1 files changed, 693 insertions, 0 deletions
diff --git a/libraries/irrlicht-1.8/source/Irrlicht/S4DVertex.h b/libraries/irrlicht-1.8/source/Irrlicht/S4DVertex.h
new file mode 100644
index 0000000..3b80919
--- /dev/null
+++ b/libraries/irrlicht-1.8/source/Irrlicht/S4DVertex.h
@@ -0,0 +1,693 @@
+// Copyright (C) 2002-2012 Nikolaus Gebhardt / Thomas Alten
+// This file is part of the "Irrlicht Engine".
+// For conditions of distribution and use, see copyright notice in irrlicht.h
+#ifndef __S_4D_VERTEX_H_INCLUDED__
+#define __S_4D_VERTEX_H_INCLUDED__
+#include "SoftwareDriver2_compile_config.h"
+#include "SoftwareDriver2_helper.h"
+#include "irrAllocator.h"
+namespace irr
+{
+namespace video
+{
+struct sVec2
+{
+        f32 x;
+        f32 y;
+        sVec2 () {}
+        sVec2 ( f32 s) : x ( s ), y ( s ) {}
+        sVec2 ( f32 _x, f32 _y )
+                : x ( _x ), y ( _y ) {}
+        void set ( f32 _x, f32 _y )
+        {
+                x = _x;
+                y = _y;
+        }
+        // f = a * t + b * ( 1 - t )
+        void interpolate(const sVec2& a, const sVec2& b, const f32 t)
+        {
+                x = b.x + ( ( a.x - b.x ) * t );
+                y = b.y + ( ( a.y - b.y ) * t );
+        }
+        sVec2 operator-(const sVec2& other) const
+        {
+                return sVec2(x - other.x, y - other.y);
+        }
+        sVec2 operator+(const sVec2& other) const
+        {
+                return sVec2(x + other.x, y + other.y);
+        }
+        void operator+=(const sVec2& other)
+        {
+                x += other.x;
+                y += other.y;
+        }
+        sVec2 operator*(const f32 s) const
+        {
+                return sVec2(x * s , y * s);
+        }
+        void operator*=( const f32 s)
+        {
+                x *= s;
+                y *= s;
+        }
+        void operator=(const sVec2& other)
+        {
+                x = other.x;
+                y = other.y;
+        }
+};
+// A8R8G8B8
+struct sVec4;
+struct sCompressedVec4
+{
+        u32 argb;
+        void setA8R8G8B8 ( u32 value )
+        {
+                argb = value;
+        }
+        void setColorf ( const video::SColorf & color )
+        {
+                argb =  core::floor32 ( color.a * 255.f ) << 24 |
+                                core::floor32 ( color.r * 255.f ) << 16 |
+                                core::floor32 ( color.g * 255.f ) << 8  |
+                                core::floor32 ( color.b * 255.f );
+        }
+        void setVec4 ( const sVec4 & v );
+        // f = a * t + b * ( 1 - t )
+        void interpolate(const sCompressedVec4& a, const sCompressedVec4& b, const f32 t)
+        {
+                argb = PixelBlend32 ( b.argb, a.argb, core::floor32 ( t * 256.f ) );
+        }
+};
+struct sVec4
+{
+        union
+        {
+                struct { f32 x, y, z, w; };
+                struct { f32 a, r, g, b; };
+//              struct { sVec2 xy, zw; };       // sorry, this does not compile with gcc
+        };
+        sVec4 () {}
+        sVec4 ( f32 s) : x ( s ), y ( s ), z ( s ), w ( s ) {}
+        sVec4 ( f32 _x, f32 _y, f32 _z, f32 _w )
+                : x ( _x ), y ( _y ), z( _z ), w ( _w ){}
+        void set ( f32 _x, f32 _y, f32 _z, f32 _w )
+        {
+                x = _x;
+                y = _y;
+                z = _z;
+                w = _w;
+        }
+        void setA8R8G8B8 ( u32 argb )
+        {
+                x = ( ( argb & 0xFF000000 ) >> 24 ) * ( 1.f / 255.f );
+                y = ( ( argb & 0x00FF0000 ) >> 16 ) * ( 1.f / 255.f );
+                z = ( ( argb & 0x0000FF00 ) >>  8 ) * ( 1.f / 255.f );
+                w = ( ( argb & 0x000000FF )       ) * ( 1.f / 255.f );
+        }
+        void setColorf ( const video::SColorf & color )
+        {
+                x = color.a;
+                y = color.r;
+                z = color.g;
+                w = color.b;
+        }
+        // f = a * t + b * ( 1 - t )
+        void interpolate(const sVec4& a, const sVec4& b, const f32 t)
+        {
+                x = b.x + ( ( a.x - b.x ) * t );
+                y = b.y + ( ( a.y - b.y ) * t );
+                z = b.z + ( ( a.z - b.z ) * t );
+                w = b.w + ( ( a.w - b.w ) * t );
+        }
+        f32 dotProduct(const sVec4& other) const
+        {
+                return x*other.x + y*other.y + z*other.z + w*other.w;
+        }
+        f32 dot_xyz( const sVec4& other) const
+        {
+                return x*other.x + y*other.y + z*other.z;
+        }
+        f32 get_length_xyz_square () const
+        {
+                return x * x + y * y + z * z;
+        }
+        f32 get_length_xyz () const
+        {
