combine the masks
  4.1520 +         movq [Mask1], mm0
  4.1521 +
  4.1522 +         ;--------------------------------------------
  4.1523 +
  4.1524 +;2
  4.1525 +         ;if ((colorB == colorC) && (colorA != colorD) && (colorB == colorF) && (colorA == colorH)
  4.1526 +         movq mm0, [eax+ebx+colorB]        ;mm0 and mm1 contain colorB
  4.1527 +         movq mm2, [eax+ebx+colorA]        ;mm2 and mm3 contain colorA
  4.1528 +         movq mm1, mm0
  4.1529 +         movq mm3, mm2
  4.1530 +
  4.1531 +         pcmpeqw mm0, [eax+ebx+ebx+colorC]
  4.1532 +         pcmpeqw mm1, [eax+colorF]
  4.1533 +         pcmpeqw mm2, [eax+ebx+ebx+colorH]
  4.1534 +         pcmpeqw mm3, [eax+ebx+ebx+colorD]
  4.1535 +
  4.1536 +         pand mm0, mm1
  4.1537 +         pxor mm1, mm1
  4.1538 +         pand mm0, mm2
  4.1539 +         pcmpeqw mm3, mm1
  4.1540 +         pand mm0, mm3                 ;result in mm0
  4.1541 +
  4.1542 +         ;if ((colorB == colorE) && (colorB == colorD) && (colorA != colorF) && (colorA == colorI)
  4.1543 +         movq mm4, [eax+ebx+colorB]        ;mm4 and mm5 contain colorB
  4.1544 +         movq mm6, [eax+ebx+colorA]        ;mm6 and mm7 contain colorA
  4.1545 +         movq mm5, mm4
  4.1546 +         movq mm7, mm6
  4.1547 +
  4.1548 +         pcmpeqw mm4, [eax+ebx+ebx+colorD]
  4.1549 +         pcmpeqw mm5, [eax+colorE]
  4.1550 +         pcmpeqw mm6, [eax+colorI]
  4.1551 +         pcmpeqw mm7, [eax+colorF]
  4.1552 +
  4.1553 +         pand mm4, mm5
  4.1554 +         pxor mm5, mm5
  4.1555 +         pand mm4, mm6
  4.1556 +         pcmpeqw mm7, mm5
  4.1557 +         pand mm4, mm7                 ;result in mm4
  4.1558 +
  4.1559 +         por mm0, mm4                  ;combine the masks
  4.1560 +         movq [Mask2], mm0
  4.1561 +
  4.1562 +
  4.1563 +;interpolate colorA and colorB
  4.1564 +         movq mm0, [eax+ebx+colorA]
  4.1565 +         movq mm1, [eax+ebx+colorB]
  4.1566 +
  4.1567 +         movq mm2, mm0
  4.1568 +         movq mm3, mm1
  4.1569 +
  4.1570 +         pand mm0, [colorMask]
  4.1571 +         pand mm1, [colorMask]
  4.1572 +
  4.1573 +         psrlw mm0, 1
  4.1574 +         psrlw mm1, 1
  4.1575 +
  4.1576 +         pand mm3, [lowPixelMask]
  4.1577 +         paddw mm0, mm1
  4.1578 +
  4.1579 +         pand mm3, mm2
  4.1580 +         paddw mm0, mm3                ;mm0 contains the interpolated values
  4.1581 +
  4.1582 +         ;assemble the pixels
  4.1583 +         movq mm1, [eax+ebx+colorA]
  4.1584 +         movq mm2, [eax+ebx+colorB]
  4.1585 +
  4.1586 +         movq mm3, [Mask1]
  4.1587 +         movq mm5, mm1
  4.1588 +         movq mm4, [Mask2]
  4.1589 +         movq mm6, mm1
  4.1590 +
  4.1591 +         pand mm1, mm3
  4.1592 +         por mm3, mm4
  4.1593 +         pxor mm7, mm7
  4.1594 +         pand mm2, mm4
  4.1595 +
  4.1596 +         pcmpeqw mm3, mm7
  4.1597 +         por mm1, mm2
  4.1598 +         pand mm0, mm3
  4.1599 +
  4.1600 +         por mm0, mm1
  4.1601 +
  4.1602 +         punpcklwd mm5, mm0
  4.1603 +         punpckhwd mm6, mm0
  4.1604 +
  4.1605 +%ifdef FAR_POINTER
  4.1606 +         movq [fs:edx], mm5
  4.1607 +         movq [fs:edx+8], mm6
  4.1608 +%else
  4.1609 +         movq [edx], mm5
  4.1610 +         movq [edx+8], mm6
  4.1611 +%endif
  4.1612 +
  4.1613 +;------------------------------------------------
  4.1614 +;        Create the Nextline
  4.1615 +;------------------------------------------------
  4.1616 +;3       ;if ((colorA == colorD) && (colorB != colorC) && (colorA == colorG) && (colorC == colorO)
  4.1617 +         movq mm0, [eax+ebx+colorA]        ;mm0 and mm1 contain colorA
  4.1618 +         movq mm2, [eax+ebx+ebx+colorC]        ;mm2 and mm3 contain colorC
  4.1619 +         movq mm1, mm0
  4.1620 +         movq mm3, mm2
  4.1621 +
  4.1622 +         push eax
  4.1623 +         add eax, ebx
  4.1624 +         pcmpeqw mm0, [eax+ebx+colorD]
  4.1625 +         pcmpeqw mm1, [eax+colorG]
  4.1626 +         pcmpeqw mm2, [eax+ebx+ebx+colorO]
  4.1627 +         pcmpeqw mm3, [eax+colorB]
  4.1628 +         pop eax
  4.1629 +
  4.1630 +         pand mm0, mm1
  4.1631 +         pxor mm1, mm1
  4.1632 +         pand mm0, mm2
  4.1633 +         pcmpeqw mm3, mm1
  4.1634 +         pand mm0, mm3                 ;result in mm0
  4.1635 +
  4.1636 +         ;if ((colorA == colorB) && (colorG != colorC) && (colorA == colorH) && (colorC == colorM)
  4.1637 +         movq mm4, [eax+ebx+colorA]        ;mm4 and mm5 contain colorA
  4.1638 +         movq mm6, [eax+ebx+ebx+colorC]        ;mm6 and mm7 contain colorC
  4.1639 +         movq mm5, mm4
  4.1640 +         movq mm7, mm6
  4.1641 +
  4.1642 +         push eax
  4.1643 +         add eax, ebx
  4.1644 +         pcmpeqw mm4, [eax+ebx+colorH]
  4.1645 +         pcmpeqw mm5, [eax+colorB]
  4.1646 +         pcmpeqw mm6, [eax+ebx+ebx+colorM]
  4.1647 +         pcmpeqw mm7, [eax+colorG]
  4.1648 +         pop eax
  4.1649 +
  4.1650 +         pand mm4, mm5
  4.1651 +         pxor mm5, mm5
  4.1652 +         pand mm4, mm6
  4.1653 +         pcmpeqw mm7, mm5
  4.1654 +         pand mm4, mm7                 ;result in mm4
  4.1655 +
  4.1656 +         por mm0, mm4                  ;combine the masks
  4.1657 +         movq [Mask1], mm0
  4.1658 +         ;--------------------------------------------
  4.1659 +
  4.1660 +;4
