dolphin/Source/Core/VideoCommon/Src/VertexShaderGen.cpp

// Copyright (C) 2003 Dolphin Project.

// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, version 2.0.

// This program 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 2.0 for more details.

// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/

// Official SVN repository and contact information can be found at
// http://code.google.com/p/dolphin-emu/

#include <math.h>
#include <locale.h>

#include "NativeVertexFormat.h"

#include "BPMemory.h"
#include "CPMemory.h"
#include "LightingShaderGen.h"
#include "VertexShaderGen.h"
#include "VideoConfig.h"

VERTEXSHADERUID  last_vertex_shader_uid;

// Mash together all the inputs that contribute to the code of a generated vertex shader into
// a unique identifier, basically containing all the bits. Yup, it's a lot ....
void GetVertexShaderId(VERTEXSHADERUID *uid, u32 components)
{
	uid->values[0] = components |
		(xfregs.numTexGen.numTexGens << 23) |
		(xfregs.numChan.numColorChans << 27) |
		(xfregs.dualTexTrans.enabled << 29);

	for (int i = 0; i < 2; ++i) {
		uid->values[1+i] = xfregs.color[i].enablelighting ?
			(u32)xfregs.color[i].hex :
			(u32)xfregs.color[i].matsource;
		uid->values[1+i] |= (xfregs.alpha[i].enablelighting ?
			(u32)xfregs.alpha[i].hex :
			(u32)xfregs.alpha[i].matsource) << 15;
	}
	uid->values[2] |= (g_ActiveConfig.bEnablePixelLigting && g_ActiveConfig.backend_info.bSupportsPixelLighting) << 31;
	u32 *pcurvalue = &uid->values[3];
	for (unsigned int i = 0; i < xfregs.numTexGen.numTexGens; ++i) {
		TexMtxInfo tinfo = xfregs.texMtxInfo[i];
		if (tinfo.texgentype != XF_TEXGEN_EMBOSS_MAP)
			tinfo.hex &= 0x7ff;
		if (tinfo.texgentype != XF_TEXGEN_REGULAR)
			tinfo.projection = 0;

		u32 val = ((tinfo.hex >> 1) & 0x1ffff);
		if (xfregs.dualTexTrans.enabled && tinfo.texgentype == XF_TEXGEN_REGULAR) {
			// rewrite normalization and post index
			val |= ((u32)xfregs.postMtxInfo[i].index << 17) | ((u32)xfregs.postMtxInfo[i].normalize << 23);
		}

		switch (i & 3) {
			case 0: pcurvalue[0] |= val; break;
			case 1: pcurvalue[0] |= val << 24; pcurvalue[1] = val >> 8; ++pcurvalue; break;
			case 2: pcurvalue[0] |= val << 16; pcurvalue[1] = val >> 16; ++pcurvalue; break;
			case 3: pcurvalue[0] |= val << 8; ++pcurvalue; break;
		}
	}
}

static char text[16384];

#define WRITE p+=sprintf

const char *GenerateVertexShaderCode(u32 components, API_TYPE api_type)
{
	setlocale(LC_NUMERIC, "C"); // Reset locale for compilation
	text[sizeof(text) - 1] = 0x7C;  // canary

	_assert_(bpmem.genMode.numtexgens == xfregs.numTexGen.numTexGens);
	_assert_(bpmem.genMode.numcolchans == xfregs.numChan.numColorChans);
	
	bool is_d3d = (api_type == API_D3D9 || api_type == API_D3D11);
	u32 lightMask = 0;
	if (xfregs.numChan.numColorChans > 0)
		lightMask |= xfregs.color[0].GetFullLightMask() | xfregs.alpha[0].GetFullLightMask();
	if (xfregs.numChan.numColorChans > 1)
		lightMask |= xfregs.color[1].GetFullLightMask() | xfregs.alpha[1].GetFullLightMask();

	char *p = text;
	WRITE(p, "//Vertex Shader: comp:%x, \n", components);
	WRITE(p, "typedef struct { float4 T0, T1, T2; float4 N0, N1, N2; } s_"I_POSNORMALMATRIX";\n"
		"typedef struct { float4 t; } FLT4;\n"
		"typedef struct { FLT4 T[24]; } s_"I_TEXMATRICES";\n"
		"typedef struct { FLT4 T[64]; } s_"I_TRANSFORMMATRICES";\n"
		"typedef struct { FLT4 T[32]; } s_"I_NORMALMATRICES";\n"
		"typedef struct { FLT4 T[64]; } s_"I_POSTTRANSFORMMATRICES";\n"
		"typedef struct { float4 col; float4 cosatt; float4 distatt; float4 pos; float4 dir; } Light;\n"
		"typedef struct { Light lights[8]; } s_"I_LIGHTS";\n"
		"typedef struct { float4 C0, C1, C2, C3; } s_"I_MATERIALS";\n"
		"typedef struct { float4 T0, T1, T2, T3; } s_"I_PROJECTION";\n"
		);