+                return core::squareroot ( x * x + y * y + z * z );
+        }
+        void normalize_xyz ()
+        {
+                const f32 l = core::reciprocal_squareroot ( x * x + y * y + z * z );
+                x *= l;
+                y *= l;
+                z *= l;
+        }
+        void project_xyz ()
+        {
+                w = core::reciprocal ( w );
+                x *= w;
+                y *= w;
+                z *= w;
+        }
+        sVec4 operator-(const sVec4& other) const
+        {
+                return sVec4(x - other.x, y - other.y, z - other.z,w - other.w);
+        }
+        sVec4 operator+(const sVec4& other) const
+        {
+                return sVec4(x + other.x, y + other.y, z + other.z,w + other.w);
+        }
+        void operator+=(const sVec4& other)
+        {
+                x += other.x;
+                y += other.y;
+                z += other.z;
+                w += other.w;
+        }
+        sVec4 operator*(const f32 s) const
+        {
+                return sVec4(x * s , y * s, z * s,w * s);
+        }
+        sVec4 operator*(const sVec4 &other) const
+        {
+                return sVec4(x * other.x , y * other.y, z * other.z,w * other.w);
+        }
+        void mulReciprocal ( f32 s )
+        {
+                const f32 i = core::reciprocal ( s );
+                x = (f32) ( x * i );
+                y = (f32) ( y * i );
+                z = (f32) ( z * i );
+                w = (f32) ( w * i );
+        }
+        void mul ( const f32 s )
+        {
+                x *= s;
+                y *= s;
+                z *= s;
+                w *= s;
+        }
+/*
+        void operator*=(f32 s)
+        {
+                x *= s;
+                y *= s;
+                z *= s;
+                w *= s;
+        }
+*/
+        void operator*=(const sVec4 &other)
+        {
+                x *= other.x;
+                y *= other.y;
+                z *= other.z;
+                w *= other.w;
+        }
+        void operator=(const sVec4& other)
+        {
+                x = other.x;
+                y = other.y;
+                z = other.z;
+                w = other.w;
+        }
+};
+struct sVec3
+{
+        union
+        {
+                struct { f32 r, g, b; };
+                struct { f32 x, y, z; };
+        };
+        sVec3 () {}
+        sVec3 ( f32 _x, f32 _y, f32 _z )
+                : r ( _x ), g ( _y ), b( _z ) {}
+        sVec3 ( const sVec4 &v )
+                : r ( v.x ), g ( v.y ), b( v.z ) {}
+        void set ( f32 _r, f32 _g, f32 _b )
+        {
+                r = _r;
+                g = _g;
+                b = _b;
+        }
+        void setR8G8B8 ( u32 argb )
+        {
+                r = ( ( argb & 0x00FF0000 ) >> 16 ) * ( 1.f / 255.f );
+                g = ( ( argb & 0x0000FF00 ) >>  8 ) * ( 1.f / 255.f );
+                b = ( ( argb & 0x000000FF )       ) * ( 1.f / 255.f );
+        }
+        void setColorf ( const video::SColorf & color )
+        {
+                r = color.r;
+                g = color.g;
+                b = color.b;
+        }
+        void add (const sVec3& other)
+        {
+                r += other.r;
+                g += other.g;
+                b += other.b;
+        }
+        void mulAdd(const sVec3& other, const f32 v)
+        {
+                r += other.r * v;
+                g += other.g * v;
+                b += other.b * v;
+        }
+        void mulAdd(const sVec3& v0, const sVec3& v1)
+        {
+                r += v0.r * v1.r;
+                g += v0.g * v1.g;
+                b += v0.b * v1.b;
+        }
+        void saturate ( sVec4 &dest, u32 argb )
+        {
+                dest.x = ( ( argb & 0xFF000000 ) >> 24 ) * ( 1.f / 255.f );
+                dest.y = core::min_ ( r, 1.f );
+                dest.z = core::min_ ( g, 1.f );
+                dest.w = core::min_ ( b, 1.f );
+        }
+        // f = a * t + b * ( 1 - t )
+        void interpolate(const sVec3& v0, const sVec3& v1, const f32 t)
+        {
+                r = v1.r + ( ( v0.r - v1.r ) * t );
+                g = v1.g + ( ( v0.g - v1.g ) * t );
+                b = v1.b + ( ( v0.b - v1.b ) * t );
+        }
+        sVec3 operator-(const sVec3& other) const
+        {
+                return sVec3(r - other.r, b - other.b, g - other.g);
+        }
+        sVec3 operator+(const sVec3& other) const
+        {
+                return sVec3(r + other.r, g + other.g, b + other.b);
+        }
+        sVec3 operator*(const f32 s) const
+        {
+                return sVec3(r * s , g * s, b * s);
+        }
+        sVec3 operator/(const f32 s) const
+        {
