dolphin/Source/Plugins/Plugin_VideoDX9/Src/VertexLoader.cpp

400 lines
13 KiB
C++
Raw Normal View History

// Copyright (C) 2003-2008 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 <stdio.h>
#include "x64Emitter.h"
#include "Common.h"
#include "LookUpTables.h"
#include "Profiler.h"
#include "VertexHandler.h"
#include "VertexLoader.h"
#include "XFStructs.h"
#include "BPStructs.h"
#include "DataReader.h"
#include "DecodedVArray.h"
//these don't need to be saved
float posScale;
float tcScale[8];
int tcElements[8];
int tcFormat[8];
int colElements[2];
float tcScaleU[8];
float tcScaleV[8];
int tcIndex;
int colIndex;
u32 addr;
DecodedVArray *varray;
int ComputeVertexSize(u32 comp)
{
int size = 0;
if (comp & VertexLoader::VB_HAS_POSMTXIDX)
size += 4;
if (comp & (VertexLoader::VB_HAS_TEXMTXIDX0 | VertexLoader::VB_HAS_TEXMTXIDX1 | VertexLoader::VB_HAS_TEXMTXIDX2 | VertexLoader::VB_HAS_TEXMTXIDX3))
size += 4;
if (comp & (VertexLoader::VB_HAS_TEXMTXIDX4 | VertexLoader::VB_HAS_TEXMTXIDX5 | VertexLoader::VB_HAS_TEXMTXIDX6 | VertexLoader::VB_HAS_TEXMTXIDX7))
size += 4;
if (comp & VertexLoader::VB_HAS_NRM0)
size += 4;
if (comp & (VertexLoader::VB_HAS_NRM1 | VertexLoader::VB_HAS_NRM2)) //combine into single check for speed
size += 8;
if (comp & VertexLoader::VB_HAS_COL0)
size += 4;
if (comp & VertexLoader::VB_HAS_COL1)
size += 4;
for (int i = 0; i < 8; i++)
if (comp & (VertexLoader::VB_HAS_UV0 << i))
size += 8;
return size;
}
void VertexLoader::SetVArray(DecodedVArray *_varray)
{
varray = _varray;
}
#include "VertexLoader_MtxIndex.h"
#include "VertexLoader_Position.h"
#include "VertexLoader_Normal.h"
#include "VertexLoader_Color.h"
#include "VertexLoader_TextCoord.h"
VertexLoader g_VertexLoaders[8];
TVtxDesc VertexLoader::m_VtxDesc;
bool VertexLoader::m_DescDirty = true;
VertexLoader::VertexLoader()
{
m_numPipelineStates = 0;
m_VertexSize = 0;
m_AttrDirty = true;
VertexLoader_Normal::Init();
}
VertexLoader::~VertexLoader()
{
}
void VertexLoader::Setup()
{
if (!m_AttrDirty && !m_DescDirty)
return;
DVSTARTPROFILE();
// Reset pipeline
m_VertexSize = 0;
m_numPipelineStates = 0;
m_components = 0;
// Position Matrix Index
if (m_VtxDesc.PosMatIdx)
{
m_PipelineStates[m_numPipelineStates++] = PosMtx_ReadDirect_UByte;
m_VertexSize += 1;
m_components |= VB_HAS_POSMTXIDX;
}
// Texture matrix indices
if (m_VtxDesc.Tex0MatIdx) {m_components|=VB_HAS_TEXMTXIDX0; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex1MatIdx) {m_components|=VB_HAS_TEXMTXIDX1; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex2MatIdx) {m_components|=VB_HAS_TEXMTXIDX2; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex3MatIdx) {m_components|=VB_HAS_TEXMTXIDX3; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex4MatIdx) {m_components|=VB_HAS_TEXMTXIDX4; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex5MatIdx) {m_components|=VB_HAS_TEXMTXIDX5; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex6MatIdx) {m_components|=VB_HAS_TEXMTXIDX6; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