	WRITE(p, "struct VS_OUTPUT {\n");
	WRITE(p, "  float4 pos : POSITION;\n");
	WRITE(p, "  float4 colors_0 : COLOR0;\n");
	WRITE(p, "  float4 colors_1 : COLOR1;\n");

	if (xfregs.numTexGen.numTexGens < 7) {
		for (unsigned int i = 0; i < xfregs.numTexGen.numTexGens; ++i)
			WRITE(p, "  float3 tex%d : TEXCOORD%d;\n", i, i);
		WRITE(p, "  float4 clipPos : TEXCOORD%d;\n", xfregs.numTexGen.numTexGens);
		if(g_ActiveConfig.bEnablePixelLigting && g_ActiveConfig.backend_info.bSupportsPixelLighting)
			WRITE(p, "  float4 Normal : TEXCOORD%d;\n", xfregs.numTexGen.numTexGens + 1);
	} else {
		// clip position is in w of first 4 texcoords
		if(g_ActiveConfig.bEnablePixelLigting && g_ActiveConfig.backend_info.bSupportsPixelLighting)
		{
			for (int i = 0; i < 8; ++i)
				WRITE(p, "  float4 tex%d : TEXCOORD%d;\n", i, i);
		}
		else
		{
			for (unsigned int i = 0; i < xfregs.numTexGen.numTexGens; ++i)
				WRITE(p, "  float%d tex%d : TEXCOORD%d;\n", i < 4 ? 4 : 3 , i, i);
		}
	}	
	WRITE(p, "};\n");

	// uniforms

	WRITE(p, "uniform s_"I_TRANSFORMMATRICES" "I_TRANSFORMMATRICES" : register(c%d);\n", C_TRANSFORMMATRICES);
	WRITE(p, "uniform s_"I_TEXMATRICES" "I_TEXMATRICES" : register(c%d);\n", C_TEXMATRICES); // also using tex matrices
	WRITE(p, "uniform s_"I_NORMALMATRICES" "I_NORMALMATRICES" : register(c%d);\n", C_NORMALMATRICES);
	WRITE(p, "uniform s_"I_POSNORMALMATRIX" "I_POSNORMALMATRIX" : register(c%d);\n", C_POSNORMALMATRIX);
	WRITE(p, "uniform s_"I_POSTTRANSFORMMATRICES" "I_POSTTRANSFORMMATRICES" : register(c%d);\n", C_POSTTRANSFORMMATRICES);
	WRITE(p, "uniform s_"I_LIGHTS" "I_LIGHTS" : register(c%d);\n", C_LIGHTS);
	WRITE(p, "uniform s_"I_MATERIALS" "I_MATERIALS" : register(c%d);\n", C_MATERIALS);
	WRITE(p, "uniform s_"I_PROJECTION" "I_PROJECTION" : register(c%d);\n", C_PROJECTION);
	WRITE(p, "uniform float4 "I_DEPTHPARAMS" : register(c%d);\n", C_DEPTHPARAMS);

	WRITE(p, "VS_OUTPUT main(\n");	
	