+                f32 inv = 1.f / s;
+                return sVec3(r * inv , g * inv, b * inv);
+        }
+        sVec3 operator*(const sVec3 &other) const
+        {
+                return sVec3(r * other.r , b * other.b, g * other.g);
+        }
+        void operator+=(const sVec3& other)
+        {
+                r += other.r;
+                g += other.g;
+                b += other.b;
+        }
+        void setLength ( f32 len )
+        {
+                const f32 l = len * core::reciprocal_squareroot ( r * r + g * g + b * b );
+                r *= l;
+                g *= l;
+                b *= l;
+        }
+};
+inline void sCompressedVec4::setVec4 ( const sVec4 & v )
+{
+        argb =  core::floor32 ( v.x * 255.f ) << 24 |
+                        core::floor32 ( v.y * 255.f ) << 16 |
+                        core::floor32 ( v.z * 255.f ) << 8  |
+                        core::floor32 ( v.w * 255.f );
+}
+enum e4DVertexFlag
+{
+        VERTEX4D_INSIDE         = 0x0000003F,
+        VERTEX4D_CLIPMASK       = 0x0000003F,
+        VERTEX4D_PROJECTED      = 0x00000100,
+        VERTEX4D_FORMAT_MASK                    = 0xFFFF0000,
+        VERTEX4D_FORMAT_MASK_TEXTURE    = 0x000F0000,
+        VERTEX4D_FORMAT_TEXTURE_1               = 0x00010000,
+        VERTEX4D_FORMAT_TEXTURE_2               = 0x00020000,
+        VERTEX4D_FORMAT_TEXTURE_3               = 0x00030000,
+        VERTEX4D_FORMAT_TEXTURE_4               = 0x00040000,
+        VERTEX4D_FORMAT_MASK_COLOR              = 0x00F00000,
+        VERTEX4D_FORMAT_COLOR_1                 = 0x00100000,
+        VERTEX4D_FORMAT_COLOR_2                 = 0x00200000,
+        VERTEX4D_FORMAT_MASK_BUMP               = 0x0F000000,
+        VERTEX4D_FORMAT_BUMP_DOT3               = 0x01000000,
+};
+const u32 MATERIAL_MAX_COLORS = 1;
+const u32 BURNING_MATERIAL_MAX_TEXTURES = 2;
+const u32 BURNING_MATERIAL_MAX_TANGENT = 1;
+// dummy Vertex. used for calculation vertex memory size
+struct s4DVertex_proxy
+{
+        u32 flag;
+        sVec4 Pos;
+        sVec2 Tex[BURNING_MATERIAL_MAX_TEXTURES];
+#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
+        sVec4 Color[MATERIAL_MAX_COLORS];
+#endif
+        sVec3 LightTangent[BURNING_MATERIAL_MAX_TANGENT];
+};
+#define SIZEOF_SVERTEX  64
+#define SIZEOF_SVERTEX_LOG2     6
+/*!
+        Internal BurningVideo Vertex
+*/
+struct s4DVertex
+{
+        u32 flag;
+        sVec4 Pos;
+        sVec2 Tex[ BURNING_MATERIAL_MAX_TEXTURES ];
+#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
+        sVec4 Color[ MATERIAL_MAX_COLORS ];
+#endif
+        sVec3 LightTangent[BURNING_MATERIAL_MAX_TANGENT];
+        //u8 fill [ SIZEOF_SVERTEX - sizeof (s4DVertex_proxy) ];
+        // f = a * t + b * ( 1 - t )
+        void interpolate(const s4DVertex& b, const s4DVertex& a, const f32 t)
+        {
+                u32 i;
+                u32 size;
+                Pos.interpolate ( a.Pos, b.Pos, t );
+#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
+                size = (flag & VERTEX4D_FORMAT_MASK_COLOR) >> 20;
+                for ( i = 0; i!= size; ++i )
+                {
+                        Color[i].interpolate ( a.Color[i], b.Color[i], t );
+                }
+#endif
+                size = (flag & VERTEX4D_FORMAT_MASK_TEXTURE) >> 16;
+                for ( i = 0; i!= size; ++i )
+                {
+                        Tex[i].interpolate ( a.Tex[i], b.Tex[i], t );
+                }
+                size = (flag & VERTEX4D_FORMAT_MASK_BUMP) >> 24;
+                for ( i = 0; i!= size; ++i )
+                {
+                        LightTangent[i].interpolate ( a.LightTangent[i], b.LightTangent[i], t );
+                }
+        }
+};
+// ----------------- Vertex Cache ---------------------------
+struct SAlignedVertex
+{
+        SAlignedVertex ( u32 element, u32 aligned )
+                : ElementSize ( element )
+        {
+                u32 byteSize = (ElementSize << SIZEOF_SVERTEX_LOG2 ) + aligned;
+                mem = new u8 [ byteSize ];
+                data = (s4DVertex*) mem;
+        }
+        virtual ~SAlignedVertex ()
+        {
+                delete [] mem;
+        }
+        s4DVertex *data;
+        u8 *mem;
+        u32 ElementSize;
+};
+// hold info for different Vertex Types
+struct SVSize
+{
+        u32 Format;
+        u32 Pitch;
+        u32 TexSize;
+};
+// a cache info
+struct SCacheInfo
+{
+        u32 index;
+        u32 hit;
+};
+#define VERTEXCACHE_ELEMENT     16
+#define VERTEXCACHE_MISS 0xFFFFFFFF
+struct SVertexCache
+{
+        SVertexCache (): mem ( VERTEXCACHE_ELEMENT * 2, 128 ) {}
+        SCacheInfo info[VERTEXCACHE_ELEMENT];
+        // Transformed and lite, clipping state