if (m_VtxDesc.Tex7MatIdx) {m_components|=VB_HAS_TEXMTXIDX7; WriteCall(TexMtx_ReadDirect_UByte); m_VertexSize+=1;}
// Position
switch (m_VtxDesc.Position)
{
case NOT_PRESENT: {_assert_msg_(0,"Vertex descriptor without position!","WTF?");} break;
case DIRECT:
{
int SizePro = 0;
switch (m_VtxAttr.PosFormat)
{
case FORMAT_UBYTE: SizePro=1; WriteCall(Pos_ReadDirect_UByte); break;
case FORMAT_BYTE: SizePro=1; WriteCall(Pos_ReadDirect_Byte); break;
case FORMAT_USHORT: SizePro=2; WriteCall(Pos_ReadDirect_UShort); break;
case FORMAT_SHORT: SizePro=2; WriteCall(Pos_ReadDirect_Short); break;
case FORMAT_FLOAT: SizePro=4; WriteCall(Pos_ReadDirect_Float); break;
default: _assert_(0); break;
}
if (m_VtxAttr.PosElements == 1)
m_VertexSize += SizePro * 3;
else
m_VertexSize += SizePro * 2;
}
break;
case INDEX8:
m_VertexSize+=1;
switch (m_VtxAttr.PosFormat)
{
case FORMAT_UBYTE: WriteCall(Pos_ReadIndex8_UByte); break; //WTF?
case FORMAT_BYTE: WriteCall(Pos_ReadIndex8_Byte); break;
case FORMAT_USHORT: WriteCall(Pos_ReadIndex8_UShort); break;
case FORMAT_SHORT: WriteCall(Pos_ReadIndex8_Short); break;
case FORMAT_FLOAT: WriteCall(Pos_ReadIndex8_Float); break;
default: _assert_(0); break;
}
break;
case INDEX16:
m_VertexSize+=2;
switch (m_VtxAttr.PosFormat)
{
case FORMAT_UBYTE: WriteCall(Pos_ReadIndex16_UByte); break;
case FORMAT_BYTE: WriteCall(Pos_ReadIndex16_Byte); break;
case FORMAT_USHORT: WriteCall(Pos_ReadIndex16_UShort); break;
case FORMAT_SHORT: WriteCall(Pos_ReadIndex16_Short); break;
case FORMAT_FLOAT: WriteCall(Pos_ReadIndex16_Float); break;
default: _assert_(0); break;
}
break;
}
// Normals
if (m_VtxDesc.Normal != NOT_PRESENT)
{
VertexLoader_Normal::index3 = m_VtxAttr.NormalIndex3 ? true : false;
unsigned int uSize = VertexLoader_Normal::GetSize(m_VtxDesc.Normal, m_VtxAttr.NormalFormat, m_VtxAttr.NormalElements);
TPipelineFunction pFunc = VertexLoader_Normal::GetFunction(m_VtxDesc.Normal, m_VtxAttr.NormalFormat, m_VtxAttr.NormalElements);
if (pFunc == 0)
{
char temp[256];
sprintf(temp,"%i %i %i", m_VtxDesc.Normal, m_VtxAttr.NormalFormat, m_VtxAttr.NormalElements);
MessageBox(0,"VertexLoader_Normal::GetFunction returned zero!",temp,0);
}
WriteCall(pFunc);
m_VertexSize += uSize;
int m_numNormals = (m_VtxAttr.NormalElements == 1) ? NRM_THREE : NRM_ONE;
m_components |= VB_HAS_NRM0;
if (m_numNormals == NRM_THREE)
m_components |= VB_HAS_NRM1 | VB_HAS_NRM2;
}
// Colors
int col[2] = {m_VtxDesc.Color0, m_VtxDesc.Color1};
for (int i = 0; i < 2; i++)
SetupColor(i,col[i], m_VtxAttr.color[i].Comp, m_VtxAttr.color[i].Elements);
// TextureCoord
int tc[8] = {
m_VtxDesc.Tex0Coord, m_VtxDesc.Tex1Coord, m_VtxDesc.Tex2Coord, m_VtxDesc.Tex3Coord,
