//----------------------------------------------------------------------------- // Copyright (c) 2015 GarageGames, LLC // // Permission is hereby granted, free of charge, to any person obtaining a copy // of this software and associated documentation files (the "Software"), to // deal in the Software without restriction, including without limitation the // rights to use, copy, modify, merge, publish, distribute, sublicense, and/or // sell copies of the Software, and to permit persons to whom the Software is // furnished to do so, subject to the following conditions: // // The above copyright notice and this permission notice shall be included in // all copies or substantial portions of the Software. // // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR // IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, // FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE // AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER // LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING // FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS // IN THE SOFTWARE. //----------------------------------------------------------------------------- #include "platform/platform.h" #include "gfx/D3D11/gfxD3D11Shader.h" #include "core/frameAllocator.h" #include "core/stream/fileStream.h" #include "core/util/safeDelete.h" #include "console/console.h" extern bool gDisassembleAllShaders; #pragma comment(lib, "d3dcompiler.lib") gfxD3DIncludeRef GFXD3D11Shader::smD3DInclude = NULL; class gfxD3D11Include : public ID3DInclude, public StrongRefBase { private: Vector mLastPath; public: void setPath(const String &path) { mLastPath.clear(); mLastPath.push_back(path); } gfxD3D11Include() {} virtual ~gfxD3D11Include() {} STDMETHOD(Open)(THIS_ D3D_INCLUDE_TYPE IncludeType, LPCSTR pFileName, LPCVOID pParentData, LPCVOID *ppData, UINT *pBytes); STDMETHOD(Close)(THIS_ LPCVOID pData); }; HRESULT gfxD3D11Include::Open(THIS_ D3D_INCLUDE_TYPE IncludeType, LPCSTR pFileName, LPCVOID pParentData, LPCVOID *ppData, UINT *pBytes) { using namespace Torque; // First try making the path relative to the parent. Torque::Path path = Torque::Path::Join( mLastPath.last(), '/', pFileName ); path = Torque::Path::CompressPath( path ); if ( !Torque::FS::ReadFile( path, (void *&)*ppData, *pBytes, true ) ) { // Ok... now try using the path as is. path = String( pFileName ); path = Torque::Path::CompressPath( path ); if ( !Torque::FS::ReadFile( path, (void *&)*ppData, *pBytes, true ) ) { AssertISV(false, avar( "Failed to open include '%s'.", pFileName)); return E_FAIL; } } // If the data was of zero size then we cannot recurse // into this file and DX won't call Close() below. // // So in this case don't push on the path. if ( *pBytes > 0 ) mLastPath.push_back( path.getRootAndPath() ); return S_OK; } HRESULT gfxD3D11Include::Close( THIS_ LPCVOID pData ) { // Free the data file and pop its path off the stack. delete [] (U8*)pData; mLastPath.pop_back(); return S_OK; } GFXD3D11ShaderConstHandle::GFXD3D11ShaderConstHandle(GFXD3D11Shader* shader) : mShader(shader), mInstancingConstant(false) { dMemset(&mDesc, 0, sizeof(mDesc)); mValid = false; mStageFlags = 0; } GFXD3D11ShaderConstHandle::GFXD3D11ShaderConstHandle(GFXD3D11Shader* shader, const GFXShaderConstDesc& desc) : mShader(shader), mDesc(desc), mInstancingConstant(false) { if (desc.constType == GFXSCT_ConstBuffer) mValid = false; else mValid = true; addDesc(desc.shaderStage, desc); mStageFlags = desc.shaderStage; } GFXD3D11ShaderConstHandle::~GFXD3D11ShaderConstHandle() { } void GFXD3D11ShaderConstHandle::addDesc(GFXShaderStage stage, const GFXShaderConstDesc& desc) { // just add for now. mDescMap[stage] = desc; } const GFXShaderConstDesc GFXD3D11ShaderConstHandle::getDesc(GFXShaderStage stage) { return mDescMap[stage]; } //------------------------------------------------------------------------------ GFXD3D11ShaderConstBuffer::GFXD3D11ShaderConstBuffer(GFXD3D11Shader* shader) { mShader = shader; mDeviceContext = D3D11DEVICECONTEXT; for (U32 i = 0; i < 6; i++) { for (U32 j = 0; j < 