--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/src/java.desktop/share/native/libawt/java2d/loops/LoopMacros.h Tue Sep 12 19:03:39 2017 +0200
@@ -0,0 +1,2276 @@
+/*
+ * Copyright (c) 2000, 2016, Oracle and/or its affiliates. All rights reserved.
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This code is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 only, as
+ * published by the Free Software Foundation. Oracle designates this
+ * particular file as subject to the "Classpath" exception as provided
+ * by Oracle in the LICENSE file that accompanied this code.
+ *
+ * This code is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * version 2 for more details (a copy is included in the LICENSE file that
+ * accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License version
+ * 2 along with this work; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
+ * or visit www.oracle.com if you need additional information or have any
+ * questions.
+ */
+
+#ifndef LoopMacros_h_Included
+#define LoopMacros_h_Included
+
+#include "j2d_md.h"
+
+#include "LineUtils.h"
+
+/*
+ * This file contains macros to aid in defining native graphics
+ * primitive functions.
+ *
+ * A number of useful building block macros are defined, but the
+ * vast majority of primitives are defined completely by a single
+ * macro expansion which uses macro names in the argument list to
+ * choose not only from a small number of strategies but also to
+ * choose macro packages specific to the source and destination
+ * pixel formats - greatly simplifying all aspects of creating
+ * a new loop.
+ *
+ * See the following macros which define entire functions with
+ * just one or two surface names and sometimes a strategy name:
+ * DEFINE_ISOCOPY_BLIT(ANYTYPE)
+ * DEFINE_ISOXOR_BLIT(ANYTYPE)
+ * DEFINE_CONVERT_BLIT(SRC, DST, CONV_METHOD)
+ * DEFINE_CONVERT_BLIT_LUT(SRC, DST, LUT_STRATEGY)
+ * DEFINE_XPAR_CONVERT_BLIT_LUT(SRC, DST, LUT_STRATEGY)
+ * DEFINE_XPAR_BLITBG_LUT(SRC, DST, LUT_STRATEGY)
+ * DEFINE_SOLID_FILLRECT(DST)
+ * DEFINE_SOLID_FILLSPANS(DST)
+ * DEFINE_SOLID_DRAWLINE(DST)
+ *
+ * Many of these loop macros take the name of a SurfaceType as
+ * an argument and use the ANSI CPP token concatenation operator
+ * "##" to reference macro and type definitions that are specific
+ * to that type of surface.
+ *
+ * A description of the various surface specific macro utilities
+ * that are used by these loop macros appears at the end of the
+ * file. The definitions of these surface-specific macros will
+ * usually appear in a header file named after the SurfaceType
+ * name (i.e. IntArgb.h, ByteGray.h, etc.).
+ */
+
+/*
+ * This loop is the standard "while (--height > 0)" loop used by
+ * some of the blits below.
+ */
+#define BlitLoopHeight(SRCTYPE, SRCPTR, SRCBASE, SRCINFO, \
+ DSTTYPE, DSTPTR, DSTBASE, DSTINFO, DSTPREFIX, \
+ HEIGHT, BODY) \
+ do { \
+ SRCTYPE ## DataType *SRCPTR = (SRCTYPE ## DataType *) (SRCBASE); \
+ DSTTYPE ## DataType *DSTPTR = (DSTTYPE ## DataType *) (DSTBASE); \
+ jint srcScan = (SRCINFO)->scanStride; \
+ jint dstScan = (DSTINFO)->scanStride; \
+ Init ## DSTTYPE ## StoreVarsY(DSTPREFIX, DSTINFO); \
+ do { \
+ BODY; \
+ SRCPTR = PtrAddBytes(SRCPTR, srcScan); \
+ DSTPTR = PtrAddBytes(DSTPTR, dstScan); \
+ Next ## DSTTYPE ## StoreVarsY(DSTPREFIX); \
+ } while (--HEIGHT > 0); \
+ } while (0)
+
+/*
+ * This loop is the standard nested "while (--width/height > 0)" loop
+ * used by most of the basic blits below.
+ */
+#define BlitLoopWidthHeight(SRCTYPE, SRCPTR, SRCBASE, SRCINFO, \
+ DSTTYPE, DSTPTR, DSTBASE, DSTINFO, DSTPREFIX, \
+ WIDTH, HEIGHT, BODY) \
+ do { \
+ SRCTYPE ## DataType *SRCPTR = (SRCTYPE ## DataType *) (SRCBASE); \
+ DSTTYPE ## DataType *DSTPTR = (DSTTYPE ## DataType *) (DSTBASE); \
+ jint srcScan = (SRCINFO)->scanStride; \
+ jint dstScan = (DSTINFO)->scanStride; \
+ Init ## DSTTYPE ## StoreVarsY(DSTPREFIX, DSTINFO); \
+ srcScan -= (WIDTH) * SRCTYPE ## PixelStride; \
+ dstScan -= (WIDTH) * DSTTYPE ## PixelStride; \
+ do { \
+ juint w = WIDTH; \
+ Init ## DSTTYPE ## StoreVarsX(DSTPREFIX, DSTINFO); \
+ do { \
+ BODY; \
+ SRCPTR = PtrAddBytes(SRCPTR, SRCTYPE ## PixelStride); \
+ DSTPTR = PtrAddBytes(DSTPTR, DSTTYPE ## PixelStride); \
+ Next ## DSTTYPE ## StoreVarsX(DSTPREFIX); \
+ } while (--w > 0); \
+ SRCPTR = PtrAddBytes(SRCPTR, srcScan); \
+ DSTPTR = PtrAddBytes(DSTPTR, dstScan); \
+ Next ## DSTTYPE ## StoreVarsY(DSTPREFIX); \
+ } while (--HEIGHT > 0); \
+ } while (0)
+
+/*
+ * This loop is the standard nested "while (--width/height > 0)" loop
+ * used by most of the scaled blits below. It calculates the proper
+ * X source variable
+ */
+#define BlitLoopScaleWidthHeight(SRCTYPE, SRCPTR, SRCBASE, SRCINFO, \
+ DSTTYPE, DSTPTR, DSTBASE, DSTINFO, DSTPREFIX, \
+ XVAR, WIDTH, HEIGHT, \
+ SXLOC, SYLOC, SXINC, SYINC, SHIFT, \
+ BODY) \
+ do { \
+ SRCTYPE ## DataType *SRCPTR; \
+ DSTTYPE ## DataType *DSTPTR = (DSTTYPE ## DataType *) (DSTBASE); \
+ jint srcScan = (SRCINFO)->scanStride; \
+ jint dstScan = (DSTINFO)->scanStride; \
+ Init ## DSTTYPE ## StoreVarsY(DSTPREFIX, DSTINFO); \
+ dstScan -= (WIDTH) * DSTTYPE ## PixelStride; \
+ do { \
+ juint w = WIDTH; \
+ jint tmpsxloc = SXLOC; \
+ SRCPTR = PtrAddBytes(SRCBASE, ((SYLOC >> SHIFT) * srcScan)); \
+ Init ## DSTTYPE ## StoreVarsX(DSTPREFIX, DSTINFO); \
+ do { \
+ jint XVAR = (tmpsxloc >> SHIFT); \
+ BODY; \
+ DSTPTR = PtrAddBytes(DSTPTR, DSTTYPE ## PixelStride); \
+ Next ## DSTTYPE ## StoreVarsX(DSTPREFIX); \
+ tmpsxloc += SXINC; \
+ } while (--w > 0); \
+ DSTPTR = PtrAddBytes(DSTPTR, dstScan); \
+ Next ## DSTTYPE ## StoreVarsY(DSTPREFIX); \
+ SYLOC += SYINC; \
+ } while (--HEIGHT > 0); \
+ } while (0)
+
+/*
+ * This loop is a standard horizontal loop iterating with a "relative"
+ * X coordinate (0 <= X < WIDTH) used primarily by the LUT conversion
+ * preprocessing loops below.
+ */
+#define BlitLoopXRel(DSTTYPE, DSTINFO, DSTPREFIX, \
+ XVAR, WIDTH, BODY) \
+ do { \
+ juint XVAR = 0; \
+ Init ## DSTTYPE ## StoreVarsX(DSTPREFIX, DSTINFO); \
+ do { \
+ BODY; \
+ Next ## DSTTYPE ## StoreVarsX(DSTPREFIX); \
+ } while (++XVAR < WIDTH); \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_CONVERT_BLIT
+ * macros. It converts from the source pixel format to the destination
+ * via an intermediate "jint rgb" format.
+ */
+#define ConvertVia1IntRgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ int rgb; \
+ Load ## SRCTYPE ## To1IntRgb(SRCPTR, SRCPREFIX, SXVAR, rgb); \
+ Store ## DSTTYPE ## From1IntRgb(DSTPTR, DSTPREFIX, DXVAR, rgb); \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_CONVERT_BLIT
+ * macros. It converts from the source pixel format to the destination
+ * via an intermediate "jint argb" format.
+ */
+#define ConvertVia1IntArgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ int argb; \
+ Load ## SRCTYPE ## To1IntArgb(SRCPTR, SRCPREFIX, SXVAR, argb); \
+ Store ## DSTTYPE ## From1IntArgb(DSTPTR, DSTPREFIX, DXVAR, argb); \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_CONVERT_BLIT
+ * macros. It converts from the source pixel format to the destination
+ * via an intermediate set of 3 component variables "jint r, g, b".
+ */
+#define ConvertVia3ByteRgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ jint r, g, b; \
+ Load ## SRCTYPE ## To3ByteRgb(SRCPTR, SRCPREFIX, SXVAR, r, g, b); \
+ Store ## DSTTYPE ## From3ByteRgb(DSTPTR, DSTPREFIX, DXVAR, r, g, b); \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_CONVERT_BLIT
+ * macros. It converts from the source pixel format to the destination
+ * via an intermediate set of 4 component variables "jint a, r, g, b".
+ */
+#define ConvertVia4ByteArgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ jint a, r, g, b; \
+ Load ## SRCTYPE ## To4ByteArgb(SRCPTR, SRCPREFIX, SXVAR, a, r, g, b); \
+ Store ## DSTTYPE ## From4ByteArgb(DSTPTR, DSTPREFIX, DXVAR, \
+ a, r, g, b); \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_CONVERT_BLIT
+ * macros. It converts from the source pixel format to the destination
+ * via an intermediate "jint gray" format.
+ */
+#define ConvertVia1ByteGray(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ jint gray; \
+ Load ## SRCTYPE ## To1ByteGray(SRCPTR, SRCPREFIX, SXVAR, gray); \
+ Store ## DSTTYPE ## From1ByteGray(DSTPTR, DSTPREFIX, DXVAR, gray); \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_XPAR_CONVERT_BLIT
+ * macros. It converts from the source pixel format to the destination
+ * via the specified intermediate format while testing for transparent pixels.
+ */
+#define ConvertXparVia1IntRgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ Declare ## SRCTYPE ## Data(XparLoad); \
+ Load ## SRCTYPE ## Data(SRCPTR, SRCPREFIX, SXVAR, XparLoad); \
+ if (! (Is ## SRCTYPE ## DataTransparent(XparLoad))) { \
+ int rgb; \
+ Convert ## SRCTYPE ## DataTo1IntRgb(XparLoad, rgb); \
+ Store ## DSTTYPE ## From1IntRgb(DSTPTR, DSTPREFIX, DXVAR, rgb); \
+ } \
+ } while (0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_XPAR_BLITBG
+ * macros. It converts from the source pixel format to the destination
+ * via the specified intermediate format while substituting the specified
+ * bgcolor for transparent pixels.
+ */
+#define BgCopyXparVia1IntRgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR, BGPIXEL, BGPREFIX) \
+ do { \
+ Declare ## SRCTYPE ## Data(XparLoad); \
+ Load ## SRCTYPE ## Data(SRCPTR, SRCPREFIX, SXVAR, XparLoad); \
+ if (Is ## SRCTYPE ## DataTransparent(XparLoad)) { \
+ Store ## DSTTYPE ## PixelData(DSTPTR, DXVAR, BGPIXEL, BGPREFIX); \
+ } else { \
+ int rgb; \
+ Convert ## SRCTYPE ## DataTo1IntRgb(XparLoad, rgb); \
+ Store ## DSTTYPE ## From1IntRgb(DSTPTR, DSTPREFIX, DXVAR, rgb); \
+ } \
+ } while (0)
+
+/*
+ * This macro determines whether or not the given pixel is considered
+ * "transparent" for XOR purposes. The ARGB pixel is considered
+ * "transparent" if the alpha value is < 0.5.
+ */
+#define IsArgbTransparent(pixel) \
+ (((jint) pixel) >= 0)
+
+/*
+ * This is a "conversion strategy" for use with the DEFINE_XOR_BLIT macro. It
+ * converts the source pixel to an intermediate ARGB value and then converts
+ * the ARGB value to the pixel representation for the destination surface. It
+ * then XORs the srcpixel, xorpixel, and destination pixel together and stores
+ * the result in the destination surface.