+        // + Clipped, Projected
+        SAlignedVertex mem;
+        // source
+        const void* vertices;
+        u32 vertexCount;
+        const void* indices;
+        u32 indexCount;
+        u32 indicesIndex;
+        u32 indicesRun;
+        // primitives consist of x vertices
+        u32 primitivePitch;
+        u32 vType;              //E_VERTEX_TYPE
+        u32 pType;              //scene::E_PRIMITIVE_TYPE
+        u32 iType;              //E_INDEX_TYPE iType
+};
+// swap 2 pointer
+REALINLINE void swapVertexPointer(const s4DVertex** v1, const s4DVertex** v2)
+{
+        const s4DVertex* b = *v1;
+        *v1 = *v2;
+        *v2 = b;
+}
+// ------------------------ Internal Scanline Rasterizer -----------------------------
+// internal scan convert
+struct sScanConvertData
+{
+        u8 left;                        // major edge left/right
+        u8 right;                       // !left
+        f32 invDeltaY[3];       // inverse edge delta y
+        f32 x[2];                       // x coordinate
+        f32 slopeX[2];          // x slope along edges
+#if defined ( SOFTWARE_DRIVER_2_USE_WBUFFER ) || defined ( SOFTWARE_DRIVER_2_PERSPECTIVE_CORRECT )
+        f32 w[2];                       // w coordinate
+        fp24 slopeW[2];         // w slope along edges
+#else
+        f32 z[2];                       // z coordinate
+        f32 slopeZ[2];          // z slope along edges
+#endif
+        sVec4 c[MATERIAL_MAX_COLORS][2];                        // color
+        sVec4 slopeC[MATERIAL_MAX_COLORS][2];   // color slope along edges
+        sVec2 t[BURNING_MATERIAL_MAX_TEXTURES][2];              // texture
+        sVec2 slopeT[BURNING_MATERIAL_MAX_TEXTURES][2]; // texture slope along edges
+        sVec3 l[BURNING_MATERIAL_MAX_TANGENT][2];               // Light Tangent
+        sVec3 slopeL[BURNING_MATERIAL_MAX_TEXTURES][2]; // tanget slope along edges
+};
+// passed to scan Line
+struct sScanLineData
+{
+        s32 y;                          // y position of scanline
+        f32 x[2];                       // x start, x end of scanline
+#if defined ( SOFTWARE_DRIVER_2_USE_WBUFFER ) || defined ( SOFTWARE_DRIVER_2_PERSPECTIVE_CORRECT )
+        f32 w[2];                       // w start, w end of scanline
+#else
+        f32 z[2];                       // z start, z end of scanline
+#endif
+#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
+        sVec4 c[MATERIAL_MAX_COLORS][2];                        // color start, color end of scanline
+#endif
+        sVec2 t[BURNING_MATERIAL_MAX_TEXTURES][2];              // texture start, texture end of scanline
+        sVec3 l[BURNING_MATERIAL_MAX_TANGENT][2];               // Light Tangent start, end
+};
+// passed to pixel Shader
+struct sPixelShaderData
+{
+        tVideoSample *dst;
+        fp24 *z;
+        s32 xStart;
+        s32 xEnd;
+        s32 dx;
+        s32 i;
+};
+/*
+        load a color value
+*/
+inline void getTexel_plain2 (   tFixPoint &r, tFixPoint &g, tFixPoint &b,
+                                                        const sVec4 &v
+                                                        )
+{
+        r = tofix ( v.y );
+        g = tofix ( v.z );
+        b = tofix ( v.w );
+}
+/*
+        load a color value
+*/
+inline void getSample_color (   tFixPoint &a, tFixPoint &r, tFixPoint &g, tFixPoint &b,
+                                                        const sVec4 &v
+                                                        )
+{
+        a = tofix ( v.x );
+        r = tofix ( v.y, COLOR_MAX * FIX_POINT_F32_MUL);
+        g = tofix ( v.z, COLOR_MAX * FIX_POINT_F32_MUL);
+        b = tofix ( v.w, COLOR_MAX * FIX_POINT_F32_MUL);
+}
+/*
+        load a color value
+*/
+inline void getSample_color ( tFixPoint &r, tFixPoint &g, tFixPoint &b,const sVec4 &v )
+{
+        r = tofix ( v.y, COLOR_MAX * FIX_POINT_F32_MUL);
+        g = tofix ( v.z, COLOR_MAX * FIX_POINT_F32_MUL);
+        b = tofix ( v.w, COLOR_MAX * FIX_POINT_F32_MUL);
+}
+/*
+        load a color value
+*/
+inline void getSample_color (   tFixPoint &r, tFixPoint &g, tFixPoint &b,
+                                                                const sVec4 &v, const f32 mulby )
+{
+        r = tofix ( v.y, mulby);
+        g = tofix ( v.z, mulby);
+        b = tofix ( v.w, mulby);
+}
+}
+}
+#endif

diff --git a/libraries/irrlicht-1.8/source/Irrlicht/S4DVertex.h b/libraries/irrlicht-1.8/source/Irrlicht/S4DVertex.h new file mode 100644 index 0000000..3b80919 --- /dev/null +++ b/libraries/irrlicht-1.8/source/Irrlicht/S4DVertex.h
@@ -0,0 +1,693 @@
	1	// Copyright (C) 2002-2012 Nikolaus Gebhardt / Thomas Alten
	2	// This file is part of the "Irrlicht Engine".