16; j++) { mBoundBuffers[i][j] = nullptr; } } } GFXD3D11ShaderConstBuffer::~GFXD3D11ShaderConstBuffer() { for (auto& pair : mBufferMap) { delete[] pair.value.data; } mBufferMap.clear(); // Clear the map for (U32 i = 0; i < 6; i++) { for (U32 j = 0; j < 16; j++) { mBoundBuffers[i][j] = nullptr; } } if (mShader) mShader->_unlinkBuffer(this); } GFXShader* GFXD3D11ShaderConstBuffer::getShader() { return mShader; } void GFXD3D11ShaderConstBuffer::setMatrix(const GFXShaderConstDesc& constDesc, const U32 inSize, const void* data, U8* basePointer) { U8* buf = basePointer; if (constDesc.constType == GFXSCT_Float4x4) { // Special case, we can just blast this guy. if (dMemcmp(buf + constDesc.offset, data, inSize) != 0) { dMemcpy(buf + constDesc.offset, data, inSize); } return; } else { PROFILE_SCOPE(GFXD3D11ConstBufferLayout_setMatrix_not4x4); // Figure out how big of a chunk we are copying. We're going to copy 4 columns by N rows of data U32 csize; switch (constDesc.constType) { case GFXSCT_Float2x2: csize = 24; //this takes up 16+8 break; case GFXSCT_Float3x3: csize = 44; //This takes up 16+16+12 break; case GFXSCT_Float4x3: csize = 48; break; default: AssertFatal(false, "Unhandled case!"); return; break; } // Loop through and copy bool ret = false; U8* currDestPointer = buf + constDesc.offset; const U8* currSourcePointer = static_cast(data); const U8* endData = currSourcePointer + inSize; while (currSourcePointer < endData) { if (dMemcmp(currDestPointer, currSourcePointer, csize) != 0) { dMemcpy(currDestPointer, currSourcePointer, csize); ret = true; } currDestPointer += csize; currSourcePointer += sizeof(MatrixF); } } } void GFXD3D11ShaderConstBuffer::internalSet(GFXShaderConstHandle* handle, const U32 inSize, const void* data) { AssertFatal(handle, "GFXD3D11ShaderConstBuffer::internalSet - Handle is NULL!"); AssertFatal(handle->isValid(), "GFXD3D11ShaderConstBuffer::internalSet - Handle is not valid!"); AssertFatal(dynamic_cast(handle), "GFXD3D11ShaderConstBuffer::internalSet - Incorrect const buffer type"); GFXD3D11ShaderConstHandle* _dxHandle = static_cast(handle); AssertFatal(mShader == _dxHandle->mShader, "GFXD3D11ShaderConstBuffer::internalSet - Should only set handles which are owned by our shader"); S32 size = inSize; for (U32 i = VERTEX_SHADER; i <= COMPUTE_SHADER; i = GFXShaderStage(i << 1)) { if (_dxHandle->mStageFlags & i) { S32 shaderStageID = -1; // Initialize to -1 (bit not found) for (int j = 0; j < sizeof(S32) * 8; ++j) { if (i & (1 << j)) { shaderStageID = j; break; } } GFXShaderConstDesc constDesc = _dxHandle->getDesc((GFXShaderStage)i); BufferKey bufDesc(constDesc.bindPoint, shaderStageID); U8* basePointer = mBufferMap[bufDesc].data; if (_dxHandle->mInstancingConstant) { dMemcpy(mInstPtr + constDesc.offset, data, size); return; } switch (constDesc.constType) { case GFXSCT_Float2x2: case GFXSCT_Float3x3: case GFXSCT_Float4x3: case GFXSCT_Float4x4: setMatrix(constDesc, size, data, basePointer); return; break; // TODO add other AlignedVector here case GFXSCT_Float2: if (size > sizeof(Point2F)) size = constDesc.size; break; default: break; } if (dMemcmp(basePointer + constDesc.offset, data, size) != 0) { dMemcpy(basePointer + constDesc.offset, data, size); } } } } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const F32 fv) { internalSet(handle, sizeof(F32), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const Point2F& fv) { internalSet(handle, sizeof(Point2F), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const Point3F& fv) { internalSet(handle, sizeof(Point3F), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const Point4F& fv) { internalSet(handle, sizeof(Point4F), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const PlaneF& fv) { internalSet(handle, sizeof(PlaneF), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const LinearColorF& fv) { internalSet(handle, sizeof(Point4F), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const S32 fv) { // This is the only type that is allowed to be used // with a sampler shader constant type, but it is only // allowed to be set from GLSL. // // So we ignore it here... all other cases will assert. // if (((GFXD3D11ShaderConstHandle*)handle)->isSampler()) return; internalSet(handle, sizeof(S32), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const Point2I& fv) { internalSet(handle, sizeof(Point2I), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const Point3I& fv) { internalSet(handle, sizeof(Point3I), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const Point4I& fv) { internalSet(handle, sizeof(Point4I), &fv); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const AlignedArray& fv) { internalSet(handle, fv.getElementSize() * fv.size(), fv.getBuffer()); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const MatrixF& mat, const GFXShaderConstType matrixType) { AssertFatal(handle, "Handle is NULL!"); AssertFatal(handle->isValid(), "Handle is not valid!"); AssertFatal(static_cast(handle), "Incorrect const buffer type!"); GFXD3D11ShaderConstHandle* _dxHandle = static_cast(handle); AssertFatal(!_dxHandle->isSampler(), "Handle is sampler constant!"); AssertFatal(_dxHandle->mShader == mShader, "Mismatched shaders!"); MatrixF transposed; if (matrixType == GFXSCT_Float4x3) { transposed = mat; } else { mat.transposeTo(transposed); } if (_dxHandle->mInstancingConstant) { GFXShaderConstDesc constDesc = _dxHandle->getDesc(GFXShaderStage::PIXEL_SHADER); if (matrixType == GFXSCT_Float4x4) dMemcpy(mInstPtr + constDesc.offset, mat, sizeof(mat)); // TODO: Support 3x3 and 2x2 matricies? return; } internalSet(handle, sizeof(MatrixF), &transposed); } void GFXD3D11ShaderConstBuffer::set(GFXShaderConstHandle* handle, const MatrixF* mat, const U32 arraySize, const GFXShaderConstType matrixType) { AssertFatal(handle, "Handle is NULL!"); AssertFatal(handle->isValid(), "Handle is not valid!"); AssertFatal(static_cast(handle), "Incorrect const buffer type!"); const GFXD3D11ShaderConstHandle* _dxHandle = static_cast(handle); AssertFatal(!_dxHandle->isSampler(), "Handle is sampler constant!"); AssertFatal(_dxHandle->mShader == mShader, "Mismatched shaders!"); static Vector transposed; if (arraySize > transposed.size()) transposed.setSize(arraySize); if (matrixType == GFXSCT_Float4x3) { dMemcpy(transposed.address(), mat, arraySize * sizeof(MatrixF)); } else { for (U32 i = 0; i < arraySize; i++) mat[i].transposeTo(transposed[i]); } // TODO: Maybe support this in the future? if (_dxHandle->mInstancingConstant) return; internalSet(handle, sizeof(MatrixF) * arraySize, transposed.begin()); } const String GFXD3D11ShaderConstBuffer::describeSelf() const { String ret; ret = String(" GFXD3D11ShaderConstBuffer\n"); return ret; } void GFXD3D11ShaderConstBuffer::addBuffer(const GFXShaderConstDesc desc) { S32 shaderStageID = -1; // Initialize to -1 (bit not found) for (int i = 0; i < sizeof(S32) * 8; ++i) { if (desc.shaderStage & (1 << i)) { shaderStageID = i; break; } } // no shaderstage defined? cannot be!!! if (shaderStageID == -1) { AssertFatal(false, "DX Const buffer requires a shaderStage flag."); } const BufferKey bufKey(desc.bindPoint, shaderStageID); // doesnt matter if its already added. U8* buf = new U8[desc.size]; dMemset(buf, 0, desc.size); mBufferMap[bufKey].data = buf; mBufferMap[bufKey].size = desc.size; mBufferMap[bufKey].isDirty = true; mBoundBuffers[(U32)shaderStageID][desc.bindPoint] = D3D11->getDeviceBuffer(desc); } void GFXD3D11ShaderConstBuffer::activate( GFXD3D11ShaderConstBuffer *prevShaderBuffer ) { PROFILE_SCOPE(GFXD3D11ShaderConstBuffer_activate); BufferRange bufRanges[6]; for (BufferMap::Iterator i = mBufferMap.begin(); i != mBufferMap.end(); ++i) { const BufferKey thisBufferDesc = i->key; ConstantBuffer thisBuff = i->value; if (prevShaderBuffer && prevShaderBuffer != this) { const ConstantBuffer prevBuffer = prevShaderBuffer->mBufferMap[i->key]; if (prevBuffer.data && !prevBuffer.isDirty) { if (prevBuffer.size != thisBuff.size) { thisBuff.isDirty = true; } else { if (dMemcmp(prevBuffer.data, thisBuff.data, thisBuff.size) != 0) { thisBuff.isDirty = true; } else { thisBuff.isDirty = false; } } } else { thisBuff.isDirty = true; } } else { thisBuff.isDirty = true; } if (thisBuff.data && thisBuff.isDirty) { D3D11DEVICECONTEXT->UpdateSubresource(mBoundBuffers[thisBufferDesc.key2][thisBufferDesc.key1], 0, NULL, thisBuff.data, thisBuff.size, 0); bufRanges[thisBufferDesc.key2].addSlot(thisBufferDesc.key1); } } if (mShader->mVertShader && bufRanges[0].isValid()) { const U32 bufStartSlot = bufRanges[0].mBufMin; const U32 numBufs = bufRanges[0].mBufMax - bufRanges[0].mBufMin + 1; ID3D11Buffer** vsBuffers = mBoundBuffers[0] + bufStartSlot; D3D11DEVICECONTEXT->VSSetConstantBuffers(bufStartSlot, numBufs, vsBuffers); } if (mShader->mPixShader && bufRanges[1].isValid()) { const U32 bufStartSlot = bufRanges[1].mBufMin; const U32 numBufs = bufRanges[1].mBufMax - bufRanges[1].mBufMin + 1; ID3D11Buffer** psBuffers = mBoundBuffers[1] + bufStartSlot; D3D11DEVICECONTEXT->PSSetConstantBuffers(bufStartSlot, numBufs, psBuffers); } if (mShader->mGeoShader && bufRanges[2].isValid()) { const U32 bufStartSlot = bufRanges[2].mBufMin; const U32 numBufs = bufRanges[2].mBufMax - bufRanges[2].mBufMin + 1; ID3D11Buffer** psBuffers = mBoundBuffers[2] + bufStartSlot; D3D11DEVICECONTEXT->GSSetConstantBuffers(bufStartSlot, numBufs, psBuffers); } mWasLost = false; } void GFXD3D11ShaderConstBuffer::onShaderReload( GFXD3D11Shader *shader ) { AssertFatal( shader == mShader, "GFXD3D11ShaderConstBuffer::onShaderReload is hosed!" ); for (auto& pair : mBufferMap) { delete[] pair.value.data; } mBufferMap.clear(); // Clear the map for (GFXD3D11Shader::BufferMap::Iterator i = shader->mBuffers.begin(); i != shader->mBuffers.end(); ++i) { // add our buffer descriptions to the full const buffer. this->addBuffer(i->value); } // Set the lost state. mWasLost = true; } //------------------------------------------------------------------------------ GFXD3D11Shader::GFXD3D11Shader() { VECTOR_SET_ASSOCIATION( mShaderConsts ); AssertFatal(D3D11DEVICE, "Invalid device for shader."); mVertShader = NULL; mPixShader = NULL; mGeoShader = NULL; if( smD3DInclude == NULL ) smD3DInclude = new gfxD3D11Include; } //------------------------------------------------------------------------------ GFXD3D11Shader::~GFXD3D11Shader() { for (auto& pair : mHandles) { if (pair.value != nullptr) { delete pair.value; pair.value = nullptr; } } mHandles.clear(); // release shaders SAFE_RELEASE(mVertShader); SAFE_RELEASE(mPixShader); SAFE_RELEASE(mGeoShader); //maybe add SAFE_RELEASE(mVertexCode) ? } bool GFXD3D11Shader::_init() { PROFILE_SCOPE( GFXD3D11Shader_Init ); SAFE_RELEASE(mVertShader); SAFE_RELEASE(mPixShader); SAFE_RELEASE(mGeoShader); // Create the macro array including the system wide macros. const U32 macroCount = smGlobalMacros.size() + mMacros.size() + 2; FrameTemp d3dMacros( macroCount ); for ( U32 i=0; i < smGlobalMacros.size(); i++ ) { d3dMacros[i].Name = smGlobalMacros[i].name.c_str(); d3dMacros[i].Definition = smGlobalMacros[i].value.c_str(); } for ( U32 i=0; i < mMacros.size(); i++ ) { d3dMacros[i+smGlobalMacros.size()].Name = mMacros[i].name.c_str(); d3dMacros[i+smGlobalMacros.size()].Definition = mMacros[i].value.c_str(); } d3dMacros[macroCount - 2].Name = "TORQUE_SM"; d3dMacros[macroCount - 2].Definition = D3D11->getShaderModel().c_str(); memset(&d3dMacros[macroCount - 1], 0, sizeof(D3D_SHADER_MACRO)); mShaderConsts.clear(); mSamplerDescriptions.clear(); if (!mVertexFile.isEmpty() && !_compileShader( mVertexFile, GFXShaderStage::VERTEX_SHADER, d3dMacros) ) return false; if (!mPixelFile.isEmpty() && !_compileShader( mPixelFile, GFXShaderStage::PIXEL_SHADER, d3dMacros)) return false; if (!mGeometryFile.isEmpty()) { if (!