+ */
+#define XorVia1IntArgb(SRCPTR, SRCTYPE, SRCPREFIX, \
+ DSTPTR, DSTTYPE, DSTANYTYPE, \
+ XVAR, XORPIXEL, XORPREFIX, \
+ MASK, MASKPREFIX, DSTINFOPTR) \
+ do { \
+ jint srcpixel; \
+ Declare ## DSTANYTYPE ## PixelData(pix) \
+ Load ## SRCTYPE ## To1IntArgb(SRCPTR, SRCPREFIX, XVAR, srcpixel); \
+ \
+ if (IsArgbTransparent(srcpixel)) { \
+ break; \
+ } \
+ \
+ DSTTYPE ## PixelFromArgb(srcpixel, srcpixel, DSTINFOPTR); \
+ \
+ Extract ## DSTANYTYPE ## PixelData(srcpixel, pix); \
+ Xor ## DSTANYTYPE ## PixelData(srcpixel, pix, DSTPTR, XVAR, \
+ XORPIXEL, XORPREFIX, \
+ MASK, MASKPREFIX); \
+ } while (0)
+
+/*
+ * "LUT_STRATEGY" macro sets.
+ *
+ * There are 2 major strategies for dealing with luts and 3
+ * implementations of those strategies.
+ *
+ * The 2 strategies are "PreProcessLut" and "ConvertOnTheFly".
+ *
+ * For the "PreProcessLut" strategy, the raw ARGB lut supplied
+ * by the SD_LOCK_LUT flag is converted at the beginning into a
+ * form that is more suited for storing into the destination
+ * pixel format. The inner loop consists of a series of table
+ * lookups with very little conversion from that intermediate
+ * pixel format.
+ *
+ * For the "ConvertOnTheFly" strategy, the raw ARGB values are
+ * converted on a pixel by pixel basis in the inner loop itself.
+ * This strategy is most useful for formats which tend to use
+ * the ARGB color format as their pixel format also.
+ *
+ * Each of these strategies has 3 implementations which are needed
+ * for the special cases:
+ * - straight conversion (invoked from DEFINE_CONVERT_BLIT_LUT)
+ * - straight conversion with transparency handling (invoked from
+ * DEFINE_XPAR_CONVERT_BLIT_LUT)
+ * - straight conversion with a bgcolor for the transparent pixels
+ * (invoked from DEFINE_XPAR_BLITBG_LUT)
+ */
+
+/***
+ * Start of PreProcessLut strategy macros, CONVERT_BLIT implementation.
+ */
+#define LutSize(TYPE) \
+ (1 << TYPE ## BitsPerPixel)
+
+#define DeclarePreProcessLutLut(SRC, DST, PIXLUT) \
+ DST ## PixelType PIXLUT[LutSize(SRC)];
+
+#define SetupPreProcessLutLut(SRC, DST, PIXLUT, SRCINFO, DSTINFO) \
+ do { \
+ jint *srcLut = (SRCINFO)->lutBase; \
+ juint lutSize = (SRCINFO)->lutSize; \
+ Declare ## DST ## StoreVars(PreLut) \
+ Init ## DST ## StoreVarsY(PreLut, DSTINFO); \
+ if (lutSize >= LutSize(SRC)) { \
+ lutSize = LutSize(SRC); \
+ } else { \
+ DST ## PixelType *pPIXLUT = &PIXLUT[lutSize]; \
+ do { \
+ Store ## DST ## From1IntArgb(pPIXLUT, PreLut, 0, 0); \
+ } while (++pPIXLUT < &PIXLUT[LutSize(SRC)]); \
+ } \
+ BlitLoopXRel(DST, DSTINFO, PreLut, x, lutSize, \
+ do { \
+ jint argb = srcLut[x]; \
+ Store ## DST ## From1IntArgb(PIXLUT, PreLut, x, argb); \
+ } while (0)); \
+ } while (0)
+
+#define BodyPreProcessLutLut(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ DSTPTR[DXVAR] = PIXLUT[SRCPTR[SXVAR]]
+
+/*
+ * End of PreProcessLut/CONVERT_BLIT macros.
+ ***/
+
+/***
+ * Start of ConvertOnTheFly strategy macros, CONVERT_BLIT implementation.
+ */
+#define DeclareConvertOnTheFlyLut(SRC, DST, PIXLUT) \
+ Declare ## SRC ## LoadVars(PIXLUT)
+
+#define SetupConvertOnTheFlyLut(SRC, DST, PIXLUT, SRCINFO, DSTINFO) \
+ Init ## SRC ## LoadVars(PIXLUT, SRCINFO)
+
+#define BodyConvertOnTheFlyLut(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ ConvertVia1IntArgb(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR)
+
+/*
+ * End of ConvertOnTheFly/CONVERT_BLIT macros.
+ ***/
+
+/***
+ * Start of PreProcessLut strategy macros, XPAR_CONVERT_BLIT implementation.
+ */
+#define DeclarePreProcessLutXparLut(SRC, DST, PIXLUT) \
+ jint PIXLUT[LutSize(SRC)];
+
+#define SetupPreProcessLutXparLut(SRC, DST, PIXLUT, SRCINFO, DSTINFO) \
+ do { \
+ jint *srcLut = (SRCINFO)->lutBase; \
+ juint lutSize = (SRCINFO)->lutSize; \
+ Declare ## DST ## StoreVars(PreLut) \
+ Init ## DST ## StoreVarsY(PreLut, DSTINFO); \
+ if (lutSize >= LutSize(SRC)) { \
+ lutSize = LutSize(SRC); \
+ } else { \
+ jint *pPIXLUT = &PIXLUT[lutSize]; \
+ do { \
+ pPIXLUT[0] = DST ## XparLutEntry; \
+ } while (++pPIXLUT < &PIXLUT[LutSize(SRC)]); \
+ } \
+ BlitLoopXRel(DST, DSTINFO, PreLut, x, lutSize, \
+ do { \
+ jint argb = srcLut[x]; \
+ if (argb < 0) { \
+ Store ## DST ## NonXparFromArgb \
+ (PIXLUT, PreLut, x, argb); \
+ } else { \
+ PIXLUT[x] = DST ## XparLutEntry; \
+ } \
+ } while (0)); \
+ } while (0)
+
+#define BodyPreProcessLutXparLut(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ jint pix = PIXLUT[SRCPTR[SXVAR]]; \
+ if (! DSTTYPE ## IsXparLutEntry(pix)) { \
+ DSTPTR[DXVAR] = (DSTTYPE ## PixelType) pix; \
+ } \
+ } while (0)
+
+/*
+ * End of PreProcessLut/XPAR_CONVERT_BLIT macros.
+ ***/
+
+/***
+ * Start of ConvertOnTheFly strategy macros, CONVERT_BLIT implementation.
+ */
+#define DeclareConvertOnTheFlyXparLut(SRC, DST, PIXLUT) \
+ Declare ## SRC ## LoadVars(PIXLUT)
+
+#define SetupConvertOnTheFlyXparLut(SRC, DST, PIXLUT, SRCINFO, DSTINFO) \
+ Init ## SRC ## LoadVars(PIXLUT, SRCINFO)
+
+#define BodyConvertOnTheFlyXparLut(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR) \
+ do { \
+ jint argb; \
+ Load ## SRCTYPE ## To1IntArgb(SRCPTR, PIXLUT, SXVAR, argb); \
+ if (argb < 0) { \
+ Store ## DSTTYPE ## From1IntArgb(DSTPTR, DSTPREFIX, DXVAR, argb); \
+ } \
+ } while (0)
+
+/*
+ * End of ConvertOnTheFly/CONVERT_BLIT macros.
+ ***/
+
+/***
+ * Start of PreProcessLut strategy macros, BLITBG implementation.
+ */
+#define DeclarePreProcessLutBgLut(SRC, DST, PIXLUT) \
+ jint PIXLUT[LutSize(SRC)];
+
+#define SetupPreProcessLutBgLut(SRC, DST, PIXLUT, SRCINFO, DSTINFO, BGPIXEL) \
+ do { \
+ jint *srcLut = (SRCINFO)->lutBase; \
+ juint lutSize = (SRCINFO)->lutSize; \
+ Declare ## DST ## StoreVars(PreLut) \
+ Init ## DST ## StoreVarsY(PreLut, DSTINFO); \
+ if (lutSize >= LutSize(SRC)) { \
+ lutSize = LutSize(SRC); \
+ } else { \
+ jint *pPIXLUT = &PIXLUT[lutSize]; \
+ do { \
+ pPIXLUT[0] = BGPIXEL; \
+ } while (++pPIXLUT < &PIXLUT[LutSize(SRC)]); \
+ } \
+ BlitLoopXRel(DST, DSTINFO, PreLut, x, lutSize, \
+ do { \
+ jint argb = srcLut[x]; \
+ if (argb < 0) { \
+ Store ## DST ## From1IntArgb(PIXLUT, PreLut, \
+ x, argb); \
+ } else { \
+ PIXLUT[x] = BGPIXEL; \
+ } \
+ } while (0)); \
+ } while (0)
+
+#define BodyPreProcessLutBgLut(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR, BGPIXEL) \
+ do { \
+ jint pix = PIXLUT[SRCPTR[SXVAR]]; \
+ Store ## DSTTYPE ## Pixel(DSTPTR, DXVAR, pix); \
+ } while (0)
+
+/*
+ * End of PreProcessLut/BLITBG implementation.
+ ***/
+
+/***
+ * Start of ConvertOnTheFly strategy macros, BLITBG implementation.
+ */
+#define DeclareConvertOnTheFlyBgLut(SRC, DST, PIXLUT) \
+ Declare ## SRC ## LoadVars(PIXLUT) \
+ Declare ## DST ## PixelData(bgpix);
+
+#define SetupConvertOnTheFlyBgLut(SRC, DST, PIXLUT, SRCINFO, DSTINFO, BGPIXEL) \
+ do { \
+ Init ## SRC ## LoadVars(PIXLUT, SRCINFO); \
+ Extract ## DST ## PixelData(BGPIXEL, bgpix); \
+ } while (0)
+
+#define BodyConvertOnTheFlyBgLut(SRCPTR, SRCTYPE, PIXLUT, \
+ DSTPTR, DSTTYPE, DSTPREFIX, \
+ SXVAR, DXVAR, BGPIXEL) \
+ do { \
+ jint argb; \
+ Load ## SRCTYPE ## To1IntArgb(SRCPTR, PIXLUT, SXVAR, argb); \
+ if (argb < 0) { \
+ Store ## DSTTYPE ## From1IntArgb(DSTPTR, DSTPREFIX, DXVAR, argb); \
+ } else { \
+ Store ## DSTTYPE ## PixelData(DSTPTR, DXVAR, BGPIXEL, bgpix); \
+ } \
+ } while (0)
+
+/*
+ * End of ConvertOnTheFly/BLITBG macros.
+ ***/
+
+/*
+ * These macros provide consistent naming conventions for the
+ * various types of native primitive inner loop functions.
+ * The names are mechanically constructed from the SurfaceType names.
+ */
+#define NAME_CONVERT_BLIT(SRC, DST) SRC ## To ## DST ## Convert
+
+#define NAME_SCALE_BLIT(SRC, DST) SRC ## To ## DST ## ScaleConvert
+
+#define NAME_XPAR_CONVERT_BLIT(SRC, DST) SRC ## To ## DST ## XparOver
+
+#define NAME_XPAR_SCALE_BLIT(SRC, DST) SRC ## To ## DST ## ScaleXparOver
+
+#define NAME_XPAR_BLITBG(SRC, DST) SRC ## To ## DST ## XparBgCopy
+
+#define NAME_XOR_BLIT(SRC, DST) SRC ## To ## DST ## XorBlit
+
+#define NAME_ISOCOPY_BLIT(ANYTYPE) ANYTYPE ## IsomorphicCopy
+
+#define NAME_ISOSCALE_BLIT(ANYTYPE) ANYTYPE ## IsomorphicScaleCopy
+
+#define NAME_ISOXOR_BLIT(ANYTYPE) ANYTYPE ## IsomorphicXorCopy
+
+#define NAME_SOLID_FILLRECT(TYPE) TYPE ## SetRect
+
+#define NAME_SOLID_FILLSPANS(TYPE) TYPE ## SetSpans
+
+#define NAME_SOLID_DRAWLINE(TYPE) TYPE ## SetLine
+
+#define NAME_XOR_FILLRECT(TYPE) TYPE ## XorRect
+
+#define NAME_XOR_FILLSPANS(TYPE) TYPE ## XorSpans
+
+#define NAME_XOR_DRAWLINE(TYPE) TYPE ## XorLine
+
+#define NAME_SRC_MASKFILL(TYPE) TYPE ## SrcMaskFill
+
+#define NAME_SRCOVER_MASKFILL(TYPE) TYPE ## SrcOverMaskFill
+
+#define NAME_ALPHA_MASKFILL(TYPE) TYPE ## AlphaMaskFill
+
+#define NAME_SRCOVER_MASKBLIT(SRC, DST) SRC ## To ## DST ## SrcOverMaskBlit
+
+#define NAME_ALPHA_MASKBLIT(SRC, DST) SRC ## To ## DST ## AlphaMaskBlit
+
+#define NAME_SOLID_DRAWGLYPHLIST(TYPE) TYPE ## DrawGlyphList
+
+#define NAME_SOLID_DRAWGLYPHLISTAA(TYPE) TYPE ## DrawGlyphListAA
+
+#define NAME_SOLID_DRAWGLYPHLISTLCD(TYPE) TYPE ## DrawGlyphListLCD
+
+#define NAME_XOR_DRAWGLYPHLIST(TYPE) TYPE ## DrawGlyphListXor
+
+#define NAME_TRANSFORMHELPER(TYPE, MODE) TYPE ## MODE ## TransformHelper
+
+#define NAME_TRANSFORMHELPER_NN(TYPE) NAME_TRANSFORMHELPER(TYPE, NrstNbr)
+#define NAME_TRANSFORMHELPER_BL(TYPE) NAME_TRANSFORMHELPER(TYPE, Bilinear)
+#define NAME_TRANSFORMHELPER_BC(TYPE) NAME_TRANSFORMHELPER(TYPE, Bicubic)
+
+#define NAME_TRANSFORMHELPER_FUNCS(TYPE) TYPE ## TransformHelperFuncs
+
+#define NAME_SOLID_FILLPGRAM(TYPE) TYPE ## SetParallelogram
+#define NAME_SOLID_PGRAM_FUNCS(TYPE) TYPE ## SetParallelogramFuncs
+
+#define NAME_XOR_FILLPGRAM(TYPE) TYPE ## XorParallelogram
+#define NAME_XOR_PGRAM_FUNCS(TYPE) TYPE ## XorParallelogramFuncs
+
+/*
+ * These macros conveniently name and declare the indicated native
+ * primitive loop function for forward referencing.