	3	// For conditions of distribution and use, see copyright notice in irrlicht.h
	4
	5
	6	#ifndef __S_4D_VERTEX_H_INCLUDED__
	7	#define __S_4D_VERTEX_H_INCLUDED__
	8
	9	#include "SoftwareDriver2_compile_config.h"
	10	#include "SoftwareDriver2_helper.h"
	11	#include "irrAllocator.h"
	12
	13	namespace irr
	14	{
	15
	16	namespace video
	17	{
	18
	19	struct sVec2
	20	{
	21	f32 x;
	22	f32 y;
	23
	24	sVec2 () {}
	25
	26	sVec2 ( f32 s) : x ( s ), y ( s ) {}
	27	sVec2 ( f32 _x, f32 _y )
	28	: x ( _x ), y ( _y ) {}
	29
	30	void set ( f32 _x, f32 _y )
	31	{
	32	x = _x;
	33	y = _y;
	34	}
	35
	36	// f = a * t + b * ( 1 - t )
	37	void interpolate(const sVec2& a, const sVec2& b, const f32 t)
	38	{
	39	x = b.x + ( ( a.x - b.x ) * t );
	40	y = b.y + ( ( a.y - b.y ) * t );
	41	}
	42
	43	sVec2 operator-(const sVec2& other) const
	44	{
	45	return sVec2(x - other.x, y - other.y);
	46	}
	47
	48	sVec2 operator+(const sVec2& other) const
	49	{
	50	return sVec2(x + other.x, y + other.y);
	51	}
	52
	53	void operator+=(const sVec2& other)
	54	{
	55	x += other.x;
	56	y += other.y;
	57	}
	58
	59	sVec2 operator*(const f32 s) const
	60	{
	61	return sVec2(x * s , y * s);
	62	}
	63
	64	void operator*=( const f32 s)
	65	{
	66	x *= s;
	67	y *= s;
	68	}
	69
	70	void operator=(const sVec2& other)
	71	{
	72	x = other.x;
	73	y = other.y;
	74	}
	75
	76	};
	77
	78	// A8R8G8B8
	79	struct sVec4;
	80	struct sCompressedVec4
	81	{
	82	u32 argb;
	83
	84	void setA8R8G8B8 ( u32 value )
	85	{
	86	argb = value;
	87	}
	88
	89	void setColorf ( const video::SColorf & color )
	90	{
	91	argb = core::floor32 ( color.a * 255.f ) << 24 \|
	92	core::floor32 ( color.r * 255.f ) << 16 \|
	93	core::floor32 ( color.g * 255.f ) << 8 \|
	94	core::floor32 ( color.b * 255.f );
	95	}
	96
	97	void setVec4 ( const sVec4 & v );
	98
	99	// f = a * t + b * ( 1 - t )
	100	void interpolate(const sCompressedVec4& a, const sCompressedVec4& b, const f32 t)
	101	{
	102	argb = PixelBlend32 ( b.argb, a.argb, core::floor32 ( t * 256.f ) );
	103	}
	104
	105
	106	};
	107
	108
	109	struct sVec4
	110	{
	111	union
	112	{
	113	struct { f32 x, y, z, w; };
	114	struct { f32 a, r, g, b; };
	115	// struct { sVec2 xy, zw; }; // sorry, this does not compile with gcc
	116	};
	117
	118
	119
	120	sVec4 () {}
	121
	122	sVec4 ( f32 s) : x ( s ), y ( s ), z ( s ), w ( s ) {}
	123
	124	sVec4 ( f32 _x, f32 _y, f32 _z, f32 _w )
	125	: x ( _x ), y ( _y ), z( _z ), w ( _w ){}
	126
	127	void set ( f32 _x, f32 _y, f32 _z, f32 _w )
	128	{
	129	x = _x;
	130	y = _y;
	131	z = _z;
	132	w = _w;
	133	}
	134
	135	void setA8R8G8B8 ( u32 argb )
	136	{
	137	x = ( ( argb & 0xFF000000 ) >> 24 ) * ( 1.f / 255.f );
	138	y = ( ( argb & 0x00FF0000 ) >> 16 ) * ( 1.f / 255.f );
	139	z = ( ( argb & 0x0000FF00 ) >> 8 ) * ( 1.f / 255.f );
	140	w = ( ( argb & 0x000000FF ) ) * ( 1.f / 255.f );
	141	}
	142
	143
	144	void setColorf ( const video::SColorf & color )
	145	{
	146	x = color.a;
	147	y = color.r;
	148	z = color.g;
	149	w = color.b;
	150	}
	151
	152
	153	// f = a * t + b * ( 1 - t )
	154	void interpolate(const sVec4& a, const sVec4& b, const f32 t)
	155	{
	156	x = b.x + ( ( a.x - b.x ) * t );
	157	y = b.y + ( ( a.y - b.y ) * t );
	158	z = b.z + ( ( a.z - b.z ) * t );
	159	w = b.w + ( ( a.w - b.w ) * t );
	160	}
	161
	162
	163	f32 dotProduct(const sVec4& other) const
	164	{
	165	return xother.x + yother.y + zother.z + wother.w;
	166	}
	167
	168	f32 dot_xyz( const sVec4& other) const
	169	{
	170	return xother.x + yother.y + z*other.z;
	171	}
	172
	173	f32 get_length_xyz_square () const
	174	{
	175	return x * x + y * y + z * z;
	176	}
	177
	178	f32 get_length_xyz () const
	179	{
	180	return core::squareroot ( x * x + y * y + z * z );
	181	}
	182
	183	void normalize_xyz ()
	184	{
	185	const f32 l = core::reciprocal_squareroot ( x * x + y * y + z * z );
	186
	187	x *= l;
	188	y *= l;
	189	z *= l;