_compileShader(mGeometryFile, GFXShaderStage::GEOMETRY_SHADER, d3dMacros)) return false; } // Mark all existing handles as invalid. // Those that are found when parsing the descriptions will then be marked valid again. for (auto& pair : mHandles) { pair.value->clear(); } _buildShaderConstantHandles(); // Notify any existing buffers that the buffer // layouts have changed and they need to update. Vector::iterator biter = mActiveBuffers.begin(); for ( ; biter != mActiveBuffers.end(); biter++ ) ((GFXD3D11ShaderConstBuffer*)(*biter))->onShaderReload( this ); return true; } bool GFXD3D11Shader::_compileShader( const Torque::Path &filePath, GFXShaderStage shaderStage, const D3D_SHADER_MACRO *defines) { PROFILE_SCOPE( GFXD3D11Shader_CompileShader ); using namespace Torque; HRESULT res = E_FAIL; ID3DBlob* code = NULL; ID3DBlob* errorBuff = NULL; ID3D11ShaderReflection* reflectionTable = NULL; #ifdef TORQUE_GFX_VISUAL_DEBUG //for use with NSight, GPU Perf studio, VS graphics debugger U32 flags = D3DCOMPILE_DEBUG | D3DCOMPILE_ENABLE_STRICTNESS | D3DCOMPILE_PREFER_FLOW_CONTROL | D3DCOMPILE_SKIP_OPTIMIZATION; #elif defined(TORQUE_DEBUG) //debug build U32 flags = D3DCOMPILE_DEBUG | D3DCOMPILE_ENABLE_STRICTNESS | D3DCOMPILE_WARNINGS_ARE_ERRORS; #else //release build U32 flags = D3DCOMPILE_ENABLE_STRICTNESS | D3DCOMPILE_OPTIMIZATION_LEVEL3; #endif #ifdef D3D11_DEBUG_SPEW Con::printf( "Compiling Shader: '%s'", filePath.getFullPath().c_str() ); #endif // Is it an HLSL shader? if(filePath.getExtension().equal("hlsl", String::NoCase)) { // Set this so that the D3DInclude::Open will have this // information for relative paths. smD3DInclude->setPath(filePath.getRootAndPath()); FileStream s; if (!s.open(filePath, Torque::FS::File::Read)) { AssertISV(false, avar("GFXD3D11Shader::initShader - failed to open shader '%s'.", filePath.getFullPath().c_str())); if ( smLogErrors ) Con::errorf( "GFXD3D11Shader::_compileShader - Failed to open shader file '%s'.", filePath.getFullPath().c_str() ); return false; } // Convert the path which might have virtualized // mount paths to a real file system path. Torque::Path realPath; if (!FS::GetFSPath( filePath, realPath)) realPath = filePath; U32 bufSize = s.getStreamSize(); FrameAllocatorMarker fam; char *buffer = NULL; buffer = (char*)fam.alloc(bufSize + 1); s.read(bufSize, buffer); buffer[bufSize] = 0; String target; switch (shaderStage) { case VERTEX_SHADER: target = D3D11->getVertexShaderTarget(); break; case PIXEL_SHADER: target = D3D11->getPixelShaderTarget(); break; case GEOMETRY_SHADER: target = D3D11->getGeometryShaderTarget(); break; case DOMAIN_SHADER: break; case HULL_SHADER: break; case COMPUTE_SHADER: break; default: break; } res = D3DCompile(buffer, bufSize, realPath.getFullPath().c_str(), defines, smD3DInclude, "main", target, flags, 0, &code, &errorBuff); } if(errorBuff) { // remove \n at end of buffer U8 *buffPtr = (U8*) errorBuff->GetBufferPointer(); U32 len = dStrlen( (const char*) buffPtr ); buffPtr[len-1] = '\0'; if(FAILED(res)) { if(smLogErrors) Con::errorf("failed to compile shader: %s", buffPtr); } else { if(smLogWarnings) Con::errorf("shader compiled with warning(s): %s", buffPtr); } } else if (code == NULL && smLogErrors) Con::errorf( "GFXD3D11Shader::_compileShader - no compiled code produced; possibly missing file '%s'.", filePath.getFullPath().c_str() ); AssertISV(SUCCEEDED(res), "Unable to compile shader!"); if(code != NULL) { switch (shaderStage) { case VERTEX_SHADER: res = D3D11DEVICE->CreateVertexShader(code->GetBufferPointer(), code->GetBufferSize(), NULL, &mVertShader); break; case PIXEL_SHADER: res = D3D11DEVICE->CreatePixelShader(code->GetBufferPointer(), code->GetBufferSize(), NULL, &mPixShader); break; case GEOMETRY_SHADER: res = D3D11DEVICE->CreateGeometryShader(code->GetBufferPointer(), code->GetBufferSize(), NULL, &mGeoShader); break; case DOMAIN_SHADER: break; case HULL_SHADER: break; case COMPUTE_SHADER: break; default: break; } if (FAILED(res)) { AssertFatal(false, "D3D11Shader::_compilershader- failed to create shader"); } if(res == S_OK) { HRESULT reflectionResult = D3DReflect(code->GetBufferPointer(), code->GetBufferSize(), IID_ID3D11ShaderReflection, (void**)&reflectionTable); if(FAILED(reflectionResult)) AssertFatal(false, "D3D11Shader::_compilershader - Failed to get shader reflection table interface"); } if(res == S_OK) _getShaderConstants(reflectionTable, shaderStage); if(FAILED(res) && smLogErrors) Con::errorf("GFXD3D11Shader::_compileShader - Unable to create shader for '%s'.", filePath.getFullPath().c_str()); } //bool result = code && SUCCEEDED(res) && HasValidConstants; bool result = code && SUCCEEDED(res); #ifdef TORQUE_DEBUG String shader; switch (shaderStage) { case