  4.1661 +         ;if ((colorB == colorC) && (colorA != colorD) && (colorC == colorH) && (colorA == colorF)
  4.1662 +         movq mm0, [eax+ebx+ebx+colorC]        ;mm0 and mm1 contain colorC
  4.1663 +         movq mm2, [eax+ebx+colorA]        ;mm2 and mm3 contain colorA
  4.1664 +         movq mm1, mm0
  4.1665 +         movq mm3, mm2
  4.1666 +
  4.1667 +         pcmpeqw mm0, [eax+ebx+colorB]
  4.1668 +         pcmpeqw mm1, [eax+ebx+ebx+colorH]
  4.1669 +         pcmpeqw mm2, [eax+colorF]
  4.1670 +         pcmpeqw mm3, [eax+ebx+ebx+colorD]
  4.1671 +
  4.1672 +         pand mm0, mm1
  4.1673 +         pxor mm1, mm1
  4.1674 +         pand mm0, mm2
  4.1675 +         pcmpeqw mm3, mm1
  4.1676 +         pand mm0, mm3                 ;result in mm0
  4.1677 +
  4.1678 +         ;if ((colorC == colorG) && (colorC == colorD) && (colorA != colorH) && (colorA == colorI)
  4.1679 +         movq mm4, [eax+ebx+ebx+colorC]        ;mm4 and mm5 contain colorC
  4.1680 +         movq mm6, [eax+ebx+colorA]        ;mm6 and mm7 contain colorA
  4.1681 +         movq mm5, mm4
  4.1682 +         movq mm7, mm6
  4.1683 +
  4.1684 +         pcmpeqw mm4, [eax+ebx+ebx+colorD]
  4.1685 +         pcmpeqw mm5, [eax+ebx+colorG]
  4.1686 +         pcmpeqw mm6, [eax+colorI]
  4.1687 +         pcmpeqw mm7, [eax+ebx+ebx+colorH]
  4.1688 +
  4.1689 +         pand mm4, mm5
  4.1690 +         pxor mm5, mm5
  4.1691 +         pand mm4, mm6
  4.1692 +         pcmpeqw mm7, mm5
  4.1693 +         pand mm4, mm7                 ;result in mm4
  4.1694 +
  4.1695 +         por mm0, mm4                  ;combine the masks
  4.1696 +         movq [Mask2], mm0
  4.1697 +         ;----------------------------------------------
  4.1698 +
  4.1699 +;interpolate colorA and colorC
  4.1700 +         movq mm0, [eax+ebx+colorA]
  4.1701 +         movq mm1, [eax+ebx+ebx+colorC]
  4.1702 +
  4.1703 +         movq mm2, mm0
  4.1704 +         movq mm3, mm1
  4.1705 +
  4.1706 +         pand mm0, [colorMask]
  4.1707 +         pand mm1, [colorMask]
  4.1708 +
  4.1709 +         psrlw mm0, 1
  4.1710 +         psrlw mm1, 1
  4.1711 +
  4.1712 +         pand mm3, [lowPixelMask]
  4.1713 +         paddw mm0, mm1
  4.1714 +
  4.1715 +         pand mm3, mm2
  4.1716 +         paddw mm0, mm3                ;mm0 contains the interpolated values
  4.1717 +         ;-------------
  4.1718 +
  4.1719 +         ;assemble the pixels
  4.1720 +         movq mm1, [eax+ebx+colorA]
  4.1721 +         movq mm2, [eax+ebx+ebx+colorC]
  4.1722 +
  4.1723 +         movq mm3, [Mask1]
  4.1724 +         movq mm4, [Mask2]
  4.1725 +
  4.1726 +         pand mm1, mm3
  4.1727 +         pand mm2, mm4
  4.1728 +
  4.1729 +         por mm3, mm4
  4.1730 +         pxor mm7, mm7
  4.1731 +         por mm1, mm2
  4.1732 +
  4.1733 +         pcmpeqw mm3, mm7
  4.1734 +         pand mm0, mm3
  4.1735 +         por mm0, mm1
  4.1736 +         movq [ACPixel], mm0
  4.1737 +
  4.1738 +;////////////////////////////////
  4.1739 +; Decide which "branch" to take
  4.1740 +;--------------------------------
  4.1741 +         movq mm0, [eax+ebx+colorA]
  4.1742 +         movq mm1, [eax+ebx+colorB]
  4.1743 +         movq mm6, mm0
  4.1744 +         movq mm7, mm1
  4.1745 +         pcmpeqw mm0, [eax+ebx+ebx+colorD]
  4.1746 +         pcmpeqw mm1, [eax+ebx+ebx+colorC]
  4.1747 +         pcmpeqw mm6, mm7
  4.1748 +
  4.1749 +         movq mm2, mm0
  4.1750 +         movq mm3, mm0
  4.1751 +
  4.1752 +         pand mm0, mm1       ;colorA == colorD && colorB == colorC
  4.1753 +         pxor mm7, mm7
  4.1754 +
  4.1755 +         pcmpeqw mm2, mm7
  4.1756 +         pand mm6, mm0
  4.1757 +         pand mm2, mm1       ;colorA != colorD && colorB == colorC
  4.1758 +
  4.1759 +         pcmpeqw mm1, mm7
  4.1760 +
  4.1761 +         pand mm1, mm3       ;colorA == colorD && colorB != colorC
  4.1762 +         pxor mm0, mm6
  4.1763 +         por mm1, mm6
  4.1764 +         movq mm7, mm0
  4.1765 +         movq [Mask2], mm2
  4.1766 +         packsswb mm7, mm7
  4.1767 +         movq [Mask1], mm1
  4.1768 +
  4.1769 +         movd ecx, mm7
  4.1770 +         test ecx, ecx
  4.1771 +         jz near .SKIP_GUESS
  4.1772 +
  4.1773 +;---------------------------------------------
  4.1774 +; Map of the pixels:                    I|E F|J
  4.1775 +;                                       G|A B|K
  4.1776 +;                                       H|C D|L
  4.1777 +;                                       M|N O|P
  4.1778 +         movq mm6, mm0
  4.1779 +         movq mm4, [eax+ebx+colorA]
  4.1780 +         movq mm5, [eax+ebx+colorB]
  4.1781 +         pxor mm7, mm7
  4.1782 +         pand mm6, [ONE]
  4.1783 +
  4.1784 +         movq mm0, [eax+colorE]
  4.1785 +         movq mm1, [eax+ebx+colorG]
  4.1786 +         movq mm2, mm0
  4.1787 +         movq mm3, mm1
  4.1788 +         pcmpeqw mm0, mm4
  4.1789 +         pcmpeqw mm1, mm4
  4.1790 +         pcmpeqw mm2, mm5
  4.1791 +         pcmpeqw mm3, mm5
  4.1792 +         pand mm0, mm6
  4.1793 +         pand mm1, mm6
  4.1794 +         pand mm2, mm6
  4.1795 +         pand mm3, mm6
  4.1796 +         paddw mm0, mm1
  4.1797 +         paddw mm2, mm3
  4.1798 +
  4.1799 +         pxor mm3, mm3
  4.1800 +         pcmpgtw mm0, mm6
  4.1801 +         pcmpgtw mm2, mm6
  4.1802 +         pcmpeqw mm0, mm3
  4.1803 +         pcmpeqw mm2, mm3
  4.1804 +         pand mm0, mm6
  4.1805 +         pand mm2, mm6
  4.1806 +         paddw mm7, mm0