	190	}
	191
	192	void project_xyz ()
	193	{
	194	w = core::reciprocal ( w );
	195	x *= w;
	196	y *= w;
	197	z *= w;
	198	}
	199
	200	sVec4 operator-(const sVec4& other) const
	201	{
	202	return sVec4(x - other.x, y - other.y, z - other.z,w - other.w);
	203	}
	204
	205	sVec4 operator+(const sVec4& other) const
	206	{
	207	return sVec4(x + other.x, y + other.y, z + other.z,w + other.w);
	208	}
	209
	210	void operator+=(const sVec4& other)
	211	{
	212	x += other.x;
	213	y += other.y;
	214	z += other.z;
	215	w += other.w;
	216	}
	217
	218	sVec4 operator*(const f32 s) const
	219	{
	220	return sVec4(x * s , y * s, z * s,w * s);
	221	}
	222
	223	sVec4 operator*(const sVec4 &other) const
	224	{
	225	return sVec4(x * other.x , y * other.y, z * other.z,w * other.w);
	226	}
	227
	228	void mulReciprocal ( f32 s )
	229	{
	230	const f32 i = core::reciprocal ( s );
	231	x = (f32) ( x * i );
	232	y = (f32) ( y * i );
	233	z = (f32) ( z * i );
	234	w = (f32) ( w * i );
	235	}
	236
	237	void mul ( const f32 s )
	238	{
	239	x *= s;
	240	y *= s;
	241	z *= s;
	242	w *= s;
	243	}
	244
	245	/*
	246	void operator*=(f32 s)
	247	{
	248	x *= s;
	249	y *= s;
	250	z *= s;
	251	w *= s;
	252	}
	253	*/
	254	void operator*=(const sVec4 &other)
	255	{
	256	x *= other.x;
	257	y *= other.y;
	258	z *= other.z;
	259	w *= other.w;
	260	}
	261
	262	void operator=(const sVec4& other)
	263	{
	264	x = other.x;
	265	y = other.y;
	266	z = other.z;
	267	w = other.w;
	268	}
	269	};
	270
	271	struct sVec3
	272	{
	273	union
	274	{
	275	struct { f32 r, g, b; };
	276	struct { f32 x, y, z; };
	277	};
	278
	279
	280	sVec3 () {}
	281	sVec3 ( f32 _x, f32 _y, f32 _z )
	282	: r ( _x ), g ( _y ), b( _z ) {}
	283
	284	sVec3 ( const sVec4 &v )
	285	: r ( v.x ), g ( v.y ), b( v.z ) {}
	286
	287	void set ( f32 _r, f32 _g, f32 _b )
	288	{
	289	r = _r;
	290	g = _g;
	291	b = _b;
	292	}
	293
	294	void setR8G8B8 ( u32 argb )
	295	{
	296	r = ( ( argb & 0x00FF0000 ) >> 16 ) * ( 1.f / 255.f );
	297	g = ( ( argb & 0x0000FF00 ) >> 8 ) * ( 1.f / 255.f );
	298	b = ( ( argb & 0x000000FF ) ) * ( 1.f / 255.f );
	299	}
	300
	301	void setColorf ( const video::SColorf & color )
	302	{
	303	r = color.r;
	304	g = color.g;
	305	b = color.b;
	306	}
	307
	308	void add (const sVec3& other)
	309	{
	310	r += other.r;
	311	g += other.g;
	312	b += other.b;
	313	}
	314
	315	void mulAdd(const sVec3& other, const f32 v)
	316	{
	317	r += other.r * v;
	318	g += other.g * v;
	319	b += other.b * v;
	320	}
	321
	322	void mulAdd(const sVec3& v0, const sVec3& v1)
	323	{
	324	r += v0.r * v1.r;
	325	g += v0.g * v1.g;
	326	b += v0.b * v1.b;
	327	}
	328
	329	void saturate ( sVec4 &dest, u32 argb )
	330	{
	331	dest.x = ( ( argb & 0xFF000000 ) >> 24 ) * ( 1.f / 255.f );
	332	dest.y = core::min_ ( r, 1.f );
	333	dest.z = core::min_ ( g, 1.f );
	334	dest.w = core::min_ ( b, 1.f );
	335	}
	336
	337	// f = a * t + b * ( 1 - t )
	338	void interpolate(const sVec3& v0, const sVec3& v1, const f32 t)
	339	{
	340	r = v1.r + ( ( v0.r - v1.r ) * t );
	341	g = v1.g + ( ( v0.g - v1.g ) * t );
	342	b = v1.b + ( ( v0.b - v1.b ) * t );
	343	}
	344
	345	sVec3 operator-(const sVec3& other) const
	346	{
	347	return sVec3(r - other.r, b - other.b, g - other.g);
	348	}
	349
	350	sVec3 operator+(const sVec3& other) const
	351	{
	352	return sVec3(r + other.r, g + other.g, b + other.b);
	353	}
	354
	355	sVec3 operator*(const f32 s) const
	356	{
	357	return sVec3(r * s , g * s, b * s);
	358	}
	359
	360	sVec3 operator/(const f32 s) const
	361	{
	362	f32 inv = 1.f / s;
	363	return sVec3(r * inv , g * inv, b * inv);
	364	}
	365
	366	sVec3 operator*(const sVec3 &other) const
	367	{
	368	return sVec3(r * other.r , b * other.b, g * other.g);
	369	}
	370
	371	void operator+=(const sVec3& other)
	372	{
	373	r += other.r;
	374	g += other.g;
	375	b += other.b;
	376	}
	377
	378	void setLength ( f32 len )