+ */
+#define DECLARE_CONVERT_BLIT(SRC, DST) \
+ BlitFunc NAME_CONVERT_BLIT(SRC, DST)
+
+#define DECLARE_SCALE_BLIT(SRC, DST) \
+ ScaleBlitFunc NAME_SCALE_BLIT(SRC, DST)
+
+#define DECLARE_XPAR_CONVERT_BLIT(SRC, DST) \
+ BlitFunc NAME_XPAR_CONVERT_BLIT(SRC, DST)
+
+#define DECLARE_XPAR_SCALE_BLIT(SRC, DST) \
+ ScaleBlitFunc NAME_XPAR_SCALE_BLIT(SRC, DST)
+
+#define DECLARE_XPAR_BLITBG(SRC, DST) \
+ BlitBgFunc NAME_XPAR_BLITBG(SRC, DST)
+
+#define DECLARE_XOR_BLIT(SRC, DST) \
+ BlitFunc NAME_XOR_BLIT(SRC, DST)
+
+#define DECLARE_ISOCOPY_BLIT(ANYTYPE) \
+ BlitFunc NAME_ISOCOPY_BLIT(ANYTYPE)
+
+#define DECLARE_ISOSCALE_BLIT(ANYTYPE) \
+ ScaleBlitFunc NAME_ISOSCALE_BLIT(ANYTYPE)
+
+#define DECLARE_ISOXOR_BLIT(ANYTYPE) \
+ BlitFunc NAME_ISOXOR_BLIT(ANYTYPE)
+
+#define DECLARE_SOLID_FILLRECT(TYPE) \
+ FillRectFunc NAME_SOLID_FILLRECT(TYPE)
+
+#define DECLARE_SOLID_FILLSPANS(TYPE) \
+ FillSpansFunc NAME_SOLID_FILLSPANS(TYPE)
+
+#define DECLARE_SOLID_DRAWLINE(TYPE) \
+ DrawLineFunc NAME_SOLID_DRAWLINE(TYPE)
+
+#define DECLARE_XOR_FILLRECT(TYPE) \
+ FillRectFunc NAME_XOR_FILLRECT(TYPE)
+
+#define DECLARE_XOR_FILLSPANS(TYPE) \
+ FillSpansFunc NAME_XOR_FILLSPANS(TYPE)
+
+#define DECLARE_XOR_DRAWLINE(TYPE) \
+ DrawLineFunc NAME_XOR_DRAWLINE(TYPE)
+
+#define DECLARE_ALPHA_MASKFILL(TYPE) \
+ MaskFillFunc NAME_ALPHA_MASKFILL(TYPE)
+
+#define DECLARE_SRC_MASKFILL(TYPE) \
+ MaskFillFunc NAME_SRC_MASKFILL(TYPE)
+
+#define DECLARE_SRCOVER_MASKFILL(TYPE) \
+ MaskFillFunc NAME_SRCOVER_MASKFILL(TYPE)
+
+#define DECLARE_SRCOVER_MASKBLIT(SRC, DST) \
+ MaskBlitFunc NAME_SRCOVER_MASKBLIT(SRC, DST)
+
+#define DECLARE_ALPHA_MASKBLIT(SRC, DST) \
+ MaskBlitFunc NAME_ALPHA_MASKBLIT(SRC, DST)
+
+#define DECLARE_SOLID_DRAWGLYPHLIST(TYPE) \
+ DrawGlyphListFunc NAME_SOLID_DRAWGLYPHLIST(TYPE)
+
+#define DECLARE_SOLID_DRAWGLYPHLISTAA(TYPE) \
+ DrawGlyphListAAFunc NAME_SOLID_DRAWGLYPHLISTAA(TYPE)
+
+#define DECLARE_SOLID_DRAWGLYPHLISTLCD(TYPE) \
+ DrawGlyphListLCDFunc NAME_SOLID_DRAWGLYPHLISTLCD(TYPE)
+
+#define DECLARE_XOR_DRAWGLYPHLIST(TYPE) \
+ DrawGlyphListFunc NAME_XOR_DRAWGLYPHLIST(TYPE)
+
+#define DECLARE_TRANSFORMHELPER_FUNCS(TYPE) \
+ TransformHelperFunc NAME_TRANSFORMHELPER_NN(TYPE); \
+ TransformHelperFunc NAME_TRANSFORMHELPER_BL(TYPE); \
+ TransformHelperFunc NAME_TRANSFORMHELPER_BC(TYPE); \
+ TransformHelperFuncs NAME_TRANSFORMHELPER_FUNCS(TYPE)
+
+#define DECLARE_SOLID_PARALLELOGRAM(TYPE) \
+ FillParallelogramFunc NAME_SOLID_FILLPGRAM(TYPE); \
+ DECLARE_SOLID_DRAWLINE(TYPE); \
+ DrawParallelogramFuncs NAME_SOLID_PGRAM_FUNCS(TYPE)
+
+#define DECLARE_XOR_PARALLELOGRAM(TYPE) \
+ FillParallelogramFunc NAME_XOR_FILLPGRAM(TYPE); \
+ DECLARE_XOR_DRAWLINE(TYPE); \
+ DrawParallelogramFuncs NAME_XOR_PGRAM_FUNCS(TYPE)
+
+/*
+ * These macros construct the necessary NativePrimitive structure
+ * for the indicated native primitive loop function which will be
+ * declared somewhere prior and defined elsewhere (usually after).
+ */
+#define REGISTER_CONVERT_BLIT(SRC, DST) \
+ REGISTER_BLIT(SRC, SrcNoEa, DST, NAME_CONVERT_BLIT(SRC, DST))
+
+#define REGISTER_CONVERT_BLIT_FLAGS(SRC, DST, SFLAGS, DFLAGS) \
+ REGISTER_BLIT_FLAGS(SRC, SrcNoEa, DST, NAME_CONVERT_BLIT(SRC, DST), \
+ SFLAGS, DFLAGS)
+
+#define REGISTER_CONVERT_BLIT_EQUIV(SRC, DST, FUNC) \
+ REGISTER_BLIT(SRC, SrcNoEa, DST, FUNC)
+
+#define REGISTER_SCALE_BLIT(SRC, DST) \
+ REGISTER_SCALEBLIT(SRC, SrcNoEa, DST, NAME_SCALE_BLIT(SRC, DST))
+
+#define REGISTER_SCALE_BLIT_FLAGS(SRC, DST, SFLAGS, DFLAGS) \
+ REGISTER_SCALEBLIT_FLAGS(SRC, SrcNoEa, DST, NAME_SCALE_BLIT(SRC, DST), \
+ SFLAGS, DFLAGS)
+
+#define REGISTER_SCALE_BLIT_EQUIV(SRC, DST, FUNC) \
+ REGISTER_SCALEBLIT(SRC, SrcNoEa, DST, FUNC)
+
+#define REGISTER_XPAR_CONVERT_BLIT(SRC, DST) \
+ REGISTER_BLIT(SRC, SrcOverBmNoEa, DST, NAME_XPAR_CONVERT_BLIT(SRC, DST))
+
+#define REGISTER_XPAR_CONVERT_BLIT_EQUIV(SRC, DST, FUNC) \
+ REGISTER_BLIT(SRC, SrcOverBmNoEa, DST, FUNC)
+
+#define REGISTER_XPAR_SCALE_BLIT(SRC, DST) \
+ REGISTER_SCALEBLIT(SRC, SrcOverBmNoEa, DST, NAME_XPAR_SCALE_BLIT(SRC, DST))
+
+#define REGISTER_XPAR_SCALE_BLIT_EQUIV(SRC, DST, FUNC) \
+ REGISTER_SCALEBLIT(SRC, SrcOverBmNoEa, DST, FUNC)
+
+#define REGISTER_XPAR_BLITBG(SRC, DST) \
+ REGISTER_BLITBG(SRC, SrcNoEa, DST, NAME_XPAR_BLITBG(SRC, DST))
+
+#define REGISTER_XPAR_BLITBG_EQUIV(SRC, DST, FUNC) \
+ REGISTER_BLITBG(SRC, SrcNoEa, DST, FUNC)
+
+#define REGISTER_XOR_BLIT(SRC, DST) \
+ REGISTER_BLIT(SRC, Xor, DST, NAME_XOR_BLIT(SRC, DST))
+
+#define REGISTER_ISOCOPY_BLIT(THISTYPE, ANYTYPE) \
+ REGISTER_BLIT(THISTYPE, SrcNoEa, THISTYPE, NAME_ISOCOPY_BLIT(ANYTYPE))
+
+#define REGISTER_ISOSCALE_BLIT(THISTYPE, ANYTYPE) \
+ REGISTER_SCALEBLIT(THISTYPE, SrcNoEa, THISTYPE, NAME_ISOSCALE_BLIT(ANYTYPE))
+
+#define REGISTER_ISOXOR_BLIT(THISTYPE, ANYTYPE) \
+ REGISTER_BLIT(THISTYPE, Xor, THISTYPE, NAME_ISOXOR_BLIT(ANYTYPE))
+
+#define REGISTER_SOLID_FILLRECT(TYPE) \
+ REGISTER_FILLRECT(AnyColor, SrcNoEa, TYPE, NAME_SOLID_FILLRECT(TYPE))
+
+#define REGISTER_SOLID_FILLSPANS(TYPE) \
+ REGISTER_FILLSPANS(AnyColor, SrcNoEa, TYPE, NAME_SOLID_FILLSPANS(TYPE))
+
+#define REGISTER_SOLID_LINE_PRIMITIVES(TYPE) \
+ REGISTER_LINE_PRIMITIVES(AnyColor, SrcNoEa, TYPE, \
+ NAME_SOLID_DRAWLINE(TYPE))
+
+#define REGISTER_XOR_FILLRECT(TYPE) \
+ REGISTER_FILLRECT(AnyColor, Xor, TYPE, NAME_XOR_FILLRECT(TYPE))
+
+#define REGISTER_XOR_FILLSPANS(TYPE) \
+ REGISTER_FILLSPANS(AnyColor, Xor, TYPE, NAME_XOR_FILLSPANS(TYPE))
+
+#define REGISTER_XOR_LINE_PRIMITIVES(TYPE) \
+ REGISTER_LINE_PRIMITIVES(AnyColor, Xor, TYPE, NAME_XOR_DRAWLINE(TYPE))
+
+#define REGISTER_ALPHA_MASKFILL(TYPE) \
+ REGISTER_MASKFILL(AnyColor, AnyAlpha, TYPE, NAME_ALPHA_MASKFILL(TYPE))
+
+#define REGISTER_SRC_MASKFILL(TYPE) \
+ REGISTER_MASKFILL(AnyColor, Src, TYPE, NAME_SRC_MASKFILL(TYPE))
+
+#define REGISTER_SRCOVER_MASKFILL(TYPE) \
+ REGISTER_MASKFILL(AnyColor, SrcOver, TYPE, NAME_SRCOVER_MASKFILL(TYPE))
+
+#define REGISTER_SRCOVER_MASKBLIT(SRC, DST) \
+ REGISTER_MASKBLIT(SRC, SrcOver, DST, NAME_SRCOVER_MASKBLIT(SRC, DST))
+
+#define REGISTER_ALPHA_MASKBLIT(SRC, DST) \
+ REGISTER_MASKBLIT(SRC, AnyAlpha, DST, NAME_ALPHA_MASKBLIT(SRC, DST))
+
+#define REGISTER_SOLID_DRAWGLYPHLIST(TYPE) \
+ REGISTER_DRAWGLYPHLIST(AnyColor, SrcNoEa, TYPE, \
+ NAME_SOLID_DRAWGLYPHLIST(TYPE))
+
+#define REGISTER_SOLID_DRAWGLYPHLISTAA(TYPE) \
+ REGISTER_DRAWGLYPHLISTAA(AnyColor, SrcNoEa, TYPE, \
+ NAME_SOLID_DRAWGLYPHLISTAA(TYPE))
+
+#define REGISTER_SOLID_DRAWGLYPHLISTLCD(TYPE) \
+ REGISTER_DRAWGLYPHLISTLCD(AnyColor, SrcNoEa, TYPE, \
+ NAME_SOLID_DRAWGLYPHLISTLCD(TYPE))
+
+#define REGISTER_XOR_DRAWGLYPHLIST(TYPE) \
+ REGISTER_DRAWGLYPHLIST(AnyColor, Xor, TYPE, \
+ NAME_XOR_DRAWGLYPHLIST(TYPE)), \
+ REGISTER_DRAWGLYPHLISTAA(AnyColor, Xor, TYPE, \
+ NAME_XOR_DRAWGLYPHLIST(TYPE))
+
+#define REGISTER_TRANSFORMHELPER_FUNCS(TYPE) \
+ REGISTER_PRIMITIVE(TransformHelper, TYPE, SrcNoEa, IntArgbPre, \
+ (AnyFunc *) &NAME_TRANSFORMHELPER_FUNCS(TYPE))
+
+#define REGISTER_SOLID_PARALLELOGRAM(TYPE) \
+ REGISTER_PRIMITIVE(FillParallelogram, AnyColor, SrcNoEa, TYPE, \
+ NAME_SOLID_FILLPGRAM(TYPE)), \
+ REGISTER_PRIMITIVE(DrawParallelogram, AnyColor, SrcNoEa, TYPE, \
+ (AnyFunc *) &NAME_SOLID_PGRAM_FUNCS(TYPE))
+
+#define REGISTER_XOR_PARALLELOGRAM(TYPE) \
+ REGISTER_PRIMITIVE(FillParallelogram, AnyColor, Xor, TYPE, \
+ NAME_XOR_FILLPGRAM(TYPE)), \
+ REGISTER_PRIMITIVE(DrawParallelogram, AnyColor, Xor, TYPE, \
+ (AnyFunc *) &NAME_XOR_PGRAM_FUNCS(TYPE))
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for copying pixels of a common type from one buffer to another.