  4.1807 +         psubw mm7, mm2
  4.1808 +
  4.1809 +         movq mm0, [eax+colorF]
  4.1810 +         movq mm1, [eax+ebx+colorK]
  4.1811 +         movq mm2, mm0
  4.1812 +         movq mm3, mm1
  4.1813 +         pcmpeqw mm0, mm4
  4.1814 +         pcmpeqw mm1, mm4
  4.1815 +         pcmpeqw mm2, mm5
  4.1816 +         pcmpeqw mm3, mm5
  4.1817 +         pand mm0, mm6
  4.1818 +         pand mm1, mm6
  4.1819 +         pand mm2, mm6
  4.1820 +         pand mm3, mm6
  4.1821 +         paddw mm0, mm1
  4.1822 +         paddw mm2, mm3
  4.1823 +
  4.1824 +         pxor mm3, mm3
  4.1825 +         pcmpgtw mm0, mm6
  4.1826 +         pcmpgtw mm2, mm6
  4.1827 +         pcmpeqw mm0, mm3
  4.1828 +         pcmpeqw mm2, mm3
  4.1829 +         pand mm0, mm6
  4.1830 +         pand mm2, mm6
  4.1831 +         paddw mm7, mm0
  4.1832 +         psubw mm7, mm2
  4.1833 +
  4.1834 +         push eax
  4.1835 +         add eax, ebx
  4.1836 +         movq mm0, [eax+ebx+colorH]
  4.1837 +         movq mm1, [eax+ebx+ebx+colorN]
  4.1838 +         movq mm2, mm0
  4.1839 +         movq mm3, mm1
  4.1840 +         pcmpeqw mm0, mm4
  4.1841 +         pcmpeqw mm1, mm4
  4.1842 +         pcmpeqw mm2, mm5
  4.1843 +         pcmpeqw mm3, mm5
  4.1844 +         pand mm0, mm6
  4.1845 +         pand mm1, mm6
  4.1846 +         pand mm2, mm6
  4.1847 +         pand mm3, mm6
  4.1848 +         paddw mm0, mm1
  4.1849 +         paddw mm2, mm3
  4.1850 +
  4.1851 +         pxor mm3, mm3
  4.1852 +         pcmpgtw mm0, mm6
  4.1853 +         pcmpgtw mm2, mm6
  4.1854 +         pcmpeqw mm0, mm3
  4.1855 +         pcmpeqw mm2, mm3
  4.1856 +         pand mm0, mm6
  4.1857 +         pand mm2, mm6
  4.1858 +         paddw mm7, mm0
  4.1859 +         psubw mm7, mm2
  4.1860 +
  4.1861 +         movq mm0, [eax+ebx+colorL]
  4.1862 +         movq mm1, [eax+ebx+ebx+colorO]
  4.1863 +         movq mm2, mm0
  4.1864 +         movq mm3, mm1
  4.1865 +         pcmpeqw mm0, mm4
  4.1866 +         pcmpeqw mm1, mm4
  4.1867 +         pcmpeqw mm2, mm5
  4.1868 +         pcmpeqw mm3, mm5
  4.1869 +         pand mm0, mm6
  4.1870 +         pand mm1, mm6
  4.1871 +         pand mm2, mm6
  4.1872 +         pand mm3, mm6
  4.1873 +         paddw mm0, mm1
  4.1874 +         paddw mm2, mm3
  4.1875 +
  4.1876 +         pxor mm3, mm3
  4.1877 +         pcmpgtw mm0, mm6
  4.1878 +         pcmpgtw mm2, mm6
  4.1879 +         pcmpeqw mm0, mm3
  4.1880 +         pcmpeqw mm2, mm3
  4.1881 +         pand mm0, mm6
  4.1882 +         pand mm2, mm6
  4.1883 +         paddw mm7, mm0
  4.1884 +         psubw mm7, mm2
  4.1885 +
  4.1886 +         pop eax
  4.1887 +         movq mm1, mm7
  4.1888 +         pxor mm0, mm0
  4.1889 +         pcmpgtw mm7, mm0
  4.1890 +         pcmpgtw mm0, mm1
  4.1891 +
  4.1892 +         por mm7, [Mask1]
  4.1893 +         por mm0, [Mask2]
  4.1894 +         movq [Mask1], mm7
  4.1895 +         movq [Mask2], mm0
  4.1896 +
  4.1897 +.SKIP_GUESS:
  4.1898 +         ;----------------------------
  4.1899 +         ;interpolate A, B, C and D
  4.1900 +         movq mm0, [eax+ebx+colorA]
  4.1901 +         movq mm1, [eax+ebx+colorB]
  4.1902 +         movq mm4, mm0
  4.1903 +         movq mm2, [eax+ebx+ebx+colorC]
  4.1904 +         movq mm5, mm1
  4.1905 +         movq mm3, [qcolorMask]
  4.1906 +         movq mm6, mm2
  4.1907 +         movq mm7, [qlowpixelMask]
  4.1908 +
  4.1909 +         pand mm0, mm3
  4.1910 +         pand mm1, mm3
  4.1911 +         pand mm2, mm3
  4.1912 +         pand mm3, [eax+ebx+ebx+colorD]
  4.1913 +
  4.1914 +         psrlw mm0, 2
  4.1915 +         pand mm4, mm7
  4.1916 +         psrlw mm1, 2
  4.1917 +         pand mm5, mm7
  4.1918 +         psrlw mm2, 2
  4.1919 +         pand mm6, mm7
  4.1920 +         psrlw mm3, 2
  4.1921 +         pand mm7, [eax+ebx+ebx+colorD]
  4.1922 +
  4.1923 +         paddw mm0, mm1
  4.1924 +         paddw mm2, mm3
  4.1925 +
  4.1926 +         paddw mm4, mm5
  4.1927 +         paddw mm6, mm7
  4.1928 +
  4.1929 +         paddw mm4, mm6
  4.1930 +         paddw mm0, mm2
  4.1931 +         psrlw mm4, 2
  4.1932 +         pand mm4, [qlowpixelMask]
  4.1933 +         paddw mm0, mm4      ;mm0 contains the interpolated value of A, B, C and D
  4.1934 +
  4.1935 +;\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
  4.1936 +         ;assemble the pixels
  4.1937 +         movq mm1, [Mask1]
  4.1938 +         movq mm2, [Mask2]
  4.1939 +         movq mm4, [eax+ebx+colorA]
  4.1940 +         movq mm5, [eax+ebx+colorB]
  4.1941 +         pand mm4, mm1
  4.1942 +         pand mm5, mm2
  4.1943 +
  4.1944 +         pxor mm7, mm7
  4.1945 +         por mm1, mm2
  4.1946 +         por mm4, mm5
  4.1947 +         pcmpeqw mm1, mm7
  4.1948 +         pand mm0, mm1
  4.1949 +         por mm4, mm0        ;mm4 contains the diagonal pixels
  4.1950 +
  4.1951 +         movq mm0, [ACPixel]
  4.1952 +         movq mm1, mm0
  4.1953 +         punpcklwd mm0, mm4
  4.1954 +         punpckhwd mm1, mm4
  4.1955 +
  4.1956 +         push edx
  4.1957 +         add edx, [ebp+dstPitch]
  4.1958 +
  4.1959 +%ifdef FAR_POINTER
  4.1960 +         movq [fs:edx], mm0
  4.1961 +         movq [fs:edx+8], mm1
  4.1962 +%else
  4.1963 +         movq [edx], mm0
  4.1964 +         movq [edx+8], mm1
  4.1965 +%endif
  4.1966 +         pop edx
  4.1967 +
  4.1968 +.SKIP_PROCESS:
  4.1969 +         mov ecx, [ebp+deltaPtr]