m_VtxDesc.Tex4Coord, m_VtxDesc.Tex5Coord, m_VtxDesc.Tex6Coord, m_VtxDesc.Tex7Coord,
};
for (int i = 0; i < 8; i++)
SetupTexCoord(i, tc[i],
m_VtxAttr.texCoord[i].Format,
m_VtxAttr.texCoord[i].Elements,
m_VtxAttr.texCoord[i].Frac);
Compile();
//RET();
/*
char temp[256];
sprintf(temp,"Size: %08x Pos: %i Norm:%i Col0: %i Col1: %i",m_VertexSize,m_VtxDesc.Position,m_VtxDesc.Normal,m_VtxDesc.Color0,m_VtxDesc.Color1);
g_VideoInitialize.pLog(temp);
sprintf(temp,"Pos: %i Norm:%i Col0: %i Col1: %i",_VtxAttr.PosFormat,_VtxAttr.NormalFormat,_VtxAttr.color[0].Comp,_VtxAttr.color[1].Comp);
g_VideoInitialize.pLog(temp);*/
}
void VertexLoader::SetupColor(int num, int mode, int format, int elements)
{
m_components |= VB_HAS_COL0 << num;
switch (mode)
{
case NOT_PRESENT:
m_components &= ~(VB_HAS_COL0 << num);
break;
case DIRECT:
switch (format)
{
case FORMAT_16B_565: m_VertexSize+=2; WriteCall(Color_ReadDirect_16b_565); break;
case FORMAT_24B_888: m_VertexSize+=3; WriteCall(Color_ReadDirect_24b_888); break;
case FORMAT_32B_888x: m_VertexSize+=4; WriteCall(Color_ReadDirect_32b_888x); break;
case FORMAT_16B_4444: m_VertexSize+=2; WriteCall(Color_ReadDirect_16b_4444); break;
case FORMAT_24B_6666: m_VertexSize+=3; WriteCall(Color_ReadDirect_24b_6666); break;
case FORMAT_32B_8888: m_VertexSize+=4; WriteCall(Color_ReadDirect_32b_8888); break;
default: _assert_(0); break;
}
break;
case INDEX8:
switch (format)
{
case FORMAT_16B_565: WriteCall(Color_ReadIndex8_16b_565); break;
case FORMAT_24B_888: WriteCall(Color_ReadIndex8_24b_888); break;
case FORMAT_32B_888x: WriteCall(Color_ReadIndex8_32b_888x); break;
case FORMAT_16B_4444: WriteCall(Color_ReadIndex8_16b_4444); break;
case FORMAT_24B_6666: WriteCall(Color_ReadIndex8_24b_6666); break;
case FORMAT_32B_8888: WriteCall(Color_ReadIndex8_32b_8888); break;
default: _assert_(0); break;
}
m_VertexSize+=1;
break;
case INDEX16:
switch (format)
{
case FORMAT_16B_565: WriteCall(Color_ReadIndex16_16b_565); break;
case FORMAT_24B_888: WriteCall(Color_ReadIndex16_24b_888); break;
case FORMAT_32B_888x: WriteCall(Color_ReadIndex16_32b_888x); break;
case FORMAT_16B_4444: WriteCall(Color_ReadIndex16_16b_4444); break;
case FORMAT_24B_6666: WriteCall(Color_ReadIndex16_24b_6666); break;
case FORMAT_32B_8888: WriteCall(Color_ReadIndex16_32b_8888); break;
default: _assert_(0); break;
}
m_VertexSize+=2;
break;
}
}
void VertexLoader::SetupTexCoord(int num, int mode, int format, int elements, int _iFrac)
{
m_components |= VB_HAS_UV0 << num;
switch (mode)
{
case NOT_PRESENT:
m_components &= ~(VB_HAS_UV0 << num);
break;
case DIRECT:
{
int sizePro=0;
switch (format)
{
case FORMAT_UBYTE: sizePro=1; WriteCall(TexCoord_ReadDirect_UByte); break;
case FORMAT_BYTE: sizePro=1; WriteCall(TexCoord_ReadDirect_Byte); break;
case FORMAT_USHORT: sizePro=2; WriteCall(TexCoord_ReadDirect_UShort); break;
