piwalker/dolphin
1
0
Fork 0
mirror of https://github.com/dolphin-emu/dolphin.git synced 2025-07-21 05:09:34 -06:00
Code Issues Packages Projects Releases Wiki Activity

Move out Cg shader generators to VideoCommon (hope to use this in the DX plugin in the future). Also move out stats code. Comment a lot and cleanup. Kill DX9 Globals.cpp.

Browse Source 
git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@938 8ced0084-cf51-0410-be5f-012b33b47a6e
This commit is contained in:
hrydgard
2008-10-22 20:54:40 +00:00
parent 32820d2036
commit 39df6c5624
41 changed files with 416 additions and 416 deletions
Download Patch File Download Diff File Expand all files Collapse all files

93
Source/Plugins/Plugin_VideoOGL/Src/VertexLoader.cpp
View File

@ -28,6 +28,7 @@
#include "StringUtil.h"
#include "Render.h"
#include "VertexShader.h"
#include "VertexManager.h"
#include "VertexLoader.h"
#include "BPStructs.h"
@ -55,7 +56,6 @@ static int colIndex;
#define inline
#endif
TVtxDesc VertexManager::s_GlobalVtxDesc;
// ==============================================================================
// Direct
@ -120,7 +120,7 @@ VertexLoader::VertexLoader()
{
m_numPipelineStages = 0;
m_VertexSize = 0;
m_AttrDirty = 1;
m_AttrDirty = AD_DIRTY;
VertexLoader_Normal::Init();
m_compiledCode = (u8 *)AllocateExecutableMemory(COMPILED_CODE_SIZE, false);
@ -136,7 +136,7 @@ VertexLoader::~VertexLoader()
int VertexLoader::ComputeVertexSize()
{
if (!m_AttrDirty) {
if (m_AttrDirty == AD_CLEAN) {
// Compare the 33 desc bits.
if (m_VtxDesc.Hex0 == VertexManager::GetVtxDesc().Hex0 &&
(m_VtxDesc.Hex1 & 1) == (VertexManager::GetVtxDesc().Hex1 & 1))
@ -152,7 +152,7 @@ int VertexLoader::ComputeVertexSize()
if (fnSetupVertexPointers != NULL && fnSetupVertexPointers == (void (*)())(void*)m_compiledCode)
VertexManager::Flush();
m_AttrDirty = 1;
m_AttrDirty = AD_DIRTY;
m_VertexSize = 0;
// Position Matrix Index
if (m_VtxDesc.PosMatIdx)
@ -257,6 +257,7 @@ int VertexLoader::ComputeVertexSize()
return m_VertexSize;
}
// Note the use of CallCdeclFunction3I etc.
// This is a horrible hack that is necessary because in 64-bit mode, Opengl32.dll is based way, way above the 32-bit
// address space that is within reach of a CALL, and just doing &fn gives us these high uncallable addresses. So we
@ -269,20 +270,24 @@ DECLARE_IMPORT(glVertexPointer);
DECLARE_IMPORT(glColorPointer);
DECLARE_IMPORT(glTexCoordPointer);
void VertexLoader::ProcessFormat()
void VertexLoader::PrepareForVertexFormat()
{
using namespace Gen;
//_assert_( VertexManager::s_pCurBufferPointer == s_pBaseBufferPointer );
if (!m_AttrDirty)
if (m_AttrDirty == AD_CLEAN)
{
// Check if local cached desc (in this VL) matches global desc
if (m_VtxDesc.Hex0 == VertexManager::GetVtxDesc().Hex0 && (m_VtxDesc.Hex1 & 1)==(VertexManager::GetVtxDesc().Hex1 & 1))
return; // same
{
return; // same
}
}
else
m_AttrDirty = 0;
else
{
m_AttrDirty = AD_CLEAN;
}
m_VtxDesc.Hex = VertexManager::GetVtxDesc().Hex;
DVSTARTPROFILE();
@ -315,7 +320,7 @@ void VertexLoader::ProcessFormat()
m_VBVertexStride += 12;
switch (m_VtxDesc.Position) {
case NOT_PRESENT: {_assert_msg_(0,"Vertex descriptor without position!","WTF?");} break;
case NOT_PRESENT: {_assert_msg_(0, "Vertex descriptor without position!", "WTF?");} break;
case DIRECT:
{
switch (m_VtxAttr.PosFormat) {
@ -409,7 +414,6 @@ void VertexLoader::ProcessFormat()
m_VBVertexStride += 6; // still include the texture coordinate, but this time as 6 bytes
m_components |= VB_HAS_UV0 << i; // have to include since using now
}
}
else {
if (tc[i] != NOT_PRESENT)
@ -425,8 +429,7 @@ void VertexLoader::ProcessFormat()
break;
}
}
if (j == 8 && !((m_components&VB_HAS_TEXMTXIDXALL)&(VB_HAS_TEXMTXIDXALL<<(i+1)))) // no more tex coords and tex matrices, so exit loop
if (j == 8 && !((m_components&VB_HAS_TEXMTXIDXALL) & (VB_HAS_TEXMTXIDXALL<<(i+1)))) // no more tex coords and tex matrices, so exit loop
break;
}
}
@ -438,18 +441,18 @@ void VertexLoader::ProcessFormat()
if (m_VBVertexStride & 3) {
// make sure all strides are at least divisible by 4 (some gfx cards experience a 3x speed boost)
m_VBStridePad = 4 - (m_VBVertexStride&3);
m_VBStridePad = 4 - (m_VBVertexStride & 3);
m_VBVertexStride += m_VBStridePad;
}
// compile the pointer set function
// compile the pointer set function - why?
u8 *old_code_ptr = GetWritableCodePtr();