  4.1970 +         add ecx, 8
  4.1971 +         mov [ebp+deltaPtr], ecx
  4.1972 +         add edx, 16
  4.1973 +         add eax, 8
  4.1974 +
  4.1975 +         pop ecx
  4.1976 +         sub ecx, 4
  4.1977 +         cmp ecx, 0
  4.1978 +         jg  near .Loop
  4.1979 +
  4.1980 +; Restore some stuff
  4.1981 +         popad
  4.1982 +         mov esp, ebp
  4.1983 +         pop ebp
  4.1984 +         emms
  4.1985 +         ret
  4.1986 +
  4.1987 +;-------------------------------------------------------------------------
  4.1988 +;-------------------------------------------------------------------------
  4.1989 +;-------------------------------------------------------------------------
  4.1990 +;-------------------------------------------------------------------------
  4.1991 +;-------------------------------------------------------------------------
  4.1992 +;-------------------------------------------------------------------------
  4.1993 +;-------------------------------------------------------------------------
  4.1994 +
  4.1995 +%ifdef __DJGPP__
  4.1996 +_Init_2xSaIMMX:
  4.1997 +%else
  4.1998 +Init_2xSaIMMX:
  4.1999 +%endif
  4.2000 +; Store some stuff
  4.2001 +         push ebp
  4.2002 +         mov ebp, esp
  4.2003 +         push edx
  4.2004 +
  4.2005 +
  4.2006 +;Damn thing doesn't work
  4.2007 +;        mov eax,1
  4.2008 +;        cpuid
  4.2009 +;        test edx, 0x00800000     ;test bit 23
  4.2010 +;        jz end2 ;bit not set => no MMX detected
  4.2011 +
  4.2012 +         mov eax, [ebp+8]         ;PixelFormat
  4.2013 +         cmp eax, 555
  4.2014 +         jz Bits555
  4.2015 +         cmp eax, 565
  4.2016 +         jz Bits565
  4.2017 +end2:
  4.2018 +         mov eax, 1
  4.2019 +         jmp end3
  4.2020 +Bits555:
  4.2021 +         mov edx, 0x7BDE7BDE
  4.2022 +         mov eax, colorMask
  4.2023 +         mov [eax], edx
  4.2024 +         mov [eax+4], edx
  4.2025 +         mov edx, 0x04210421
  4.2026 +         mov eax, lowPixelMask
  4.2027 +         mov [eax], edx
  4.2028 +         mov [eax+4], edx
  4.2029 +         mov edx, 0x739C739C
  4.2030 +         mov eax, qcolorMask
  4.2031 +         mov [eax], edx
  4.2032 +         mov [eax+4], edx
  4.2033 +         mov edx, 0x0C630C63
  4.2034 +         mov eax, qlowpixelMask
  4.2035 +         mov [eax], edx
  4.2036 +         mov [eax+4], edx
  4.2037 +         mov eax, 0
  4.2038 +         jmp end3
  4.2039 +Bits565:
  4.2040 +         mov edx, 0xF7DEF7DE
  4.2041 +         mov eax, colorMask
  4.2042 +         mov [eax], edx
  4.2043 +         mov [eax+4], edx
  4.2044 +         mov edx, 0x08210821
  4.2045 +         mov eax, lowPixelMask
  4.2046 +         mov [eax], edx
  4.2047 +         mov [eax+4], edx
  4.2048 +         mov edx, 0xE79CE79C
  4.2049 +         mov eax, qcolorMask
  4.2050 +         mov [eax], edx
  4.2051 +         mov [eax+4], edx
  4.2052 +         mov edx, 0x18631863
  4.2053 +         mov eax, qlowpixelMask
  4.2054 +         mov [eax], edx
  4.2055 +         mov [eax+4], edx
  4.2056 +         mov eax, 0
  4.2057 +         jmp end3
  4.2058 +end3:   
  4.2059 +         pop edx
  4.2060 +         mov esp, ebp
  4.2061 +         pop ebp
  4.2062 +         ret
  4.2063 +
  4.2064 +
  4.2065 +;-------------------------------------------------------------------------
  4.2066 +;-------------------------------------------------------------------------
  4.2067 +;-------------------------------------------------------------------------
  4.2068 +;-------------------------------------------------------------------------
  4.2069 +;-------------------------------------------------------------------------
  4.2070 +;-------------------------------------------------------------------------
  4.2071 +;-------------------------------------------------------------------------
  4.2072 +
  4.2073 +        SECTION .data ALIGN = 32
  4.2074 +;Some constants
  4.2075 +colorMask     dd 0xF7DEF7DE,0xF7DEF7DE
  4.2076 +lowPixelMask  dd 0x08210821,0x08210821
  4.2077 +
  4.2078 +qcolorMask    dd 0xE79CE79C,0xE79CE79C
  4.2079 +qlowpixelMask dd 0x18631863,0x18631863
  4.2080 +
  4.2081 +darkenMask    dd 0xC718C718,0xC718C718
  4.2082 +GreenMask     dd 0x07E007E0,0x07E007E0
  4.2083 +RedBlueMask   dd 0xF81FF81F,0xF81FF81F
  4.2084 +
  4.2085 +FALSE         dd 0x00000000,0x00000000
  4.2086 +TRUE          dd 0xffffffff,0xffffffff
  4.2087 +ONE           dd 0x00010001,0x00010001
  4.2088 +
  4.2089 +
  4.2090 +        SECTION .bss ALIGN = 32
  4.2091 +ACPixel       resb 8
  4.2092 +Mask1         resb 8
  4.2093 +Mask2         resb 8
  4.2094 +
  4.2095 +I56Pixel      resb 8
  4.2096 +I23Pixel      resb 8
  4.2097 +I5556Pixel    resb 8
  4.2098 +I2223Pixel    resb 8
  4.2099 +I5666Pixel    resb 8
  4.2100 +I2333Pixel    resb 8
  4.2101 +Mask26        resb 8
  4.2102 +Mask35        resb 8
  4.2103 +Mask26b       resb 8
  4.2104 +Mask35b       resb 8
  4.2105 +product1a     resb 8
  4.2106 +product1b     resb 8
  4.2107 +product2a     resb 8
  4.2108 +product2b     resb 8
  4.2109 +final1a       resb 8
  4.2110 +final1b       resb 8
  4.2111 +final2a       resb 8
  4.2112 +final2b       resb 8

     5.1 --- /dev/null	Thu Jan 01 00:00:00 1970 +0000
     5.2 +++ b/src/filters/Makefile	Sun Mar 04 20:32:31 2012 -0600
     5.3 @@ -0,0 +1,488 @@
     5.4 +# Makefile.in generated by automake 1.10.1 from Makefile.am.