+ */
+#define DEFINE_ISOCOPY_BLIT(ANYTYPE) \
+void NAME_ISOCOPY_BLIT(ANYTYPE)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## ANYTYPE ## StoreVars(DstWrite) \
+ BlitLoopHeight(ANYTYPE, pSrc, srcBase, pSrcInfo, \
+ ANYTYPE, pDst, dstBase, pDstInfo, DstWrite, \
+ height, \
+ memcpy(pDst, pSrc, width * ANYTYPE ## PixelStride)); \
+}
+
+/*
+ * This macro defines an entire function to implement a ScaleBlit inner loop
+ * for scaling pixels of a common type from one buffer to another.
+ */
+#define DEFINE_ISOSCALE_BLIT(ANYTYPE) \
+void NAME_ISOSCALE_BLIT(ANYTYPE)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint sxloc, jint syloc, \
+ jint sxinc, jint syinc, jint shift, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## ANYTYPE ## StoreVars(DstWrite) \
+ BlitLoopScaleWidthHeight(ANYTYPE, pSrc, srcBase, pSrcInfo, \
+ ANYTYPE, pDst, dstBase, pDstInfo, DstWrite, \
+ x, width, height, \
+ sxloc, syloc, sxinc, syinc, shift, \
+ Copy ## ANYTYPE ## PixelData(pSrc, x, pDst, 0)); \
+}
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for XORing pixels of a common type from one buffer into another.
+ */
+#define DEFINE_ISOXOR_BLIT(ANYTYPE) \
+void NAME_ISOXOR_BLIT(ANYTYPE)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint xorpixel = pCompInfo->details.xorPixel; \
+ Declare ## ANYTYPE ## PixelData(xor) \
+ Declare ## ANYTYPE ## StoreVars(DstWrite) \
+ \
+ Extract ## ANYTYPE ## PixelData(xorpixel, xor); \
+ \
+ BlitLoopWidthHeight(ANYTYPE, pSrc, srcBase, pSrcInfo, \
+ ANYTYPE, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ XorCopy ## ANYTYPE ## PixelData(pSrc, pDst, 0, \
+ xorpixel, xor)); \
+}
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for converting pixels from a buffer of one type into a buffer of
+ * another type. No blending is done of the pixels.
+ */
+#define DEFINE_CONVERT_BLIT(SRC, DST, STRATEGY) \
+void NAME_CONVERT_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ Declare ## DST ## StoreVars(DstWrite) \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ ConvertVia ## STRATEGY(pSrc, SRC, SrcRead, \
+ pDst, DST, DstWrite, \
+ 0, 0)); \
+}
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for converting pixels from a buffer of byte pixels with a lookup
+ * table into a buffer of another type. No blending is done of the pixels.
+ */
+#define DEFINE_CONVERT_BLIT_LUT(SRC, DST, LUT_STRATEGY) \
+void NAME_CONVERT_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## StoreVars(DstWrite) \
+ Declare ## LUT_STRATEGY ## Lut(SRC, DST, pixLut) \
+ \
+ Setup ## LUT_STRATEGY ## Lut(SRC, DST, pixLut,\
+ pSrcInfo, pDstInfo); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ Body ## LUT_STRATEGY ## Lut(pSrc, SRC, \
+ pixLut, \
+ pDst, DST, \
+ DstWrite, 0, 0));\
+}
+#define DEFINE_CONVERT_BLIT_LUT8(SRC, DST, LUT_STRATEGY) \
+ DEFINE_CONVERT_BLIT_LUT(SRC, DST, LUT_STRATEGY)
+
+/*
+ * This macro defines an entire function to implement a ScaleBlit inner
+ * loop for scaling and converting pixels from a buffer of one type into
+ * a buffer of another type. No blending is done of the pixels.
+ */
+#define DEFINE_SCALE_BLIT(SRC, DST, STRATEGY) \
+void NAME_SCALE_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint sxloc, jint syloc, \
+ jint sxinc, jint syinc, jint shift, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ Declare ## DST ## StoreVars(DstWrite) \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ BlitLoopScaleWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ x, width, height, \
+ sxloc, syloc, sxinc, syinc, shift, \
+ ConvertVia ## STRATEGY(pSrc, SRC, SrcRead, \
+ pDst, DST, DstWrite, \
+ x, 0)); \
+}
+
+/*
+ * This macro defines an entire function to implement a ScaleBlit inner
+ * loop for scaling and converting pixels from a buffer of byte pixels
+ * with a lookup table into a buffer of another type. No blending is
+ * done of the pixels.
+ */
+#define DEFINE_SCALE_BLIT_LUT(SRC, DST, LUT_STRATEGY) \
+void NAME_SCALE_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint sxloc, jint syloc, \
+ jint sxinc, jint syinc, jint shift, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## StoreVars(DstWrite) \
+ Declare ## LUT_STRATEGY ## Lut(SRC, DST, pixLut) \
+ \
+ Setup ## LUT_STRATEGY ## Lut(SRC, DST, pixLut, pSrcInfo, pDstInfo); \
+ BlitLoopScaleWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ x, width, height, \
+ sxloc, syloc, sxinc, syinc, shift, \
+ Body ## LUT_STRATEGY ## Lut(pSrc, SRC, pixLut, \
+ pDst, DST, \
+ DstWrite, x, 0));\
+}
+#define DEFINE_SCALE_BLIT_LUT8(SRC, DST, LUT_STRATEGY) \
+ DEFINE_SCALE_BLIT_LUT(SRC, DST, LUT_STRATEGY)
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for drawing opaque pixels from a buffer of one type onto a buffer of
+ * another type, ignoring the transparent pixels in the source buffer.
+ * No blending is done of the pixels - the converted pixel value is
+ * either copied or the destination is left untouched.
+ */
+#define DEFINE_XPAR_CONVERT_BLIT(SRC, DST, STRATEGY) \
+void NAME_XPAR_CONVERT_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ Declare ## DST ## StoreVars(DstWrite) \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ ConvertXparVia ## STRATEGY(pSrc, SRC, SrcRead, \
+ pDst, DST, DstWrite, \
+ 0, 0)); \
+}
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for converting pixels from a buffer of byte pixels with a lookup
+ * table containing transparent pixels into a buffer of another type.
+ * No blending is done of the pixels - the converted pixel value is
+ * either copied or the destination is left untouched.
+ */
+#define DEFINE_XPAR_CONVERT_BLIT_LUT(SRC, DST, LUT_STRATEGY) \
+void NAME_XPAR_CONVERT_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## StoreVars(DstWrite) \
+ Declare ## LUT_STRATEGY ## XparLut(SRC, DST, pixLut) \
+ \
+ Setup ## LUT_STRATEGY ## XparLut(SRC, DST, pixLut, pSrcInfo, pDstInfo); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ Body ## LUT_STRATEGY ## XparLut(pSrc, SRC, pixLut, \
+ pDst, DST, \
+ DstWrite, 0, 0)); \
+}
+#define DEFINE_XPAR_CONVERT_BLIT_LUT8(SRC, DST, LUT_STRATEGY) \
+ DEFINE_XPAR_CONVERT_BLIT_LUT(SRC, DST, LUT_STRATEGY)
+
+/*
+ * This macro defines an entire function to implement a ScaleBlit inner
+ * loop for scaling and converting pixels from a buffer of byte pixels
+ * with a lookup table containing transparent pixels into a buffer of
+ * another type.
+ * No blending is done of the pixels - the converted pixel value is
+ * either copied or the destination is left untouched.
+ */
+#define DEFINE_XPAR_SCALE_BLIT_LUT(SRC, DST, LUT_STRATEGY) \
+void NAME_XPAR_SCALE_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint sxloc, jint syloc, \
+ jint sxinc, jint syinc, jint shift, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## StoreVars(DstWrite) \
+ Declare ## LUT_STRATEGY ## XparLut(SRC, DST, pixLut) \
+ \
+ Setup ## LUT_STRATEGY ## XparLut(SRC, DST, pixLut, pSrcInfo, pDstInfo); \
+ BlitLoopScaleWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ x, width, height, \
+ sxloc, syloc, sxinc, syinc, shift, \
+ Body ## LUT_STRATEGY ## XparLut(pSrc, SRC, pixLut, \
+ pDst, DST, \
+ DstWrite, \
+ x, 0)); \
+}
+#define DEFINE_XPAR_SCALE_BLIT_LUT8(SRC, DST, LUT_STRATEGY) \
+ DEFINE_XPAR_SCALE_BLIT_LUT(SRC, DST, LUT_STRATEGY)
+
+/*
+ * This macro defines an entire function to implement a ScaleBlit inner
+ * loop for scaling and converting pixels from a buffer of one type
+ * containing transparent pixels into a buffer of another type.
+ *
+ * No blending is done of the pixels - the converted pixel value is
+ * either copied or the destination is left untouched.
+ */
+#define DEFINE_XPAR_SCALE_BLIT(SRC, DST, STRATEGY) \
+void NAME_XPAR_SCALE_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint sxloc, jint syloc, \
+ jint sxinc, jint syinc, jint shift, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ Declare ## DST ## StoreVars(DstWrite) \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ BlitLoopScaleWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ x, width, height, \
+ sxloc, syloc, sxinc, syinc, shift, \
+ ConvertXparVia ## STRATEGY(pSrc, SRC, SrcRead, \
+ pDst, DST, DstWrite, \
+ x, 0)); \
+}
+
+/*
+ * This macro defines an entire function to implement a BlitBg inner loop
+ * for converting pixels from a buffer of one type containing transparent
+ * pixels into a buffer of another type with a specified bgcolor for the
+ * transparent pixels.
+ * No blending is done of the pixels other than to substitute the
+ * bgcolor for any transparent pixels.
+ */
+#define DEFINE_XPAR_BLITBG(SRC, DST, STRATEGY) \
+void NAME_XPAR_BLITBG(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint bgpixel, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ Declare ## DST ## StoreVars(DstWrite) \
+ Declare ## DST ## PixelData(bgdata) \
+ \
+ Extract ## DST ## PixelData(bgpixel, bgdata); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ BgCopyXparVia ## STRATEGY(pSrc, SRC, SrcRead, \
+ pDst, DST, DstWrite, \
+ 0, 0, bgpixel, bgdata)); \
+}
+
+/*
+ * This macro defines an entire function to implement a BlitBg inner loop
+ * for converting pixels from a buffer of byte pixels with a lookup
+ * table containing transparent pixels into a buffer of another type
+ * with a specified bgcolor for the transparent pixels.
+ * No blending is done of the pixels other than to substitute the
+ * bgcolor for any transparent pixels.