case FORMAT_SHORT: sizePro=2; WriteCall(TexCoord_ReadDirect_Short); break;
case FORMAT_FLOAT: sizePro=4; WriteCall(TexCoord_ReadDirect_Float); break;
default: _assert_(0); break;
}
m_VertexSize += sizePro * (elements?2:1);
}
break;
case INDEX8:
switch (format)
{
case FORMAT_UBYTE: WriteCall(TexCoord_ReadIndex8_UByte); break;
case FORMAT_BYTE: WriteCall(TexCoord_ReadIndex8_Byte); break;
case FORMAT_USHORT: WriteCall(TexCoord_ReadIndex8_UShort); break;
case FORMAT_SHORT: WriteCall(TexCoord_ReadIndex8_Short); break;
case FORMAT_FLOAT: WriteCall(TexCoord_ReadIndex8_Float); break;
default: _assert_(0); break;
}
m_VertexSize+=1;
break;
case INDEX16:
switch (format)
{
case FORMAT_UBYTE: WriteCall(TexCoord_ReadIndex16_UByte); break;
case FORMAT_BYTE: WriteCall(TexCoord_ReadIndex16_Byte); break;
case FORMAT_USHORT: WriteCall(TexCoord_ReadIndex16_UShort); break;
case FORMAT_SHORT: WriteCall(TexCoord_ReadIndex16_Short); break;
case FORMAT_FLOAT: WriteCall(TexCoord_ReadIndex16_Float); break;
default: _assert_(0);
}
m_VertexSize+=2;
break;
}
}
void VertexLoader::WriteCall(TPipelineFunction func)
{
m_PipelineStates[m_numPipelineStates++] = func;
}
using namespace Gen;
// See comment in RunVertices
void VertexLoader::Compile()
{
return;
/* Gen::SetCodePtr(m_compiledCode);
//INT3();
Gen::Util::EmitPrologue(0);
const u8 *loopStart = Gen::GetCodePtr();
MOV(32, M(&tcIndex), Imm32(0));
MOV(32, M(&colIndex), Imm32(0));
for (int i = 0; i < m_numPipelineStates; i++)
{
ABI_AlignStack(1 * 4);
PUSH(32, Imm32((u32)&m_VtxAttr));
CALL(m_PipelineStates[i]);
ABI_RestoreStack(1 * 4);
}
ADD(32, M(&varray->count), Imm8(1));
SUB(32, M(&this->m_counter), Imm8(1));
J_CC(CC_NZ, loopStart, true);
// Epilogue
Gen::Util::EmitEpilogue(0);
if (Gen::GetCodePtr()-(u8*)m_compiledCode > sizeof(m_compiledCode))
{
Crash();
}*/
}
void VertexLoader::PrepareRun()
{
posScale = shiftLookup[m_VtxAttr.PosFrac];
for (int i = 0; i < 8; i++)
{
tcScaleU[i] = shiftLookup[m_VtxAttr.texCoord[i].Frac];
tcScaleV[i] = shiftLookup[m_VtxAttr.texCoord[i].Frac];
tcElements[i] = m_VtxAttr.texCoord[i].Elements;
tcFormat[i] = m_VtxAttr.texCoord[i].Format;
}
for (int i = 0; i < 2; i++)
colElements[i] = m_VtxAttr.color[i].Elements;
varray->SetComponents(m_components);
}
void VertexLoader::RunVertices(int count)
{
DVSTARTPROFILE();
for (int v = 0; v < count; v++)
{
tcIndex = 0;
colIndex = 0;
s_texmtxread = 0;
for (int i = 0; i < m_numPipelineStates; i++)
{
m_PipelineStates[i](&m_VtxAttr);
}
varray->Next();
}
/*
This is not the bottleneck ATM, so compiling etc doesn't really help.
At least not when all we do is compile it to a list of function calls.
Should help more when we inline, but this requires the new vertex format.
Maybe later, and with smarter caching.
if (count)
{
this->m_counter = count;
((void (*)())((void*)&m_compiledCode[0]))();
}*/
}