VERTEX_SHADER: shader = mVertexFile.getFileName(); mVertShader->SetPrivateData(WKPDID_D3DDebugObjectName, shader.size(), shader.c_str()); break; case PIXEL_SHADER: shader = mPixelFile.getFileName(); mPixShader->SetPrivateData(WKPDID_D3DDebugObjectName, shader.size(), shader.c_str()); break; case GEOMETRY_SHADER: shader = mGeometryFile.getFileName(); mGeoShader->SetPrivateData(WKPDID_D3DDebugObjectName, shader.size(), shader.c_str()); break; case DOMAIN_SHADER: break; case HULL_SHADER: break; case COMPUTE_SHADER: break; default: break; } #endif SAFE_RELEASE(code); SAFE_RELEASE(reflectionTable); SAFE_RELEASE(errorBuff); return result; } void GFXD3D11Shader::_getShaderConstants( ID3D11ShaderReflection* refTable, GFXShaderStage shaderStage) { PROFILE_SCOPE(GFXD3D11Shader_GetShaderConstants); AssertFatal(refTable, "NULL constant table not allowed, is this an assembly shader?"); D3D11_SHADER_DESC shaderDesc; if (refTable->GetDesc(&shaderDesc) != S_OK) { AssertFatal(false, "Shader Reflection table unable to be created"); } // we loop through and account for the most common data types. for (U32 i = 0; i < shaderDesc.ConstantBuffers; i++) { GFXShaderConstDesc desc; ID3D11ShaderReflectionConstantBuffer* constantBuffer = refTable->GetConstantBufferByIndex(i); D3D11_SHADER_BUFFER_DESC constantBufferDesc; if (constantBuffer->GetDesc(&constantBufferDesc) == S_OK) { desc.name = String(constantBufferDesc.Name); desc.size = constantBufferDesc.Size; desc.constType = GFXSCT_ConstBuffer; desc.shaderStage = shaderStage; desc.samplerReg = -1; // get our binding point. D3D11_SHADER_INPUT_BIND_DESC shaderInputBind; refTable->GetResourceBindingDescByName(constantBufferDesc.Name, &shaderInputBind); desc.bindPoint = shaderInputBind.BindPoint; if (String::compare(desc.name, "$Globals") == 0 || String::compare(desc.name, "$Params") == 0) { switch (shaderStage) { case VERTEX_SHADER: desc.name = desc.name + "_" + mVertexFile.getFileName(); break; case PIXEL_SHADER: desc.name = desc.name + "_" + mPixelFile.getFileName(); break; case GEOMETRY_SHADER: desc.name = desc.name + "_" + mGeometryFile.getFileName(); break; case DOMAIN_SHADER: break; case HULL_SHADER: break; case COMPUTE_SHADER: break; default: break; } } mBuffers[desc.name] = desc; // now loop vars and add them to mShaderConsts. for (U32 j = 0; j < constantBufferDesc.Variables; j++) { ID3D11ShaderReflectionVariable* bufferVar = constantBuffer->GetVariableByIndex(j); D3D11_SHADER_VARIABLE_DESC shaderVarDesc; bufferVar->GetDesc(&shaderVarDesc); D3D11_SHADER_TYPE_DESC shaderTypeDesc; bufferVar->GetType()->GetDesc(&shaderTypeDesc); if (shaderTypeDesc.Class == D3D_SVC_STRUCT) { // we gotta loop through its variables =/ add support in future. for now continue so it skips. // no idea how to handle arrays of structs.... /*for (U32 j = 0; j < shaderTypeDesc.Members; j++) { GFXShaderConstDesc memVarDesc; ID3D11ShaderReflectionType* memType = bufferVar->GetType()->GetMemberTypeByIndex(j); D3D11_SHADER_TYPE_DESC memTypeDesc; memType->GetDesc(&memTypeDesc); memVarDesc.name = String(shaderVarDesc.Name) + "." + String(memTypeDesc.Name); if (memVarDesc.name.find("$") != 0) memVarDesc.name = String::ToString("$%s", memVarDesc.name.c_str()); #ifdef D3D11_DEBUG_SPEW Con::printf("Variable Name %s:, offset: %d", memVarDesc.name.c_str(), memVarDesc.Offset); #endif }*/ continue; } else { GFXShaderConstDesc varDesc; varDesc.name = String(shaderVarDesc.Name); if (varDesc.name.find("$") != 0) varDesc.name = String::ToString("$%s", varDesc.name.c_str()); // set the bind point to the same as the const buffer. varDesc.bindPoint = desc.bindPoint; varDesc.offset = shaderVarDesc.StartOffset; varDesc.arraySize = mMax(shaderTypeDesc.Elements, 1); varDesc.size = shaderVarDesc.Size; varDesc.shaderStage = shaderStage; varDesc.samplerReg = -1; varDesc.constType = convertConstType(shaderTypeDesc); #ifdef D3D11_DEBUG_SPEW Con::printf("Variable Name %s:, offset: %d, size: %d, constantDesc.Elements: %d", varDesc.name.c_str(), varDesc.offset, varDesc.size, varDesc.arraySize); #endif mShaderConsts.push_back(varDesc); } } } else { AssertFatal(false, "Unable to get shader constant description! (may need more elements of constantDesc"); } } for (U32 i = 0; i < shaderDesc.BoundResources; i++) { GFXShaderConstDesc desc; D3D11_SHADER_INPUT_BIND_DESC shaderInputBind; refTable->GetResourceBindingDesc(i, &shaderInputBind); if (shaderInputBind.Type == D3D_SIT_TEXTURE || shaderInputBind.Type == D3D_SIT_UAV_RWTYPED) { // these should return shaderResourceViews and add them to shaderResources. /*switch (shaderInputBind.Dimension) { case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE1D: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE1DARRAY: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE2D: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE2DARRAY: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE2DMS: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE2DMSARRAY: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURE3D: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURECUBE: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_TEXTURECUBEARRAY: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_BUFFER: break; case D3D_SRV_DIMENSION::D3D_SRV_DIMENSION_BUFFEREX: break; default: break; }*/ } else if (shaderInputBind.Type == D3D_SIT_SAMPLER) { // Prepend a "$" if it doesn't exist. Just to make things consistent. desc.name = String(shaderInputBind.Name); if (desc.name.find("$") != 0) desc.name = String::ToString("$%s", desc.name.c_str()); desc.constType = GFXSCT_Sampler; desc.samplerReg = shaderInputBind.BindPoint; desc.bindPoint = -1; desc.shaderStage = shaderStage; desc.arraySize = shaderInputBind.BindCount; mSamplerDescriptions.push_back(desc); } else if (shaderInputBind.Type == D3D_SIT_UAV_RWSTRUCTURED || shaderInputBind.Type == D3D_SIT_UAV_RWBYTEADDRESS || shaderInputBind.Type == D3D_SIT_UAV_APPEND_STRUCTURED || shaderInputBind.Type == D3D_SIT_UAV_CONSUME_STRUCTURED || shaderInputBind.Type == D3D_SIT_UAV_RWSTRUCTURED_WITH_COUNTER) { // these should return an unorderedAccessViews and add them to shaderResources. } else if (shaderInputBind.Type == D3D_SIT_STRUCTURED || shaderInputBind.Type == D3D_SIT_BYTEADDRESS) { // these should return shaderResourceViews and add them to shaderResources. } } } GFXShaderConstType GFXD3D11Shader::convertConstType(const D3D11_SHADER_TYPE_DESC typeDesc) { if (typeDesc.Class == D3D_SVC_SCALAR || typeDesc.Class == D3D_SVC_VECTOR) { switch (typeDesc.Type) { case D3D_SVT_BOOL: return (GFXShaderConstType)((U32)GFXSCT_Bool + typeDesc.Columns - 1); break; case D3D_SVT_INT: return (GFXShaderConstType)((U32)GFXSCT_Int + typeDesc.Columns - 1); break; case D3D_SVT_FLOAT: return (GFXShaderConstType)((U32)GFXSCT_Float + typeDesc.Columns - 1); break; case D3D_SVT_UINT: return (GFXShaderConstType)((U32)GFXSCT_UInt + typeDesc.Columns - 1); break; default: AssertFatal(false, "Unknown shader constant class enum, maybe you could add it?"); break; } } else if (typeDesc.Class == D3D_SVC_MATRIX_COLUMNS || typeDesc.Class == D3D_SVC_MATRIX_ROWS) { if (typeDesc.Type != D3D_SVT_FLOAT) { AssertFatal(false, "Only Float matrices are supported for now. Support for other types needs to be added."); } switch (typeDesc.Rows) { case 3: return typeDesc.Columns == 4 ? GFXSCT_Float3x4 : GFXSCT_Float3x3; break; case 4: return typeDesc.Columns == 3 ? GFXSCT_Float4x3 : GFXSCT_Float4x4; break; } } return GFXSCT_Uknown; } void GFXD3D11Shader::_buildShaderConstantHandles() { // loop through all constants, add them to the handle map // and add the const buffers to the buffer map. for (U32 i = 0; i < mShaderConsts.size(); i++) { GFXD3D11ShaderConstHandle* handle; GFXShaderConstDesc& desc = mShaderConsts[i]; HandleMap::Iterator j = mHandles.find(desc.name); // already added? reinit just in case.. // not sure if we need to do anything different with samplers. if (j != mHandles.end()) { handle = j->value; handle->mShader = this; if((handle->mStageFlags & desc.shaderStage) == 0) handle->mStageFlags |= (U32)desc.shaderStage; handle->addDesc(desc.shaderStage, desc); handle->setValid(true); } else { handle = new GFXD3D11ShaderConstHandle(this, desc); mHandles[desc.name] = handle; } } for (U32 j = 0; j < mSamplerDescriptions.size(); j++) { const GFXShaderConstDesc& desc = mSamplerDescriptions[j]; AssertFatal(desc.constType == GFXSCT_Sampler || desc.constType == GFXSCT_SamplerCube || desc.constType == GFXSCT_SamplerCubeArray || desc.constType == GFXSCT_SamplerTextureArray, "GFXD3D11Shader::_buildShaderConstantHandles - Invalid samplerDescription type!"); GFXD3D11ShaderConstHandle* handle; HandleMap::Iterator k = mHandles.find(desc.name); // already added? reinit just in case.. // not sure if we need to do anything different with samplers. if (k != mHandles.end()) { handle = k->value; } else { handle = new GFXD3D11ShaderConstHandle(this, desc); mHandles[desc.name] = handle; } handle->mShader = this; handle->setValid(true); } _buildInstancingShaderConstantHandles(); } void GFXD3D11Shader::_buildInstancingShaderConstantHandles() { // If we have no instancing than just return if (!mInstancingFormat) return; U32 offset = 0; for ( U32 i=0; i < mInstancingFormat->getElementCount(); i++ ) { const GFXVertexElement &element = mInstancingFormat->getElement( i ); String constName = String::ToString( "$%s", element.getSemantic().c_str() ); GFXD3D11ShaderConstHandle *handle; HandleMap::Iterator j = mHandles.find( constName ); GFXShaderConstDesc desc; desc.name = constName; desc.offset = offset; switch (element.getType()) { case GFXDeclType_Float4: desc.constType = GFXSCT_Float4; break; default: desc.constType = GFXSCT_Float; break; } desc.shaderStage = GFXShaderStage::PIXEL_SHADER; desc.samplerReg = -1; desc.size = 0; desc.arraySize = 1; if ( j != mHandles.end() ) handle = j->value; else { handle = new GFXD3D11ShaderConstHandle(this, desc); mHandles[ constName ] = handle; } handle->mShader = this; handle->setValid( true ); handle->mInstancingConstant = true; // If this is a matrix we will have 2 or 3 more of these // semantics with the same name after it. for ( ; i < mInstancingFormat->getElementCount(); i++ ) { const GFXVertexElement &nextElement = mInstancingFormat->getElement( i ); if ( nextElement.getSemantic() != element.getSemantic() ) { i--; break; } offset += nextElement.getSizeInBytes(); } } } GFXShaderConstBufferRef GFXD3D11Shader::allocConstBuffer() { GFXD3D11ShaderConstBuffer* buffer = new GFXD3D11ShaderConstBuffer(this); for (BufferMap::Iterator i = mBuffers.begin(); i != mBuffers.end(); ++i) { // add our buffer descriptions to the full const buffer. buffer->addBuffer(i->value); } mActiveBuffers.push_back( buffer ); buffer->registerResourceWithDevice(getOwningDevice()); return buffer; } /// Returns a shader constant handle for name, if the variable doesn't exist NULL is returned. GFXShaderConstHandle* GFXD3D11Shader::getShaderConstHandle(const String& name) { HandleMap::Iterator i = mHandles.find(name); if ( i != mHandles.end() ) { return i->value; } else { GFXD3D11ShaderConstHandle *handle = new GFXD3D11ShaderConstHandle(this); handle->setValid( false ); mHandles[name] = handle; return handle; } } GFXShaderConstHandle* GFXD3D11Shader::findShaderConstHandle(const String& name) { HandleMap::Iterator i = mHandles.find(name); if(i != mHandles.end()) return i->value; else { return NULL; } } const Vector& GFXD3D11Shader::getShaderConstDesc() const { return mShaderConsts; } U32 GFXD3D11Shader::getAlignmentValue(const GFXShaderConstType constType) const { const U32 mRowSizeF = 16; const U32 mRowSizeI = 16; switch (constType) { case GFXSCT_Float : case GFXSCT_Float2 : case GFXSCT_Float3 : case GFXSCT_Float4 : return mRowSizeF; break; // Matrices case GFXSCT_Float2x2 : return mRowSizeF * 2; break; case GFXSCT_Float3x3 : return mRowSizeF * 3; break; case GFXSCT_Float4x3: return mRowSizeF * 3; break; case GFXSCT_Float4x4 : return mRowSizeF * 4; break; //// Scalar case GFXSCT_Int : case GFXSCT_Int2 : case GFXSCT_Int3 : case GFXSCT_Int4 : return mRowSizeI; break; default: AssertFatal(false, "Unsupported type!"); return 0; break; } } void GFXD3D11Shader::zombify() { // Shaders don't need zombification } void GFXD3D11Shader::resurrect() { // Shaders are never zombies, and therefore don't have to be brought back. }