+ */
+#define DEFINE_XPAR_BLITBG_LUT(SRC, DST, LUT_STRATEGY) \
+void NAME_XPAR_BLITBG(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ jint bgpixel, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## StoreVars(DstWrite) \
+ Declare ## LUT_STRATEGY ## BgLut(SRC, DST, pixLut) \
+ \
+ Setup ## LUT_STRATEGY ## BgLut(SRC, DST, pixLut, pSrcInfo, pDstInfo, \
+ bgpixel); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ Body ## LUT_STRATEGY ## BgLut(pSrc, SRC, pixLut, \
+ pDst, DST, \
+ DstWrite, 0, 0, \
+ bgpixel)); \
+}
+#define DEFINE_XPAR_BLITBG_LUT8(SRC, DST, LUT_STRATEGY) \
+ DEFINE_XPAR_BLITBG_LUT(SRC, DST, LUT_STRATEGY)
+
+/*
+ * This macro defines an entire function to implement a Blit inner loop
+ * for converting pixels from a buffer of one type into a buffer of
+ * another type. Each source pixel is XORed with the current XOR color value.
+ * That result is then XORed with the destination pixel and the final
+ * result is stored in the destination surface.
+ */
+#define DEFINE_XOR_BLIT(SRC, DST, DSTANYTYPE) \
+void NAME_XOR_BLIT(SRC, DST)(void *srcBase, void *dstBase, \
+ juint width, juint height, \
+ SurfaceDataRasInfo *pSrcInfo, \
+ SurfaceDataRasInfo *pDstInfo, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint xorpixel = pCompInfo->details.xorPixel; \
+ juint alphamask = pCompInfo->alphaMask; \
+ Declare ## DSTANYTYPE ## PixelData(xor) \
+ Declare ## DSTANYTYPE ## PixelData(mask) \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ Declare ## DST ## StoreVars(DstWrite) \
+ \
+ Extract ## DSTANYTYPE ## PixelData(xorpixel, xor); \
+ Extract ## DSTANYTYPE ## PixelData(alphamask, mask); \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ BlitLoopWidthHeight(SRC, pSrc, srcBase, pSrcInfo, \
+ DST, pDst, dstBase, pDstInfo, DstWrite, \
+ width, height, \
+ XorVia1IntArgb(pSrc, SRC, SrcRead, \
+ pDst, DST, DSTANYTYPE, \
+ 0, xorpixel, xor, \
+ alphamask, mask, pDstInfo)); \
+}
+
+/*
+ * This macro defines an entire function to implement a FillRect inner loop
+ * for setting a rectangular region of pixels to a specific pixel value.
+ * No blending of the fill color is done with the pixels.
+ */
+#define DEFINE_SOLID_FILLRECT(DST) \
+void NAME_SOLID_FILLRECT(DST)(SurfaceDataRasInfo *pRasInfo, \
+ jint lox, jint loy, \
+ jint hix, jint hiy, \
+ jint pixel, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## PixelData(pix) \
+ DST ## DataType *pPix; \
+ jint scan = pRasInfo->scanStride; \
+ juint height = hiy - loy; \
+ juint width = hix - lox; \
+ \
+ pPix = PtrCoord(pRasInfo->rasBase, lox, DST ## PixelStride, loy, scan); \
+ Extract ## DST ## PixelData(pixel, pix); \
+ do { \
+ juint x = 0; \
+ do { \
+ Store ## DST ## PixelData(pPix, x, pixel, pix); \
+ } while (++x < width); \
+ pPix = PtrAddBytes(pPix, scan); \
+ } while (--height > 0); \
+}
+
+/*
+ * This macro defines an entire function to implement a FillSpans inner loop
+ * for iterating through a list of spans and setting those regions of pixels
+ * to a specific pixel value. No blending of the fill color is done with
+ * the pixels.
+ */
+#define DEFINE_SOLID_FILLSPANS(DST) \
+void NAME_SOLID_FILLSPANS(DST)(SurfaceDataRasInfo *pRasInfo, \
+ SpanIteratorFuncs *pSpanFuncs, void *siData, \
+ jint pixel, NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ void *pBase = pRasInfo->rasBase; \
+ Declare ## DST ## PixelData(pix) \
+ jint scan = pRasInfo->scanStride; \
+ jint bbox[4]; \
+ \
+ Extract ## DST ## PixelData(pixel, pix); \
+ while ((*pSpanFuncs->nextSpan)(siData, bbox)) { \
+ jint x = bbox[0]; \
+ jint y = bbox[1]; \
+ juint w = bbox[2] - x; \
+ juint h = bbox[3] - y; \
+ DST ## DataType *pPix = PtrCoord(pBase, \
+ x, DST ## PixelStride, \
+ y, scan); \
+ do { \
+ juint relx; \
+ for (relx = 0; relx < w; relx++) { \
+ Store ## DST ## PixelData(pPix, relx, pixel, pix); \
+ } \
+ pPix = PtrAddBytes(pPix, scan); \
+ } while (--h > 0); \
+ } \
+}
+
+/*
+ * This macro defines an entire function to implement a FillParallelogram
+ * inner loop for tracing 2 diagonal edges (left and right) and setting
+ * those regions of pixels between them to a specific pixel value.
+ * No blending of the fill color is done with the pixels.
+ */
+#define DEFINE_SOLID_FILLPGRAM(DST) \
+void NAME_SOLID_FILLPGRAM(DST)(SurfaceDataRasInfo *pRasInfo, \
+ jint lox, jint loy, jint hix, jint hiy, \
+ jlong leftx, jlong dleftx, \
+ jlong rightx, jlong drightx, \
+ jint pixel, struct _NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## PixelData(pix) \
+ jint scan = pRasInfo->scanStride; \
+ DST ## DataType *pPix = PtrCoord(pRasInfo->rasBase, 0, 0, loy, scan); \
+ \
+ Extract ## DST ## PixelData(pixel, pix); \
+ while (loy < hiy) { \
+ jint lx = WholeOfLong(leftx); \
+ jint rx = WholeOfLong(rightx); \
+ if (lx < lox) lx = lox; \
+ if (rx > hix) rx = hix; \
+ while (lx < rx) { \
+ Store ## DST ## PixelData(pPix, lx, pixel, pix); \
+ lx++; \
+ } \
+ pPix = PtrAddBytes(pPix, scan); \
+ leftx += dleftx; \
+ rightx += drightx; \
+ loy++; \
+ } \
+}
+
+#define DEFINE_SOLID_DRAWPARALLELOGRAM_FUNCS(DST) \
+ DrawParallelogramFuncs NAME_SOLID_PGRAM_FUNCS(DST) = { \
+ NAME_SOLID_FILLPGRAM(DST), \
+ NAME_SOLID_DRAWLINE(DST), \
+ };
+
+#define DEFINE_SOLID_PARALLELOGRAM(DST) \
+ DEFINE_SOLID_FILLPGRAM(DST) \
+ DEFINE_SOLID_DRAWPARALLELOGRAM_FUNCS(DST)
+
+/*
+ * This macro declares the bumpmajor and bumpminor variables used for the
+ * DrawLine functions.
+ */
+#define DeclareBumps(BUMPMAJOR, BUMPMINOR) \
+ jint BUMPMAJOR, BUMPMINOR;
+
+/*
+ * This macro extracts "instructions" from the bumpmajor and bumpminor masks
+ * that determine the initial bumpmajor and bumpminor values. The bumpmajor
+ * and bumpminor masks are laid out in the following format:
+ *
+ * bumpmajormask: bumpminormask:
+ * bit0: bumpmajor = pixelStride bit0: bumpminor = pixelStride
+ * bit1: bumpmajor = -pixelStride bit1: bumpminor = -pixelStride
+ * bit2: bumpmajor = scanStride bit2: bumpminor = scanStride
+ * bit3: bumpmajor = -scanStride bit3: bumpminor = -scanStride
+ */
+#define InitBumps(BUMPMAJOR, BUMPMINOR, \
+ BUMPMAJORMASK, BUMPMINORMASK, \
+ PIXELSTRIDE, SCANSTRIDE) \
+ BUMPMAJOR = (BUMPMAJORMASK & BUMP_POS_PIXEL) ? PIXELSTRIDE : \
+ (BUMPMAJORMASK & BUMP_NEG_PIXEL) ? -PIXELSTRIDE : \
+ (BUMPMAJORMASK & BUMP_POS_SCAN) ? SCANSTRIDE : \
+ -SCANSTRIDE; \
+ BUMPMINOR = (BUMPMINORMASK & BUMP_POS_PIXEL) ? PIXELSTRIDE : \
+ (BUMPMINORMASK & BUMP_NEG_PIXEL) ? -PIXELSTRIDE : \
+ (BUMPMINORMASK & BUMP_POS_SCAN) ? SCANSTRIDE : \
+ (BUMPMINORMASK & BUMP_NEG_SCAN) ? -SCANSTRIDE : \
+ 0; \
+ BUMPMINOR += BUMPMAJOR;
+
+/*
+ * This macro defines an entire function to implement a DrawLine inner loop
+ * for iterating along a horizontal or vertical line and setting the pixels
+ * on that line to a specific pixel value. No blending of the fill color
+ * is done with the pixels.
+ */
+#define DEFINE_SOLID_DRAWLINE(DST) \
+void NAME_SOLID_DRAWLINE(DST)(SurfaceDataRasInfo *pRasInfo, \
+ jint x1, jint y1, jint pixel, \
+ jint steps, jint error, \
+ jint bumpmajormask, jint errmajor, \
+ jint bumpminormask, jint errminor, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ Declare ## DST ## PixelData(pix) \
+ jint scan = pRasInfo->scanStride; \
+ DST ## DataType *pPix = PtrCoord(pRasInfo->rasBase, \
+ x1, DST ## PixelStride, \
+ y1, scan); \
+ DeclareBumps(bumpmajor, bumpminor) \
+ \
+ InitBumps(bumpmajor, bumpminor, bumpmajormask, bumpminormask, \
+ DST ## PixelStride, scan); \
+ Extract ## DST ## PixelData(pixel, pix); \
+ if (errmajor == 0) { \
+ do { \
+ Store ## DST ## PixelData(pPix, 0, pixel, pix); \
+ pPix = PtrAddBytes(pPix, bumpmajor); \
+ } while (--steps > 0); \
+ } else { \
+ do { \
+ Store ## DST ## PixelData(pPix, 0, pixel, pix); \
+ if (error < 0) { \
+ pPix = PtrAddBytes(pPix, bumpmajor); \
+ error += errmajor; \
+ } else { \
+ pPix = PtrAddBytes(pPix, bumpminor); \
+ error -= errminor; \
+ } \
+ } while (--steps > 0); \
+ } \
+}
+
+/*
+ * This macro defines an entire function to implement a FillRect inner loop
+ * for setting a rectangular region of pixels to a specific pixel value.
+ * Each destination pixel is XORed with the current XOR mode color as well as
+ * the current fill color.
+ */
+#define DEFINE_XOR_FILLRECT(DST) \
+void NAME_XOR_FILLRECT(DST)(SurfaceDataRasInfo *pRasInfo, \
+ jint lox, jint loy, \
+ jint hix, jint hiy, \
+ jint pixel, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint xorpixel = pCompInfo->details.xorPixel; \
+ juint alphamask = pCompInfo->alphaMask; \
+ Declare ## DST ## PixelData(xor) \
+ Declare ## DST ## PixelData(pix) \
+ Declare ## DST ## PixelData(mask) \
+ DST ## DataType *pPix; \
+ jint scan = pRasInfo->scanStride; \
+ juint height = hiy - loy; \
+ juint width = hix - lox; \
+ \
+ pPix = PtrCoord(pRasInfo->rasBase, lox, DST ## PixelStride, loy, scan); \
+ Extract ## DST ## PixelData(xorpixel, xor); \
+ Extract ## DST ## PixelData(pixel, pix); \
+ Extract ## DST ## PixelData(alphamask, mask); \
+ \
+ do { \
+ juint x = 0; \
+ do { \
+ Xor ## DST ## PixelData(pixel, pix, pPix, x, \
+ xorpixel, xor, alphamask, mask); \
+ } while (++x < width); \
+ pPix = PtrAddBytes(pPix, scan); \
+ } while (--height > 0); \
+}
+
+/*
+ * This macro defines an entire function to implement a FillSpans inner loop
+ * for iterating through a list of spans and setting those regions of pixels
+ * to a specific pixel value. Each destination pixel is XORed with the
+ * current XOR mode color as well as the current fill color.