	379	{
	380	const f32 l = len * core::reciprocal_squareroot ( r * r + g * g + b * b );
	381
	382	r *= l;
	383	g *= l;
	384	b *= l;
	385	}
	386
	387	};
	388
	389
	390
	391	inline void sCompressedVec4::setVec4 ( const sVec4 & v )
	392	{
	393	argb = core::floor32 ( v.x * 255.f ) << 24 \|
	394	core::floor32 ( v.y * 255.f ) << 16 \|
	395	core::floor32 ( v.z * 255.f ) << 8 \|
	396	core::floor32 ( v.w * 255.f );
	397	}
	398
	399
	400	enum e4DVertexFlag
	401	{
	402	VERTEX4D_INSIDE = 0x0000003F,
	403	VERTEX4D_CLIPMASK = 0x0000003F,
	404	VERTEX4D_PROJECTED = 0x00000100,
	405
	406	VERTEX4D_FORMAT_MASK = 0xFFFF0000,
	407
	408	VERTEX4D_FORMAT_MASK_TEXTURE = 0x000F0000,
	409	VERTEX4D_FORMAT_TEXTURE_1 = 0x00010000,
	410	VERTEX4D_FORMAT_TEXTURE_2 = 0x00020000,
	411	VERTEX4D_FORMAT_TEXTURE_3 = 0x00030000,
	412	VERTEX4D_FORMAT_TEXTURE_4 = 0x00040000,
	413
	414	VERTEX4D_FORMAT_MASK_COLOR = 0x00F00000,
	415	VERTEX4D_FORMAT_COLOR_1 = 0x00100000,
	416	VERTEX4D_FORMAT_COLOR_2 = 0x00200000,
	417
	418	VERTEX4D_FORMAT_MASK_BUMP = 0x0F000000,
	419	VERTEX4D_FORMAT_BUMP_DOT3 = 0x01000000,
	420
	421	};
	422
	423	const u32 MATERIAL_MAX_COLORS = 1;
	424	const u32 BURNING_MATERIAL_MAX_TEXTURES = 2;
	425	const u32 BURNING_MATERIAL_MAX_TANGENT = 1;
	426
	427	// dummy Vertex. used for calculation vertex memory size
	428	struct s4DVertex_proxy
	429	{
	430	u32 flag;
	431	sVec4 Pos;
	432	sVec2 Tex[BURNING_MATERIAL_MAX_TEXTURES];
	433
	434	#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
	435	sVec4 Color[MATERIAL_MAX_COLORS];
	436	#endif
	437
	438	sVec3 LightTangent[BURNING_MATERIAL_MAX_TANGENT];
	439
	440	};
	441
	442	#define SIZEOF_SVERTEX 64
	443	#define SIZEOF_SVERTEX_LOG2 6
	444
	445	/*!
	446	Internal BurningVideo Vertex
	447	*/
	448	struct s4DVertex
	449	{
	450	u32 flag;
	451
	452	sVec4 Pos;
	453	sVec2 Tex[ BURNING_MATERIAL_MAX_TEXTURES ];
	454
	455	#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
	456	sVec4 Color[ MATERIAL_MAX_COLORS ];
	457	#endif
	458
	459	sVec3 LightTangent[BURNING_MATERIAL_MAX_TANGENT];
	460
	461	//u8 fill [ SIZEOF_SVERTEX - sizeof (s4DVertex_proxy) ];
	462
	463	// f = a * t + b * ( 1 - t )
	464	void interpolate(const s4DVertex& b, const s4DVertex& a, const f32 t)
	465	{
	466	u32 i;
	467	u32 size;
	468
	469	Pos.interpolate ( a.Pos, b.Pos, t );
	470
	471	#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
	472	size = (flag & VERTEX4D_FORMAT_MASK_COLOR) >> 20;
	473	for ( i = 0; i!= size; ++i )
	474	{
	475	Color[i].interpolate ( a.Color[i], b.Color[i], t );
	476	}
	477	#endif
	478
	479	size = (flag & VERTEX4D_FORMAT_MASK_TEXTURE) >> 16;
	480	for ( i = 0; i!= size; ++i )
	481	{
	482	Tex[i].interpolate ( a.Tex[i], b.Tex[i], t );
	483	}
	484
	485	size = (flag & VERTEX4D_FORMAT_MASK_BUMP) >> 24;
	486	for ( i = 0; i!= size; ++i )
	487	{
	488	LightTangent[i].interpolate ( a.LightTangent[i], b.LightTangent[i], t );
	489	}
	490
	491	}
	492	};
	493
	494	// ----------------- Vertex Cache ---------------------------
	495
	496	struct SAlignedVertex
	497	{
	498	SAlignedVertex ( u32 element, u32 aligned )
	499	: ElementSize ( element )
	500	{
	501	u32 byteSize = (ElementSize << SIZEOF_SVERTEX_LOG2 ) + aligned;
	502	mem = new u8 [ byteSize ];
	503	data = (s4DVertex*) mem;
	504	}
	505
	506	virtual ~SAlignedVertex ()
	507	{
	508	delete [] mem;
	509	}
	510
	511	s4DVertex *data;
	512	u8 *mem;
	513	u32 ElementSize;
	514	};
	515
	516
	517	// hold info for different Vertex Types
	518	struct SVSize
	519	{
	520	u32 Format;
	521	u32 Pitch;
	522	u32 TexSize;
	523	};
	524
	525
	526	// a cache info
	527	struct SCacheInfo
	528	{
	529	u32 index;
	530	u32 hit;
	531	};
	532
	533	#define VERTEXCACHE_ELEMENT 16
	534	#define VERTEXCACHE_MISS 0xFFFFFFFF
	535	struct SVertexCache
	536	{
	537	SVertexCache (): mem ( VERTEXCACHE_ELEMENT * 2, 128 ) {}
	538
	539	SCacheInfo info[VERTEXCACHE_ELEMENT];
	540
	541
