piwalker/dolphin
1
0
Fork 0
mirror of https://github.com/dolphin-emu/dolphin.git synced 2025-07-24 14:49:42 -06:00
Code Issues Packages Projects Releases Wiki Activity

Turn the X86 emitter into a class, so the code pointer is no longer a global, yay! Created XCodeBlock that derives from XEmitter, and the Jit now derives from XCodeBlock so it can call all ADD SUB JNZ etc without having to prefix them with "emit.". I think someone's gonna like this.

Browse Source 
There's some cleanup still to be done, but hey, it works. There shouldn't be a noticable speed difference.

I hope GCC doesn't have a problem with the "member function pointers" I used.

git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@1594 8ced0084-cf51-0410-be5f-012b33b47a6e
This commit is contained in:
hrydgard
2008-12-19 21:24:52 +00:00
parent b5dcdcf779
commit 104acd5bc1
31 changed files with 1297 additions and 1153 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
Source/Plugins/Plugin_VideoOGL/Src/BPStructs.cpp
View File

@ -463,7 +463,7 @@ void BPWritten(int addr, int changes, int newval)
{
// the number of lines copied is determined by the y scale * source efb height
float yScale = bpmem.dispcopyyscale / 256.0f;
float xfbLines = bpmem.copyTexSrcWH.y + 1.0 * yScale;
float xfbLines = bpmem.copyTexSrcWH.y + 1.0f * yScale;
XFB_Write(Memory_GetPtr(bpmem.copyTexDest<<5), multirc, (bpmem.copyMipMapStrideChannels << 4), (int)xfbLines);
}
else

44
Source/Plugins/Plugin_VideoOGL/Src/NativeVertexFormat.cpp
View File

@ -82,68 +82,68 @@ void NativeVertexFormat::Initialize(const PortableVertexDeclaration &_vtx_decl)
}
#ifdef USE_JIT
Gen::XEmitter emit(m_compiledCode);
// Alright, we have our vertex declaration. Compile some crazy code to set it quickly using GL.
u8 *old_code_ptr = GetWritableCodePtr();
SetCodePtr(m_compiledCode);
ABI_EmitPrologue(6);
emit.ABI_EmitPrologue(6);
CallCdeclFunction4_I(glVertexPointer, 3, GL_FLOAT, _vtx_decl.stride, 0);
emit.CallCdeclFunction4_I(glVertexPointer, 3, GL_FLOAT, _vtx_decl.stride, 0);
if (_vtx_decl.num_normals >= 1) {
CallCdeclFunction3_I(glNormalPointer, VarToGL(_vtx_decl.normal_gl_type), _vtx_decl.stride, _vtx_decl.normal_offset[0]);
emit.CallCdeclFunction3_I(glNormalPointer, VarToGL(_vtx_decl.normal_gl_type), _vtx_decl.stride, _vtx_decl.normal_offset[0]);
if (_vtx_decl.num_normals == 3) {
CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_NORM1_ATTRIB, _vtx_decl.normal_gl_size, VarToGL(_vtx_decl.normal_gl_type), GL_TRUE, _vtx_decl.stride, _vtx_decl.normal_offset[1]);
CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_NORM2_ATTRIB, _vtx_decl.normal_gl_size, VarToGL(_vtx_decl.normal_gl_type), GL_TRUE, _vtx_decl.stride, _vtx_decl.normal_offset[2]);
emit.CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_NORM1_ATTRIB, _vtx_decl.normal_gl_size, VarToGL(_vtx_decl.normal_gl_type), GL_TRUE, _vtx_decl.stride, _vtx_decl.normal_offset[1]);
emit.CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_NORM2_ATTRIB, _vtx_decl.normal_gl_size, VarToGL(_vtx_decl.normal_gl_type), GL_TRUE, _vtx_decl.stride, _vtx_decl.normal_offset[2]);
}
}
for (int i = 0; i < 2; i++) {
if (_vtx_decl.color_offset[i] != -1) {
if (i == 0)
CallCdeclFunction4_I(glColorPointer, 4, GL_UNSIGNED_BYTE, _vtx_decl.stride, _vtx_decl.color_offset[i]);
emit.CallCdeclFunction4_I(glColorPointer, 4, GL_UNSIGNED_BYTE, _vtx_decl.stride, _vtx_decl.color_offset[i]);
else
CallCdeclFunction4((void *)glSecondaryColorPointer, 4, GL_UNSIGNED_BYTE, _vtx_decl.stride, _vtx_decl.color_offset[i]);
emit.CallCdeclFunction4((void *)glSecondaryColorPointer, 4, GL_UNSIGNED_BYTE, _vtx_decl.stride, _vtx_decl.color_offset[i]);
}
}
for (int i = 0; i < 8; i++) {
if (_vtx_decl.texcoord_offset[i] != -1) {
for (int i = 0; i < 8; i++)
{
if (_vtx_decl.texcoord_offset[i] != -1)
{
int id = GL_TEXTURE0 + i;
#ifdef _M_X64
#ifdef _MSC_VER
MOV(32, R(RCX), Imm32(id));
emit.MOV(32, R(RCX), Imm32(id));
#else
MOV(32, R(RDI), Imm32(id));
emit.MOV(32, R(RDI), Imm32(id));
#endif
#else
ABI_AlignStack(1 * 4);
PUSH(32, Imm32(id));
emit.ABI_AlignStack(1 * 4);
emit.PUSH(32, Imm32(id));
#endif
CALL((void *)glClientActiveTexture);
emit.CALL((void *)glClientActiveTexture);
#ifndef _M_X64
#ifdef _WIN32
// don't inc stack on windows, stdcall
#else
ABI_RestoreStack(1 * 4);
emit.ABI_RestoreStack(1 * 4);
#endif
#endif
CallCdeclFunction4_I(
emit.CallCdeclFunction4_I(
glTexCoordPointer, _vtx_decl.texcoord_size[i], VarToGL(_vtx_decl.texcoord_gl_type[i]),
_vtx_decl.stride, _vtx_decl.texcoord_offset[i]);
}
}
if (_vtx_decl.posmtx_offset != -1) {
CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_POSMTX_ATTRIB, 4, GL_UNSIGNED_BYTE, GL_FALSE, _vtx_decl.stride, _vtx_decl.posmtx_offset);
emit.CallCdeclFunction6((void *)glVertexAttribPointer, SHADER_POSMTX_ATTRIB, 4, GL_UNSIGNED_BYTE, GL_FALSE, _vtx_decl.stride, _vtx_decl.posmtx_offset);
}
ABI_EmitEpilogue(6);
if (Gen::GetCodePtr() - (u8*)m_compiledCode > COMPILED_CODE_SIZE)
emit.ABI_EmitEpilogue(6);
if (emit.GetCodePtr() - (u8*)m_compiledCode > COMPILED_CODE_SIZE)
{
Crash();
}
SetCodePtr(old_code_ptr);
#endif
this->vtx_decl = _vtx_decl;
}