	// inputs
	if (components & VB_HAS_NRM0)
		WRITE(p, "  float3 rawnorm0 : NORMAL0,\n");
	if (components & VB_HAS_NRM1) {
		if (is_d3d)
			WRITE(p, "  float3 rawnorm1 : NORMAL1,\n");
		else
			WRITE(p, "  float3 rawnorm1 : ATTR%d,\n", SHADER_NORM1_ATTRIB);
	}
	if (components & VB_HAS_NRM2) {
		if (is_d3d)
			WRITE(p, "  float3 rawnorm2 : NORMAL2,\n");
		else
			WRITE(p, "  float3 rawnorm2 : ATTR%d,\n", SHADER_NORM2_ATTRIB);
	}
	if (components & VB_HAS_COL0)
		WRITE(p, "  float4 color0 : COLOR0,\n");
	if (components & VB_HAS_COL1)
		WRITE(p, "  float4 color1 : COLOR1,\n");
	for (int i = 0; i < 8; ++i) {
		u32 hastexmtx = (components & (VB_HAS_TEXMTXIDX0<<i));
		if ((components & (VB_HAS_UV0<<i)) || hastexmtx)
			WRITE(p, "  float%d tex%d : TEXCOORD%d,\n", hastexmtx ? 3 : 2, i, i);
	}
	if (components & VB_HAS_POSMTXIDX) {
		if (is_d3d)
		{
			WRITE(p, "  float4 blend_indices : BLENDINDICES,\n");
		}
		else
			WRITE(p, "  float fposmtx : ATTR%d,\n", SHADER_POSMTX_ATTRIB);
	}
	WRITE(p, "  float4 rawpos : POSITION) {\n");
	WRITE(p, "VS_OUTPUT o;\n");	
	// transforms
	if (components & VB_HAS_POSMTXIDX) {
		if (api_type == API_D3D9)
		{
			WRITE(p, "int4 indices = D3DCOLORtoUBYTE4(blend_indices);\n");
			WRITE(p, "int posmtx = indices.x;\n");
		}
		else if (api_type == API_D3D11)
		{
			WRITE(p, "int posmtx = blend_indices.x * 255.0f;\n");
		}
		else
		{
			WRITE(p, "int posmtx = fposmtx;\n");
		}

		WRITE(p, "float4 pos = float4(dot("I_TRANSFORMMATRICES".T[posmtx].t, rawpos), dot("I_TRANSFORMMATRICES".T[posmtx+1].t, rawpos), dot("I_TRANSFORMMATRICES".T[posmtx+2].t, rawpos), 1);\n");		

		if (components & VB_HAS_NRMALL) {
			WRITE(p, "int normidx = posmtx >= 32 ? (posmtx-32) : posmtx;\n");
			WRITE(p, "float3 N0 = "I_NORMALMATRICES".T[normidx].t.xyz, N1 = "I_NORMALMATRICES".T[normidx+1].t.xyz, N2 = "I_NORMALMATRICES".T[normidx+2].t.xyz;\n");
		}

		if (components & VB_HAS_NRM0)
			WRITE(p, "float3 _norm0 = normalize(float3(dot(N0, rawnorm0), dot(N1, rawnorm0), dot(N2, rawnorm0)));\n");
		if (components & VB_HAS_NRM1)
			WRITE(p, "float3 _norm1 = float3(dot(N0, rawnorm1), dot(N1, rawnorm1), dot(N2, rawnorm1));\n");
		if (components & VB_HAS_NRM2)
			WRITE(p, "float3 _norm2 = float3(dot(N0, rawnorm2), dot(N1, rawnorm2), dot(N2, rawnorm2));\n");
	}
	else
	{
		WRITE(p, "float4 pos = float4(dot("I_POSNORMALMATRIX".T0, rawpos), dot("I_POSNORMALMATRIX".T1, rawpos), dot("I_POSNORMALMATRIX".T2, rawpos), 1.0f);\n");
		if (components & VB_HAS_NRM0)
			WRITE(p, "float3 _norm0 = normalize(float3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm0), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm0), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm0)));\n");
		if (components & VB_HAS_NRM1)
			WRITE(p, "float3 _norm1 = float3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm1), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm1), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm1));\n");
		if (components & VB_HAS_NRM2)
			WRITE(p, "float3 _norm2 = float3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm2), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm2), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm2));\n");
	}

	if (!(components & VB_HAS_NRM0))
		WRITE(p, "float3 _norm0 = float3(0.0f, 0.0f, 0.0f);\n");

	

	WRITE(p, "o.pos = float4(dot("I_PROJECTION".T0, pos), dot("I_PROJECTION".T1, pos), dot("I_PROJECTION".T2, pos), dot("I_PROJECTION".T3, pos));\n");

	WRITE(p, "float4 mat, lacc;\n"
	"float3 ldir, h;\n"
	"float dist, dist2, attn;\n");

	if(xfregs.numChan.numColorChans == 0)
	{
		if (components & VB_HAS_COL0)
			WRITE(p, "o.colors_0 = color0;\n");
		else
			WRITE(p, "o.colors_0 = float4(1.0f, 1.0f, 1.0f, 1.0f);\n");		
	}
	
	p = GenerateLightingShader(p, components, I_MATERIALS, I_LIGHTS, "color", "o.colors_");

	if(xfregs.numChan.numColorChans < 2)
	{
		if (components & VB_HAS_COL1)
			WRITE(p, "o.colors_1 = color1;\n");
		else
			WRITE(p, "o.colors_1 = o.colors_0;\n");		
	}
	// special case if only pos and tex coord 0 and tex coord input is AB11
	// donko - this has caused problems in some games. removed for now.
	bool texGenSpecialCase = false;
	/*bool texGenSpecialCase =
		((g_VtxDesc.Hex & 0x60600L) == g_VtxDesc.Hex) && // only pos and tex coord 0
		(g_VtxDesc.Tex0Coord != NOT_PRESENT) &&
		(xfregs.texcoords[0].texmtxinfo.inputform == XF_TEXINPUT_AB11);
		*/