     5.5 +# src/filters/Makefile.  Generated from Makefile.in by configure.
     5.6 +
     5.7 +# Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
     5.8 +# 2003, 2004, 2005, 2006, 2007, 2008  Free Software Foundation, Inc.
     5.9 +# This Makefile.in is free software; the Free Software Foundation
    5.10 +# gives unlimited permission to copy and/or distribute it,
    5.11 +# with or without modifications, as long as this notice is preserved.
    5.12 +
    5.13 +# This program is distributed in the hope that it will be useful,
    5.14 +# but WITHOUT ANY WARRANTY, to the extent permitted by law; without
    5.15 +# even the implied warranty of MERCHANTABILITY or FITNESS FOR A
    5.16 +# PARTICULAR PURPOSE.
    5.17 +
    5.18 +
    5.19 +
    5.20 +
    5.21 +pkgdatadir = $(datadir)/VisualBoyAdvance
    5.22 +pkglibdir = $(libdir)/VisualBoyAdvance
    5.23 +pkgincludedir = $(includedir)/VisualBoyAdvance
    5.24 +am__cd = CDPATH="$${ZSH_VERSION+.}$(PATH_SEPARATOR)" && cd
    5.25 +install_sh_DATA = $(install_sh) -c -m 644
    5.26 +install_sh_PROGRAM = $(install_sh) -c
    5.27 +install_sh_SCRIPT = $(install_sh) -c
    5.28 +INSTALL_HEADER = $(INSTALL_DATA)
    5.29 +transform = $(program_transform_name)
    5.30 +NORMAL_INSTALL = :
    5.31 +PRE_INSTALL = :
    5.32 +POST_INSTALL = :
    5.33 +NORMAL_UNINSTALL = :
    5.34 +PRE_UNINSTALL = :
    5.35 +POST_UNINSTALL = :
    5.36 +build_triplet = x86_64-unknown-linux-gnu
    5.37 +host_triplet = x86_64-unknown-linux-gnu
    5.38 +target_triplet = x86_64-unknown-linux-gnu
    5.39 +subdir = src/filters
    5.40 +DIST_COMMON = $(srcdir)/Makefile.am $(srcdir)/Makefile.in
    5.41 +OBJDIR = $(top_srcdir)/src/obj
    5.42 +ACLOCAL_M4 = $(top_srcdir)/aclocal.m4
    5.43 +am__aclocal_m4_deps = $(top_srcdir)/m4/gettext.m4 \
    5.44 +	$(top_srcdir)/m4/iconv.m4 $(top_srcdir)/m4/lib-ld.m4 \
    5.45 +	$(top_srcdir)/m4/lib-link.m4 $(top_srcdir)/m4/lib-prefix.m4 \
    5.46 +	$(top_srcdir)/m4/nls.m4 $(top_srcdir)/m4/po.m4 \
    5.47 +	$(top_srcdir)/m4/progtest.m4 $(top_srcdir)/acinclude.m4 \
    5.48 +	$(top_srcdir)/configure.in
    5.49 +am__configure_deps = $(am__aclocal_m4_deps) $(CONFIGURE_DEPENDENCIES) \
    5.50 +	$(ACLOCAL_M4)
    5.51 +mkinstalldirs = $(SHELL) $(top_srcdir)/mkinstalldirs
    5.52 +CONFIG_CLEAN_FILES =
    5.53 +LIBRARIES = $(noinst_LIBRARIES)
    5.54 +AR = ar
    5.55 +ARFLAGS = cru
    5.56 +lib386_a_AR = $(AR) $(ARFLAGS)
    5.57 +lib386_a_LIBADD =
    5.58 +am_lib386_a_OBJECTS = 2xSaImmx.$(OBJEXT)
    5.59 +lib386_a_OBJECTS = $(patsubst %,$(OBJDIR)/%,$(am_lib386_a_OBJECTS))
    5.60 +libfilter_a_AR = $(AR) $(ARFLAGS)
    5.61 +libfilter_a_LIBADD =
    5.62 +am_libfilter_a_OBJECTS = 2xSaI.$(OBJEXT) admame.$(OBJEXT) \
    5.63 +	bilinear.$(OBJEXT) hq2x.$(OBJEXT) interframe.$(OBJEXT) \
    5.64 +	motionblur.$(OBJEXT) pixel.$(OBJEXT) scanline.$(OBJEXT) \
    5.65 +	simple2x.$(OBJEXT)
    5.66 +libfilter_a_OBJECTS = $(patsubst %,$(OBJDIR)/%,$(am_libfilter_a_OBJECTS))
    5.67 +DEFAULT_INCLUDES = -I.