SetCodePtr(m_compiledCode);
Util::EmitPrologue(6);
int offset = 0;
// Position
if (m_VtxDesc.Position != NOT_PRESENT) {
if (m_VtxDesc.Position != NOT_PRESENT) { // TODO: Why the check? Always present, AFAIK!
CallCdeclFunction4_I(glVertexPointer, 3, GL_FLOAT, m_VBVertexStride, offset);
offset += 12;
}
@ -484,6 +487,8 @@ void VertexLoader::ProcessFormat()
}
}
// TODO : With byte or short normals above, offset will be misaligned (not 4byte aligned)! Ugh!
for (int i = 0; i < 2; i++) {
if (col[i] != NOT_PRESENT) {
if (i)
@ -496,9 +501,8 @@ void VertexLoader::ProcessFormat()
// TextureCoord
for (int i = 0; i < 8; i++) {
if (tc[i] != NOT_PRESENT || (m_components&(VB_HAS_TEXMTXIDX0<<i))) {
int id = GL_TEXTURE0+i;
if (tc[i] != NOT_PRESENT || (m_components & (VB_HAS_TEXMTXIDX0 << i))) {
int id = GL_TEXTURE0 + i;
#ifdef _M_X64
#ifdef _MSC_VER
MOV(32, R(RCX), Imm32(id));
@ -517,6 +521,7 @@ void VertexLoader::ProcessFormat()
ABI_RestoreStack(1 * 4);
#endif
#endif
// TODO : More potential disalignment!
if (m_components & (VB_HAS_TEXMTXIDX0 << i)) {
if (tc[i] != NOT_PRESENT) {
CallCdeclFunction4_I(glTexCoordPointer, 3, GL_FLOAT, m_VBVertexStride, offset);
@ -528,14 +533,14 @@ void VertexLoader::ProcessFormat()
}
}
else {
CallCdeclFunction4_I(glTexCoordPointer, m_VtxAttr.texCoord[i].Elements?2:1, GL_FLOAT, m_VBVertexStride, offset);
CallCdeclFunction4_I(glTexCoordPointer, m_VtxAttr.texCoord[i].Elements ? 2 : 1, GL_FLOAT, m_VBVertexStride, offset);
offset += 4 * (m_VtxAttr.texCoord[i].Elements?2:1);
}
}
}
if (m_VtxDesc.PosMatIdx) {
CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_POSMTX_ATTRIB,1,GL_UNSIGNED_BYTE, GL_FALSE, m_VBVertexStride, offset);
CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_POSMTX_ATTRIB, 1, GL_UNSIGNED_BYTE, GL_FALSE, m_VBVertexStride, offset);
offset += 1;
}
@ -551,19 +556,6 @@ void VertexLoader::ProcessFormat()
SetCodePtr(old_code_ptr);
}
void VertexLoader::PrepareRun()
{
posScale = shiftLookup[m_VtxAttr.PosFrac];
if (m_components & VB_HAS_UVALL) {
for (int i = 0; i < 8; i++) {
tcScaleU[i] = shiftLookup[m_VtxAttr.texCoord[i].Frac];
tcScaleV[i] = shiftLookup[m_VtxAttr.texCoord[i].Frac];
}
}
for (int i = 0; i < 2; i++)
colElements[i] = m_VtxAttr.color[i].Elements;
}
void VertexLoader::SetupColor(int num, int mode, int format, int elements)
{
// if COL0 not present, then embed COL1 into COL0
@ -669,38 +661,49 @@ void VertexLoader::RunVertices(int primitive, int count)
{
DVSTARTPROFILE();
ComputeVertexSize(); // HACK for underruns in Super Monkey Ball etc. !!!! dirty handling must be wrong.
if (count <= 0)
return;
// This has dirty handling - won't actually recompute unless necessary.
ComputeVertexSize();
// Figure out a better check. Also, jitting fnSetupVertexPointers seems pretty silly - not likely to be a bottleneck.
if (fnSetupVertexPointers != NULL && fnSetupVertexPointers != (void (*)())(void*)m_compiledCode)
VertexManager::Flush();
if (bpmem.genMode.cullmode == 3 && primitive < 5)
{
// if cull mode is none, ignore triangles and quads
DataSkip(count*m_VertexSize);
DataSkip(count * m_VertexSize);
return;
}
ProcessFormat();
fnSetupVertexPointers = (void (*)())(void*)m_compiledCode;
// This has dirty handling - won't actually recompute unless necessary.
PrepareForVertexFormat();
fnSetupVertexPointers = (void (*)())(void*)m_compiledCode;
VertexManager::EnableComponents(m_components);
PrepareRun();
// Load position and texcoord scale factors.
// Hm, this could be done when the VtxAttr is set, instead.
posScale = shiftLookup[m_VtxAttr.PosFrac];
if (m_components & VB_HAS_UVALL) {
for (int i = 0; i < 8; i++) {
tcScaleU[i] = shiftLookup[m_VtxAttr.texCoord[i].Frac];
tcScaleV[i] = shiftLookup[m_VtxAttr.texCoord[i].Frac];
}
}
for (int i = 0; i < 2; i++)
colElements[i] = m_VtxAttr.color[i].Elements;
// if strips or fans, make sure all vertices can fit in buffer, otherwise flush
int granularity = 1;
switch(primitive) {
switch (primitive) {
case 3: // strip
case 4: // fan
if (VertexManager::GetRemainingSize() < 3*m_VBVertexStride )
if (VertexManager::GetRemainingSize() < 3 * m_VBVertexStride )
VertexManager::Flush();
break;
case 6: // line strip
if (VertexManager::GetRemainingSize() < 2*m_VBVertexStride )
if (VertexManager::GetRemainingSize() < 2 * m_VBVertexStride )
VertexManager::Flush();
break;
case 0: // quads