	// transform texcoords
	WRITE(p, "float4 coord = float4(0.0f, 0.0f, 1.0f, 1.0f);\n");
	for (unsigned int i = 0; i < xfregs.numTexGen.numTexGens; ++i) {
		TexMtxInfo& texinfo = xfregs.texMtxInfo[i];

		WRITE(p, "{\n");
		WRITE(p, "coord = float4(0.0f, 0.0f, 1.0f, 1.0f);\n");
		switch (texinfo.sourcerow) {
		case XF_SRCGEOM_INROW:
			_assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
			WRITE(p, "coord = rawpos;\n"); // pos.w is 1
			break;
		case XF_SRCNORMAL_INROW:
			if (components & VB_HAS_NRM0) {
				_assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
				WRITE(p, "coord = float4(rawnorm0.xyz, 1.0f);\n");
			}
			break;
		case XF_SRCCOLORS_INROW:
			_assert_( texinfo.texgentype == XF_TEXGEN_COLOR_STRGBC0 || texinfo.texgentype == XF_TEXGEN_COLOR_STRGBC1 );
			break;
		case XF_SRCBINORMAL_T_INROW:
			if (components & VB_HAS_NRM1) {
				_assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
				WRITE(p, "coord = float4(rawnorm1.xyz, 1.0f);\n");
			}
			break;
		case XF_SRCBINORMAL_B_INROW:
			if (components & VB_HAS_NRM2) {
				_assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
				WRITE(p, "coord = float4(rawnorm2.xyz, 1.0f);\n");
			}
			break;
		default:
			_assert_(texinfo.sourcerow <= XF_SRCTEX7_INROW);
			if (components & (VB_HAS_UV0<<(texinfo.sourcerow - XF_SRCTEX0_INROW)) )
				WRITE(p, "coord = float4(tex%d.x, tex%d.y, 1.0f, 1.0f);\n", texinfo.sourcerow - XF_SRCTEX0_INROW, texinfo.sourcerow - XF_SRCTEX0_INROW);
			break;
		}

		// first transformation
		switch (texinfo.texgentype) {
			case XF_TEXGEN_EMBOSS_MAP: // calculate tex coords into bump map

				if (components & (VB_HAS_NRM1|VB_HAS_NRM2)) {
					// transform the light dir into tangent space
					WRITE(p, "ldir = normalize("I_LIGHTS".lights[%d].pos.xyz - pos.xyz);\n", texinfo.embosslightshift);
					WRITE(p, "o.tex%d.xyz = o.tex%d.xyz + float3(dot(ldir, _norm1), dot(ldir, _norm2), 0.0f);\n", i, texinfo.embosssourceshift);
				}
				else
				{
					_assert_(0); // should have normals
					WRITE(p, "o.tex%d.xyz = o.tex%d.xyz;\n", i, texinfo.embosssourceshift);
				}

				break;
			case XF_TEXGEN_COLOR_STRGBC0:
				_assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);
				WRITE(p, "o.tex%d.xyz = float3(o.colors_0.x, o.colors_0.y, 1);\n", i);
				break;
			case XF_TEXGEN_COLOR_STRGBC1:
				_assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);
				WRITE(p, "o.tex%d.xyz = float3(o.colors_1.x, o.colors_1.y, 1);\n", i);
				break;
			case XF_TEXGEN_REGULAR:
			default:
				if (components & (VB_HAS_TEXMTXIDX0<<i)) {
					if (texinfo.projection == XF_TEXPROJ_STQ)
						WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+1].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+2].t));\n", i, i, i, i);
					else {
						WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+1].t), 1);\n", i, i, i);
					}
				}
				else {
					if (texinfo.projection == XF_TEXPROJ_STQ)
						WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t));\n", i, 3*i, 3*i+1, 3*i+2);
					else
						WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t), 1);\n", i, 3*i, 3*i+1);
				}
				break;
		}

		if (xfregs.dualTexTrans.enabled && texinfo.texgentype == XF_TEXGEN_REGULAR) { // only works for regular tex gen types?
			const PostMtxInfo& postInfo = xfregs.postMtxInfo[i];

			int postidx = postInfo.index;
			WRITE(p, "float4 P0 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n"
				"float4 P1 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n"
				"float4 P2 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n",
				postidx&0x3f, (postidx+1)&0x3f, (postidx+2)&0x3f);

			if (texGenSpecialCase) {
				// no normalization
				// q of input is 1
				// q of output is unknown