    5.68 +depcomp = $(SHELL) $(top_srcdir)/depcomp
    5.69 +am__depfiles_maybe = depfiles
    5.70 +CXXCOMPILE = $(CXX) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \
    5.71 +	$(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CXXFLAGS) $(CXXFLAGS)
    5.72 +CXXLD = $(CXX)
    5.73 +CXXLINK = $(CXXLD) $(AM_CXXFLAGS) $(CXXFLAGS) $(AM_LDFLAGS) $(LDFLAGS) \
    5.74 +	-o $@
    5.75 +COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \
    5.76 +	$(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)
    5.77 +CCLD = $(CC)
    5.78 +LINK = $(CCLD) $(AM_CFLAGS) $(CFLAGS) $(AM_LDFLAGS) $(LDFLAGS) -o $@
    5.79 +SOURCES = $(lib386_a_SOURCES) $(libfilter_a_SOURCES)
    5.80 +DIST_SOURCES = $(lib386_a_SOURCES) $(libfilter_a_SOURCES)
    5.81 +ETAGS = etags
    5.82 +CTAGS = ctags
    5.83 +DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
    5.84 +ACLOCAL = ${SHELL} /home/r/proj/vba/trunk/missing --run aclocal-1.10
    5.85 +AMTAR = ${SHELL} /home/r/proj/vba/trunk/missing --run tar
    5.86 +AUTOCONF = ${SHELL} /home/r/proj/vba/trunk/missing --run autoconf
    5.87 +AUTOHEADER = ${SHELL} /home/r/proj/vba/trunk/missing --run autoheader
    5.88 +AUTOMAKE = ${SHELL} /home/r/proj/vba/trunk/missing --run automake-1.10
    5.89 +AWK = gawk
    5.90 +CC = gcc
    5.91 +CCDEPMODE = depmode=gcc3
    5.92 +CFLAGS = -g -O2
    5.93 +CPP = gcc -E
    5.94 +CPPFLAGS = 
    5.95 +CXX = g++
    5.96 +CXXDEPMODE = depmode=gcc3
    5.97 +CXXFLAGS = -g -O2 -DC_CORE -DPROFILING -DDEV_VERSION
    5.98 +CYGPATH_W = echo
    5.99 +DEFS = -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE_URL=\"\" -DPACKAGE=\"VisualBoyAdvance\" -DVERSION=\"1.7.2\" -DYYTEXT_POINTER=1 -DHAVE_LIBZ=1 -DHAVE_LIBPNG=1 -DHAVE_LIBPTHREAD=1 -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_MALLOC_H=1 -DHAVE_STRINGS_H=1 -DHAVE_UNISTD_H=1 -DHAVE_ARPA_INET_H=1 -DHAVE_NETINET_IN_H=1
   5.100 +DEPDIR = .deps
   5.101 +ECHO_C = 
   5.102 +ECHO_N = -n
   5.103 +ECHO_T = 
   5.104 +EGREP = /bin/grep -E
   5.105 +EXEEXT = 
   5.106 +GETTEXT_PACKAGE = 
   5.107 +GMSGFMT = 
   5.108 +GREP = /bin/grep
   5.109 +GTKMM_CFLAGS = 
   5.110 +GTKMM_CPPFLAGS = 
   5.111 +GTKMM_LIBS = 
   5.112 +INSTALL = /usr/bin/install -c
   5.113 +INSTALL_DATA = ${INSTALL} -m 644
   5.114 +INSTALL_PROGRAM = ${INSTALL}
   5.115 +INSTALL_SCRIPT = ${INSTALL}
   5.116 +INSTALL_STRIP_PROGRAM = $(install_sh) -c -s
   5.117 +INTLLIBS = 
   5.118 +LDFLAGS = 
   5.119 +LEX = flex
   5.120 +LEXLIB = -lfl
   5.121 +LEX_OUTPUT_ROOT = lex.yy
   5.122 +LIBICONV = 
   5.123 +LIBINTL = 
   5.124 +LIBOBJS = 
   5.125 +LIBS = -lpthread -lpng -lz 
   5.126 +LTLIBICONV = 
   5.127 +LTLIBINTL = 
   5.128 +LTLIBOBJS = 
   5.129 +MAKEINFO = ${SHELL} /home/r/proj/vba/trunk/missing --run makeinfo
   5.130 +MKDIR_P = /bin/mkdir -p
   5.131 +MKINSTALLDIRS = 
   5.132 +MSGFMT = 
   5.133 +MSGMERGE = 
   5.134 +NASM = /usr/bin/nasm
   5.135 +OBJEXT = o
   5.136 +PACKAGE = VisualBoyAdvance
   5.137 +PACKAGE_BUGREPORT = 
   5.138 +PACKAGE_NAME = 
   5.139 +PACKAGE_STRING = 
   5.140 +PACKAGE_TARNAME = 
   5.141 +PACKAGE_VERSION = 
   5.142 +PATH_SEPARATOR = :
   5.143 +PKG_CONFIG = 
   5.144 +POSUB = 
   5.145 +RANLIB = ranlib
   5.146 +SDL_CFLAGS = -I/usr/include/SDL -D_GNU_SOURCE=1 -D_REENTRANT
   5.147 +SDL_CONFIG = /usr/bin/sdl-config
   5.148 +SDL_LIBS = -L/usr/lib -Wl,-rpath,/usr/lib -lSDL -lpthread
   5.149 +SET_MAKE = 
   5.150 +SHELL = /bin/sh
   5.151 +STRIP = 
   5.152 +USE_NLS = 
   5.153 +VBA_EXTRA = 
   5.154 +VBA_LIBS = ../gba/libgba.a ../gb/libgb.a ../common/libgbcom.a ../filters/libfilter.a ../lua/libgblua.a ../prof/libprof.a
   5.155 +VBA_SRC_EXTRA =  lua prof sdl
   5.156 +VERSION = 1.7.2
   5.157 +XGETTEXT = 
   5.158 +XMKMF = 
   5.159 +YACC = bison -y
   5.160 +YFLAGS = 
   5.161 +abs_builddir = /home/r/proj/vba/trunk/src/filters
   5.162 +abs_srcdir = /home/r/proj/vba/trunk/src/filters
   5.163 +abs_top_builddir = /home/r/proj/vba/trunk
   5.164 +abs_top_srcdir = /home/r/proj/vba/trunk
   5.165 +ac_ct_CC = gcc
   5.166 +ac_ct_CXX = g++
   5.167 +am__include = include
   5.168 +am__leading_dot = .
   5.169 +am__quote = 
   5.170 +am__tar = ${AMTAR} chof - "$$tardir"
   5.171 +am__untar = ${AMTAR} xf -
   5.172 +bindir = ${exec_prefix}/bin
   5.173 +build = x86_64-unknown-linux-gnu
   5.174 +build_alias = 
   5.175 +build_cpu = x86_64
   5.176 +build_os = linux-gnu
   5.177 +build_vendor = unknown
   5.178 +builddir = .
   5.179 +datadir = ${datarootdir}
   5.180 +datarootdir = ${prefix}/share
   5.181 +docdir = ${datarootdir}/doc/${PACKAGE}
   5.182 +dvidir = ${docdir}
   5.183 +exec_prefix = ${prefix}
   5.184 +host = x86_64-unknown-linux-gnu
   5.185 +host_alias = 
   5.186 +host_cpu = x86_64
   5.187 +host_os = linux-gnu
   5.188 +host_vendor = unknown
   5.189 +htmldir = ${docdir}
   5.190 +includedir = ${prefix}/include
   5.191 +infodir = ${datarootdir}/info
   5.192 +install_sh = $(SHELL) /home/r/proj/vba/trunk/install-sh
   5.193 +libdir = ${exec_prefix}/lib
   5.194 +libexecdir = ${exec_prefix}/libexec
   5.195 +localedir = ${datarootdir}/locale
   5.196 +localstatedir = ${prefix}/var
   5.197 +mandir = ${datarootdir}/man
   5.198 +mkdir_p = /bin/mkdir -p
   5.199 +oldincludedir = /usr/include
   5.200 +pdfdir = ${docdir}
   5.201 +prefix = /usr/local
   5.202 +program_transform_name = s,x,x,
   5.203 +psdir = ${docdir}
   5.204 +sbindir = ${exec_prefix}/sbin
   5.205 +sharedstatedir = ${prefix}/com
   5.206 +srcdir = .
   5.207 +sysconfdir = ${prefix}/etc
   5.208 +target = x86_64-unknown-linux-gnu
   5.209 +target_alias = 
   5.210 +target_cpu = x86_64
   5.211 +target_os = linux-gnu
   5.212 +target_vendor = unknown
   5.213 +top_builddir = ../..
   5.214 +top_srcdir = ../..