	542	// Transformed and lite, clipping state
	543	// + Clipped, Projected
	544	SAlignedVertex mem;
	545
	546	// source
	547	const void* vertices;
	548	u32 vertexCount;
	549
	550	const void* indices;
	551	u32 indexCount;
	552	u32 indicesIndex;
	553
	554	u32 indicesRun;
	555
	556	// primitives consist of x vertices
	557	u32 primitivePitch;
	558
	559	u32 vType; //E_VERTEX_TYPE
	560	u32 pType; //scene::E_PRIMITIVE_TYPE
	561	u32 iType; //E_INDEX_TYPE iType
	562
	563	};
	564
	565
	566	// swap 2 pointer
	567	REALINLINE void swapVertexPointer(const s4DVertex v1, const s4DVertex v2)
	568	{
	569	const s4DVertex* b = *v1;
	570	v1 = v2;
	571	*v2 = b;
	572	}
	573
	574
	575	// ------------------------ Internal Scanline Rasterizer -----------------------------
	576
	577
	578
	579	// internal scan convert
	580	struct sScanConvertData
	581	{
	582	u8 left; // major edge left/right
	583	u8 right; // !left
	584
	585	f32 invDeltaY[3]; // inverse edge delta y
	586
	587	f32 x[2]; // x coordinate
	588	f32 slopeX[2]; // x slope along edges
	589
	590	#if defined ( SOFTWARE_DRIVER_2_USE_WBUFFER ) \|\| defined ( SOFTWARE_DRIVER_2_PERSPECTIVE_CORRECT )
	591	f32 w[2]; // w coordinate
	592	fp24 slopeW[2]; // w slope along edges
	593	#else
	594	f32 z[2]; // z coordinate
	595	f32 slopeZ[2]; // z slope along edges
	596	#endif
	597
	598	sVec4 c[MATERIAL_MAX_COLORS][2]; // color
	599	sVec4 slopeC[MATERIAL_MAX_COLORS][2]; // color slope along edges
	600
	601	sVec2 t[BURNING_MATERIAL_MAX_TEXTURES][2]; // texture
	602	sVec2 slopeT[BURNING_MATERIAL_MAX_TEXTURES][2]; // texture slope along edges
	603
	604	sVec3 l[BURNING_MATERIAL_MAX_TANGENT][2]; // Light Tangent
	605	sVec3 slopeL[BURNING_MATERIAL_MAX_TEXTURES][2]; // tanget slope along edges
	606	};
	607
	608	// passed to scan Line
	609	struct sScanLineData
	610	{
	611	s32 y; // y position of scanline
	612	f32 x[2]; // x start, x end of scanline
	613
	614	#if defined ( SOFTWARE_DRIVER_2_USE_WBUFFER ) \|\| defined ( SOFTWARE_DRIVER_2_PERSPECTIVE_CORRECT )
	615	f32 w[2]; // w start, w end of scanline
	616	#else
	617	f32 z[2]; // z start, z end of scanline
	618	#endif
	619
	620	#ifdef SOFTWARE_DRIVER_2_USE_VERTEX_COLOR
	621	sVec4 c[MATERIAL_MAX_COLORS][2]; // color start, color end of scanline
	622	#endif
	623
	624	sVec2 t[BURNING_MATERIAL_MAX_TEXTURES][2]; // texture start, texture end of scanline
	625	sVec3 l[BURNING_MATERIAL_MAX_TANGENT][2]; // Light Tangent start, end
	626	};
	627
	628	// passed to pixel Shader
	629	struct sPixelShaderData
	630	{
	631	tVideoSample *dst;
	632	fp24 *z;
	633
	634	s32 xStart;
	635	s32 xEnd;
	636	s32 dx;
	637	s32 i;
	638	};
	639
	640	/*
	641	load a color value
	642	*/
	643	inline void getTexel_plain2 ( tFixPoint &r, tFixPoint &g, tFixPoint &b,
	644	const sVec4 &v
	645	)
	646	{
	647	r = tofix ( v.y );
	648	g = tofix ( v.z );
	649	b = tofix ( v.w );
	650	}
	651
	652	/*
	653	load a color value
	654	*/
	655	inline void getSample_color ( tFixPoint &a, tFixPoint &r, tFixPoint &g, tFixPoint &b,
	656	const sVec4 &v
	657	)
	658	{
	659	a = tofix ( v.x );
	660	r = tofix ( v.y, COLOR_MAX * FIX_POINT_F32_MUL);
	661	g = tofix ( v.z, COLOR_MAX * FIX_POINT_F32_MUL);
	662	b = tofix ( v.w, COLOR_MAX * FIX_POINT_F32_MUL);
	663	}
	664
	665	/*
	666	load a color value
	667	*/
	668	inline void getSample_color ( tFixPoint &r, tFixPoint &g, tFixPoint &b,const sVec4 &v )
	669	{
	670	r = tofix ( v.y, COLOR_MAX * FIX_POINT_F32_MUL);
	671	g = tofix ( v.z, COLOR_MAX * FIX_POINT_F32_MUL);
	672	b = tofix ( v.w, COLOR_MAX * FIX_POINT_F32_MUL);
	673	}
	674
	675	/*
	676	load a color value
	677	*/
	678	inline void getSample_color ( tFixPoint &r, tFixPoint &g, tFixPoint &b,
	679	const sVec4 &v, const f32 mulby )
	680	{
	681	r = tofix ( v.y, mulby);
	682	g = tofix ( v.z, mulby);
	683	b = tofix ( v.w, mulby);
	684	}
	685
	686
	687
	688	}
	689
	690	}
	691
	692	#endif
	693