				// multiply by postmatrix
				WRITE(p, "o.tex%d.xyz = float3(dot(P0.xy, o.tex%d.xy) + P0.z + P0.w, dot(P1.xy, o.tex%d.xy) + P1.z + P1.w, 0.0f);\n", i, i, i);
			}
			else
			{
				if (postInfo.normalize)
					WRITE(p, "o.tex%d.xyz = normalize(o.tex%d.xyz);\n", i, i);

				// multiply by postmatrix
				WRITE(p, "o.tex%d.xyz = float3(dot(P0.xyz, o.tex%d.xyz) + P0.w, dot(P1.xyz, o.tex%d.xyz) + P1.w, dot(P2.xyz, o.tex%d.xyz) + P2.w);\n", i, i, i, i);
			}
		}

		WRITE(p, "}\n");
	}

	// clipPos/w needs to be done in pixel shader, not here
	if (xfregs.numTexGen.numTexGens < 7) {
		WRITE(p, "o.clipPos = float4(pos.x,pos.y,o.pos.z,o.pos.w);\n");
	} else {
		WRITE(p, "o.tex0.w = pos.x;\n");
		WRITE(p, "o.tex1.w = pos.y;\n");
		WRITE(p, "o.tex2.w = o.pos.z;\n");
		WRITE(p, "o.tex3.w = o.pos.w;\n");
	}

	if(g_ActiveConfig.bEnablePixelLigting && g_ActiveConfig.backend_info.bSupportsPixelLighting)
	{
		if (xfregs.numTexGen.numTexGens < 7) {
			WRITE(p, "o.Normal = float4(_norm0.x,_norm0.y,_norm0.z,pos.z);\n");
		} else {
			WRITE(p, "o.tex4.w = _norm0.x;\n");
			WRITE(p, "o.tex5.w = _norm0.y;\n");
			WRITE(p, "o.tex6.w = _norm0.z;\n");
			if (xfregs.numTexGen.numTexGens < 8)
				WRITE(p, "o.tex7 = pos.xyzz;\n");
			else
				WRITE(p, "o.tex7.w = pos.z;\n");
		}		
		if (components & VB_HAS_COL0)
			WRITE(p, "o.colors_0 = color0;\n");		

		if (components & VB_HAS_COL1)
			WRITE(p, "o.colors_1 = color1;\n");
	}

	//write the true depth value, if the game uses depth textures pixel shaders will override with the correct values
	//if not early z culling will improve speed
	if (is_d3d) {
		WRITE(p, "o.pos.z = "I_DEPTHPARAMS".x * o.pos.w + o.pos.z * "I_DEPTHPARAMS".y;\n");
	} else {
	    // this results in a scale from -1..0 to -1..1 after perspective
	    // divide
	    WRITE(p, "o.pos.z = o.pos.w + o.pos.z * 2.0f;\n");

	    // Sonic Unleashed puts its final rendering at the near or
	    // far plane of the viewing frustrum(actually box, they use
	    // orthogonal projection for that), and we end up putting it
	    // just beyond, and the rendering gets clipped away. (The
	    // primitive gets dropped)
	    WRITE(p, "o.pos.z = o.pos.z * 1048575.0f/1048576.0f;\n");

	    // the next steps of the OGL pipeline are:
	    // (x_c,y_c,z_c,w_c) = o.pos  //switch to OGL spec terminology
	    // clipping to -w_c <= (x_c,y_c,z_c) <= w_c
	    // (x_d,y_d,z_d) = (x_c,y_c,z_c)/w_c//perspective divide
	    // z_w = (f-n)/2*z_d + (n+f)/2
	    // z_w now contains the value to go to the 0..1 depth buffer
	    
	    //trying to get the correct semantic while not using glDepthRange
	    //seems to get rather complicated
	}

	WRITE(p, "return o;\n}\n");


	if (text[sizeof(text) - 1] != 0x7C)
		PanicAlert("VertexShader generator - buffer too small, canary has been eaten!");
	setlocale(LC_NUMERIC, ""); // restore locale
	return text;
}