   5.215 +SUFFIXES = .asm
   5.216 +noinst_LIBRARIES = lib386.a libfilter.a
   5.217 +lib386_a_SOURCES = 2xSaImmx.asm
   5.218 +libfilter_a_SOURCES = \
   5.219 +	2xSaI.cpp		\
   5.220 +	admame.cpp		\
   5.221 +	bilinear.cpp		\
   5.222 +	hq2x.cpp		\
   5.223 +	hq2x.h			\
   5.224 +	interframe.cpp		\
   5.225 +	interp.h		\
   5.226 +	lq2x.h			\
   5.227 +	motionblur.cpp		\
   5.228 +	pixel.cpp		\
   5.229 +	scanline.cpp		\
   5.230 +	simple2x.cpp
   5.231 +
   5.232 +all: all-am
   5.233 +
   5.234 +.SUFFIXES:
   5.235 +.SUFFIXES: .asm .cpp .o .obj
   5.236 +$(srcdir)/Makefile.in:  $(srcdir)/Makefile.am  $(am__configure_deps)
   5.237 +	@for dep in $?; do \
   5.238 +	  case '$(am__configure_deps)' in \
   5.239 +	    *$$dep*) \
   5.240 +	      cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh \
   5.241 +		&& exit 0; \
   5.242 +	      exit 1;; \
   5.243 +	  esac; \
   5.244 +	done; \
   5.245 +	echo ' cd $(top_srcdir) && $(AUTOMAKE) --gnu  src/filters/Makefile'; \
   5.246 +	cd $(top_srcdir) && \
   5.247 +	  $(AUTOMAKE) --gnu  src/filters/Makefile
   5.248 +.PRECIOUS: Makefile
   5.249 +Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status
   5.250 +	@case '$?' in \
   5.251 +	  *config.status*) \
   5.252 +	    cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh;; \
   5.253 +	  *) \
   5.254 +	    echo ' cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__depfiles_maybe)'; \
   5.255 +	    cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__depfiles_maybe);; \
   5.256 +	esac;
   5.257 +
   5.258 +$(top_builddir)/config.status: $(top_srcdir)/configure $(CONFIG_STATUS_DEPENDENCIES)
   5.259 +	cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
   5.260 +
   5.261 +$(top_srcdir)/configure:  $(am__configure_deps)
   5.262 +	cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
   5.263 +$(ACLOCAL_M4):  $(am__aclocal_m4_deps)
   5.264 +	cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
   5.265 +
   5.266 +clean-noinstLIBRARIES:
   5.267 +	-test -z "$(noinst_LIBRARIES)" || rm -f $(noinst_LIBRARIES)
   5.268 +lib386.a: $(lib386_a_OBJECTS) $(lib386_a_DEPENDENCIES) 
   5.269 +	-rm -f lib386.a
   5.270 +	$(lib386_a_AR) lib386.a $(lib386_a_OBJECTS) $(lib386_a_LIBADD)
   5.271 +	$(RANLIB) lib386.a
   5.272 +libfilter.a: $(libfilter_a_OBJECTS) $(libfilter_a_DEPENDENCIES) 
   5.273 +	-rm -f libfilter.a
   5.274 +	$(libfilter_a_AR) libfilter.a $(libfilter_a_OBJECTS) $(libfilter_a_LIBADD)
   5.275 +	$(RANLIB) libfilter.a
   5.276 +
   5.277 +mostlyclean-compile:
   5.278 +	-rm -f *.$(OBJEXT)
   5.279 +
   5.280 +distclean-compile:
   5.281 +	-rm -f *.tab.c
   5.282 +
   5.283 +include ./$(DEPDIR)/2xSaI.Po
   5.284 +include ./$(DEPDIR)/admame.Po
   5.285 +include ./$(DEPDIR)/bilinear.Po
   5.286 +include ./$(DEPDIR)/hq2x.Po
   5.287 +include ./$(DEPDIR)/interframe.Po
   5.288 +include ./$(DEPDIR)/motionblur.Po
   5.289 +include ./$(DEPDIR)/pixel.Po
   5.290 +include ./$(DEPDIR)/scanline.Po
   5.291 +include ./$(DEPDIR)/simple2x.Po
   5.292 +
   5.293 +$(OBJDIR)/%.o: %.cpp
   5.294 +	$(CXXCOMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ $<
   5.295 +	mv -f $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
   5.296 +#	source='$<' object='$@' libtool=no \
   5.297 +#	DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) \
   5.298 +#	$(CXXCOMPILE) -c -o $@ $<
   5.299 +
   5.300 +.cpp.obj:
   5.301 +	$(CXXCOMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ `$(CYGPATH_W) '$<'`
   5.302 +	mv -f $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
   5.303 +#	source='$<' object='$@' libtool=no \
   5.304 +#	DEPDIR=$(DEPDIR) $(CXXDEPMODE) $(depcomp) \
   5.305 +#	$(CXXCOMPILE) -c -o $@ `$(CYGPATH_W) '$<'`
   5.306 +
   5.307 +ID: $(HEADERS) $(SOURCES) $(LISP) $(TAGS_FILES)
   5.308 +	list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \
   5.309 +	unique=`for i in $$list; do \
   5.310 +	    if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \
   5.311 +	  done | \
   5.312 +	  $(AWK) '{ files[$$0] = 1; nonemtpy = 1; } \
   5.313 +	      END { if (nonempty) { for (i in files) print i; }; }'`; \
   5.314 +	mkid -fID $$unique
   5.315 +tags: TAGS
   5.316 +
   5.317 +TAGS:  $(HEADERS) $(SOURCES)  $(TAGS_DEPENDENCIES) \
   5.318 +		$(TAGS_FILES) $(LISP)
   5.319 +	tags=; \
   5.320 +	here=`pwd`; \
   5.321 +	list='$(SOURCES) $(HEADERS)  $(LISP) $(TAGS_FILES)'; \
   5.322 +	unique=`for i in $$list; do \
   5.323 +	    if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \
   5.324 +	  done | \
   5.325 +	  $(AWK) '{ files[$$0] = 1; nonempty = 1; } \
   5.326 +	      END { if (nonempty) { for (i in files) print i; }; }'`; \
   5.327 +	if test -z "$(ETAGS_ARGS)$$tags$$unique"; then :; else \
   5.328 +	  test -n "$$unique" || unique=$$empty_fix; \
   5.329 +	  $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \
   5.330 +	    $$tags $$unique; \
   5.331 +	fi
   5.332 +ctags: CTAGS
   5.333 +CTAGS:  $(HEADERS) $(SOURCES)  $(TAGS_DEPENDENCIES) \
   5.334 +		$(TAGS_FILES) $(LISP)
   5.335 +	tags=; \
   5.336 +	list='$(SOURCES) $(HEADERS)  $(LISP) $(TAGS_FILES)'; \
   5.337 +	unique=`for i in $$list; do \
   5.338 +	    if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \
   5.339 +	  done | \
   5.340 +	  $(AWK) '{ files[$$0] = 1; nonempty = 1; } \
   5.341 +	      END { if (nonempty) { for (i in files) print i; }; }'`; \