21
Source/Plugins/Plugin_VideoOGL/Src/VertexLoader.cpp
View File

@ -44,7 +44,7 @@
#define USE_JIT
#define COMPILED_CODE_SIZE 4096*4
#define COMPILED_CODE_SIZE 4096
NativeVertexFormat *g_nativeVertexFmt;
@ -116,6 +116,7 @@ void LOADERDECL TexMtx_Write_Short3()
VertexLoader::VertexLoader(const TVtxDesc &vtx_desc, const VAT &vtx_attr)
{
m_compiledCode = NULL;
m_numLoadedVertices = 0;
m_VertexSize = 0;
m_numPipelineStages = 0;
@ -126,16 +127,14 @@ VertexLoader::VertexLoader(const TVtxDesc &vtx_desc, const VAT &vtx_attr)
m_VtxDesc = vtx_desc;
SetVAT(vtx_attr.g0.Hex, vtx_attr.g1.Hex, vtx_attr.g2.Hex);
m_compiledCode = (u8 *)AllocateExecutableMemory(COMPILED_CODE_SIZE, false);
if (m_compiledCode) {
memset(m_compiledCode, 0, COMPILED_CODE_SIZE);
}
AllocCodeSpace(COMPILED_CODE_SIZE);
CompileVertexTranslator();
WriteProtect();
}
VertexLoader::~VertexLoader()
{
FreeMemoryPages(m_compiledCode, COMPILED_CODE_SIZE);
FreeCodeSpace();
delete m_NativeFmt;
}
@ -143,13 +142,14 @@ void VertexLoader::CompileVertexTranslator()
{
m_VertexSize = 0;
const TVtxAttr &vtx_attr = m_VtxAttr;
//const TVtxDesc &vtx_desc = m_VtxDesc;
#ifdef USE_JIT
u8 *old_code_ptr = GetWritableCodePtr();
SetCodePtr(m_compiledCode);
if (m_compiledCode)
PanicAlert("trying to recompile a vtx translator");
m_compiledCode = GetCodePtr();
ABI_EmitPrologue(4);
// MOV(32, R(EBX), M(&loop_counter));
// Start loop here
const u8 *loop_start = GetCodePtr();
@ -477,7 +477,6 @@ void VertexLoader::CompileVertexTranslator()
//SUB(32, R(EBX), Imm8(1));
J_CC(CC_NZ, loop_start, true);
ABI_EmitEpilogue(4);
SetCodePtr(old_code_ptr);
#endif
m_NativeFmt->Initialize(vtx_decl);
}

7
Source/Plugins/Plugin_VideoOGL/Src/VertexLoader.h
View File

@ -22,9 +22,10 @@
#include "CPMemory.h"
#include "DataReader.h"
#include "NativeVertexFormat.h"
#include "x64Emitter.h"
class VertexLoaderUID
{
u32 vid[5];
@ -52,7 +53,7 @@ public:
}
};
class VertexLoader
class VertexLoader : public Gen::XCodeBlock
{
public:
VertexLoader(const TVtxDesc &vtx_desc, const VAT &vtx_attr);
@ -86,7 +87,7 @@ private:
TPipelineFunction m_PipelineStages[64]; // TODO - figure out real max. it's lower.
int m_numPipelineStages;
u8 *m_compiledCode;
const u8 *m_compiledCode;
int m_numLoadedVertices;