+ */
+#define DEFINE_XOR_FILLSPANS(DST) \
+void NAME_XOR_FILLSPANS(DST)(SurfaceDataRasInfo *pRasInfo, \
+ SpanIteratorFuncs *pSpanFuncs, \
+ void *siData, jint pixel, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ void *pBase = pRasInfo->rasBase; \
+ jint xorpixel = pCompInfo->details.xorPixel; \
+ juint alphamask = pCompInfo->alphaMask; \
+ Declare ## DST ## PixelData(xor) \
+ Declare ## DST ## PixelData(pix) \
+ Declare ## DST ## PixelData(mask) \
+ jint scan = pRasInfo->scanStride; \
+ jint bbox[4]; \
+ \
+ Extract ## DST ## PixelData(xorpixel, xor); \
+ Extract ## DST ## PixelData(pixel, pix); \
+ Extract ## DST ## PixelData(alphamask, mask); \
+ \
+ while ((*pSpanFuncs->nextSpan)(siData, bbox)) { \
+ jint x = bbox[0]; \
+ jint y = bbox[1]; \
+ juint w = bbox[2] - x; \
+ juint h = bbox[3] - y; \
+ DST ## DataType *pPix = PtrCoord(pBase, \
+ x, DST ## PixelStride, \
+ y, scan); \
+ do { \
+ juint relx; \
+ for (relx = 0; relx < w; relx++) { \
+ Xor ## DST ## PixelData(pixel, pix, pPix, relx, \
+ xorpixel, xor, alphamask, mask); \
+ } \
+ pPix = PtrAddBytes(pPix, scan); \
+ } while (--h > 0); \
+ } \
+}
+
+/*
+ * This macro defines an entire function to implement a DrawLine inner loop
+ * for iterating along a horizontal or vertical line and setting the pixels
+ * on that line to a specific pixel value. Each destination pixel is XORed
+ * with the current XOR mode color as well as the current draw color.
+ */
+#define DEFINE_XOR_DRAWLINE(DST) \
+void NAME_XOR_DRAWLINE(DST)(SurfaceDataRasInfo *pRasInfo, \
+ jint x1, jint y1, jint pixel, \
+ jint steps, jint error, \
+ jint bumpmajormask, jint errmajor, \
+ jint bumpminormask, jint errminor, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint xorpixel = pCompInfo->details.xorPixel; \
+ juint alphamask = pCompInfo->alphaMask; \
+ Declare ## DST ## PixelData(xor) \
+ Declare ## DST ## PixelData(pix) \
+ Declare ## DST ## PixelData(mask) \
+ jint scan = pRasInfo->scanStride; \
+ DST ## DataType *pPix = PtrCoord(pRasInfo->rasBase, \
+ x1, DST ## PixelStride, \
+ y1, scan); \
+ DeclareBumps(bumpmajor, bumpminor) \
+ \
+ InitBumps(bumpmajor, bumpminor, bumpmajormask, bumpminormask, \
+ DST ## PixelStride, scan); \
+ Extract ## DST ## PixelData(xorpixel, xor); \
+ Extract ## DST ## PixelData(pixel, pix); \
+ Extract ## DST ## PixelData(alphamask, mask); \
+ \
+ if (errmajor == 0) { \
+ do { \
+ Xor ## DST ## PixelData(pixel, pix, pPix, 0, \
+ xorpixel, xor, alphamask, mask); \
+ pPix = PtrAddBytes(pPix, bumpmajor); \
+ } while (--steps > 0); \
+ } else { \
+ do { \
+ Xor ## DST ## PixelData(pixel, pix, pPix, 0, \
+ xorpixel, xor, alphamask, mask); \
+ if (error < 0) { \
+ pPix = PtrAddBytes(pPix, bumpmajor); \
+ error += errmajor; \
+ } else { \
+ pPix = PtrAddBytes(pPix, bumpminor); \
+ error -= errminor; \
+ } \
+ } while (--steps > 0); \
+ } \
+}
+
+/*
+ * This macro is used to declare the variables needed by the glyph clipping
+ * macro.
+ */
+#define DeclareDrawGlyphListClipVars(PIXELS, ROWBYTES, WIDTH, HEIGHT, \
+ LEFT, TOP, RIGHT, BOTTOM) \
+ const jubyte * PIXELS; \
+ int ROWBYTES; \
+ int LEFT, TOP; \
+ int WIDTH, HEIGHT; \
+ int RIGHT, BOTTOM;
+
+/*
+ * This macro represents the glyph clipping code used in the various
+ * DRAWGLYPHLIST macros. This macro is typically used within a loop. Note
+ * that the body of this macro is NOT wrapped in a do..while block due to
+ * the use of continue statements within the block (those continue statements
+ * are intended skip the outer loop, not the do..while loop). To combat this
+ * problem, pass in the code (typically a continue statement) that should be
+ * executed when a null glyph is encountered.
+ */
+#define ClipDrawGlyphList(DST, PIXELS, BYTESPERPIXEL, ROWBYTES, WIDTH, HEIGHT,\
+ LEFT, TOP, RIGHT, BOTTOM, \
+ CLIPLEFT, CLIPTOP, CLIPRIGHT, CLIPBOTTOM, \
+ GLYPHS, GLYPHCOUNTER, NULLGLYPHCODE) \
+ PIXELS = (const jubyte *)GLYPHS[GLYPHCOUNTER].pixels; \
+ if (!PIXELS) { \
+ NULLGLYPHCODE; \
+ } \
+ ROWBYTES = GLYPHS[GLYPHCOUNTER].rowBytes; \
+ LEFT = GLYPHS[GLYPHCOUNTER].x; \
+ TOP = GLYPHS[GLYPHCOUNTER].y; \
+ WIDTH = GLYPHS[GLYPHCOUNTER].width; \
+ HEIGHT = GLYPHS[GLYPHCOUNTER].height; \
+\
+ /* if any clipping required, modify parameters now */ \
+ RIGHT = LEFT + WIDTH; \
+ BOTTOM = TOP + HEIGHT; \
+ if (LEFT < CLIPLEFT) { \
+ /* Multiply needed for LCD text as PIXELS is really BYTES */ \
+ PIXELS += (CLIPLEFT - LEFT) * BYTESPERPIXEL ; \
+ LEFT = CLIPLEFT; \
+ } \
+ if (TOP < CLIPTOP) { \
+ PIXELS += (CLIPTOP - TOP) * ROWBYTES; \
+ TOP = CLIPTOP; \
+ } \
+ if (RIGHT > CLIPRIGHT) { \
+ RIGHT = CLIPRIGHT; \
+ } \
+ if (BOTTOM > CLIPBOTTOM) { \
+ BOTTOM = CLIPBOTTOM; \
+ } \
+ if (RIGHT <= LEFT || BOTTOM <= TOP) { \
+ NULLGLYPHCODE; \
+ } \
+ WIDTH = RIGHT - LEFT; \
+ HEIGHT = BOTTOM - TOP;
+
+#define DEFINE_SOLID_DRAWGLYPHLIST(DST) \
+void NAME_SOLID_DRAWGLYPHLIST(DST)(SurfaceDataRasInfo *pRasInfo, \
+ ImageRef *glyphs, \
+ jint totalGlyphs, jint fgpixel, \
+ jint argbcolor, \
+ jint clipLeft, jint clipTop, \
+ jint clipRight, jint clipBottom, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint glyphCounter; \
+ jint scan = pRasInfo->scanStride; \
+ Declare ## DST ## PixelData(pix) \
+ DST ## DataType *pPix; \
+\
+ Extract ## DST ## PixelData(fgpixel, pix); \
+ for (glyphCounter = 0; glyphCounter < totalGlyphs; glyphCounter++) { \
+ DeclareDrawGlyphListClipVars(pixels, rowBytes, width, height, \
+ left, top, right, bottom) \
+ ClipDrawGlyphList(DST, pixels, 1, rowBytes, width, height, \
+ left, top, right, bottom, \
+ clipLeft, clipTop, clipRight, clipBottom, \
+ glyphs, glyphCounter, continue) \
+ pPix = PtrCoord(pRasInfo->rasBase,left,DST ## PixelStride,top,scan); \
+\
+ do { \
+ int x = 0; \
+ do { \
+ if (pixels[x]) { \
+ Store ## DST ## PixelData(pPix, x, fgpixel, pix); \
+ } \
+ } while (++x < width); \
+ pPix = PtrAddBytes(pPix, scan); \
+ pixels += rowBytes; \
+ } while (--height > 0); \
+ } \
+}
+
+#define GlyphListAABlend3ByteRgb(DST, GLYPH_PIXELS, PIXEL_INDEX, DST_PTR, \
+ FG_PIXEL, PREFIX, SRC_PREFIX) \
+ do { \
+ DeclareCompVarsFor3ByteRgb(dst) \
+ jint mixValSrc = GLYPH_PIXELS[PIXEL_INDEX]; \
+ if (mixValSrc) { \
+ if (mixValSrc < 255) { \
+ jint mixValDst = 255 - mixValSrc; \
+ Load ## DST ## To3ByteRgb(DST_PTR, pix, PIXEL_INDEX, \
+ dstR, dstG, dstB); \
+ MultMultAddAndStore3ByteRgbComps(dst, mixValDst, dst, \
+ mixValSrc, SRC_PREFIX); \
+ Store ## DST ## From3ByteRgb(DST_PTR, pix, PIXEL_INDEX, \
+ dstR, dstG, dstB); \
+ } else { \
+ Store ## DST ## PixelData(DST_PTR, PIXEL_INDEX, \
+ FG_PIXEL, PREFIX); \
+ } \
+ } \
+ } while (0);
+
+/*
+ * Antialiased glyph drawing results in artifacts around the character edges
+ * when text is drawn ontop of translucent background color. The standard
+ * blending equation for two colors:
+ * destColor = srcColor * glyphAlpha + destColor * (1 - glyphAlpha)
+ * works only when srcColor and destColor are opaque. For translucent srcColor
+ * and destColor, the respective alpha components in each color will influence
+ * the visibility of the color and the visibility of the color below it. Hence
+ * the equation for blending is given as:
+ * resA = srcAlpha + dstAlpha * (1 - srcAlpha)
+ * resCol = (srcColor * srcAlpha + destColor * destAlpha * (1- srcAlpha))/resA
+ * In addition, srcAlpha is multiplied with the glyphAlpha- that indicates the
+ * grayscale mask value of the glyph being drawn. The combined result provides
+ * smooth antialiased text on the buffer without any artifacts. Since the
+ * logic is executed for every pixel in a glyph, the implementation is further
+ * optimized to reduce computation and improve execution time.
+ */
+#define GlyphListAABlend4ByteArgb(DST, GLYPH_PIXELS, PIXEL_INDEX, DST_PTR, \
+ FG_PIXEL, PREFIX, SRC_PREFIX) \
+ do { \
+ DeclareAlphaVarFor4ByteArgb(resA) \
+ DeclareCompVarsFor4ByteArgb(res) \
+ jint mixValSrc = GLYPH_PIXELS[PIXEL_INDEX]; \
+ if (mixValSrc) { \
+ if (mixValSrc != 0xff) { \
+ PromoteByteAlphaFor4ByteArgb(mixValSrc); \
+ resA = MultiplyAlphaFor4ByteArgb(mixValSrc, SRC_PREFIX ## A); \
+ } else { \
+ resA = SRC_PREFIX ## A; \
+ } \
+ if (resA != MaxValFor4ByteArgb) { \
+ DeclareAndInvertAlphaVarFor4ByteArgb(dstF, resA) \
+ DeclareAndClearAlphaVarFor4ByteArgb(dstA) \
+ DeclareCompVarsFor4ByteArgb(dst) \
+ DeclareCompVarsFor4ByteArgb(tmp) \
+ MultiplyAndStore4ByteArgbComps(res, resA, SRC_PREFIX); \
+ if (!(DST ## IsPremultiplied)) { \
+ Load ## DST ## To4ByteArgb(DST_PTR, pix, PIXEL_INDEX, \
+ dstA, dstR, dstG, dstB); \
+ Store4ByteArgbCompsUsingOp(tmp, =, dst); \
+ } else { \
+ Declare ## DST ## AlphaLoadData(DstPix) \
+ jint pixelOffset = PIXEL_INDEX * (DST ## PixelStride); \
+ DST ## DataType *pixelAddress = PtrAddBytes(DST_PTR, \
+ pixelOffset); \
+ LoadAlphaFrom ## DST ## For4ByteArgb(pixelAddress, \
+ DstPix, \
+ dst); \
+ Postload4ByteArgbFrom ## DST(pixelAddress, \
+ DstPix, \
+ tmp); \
+ } \
+ if (dstA) { \
+ DeclareAlphaVarFor4ByteArgb(blendF) \
+ dstA = MultiplyAlphaFor4ByteArgb(dstF, dstA); \
+ resA += dstA; \
+ blendF = SrcOver ## DST ## BlendFactor(dstF, dstA); \
+ if (blendF != MaxValFor4ByteArgb) { \
+ MultiplyAndStore4ByteArgbComps(tmp, \
+ blendF, \
+ tmp); \
+ } \
+ Store4ByteArgbCompsUsingOp(res, +=, tmp); \
+ } \
+ } else { \
+ Store ## DST ## PixelData(DST_PTR, PIXEL_INDEX, \
+ FG_PIXEL, PREFIX); \
+ break; \
+ } \
+ if (!(DST ## IsOpaque) && \
+ !(DST ## IsPremultiplied) && resA && \
+ resA < MaxValFor4ByteArgb) \
+ { \
+ DivideAndStore4ByteArgbComps(res, res, resA); \
+ } \
+ Store ## DST ## From4ByteArgbComps(DST_PTR, pix, \
+ PIXEL_INDEX, res); \
+ } \
+ } while (0);
+
+#define GlyphListAABlend1ByteGray(DST, GLYPH_PIXELS, PIXEL_INDEX, DST_PTR, \
+ FG_PIXEL, PREFIX, SRC_PREFIX) \
+ do { \
+ DeclareCompVarsFor1ByteGray(dst) \
+ jint mixValSrc = GLYPH_PIXELS[PIXEL_INDEX]; \
+ if (mixValSrc) { \
+ if (mixValSrc < 255) { \
+ jint mixValDst = 255 - mixValSrc; \
+ Load ## DST ## To1ByteGray(DST_PTR, pix, PIXEL_INDEX, \
+ dstG); \
+ MultMultAddAndStore1ByteGrayComps(dst, mixValDst, dst, \
+ mixValSrc, SRC_PREFIX); \
+ Store ## DST ## From1ByteGray(DST_PTR, pix, PIXEL_INDEX, \
+ dstG); \
+ } else { \
+ Store ## DST ## PixelData(DST_PTR, PIXEL_INDEX, \
+ FG_PIXEL, PREFIX); \
+ } \
+ } \
+ } while (0);
+
+#define GlyphListAABlend1ShortGray(DST, GLYPH_PIXELS, PIXEL_INDEX, DST_PTR, \
+ FG_PIXEL, PREFIX, SRC_PREFIX) \
+ do { \
+ DeclareCompVarsFor1ShortGray(dst) \
+ juint mixValSrc = GLYPH_PIXELS[PIXEL_INDEX]; \
+ if (mixValSrc) { \
+ if (mixValSrc < 255) { \
+ juint mixValDst; \
+ PromoteByteAlphaFor1ShortGray(mixValSrc); \
+ mixValDst = 0xffff - mixValSrc; \
+ Load ## DST ## To1ShortGray(DST_PTR, pix, PIXEL_INDEX, \
+ dstG); \
+ MultMultAddAndStore1ShortGrayComps(dst, mixValDst, dst, \
+ mixValSrc, SRC_PREFIX); \
+ Store ## DST ## From1ShortGray(DST_PTR, pix, PIXEL_INDEX, \
+ dstG); \
+ } else { \
+ Store ## DST ## PixelData(DST_PTR, PIXEL_INDEX, \
+ FG_PIXEL, PREFIX); \
+ } \
+ } \
+ } while (0);
+
+#define DEFINE_SOLID_DRAWGLYPHLISTAA(DST, STRATEGY) \
+void NAME_SOLID_DRAWGLYPHLISTAA(DST)(SurfaceDataRasInfo *pRasInfo, \
+ ImageRef *glyphs, \
+ jint totalGlyphs, jint fgpixel, \
+ jint argbcolor, \
+ jint clipLeft, jint clipTop, \
+ jint clipRight, jint clipBottom, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint glyphCounter; \
+ jint scan = pRasInfo->scanStride; \
+ DST ## DataType *pPix; \
+ Declare ## DST ## PixelData(solidpix) \
+ DeclareAlphaVarFor ## STRATEGY(srcA) \
+ DeclareCompVarsFor ## STRATEGY(src) \
+\
+ Declare ## DST ## LoadVars(pix) \
+ Declare ## DST ## StoreVars(pix) \
+\
+ Init ## DST ## LoadVars(pix, pRasInfo); \
+ Init ## DST ## StoreVarsY(pix, pRasInfo); \
+ Init ## DST ## StoreVarsX(pix, pRasInfo); \
+ Extract ## STRATEGY ## CompsAndAlphaFromArgb(argbcolor, src); \
+ Extract ## DST ## PixelData(fgpixel, solidpix); \
+\
+ for (glyphCounter = 0; glyphCounter < totalGlyphs; glyphCounter++) { \
+ DeclareDrawGlyphListClipVars(pixels, rowBytes, width, height, \
+ left, top, right, bottom) \
+ ClipDrawGlyphList(DST, pixels, 1, rowBytes, width, height, \
+ left, top, right, bottom, \
+ clipLeft, clipTop, clipRight, clipBottom, \
+ glyphs, glyphCounter, continue) \
+ pPix = PtrCoord(pRasInfo->rasBase,left,DST ## PixelStride,top,scan); \
+\
+ Set ## DST ## StoreVarsYPos(pix, pRasInfo, top); \
+ do { \
+ int x = 0; \
+ Set ## DST ## StoreVarsXPos(pix, pRasInfo, left); \
+ do { \
+ GlyphListAABlend ## STRATEGY(DST, pixels, x, pPix, \
+ fgpixel, solidpix, src); \
+ Next ## DST ## StoreVarsX(pix); \
+ } while (++x < width); \
+ pPix = PtrAddBytes(pPix, scan); \
+ pixels += rowBytes; \
+ Next ## DST ## StoreVarsY(pix); \
+ } while (--height > 0); \
+ } \
+}
+
+
+#define GlyphListLCDBlend3ByteRgb(DST, GLYPH_PIXELS, PIXEL_INDEX, DST_PTR, \
+ FG_PIXEL, PREFIX, SRC_PREFIX) \
+ do { \
+ DeclareCompVarsFor3ByteRgb(dst) \
+ jint mixValSrcG = GLYPH_PIXELS[PIXEL_INDEX*3+1]; \
+ jint mixValSrcR, mixValSrcB; \
+ if (rgbOrder) { \
+ mixValSrcR = GLYPH_PIXELS[PIXEL_INDEX*3]; \
+ mixValSrcB = GLYPH_PIXELS[PIXEL_INDEX*3+2]; \
+ } else { \
+ mixValSrcR = GLYPH_PIXELS[PIXEL_INDEX*3+2]; \
+ mixValSrcB = GLYPH_PIXELS[PIXEL_INDEX*3]; \
+ } \
+ if ((mixValSrcR | mixValSrcG | mixValSrcB) != 0) { \
+ if ((mixValSrcR & mixValSrcG & mixValSrcB) < 255) { \
+ jint mixValDstR = 255 - mixValSrcR; \
+ jint mixValDstG = 255 - mixValSrcG; \
+ jint mixValDstB = 255 - mixValSrcB; \
+ Load ## DST ## To3ByteRgb(DST_PTR, pix, PIXEL_INDEX, \
+ dstR, dstG, dstB); \
+ dstR = invGammaLut[dstR]; \
+ dstG = invGammaLut[dstG]; \
+ dstB = invGammaLut[dstB]; \
+ MultMultAddAndStoreLCD3ByteRgbComps(dst, mixValDst, dst, \
+ mixValSrc, SRC_PREFIX); \
+ dstR = gammaLut[dstR]; \
+ dstG = gammaLut[dstG]; \
+ dstB = gammaLut[dstB]; \
+ Store ## DST ## From3ByteRgb(DST_PTR, pix, PIXEL_INDEX, \
+ dstR, dstG, dstB); \
+ } else { \
+ Store ## DST ## PixelData(DST_PTR, PIXEL_INDEX, \
+ FG_PIXEL, PREFIX); \
+ } \
+ } \
+ } while (0)
+
+
+/* There is no alpha channel in the glyph data with which to interpolate
+ * between the src and dst alphas, but a reasonable approximation is to
+ * sum the coverage alphas of the colour channels and divide by 3.
+ * We can approximate division by 3 using mult and shift. See
+ * sun/font/scalerMethods.c for a detailed explanation of why "21931"
+ */
+#define GlyphListLCDBlend4ByteArgb(DST, GLYPH_PIXELS, PIXEL_INDEX, DST_PTR, \
+ FG_PIXEL, PREFIX, SRC_PREFIX) \
+ do { \
+ DeclareAlphaVarFor4ByteArgb(dstA) \
+ DeclareCompVarsFor4ByteArgb(dst) \
+ jint mixValSrcG = GLYPH_PIXELS[PIXEL_INDEX*3+1]; \
+ jint mixValSrcR, mixValSrcB; \
+ if (rgbOrder) { \
+ mixValSrcR = GLYPH_PIXELS[PIXEL_INDEX*3]; \
+ mixValSrcB = GLYPH_PIXELS[PIXEL_INDEX*3+2]; \
+ } else { \
+ mixValSrcR = GLYPH_PIXELS[PIXEL_INDEX*3+2]; \
+ mixValSrcB = GLYPH_PIXELS[PIXEL_INDEX*3]; \
+ } \
+ if ((mixValSrcR | mixValSrcG | mixValSrcB) != 0) { \
+ if ((mixValSrcR & mixValSrcG & mixValSrcB) < 255) { \
+ jint mixValDstR = 255 - mixValSrcR; \
+ jint mixValDstG = 255 - mixValSrcG; \
+ jint mixValDstB = 255 - mixValSrcB; \
+ jint mixValSrcA = ((mixValSrcR + mixValSrcG + mixValSrcB) \
+ * 21931) >> 16;\
+ jint mixValDstA = 255 - mixValSrcA; \
+ Load ## DST ## To4ByteArgb(DST_PTR, pix, PIXEL_INDEX, \
+ dstA, dstR, dstG, dstB); \
+ dstR = invGammaLut[dstR]; \
+ dstG = invGammaLut[dstG]; \
+ dstB = invGammaLut[dstB]; \
+ dstA = MUL8(dstA, mixValDstA) + \
+ MUL8(SRC_PREFIX ## A, mixValSrcA); \
+ MultMultAddAndStoreLCD4ByteArgbComps(dst, mixValDst, dst, \
+ mixValSrc, SRC_PREFIX); \
+ dstR = gammaLut[dstR]; \
+ dstG = gammaLut[dstG]; \
+ dstB = gammaLut[dstB]; \
+ if (!(DST ## IsOpaque) && \
+ !(DST ## IsPremultiplied) && dstA && dstA < 255) { \
+ DivideAndStore4ByteArgbComps(dst, dst, dstA); \
+ } \
+ Store ## DST ## From4ByteArgbComps(DST_PTR, pix, \
+ PIXEL_INDEX, dst); \
+ } else { \
+ Store ## DST ## PixelData(DST_PTR, PIXEL_INDEX, \
+ FG_PIXEL, PREFIX); \
+ } \
+ } \
+ } while (0);
+
+#define DEFINE_SOLID_DRAWGLYPHLISTLCD(DST, STRATEGY) \
+void NAME_SOLID_DRAWGLYPHLISTLCD(DST)(SurfaceDataRasInfo *pRasInfo, \
+ ImageRef *glyphs, \
+ jint totalGlyphs, jint fgpixel, \
+ jint argbcolor, \
+ jint clipLeft, jint clipTop, \
+ jint clipRight, jint clipBottom, \
+ jint rgbOrder, \
+ unsigned char *gammaLut, \
+ unsigned char * invGammaLut, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint glyphCounter, bpp; \
+ jint scan = pRasInfo->scanStride; \
+ DST ## DataType *pPix; \
+ Declare ## DST ## PixelData(solidpix) \
+ DeclareAlphaVarFor ## STRATEGY(srcA) \
+ DeclareCompVarsFor ## STRATEGY(src) \
+\
+ Declare ## DST ## LoadVars(pix) \
+ Declare ## DST ## StoreVars(pix) \
+\
+ Init ## DST ## LoadVars(pix, pRasInfo); \
+ Init ## DST ## StoreVarsY(pix, pRasInfo); \
+ Init ## DST ## StoreVarsX(pix, pRasInfo); \
+ Extract ## STRATEGY ## CompsAndAlphaFromArgb(argbcolor, src); \
+ Extract ## DST ## PixelData(fgpixel, solidpix); \
+ srcR = invGammaLut[srcR]; \
+ srcG = invGammaLut[srcG]; \
+ srcB = invGammaLut[srcB]; \
+\
+ for (glyphCounter = 0; glyphCounter < totalGlyphs; glyphCounter++) { \
+ DeclareDrawGlyphListClipVars(pixels, rowBytes, width, height, \
+ left, top, right, bottom) \
+ bpp = \
+ (glyphs[glyphCounter].rowBytes == glyphs[glyphCounter].width) ? 1 : 3;\
+ ClipDrawGlyphList(DST, pixels, bpp, rowBytes, width, height, \
+ left, top, right, bottom, \
+ clipLeft, clipTop, clipRight, clipBottom, \
+ glyphs, glyphCounter, continue) \
+ pPix = PtrCoord(pRasInfo->rasBase,left,DST ## PixelStride,top,scan); \
+\
+ Set ## DST ## StoreVarsYPos(pix, pRasInfo, top); \
+ if (bpp!=1) { \
+ /* subpixel positioning adjustment */ \
+ pixels += glyphs[glyphCounter].rowBytesOffset; \
+ } \
+ do { \
+ int x = 0; \
+ Set ## DST ## StoreVarsXPos(pix, pRasInfo, left); \
+ if (bpp==1) { \
+ do { \
+ if (pixels[x]) { \
+ Store ## DST ## PixelData(pPix, x, fgpixel, solidpix);\
+ } \
+ } while (++x < width); \
+ } else { \
+ do { \
+ GlyphListLCDBlend ## STRATEGY(DST, pixels, x, pPix, \
+ fgpixel, solidpix, src); \
+ Next ## DST ## StoreVarsX(pix); \
+ } while (++x < width); \
+ } \
+ pPix = PtrAddBytes(pPix, scan); \
+ pixels += rowBytes; \
+ Next ## DST ## StoreVarsY(pix); \
+ } while (--height > 0); \
+ } \
+}
+
+#define DEFINE_XOR_DRAWGLYPHLIST(DST) \
+void NAME_XOR_DRAWGLYPHLIST(DST)(SurfaceDataRasInfo *pRasInfo, \
+ ImageRef *glyphs, \
+ jint totalGlyphs, jint fgpixel, \
+ jint argbcolor, \
+ jint clipLeft, jint clipTop, \
+ jint clipRight, jint clipBottom, \
+ NativePrimitive *pPrim, \
+ CompositeInfo *pCompInfo) \
+{ \
+ jint glyphCounter; \
+ jint scan = pRasInfo->scanStride; \
+ jint xorpixel = pCompInfo->details.xorPixel; \
+ juint alphamask = pCompInfo->alphaMask; \
+ Declare ## DST ## PixelData(xor) \
+ Declare ## DST ## PixelData(pix) \
+ Declare ## DST ## PixelData(mask) \
+ DST ## DataType *pPix; \
+ \
+ Extract ## DST ## PixelData(xorpixel, xor); \
+ Extract ## DST ## PixelData(fgpixel, pix); \
+ Extract ## DST ## PixelData(alphamask, mask); \
+ for (glyphCounter = 0; glyphCounter < totalGlyphs; glyphCounter++) { \
+ DeclareDrawGlyphListClipVars(pixels, rowBytes, width, height, \
+ left, top, right, bottom) \
+ ClipDrawGlyphList(DST, pixels, 1, rowBytes, width, height, \
+ left, top, right, bottom, \
+ clipLeft, clipTop, clipRight, clipBottom, \
+ glyphs, glyphCounter, continue) \
+ pPix = PtrCoord(pRasInfo->rasBase,left,DST ## PixelStride,top,scan); \
+ \
+ do { \
+ int x = 0; \
+ do { \
+ if (pixels[x]) { \
+ Xor ## DST ## PixelData(fgpixel, pix, pPix, x, \
+ xorpixel, xor, alphamask, mask); \
+ } \
+ } while (++x < width); \
+ pPix = PtrAddBytes(pPix, scan); \
+ pixels += rowBytes; \
+ } while (--height > 0); \
+ } \
+}
+
+#define DEFINE_TRANSFORMHELPER_NN(SRC) \
+void NAME_TRANSFORMHELPER_NN(SRC)(SurfaceDataRasInfo *pSrcInfo, \
+ jint *pRGB, jint numpix, \
+ jlong xlong, jlong dxlong, \
+ jlong ylong, jlong dylong) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ SRC ## DataType *pBase = pSrcInfo->rasBase; \
+ jint scan = pSrcInfo->scanStride; \
+ jint *pEnd = pRGB + numpix; \
+ \
+ xlong += IntToLong(pSrcInfo->bounds.x1); \
+ ylong += IntToLong(pSrcInfo->bounds.y1); \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ while (pRGB < pEnd) { \
+ SRC ## DataType *pRow = PtrAddBytes(pBase, WholeOfLong(ylong) * scan); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 0, \
+ SrcRead, pRow, WholeOfLong(xlong)); \
+ pRGB++; \
+ xlong += dxlong; \
+ ylong += dylong; \
+ } \
+}
+
+#define DEFINE_TRANSFORMHELPER_BL(SRC) \
+void NAME_TRANSFORMHELPER_BL(SRC)(SurfaceDataRasInfo *pSrcInfo, \
+ jint *pRGB, jint numpix, \
+ jlong xlong, jlong dxlong, \
+ jlong ylong, jlong dylong) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ jint scan = pSrcInfo->scanStride; \
+ jint cx, cy, cw, ch; \
+ jint *pEnd = pRGB + numpix*4; \
+ \
+ cx = pSrcInfo->bounds.x1; \
+ cw = pSrcInfo->bounds.x2-cx; \
+ \
+ cy = pSrcInfo->bounds.y1; \
+ ch = pSrcInfo->bounds.y2-cy; \
+ \
+ xlong -= LongOneHalf; \
+ ylong -= LongOneHalf; \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ while (pRGB < pEnd) { \
+ jint xwhole = WholeOfLong(xlong); \
+ jint ywhole = WholeOfLong(ylong); \
+ jint xdelta, ydelta, isneg; \
+ SRC ## DataType *pRow; \
+ \
+ xdelta = ((juint) (xwhole + 1 - cw)) >> 31; \
+ isneg = xwhole >> 31; \
+ xwhole -= isneg; \
+ xdelta += isneg; \
+ \
+ ydelta = ((ywhole + 1 - ch) >> 31); \
+ isneg = ywhole >> 31; \
+ ywhole -= isneg; \
+ ydelta -= isneg; \
+ ydelta &= scan; \
+ \
+ xwhole += cx; \
+ pRow = PtrAddBytes(pSrcInfo->rasBase, (ywhole + cy) * scan); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 0, SrcRead, pRow, xwhole); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 1, SrcRead, pRow, xwhole+xdelta); \
+ pRow = PtrAddBytes(pRow, ydelta); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 2, SrcRead, pRow, xwhole); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 3, SrcRead, pRow, xwhole+xdelta); \
+ \
+ pRGB += 4; \
+ xlong += dxlong; \
+ ylong += dylong; \
+ } \
+}
+
+#define DEFINE_TRANSFORMHELPER_BC(SRC) \
+void NAME_TRANSFORMHELPER_BC(SRC)(SurfaceDataRasInfo *pSrcInfo, \
+ jint *pRGB, jint numpix, \
+ jlong xlong, jlong dxlong, \
+ jlong ylong, jlong dylong) \
+{ \
+ Declare ## SRC ## LoadVars(SrcRead) \
+ jint scan = pSrcInfo->scanStride; \
+ jint cx, cy, cw, ch; \
+ jint *pEnd = pRGB + numpix*16; \
+ \
+ cx = pSrcInfo->bounds.x1; \
+ cw = pSrcInfo->bounds.x2-cx; \
+ \
+ cy = pSrcInfo->bounds.y1; \
+ ch = pSrcInfo->bounds.y2-cy; \
+ \
+ xlong -= LongOneHalf; \
+ ylong -= LongOneHalf; \
+ \
+ Init ## SRC ## LoadVars(SrcRead, pSrcInfo); \
+ while (pRGB < pEnd) { \
+ jint xwhole = WholeOfLong(xlong); \
+ jint ywhole = WholeOfLong(ylong); \
+ jint xdelta0, xdelta1, xdelta2; \
+ jint ydelta0, ydelta1, ydelta2; \
+ jint isneg; \
+ SRC ## DataType *pRow; \
+ \
+ xdelta0 = (-xwhole) >> 31; \
+ xdelta1 = ((juint) (xwhole + 1 - cw)) >> 31; \
+ xdelta2 = ((juint) (xwhole + 2 - cw)) >> 31; \
+ isneg = xwhole >> 31; \
+ xwhole -= isneg; \
+ xdelta1 += isneg; \
+ xdelta2 += xdelta1; \
+ \
+ ydelta0 = ((-ywhole) >> 31) & (-scan); \
+ ydelta1 = ((ywhole + 1 - ch) >> 31) & scan; \
+ ydelta2 = ((ywhole + 2 - ch) >> 31) & scan; \
+ isneg = ywhole >> 31; \
+ ywhole -= isneg; \
+ ydelta1 += (isneg & -scan); \
+ \
+ xwhole += cx; \
+ pRow = PtrAddBytes(pSrcInfo->rasBase, (ywhole + cy) * scan); \
+ pRow = PtrAddBytes(pRow, ydelta0); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 0, SrcRead, pRow, xwhole+xdelta0); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 1, SrcRead, pRow, xwhole ); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 2, SrcRead, pRow, xwhole+xdelta1); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 3, SrcRead, pRow, xwhole+xdelta2); \
+ pRow = PtrAddBytes(pRow, -ydelta0); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 4, SrcRead, pRow, xwhole+xdelta0); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 5, SrcRead, pRow, xwhole ); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 6, SrcRead, pRow, xwhole+xdelta1); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 7, SrcRead, pRow, xwhole+xdelta2); \
+ pRow = PtrAddBytes(pRow, ydelta1); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 8, SrcRead, pRow, xwhole+xdelta0); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 9, SrcRead, pRow, xwhole ); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 10, SrcRead, pRow, xwhole+xdelta1); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 11, SrcRead, pRow, xwhole+xdelta2); \
+ pRow = PtrAddBytes(pRow, ydelta2); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 12, SrcRead, pRow, xwhole+xdelta0); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 13, SrcRead, pRow, xwhole ); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 14, SrcRead, pRow, xwhole+xdelta1); \
+ Copy ## SRC ## ToIntArgbPre(pRGB, 15, SrcRead, pRow, xwhole+xdelta2); \
+ \
+ pRGB += 16; \
+ xlong += dxlong; \
+ ylong += dylong; \
+ } \
+}
+
+#define DEFINE_TRANSFORMHELPER_FUNCS(SRC) \
+ TransformHelperFuncs NAME_TRANSFORMHELPER_FUNCS(SRC) = { \
+ NAME_TRANSFORMHELPER_NN(SRC), \
+ NAME_TRANSFORMHELPER_BL(SRC), \
+ NAME_TRANSFORMHELPER_BC(SRC), \
+ };
+
+#define DEFINE_TRANSFORMHELPERS(SRC) \
+ DEFINE_TRANSFORMHELPER_NN(SRC) \
+ DEFINE_TRANSFORMHELPER_BL(SRC) \
+ DEFINE_TRANSFORMHELPER_BC(SRC) \
+ DEFINE_TRANSFORMHELPER_FUNCS(SRC)
+
+/*
+ * The macros defined above use the following macro definitions supplied
+ * for the various surface types to manipulate pixels and pixel data.
+ * The surface-specific macros are typically supplied by header files
+ * named after the SurfaceType name (i.e. IntArgb.h, ByteGray.h, etc.).
+ *
+ * In the macro names in the following definitions, the string <stype>
+ * is used as a place holder for the SurfaceType name (i.e. IntArgb).
+ * The macros above access these type specific macros using the ANSI
+ * CPP token concatenation operator "##".
+ *
+ * <stype>DataType A typedef for the type of the pointer
+ * that is used to access the raster data
+ * for the given surface type.
+ * <stype>PixelStride Pixel stride for the surface type.
+ *
+ * Declare<stype>LoadVars Declare the variables needed to control
+ * loading color information from an stype
+ * raster (i.e. lookup tables).
+ * Init<stype>LoadVars Init the lookup table variables.
+ * Declare<stype>StoreVars Declare the storage variables needed to
+ * control storing pixel data based on the
+ * pixel coordinate (i.e. dithering variables).
+ * Init<stype>StoreVarsY Init the dither variables for starting Y.
+ * Next<stype>StoreVarsY Increment the dither variables for next Y.
+ * Init<stype>StoreVarsX Init the dither variables for starting X.
+ * Next<stype>StoreVarsX Increment the dither variables for next X.
+ *
+ * Load<stype>To1IntRgb Load a pixel and form an INT_RGB integer.
+ * Store<stype>From1IntRgb Store a pixel from an INT_RGB integer.
+ * Load<stype>To1IntArgb Load a pixel and form an INT_ARGB integer.
+ * Store<stype>From1IntArgb Store a pixel from an INT_ARGB integer.
+ * Load<stype>To3ByteRgb Load a pixel into R, G, and B components.
+ * Store<stype>From3ByteRgb Store a pixel from R, G, and B components.
+ * Load<stype>To4ByteArgb Load a pixel into A, R, G, and B components.
+ * Store<stype>From4ByteArgb Store a pixel from A, R, G, and B components.
+ * Load<stype>To1ByteGray Load a pixel and form a BYTE_GRAY byte.
+ * Store<stype>From1ByteGray Store a pixel from a BYTE_GRAY byte.
+ *
+ * <stype>PixelType Typedef for a "single quantity pixel" (SQP)
+ * that can hold the data for one stype pixel.
+ * <stype>XparLutEntry An SQP that can be used to represent a
+ * transparent pixel for stype.
+ * Store<stype>NonXparFromArgb Store an SQP from an INT_ARGB integer in
+ * such a way that it would not be confused
+ * with the XparLutEntry value for stype.
+ * <stype>IsXparLutEntry Test an SQP for the XparLutEntry value.
+ * Store<stype>Pixel Store the pixel data from an SQP.
+ * <stype>PixelFromArgb Converts an INT_ARGB value into the specific
+ * pixel representation for the surface type.
+ *
+ * Declare<stype>PixelData Declare the pixel data variables (PDV) needed
+ * to hold the elements of pixel data ready to
+ * store into an stype raster (may be empty for
+ * stypes whose SQP format is their data format).
+ * Extract<stype>PixelData Extract an SQP value into the PDVs.
+ * Store<stype>PixelData Store the PDVs into an stype raster.
+ * XorCopy<stype>PixelData Xor the PDVs into an stype raster.
+ */
+#endif /* LoopMacros_h_Included */