   5.342 +	test -z "$(CTAGS_ARGS)$$tags$$unique" \
   5.343 +	  || $(CTAGS) $(CTAGSFLAGS) $(AM_CTAGSFLAGS) $(CTAGS_ARGS) \
   5.344 +	     $$tags $$unique
   5.345 +
   5.346 +GTAGS:
   5.347 +	here=`$(am__cd) $(top_builddir) && pwd` \
   5.348 +	  && cd $(top_srcdir) \
   5.349 +	  && gtags -i $(GTAGS_ARGS) $$here
   5.350 +
   5.351 +distclean-tags:
   5.352 +	-rm -f TAGS ID GTAGS GRTAGS GSYMS GPATH tags
   5.353 +
   5.354 +distdir: $(DISTFILES)
   5.355 +	@srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \
   5.356 +	topsrcdirstrip=`echo "$(top_srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \
   5.357 +	list='$(DISTFILES)'; \
   5.358 +	  dist_files=`for file in $$list; do echo $$file; done | \
   5.359 +	  sed -e "s|^$$srcdirstrip/||;t" \
   5.360 +	      -e "s|^$$topsrcdirstrip/|$(top_builddir)/|;t"`; \
   5.361 +	case $$dist_files in \
   5.362 +	  */*) $(MKDIR_P) `echo "$$dist_files" | \
   5.363 +			   sed '/\//!d;s|^|$(distdir)/|;s,/[^/]*$$,,' | \
   5.364 +			   sort -u` ;; \
   5.365 +	esac; \
   5.366 +	for file in $$dist_files; do \
   5.367 +	  if test -f $$file || test -d $$file; then d=.; else d=$(srcdir); fi; \
   5.368 +	  if test -d $$d/$$file; then \
   5.369 +	    dir=`echo "/$$file" | sed -e 's,/[^/]*$$,,'`; \
   5.370 +	    if test -d $(srcdir)/$$file && test $$d != $(srcdir); then \
   5.371 +	      cp -pR $(srcdir)/$$file $(distdir)$$dir || exit 1; \
   5.372 +	    fi; \
   5.373 +	    cp -pR $$d/$$file $(distdir)$$dir || exit 1; \
   5.374 +	  else \
   5.375 +	    test -f $(distdir)/$$file \
   5.376 +	    || cp -p $$d/$$file $(distdir)/$$file \
   5.377 +	    || exit 1; \
   5.378 +	  fi; \
   5.379 +	done
   5.380 +check-am: all-am
   5.381 +check: check-am
   5.382 +all-am: Makefile $(lib386_a_OBJECTS) $(libfilter_a_OBJECTS)
   5.383 +installdirs:
   5.384 +install: install-am
   5.385 +install-exec: install-exec-am
   5.386 +install-data: install-data-am
   5.387 +uninstall: uninstall-am
   5.388 +
   5.389 +install-am: all-am
   5.390 +	@$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am
   5.391 +
   5.392 +installcheck: installcheck-am
   5.393 +install-strip:
   5.394 +	$(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \
   5.395 +	  install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \
   5.396 +	  `test -z '$(STRIP)' || \
   5.397 +	    echo "INSTALL_PROGRAM_ENV=STRIPPROG='$(STRIP)'"` install
   5.398 +mostlyclean-generic:
   5.399 +
   5.400 +clean-generic:
   5.401 +
   5.402 +distclean-generic:
   5.403 +	-test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES)
   5.404 +
   5.405 +maintainer-clean-generic:
   5.406 +	@echo "This command is intended for maintainers to use"
   5.407 +	@echo "it deletes files that may require special tools to rebuild."
   5.408 +clean: clean-am
   5.409 +
   5.410 +clean-am: clean-generic clean-noinstLIBRARIES mostlyclean-am
   5.411 +
   5.412 +distclean: distclean-am
   5.413 +	-rm -rf ./$(DEPDIR)
   5.414 +	-rm -f Makefile
   5.415 +distclean-am: clean-am distclean-compile distclean-generic \
   5.416 +	distclean-tags
   5.417 +
   5.418 +dvi: dvi-am
   5.419 +
   5.420 +dvi-am:
   5.421 +
   5.422 +html: html-am
   5.423 +
   5.424 +info: info-am
   5.425 +
   5.426 +info-am:
   5.427 +
   5.428 +install-data-am:
   5.429 +
   5.430 +install-dvi: install-dvi-am
   5.431 +
   5.432 +install-exec-am:
   5.433 +
   5.434 +install-html: install-html-am
   5.435 +
   5.436 +install-info: install-info-am
   5.437 +
   5.438 +install-man:
   5.439 +
   5.440 +install-pdf: install-pdf-am
   5.441 +
   5.442 +install-ps: install-ps-am
   5.443 +
   5.444 +installcheck-am:
   5.445 +
   5.446 +maintainer-clean: maintainer-clean-am
   5.447 +	-rm -rf ./$(DEPDIR)
   5.448 +	-rm -f Makefile
   5.449 +maintainer-clean-am: distclean-am maintainer-clean-generic
   5.450 +
   5.451 +mostlyclean: mostlyclean-am
   5.452 +
   5.453 +mostlyclean-am: mostlyclean-compile mostlyclean-generic
   5.454 +
   5.455 +pdf: pdf-am
   5.456 +
   5.457 +pdf-am:
   5.458 +
   5.459 +ps: ps-am
   5.460 +
   5.461 +ps-am:
   5.462 +
   5.463 +uninstall-am:
   5.464 +
   5.465 +.MAKE: install-am install-strip
   5.466 +
   5.467 +.PHONY: CTAGS GTAGS all all-am check check-am clean clean-generic \
   5.468 +	clean-noinstLIBRARIES ctags distclean distclean-compile \
   5.469 +	distclean-generic distclean-tags distdir dvi dvi-am html \
   5.470 +	html-am info info-am install install-am install-data \
   5.471 +	install-data-am install-dvi install-dvi-am install-exec \
   5.472 +	install-exec-am install-html install-html-am install-info \
   5.473 +	install-info-am install-man install-pdf install-pdf-am \
   5.474 +	install-ps install-ps-am install-strip installcheck \
   5.475 +	installcheck-am installdirs maintainer-clean \
   5.476 +	maintainer-clean-generic mostlyclean mostlyclean-compile \
   5.477 +	mostlyclean-generic pdf pdf-am ps ps-am tags uninstall \
   5.478 +	uninstall-am
   5.479 +
   5.480 +LBITS := $(shell getconf LONG_BIT)
   5.481 +ifeq ($(LBITS),64)
   5.482 +   ELFFLAG = elf64
   5.483 +else
   5.484 +   ELFFLAG = elf
   5.485 +endif
   5.486 +
   5.487 +$(OBJDIR)/%.o: %.asm
   5.488 +	$(NASM) -f $(ELFFLAG) -o $@ $<
   5.489 +# Tell versions [3.59,3.63) of GNU make to not export all variables.
   5.490 +# Otherwise a system limit (for SysV at least) may be exceeded.
   5.491 +.NOEXPORT: