void Model::CreatePipelineState()
{
Microsoft::WRL::ComPtr<ID3DBlob> VertexShader;
Microsoft::WRL::ComPtr<ID3DBlob> PixelShader;
LoadAndCompileShader(VertexShader, PixelShader);
std::array<D3D12_INPUT_ELEMENT_DESC, 2> InputElementDesc
{ {
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
} };
D3D12_GRAPHICS_PIPELINE_STATE_DESC PipelineStateDesc = {};
PipelineStateDesc.InputLayout = { InputElementDesc.data(), static_cast<UINT>(InputElementDesc.size()) };
PipelineStateDesc.pRootSignature = RootSignature.Get();
PipelineStateDesc.VS = CD3DX12_SHADER_BYTECODE(VertexShader.Get());
PipelineStateDesc.PS = CD3DX12_SHADER_BYTECODE(PixelShader.Get());
PipelineStateDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
PipelineStateDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
PipelineStateDesc.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
PipelineStateDesc.SampleMask = UINT_MAX;
PipelineStateDesc.SampleDesc.Count = 1;
PipelineStateDesc.NumRenderTargets = 1;
PipelineStateDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
PipelineStateDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
PipelineStateDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
Utility::ThrowOnFail(DeviceContext.GetDevice()->CreateGraphicsPipelineState(&PipelineStateDesc, IID_PPV_ARGS(&PipelineState)));
}
Pipeline::Pipeline() :
m_pPipelineState( nullptr )
{
m_pipelineDesc = {};
m_pipelineDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
m_pipelineDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
m_pipelineDesc.DepthStencilState.DepthEnable = false;
m_pipelineDesc.DepthStencilState.StencilEnable = false;
m_pipelineDesc.SampleMask = UINT_MAX;
m_pipelineDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_PATCH;
m_pipelineDesc.NumRenderTargets = 1;
m_pipelineDesc.RTVFormats[0] = DXGI_FORMAT_B8G8R8A8_UNORM;
m_pipelineDesc.SampleDesc.Count = 1;
}
void TextureStore::init() {
// Create an empty root signature.
{
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(xapp().device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&rootSignature)));
}
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = rootSignature.Get();
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
//ThrowIfFailed(device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&pipelineState)));
#include "CompiledShaders/PostVS.h"
//#include "CompiledShaders/PostPS.h"
// test shade library functions
//{
//D3DLoadModule() uses ID3D11Module
//ComPtr<ID3DBlob> vShader;
//ThrowIfFailed(D3DReadFileToBlob(L"", &vShader));
psoDesc.VS = { binShader_PostVS, sizeof(binShader_PostVS) };
ThrowIfFailed(xapp().device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&pipelineState)));
ThrowIfFailed(xapp().device->CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT, IID_PPV_ARGS(&commandAllocator)));
ThrowIfFailed(xapp().device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, commandAllocator.Get(), pipelineState.Get(), IID_PPV_ARGS(&commandList)));
ThrowIfFailed(xapp().device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&updateFrameData.fence)));
updateFrameData.fence->SetName(L"fence_texture_update");
updateFrameData.fenceValue = 0;
updateFrameData.fenceEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (updateFrameData.fenceEvent == nullptr) {
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
}
void CrossNodeResources::LoadAssets()
{
// Create the root signatures.
// Root signatures may be shared across GPU nodes.
{
D3D12_FEATURE_DATA_ROOT_SIGNATURE featureData = {};
// This is the highest version the sample supports. If CheckFeatureSupport succeeds, the HighestVersion returned will not be greater than this.
featureData.HighestVersion = D3D_ROOT_SIGNATURE_VERSION_1_1;
if (FAILED(m_device->CheckFeatureSupport(D3D12_FEATURE_ROOT_SIGNATURE, &featureData, sizeof(featureData))))
{
featureData.HighestVersion = D3D_ROOT_SIGNATURE_VERSION_1_0;
}
// Create a root signature for rendering the triangle scene.
{
CD3DX12_DESCRIPTOR_RANGE1 sceneRanges[1];
sceneRanges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0, 0, D3D12_DESCRIPTOR_RANGE_FLAG_DATA_STATIC);
CD3DX12_ROOT_PARAMETER1 sceneRootParameters[2];
sceneRootParameters[0].InitAsDescriptorTable(1, &sceneRanges[0], D3D12_SHADER_VISIBILITY_VERTEX);
sceneRootParameters[1].InitAsConstants(1, 1, 0, D3D12_SHADER_VISIBILITY_VERTEX);
CD3DX12_VERSIONED_ROOT_SIGNATURE_DESC sceneRootSignatureDesc;
sceneRootSignatureDesc.Init_1_1(_countof(sceneRootParameters), sceneRootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3DX12SerializeVersionedRootSignature(&sceneRootSignatureDesc, featureData.HighestVersion, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(Settings::SharedNodeMask, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_sceneRootSignature)));
}
// Create a root signature for the post-process pass.
{
CD3DX12_DESCRIPTOR_RANGE1 postRanges[2];
postRanges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, Settings::SceneHistoryCount, 0, 0, D3D12_DESCRIPTOR_RANGE_FLAG_DATA_STATIC_WHILE_SET_AT_EXECUTE);
postRanges[1].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SAMPLER, 1, 0);
CD3DX12_ROOT_PARAMETER1 postRootParameters[3];
postRootParameters[0].InitAsDescriptorTable(1, &postRanges[0], D3D12_SHADER_VISIBILITY_PIXEL);
postRootParameters[1].InitAsDescriptorTable(1, &postRanges[1], D3D12_SHADER_VISIBILITY_PIXEL);
postRootParameters[2].InitAsConstants(2, 0, 0, D3D12_SHADER_VISIBILITY_PIXEL);
CD3DX12_VERSIONED_ROOT_SIGNATURE_DESC postRootSignatureDesc;
postRootSignatureDesc.Init_1_1(_countof(postRootParameters), postRootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3DX12SerializeVersionedRootSignature(&postRootSignatureDesc, featureData.HighestVersion, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(Settings::SharedNodeMask, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_postRootSignature)));
}
}
// Create the pipeline state, which includes compiling and loading shaders.
// Pipeline states may be shared across GPU nodes.
{
// Define the vertex input layout for the triangle scene.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
// Describe and create the graphics pipeline state objects (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_sceneRootSignature.Get();
psoDesc.VS = { g_SceneVS, sizeof(g_SceneVS) };
psoDesc.PS = { g_ScenePS, sizeof(g_ScenePS) };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
psoDesc.NodeMask = Settings::SharedNodeMask;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_scenePipelineState)));
// Define the vertex input layout for the post-process fullscreen quad.
D3D12_INPUT_ELEMENT_DESC postInputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
// Describe and create the PSO for the post-process pass.
D3D12_GRAPHICS_PIPELINE_STATE_DESC postPsoDesc = {};
postPsoDesc.InputLayout = { postInputElementDescs, _countof(postInputElementDescs) };
postPsoDesc.pRootSignature = m_postRootSignature.Get();
postPsoDesc.VS = { g_PostVS, sizeof(g_PostVS) };
postPsoDesc.PS = { g_PostPS, sizeof(g_PostPS) };
postPsoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
postPsoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
postPsoDesc.DepthStencilState.DepthEnable = FALSE;
postPsoDesc.DepthStencilState.StencilEnable = FALSE;
postPsoDesc.SampleMask = UINT_MAX;
postPsoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
postPsoDesc.NumRenderTargets = 1;
postPsoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
postPsoDesc.SampleDesc.Count = 1;
postPsoDesc.NodeMask = Settings::SharedNodeMask;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&postPsoDesc, IID_PPV_ARGS(&m_postPipelineState)));
}
// Create and map the constant buffers.
// Upload heaps live in system memory and can be made visible to all GPU nodes.
{
const UINT constantBufferDataSize = Settings::TriangleCount * Settings::SceneConstantBufferFrames * sizeof(SceneConstantBuffer);
D3D12_HEAP_PROPERTIES uploadHeapProps = CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD);
uploadHeapProps.VisibleNodeMask = Settings::SharedNodeMask;
ThrowIfFailed(m_device->CreateCommittedResource(
&uploadHeapProps,
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(constantBufferDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_sceneConstantBuffer)));
// Map the constant buffers. We don't unmap this until the app closes.
// Keeping things mapped for the lifetime of the resource is okay.
ThrowIfFailed(m_sceneConstantBuffer->Map(0, nullptr, reinterpret_cast<void**>(&m_mappedConstantBuffer)));
ZeroMemory(m_mappedConstantBuffer, constantBufferDataSize);
}
}
// Load the sample assets.
void D3D12Bundles::LoadAssets()
{
// These upload heaps are only needed during loading.
ComPtr<ID3D12Resource> vertexBufferUploadHeap;
ComPtr<ID3D12Resource> indexBufferUploadHeap;
ComPtr<ID3D12Resource> textureUploadHeap;
// Create the root signature.
{
CD3DX12_DESCRIPTOR_RANGE ranges[3];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
ranges[1].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SAMPLER, 1, 0);
ranges[2].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
CD3DX12_ROOT_PARAMETER rootParameters[3];
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_PIXEL);
rootParameters[1].InitAsDescriptorTable(1, &ranges[1], D3D12_SHADER_VISIBILITY_PIXEL);
rootParameters[2].InitAsDescriptorTable(1, &ranges[2], D3D12_SHADER_VISIBILITY_ALL);
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state, which includes loading shaders.
{
UINT8* pVertexShaderData;
UINT8* pPixelShaderData1;
UINT8* pPixelShaderData2;
UINT vertexShaderDataLength;
UINT pixelShaderDataLength1;
UINT pixelShaderDataLength2;
// Load pre-compiled shaders.
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(L"shader_mesh_simple_vert.cso").c_str(), &pVertexShaderData, &vertexShaderDataLength));
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(L"shader_mesh_simple_pixel.cso").c_str(), &pPixelShaderData1, &pixelShaderDataLength1));
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(L"shader_mesh_alt_pixel.cso").c_str(), &pPixelShaderData2, &pixelShaderDataLength2));
CD3DX12_RASTERIZER_DESC rasterizerStateDesc(D3D12_DEFAULT);
rasterizerStateDesc.CullMode = D3D12_CULL_MODE_NONE;
// Describe and create the graphics pipeline state objects (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { SampleAssets::StandardVertexDescription, SampleAssets::StandardVertexDescriptionNumElements };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { pVertexShaderData, vertexShaderDataLength };
psoDesc.PS = { pPixelShaderData1, pixelShaderDataLength1 };
psoDesc.RasterizerState = rasterizerStateDesc;
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState1)));
// Modify the description to use an alternate pixel shader and create
// a second PSO.
psoDesc.PS = { pPixelShaderData2, pixelShaderDataLength2 };
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState2)));
delete pVertexShaderData;
delete pPixelShaderData1;
delete pPixelShaderData2;
}
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocator.Get(), nullptr, IID_PPV_ARGS(&m_commandList)));
// Create render target views (RTVs).
CD3DX12_CPU_DESCRIPTOR_HANDLE rtvHandle(m_rtvHeap->GetCPUDescriptorHandleForHeapStart());
for (UINT i = 0; i < FrameCount; i++)
{
ThrowIfFailed(m_swapChain->GetBuffer(i, IID_PPV_ARGS(&m_renderTargets[i])));
m_device->CreateRenderTargetView(m_renderTargets[i].Get(), nullptr, rtvHandle);
rtvHandle.Offset(1, m_rtvDescriptorSize);
}
// Read in mesh data for vertex/index buffers.
UINT8* pMeshData;
UINT meshDataLength;
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(SampleAssets::DataFileName).c_str(), &pMeshData, &meshDataLength));
// Create the vertex buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::VertexDataSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::VertexDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&vertexBufferUploadHeap)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = pMeshData + SampleAssets::VertexDataOffset;
vertexData.RowPitch = SampleAssets::VertexDataSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_vertexBuffer.Get(), vertexBufferUploadHeap.Get(), 0, 0, 1, &vertexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = SampleAssets::StandardVertexStride;
m_vertexBufferView.SizeInBytes = SampleAssets::VertexDataSize;
}
// Create the index buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::IndexDataSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_indexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::IndexDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&indexBufferUploadHeap)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the index buffer.
D3D12_SUBRESOURCE_DATA indexData = {};
indexData.pData = pMeshData + SampleAssets::IndexDataOffset;
indexData.RowPitch = SampleAssets::IndexDataSize;
indexData.SlicePitch = indexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_indexBuffer.Get(), indexBufferUploadHeap.Get(), 0, 0, 1, &indexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_indexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_INDEX_BUFFER));
// Describe the index buffer view.
m_indexBufferView.BufferLocation = m_indexBuffer->GetGPUVirtualAddress();
m_indexBufferView.Format = SampleAssets::StandardIndexFormat;
m_indexBufferView.SizeInBytes = SampleAssets::IndexDataSize;
m_numIndices = SampleAssets::IndexDataSize / 4; // R32_UINT (SampleAssets::StandardIndexFormat) = 4 bytes each.
}
// Create the texture and sampler.
{
// Describe and create a Texture2D.
D3D12_RESOURCE_DESC textureDesc = {};
textureDesc.MipLevels = SampleAssets::Textures[0].MipLevels;
textureDesc.Format = SampleAssets::Textures[0].Format;
textureDesc.Width = SampleAssets::Textures[0].Width;
textureDesc.Height = SampleAssets::Textures[0].Height;
textureDesc.Flags = D3D12_RESOURCE_FLAG_NONE;
textureDesc.DepthOrArraySize = 1;
textureDesc.SampleDesc.Count = 1;
textureDesc.SampleDesc.Quality = 0;
textureDesc.Dimension = D3D12_RESOURCE_DIMENSION_TEXTURE2D;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&textureDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_texture)));
const UINT subresourceCount = textureDesc.DepthOrArraySize * textureDesc.MipLevels;
const UINT64 uploadBufferSize = GetRequiredIntermediateSize(m_texture.Get(), 0, subresourceCount);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(uploadBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&textureUploadHeap)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the Texture2D.
D3D12_SUBRESOURCE_DATA textureData = {};
textureData.pData = pMeshData + SampleAssets::Textures[0].Data[0].Offset;
textureData.RowPitch = SampleAssets::Textures[0].Data[0].Pitch;
textureData.SlicePitch = SampleAssets::Textures[0].Data[0].Size;
UpdateSubresources(m_commandList.Get(), m_texture.Get(), textureUploadHeap.Get(), 0, 0, subresourceCount, &textureData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_texture.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE));
// Describe and create a sampler.
D3D12_SAMPLER_DESC samplerDesc = {};
samplerDesc.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
samplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.MinLOD = 0;
samplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
samplerDesc.MipLODBias = 0.0f;
samplerDesc.MaxAnisotropy = 1;
samplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
m_device->CreateSampler(&samplerDesc, m_samplerHeap->GetCPUDescriptorHandleForHeapStart());
// Describe and create a SRV for the texture.
D3D12_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
srvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
srvDesc.Format = SampleAssets::Textures->Format;
srvDesc.ViewDimension = D3D12_SRV_DIMENSION_TEXTURE2D;
srvDesc.Texture2D.MipLevels = 1;
m_device->CreateShaderResourceView(m_texture.Get(), &srvDesc, m_cbvSrvHeap->GetCPUDescriptorHandleForHeapStart());
}
delete pMeshData;
// Create the depth stencil view.
{
D3D12_DEPTH_STENCIL_VIEW_DESC depthStencilDesc = {};
depthStencilDesc.Format = DXGI_FORMAT_D32_FLOAT;
depthStencilDesc.ViewDimension = D3D12_DSV_DIMENSION_TEXTURE2D;
depthStencilDesc.Flags = D3D12_DSV_FLAG_NONE;
D3D12_CLEAR_VALUE depthOptimizedClearValue = {};
depthOptimizedClearValue.Format = DXGI_FORMAT_D32_FLOAT;
depthOptimizedClearValue.DepthStencil.Depth = 1.0f;
depthOptimizedClearValue.DepthStencil.Stencil = 0;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Tex2D(DXGI_FORMAT_D32_FLOAT, m_width, m_height, 1, 0, 1, 0, D3D12_RESOURCE_FLAG_ALLOW_DEPTH_STENCIL),
D3D12_RESOURCE_STATE_DEPTH_WRITE,
&depthOptimizedClearValue,
IID_PPV_ARGS(&m_depthStencil)
));
m_device->CreateDepthStencilView(m_depthStencil.Get(), &depthStencilDesc, m_dsvHeap->GetCPUDescriptorHandleForHeapStart());
}
// Close the command list and execute it to begin the initial GPU setup.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
CreateFrameResources();
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValue, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValue++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
// Signal and increment the fence value.
const UINT64 fenceToWaitFor = m_fenceValue;
ThrowIfFailed(m_commandQueue->Signal(m_fence.Get(), fenceToWaitFor));
m_fenceValue++;
// Wait until the fence is completed.
ThrowIfFailed(m_fence->SetEventOnCompletion(fenceToWaitFor, m_fenceEvent));
WaitForSingleObject(m_fenceEvent, INFINITE);
}
}
// Load the sample assets.
void D3D12Fullscreen::LoadAssets()
{
// Create an empty root signature.
{
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
NAME_D3D12_OBJECT(m_rootSignature);
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
ComPtr<ID3DBlob> error;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, &error));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(vertexShader.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pixelShader.Get());
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
NAME_D3D12_OBJECT(m_pipelineState);
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocators[m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
NAME_D3D12_OBJECT(m_commandList);
LoadSizeDependentResources();
// Close the command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu();
}
}
// Load the sample assets.
void D3D12HelloConstBuffers::LoadAssets()
{
// Create a root signature consisting of a single CBV parameter.
{
CD3DX12_DESCRIPTOR_RANGE ranges[1];
CD3DX12_ROOT_PARAMETER rootParameters[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_VERTEX);
// Allow input layout and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_PIXEL_SHADER_ROOT_ACCESS;
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#if DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocator.Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
// Command lists are created in the recording state, but there is nothing
// to record yet. The main loop expects it to be closed, so close it now.
ThrowIfFailed(m_commandList->Close());
// Create the vertex buffer.
{
// Define the geometry for a triangle.
Vertex triangleVertices[] =
{
{ { 0.0f, 0.25f * m_aspectRatio, 0.0f }, { 1.0f, 0.0f, 0.0f, 1.0f } },
{ { 0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 1.0f, 0.0f, 1.0f } },
{ { -0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 0.0f, 1.0f, 1.0f } }
};
const UINT vertexBufferSize = sizeof(triangleVertices);
// Note: using upload heaps to transfer static data like vert buffers is not
// recommended. Every time the GPU needs it, the upload heap will be marshalled
// over. Please read up on Default Heap usage. An upload heap is used here for
// code simplicity and because there are very few verts to actually transfer.
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
// Copy the triangle data to the vertex buffer.
UINT8* pVertexDataBegin;
ThrowIfFailed(m_vertexBuffer->Map(0, nullptr, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, triangleVertices, sizeof(triangleVertices));
m_vertexBuffer->Unmap(0, nullptr);
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create the constant buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(1024 * 64),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_constantBuffer)));
// Describe and create a constant buffer view.
D3D12_CONSTANT_BUFFER_VIEW_DESC cbvDesc = {};
cbvDesc.BufferLocation = m_constantBuffer->GetGPUVirtualAddress();
cbvDesc.SizeInBytes = (sizeof(ConstantBuffer) + 255) & ~255; // CB size is required to be 256-byte aligned.
m_device->CreateConstantBufferView(&cbvDesc, m_cbvHeap->GetCPUDescriptorHandleForHeapStart());
// Initialize and map the constant buffers. We don't unmap this until the
// app closes. Keeping things mapped for the lifetime of the resource is okay.
ZeroMemory(&m_constantBufferData, sizeof(m_constantBufferData));
ThrowIfFailed(m_constantBuffer->Map(0, nullptr, reinterpret_cast<void**>(&m_pCbvDataBegin)));
memcpy(m_pCbvDataBegin, &m_constantBufferData, sizeof(m_constantBufferData));
}
// Create and record the bundle.
{
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_BUNDLE, m_bundleAllocator.Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_bundle)));
m_bundle->SetDescriptorHeaps(1, m_cbvHeap.GetAddressOf());
m_bundle->SetGraphicsRootSignature(m_rootSignature.Get());
m_bundle->IASetPrimitiveTopology(D3D_PRIMITIVE_TOPOLOGY_TRIANGLELIST);
m_bundle->IASetVertexBuffers(0, 1, &m_vertexBufferView);
m_bundle->SetGraphicsRootDescriptorTable(0, m_cbvHeap->GetGPUDescriptorHandleForHeapStart());
m_bundle->DrawInstanced(3, 1, 0, 0);
ThrowIfFailed(m_bundle->Close());
}
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValue = 1;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForPreviousFrame();
}
}
// Load the sample assets.
void D3D1211on12::LoadAssets()
{
// Create an empty root signature.
{
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_d3d12Device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#if DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_d3d12Device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
}
ThrowIfFailed(m_d3d12Device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocators[m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
// Create D2D/DWrite objects for rendering text.
{
ThrowIfFailed(m_d2dDeviceContext->CreateSolidColorBrush(D2D1::ColorF(D2D1::ColorF::Black), &m_textBrush));
ThrowIfFailed(m_dWriteFactory->CreateTextFormat(
L"Verdana",
NULL,
DWRITE_FONT_WEIGHT_NORMAL,
DWRITE_FONT_STYLE_NORMAL,
DWRITE_FONT_STRETCH_NORMAL,
50,
L"en-us",
&m_textFormat
));
ThrowIfFailed(m_textFormat->SetTextAlignment(DWRITE_TEXT_ALIGNMENT_CENTER));
ThrowIfFailed(m_textFormat->SetParagraphAlignment(DWRITE_PARAGRAPH_ALIGNMENT_CENTER));
}
ComPtr<ID3D12Resource> vertexBufferUpload;
// Create the vertex buffer.
{
// Define the geometry for a triangle.
Vertex triangleVertices[] =
{
{ { 0.0f, 0.25f * m_aspectRatio, 0.0f }, { 1.0f, 0.0f, 0.0f, 1.0f } },
{ { 0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 1.0f, 0.0f, 1.0f } },
{ { -0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 0.0f, 1.0f, 1.0f } }
};
const UINT vertexBufferSize = sizeof(triangleVertices);
ThrowIfFailed(m_d3d12Device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_d3d12Device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&vertexBufferUpload)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = reinterpret_cast<UINT8*>(triangleVertices);
vertexData.RowPitch = vertexBufferSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_vertexBuffer.Get(), vertexBufferUpload.Get(), 0, 0, 1, &vertexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = vertexBufferSize;
}
// Close the command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_d3d12Device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(m_fence.GetAddressOf())));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu();
}
}
// Load the sample assets.
void D3D12HeterogeneousMultiadapter::LoadAssets()
{
// Create the root signatures.
{
CD3DX12_ROOT_PARAMETER rootParameters[2];
rootParameters[0].InitAsConstantBufferView(0, 0, D3D12_SHADER_VISIBILITY_VERTEX);
rootParameters[1].InitAsConstantBufferView(1, 0, D3D12_SHADER_VISIBILITY_PIXEL);
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_devices[Primary]->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
CD3DX12_DESCRIPTOR_RANGE ranges[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
CD3DX12_ROOT_PARAMETER blurRootParameters[3];
blurRootParameters[0].InitAsConstantBufferView(0, 0, D3D12_SHADER_VISIBILITY_PIXEL);
blurRootParameters[1].InitAsDescriptorTable(_countof(ranges), ranges, D3D12_SHADER_VISIBILITY_PIXEL);
blurRootParameters[2].InitAsConstantBufferView(1, 0, D3D12_SHADER_VISIBILITY_PIXEL);
CD3DX12_STATIC_SAMPLER_DESC staticPointSampler(0);
staticPointSampler.Filter = D3D12_FILTER_MIN_MAG_MIP_POINT;
staticPointSampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_PIXEL;
CD3DX12_STATIC_SAMPLER_DESC staticLinearSampler(1);
staticLinearSampler.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
staticLinearSampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_PIXEL;
D3D12_STATIC_SAMPLER_DESC staticSamplers[] = { staticPointSampler, staticLinearSampler };
rootSignatureDesc.Init(_countof(blurRootParameters), blurRootParameters, _countof(staticSamplers), staticSamplers, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_devices[Secondary]->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_blurRootSignature)));
}
// Create the pipeline states, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
ComPtr<ID3DBlob> vertexShaderBlur;
ComPtr<ID3DBlob> pixelShaderBlurU;
ComPtr<ID3DBlob> pixelShaderBlurV;
ComPtr<ID3DBlob> error;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VShader", "vs_5_0", compileFlags, 0, &vertexShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PShader", "ps_5_0", compileFlags, 0, &pixelShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"blurShaders.hlsl").c_str(), nullptr, nullptr, "VSSimpleBlur", "vs_5_0", compileFlags, 0, &vertexShaderBlur, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"blurShaders.hlsl").c_str(), nullptr, nullptr, "PSSimpleBlurU", "ps_5_0", compileFlags, 0, &pixelShaderBlurU, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"blurShaders.hlsl").c_str(), nullptr, nullptr, "PSSimpleBlurV", "ps_5_0", compileFlags, 0, &pixelShaderBlurV, &error));
// Define the vertex input layouts.
const D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
const D3D12_INPUT_ELEMENT_DESC blurInputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
// Describe and create the graphics pipeline state objects (PSOs).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(vertexShader.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pixelShader.Get());
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_devices[Primary]->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
psoDesc.InputLayout = { blurInputElementDescs, _countof(blurInputElementDescs) };
psoDesc.pRootSignature = m_blurRootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(vertexShaderBlur.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pixelShaderBlurU.Get());
psoDesc.DepthStencilState.DepthEnable = false;
psoDesc.DSVFormat = DXGI_FORMAT_UNKNOWN;
ThrowIfFailed(m_devices[Secondary]->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_blurPipelineStates[0])));
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pixelShaderBlurV.Get());
ThrowIfFailed(m_devices[Secondary]->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_blurPipelineStates[1])));
}
// Create the command lists.
ThrowIfFailed(m_devices[Primary]->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_directCommandAllocators[Primary][m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_directCommandLists[Primary])));
ThrowIfFailed(m_devices[Primary]->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_COPY, m_copyCommandAllocators[m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_copyCommandList)));
ThrowIfFailed(m_copyCommandList->Close());
ThrowIfFailed(m_devices[Secondary]->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_directCommandAllocators[Secondary][m_frameIndex].Get(), m_blurPipelineStates[0].Get(), IID_PPV_ARGS(&m_directCommandLists[Secondary])));
// Note: ComPtr's are CPU objects but these resources need to stay in scope until
// the command list that references them has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resources are not
// prematurely destroyed.
ComPtr<ID3D12Resource> vertexBufferUpload;
ComPtr<ID3D12Resource> fullscreenQuadVertexBufferUpload;
// Create the vertex buffer for the primary adapter.
{
// Define the geometry for a triangle.
Vertex triangleVertices[] =
{
{ { 0.0f, TriangleHalfWidth, TriangleDepth } },
{ { TriangleHalfWidth, -TriangleHalfWidth, TriangleDepth } },
{ { -TriangleHalfWidth, -TriangleHalfWidth, TriangleDepth } }
};
const UINT vertexBufferSize = sizeof(triangleVertices);
ThrowIfFailed(m_devices[Primary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_devices[Primary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&vertexBufferUpload)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = reinterpret_cast<UINT8*>(triangleVertices);
vertexData.RowPitch = vertexBufferSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_directCommandLists[Primary].Get(), m_vertexBuffer.Get(), vertexBufferUpload.Get(), 0, 0, 1, &vertexData);
m_directCommandLists[Primary]->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = sizeof(triangleVertices);
}
// Create the vertex buffer for the secondary adapter.
{
// Define the geometry for a fullscreen triangle.
VertexPositionUV quadVertices[] =
{
{ { -1.0f, -1.0f, 0.0f, 1.0f }, { 0.0f, 0.0f } }, // Bottom left.
{ { -1.0f, 1.0f, 0.0f, 1.0f }, { 0.0f, 1.0f } }, // Top left.
{ { 1.0f, -1.0f, 0.0f, 1.0f }, { 1.0f, 0.0f } }, // Bottom right.
{ { 1.0f, 1.0f, 0.0f, 1.0f }, { 1.0f, 1.0f } }, // Top right.
};
const UINT vertexBufferSize = sizeof(quadVertices);
ThrowIfFailed(m_devices[Secondary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_fullscreenQuadVertexBuffer)));
ThrowIfFailed(m_devices[Secondary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&fullscreenQuadVertexBufferUpload)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = reinterpret_cast<UINT8*>(quadVertices);
vertexData.RowPitch = vertexBufferSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_directCommandLists[Secondary].Get(), m_fullscreenQuadVertexBuffer.Get(), fullscreenQuadVertexBufferUpload.Get(), 0, 0, 1, &vertexData);
m_directCommandLists[Secondary]->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_fullscreenQuadVertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_fullscreenQuadVertexBufferView.BufferLocation = m_fullscreenQuadVertexBuffer->GetGPUVirtualAddress();
m_fullscreenQuadVertexBufferView.StrideInBytes = sizeof(VertexPositionUV);
m_fullscreenQuadVertexBufferView.SizeInBytes = sizeof(quadVertices);
}
// Create the depth stencil view.
{
D3D12_DEPTH_STENCIL_VIEW_DESC depthStencilDesc = {};
depthStencilDesc.Format = DXGI_FORMAT_D32_FLOAT;
depthStencilDesc.ViewDimension = D3D12_DSV_DIMENSION_TEXTURE2D;
depthStencilDesc.Flags = D3D12_DSV_FLAG_NONE;
const CD3DX12_CLEAR_VALUE clearValue(DXGI_FORMAT_D32_FLOAT, 1.0f, 0);
ThrowIfFailed(m_devices[Primary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Tex2D(DXGI_FORMAT_D32_FLOAT, m_width, m_height, 1, 0, 1, 0, D3D12_RESOURCE_FLAG_ALLOW_DEPTH_STENCIL),
D3D12_RESOURCE_STATE_DEPTH_WRITE,
&clearValue,
IID_PPV_ARGS(&m_depthStencil)
));
m_devices[Primary]->CreateDepthStencilView(m_depthStencil.Get(), &depthStencilDesc, m_dsvHeap->GetCPUDescriptorHandleForHeapStart());
}
// Create the constant buffers.
{
{
const UINT64 constantBufferSize = sizeof(ConstantBufferData) * MaxTriangleCount * FrameCount;
ThrowIfFailed(m_devices[Primary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(constantBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_constantBuffer)));
// Setup constant buffer data.
for (UINT n = 0; n < MaxTriangleCount; n++)
{
m_constantBufferData[n].velocity = XMFLOAT4(GetRandomFloat(0.01f, 0.02f), 0.0f, 0.0f, 0.0f);
m_constantBufferData[n].offset = XMFLOAT4(GetRandomFloat(-5.0f, -1.5f), GetRandomFloat(-1.0f, 1.0f), GetRandomFloat(0.0f, 2.0f), 0.0f);
m_constantBufferData[n].color = XMFLOAT4(GetRandomFloat(0.5f, 1.0f), GetRandomFloat(0.5f, 1.0f), GetRandomFloat(0.5f, 1.0f), 1.0f);
XMStoreFloat4x4(&m_constantBufferData[n].projection, XMMatrixTranspose(XMMatrixPerspectiveFovLH(XM_PIDIV4, m_aspectRatio, 0.01f, 20.0f)));
}
// Map the constant buffer. We don't unmap this until the app closes.
// Keeping things mapped for the lifetime of the resource is okay.
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_constantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pCbvDataBegin)));
memcpy(m_pCbvDataBegin, &m_constantBufferData[0], constantBufferSize / FrameCount);
}
{
const UINT64 workloadConstantBufferSize = sizeof(WorkloadConstantBufferData) * FrameCount;
ThrowIfFailed(m_devices[Primary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(workloadConstantBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_workloadConstantBuffer)));
// Setup constant buffer data.
m_workloadConstantBufferData.loopCount = m_psLoopCount;
// Map the constant buffer. We don't unmap this until the app closes.
// Keeping things mapped for the lifetime of the resource is okay.
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_workloadConstantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pWorkloadCbvDataBegin)));
memcpy(m_pWorkloadCbvDataBegin, &m_workloadConstantBufferData, workloadConstantBufferSize / FrameCount);
}
{
const UINT64 blurWorkloadConstantBufferSize = sizeof(WorkloadConstantBufferData) * FrameCount;
ThrowIfFailed(m_devices[Secondary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(blurWorkloadConstantBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_blurWorkloadConstantBuffer)));
// Setup constant buffer data.
m_blurWorkloadConstantBufferData.loopCount = m_blurPSLoopCount;
// Map the constant buffer. We don't unmap this until the app closes.
// Keeping things mapped for the lifetime of the resource is okay.
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_blurWorkloadConstantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pBlurWorkloadCbvDataBegin)));
memcpy(m_pBlurWorkloadCbvDataBegin, &m_blurWorkloadConstantBufferData, blurWorkloadConstantBufferSize / FrameCount);
}
{
ThrowIfFailed(m_devices[Secondary]->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(sizeof(BlurConstantBufferData)),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_blurConstantBuffer)));
// Map the constant buffer.
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_blurConstantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pBlurCbvDataBegin)));
// Setup constant buffer data.
m_pBlurCbvDataBegin[0].offset = 0.5f;
m_pBlurCbvDataBegin[0].textureDimensions.x = static_cast<float>(m_width);
m_pBlurCbvDataBegin[0].textureDimensions.y = static_cast<float>(m_height);
// Unmap the constant buffer because we don't update this again.
// If we ever do, it should be buffered by the number of frames like other constant buffers.
const CD3DX12_RANGE emptyRange(0, 0);
m_blurConstantBuffer->Unmap(0, &emptyRange);
m_pBlurCbvDataBegin = nullptr;
}
}
// Close the command lists and execute them to begin the vertex buffer copies into the default heaps.
for (UINT i = 0; i < GraphicsAdaptersCount; i++)
{
ThrowIfFailed(m_directCommandLists[i]->Close());
ID3D12CommandList* ppCommandLists[] = { m_directCommandLists[i].Get() };
m_directCommandQueues[i]->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
}
// Create synchronization objects and wait until assets have been uploaded to the GPU.
// We use a cross-adapter fence for handling Signals and Waits between adapters.
// We use regular fences for things that don't need to be cross adapter because they don't need the additional overhead associated with being cross-adapter.
{
// Fence used to control CPU pacing.
ThrowIfFailed(m_devices[Secondary]->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_frameFence)));
// Fence used by the primary adapter to signal its copy queue that it has completed rendering.
// When this is signaled, the primary adapter's copy queue can begin copying to the cross-adapter shared resource.
ThrowIfFailed(m_devices[Primary]->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_renderFence)));
// Cross-adapter shared fence used by both adapters.
// Used by the primary adapter to signal the secondary adapter that it has completed copying to the cross-adapter shared resource.
// When this is signaled, the secondary adapter can begin its work.
ThrowIfFailed(m_devices[Primary]->CreateFence(0, D3D12_FENCE_FLAG_SHARED | D3D12_FENCE_FLAG_SHARED_CROSS_ADAPTER, IID_PPV_ARGS(&m_crossAdapterFences[Primary])));
// For now, require GENERIC_ALL access.
HANDLE fenceHandle = nullptr;
ThrowIfFailed(m_devices[Primary]->CreateSharedHandle(
m_crossAdapterFences[Primary].Get(),
nullptr,
GENERIC_ALL,
nullptr,
&fenceHandle));
HRESULT openSharedHandleResult = m_devices[Secondary]->OpenSharedHandle(fenceHandle, IID_PPV_ARGS(&m_crossAdapterFences[Secondary]));
// We can close the handle after opening the cross-adapter shared fence.
CloseHandle(fenceHandle);
ThrowIfFailed(openSharedHandleResult);
for (UINT i = 0; i < GraphicsAdaptersCount; i++)
{
// Create an event handle to use for frame synchronization.
m_fenceEvents[i] = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (m_fenceEvents == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu(static_cast<GraphicsAdapter>(i));
}
}
}
void dx::initialise()
{
gDX.activeContextState = nullptr;
gDX.activeDisplayList = nullptr;
gDX.activeDisplayListSize = 0;
gDX.activeDisplayListOffset = 0;
for (auto i = 0; i < GX2_NUM_MRT_BUFFER; ++i) {
gDX.activeColorBuffer[i] = nullptr;
}
gDX.activeDepthBuffer = nullptr;
gDX.state.primitiveRestartIdx = 0xFFFFFFFF;
gDX.viewport.Width = static_cast<float>(platform::ui::getWindowWidth());
gDX.viewport.Height = static_cast<float>(platform::ui::getWindowHeight());
gDX.viewport.MaxDepth = 1.0f;
gDX.scissorRect.right = static_cast<LONG>(platform::ui::getWindowWidth());
gDX.scissorRect.bottom = static_cast<LONG>(platform::ui::getWindowHeight());
// Enable the D3D12 debug layer.
{
ComPtr<ID3D12Debug> debugController;
if (SUCCEEDED(D3D12GetDebugInterface(IID_PPV_ARGS(&debugController))))
{
debugController->EnableDebugLayer();
}
}
ComPtr<IDXGIFactory4> factory;
ThrowIfFailed(CreateDXGIFactory1(IID_PPV_ARGS(&factory)));
// Always use WARP for now...
static const bool USE_WARP_DEVICE = true;
if (USE_WARP_DEVICE)
{
ComPtr<IDXGIAdapter> warpAdapter;
ThrowIfFailed(factory->EnumWarpAdapter(IID_PPV_ARGS(&warpAdapter)));
ThrowIfFailed(D3D12CreateDevice(
warpAdapter.Get(),
D3D_FEATURE_LEVEL_11_0,
IID_PPV_ARGS(&gDX.device)
));
} else
{
ThrowIfFailed(D3D12CreateDevice(
nullptr,
D3D_FEATURE_LEVEL_11_0,
IID_PPV_ARGS(&gDX.device)
));
}
// Describe and create the command queue.
D3D12_COMMAND_QUEUE_DESC queueDesc = {};
queueDesc.Flags = D3D12_COMMAND_QUEUE_FLAG_NONE;
queueDesc.Type = D3D12_COMMAND_LIST_TYPE_DIRECT;
ThrowIfFailed(gDX.device->CreateCommandQueue(&queueDesc, IID_PPV_ARGS(&gDX.commandQueue)));
// Describe and create the swap chain.
DXGI_SWAP_CHAIN_DESC swapChainDesc = {};
swapChainDesc.BufferCount = gDX.FrameCount;
swapChainDesc.BufferDesc.Width = platform::ui::getWindowWidth();
swapChainDesc.BufferDesc.Height = platform::ui::getWindowHeight();
swapChainDesc.BufferDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
swapChainDesc.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT;
swapChainDesc.SwapEffect = DXGI_SWAP_EFFECT_FLIP_DISCARD;
swapChainDesc.OutputWindow = reinterpret_cast<HWND>(platform::ui::getWindowHandle());
swapChainDesc.SampleDesc.Count = 1;
swapChainDesc.Windowed = TRUE;
ComPtr<IDXGISwapChain> swapChain;
ThrowIfFailed(factory->CreateSwapChain(
gDX.commandQueue.Get(), // Swap chain needs the queue so that it can force a flush on it.
&swapChainDesc,
&swapChain
));
ThrowIfFailed(swapChain.As(&gDX.swapChain));
gDX.frameIndex = gDX.swapChain->GetCurrentBackBufferIndex();
// Create descriptor heaps.
{
// Describe and create a render target view (RTV) descriptor heap.
D3D12_DESCRIPTOR_HEAP_DESC rtvHeapDesc = {};
rtvHeapDesc.NumDescriptors = gDX.FrameCount + 128;
rtvHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_RTV;
rtvHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
gDX.rtvHeap = new DXHeap(gDX.device.Get(), rtvHeapDesc);
// Describe and create a depth stencil view (DSV) descriptor heap.
D3D12_DESCRIPTOR_HEAP_DESC dsvHeapDesc = {};
dsvHeapDesc.NumDescriptors = 128;
dsvHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_DSV;
dsvHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
gDX.dsvHeap = new DXHeap(gDX.device.Get(), dsvHeapDesc);
// Describe and create a constant buffer view (CBV), Shader resource
// view (SRV), and unordered access view (UAV) descriptor heap.
D3D12_DESCRIPTOR_HEAP_DESC srvHeapDesc = {};
srvHeapDesc.NumDescriptors = 2048;
srvHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV;
srvHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_SHADER_VISIBLE;
gDX.srvHeap = new DXHeap(gDX.device.Get(), srvHeapDesc);
}
// Create frame resources.
{
// Create a RTV for each frame.
for (UINT n = 0; n < gDX.FrameCount; n++)
{
ThrowIfFailed(gDX.swapChain->GetBuffer(n, IID_PPV_ARGS(&gDX.renderTargets[n])));
gDX.scanbufferRtv[n] = gDX.rtvHeap->alloc();
gDX.device->CreateRenderTargetView(gDX.renderTargets[n].Get(), nullptr, *gDX.scanbufferRtv[n]);
}
}
// Create command allocators.
{
for (UINT n = 0; n < gDX.FrameCount; n++)
{
ThrowIfFailed(gDX.device->CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT, IID_PPV_ARGS(&gDX.commandAllocator[n])));
}
}
// Create the root signature.
{
CD3DX12_DESCRIPTOR_RANGE ranges[GX2_NUM_SAMPLERS];
CD3DX12_ROOT_PARAMETER rootParameters[GX2_NUM_SAMPLERS + GX2_NUM_UNIFORMBLOCKS];
uint32_t paramIdx = 0;
for (auto i = 0; i < GX2_NUM_UNIFORMBLOCKS; ++i) {
gDX.cbvIndex[i] = paramIdx;
rootParameters[paramIdx++].InitAsConstantBufferView(i, 0, D3D12_SHADER_VISIBILITY_ALL);
}
for (auto i = 0; i < GX2_NUM_SAMPLERS; ++i) {
gDX.srvIndex[i] = paramIdx;
ranges[i].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, i);
rootParameters[paramIdx++].InitAsDescriptorTable(1, &ranges[i], D3D12_SHADER_VISIBILITY_ALL);
}
D3D12_STATIC_SAMPLER_DESC samplers[GX2_NUM_SAMPLERS];
for (int i = 0; i < GX2_NUM_SAMPLERS; ++i) {
auto &sampler = samplers[i] = {};
sampler.Filter = D3D12_FILTER_MIN_MAG_MIP_POINT;
sampler.AddressU = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressV = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressW = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.MipLODBias = 0;
sampler.MaxAnisotropy = 0;
sampler.ComparisonFunc = D3D12_COMPARISON_FUNC_NEVER;
sampler.BorderColor = D3D12_STATIC_BORDER_COLOR_TRANSPARENT_BLACK;
sampler.MinLOD = 0.0f;
sampler.MaxLOD = D3D12_FLOAT32_MAX;
sampler.ShaderRegister = i;
sampler.RegisterSpace = 0;
sampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_PIXEL;
}
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, GX2_NUM_SAMPLERS, samplers, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(gDX.device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&gDX.rootSignature)));
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#ifdef _DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(L"resources/shaders/screendraw.hlsl", nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(L"resources/shaders/screendraw.hlsl", nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
{
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = gDX.rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(gDX.device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&gDX.emuPipelineState)));
}
}
gDX.pipelineMgr = new DXPipelineMgr();
// Create the command list.
gDX.frameIndex = gDX.swapChain->GetCurrentBackBufferIndex();
ThrowIfFailed(gDX.device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, gDX.commandAllocator[gDX.frameIndex].Get(), gDX.emuPipelineState.Get(), IID_PPV_ARGS(&gDX.commandList)));
{
#define SCREENSPACE(x, y) { -1 + ((x) / (float)platform::ui::getWindowWidth()) * 2, 1 - ((y) / (float)platform::ui::getWindowHeight()) * 2, 0.0f }
float tvX = 0;
float tvY = 0;
float tvWidth = static_cast<float>(platform::ui::getTvWidth());
float tvHeight = static_cast<float>(platform::ui::getTvHeight());
float drcX = (tvWidth - static_cast<float>(platform::ui::getDrcWidth())) / 2.0f;
float drcY = tvHeight;
float drcWidth = static_cast<float>(platform::ui::getDrcWidth());
float drcHeight = static_cast<float>(platform::ui::getDrcHeight());
struct Vertex {
XMFLOAT3 pos;
XMFLOAT2 uv;
} triangleVertices[] =
{
{ SCREENSPACE(tvX, tvY + tvHeight),{ 0.0f, 1.0f } },
{ SCREENSPACE(tvX, tvY),{ 0.0f, 0.0f } },
{ SCREENSPACE(tvX + tvWidth, tvY + tvHeight),{ 1.0f, 1.0f } },
{ SCREENSPACE(tvX + tvWidth, tvY),{ 1.0f, 0.0f } },
{ SCREENSPACE(drcX, drcY + drcHeight),{ 0.0f, 1.0f } },
{ SCREENSPACE(drcX, drcY),{ 0.0f, 0.0f } },
{ SCREENSPACE(drcX + drcWidth, drcY + drcHeight),{ 1.0f, 1.0f } },
{ SCREENSPACE(drcX + drcWidth, drcY),{ 1.0f, 0.0f } },
};
#undef SCREENSPACE
const UINT vertexBufferSize = sizeof(triangleVertices);
// Note: using upload heaps to transfer static data like vert buffers is not
// recommended. Every time the GPU needs it, the upload heap will be marshalled
// over. Please read up on Default Heap usage. An upload heap is used here for
// code simplicity and because there are very few verts to actually transfer.
ThrowIfFailed(gDX.device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&gDX.vertexBuffer)));
// Copy the triangle data to the vertex buffer.
UINT8* pVertexDataBegin;
ThrowIfFailed(gDX.vertexBuffer->Map(0, nullptr, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, triangleVertices, sizeof(triangleVertices));
gDX.vertexBuffer->Unmap(0, nullptr);
// Initialize the vertex buffer view.
gDX.vertexBufferView.BufferLocation = gDX.vertexBuffer->GetGPUVirtualAddress();
gDX.vertexBufferView.StrideInBytes = sizeof(Vertex);
gDX.vertexBufferView.SizeInBytes = vertexBufferSize;
}
// 10MB Temporary Vertex Buffer
gDX.ppcVertexBuffer = new DXDynBuffer(gDX.device.Get(), 10 * 1024 * 1024);
// Close the command list and execute it to begin the initial GPU setup.
ThrowIfFailed(gDX.commandList->Close());
ID3D12CommandList* ppCommandLists[] = { gDX.commandList.Get() };
gDX.commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects.
{
ThrowIfFailed(gDX.device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&gDX.fence)));
// Create an event handle to use for frame synchronization.
gDX.fenceEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (gDX.fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for frame completion
gDX.swapCount++;
const uint64_t fenceValue = gDX.swapCount;
ThrowIfFailed(gDX.commandQueue->Signal(gDX.fence.Get(), fenceValue));
if (gDX.fence->GetCompletedValue() < gDX.swapCount)
{
ThrowIfFailed(gDX.fence->SetEventOnCompletion(gDX.swapCount, gDX.fenceEvent));
WaitForSingleObject(gDX.fenceEvent, INFINITE);
}
}
_beginFrame();
}
// Load the sample assets.
void D3D12SmallResources::LoadAssets()
{
// Create a root signature consisting of a single CBV parameter.
{
CD3DX12_DESCRIPTOR_RANGE ranges[1];
CD3DX12_ROOT_PARAMETER rootParameters[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_PIXEL);
// Allow input layout and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_VERTEX_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS;
CD3DX12_STATIC_SAMPLER_DESC samplerDesc(0, D3D12_FILTER_MIN_MAG_MIP_LINEAR);
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 1, &samplerDesc, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
NAME_D3D12_OBJECT(m_rootSignature);
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state objects (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(vertexShader.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pixelShader.Get());
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
NAME_D3D12_OBJECT(m_pipelineState);
}
// Create the command lists.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocators[m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
NAME_D3D12_OBJECT(m_commandList);
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_COPY, m_copyCommandAllocator.Get(), nullptr, IID_PPV_ARGS(&m_copyCommandList)));
ThrowIfFailed(m_copyCommandList->Close());
NAME_D3D12_OBJECT(m_copyCommandList);
// Note: ComPtr's are CPU objects but this resource needs to stay in scope until
// the command list that references it has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resource is not
// prematurely destroyed.
ComPtr<ID3D12Resource> vertexBufferUpload;
// Create the vertex buffer.
{
// Create quads for all of the images that will be generated and drawn to the screen.
Vertex quadVertices[TextureCount * 4];
UINT index = 0;
float offsetX = 0.15f;
float marginX = offsetX / 10.0f;;
float startX = (GridWidth / 2.0f) * -(offsetX + marginX) + marginX / 2.0f;
float offsetY = offsetX * m_aspectRatio;
float marginY = offsetY / 10.0f;
float y = (GridHeight / 2.0f) * (offsetY + marginY) - marginY / 2.0f;
for (UINT row = 0; row < GridHeight; row++)
{
float x = startX;
for (UINT column = 0; column < GridWidth; column++)
{
quadVertices[index++] = { { x, y - offsetY, 0.0f }, { 0.0f, 0.0f } };
quadVertices[index++] = { { x, y, 0.0f }, { 0.0f, 1.0f } };
quadVertices[index++] = { { x + offsetX, y - offsetY, 0.0f }, { 1.0f, 0.0f } };
quadVertices[index++] = { { x + offsetX, y, 0.0f }, { 1.0f, 1.0f } };
x += offsetX + marginX;
}
y -= offsetY + marginY;
}
const UINT vertexBufferSize = sizeof(quadVertices);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&vertexBufferUpload)));
NAME_D3D12_OBJECT(m_vertexBuffer);
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = reinterpret_cast<UINT8*>(quadVertices);
vertexData.RowPitch = vertexBufferSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_vertexBuffer.Get(), vertexBufferUpload.Get(), 0, 0, 1, &vertexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = sizeof(quadVertices);
}
// Close the command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu();
}
CreateTextures();
}
// Load the sample assets.
void D3D12ExecuteIndirect::LoadAssets()
{
// Create the root signatures.
{
CD3DX12_ROOT_PARAMETER rootParameters[GraphicsRootParametersCount];
rootParameters[Cbv].InitAsConstantBufferView(0, 0, D3D12_SHADER_VISIBILITY_VERTEX);
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
// Create compute signature.
CD3DX12_DESCRIPTOR_RANGE ranges[2];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 2, 0);
ranges[1].Init(D3D12_DESCRIPTOR_RANGE_TYPE_UAV, 1, 0);
CD3DX12_ROOT_PARAMETER computeRootParameters[ComputeRootParametersCount];
computeRootParameters[SrvUavTable].InitAsDescriptorTable(2, ranges);
computeRootParameters[RootConstants].InitAsConstants(4, 0);
CD3DX12_ROOT_SIGNATURE_DESC computeRootSignatureDesc;
computeRootSignatureDesc.Init(_countof(computeRootParameters), computeRootParameters);
ThrowIfFailed(D3D12SerializeRootSignature(&computeRootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_computeRootSignature)));
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
ComPtr<ID3DBlob> computeShader;
ComPtr<ID3DBlob> error;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"compute.hlsl").c_str(), nullptr, nullptr, "CSMain", "cs_5_0", compileFlags, 0, &computeShader, &error));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
// Describe and create the graphics pipeline state objects (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
// Describe and create the compute pipeline state object (PSO).
D3D12_COMPUTE_PIPELINE_STATE_DESC computePsoDesc = {};
computePsoDesc.pRootSignature = m_computeRootSignature.Get();
computePsoDesc.CS = { reinterpret_cast<UINT8*>(computeShader->GetBufferPointer()), computeShader->GetBufferSize() };
ThrowIfFailed(m_device->CreateComputePipelineState(&computePsoDesc, IID_PPV_ARGS(&m_computeState)));
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocators[m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_COMPUTE, m_computeCommandAllocators[m_frameIndex].Get(), m_computeState.Get(), IID_PPV_ARGS(&m_computeCommandList)));
ThrowIfFailed(m_computeCommandList->Close());
// Note: ComPtr's are CPU objects but these resources need to stay in scope until
// the command list that references them has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resources are not
// prematurely destroyed.
ComPtr<ID3D12Resource> vertexBufferUpload;
ComPtr<ID3D12Resource> commandBufferUpload;
// Create the vertex buffer.
{
// Define the geometry for a triangle.
Vertex triangleVertices[] =
{
{ { 0.0f, TriangleHalfWidth, TriangleDepth } },
{ { TriangleHalfWidth, -TriangleHalfWidth, TriangleDepth } },
{ { -TriangleHalfWidth, -TriangleHalfWidth, TriangleDepth } }
};
const UINT vertexBufferSize = sizeof(triangleVertices);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&vertexBufferUpload)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = reinterpret_cast<UINT8*>(triangleVertices);
vertexData.RowPitch = vertexBufferSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_vertexBuffer.Get(), vertexBufferUpload.Get(), 0, 0, 1, &vertexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = sizeof(triangleVertices);
}
// Create the depth stencil view.
{
D3D12_DEPTH_STENCIL_VIEW_DESC depthStencilDesc = {};
depthStencilDesc.Format = DXGI_FORMAT_D32_FLOAT;
depthStencilDesc.ViewDimension = D3D12_DSV_DIMENSION_TEXTURE2D;
depthStencilDesc.Flags = D3D12_DSV_FLAG_NONE;
D3D12_CLEAR_VALUE depthOptimizedClearValue = {};
depthOptimizedClearValue.Format = DXGI_FORMAT_D32_FLOAT;
depthOptimizedClearValue.DepthStencil.Depth = 1.0f;
depthOptimizedClearValue.DepthStencil.Stencil = 0;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Tex2D(DXGI_FORMAT_D32_FLOAT, m_width, m_height, 1, 0, 1, 0, D3D12_RESOURCE_FLAG_ALLOW_DEPTH_STENCIL),
D3D12_RESOURCE_STATE_DEPTH_WRITE,
&depthOptimizedClearValue,
IID_PPV_ARGS(&m_depthStencil)
));
m_device->CreateDepthStencilView(m_depthStencil.Get(), &depthStencilDesc, m_dsvHeap->GetCPUDescriptorHandleForHeapStart());
}
// Create the constant buffers.
{
const UINT constantBufferDataSize = TriangleResourceCount * sizeof(ConstantBufferData);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(constantBufferDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_constantBuffer)));
D3D12_CONSTANT_BUFFER_VIEW_DESC cbvDesc = {};
cbvDesc.SizeInBytes = sizeof(ConstantBufferData);
// Create constant buffer views to access the upload buffer.
for (UINT n = 0; n < TriangleCount; n++)
{
m_constantBufferData[n].velocity = XMFLOAT4(GetRandomFloat(0.01f, 0.02f), 0.0f, 0.0f, 0.0f);
m_constantBufferData[n].offset = XMFLOAT4(GetRandomFloat(-5.0f, -1.5f), GetRandomFloat(-1.0f, 1.0f), GetRandomFloat(0.0f, 2.0f), 0.0f);
m_constantBufferData[n].color = XMFLOAT4(GetRandomFloat(0.5f, 1.0f), GetRandomFloat(0.5f, 1.0f), GetRandomFloat(0.5f, 1.0f), 1.0f);
XMStoreFloat4x4(&m_constantBufferData[n].projection, XMMatrixTranspose(XMMatrixPerspectiveFovLH(XM_PIDIV4, m_aspectRatio, 0.01f, 20.0f)));
}
// Map the constant buffers. We don't unmap this until the app closes.
// Keeping things mapped for the lifetime of the resource is okay.
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_constantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pCbvDataBegin)));
memcpy(m_pCbvDataBegin, &m_constantBufferData[0], TriangleCount * sizeof(ConstantBufferData));
// Create shader resource views (SRV) of the constant buffers for the
// compute shader to read from.
D3D12_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
srvDesc.Format = DXGI_FORMAT_UNKNOWN;
srvDesc.ViewDimension = D3D12_SRV_DIMENSION_BUFFER;
srvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
srvDesc.Buffer.NumElements = TriangleCount;
srvDesc.Buffer.StructureByteStride = sizeof(ConstantBufferData);
srvDesc.Buffer.Flags = D3D12_BUFFER_SRV_FLAG_NONE;
CD3DX12_CPU_DESCRIPTOR_HANDLE cbvSrvHandle(m_cbvSrvUavHeap->GetCPUDescriptorHandleForHeapStart(), CbvSrvOffset, m_cbvSrvUavDescriptorSize);
for (UINT frame = 0; frame < FrameCount; frame++)
{
srvDesc.Buffer.FirstElement = frame * TriangleCount;
m_device->CreateShaderResourceView(m_constantBuffer.Get(), &srvDesc, cbvSrvHandle);
cbvSrvHandle.Offset(CbvSrvUavDescriptorCountPerFrame, m_cbvSrvUavDescriptorSize);
}
}
// Create the command signature used for indirect drawing.
{
// Each command consists of a CBV update and a DrawInstanced call.
D3D12_INDIRECT_ARGUMENT_DESC argumentDescs[2] = {};
argumentDescs[0].Type = D3D12_INDIRECT_ARGUMENT_TYPE_CONSTANT_BUFFER_VIEW;
argumentDescs[0].ConstantBufferView.RootParameterIndex = Cbv;
argumentDescs[1].Type = D3D12_INDIRECT_ARGUMENT_TYPE_DRAW;
D3D12_COMMAND_SIGNATURE_DESC commandSignatureDesc = {};
commandSignatureDesc.pArgumentDescs = argumentDescs;
commandSignatureDesc.NumArgumentDescs = _countof(argumentDescs);
commandSignatureDesc.ByteStride = sizeof(IndirectCommand);
ThrowIfFailed(m_device->CreateCommandSignature(&commandSignatureDesc, m_rootSignature.Get(), IID_PPV_ARGS(&m_commandSignature)));
}
// Create the command buffers and UAVs to store the results of the compute work.
{
std::vector<IndirectCommand> commands;
commands.resize(TriangleResourceCount);
const UINT commandBufferSize = CommandBufferSizePerFrame * FrameCount;
D3D12_RESOURCE_DESC commandBufferDesc = CD3DX12_RESOURCE_DESC::Buffer(commandBufferSize);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&commandBufferDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_commandBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(commandBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&commandBufferUpload)));
D3D12_GPU_VIRTUAL_ADDRESS gpuAddress = m_constantBuffer->GetGPUVirtualAddress();
UINT commandIndex = 0;
for (UINT frame = 0; frame < FrameCount; frame++)
{
for (UINT n = 0; n < TriangleCount; n++)
{
commands[commandIndex].cbv = gpuAddress;
commands[commandIndex].drawArguments.VertexCountPerInstance = 3;
commands[commandIndex].drawArguments.InstanceCount = 1;
commands[commandIndex].drawArguments.StartVertexLocation = 0;
commands[commandIndex].drawArguments.StartInstanceLocation = 0;
commandIndex++;
gpuAddress += sizeof(ConstantBufferData);
}
}
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the command buffer.
D3D12_SUBRESOURCE_DATA commandData = {};
commandData.pData = reinterpret_cast<UINT8*>(&commands[0]);
commandData.RowPitch = commandBufferSize;
commandData.SlicePitch = commandData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_commandBuffer.Get(), commandBufferUpload.Get(), 0, 0, 1, &commandData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_commandBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_NON_PIXEL_SHADER_RESOURCE));
// Create SRVs for the command buffers.
D3D12_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
srvDesc.Format = DXGI_FORMAT_UNKNOWN;
srvDesc.ViewDimension = D3D12_SRV_DIMENSION_BUFFER;
srvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
srvDesc.Buffer.NumElements = TriangleCount;
srvDesc.Buffer.StructureByteStride = sizeof(IndirectCommand);
srvDesc.Buffer.Flags = D3D12_BUFFER_SRV_FLAG_NONE;
CD3DX12_CPU_DESCRIPTOR_HANDLE commandsHandle(m_cbvSrvUavHeap->GetCPUDescriptorHandleForHeapStart(), CommandsOffset, m_cbvSrvUavDescriptorSize);
for (UINT frame = 0; frame < FrameCount; frame++)
{
srvDesc.Buffer.FirstElement = frame * TriangleCount;
m_device->CreateShaderResourceView(m_commandBuffer.Get(), &srvDesc, commandsHandle);
commandsHandle.Offset(CbvSrvUavDescriptorCountPerFrame, m_cbvSrvUavDescriptorSize);
}
// Create the unordered access views (UAVs) that store the results of the compute work.
CD3DX12_CPU_DESCRIPTOR_HANDLE processedCommandsHandle(m_cbvSrvUavHeap->GetCPUDescriptorHandleForHeapStart(), ProcessedCommandsOffset, m_cbvSrvUavDescriptorSize);
for (UINT frame = 0; frame < FrameCount; frame++)
{
// Allocate a buffer large enough to hold all of the indirect commands
// for a single frame as well as a UAV counter.
commandBufferDesc = CD3DX12_RESOURCE_DESC::Buffer(CommandBufferSizePerFrame + sizeof(UINT), D3D12_RESOURCE_FLAG_ALLOW_UNORDERED_ACCESS);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&commandBufferDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_processedCommandBuffers[frame])));
D3D12_UNORDERED_ACCESS_VIEW_DESC uavDesc = {};
uavDesc.Format = DXGI_FORMAT_UNKNOWN;
uavDesc.ViewDimension = D3D12_UAV_DIMENSION_BUFFER;
uavDesc.Buffer.FirstElement = 0;
uavDesc.Buffer.NumElements = TriangleCount;
uavDesc.Buffer.StructureByteStride = sizeof(IndirectCommand);
uavDesc.Buffer.CounterOffsetInBytes = CommandBufferSizePerFrame;
uavDesc.Buffer.Flags = D3D12_BUFFER_UAV_FLAG_NONE;
m_device->CreateUnorderedAccessView(
m_processedCommandBuffers[frame].Get(),
m_processedCommandBuffers[frame].Get(),
&uavDesc,
processedCommandsHandle);
processedCommandsHandle.Offset(CbvSrvUavDescriptorCountPerFrame, m_cbvSrvUavDescriptorSize);
}
// Allocate a buffer that can be used to reset the UAV counters and initialize
// it to 0.
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(sizeof(UINT)),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_processedCommandBufferCounterReset)));
UINT8* pMappedCounterReset = nullptr;
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_processedCommandBufferCounterReset->Map(0, &readRange, reinterpret_cast<void**>(&pMappedCounterReset)));
ZeroMemory(pMappedCounterReset, sizeof(UINT));
m_processedCommandBufferCounterReset->Unmap(0, nullptr);
}
// Close the command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_computeFence)));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu();
}
}
void SampleRenderer::Initialize()
{
m_RenderQueue = new ::Dispatch::WorkQueue("RenderQueue", ::Concurrency::ThreadPriority::High);
m_RenderQueue->Loop();
isInitialized = false;
float aspectRatio = 1900.f/700;
float fovAngleY = 70.0f * XM_PI / 180.0f;
if (aspectRatio < 1.0f)
{
fovAngleY *= 2.0f;
}
XMMATRIX perspectiveMatrix = XMMatrixPerspectiveFovRH(fovAngleY, aspectRatio, 0.01f, 100.0f);
XMStoreFloat4x4(&m_MVP.projection, XMMatrixTranspose(perspectiveMatrix));
static const XMVECTORF32 eye = { 0.0f, 0.7f, 1.5f, 0.0f };
static const XMVECTORF32 at = { 0.0f, -0.1f, 0.0f, 0.0f };
static const XMVECTORF32 up = { 0.0f, 1.0f, 0.0f, 0.0f };
XMStoreFloat4x4(&m_MVP.model, XMMatrixTranspose(XMMatrixRotationY(1.57)));
XMStoreFloat4x4(&m_MVP.view, XMMatrixTranspose(XMMatrixLookAtRH(eye, at, up)));
auto d3dDevice = gD3DDevice->GetD3DDevice();
CD3DX12_ROOT_PARAMETER parameter[3];
CD3DX12_DESCRIPTOR_RANGE range;
range.Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
parameter[0].InitAsDescriptorTable(1, &range, D3D12_SHADER_VISIBILITY_VERTEX);
CD3DX12_DESCRIPTOR_RANGE range2;
range2.Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
parameter[1].InitAsDescriptorTable(1, &range2, D3D12_SHADER_VISIBILITY_PIXEL);
CD3DX12_DESCRIPTOR_RANGE range3;
range3.Init(D3D12_DESCRIPTOR_RANGE_TYPE_SAMPLER, 1, 0);
parameter[2].InitAsDescriptorTable(1, &range3, D3D12_SHADER_VISIBILITY_PIXEL);
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_PIXEL_SHADER_ROOT_ACCESS;
CD3DX12_ROOT_SIGNATURE_DESC descRootSignature;
descRootSignature.Init(3, parameter, 0, nullptr, rootSignatureFlags);
PtrBlob pSignature, pError;
ThrowIfFailed(D3D12SerializeRootSignature(&descRootSignature, D3D_ROOT_SIGNATURE_VERSION_1, pSignature.GetInitReference(), pError.GetInitReference()));
ThrowIfFailed(d3dDevice->CreateRootSignature(0, pSignature->GetBufferPointer(), pSignature->GetBufferSize(), IID_PPV_ARGS(m_RootSignature.GetInitReference())));
m_CopyQueue = new CopyQueue_tr(d3dDevice);
m_CopyQueue->StartLoop();
m_RenderQueue->Queue(::Dispatch::Bind([this, d3dDevice]() {
m_VS.Load("/Data/Test/Test.vso");
m_PS.Load("/Data/Test/Test.pso");
static const D3D12_INPUT_ELEMENT_DESC inputLayout[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
D3D12_GRAPHICS_PIPELINE_STATE_DESC state = {};
state.InputLayout = { inputLayout, _countof(inputLayout) };
state.pRootSignature = m_RootSignature;
state.VS = { m_VS.GetBlob()->GetBufferPointer(), m_VS.GetBlob()->GetBufferSize() };
state.PS = { m_PS.GetBlob()->GetBufferPointer(), m_PS.GetBlob()->GetBufferSize() };
state.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
state.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
state.DepthStencilState.DepthEnable = FALSE;
state.DepthStencilState.StencilEnable = FALSE;
state.SampleMask = UINT_MAX;
state.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
state.NumRenderTargets = 1;
state.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
state.SampleDesc.Count = 1;
ThrowIfFailed(d3dDevice->CreateGraphicsPipelineState(&state, IID_PPV_ARGS(m_PipeLineState.GetInitReference())));
LogUtil::Out("Renderer", ::Concurrency::Thread::GetCurrentThreadName());
}));
m_RenderQueue->Queue(::Dispatch::Bind([this, d3dDevice]() {
ThrowIfFailed(d3dDevice->CreateCommandList(1, D3D12_COMMAND_LIST_TYPE_DIRECT, gD3DDevice->GetCommandAllocator(), m_PipeLineState, IID_PPV_ARGS(m_CmdList.GetInitReference())));
D3D12_DESCRIPTOR_HEAP_DESC heapDesc = {};
heapDesc.NumDescriptors = DeviceManager::GetFrameCount();
heapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV;
heapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_SHADER_VISIBLE;
ThrowIfFailed(d3dDevice->CreateDescriptorHeap(&heapDesc, IID_PPV_ARGS(m_CBVHeap.GetInitReference())));
m_CBVHeap->SetName(L"Constant Buffer Heap");
m_samplerHeap.Create(d3dDevice, D3D12_DESCRIPTOR_HEAP_TYPE::D3D12_DESCRIPTOR_HEAP_TYPE_SAMPLER, 1, true);
D3D12_SAMPLER_DESC samplerDesc = {};
samplerDesc.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
samplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.MipLODBias = 0.0f;
samplerDesc.MaxAnisotropy = 1;
samplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
samplerDesc.MinLOD = 0;
samplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
d3dDevice->CreateSampler(&samplerDesc, m_samplerHeap.hCPU(0));
m_CubeMesh = new CubeMesh(d3dDevice, m_CmdList);
this->m_CopyQueue->SubmitTexture(L"\\Data\\Test\\seafloor2bc1.dds", m_CubeMesh);
m_CBO = new UniformBuffer<ModelViewProjectionConstantBuffer>("ModeViewMatrix", d3dDevice, 1U);
m_CBO->CreateOnHeap(m_CBVHeap, d3dDevice);
m_ConstantBuffer = m_CBO->Map();
ZeroMemory(m_ConstantBuffer, DeviceManager::GetFrameCount() * m_CBO->sAlignedConstantBufferSize);
ThrowIfFailed(m_CmdList->Close());
ID3D12CommandList* ppCommandLists[] = { m_CmdList };
gD3DDevice->GetCommandQueue()->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
gD3DDevice->WaitForGPU();
isInitialized = true;
}));
}
void AppTest::LoadAssets()
{
//Load and compile the shaders
D3D12Shader vertShader = D3D12Shader(ShaderType::Vertex, L"TestShader.hlsl", "VS");
D3D12Shader pixelShader = D3D12Shader(ShaderType::Pixel, L"TestShader.hlsl", "PS");
std::vector<ShaderMacro> macros;
macros.push_back({ "BOXCOLOR", "float4(1.0f, 1.0f, 1.0f, 1.0f)" });
vertShader.Compile(¯os[0], static_cast<uint32_t>(macros.size()));
pixelShader.Compile(¯os[0], static_cast<uint32_t>(macros.size()));
D3D12_INPUT_ELEMENT_DESC layout[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "NORMAL", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA , 0}
};
UINT numElements = ARRAYSIZE(layout);
//Create root signiture with one root constant buffer view
D3D12_ROOT_PARAMETER param;
D3D12_DESCRIPTOR_RANGE descriptorRange = {};
if (UseRootLevelCBV)
{
param.ParameterType = D3D12_ROOT_PARAMETER_TYPE_CBV;
param.ShaderVisibility = D3D12_SHADER_VISIBILITY_VERTEX;
param.Descriptor = { 0, 0 };
}
else
{
//Used for if you need non root level constant buffer views
//This is somewhat slower than switching out a root level CBV
descriptorRange.BaseShaderRegister = 0;
descriptorRange.NumDescriptors = 1;
descriptorRange.OffsetInDescriptorsFromTableStart = 0;
descriptorRange.RangeType = D3D12_DESCRIPTOR_RANGE_TYPE_CBV;
descriptorRange.RegisterSpace = 0;
param.ParameterType = D3D12_ROOT_PARAMETER_TYPE_DESCRIPTOR_TABLE;
param.ShaderVisibility = D3D12_SHADER_VISIBILITY_VERTEX;
param.DescriptorTable.NumDescriptorRanges = 1;
param.DescriptorTable.pDescriptorRanges = &descriptorRange;
}
ComPtr<ID3D10Blob> pOutBlob, pErrorBlob;
D3D12_ROOT_SIGNATURE_DESC rootSigDesc;
rootSigDesc.NumParameters = 1;
rootSigDesc.pParameters = ¶m;
rootSigDesc.NumStaticSamplers = 0;
rootSigDesc.pStaticSamplers = nullptr;
rootSigDesc.Flags = D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT;
CHK(D3D12SerializeRootSignature(&rootSigDesc, D3D_ROOT_SIGNATURE_VERSION_1, pOutBlob.GetAddressOf(), pErrorBlob.GetAddressOf()));
CHK(Device->CreateRootSignature(0, pOutBlob->GetBufferPointer(), pOutBlob->GetBufferSize(), IID_PPV_ARGS(RootSignature.GetAddressOf())));
//Create the PSO
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc;
ZeroMemory(&psoDesc, sizeof(psoDesc));
psoDesc.InputLayout = { layout, numElements };
psoDesc.pRootSignature = RootSignature.Get();
psoDesc.VS = { reinterpret_cast<BYTE*>(vertShader.GetByteCodePtr()), vertShader.GetSize() };
psoDesc.PS = { reinterpret_cast<BYTE*>(pixelShader.GetByteCodePtr()), pixelShader.GetSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);;
psoDesc.RasterizerState.FrontCounterClockwise = true; //Using RH coordinate system
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = true;
psoDesc.DepthStencilState.DepthFunc = D3D12_COMPARISON_FUNC_LESS_EQUAL;
psoDesc.DepthStencilState.DepthWriteMask = D3D12_DEPTH_WRITE_MASK_ALL;
psoDesc.DepthStencilState.StencilEnable = false;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
psoDesc.SampleDesc.Quality = 0;
CHK(Device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(PSO.GetAddressOf())));
//Create descriptor heap
D3D12_DESCRIPTOR_HEAP_DESC descHeapDesc = {};
descHeapDesc.NumDescriptors = 1;
descHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_RTV;
descHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
CHK(Device->CreateDescriptorHeap(&descHeapDesc, IID_PPV_ARGS(DescriptorHeap.GetAddressOf())));
D3D12_DESCRIPTOR_HEAP_DESC DSVDescHeapDesc = {};
DSVDescHeapDesc.NumDescriptors = 1;
DSVDescHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_DSV;
DSVDescHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
CHK(Device->CreateDescriptorHeap(&DSVDescHeapDesc, IID_PPV_ARGS(DSVDescriptorHeap.GetAddressOf())));
CHK(Device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, CommandAllocator.Get(), PSO.Get(), IID_PPV_ARGS(CommandList.GetAddressOf())));
//Create the command lists for each thread
for (unsigned int i = 0; i < ThreadCount; i++)
{
CHK(Device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, CommandAllocatorArray[i].Get(), PSO.Get(), IID_PPV_ARGS(CommandListArray[i].GetAddressOf())));
CommandListArray[i].Get()->Close();
}
//Create back buffer and render target
CHK(SwapChain->GetBuffer(0, IID_PPV_ARGS(RenderTarget.GetAddressOf())));
Device->CreateRenderTargetView(RenderTarget.Get(), nullptr, DescriptorHeap->GetCPUDescriptorHandleForHeapStart());
Viewport.TopLeftX = 0.0f;
Viewport.TopLeftY = 0.0f;
Viewport.Width = static_cast<float>(ViewportWidth);
Viewport.Height = static_cast<float>(ViewportHeight);
Viewport.MinDepth = 0.0f;
Viewport.MaxDepth = 1.0f;
RectScissor.left = 0;
RectScissor.top = 0;
RectScissor.right = ViewportWidth;
RectScissor.bottom = ViewportHeight;
//Create depth buffer
D3D12_RESOURCE_DESC depthResDesc = CD3DX12_RESOURCE_DESC::Tex2D(
DXGI_FORMAT_R32_TYPELESS,
ViewportWidth,
ViewportHeight,
1, 1, 1, 0, D3D12_RESOURCE_FLAG_ALLOW_DEPTH_STENCIL, D3D12_TEXTURE_LAYOUT_UNKNOWN);
D3D12_CLEAR_VALUE dsvClearValue;
dsvClearValue.Format = DXGI_FORMAT_D32_FLOAT;
dsvClearValue.DepthStencil.Depth = 1.0f;
dsvClearValue.DepthStencil.Stencil = 0;
CHK(Device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&depthResDesc,
D3D12_RESOURCE_STATE_DEPTH_WRITE,
&dsvClearValue,
IID_PPV_ARGS(DepthBufferTexture.GetAddressOf())));
DepthBufferTexture->SetName(L"Depth Buffer");
D3D12_DEPTH_STENCIL_VIEW_DESC dsvDesc = {};
dsvDesc.ViewDimension = D3D12_DSV_DIMENSION_TEXTURE2D;
dsvDesc.Format = DXGI_FORMAT_D32_FLOAT;
dsvDesc.Texture2D.MipSlice = 0;
dsvDesc.Flags = D3D12_DSV_FLAG_NONE;
Device->CreateDepthStencilView(DepthBufferTexture.Get(), &dsvDesc, DSVDescriptorHeap->GetCPUDescriptorHandleForHeapStart());
//Allocate buffer for all constant buffers
PerObjectConstantBuffers.Init(BoxCount, sizeof(CBPerObject), Device.Get());
//Create the constant buffer descriptor heap and populate it
if (!UseRootLevelCBV)
{
ConstantBufferDescriptorHeap = std::make_unique<D3D12RHIDescriptorHeap>(BoxCount, D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV, true);
HRESULT hr = ConstantBufferDescriptorHeap->Init(Device.Get());
assert(SUCCEEDED(hr));
D3D12_CONSTANT_BUFFER_VIEW_DESC constantBufferViewDesc = {};
constantBufferViewDesc.SizeInBytes = PerObjectConstantBuffers.GetAlignedStride();
D3D12_CPU_DESCRIPTOR_HANDLE cpuDescriptorHandle;
for (unsigned int i = 0; i < BoxCount; i++)
{
cpuDescriptorHandle.ptr = reinterpret_cast<size_t>(ConstantBufferDescriptorHeap->GetDescriptorCPUPtr(i));
constantBufferViewDesc.BufferLocation = PerObjectConstantBuffers.GetGPUHandle(i);
Device->CreateConstantBufferView(&constantBufferViewDesc, cpuDescriptorHandle);
}
}
//Generate mesh
GeometryGenerator::SimpleMesh shapeMesh;
GeometryGenerator::CreateBox(1.0f, 1.0f, 1.0f, shapeMesh);
//GeometryGenerator::CreateSphere(1.0f, 10, 10, shapeMesh);
IndexCount = static_cast<uint32_t>(shapeMesh.Indices.size());
//Copy vertex data to buffer
//This should be put into its own job system using the copy engine eventually
Device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(sizeof(GeometryGenerator::SimpleVertex) * shapeMesh.Vertices.size()),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(VertBuffer.GetAddressOf()));
Device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(sizeof(uint32_t) * shapeMesh.Indices.size()),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(IndexBuffer.GetAddressOf()));
UINT8* dataBegin;
VertBuffer->Map(0, nullptr, reinterpret_cast<void**>(&dataBegin));
memcpy(dataBegin, &shapeMesh.Vertices[0], sizeof(GeometryGenerator::SimpleVertex) * shapeMesh.Vertices.size());
VertBuffer->Unmap(0, nullptr);
IndexBuffer->Map(0, nullptr, reinterpret_cast<void**>(&dataBegin));
memcpy(dataBegin, &shapeMesh.Indices[0], sizeof(uint32_t) * shapeMesh.Indices.size());
IndexBuffer->Unmap(0, nullptr);
//Create vertex buffer view
DescViewBufVert.BufferLocation = VertBuffer->GetGPUVirtualAddress();
DescViewBufVert.StrideInBytes = sizeof(GeometryGenerator::SimpleVertex);
DescViewBufVert.SizeInBytes = static_cast<UINT>(sizeof(GeometryGenerator::SimpleVertex) * shapeMesh.Vertices.size());
DescViewBufIndex.BufferLocation = IndexBuffer->GetGPUVirtualAddress();
DescViewBufIndex.Format = DXGI_FORMAT_R32_UINT;
DescViewBufIndex.SizeInBytes = static_cast<UINT>(sizeof(uint32_t) * shapeMesh.Indices.size());
//Create fencing object
Device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(Fence.GetAddressOf()));
CurrentFence = 1;
//Initialize bundles for drawing
if (UseBundles)
{
InitBundles();
}
//Close the command list and use it to execute the GPU setup
CommandList->Close();
ID3D12CommandList* ppCommandLists[] = { CommandList.Get() };
CommandQueue->ExecuteCommandLists(1, ppCommandLists);
HandleEvent = CreateEventEx(nullptr, false, false, EVENT_ALL_ACCESS);
WaitForGPU();
}
bool D3DClass::Initialize(int screenHeight, int screenWidth, HWND hwnd, bool vsync, bool fullscreen)
{
D3D_FEATURE_LEVEL featureLevel;
HRESULT result;
D3D12_COMMAND_QUEUE_DESC commandQueueDesc;
IDXGIFactory4* factory;
IDXGIAdapter* adapter;
IDXGIOutput* adapterOutput;
unsigned int numModes, i, numerator, denominator, renderTargetViewDescriptorSize;
unsigned long long stringLength;
DXGI_MODE_DESC* displayModeList;
DXGI_ADAPTER_DESC adapterDesc;
DXGI_SWAP_CHAIN_DESC swapChainDesc;
IDXGISwapChain* swapChain;
D3D12_DESCRIPTOR_HEAP_DESC renderTargetViewHeapDesc;
D3D12_CPU_DESCRIPTOR_HANDLE renderTargetViewHandle;
// Store the vsync setting.
m_vsync_enabled = vsync;
// Set the feature level to DirectX 12.1 to enable using all the DirectX 12 features.
// Note: Not all cards support full DirectX 12, this feature level may need to be reduced on some cards to 12.0.
featureLevel = D3D_FEATURE_LEVEL_11_0;
// Create the Direct3D 12 device.
result = D3D12CreateDevice(NULL, featureLevel, __uuidof(ID3D12Device), (void**)&m_device);
if (FAILED(result))
{
MessageBox(hwnd, L"Could not create a DirectX 12.1 device. The default video card does not support DirectX 12.1.", L"DirectX Device Failure", MB_OK);
return false;
}
// Initialize the description of the command queue.
ZeroMemory(&commandQueueDesc, sizeof(commandQueueDesc));
// Set up the description of the command queue.
commandQueueDesc.Type = D3D12_COMMAND_LIST_TYPE_DIRECT;
commandQueueDesc.Priority = D3D12_COMMAND_QUEUE_PRIORITY_NORMAL;
commandQueueDesc.Flags = D3D12_COMMAND_QUEUE_FLAG_NONE;
commandQueueDesc.NodeMask = 0;
// Create the command queue.
m_device->CreateCommandQueue(&commandQueueDesc, __uuidof(ID3D12CommandQueue), (void**)&m_commandQueue);
// Create a DirectX graphics interface factory.
CreateDXGIFactory1(__uuidof(IDXGIFactory4), (void**)&factory);
// Use the factory to create an adapter for the primary graphics interface (video card).
factory->EnumAdapters(0, &adapter);
// Enumerate the primary adapter output (monitor).
adapter->EnumOutputs(0, &adapterOutput);
// Get the number of modes that fit the DXGI_FORMAT_R8G8B8A8_UNORM display format for the adapter output (monitor).
adapterOutput->GetDisplayModeList(DXGI_FORMAT_R8G8B8A8_UNORM, DXGI_ENUM_MODES_INTERLACED, &numModes, NULL);
// Create a list to hold all the possible display modes for this monitor/video card combination.
displayModeList = new DXGI_MODE_DESC[numModes];
// Now fill the display mode list structures.
adapterOutput->GetDisplayModeList(DXGI_FORMAT_R8G8B8A8_UNORM, DXGI_ENUM_MODES_INTERLACED, &numModes, displayModeList);
// Now go through all the display modes and find the one that matches the screen height and width.
// When a match is found store the numerator and denominator of the refresh rate for that monitor.
for (i = 0; i<numModes; i++)
{
if (displayModeList[i].Height == (unsigned int)screenHeight)
{
if (displayModeList[i].Width == (unsigned int)screenWidth)
{
numerator = displayModeList[i].RefreshRate.Numerator;
denominator = displayModeList[i].RefreshRate.Denominator;
}
}
}
// Get the adapter (video card) description.
adapter->GetDesc(&adapterDesc);
// Store the dedicated video card memory in megabytes.
m_videoCardMemory = (int)(adapterDesc.DedicatedVideoMemory / 1024 / 1024);
// Convert the name of the video card to a character array and store it.
wcstombs_s(&stringLength, m_videoCardDescription, 128, adapterDesc.Description, 128);
// Release the display mode list.
delete[] displayModeList;
displayModeList = nullptr;
// Release the adapter output.
adapterOutput->Release();
adapterOutput = nullptr;
// Release the adapter.
adapter->Release();
adapter = nullptr;
// Initialize the swap chain description.
ZeroMemory(&swapChainDesc, sizeof(swapChainDesc));
// Set the swap chain to use double buffering.
swapChainDesc.BufferCount = 2;
// Set the height and width of the back buffers in the swap chain.
swapChainDesc.BufferDesc.Height = screenHeight;
swapChainDesc.BufferDesc.Width = screenWidth;
// Set a regular 32-bit surface for the back buffers.
swapChainDesc.BufferDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
// Set the usage of the back buffers to be render target outputs.
swapChainDesc.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT;
// Set the swap effect to discard the previous buffer contents after swapping.
swapChainDesc.SwapEffect = DXGI_SWAP_EFFECT_FLIP_DISCARD;
// Set the handle for the window to render to.
swapChainDesc.OutputWindow = hwnd;
// Set to full screen or windowed mode.
swapChainDesc.Windowed = !fullscreen;
// Set the refresh rate of the back buffer.
if (m_vsync_enabled)
{
swapChainDesc.BufferDesc.RefreshRate.Numerator = numerator;
swapChainDesc.BufferDesc.RefreshRate.Denominator = denominator;
}
else
{
swapChainDesc.BufferDesc.RefreshRate.Numerator = 0;
swapChainDesc.BufferDesc.RefreshRate.Denominator = 1;
}
// Turn multisampling off.
swapChainDesc.SampleDesc.Count = 1;
swapChainDesc.SampleDesc.Quality = 0;
// Set the scan line ordering and scaling to unspecified.
swapChainDesc.BufferDesc.ScanlineOrdering = DXGI_MODE_SCANLINE_ORDER_UNSPECIFIED;
swapChainDesc.BufferDesc.Scaling = DXGI_MODE_SCALING_UNSPECIFIED;
// Don't set the advanced flags.
swapChainDesc.Flags = 0;
// Finally create the swap chain using the swap chain description.
factory->CreateSwapChain(m_commandQueue, &swapChainDesc, &swapChain);
// Next upgrade the IDXGISwapChain to a IDXGISwapChain3 interface and store it in a private member variable named m_swapChain.
// This will allow us to use the newer functionality such as getting the current back buffer index.
swapChain->QueryInterface(__uuidof(IDXGISwapChain3), (void**)&m_swapChain);
// Clear pointer to original swap chain interface since we are using version 3 instead (m_swapChain).
swapChain = nullptr;
// Release the factory now that the swap chain has been created.
factory->Release();
factory = nullptr;
// Initialize the render target view heap description for the two back buffers.
ZeroMemory(&renderTargetViewHeapDesc, sizeof(renderTargetViewHeapDesc));
// Set the number of descriptors to two for our two back buffers. Also set the heap tyupe to render target views.
renderTargetViewHeapDesc.NumDescriptors = 2;
renderTargetViewHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_RTV;
renderTargetViewHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
// Create the render target view heap for the back buffers.
m_device->CreateDescriptorHeap(&renderTargetViewHeapDesc, __uuidof(ID3D12DescriptorHeap), (void**)&m_renderTargetViewHeap);
// Get a handle to the starting memory location in the render target view heap to identify where the render target views will be located for the two back buffers.
renderTargetViewHandle = m_renderTargetViewHeap->GetCPUDescriptorHandleForHeapStart();
// Get the size of the memory location for the render target view descriptors.
renderTargetViewDescriptorSize = m_device->GetDescriptorHandleIncrementSize(D3D12_DESCRIPTOR_HEAP_TYPE_RTV);
// Get a pointer to the first back buffer from the swap chain.
m_swapChain->GetBuffer(0, __uuidof(ID3D12Resource), (void**)&m_backBufferRenderTarget[0]);
// Create a render target view for the first back buffer.
m_device->CreateRenderTargetView(m_backBufferRenderTarget[0], NULL, renderTargetViewHandle);
// Increment the view handle to the next descriptor location in the render target view heap.
renderTargetViewHandle.ptr += renderTargetViewDescriptorSize;
// Get a pointer to the second back buffer from the swap chain.
m_swapChain->GetBuffer(1, __uuidof(ID3D12Resource), (void**)&m_backBufferRenderTarget[1]);
// Create a render target view for the second back buffer.
m_device->CreateRenderTargetView(m_backBufferRenderTarget[1], NULL, renderTargetViewHandle);
// Finally get the initial index to which buffer is the current back buffer.
m_bufferIndex = m_swapChain->GetCurrentBackBufferIndex();
// Create a command allocator.
m_device->CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT, __uuidof(ID3D12CommandAllocator), (void**)&m_commandAllocator);
// Create a fence for GPU synchronization.
m_device->CreateFence(0, D3D12_FENCE_FLAG_NONE, __uuidof(ID3D12Fence), (void**)&m_fence);
// Create an event object for the fence.
m_fenceEvent = CreateEvent(nullptr, false, false, nullptr);
// Initialize the starting fence value.
m_fenceValue = 1;
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ID3DBlob* signature;
ID3DBlob* error;
D3D12SerializeRootSignature(
&rootSignatureDesc,
D3D_ROOT_SIGNATURE_VERSION_1,
&signature,
&error);
m_device->CreateRootSignature(
0,
signature->GetBufferPointer(),
signature->GetBufferSize(),
IID_PPV_ARGS(&m_rootSignature));
//Create pipeline state, load and compiler shaders.
//Note here to change D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION for debug
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
D3DReadFileToBlob(L"DefaultVS.cso", &m_vertexShader);
D3DReadFileToBlob(L"DefaultHS.cso", &m_hullShader);
D3DReadFileToBlob(L"DefaultDS.cso", &m_domainShader);
D3DReadFileToBlob(L"DefaultPS.cso", &m_pixelShader);
std::array<D3D12_INPUT_ELEMENT_DESC, 2> inputElementDescs =
{
D3D12_INPUT_ELEMENT_DESC{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
D3D12_INPUT_ELEMENT_DESC{ "COLOR", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
D3D12_GRAPHICS_PIPELINE_STATE_DESC pipelineDesc = {};
pipelineDesc.InputLayout = { inputElementDescs.data() , (UINT)inputElementDescs.size() };
pipelineDesc.pRootSignature = m_rootSignature;
pipelineDesc.VS = { m_vertexShader->GetBufferPointer(), m_vertexShader->GetBufferSize() };
pipelineDesc.PS = { m_pixelShader->GetBufferPointer(), m_pixelShader->GetBufferSize() };
pipelineDesc.HS = { m_hullShader->GetBufferPointer(), m_hullShader->GetBufferSize() };
pipelineDesc.DS = { m_domainShader->GetBufferPointer(), m_domainShader->GetBufferSize() };
pipelineDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT_WIREFRAME);
pipelineDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
pipelineDesc.DepthStencilState.DepthEnable = false;
pipelineDesc.DepthStencilState.StencilEnable = false;
pipelineDesc.SampleMask = UINT_MAX;
pipelineDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_PATCH;
pipelineDesc.NumRenderTargets = 1;
pipelineDesc.RTVFormats[0] = DXGI_FORMAT_B8G8R8A8_UNORM;
pipelineDesc.SampleDesc.Count = 1;
m_device->CreateGraphicsPipelineState(&pipelineDesc, IID_PPV_ARGS(&m_pipelineState));
// Create a basic command list.
m_device->CreateCommandList(
0,
D3D12_COMMAND_LIST_TYPE_DIRECT,
m_commandAllocator,
m_pipelineState,
IID_PPV_ARGS(&m_commandList));
// Initially we need to close the command list during initialization as it is created in a recording state.
m_commandList->Close();
Vertex triangleVertices[] =
{
{ { 0.0f, 0.5f, 0.f }, { 1.f, 0.f, 0.f, 1.f } },
{ { 0.5f, -0.5f, 0.f },{ 0.f, 1.f, 0.f, 1.f } },
{ { -0.5f, -0.5f, 0.f },{ 0.f, 0.f, 1.f, 1.f } },
};
const UINT vertexBufferSize = sizeof(triangleVertices);
m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer));
UINT8* pVertexDataBegin;
CD3DX12_RANGE readRange(0, 0);
m_vertexBuffer->Map(0, &readRange, reinterpret_cast<void**>(&pVertexDataBegin));
memcpy(pVertexDataBegin, triangleVertices, vertexBufferSize);
m_vertexBuffer->Unmap(0, nullptr);
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = vertexBufferSize;
m_viewport.Height = screenHeight;
m_viewport.Width = screenWidth;
m_viewport.MaxDepth = 1000.f;
m_viewport.MinDepth = 0.1f;
m_viewport.TopLeftX = 0.f;
m_viewport.TopLeftY = 0.f;
m_scissorRect.left = 0;
m_scissorRect.right = screenWidth;
m_scissorRect.top = 0;
m_scissorRect.bottom = screenHeight;
unsigned long long fenceToWaitFor = m_fenceValue;
m_commandQueue->Signal(m_fence, fenceToWaitFor);
m_fenceValue++;
// Wait until the GPU is done rendering.
if (m_fence->GetCompletedValue() < fenceToWaitFor)
{
m_fence->SetEventOnCompletion(fenceToWaitFor, m_fenceEvent);
WaitForSingleObject(m_fenceEvent, INFINITE);
}
return true;
}
void DirectX12Render::DirectX12Render::InitializeAssets()
{
//Create the root sig
CD3DX12_ROOT_SIGNATURE_DESC RootDesc;
RootDesc.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> Signature;
ComPtr<ID3DBlob> Error;
ThrowFail(D3D12SerializeRootSignature(&RootDesc, D3D_ROOT_SIGNATURE_VERSION_1, &Signature, &Error), TEXT("Root Signature Failed to Create"));
ThrowFail(Device->CreateRootSignature(0, Signature->GetBufferPointer(), Signature->GetBufferSize(), IID_PPV_ARGS(&RootSignature)), TEXT("Root Signature Failed to Create"));
ComPtr<ID3DBlob> VertexShader;
ComPtr<ID3DBlob> PixelShader;
//Shader Debugging code
#if defined(_DEBUG)
UINT CompileFlags = 1 << 0 | 1 << 2;
#else
UINT CompileFlags = 0;
#endif
ThrowFail(D3DCompileFromFile(L"shaders.hlsl", nullptr, nullptr, "VSMain", "vs_5_0", CompileFlags, 0, &VertexShader, nullptr), TEXT("Failed to Compile Vertex Shader"));
ThrowFail(D3DCompileFromFile(L"shaders.hlsl", nullptr, nullptr, "PSMain", "ps_5_0", CompileFlags, 0, &PixelShader, nullptr), TEXT("Failed to Compile Pixel Shader"));
D3D12_INPUT_ELEMENT_DESC InputElementDesc[] =
{
{"POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0},
{"COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
//Create the Pipeline State Object
D3D12_GRAPHICS_PIPELINE_STATE_DESC PSO = {};
PSO.InputLayout = { InputElementDesc, _countof(InputElementDesc) };
PSO.pRootSignature = RootSignature.Get();
PSO.VS = { reinterpret_cast<UINT8*>(VertexShader->GetBufferPointer()), VertexShader->GetBufferSize() };
PSO.PS = { reinterpret_cast<UINT8*>(PixelShader->GetBufferPointer()), PixelShader->GetBufferSize() };
PSO.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
PSO.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
PSO.DepthStencilState.DepthEnable = false;
PSO.DepthStencilState.StencilEnable = false;
PSO.SampleMask = UINT_MAX;
PSO.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
PSO.NumRenderTargets = 1;
PSO.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
PSO.SampleDesc.Count = 1;
ThrowFail(Device->CreateGraphicsPipelineState(&PSO, IID_PPV_ARGS(&Pipeline)), TEXT("Failed to create Pipeline"));
//Create Command list
ThrowFail(Device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, CommandAllocator.Get(), Pipeline.Get(), IID_PPV_ARGS(&CommandList)), TEXT("Command List Creation Failed"));
ThrowFail(CommandList->Close(), TEXT("Command List Failed to Close"));
//Create a fence to sync CPU to GPU
ThrowFail(Device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&Fence)), TEXT("Failed to create fence"));
FenceHandle = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (FenceHandle == nullptr)
{
ThrowFail(HRESULT_FROM_WIN32(GetLastError()), TEXT("Failed to get Last Error, I know..."));
}
}
void loadAssets()
{
//
// handles to vert and pixel shaders
//
ComPtr<ID3DBlob> blobShaderVert, blobShaderPixel;
//
// compile shaders
//
D3DCompileFromFile(L"shaders.hlsl", nullptr, nullptr, "VShader", "vs_5_0", 0, 0, blobShaderVert.GetAddressOf(), nullptr);
D3DCompileFromFile(L"shaders.hlsl", nullptr, nullptr, "PShader", "ps_5_0", 0, 0, blobShaderPixel.GetAddressOf(), nullptr);
//
// create input layout
//
D3D12_INPUT_ELEMENT_DESC layout[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_PER_VERTEX_DATA, 0 }
};
UINT numElements = sizeof(layout) / sizeof(layout[0]);
//
// create an empty root signature
//
ComPtr<ID3DBlob> pOutBlob, pErrorBlob;
D3D12_ROOT_SIGNATURE_DESC descRootSignature;
descRootSignature.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
D3D12SerializeRootSignature(&descRootSignature, D3D_ROOT_SIGNATURE_VERSION_1, pOutBlob.GetAddressOf(), pErrorBlob.GetAddressOf());
mDevice->CreateRootSignature(0, pOutBlob->GetBufferPointer(), pOutBlob->GetBufferSize(), IID_PPV_ARGS(mRootSignature.GetAddressOf()));
//
// create a PSO description
//
D3D12_GRAPHICS_PIPELINE_STATE_DESC descPso;
ZeroMemory(&descPso, sizeof(descPso));
descPso.InputLayout = { layout, numElements };
descPso.pRootSignature = mRootSignature.Get();
descPso.VS = { reinterpret_cast<BYTE*>(blobShaderVert->GetBufferPointer()), blobShaderVert->GetBufferSize() };
descPso.PS = { reinterpret_cast<BYTE*>(blobShaderPixel->GetBufferPointer()), blobShaderPixel->GetBufferSize() };
descPso.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
descPso.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
descPso.DepthStencilState.DepthEnable = FALSE;
descPso.DepthStencilState.StencilEnable = FALSE;
descPso.SampleMask = UINT_MAX;
descPso.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
descPso.NumRenderTargets = 1;
descPso.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
descPso.SampleDesc.Count = 1;
//
// create the actual PSO
//
mDevice->CreateGraphicsPipelineState(&descPso, IID_PPV_ARGS(mPSO.GetAddressOf()));
//
// create descriptor heap
//
D3D12_DESCRIPTOR_HEAP_DESC descHeap = {};
descHeap.NumDescriptors = 1;
descHeap.Type = D3D12_DESCRIPTOR_HEAP_TYPE_RTV;
descHeap.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
mDevice->CreateDescriptorHeap(&descHeap, IID_PPV_ARGS(mDescriptorHeap.GetAddressOf()));
//
// create command list
//
mDevice->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, mCommandAllocator.Get(), mPSO.Get(), IID_PPV_ARGS(mCommandList.GetAddressOf()));
//
// create backbuffer/rendertarget
//
mSwapChain->GetBuffer(0, IID_PPV_ARGS(mRenderTarget.GetAddressOf()));
mDevice->CreateRenderTargetView(mRenderTarget.Get(), nullptr, mDescriptorHeap->GetCPUDescriptorHandleForHeapStart());
//
// set the viewport
//
mViewPort =
{
0.0f,
0.0f,
static_cast<float>(mWidth),
static_cast<float>(mHeight),
0.0f,
1.0f
};
//
// create scissor rectangle
//
mRectScissor = { 0, 0, mWidth, mHeight };
//
// create geometry for a triangle
//
VERTEX triangleVerts[] =
{
{ 0.0f, 0.5f, 0.0f,{ 1.0f, 0.0f, 0.0f, 1.0f } },
{ 0.45f, -0.5, 0.0f,{ 0.0f, 1.0f, 0.0f, 1.0f } },
{ -0.45f, -0.5f, 0.0f,{ 0.0f, 0.0f, 1.0f, 1.0f } }
};
//
// actually create the vert buffer
// Note: using upload heaps to transfer static data like vert buffers is not recommended.
// Every time the GPU needs it, the upload heap will be marshalled over. Please read up on Default Heap usage.
// An upload heap is used here for code simplicity and because there are very few verts to actually transfer
//
mDevice->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(3 * sizeof(VERTEX)),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr, // Clear value
IID_PPV_ARGS(mBufVerts.GetAddressOf()));
//
// copy the triangle data to the vertex buffer
//
UINT8* dataBegin;
mBufVerts->Map(0, nullptr, reinterpret_cast<void**>(&dataBegin));
memcpy(dataBegin, triangleVerts, sizeof(triangleVerts));
mBufVerts->Unmap(0, nullptr);
//
// create vertex buffer view
//
mDescViewBufVert.BufferLocation = mBufVerts->GetGPUVirtualAddress();
mDescViewBufVert.StrideInBytes = sizeof(VERTEX);
mDescViewBufVert.SizeInBytes = sizeof(triangleVerts);
//
// create fencing object
//
mDevice->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(mFence.GetAddressOf()));
mCurrentFence = 1;
//
// close the command list and use it to execute the initial GPU setup
//
mCommandList->Close();
ID3D12CommandList* ppCommandLists[] = { mCommandList.Get() };
mCommandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
//
// create event handle
//
mHandleEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
//
// wait for the command list to execute; we are reusing the same command list in our main loop but for now,
// we just want to wait for setup to complete before continuing
//
waitForGPU();
}
// Load the sample assets.
void D3D12Fullscreen::LoadAssets()
{
// Create a root signature consisting of a descriptor table with a single CBV.
{
CD3DX12_DESCRIPTOR_RANGE ranges[1];
CD3DX12_ROOT_PARAMETER rootParameters[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_VERTEX);
// Allow input layout and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_PIXEL_SHADER_ROOT_ACCESS;
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
NAME_D3D12_OBJECT(m_rootSignature);
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
ComPtr<ID3DBlob> error;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, &error));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(vertexShader.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pixelShader.Get());
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
NAME_D3D12_OBJECT(m_pipelineState);
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocators[m_frameIndex].Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
NAME_D3D12_OBJECT(m_commandList);
LoadSizeDependentResources();
// Create/update the vertex buffer.
{
// Define the geometry for a thin quad that will animate across the screen.
const float x = QuadWidth / 2.0f;
const float y = QuadHeight / 2.0f;
Vertex quadVertices[] =
{
{ { -x, -y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } },
{ { -x, y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } },
{ { x, -y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } },
{ { x, y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } }
};
const UINT vertexBufferSize = sizeof(quadVertices);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBufferUpload)));
NAME_D3D12_OBJECT(m_vertexBuffer);
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
UINT8* pVertexDataBegin;
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_vertexBufferUpload->Map(0, &readRange, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, quadVertices, sizeof(quadVertices));
m_vertexBufferUpload->Unmap(0, nullptr);
m_commandList->CopyBufferRegion(m_vertexBuffer.Get(), 0, m_vertexBufferUpload.Get(), 0, vertexBufferSize);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer views.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create the constant buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(sizeof(SceneConstantBuffer) * FrameCount),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_constantBuffer)));
NAME_D3D12_OBJECT(m_constantBuffer);
// Describe and create constant buffer views.
D3D12_CONSTANT_BUFFER_VIEW_DESC cbvDesc = {};
cbvDesc.BufferLocation = m_constantBuffer->GetGPUVirtualAddress();
cbvDesc.SizeInBytes = sizeof(SceneConstantBuffer);
CD3DX12_CPU_DESCRIPTOR_HANDLE cpuHandle(m_cbvHeap->GetCPUDescriptorHandleForHeapStart());
for (UINT n = 0; n < FrameCount; n++)
{
m_device->CreateConstantBufferView(&cbvDesc, cpuHandle);
cbvDesc.BufferLocation += sizeof(SceneConstantBuffer);
cpuHandle.Offset(m_cbvDescriptorSize);
}
// Initialize and map the constant buffers. We don't unmap this until the
// app closes. Keeping things mapped for the lifetime of the resource is okay.
ZeroMemory(&m_constantBufferData, sizeof(m_constantBufferData));
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_constantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pCbvDataBegin)));
memcpy(m_pCbvDataBegin, &m_constantBufferData, sizeof(m_constantBufferData));
}
// Close the command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu();
}
}
void GltfPbr::CreatePipeline(ID3D12Device* pDevice, UINT node, std::vector<std::string> semanticNames, std::vector<D3D12_INPUT_ELEMENT_DESC> layout, PBRPrimitives *pPrimitive)
{
//=================================================================================================
// let vertex shader know what buffers are present
// The Shader Code glTF20_EX.hlsl has if defs that are enable using these attributes when compiled
// and ref by the D3D12_GRAPHICS_PIPELINE_STATE_DESC
//=================================================================================================
bool Has_Normals = false;
std::map<std::string, std::string> attributeDefines;
for (unsigned int i = 0; i < layout.size(); i++)
{
layout[i].SemanticName = semanticNames[i].c_str();
attributeDefines[std::string("HAS_") + layout[i].SemanticName] = "1";
if (semanticNames[i].compare("NORMAL") == 0) Has_Normals = true;
}
// Compile shaders
//
ID3DBlob *pBlobShaderVert, *pBlobShaderPixel;
{
// build macro structure
//
std::vector<D3D_SHADER_MACRO> macros;
CompileMacros(&attributeDefines, ¯os);
CompileMacros(&pPrimitive->m_pMaterial->m_defines, ¯os);
macros.push_back(D3D_SHADER_MACRO{ NULL, NULL });
ID3DBlob *pError;
D3DCompileFromFile(L"./plugins/shaders/glTF20_EX.hlsl", macros.data(), nullptr, "mainVS", "vs_5_0", 0, 0, &pBlobShaderVert, &pError);
D3DCompileFromFile(L"./plugins/shaders/glTF20_EX.hlsl", macros.data(), nullptr, "mainPS", "ps_5_0", 0, 0, &pBlobShaderPixel, &pError);
if (pError != NULL)
{
char *msg = (char *)pError->GetBufferPointer();
MessageBoxA(0, msg, "", 0);
}
}
// Create root signature
//
{
CD3DX12_DESCRIPTOR_RANGE DescRange[4];
DescRange[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0); // b0 <- per frame
DescRange[1].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, pPrimitive->m_pMaterial->m_textureCount, 0); // t0 <- per material
DescRange[2].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 1); // b1 <- per material parameters
DescRange[3].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SAMPLER, 5, 0); // s0 <- samplers
CD3DX12_ROOT_PARAMETER RTSlot[4];
RTSlot[0].InitAsDescriptorTable(1, &DescRange[0], D3D12_SHADER_VISIBILITY_ALL);
RTSlot[1].InitAsDescriptorTable(1, &DescRange[1], D3D12_SHADER_VISIBILITY_PIXEL);
RTSlot[2].InitAsDescriptorTable(1, &DescRange[2], D3D12_SHADER_VISIBILITY_ALL);
RTSlot[3].InitAsDescriptorTable(1, &DescRange[3], D3D12_SHADER_VISIBILITY_PIXEL);
// the root signature contains 3 slots to be used
CD3DX12_ROOT_SIGNATURE_DESC descRootSignature = CD3DX12_ROOT_SIGNATURE_DESC();
descRootSignature.NumParameters = 4;
descRootSignature.pParameters = RTSlot;
descRootSignature.NumStaticSamplers = 0;
descRootSignature.pStaticSamplers = NULL;
// deny uneccessary access to certain pipeline stages
descRootSignature.Flags = D3D12_ROOT_SIGNATURE_FLAG_NONE
| D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT
//| D3D12_ROOT_SIGNATURE_FLAG_DENY_VERTEX_SHADER_ROOT_ACCESS
| D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS
| D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS
| D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS;
//| D3D12_ROOT_SIGNATURE_FLAG_DENY_PIXEL_SHADER_ROOT_ACCESS;
ID3DBlob *pOutBlob, *pErrorBlob = NULL;
ThrowIfFailed(D3D12SerializeRootSignature(&descRootSignature, D3D_ROOT_SIGNATURE_VERSION_1, &pOutBlob, &pErrorBlob));
ThrowIfFailed(
pDevice->CreateRootSignature(
node,
pOutBlob->GetBufferPointer(),
pOutBlob->GetBufferSize(),
IID_PPV_ARGS(&pPrimitive->m_RootSignature))
);
pPrimitive->m_RootSignature->SetName(L"OnCreatePrimitiveColorPass");
pOutBlob->Release();
if (pErrorBlob)
pErrorBlob->Release();
}
D3D12_RENDER_TARGET_BLEND_DESC blendingOpaque = D3D12_RENDER_TARGET_BLEND_DESC
{
FALSE,FALSE,
D3D12_BLEND_ONE, D3D12_BLEND_ZERO, D3D12_BLEND_OP_ADD,
D3D12_BLEND_ONE, D3D12_BLEND_ZERO, D3D12_BLEND_OP_ADD,
D3D12_LOGIC_OP_NOOP,
D3D12_COLOR_WRITE_ENABLE_ALL,
};
D3D12_RENDER_TARGET_BLEND_DESC blendingBlend = D3D12_RENDER_TARGET_BLEND_DESC
{
TRUE,FALSE,
D3D12_BLEND_SRC_ALPHA, D3D12_BLEND_INV_SRC_ALPHA, D3D12_BLEND_OP_ADD,
D3D12_BLEND_ONE, D3D12_BLEND_ZERO, D3D12_BLEND_OP_ADD,
D3D12_LOGIC_OP_NOOP,
D3D12_COLOR_WRITE_ENABLE_ALL,
};
// Create a PSO description
//
if (!pBlobShaderVert || !pBlobShaderPixel)
{
throw 1;
}
D3D12_GRAPHICS_PIPELINE_STATE_DESC descPso = {};
descPso.InputLayout = { layout.data(), (UINT)layout.size() };
descPso.pRootSignature = pPrimitive->m_RootSignature.Get();
descPso.VS = { reinterpret_cast<BYTE*>(pBlobShaderVert->GetBufferPointer()), pBlobShaderVert->GetBufferSize() };
descPso.PS = { reinterpret_cast<BYTE*>(pBlobShaderPixel->GetBufferPointer()), pBlobShaderPixel->GetBufferSize() };
descPso.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
descPso.RasterizerState.CullMode = D3D12_CULL_MODE_FRONT;
// Decide on default view based on attributes available
if (m_pGLTFData)
{
if(m_pGLTFData->isBinFile)
descPso.RasterizerState.FillMode = D3D12_FILL_MODE_WIREFRAME;
else {
if (Has_Normals)
{
descPso.RasterizerState.FillMode = D3D12_FILL_MODE_SOLID;
}
else
descPso.RasterizerState.FillMode = D3D12_FILL_MODE_WIREFRAME;
}
}
else
{
if (Has_Normals)
{
descPso.RasterizerState.FillMode = D3D12_FILL_MODE_SOLID;
}
else
descPso.RasterizerState.FillMode = D3D12_FILL_MODE_WIREFRAME;
}
descPso.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
descPso.BlendState.IndependentBlendEnable = TRUE;
descPso.BlendState.RenderTarget[0] = blendingBlend;
descPso.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
descPso.DepthStencilState.DepthFunc = D3D12_COMPARISON_FUNC_LESS;
descPso.SampleMask = UINT_MAX;
descPso.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
descPso.NumRenderTargets = 1;
descPso.RTVFormats[0] = DXGI_FORMAT_R16G16B16A16_UNORM;
descPso.DSVFormat = DXGI_FORMAT_D32_FLOAT;
descPso.SampleDesc.Count = 4;
descPso.NodeMask = node;
descPso.Flags = D3D12_PIPELINE_STATE_FLAG_NONE;
ThrowIfFailed(
pDevice->CreateGraphicsPipelineState(&descPso, IID_PPV_ARGS(&pPrimitive->m_PipelineRender))
);
// create samplers if not initialized (this should happen once)
if (m_sampler.GetSize()==0)
{
m_pResourceViewHeaps->AllocSamplerDescriptor(5, &m_sampler);
//for pbr materials
D3D12_SAMPLER_DESC SamplerDesc;
ZeroMemory(&SamplerDesc, sizeof(SamplerDesc));
SamplerDesc.Filter = D3D12_FILTER_MIN_MAG_LINEAR_MIP_POINT;
SamplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
SamplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
SamplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
SamplerDesc.BorderColor[0] = 0.0f;
SamplerDesc.BorderColor[1] = 0.0f;
SamplerDesc.BorderColor[2] = 0.0f;
SamplerDesc.BorderColor[3] = 0.0f;
SamplerDesc.MinLOD = 0.0f;
SamplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
SamplerDesc.MipLODBias = 0;
SamplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
SamplerDesc.MaxAnisotropy = 1;
pDevice->CreateSampler(&SamplerDesc, m_sampler.GetCPU(0));
// diffuse env map sampler
ZeroMemory(&SamplerDesc, sizeof(SamplerDesc));
SamplerDesc.Filter = D3D12_FILTER_MIN_MAG_LINEAR_MIP_POINT;
SamplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
SamplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
SamplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
SamplerDesc.BorderColor[0] = 0.0f;
SamplerDesc.BorderColor[1] = 0.0f;
SamplerDesc.BorderColor[2] = 0.0f;
SamplerDesc.BorderColor[3] = 0.0f;
SamplerDesc.MinLOD = 0.0f;
SamplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
SamplerDesc.MipLODBias = 0;
SamplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
SamplerDesc.MaxAnisotropy = 1;
pDevice->CreateSampler(&SamplerDesc, m_sampler.GetCPU(1));
// specular env map sampler
ZeroMemory(&SamplerDesc, sizeof(SamplerDesc));
SamplerDesc.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
SamplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
SamplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
SamplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
SamplerDesc.BorderColor[0] = 0.0f;
SamplerDesc.BorderColor[1] = 0.0f;
SamplerDesc.BorderColor[2] = 0.0f;
SamplerDesc.BorderColor[3] = 0.0f;
SamplerDesc.MinLOD = 0.0f;
SamplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
SamplerDesc.MipLODBias = 0;
SamplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
SamplerDesc.MaxAnisotropy = 1;
pDevice->CreateSampler(&SamplerDesc, m_sampler.GetCPU(2));
// specular BRDF lut sampler
ZeroMemory(&SamplerDesc, sizeof(SamplerDesc));
SamplerDesc.Filter = D3D12_FILTER_MIN_MAG_LINEAR_MIP_POINT;
SamplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
SamplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
SamplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
SamplerDesc.BorderColor[0] = 0.0f;
SamplerDesc.BorderColor[1] = 0.0f;
SamplerDesc.BorderColor[2] = 0.0f;
SamplerDesc.BorderColor[3] = 0.0f;
SamplerDesc.MinLOD = 0.0f;
SamplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
SamplerDesc.MipLODBias = 0;
SamplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
SamplerDesc.MaxAnisotropy = 1;
pDevice->CreateSampler(&SamplerDesc, m_sampler.GetCPU(3));
// specular BRDF lut sampler
D3D12_SAMPLER_DESC samplerShadow = {
D3D12_FILTER_COMPARISON_MIN_MAG_LINEAR_MIP_POINT,
D3D12_TEXTURE_ADDRESS_MODE_CLAMP,
D3D12_TEXTURE_ADDRESS_MODE_CLAMP,
D3D12_TEXTURE_ADDRESS_MODE_CLAMP,
0.0f,
1u,
D3D12_COMPARISON_FUNC_LESS_EQUAL,
{ 0.0f, 0.0f, 0.0f, 0.0f },
0.0f,
D3D12_FLOAT32_MAX
};
pDevice->CreateSampler(&samplerShadow, m_sampler.GetCPU(4));
}
pPrimitive->m_sampler = &m_sampler;
}
// Load the assets.
HRESULT VolumetricAnimation::LoadAssets()
{
HRESULT hr;
// Create a root signature consisting of a descriptor table with a CBV SRV and a sampler.
{
CD3DX12_DESCRIPTOR_RANGE ranges[3];
CD3DX12_ROOT_PARAMETER rootParameters[3];
ranges[0].Init( D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0 );
ranges[1].Init( D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0 );
ranges[2].Init( D3D12_DESCRIPTOR_RANGE_TYPE_UAV, 1, 0 );
rootParameters[RootParameterCBV].InitAsDescriptorTable( 1, &ranges[0], D3D12_SHADER_VISIBILITY_ALL );
rootParameters[RootParameterSRV].InitAsDescriptorTable( 1, &ranges[1], D3D12_SHADER_VISIBILITY_PIXEL );
rootParameters[RootParameterUAV].InitAsDescriptorTable( 1, &ranges[2], D3D12_SHADER_VISIBILITY_ALL );
D3D12_STATIC_SAMPLER_DESC sampler = {};
sampler.Filter = D3D12_FILTER_MIN_MAG_MIP_POINT;
sampler.AddressU = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressV = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressW = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.MipLODBias = 0;
sampler.MaxAnisotropy = 0;
sampler.ComparisonFunc = D3D12_COMPARISON_FUNC_NEVER;
sampler.BorderColor = D3D12_STATIC_BORDER_COLOR_TRANSPARENT_BLACK;
sampler.MinLOD = 0.0f;
sampler.MaxLOD = D3D12_FLOAT32_MAX;
sampler.ShaderRegister = 0;
sampler.RegisterSpace = 0;
sampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_PIXEL;
// Allow input layout and deny unnecessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS;
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init( _countof(rootParameters), rootParameters, 1, &sampler, rootSignatureFlags );
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
V( D3D12SerializeRootSignature( &rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error ) );
if ( error ) PRINTERROR( reinterpret_cast< const char* >( error->GetBufferPointer() ) );
VRET( m_device->CreateRootSignature( 0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS( &m_graphicsRootSignature ) ) );
DXDebugName( m_graphicsRootSignature );
// Create compute signature. Must change visibility for the SRV.
rootParameters[RootParameterSRV].ShaderVisibility = D3D12_SHADER_VISIBILITY_ALL;
CD3DX12_ROOT_SIGNATURE_DESC computeRootSignatureDesc( _countof( rootParameters ), rootParameters, 0, nullptr );
VRET( D3D12SerializeRootSignature( &computeRootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error ) );
VRET( m_device->CreateRootSignature( 0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS( &m_computeRootSignature ) ) );
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
ComPtr<ID3DBlob> computeShader;
UINT compileFlags = 0;
VRET( CompileShaderFromFile( GetAssetFullPath( _T( "VolumetricAnimation_shader.hlsl" ) ).c_str(), nullptr, D3D_COMPILE_STANDARD_FILE_INCLUDE, "vsmain", "vs_5_0", compileFlags, 0, &vertexShader ) );
VRET( CompileShaderFromFile( GetAssetFullPath( _T( "VolumetricAnimation_shader.hlsl" ) ).c_str(), nullptr, D3D_COMPILE_STANDARD_FILE_INCLUDE, "psmain", "ps_5_0", compileFlags, 0, &pixelShader ) );
VRET( CompileShaderFromFile( GetAssetFullPath( _T( "VolumetricAnimation_shader.hlsl" ) ).c_str(), nullptr, D3D_COMPILE_STANDARD_FILE_INCLUDE, "csmain", "cs_5_0", compileFlags, 0, &computeShader ) );
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
CD3DX12_DEPTH_STENCIL_DESC depthStencilDesc( D3D12_DEFAULT );
depthStencilDesc.DepthEnable = true;
depthStencilDesc.DepthWriteMask = D3D12_DEPTH_WRITE_MASK_ALL;
depthStencilDesc.DepthFunc = D3D12_COMPARISON_FUNC_LESS_EQUAL;
depthStencilDesc.StencilEnable = FALSE;
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof( inputElementDescs ) };
psoDesc.pRootSignature = m_graphicsRootSignature.Get();
psoDesc.VS = { reinterpret_cast< UINT8* >( vertexShader->GetBufferPointer() ), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast< UINT8* >( pixelShader->GetBufferPointer() ), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC( D3D12_DEFAULT );
psoDesc.BlendState = CD3DX12_BLEND_DESC( D3D12_DEFAULT );
psoDesc.DepthStencilState = depthStencilDesc;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
VRET( m_device->CreateGraphicsPipelineState( &psoDesc, IID_PPV_ARGS( &m_pipelineState ) ) );
DXDebugName( m_pipelineState );
// Describe and create the compute pipeline state object (PSO).
D3D12_COMPUTE_PIPELINE_STATE_DESC computePsoDesc = {};
computePsoDesc.pRootSignature = m_computeRootSignature.Get();
computePsoDesc.CS = { reinterpret_cast< UINT8* >( computeShader->GetBufferPointer() ), computeShader->GetBufferSize() };
VRET( m_device->CreateComputePipelineState( &computePsoDesc, IID_PPV_ARGS( &m_computeState ) ) );
DXDebugName( m_computeState );
}
// Create the compute command list.
VRET( m_device->CreateCommandList( 0, D3D12_COMMAND_LIST_TYPE_COMPUTE, m_computeCmdAllocator.Get(),m_computeState.Get(), IID_PPV_ARGS( &m_computeCmdList ) ) );
DXDebugName( m_computeCmdList );
VRET( m_computeCmdList->Close() );
// Create the graphics command list.
VRET( m_device->CreateCommandList( 0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_graphicCmdAllocator.Get(), m_pipelineState.Get(), IID_PPV_ARGS( &m_graphicCmdList ) ) );
DXDebugName( m_graphicCmdList );
// Note: ComPtr's are CPU objects but this resource needs to stay in scope until
// the command list that references it has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resource is not
// prematurely destroyed.
ComPtr<ID3D12Resource> volumeBufferUploadHeap;
// Create the volumeBuffer.
{
UINT volumeBufferSize = m_volumeDepth*m_volumeHeight*m_volumeWidth * 4 * sizeof( UINT8 );
D3D12_RESOURCE_DESC bufferDesc = CD3DX12_RESOURCE_DESC::Buffer( volumeBufferSize, D3D12_RESOURCE_FLAG_ALLOW_UNORDERED_ACCESS );
D3D12_RESOURCE_DESC uploadBufferDesc = CD3DX12_RESOURCE_DESC::Buffer( volumeBufferSize );
VRET( m_device->CreateCommittedResource(&CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_DEFAULT ),D3D12_HEAP_FLAG_NONE,
&bufferDesc,D3D12_RESOURCE_STATE_COPY_DEST,nullptr,IID_PPV_ARGS( &m_volumeBuffer ) ) );
const UINT64 uploadBufferSize = GetRequiredIntermediateSize( m_volumeBuffer.Get(), 0, 1 );
// Create the GPU upload buffer.
VRET( m_device->CreateCommittedResource(&CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_UPLOAD ),D3D12_HEAP_FLAG_NONE,
&uploadBufferDesc,D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,IID_PPV_ARGS( &volumeBufferUploadHeap ) ) );
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the Texture2D.
UINT8* volumeBuffer = ( UINT8* ) malloc( volumeBufferSize );
memset( volumeBuffer, 64, volumeBufferSize );
//float radius = m_volumeHeight / 2.f;
float a = m_volumeWidth / 2.f;
float b = m_volumeHeight / 2.f;
float c = m_volumeDepth / 2.f;
float radius = sqrt( a*a + b*b + c*c );
for ( UINT z = 0; z < m_volumeDepth; z++ )
for ( UINT y = 0; y < m_volumeHeight; y++ )
for ( UINT x = 0; x < m_volumeWidth; x++ )
{
float _x = x - m_volumeWidth / 2.f;
float _y = y - m_volumeHeight / 2.f;
float _z = z - m_volumeDepth / 2.f;
//float currentRaidus =abs(_x)+abs(_y)+abs(_z);
float currentRaidus = sqrt( _x*_x + _y*_y + _z*_z );
float scale = currentRaidus *3.f / radius;
UINT idx = 4 - (UINT)floor( scale );
UINT interm = ( UINT ) ( 192 * scale +0.5f );
UINT8 col = interm % 192+1;
volumeBuffer[( x + y*m_volumeWidth + z*m_volumeHeight*m_volumeWidth ) * 4 + 0] += col * m_constantBufferData.colVal[idx].x;
volumeBuffer[( x + y*m_volumeWidth + z*m_volumeHeight*m_volumeWidth ) * 4 + 1] += col * m_constantBufferData.colVal[idx].y;
volumeBuffer[( x + y*m_volumeWidth + z*m_volumeHeight*m_volumeWidth ) * 4 + 2] += col * m_constantBufferData.colVal[idx].z;
volumeBuffer[( x + y*m_volumeWidth + z*m_volumeHeight*m_volumeWidth ) * 4 + 3] = m_constantBufferData.colVal[idx].w;
}
D3D12_SUBRESOURCE_DATA volumeBufferData = {};
volumeBufferData.pData = &volumeBuffer[0];
volumeBufferData.RowPitch = volumeBufferSize;
volumeBufferData.SlicePitch = volumeBufferData.RowPitch;
UpdateSubresources( m_graphicCmdList.Get(), m_volumeBuffer.Get(), volumeBufferUploadHeap.Get(), 0, 0, 1, &volumeBufferData );
m_graphicCmdList->ResourceBarrier( 1, &CD3DX12_RESOURCE_BARRIER::Transition( m_volumeBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_UNORDERED_ACCESS ) );
// Describe and create a SRV for the volumeBuffer.
D3D12_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
srvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
srvDesc.Format = DXGI_FORMAT_UNKNOWN;
srvDesc.ViewDimension = D3D12_SRV_DIMENSION_BUFFER;
srvDesc.Buffer.FirstElement = 0;
srvDesc.Buffer.NumElements = m_volumeDepth*m_volumeHeight*m_volumeWidth;
srvDesc.Buffer.StructureByteStride = 4 * sizeof( UINT8 );
srvDesc.Buffer.Flags = D3D12_BUFFER_SRV_FLAG_NONE;
CD3DX12_CPU_DESCRIPTOR_HANDLE srvHandle( m_cbvsrvuavHeap->GetCPUDescriptorHandleForHeapStart(), RootParameterSRV, m_cbvsrvuavDescriptorSize );
m_device->CreateShaderResourceView( m_volumeBuffer.Get(), &srvDesc, srvHandle );
// Describe and create a UAV for the volumeBuffer.
D3D12_UNORDERED_ACCESS_VIEW_DESC uavDesc = {};
uavDesc.Format = DXGI_FORMAT_UNKNOWN;
uavDesc.ViewDimension = D3D12_UAV_DIMENSION_BUFFER;
uavDesc.Buffer.FirstElement = 0;
uavDesc.Buffer.NumElements = m_volumeWidth*m_volumeHeight*m_volumeDepth;
uavDesc.Buffer.StructureByteStride = 4 * sizeof( UINT8 );
uavDesc.Buffer.CounterOffsetInBytes = 0;
uavDesc.Buffer.Flags = D3D12_BUFFER_UAV_FLAG_NONE;
CD3DX12_CPU_DESCRIPTOR_HANDLE uavHandle( m_cbvsrvuavHeap->GetCPUDescriptorHandleForHeapStart(), RootParameterUAV, m_cbvsrvuavDescriptorSize );
m_device->CreateUnorderedAccessView( m_volumeBuffer.Get(), nullptr, &uavDesc, uavHandle );
free( volumeBuffer );
}
// Create the vertex buffer.
// Note: ComPtr's are CPU objects but this resource needs to stay in scope until
// the command list that references it has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resource is not
// prematurely destroyed.
ComPtr<ID3D12Resource> vertexBufferUpload;
{
// Define the geometry for a triangle.
Vertex cubeVertices[] =
{
{ XMFLOAT3( -128.f, -128.f, -128.f ) },
{ XMFLOAT3( -128.f, -128.f, 128.f ) },
{ XMFLOAT3( -128.f, 128.f, -128.f ) },
{ XMFLOAT3( -128.f, 128.f, 128.f ) },
{ XMFLOAT3( 128.f, -128.f, -128.f )},
{ XMFLOAT3( 128.f, -128.f, 128.f )},
{ XMFLOAT3( 128.f, 128.f, -128.f )},
{ XMFLOAT3( 128.f, 128.f, 128.f )},
};
const UINT vertexBufferSize = sizeof( cubeVertices );
VRET( m_device->CreateCommittedResource( &CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_UPLOAD ), D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer( vertexBufferSize ), D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr, IID_PPV_ARGS( &vertexBufferUpload ) ) );
VRET( m_device->CreateCommittedResource( &CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_DEFAULT ), D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer( vertexBufferSize ), D3D12_RESOURCE_STATE_COPY_DEST,
nullptr, IID_PPV_ARGS( &m_vertexBuffer ) ) );
DXDebugName( m_vertexBuffer );
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = reinterpret_cast< UINT8* >( cubeVertices );
vertexData.RowPitch = vertexBufferSize;
vertexData.SlicePitch = vertexBufferSize;
UpdateSubresources<1>( m_graphicCmdList.Get(), m_vertexBuffer.Get(), vertexBufferUpload.Get(), 0, 0, 1, &vertexData );
m_graphicCmdList->ResourceBarrier( 1, &CD3DX12_RESOURCE_BARRIER::Transition( m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST,
D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER ));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof( Vertex );
m_vertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create the index buffer
// Note: ComPtr's are CPU objects but this resource needs to stay in scope until
// the command list that references it has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resource is not
// prematurely destroyed.
ComPtr<ID3D12Resource> indexBufferUpload;
{
uint16_t cubeIndices[] =
{
0,2,1, 1,2,3, 4,5,6, 5,7,6, 0,1,5, 0,5,4, 2,6,7, 2,7,3, 0,4,6, 0,6,2, 1,3,7, 1,7,5,
};
const UINT indexBufferSize = sizeof( cubeIndices );
VRET( m_device->CreateCommittedResource( &CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_UPLOAD ), D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer( indexBufferSize ), D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr, IID_PPV_ARGS( &indexBufferUpload ) ) );
VRET( m_device->CreateCommittedResource( &CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_DEFAULT ), D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer( indexBufferSize ), D3D12_RESOURCE_STATE_COPY_DEST,
nullptr, IID_PPV_ARGS( &m_indexBuffer ) ) );
DXDebugName( m_indexBuffer );
D3D12_SUBRESOURCE_DATA indexData = {};
indexData.pData = reinterpret_cast< UINT8* >( cubeIndices );
indexData.RowPitch = indexBufferSize;
indexData.SlicePitch = indexBufferSize;
UpdateSubresources<1>( m_graphicCmdList.Get(), m_indexBuffer.Get(), indexBufferUpload.Get(), 0, 0, 1, &indexData );
m_graphicCmdList->ResourceBarrier( 1, &CD3DX12_RESOURCE_BARRIER::Transition( m_indexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST,
D3D12_RESOURCE_STATE_INDEX_BUFFER ) );
m_indexBufferView.BufferLocation = m_indexBuffer->GetGPUVirtualAddress();
m_indexBufferView.SizeInBytes = sizeof( cubeIndices );
m_indexBufferView.Format = DXGI_FORMAT_R16_UINT;
}
// Create the constant buffer
{
VRET( m_device->CreateCommittedResource( &CD3DX12_HEAP_PROPERTIES( D3D12_HEAP_TYPE_UPLOAD ), D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer( 1024 * 64 ), D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr, IID_PPV_ARGS( &m_constantBuffer ) ) );
DXDebugName( m_constantBuffer );
// Describe and create a constant buffer view.
D3D12_CONSTANT_BUFFER_VIEW_DESC cbvDesc = {};
cbvDesc.BufferLocation = m_constantBuffer->GetGPUVirtualAddress();
cbvDesc.SizeInBytes = ( sizeof( ConstantBuffer ) + 255 ) & ~255; // CB size is required to be 256-byte aligned.
m_device->CreateConstantBufferView( &cbvDesc, m_cbvsrvuavHeap->GetCPUDescriptorHandleForHeapStart() );
// Initialize and map the constant buffers. We don't unmap this until the
// app closes. Keeping things mapped for the lifetime of the resource is okay.
CD3DX12_RANGE readRange( 0, 0 ); // We do not intend to read from this resource on the CPU.
VRET( m_constantBuffer->Map( 0, &readRange, reinterpret_cast< void** >( &m_pCbvDataBegin ) ) );
memcpy( m_pCbvDataBegin, &m_constantBufferData, sizeof( m_constantBufferData ) );
}
// Close the command list and execute it to begin the initial GPU setup.
VRET( m_graphicCmdList->Close() );
ID3D12CommandList* ppCommandLists[] = { m_graphicCmdList.Get() };
m_graphicCmdQueue->ExecuteCommandLists( _countof( ppCommandLists ), ppCommandLists );
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
VRET( m_device->CreateFence( 0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS( &m_fence ) ) );
DXDebugName( m_fence );
m_fenceValue = 1;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent( nullptr, FALSE, FALSE, nullptr );
if ( m_fenceEvent == nullptr )
{
VRET( HRESULT_FROM_WIN32( GetLastError() ) );
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGraphicsCmd();
}
XMVECTORF32 vecEye = { 500.0f, 500.0f, -500.0f };
XMVECTORF32 vecAt = { 0.0f, 0.0f, 0.0f };
m_camera.SetViewParams( vecEye, vecAt );
m_camera.SetEnablePositionMovement( true );
m_camera.SetButtonMasks( MOUSE_RIGHT_BUTTON, MOUSE_WHEEL, MOUSE_LEFT_BUTTON );
return S_OK;
}
// Load the sample assets.
void D3D12DynamicIndexing::LoadAssets()
{
// Note: ComPtr's are CPU objects but these resources need to stay in scope until
// the command list that references them has finished executing on the GPU.
// We will flush the GPU at the end of this method to ensure the resources are not
// prematurely destroyed.
ComPtr<ID3D12Resource> vertexBufferUploadHeap;
ComPtr<ID3D12Resource> indexBufferUploadHeap;
ComPtr<ID3D12Resource> textureUploadHeap;
ComPtr<ID3D12Resource> materialsUploadHeap;
// Create the root signature.
{
CD3DX12_DESCRIPTOR_RANGE ranges[3];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1 + CityMaterialCount, 0); // Diffuse texture + array of materials.
ranges[1].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SAMPLER, 1, 0);
ranges[2].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
CD3DX12_ROOT_PARAMETER rootParameters[4];
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_PIXEL);
rootParameters[1].InitAsDescriptorTable(1, &ranges[1], D3D12_SHADER_VISIBILITY_PIXEL);
rootParameters[2].InitAsDescriptorTable(1, &ranges[2], D3D12_SHADER_VISIBILITY_VERTEX);
rootParameters[3].InitAsConstants(1, 0, 0, D3D12_SHADER_VISIBILITY_PIXEL);
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
NAME_D3D12_OBJECT(m_rootSignature);
}
// Create the pipeline state, which includes loading shaders.
{
UINT8* pVertexShaderData;
UINT8* pPixelShaderData;
UINT vertexShaderDataLength;
UINT pixelShaderDataLength;
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(L"shader_mesh_simple_vert.cso").c_str(), &pVertexShaderData, &vertexShaderDataLength));
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(L"shader_mesh_dynamic_indexing_pixel.cso").c_str(), &pPixelShaderData, &pixelShaderDataLength));
CD3DX12_RASTERIZER_DESC rasterizerStateDesc(D3D12_DEFAULT);
rasterizerStateDesc.CullMode = D3D12_CULL_MODE_NONE;
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { SampleAssets::StandardVertexDescription, SampleAssets::StandardVertexDescriptionNumElements };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(pVertexShaderData, vertexShaderDataLength);
psoDesc.PS = CD3DX12_SHADER_BYTECODE(pPixelShaderData, pixelShaderDataLength);
psoDesc.RasterizerState = rasterizerStateDesc;
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState = CD3DX12_DEPTH_STENCIL_DESC(D3D12_DEFAULT);
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
NAME_D3D12_OBJECT(m_pipelineState);
delete pVertexShaderData;
delete pPixelShaderData;
}
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocator.Get(), nullptr, IID_PPV_ARGS(&m_commandList)));
NAME_D3D12_OBJECT(m_commandList);
// Create render target views (RTVs).
CD3DX12_CPU_DESCRIPTOR_HANDLE rtvHandle(m_rtvHeap->GetCPUDescriptorHandleForHeapStart());
for (UINT i = 0; i < FrameCount; i++)
{
ThrowIfFailed(m_swapChain->GetBuffer(i, IID_PPV_ARGS(&m_renderTargets[i])));
m_device->CreateRenderTargetView(m_renderTargets[i].Get(), nullptr, rtvHandle);
rtvHandle.Offset(1, m_rtvDescriptorSize);
NAME_D3D12_OBJECT_INDEXED(m_renderTargets, i);
}
// Read in mesh data for vertex/index buffers.
UINT8* pMeshData;
UINT meshDataLength;
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(SampleAssets::DataFileName).c_str(), &pMeshData, &meshDataLength));
// Create the vertex buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::VertexDataSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::VertexDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&vertexBufferUploadHeap)));
NAME_D3D12_OBJECT(m_vertexBuffer);
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = pMeshData + SampleAssets::VertexDataOffset;
vertexData.RowPitch = SampleAssets::VertexDataSize;
vertexData.SlicePitch = vertexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_vertexBuffer.Get(), vertexBufferUploadHeap.Get(), 0, 0, 1, &vertexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = SampleAssets::StandardVertexStride;
m_vertexBufferView.SizeInBytes = SampleAssets::VertexDataSize;
}
// Create the index buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::IndexDataSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_indexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::IndexDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&indexBufferUploadHeap)));
NAME_D3D12_OBJECT(m_indexBuffer);
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the index buffer.
D3D12_SUBRESOURCE_DATA indexData = {};
indexData.pData = pMeshData + SampleAssets::IndexDataOffset;
indexData.RowPitch = SampleAssets::IndexDataSize;
indexData.SlicePitch = indexData.RowPitch;
UpdateSubresources<1>(m_commandList.Get(), m_indexBuffer.Get(), indexBufferUploadHeap.Get(), 0, 0, 1, &indexData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_indexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_INDEX_BUFFER));
// Describe the index buffer view.
m_indexBufferView.BufferLocation = m_indexBuffer->GetGPUVirtualAddress();
m_indexBufferView.Format = SampleAssets::StandardIndexFormat;
m_indexBufferView.SizeInBytes = SampleAssets::IndexDataSize;
m_numIndices = SampleAssets::IndexDataSize / 4; // R32_UINT (SampleAssets::StandardIndexFormat) = 4 bytes each.
}
// Create the textures and sampler.
{
// Procedurally generate an array of textures to use as city materials.
{
// All of these materials use the same texture desc.
D3D12_RESOURCE_DESC textureDesc = {};
textureDesc.MipLevels = 1;
textureDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
textureDesc.Width = CityMaterialTextureWidth;
textureDesc.Height = CityMaterialTextureHeight;
textureDesc.Flags = D3D12_RESOURCE_FLAG_NONE;
textureDesc.DepthOrArraySize = 1;
textureDesc.SampleDesc.Count = 1;
textureDesc.SampleDesc.Quality = 0;
textureDesc.Dimension = D3D12_RESOURCE_DIMENSION_TEXTURE2D;
// The textures evenly span the color rainbow so that each city gets
// a different material.
float materialGradStep = (1.0f / static_cast<float>(CityMaterialCount));
// Generate texture data.
std::vector<std::vector<unsigned char>> cityTextureData;
cityTextureData.resize(CityMaterialCount);
for (UINT i = 0; i < CityMaterialCount; ++i)
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&textureDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_cityMaterialTextures[i])));
NAME_D3D12_OBJECT_INDEXED(m_cityMaterialTextures, i);
// Fill the texture.
float t = i * materialGradStep;
cityTextureData[i].resize(CityMaterialTextureWidth * CityMaterialTextureHeight * CityMaterialTextureChannelCount);
for (int x = 0; x < CityMaterialTextureWidth; ++x)
{
for (int y = 0; y < CityMaterialTextureHeight; ++y)
{
// Compute the appropriate index into the buffer based on the x/y coordinates.
int pixelIndex = (y * CityMaterialTextureChannelCount * CityMaterialTextureWidth) + (x * CityMaterialTextureChannelCount);
// Determine this row's position along the rainbow gradient.
float tPrime = t + ((static_cast<float>(y) / static_cast<float>(CityMaterialTextureHeight)) * materialGradStep);
// Compute the RGB value for this position along the rainbow
// and pack the pixel value.
XMVECTOR hsl = XMVectorSet(tPrime, 0.5f, 0.5f, 1.0f);
XMVECTOR rgb = XMColorHSLToRGB(hsl);
cityTextureData[i][pixelIndex + 0] = static_cast<unsigned char>((255 * XMVectorGetX(rgb)));
cityTextureData[i][pixelIndex + 1] = static_cast<unsigned char>((255 * XMVectorGetY(rgb)));
cityTextureData[i][pixelIndex + 2] = static_cast<unsigned char>((255 * XMVectorGetZ(rgb)));
cityTextureData[i][pixelIndex + 3] = 255;
}
}
}
// Upload texture data to the default heap resources.
{
const UINT subresourceCount = textureDesc.DepthOrArraySize * textureDesc.MipLevels;
const UINT64 uploadBufferStep = GetRequiredIntermediateSize(m_cityMaterialTextures[0].Get(), 0, subresourceCount); // All of our textures are the same size in this case.
const UINT64 uploadBufferSize = uploadBufferStep * CityMaterialCount;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(uploadBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&materialsUploadHeap)));
for (int i = 0; i < CityMaterialCount; ++i)
{
// Copy data to the intermediate upload heap and then schedule
// a copy from the upload heap to the appropriate texture.
D3D12_SUBRESOURCE_DATA textureData = {};
textureData.pData = &cityTextureData[i][0];
textureData.RowPitch = static_cast<LONG_PTR>((CityMaterialTextureChannelCount * textureDesc.Width));
textureData.SlicePitch = textureData.RowPitch * textureDesc.Height;
UpdateSubresources(m_commandList.Get(), m_cityMaterialTextures[i].Get(), materialsUploadHeap.Get(), i * uploadBufferStep, 0, subresourceCount, &textureData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_cityMaterialTextures[i].Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE));
}
}
}
// Load the occcity diffuse texture with baked-in ambient lighting.
// This texture will be blended with a texture from the materials
// array in the pixel shader.
{
D3D12_RESOURCE_DESC textureDesc = {};
textureDesc.MipLevels = SampleAssets::Textures[0].MipLevels;
textureDesc.Format = SampleAssets::Textures[0].Format;
textureDesc.Width = SampleAssets::Textures[0].Width;
textureDesc.Height = SampleAssets::Textures[0].Height;
textureDesc.Flags = D3D12_RESOURCE_FLAG_NONE;
textureDesc.DepthOrArraySize = 1;
textureDesc.SampleDesc.Count = 1;
textureDesc.SampleDesc.Quality = 0;
textureDesc.Dimension = D3D12_RESOURCE_DIMENSION_TEXTURE2D;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&textureDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_cityDiffuseTexture)));
const UINT subresourceCount = textureDesc.DepthOrArraySize * textureDesc.MipLevels;
const UINT64 uploadBufferSize = GetRequiredIntermediateSize(m_cityDiffuseTexture.Get(), 0, subresourceCount);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(uploadBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&textureUploadHeap)));
NAME_D3D12_OBJECT(m_cityDiffuseTexture);
// Copy data to the intermediate upload heap and then schedule
// a copy from the upload heap to the diffuse texture.
D3D12_SUBRESOURCE_DATA textureData = {};
textureData.pData = pMeshData + SampleAssets::Textures[0].Data[0].Offset;
textureData.RowPitch = SampleAssets::Textures[0].Data[0].Pitch;
textureData.SlicePitch = SampleAssets::Textures[0].Data[0].Size;
UpdateSubresources(m_commandList.Get(), m_cityDiffuseTexture.Get(), textureUploadHeap.Get(), 0, 0, subresourceCount, &textureData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_cityDiffuseTexture.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE));
}
// Describe and create a sampler.
D3D12_SAMPLER_DESC samplerDesc = {};
samplerDesc.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
samplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
samplerDesc.MinLOD = 0;
samplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
samplerDesc.MipLODBias = 0.0f;
samplerDesc.MaxAnisotropy = 1;
samplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
m_device->CreateSampler(&samplerDesc, m_samplerHeap->GetCPUDescriptorHandleForHeapStart());
// Create SRV for the city's diffuse texture.
CD3DX12_CPU_DESCRIPTOR_HANDLE srvHandle(m_cbvSrvHeap->GetCPUDescriptorHandleForHeapStart(), 0, m_cbvSrvDescriptorSize);
D3D12_SHADER_RESOURCE_VIEW_DESC diffuseSrvDesc = {};
diffuseSrvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
diffuseSrvDesc.Format = SampleAssets::Textures->Format;
diffuseSrvDesc.ViewDimension = D3D12_SRV_DIMENSION_TEXTURE2D;
diffuseSrvDesc.Texture2D.MipLevels = 1;
m_device->CreateShaderResourceView(m_cityDiffuseTexture.Get(), &diffuseSrvDesc, srvHandle);
srvHandle.Offset(m_cbvSrvDescriptorSize);
// Create SRVs for each city material.
for (int i = 0; i < CityMaterialCount; ++i)
{
D3D12_SHADER_RESOURCE_VIEW_DESC materialSrvDesc = {};
materialSrvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
materialSrvDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
materialSrvDesc.ViewDimension = D3D12_SRV_DIMENSION_TEXTURE2D;
materialSrvDesc.Texture2D.MipLevels = 1;
m_device->CreateShaderResourceView(m_cityMaterialTextures[i].Get(), &materialSrvDesc, srvHandle);
srvHandle.Offset(m_cbvSrvDescriptorSize);
}
}
delete pMeshData;
// Create the depth stencil view.
{
D3D12_DEPTH_STENCIL_VIEW_DESC depthStencilDesc = {};
depthStencilDesc.Format = DXGI_FORMAT_D32_FLOAT;
depthStencilDesc.ViewDimension = D3D12_DSV_DIMENSION_TEXTURE2D;
depthStencilDesc.Flags = D3D12_DSV_FLAG_NONE;
D3D12_CLEAR_VALUE depthOptimizedClearValue = {};
depthOptimizedClearValue.Format = DXGI_FORMAT_D32_FLOAT;
depthOptimizedClearValue.DepthStencil.Depth = 1.0f;
depthOptimizedClearValue.DepthStencil.Stencil = 0;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Tex2D(DXGI_FORMAT_D32_FLOAT, m_width, m_height, 1, 0, 1, 0, D3D12_RESOURCE_FLAG_ALLOW_DEPTH_STENCIL),
D3D12_RESOURCE_STATE_DEPTH_WRITE,
&depthOptimizedClearValue,
IID_PPV_ARGS(&m_depthStencil)
));
NAME_D3D12_OBJECT(m_depthStencil);
m_device->CreateDepthStencilView(m_depthStencil.Get(), &depthStencilDesc, m_dsvHeap->GetCPUDescriptorHandleForHeapStart());
}
// Close the command list and execute it to begin the initial GPU setup.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
CreateFrameResources();
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValue, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValue++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
// Signal and increment the fence value.
const UINT64 fenceToWaitFor = m_fenceValue;
ThrowIfFailed(m_commandQueue->Signal(m_fence.Get(), fenceToWaitFor));
m_fenceValue++;
// Wait until the fence is completed.
ThrowIfFailed(m_fence->SetEventOnCompletion(fenceToWaitFor, m_fenceEvent));
WaitForSingleObject(m_fenceEvent, INFINITE);
}
}
// Load the sample assets.
void D3D12Fullscreen::LoadAssets()
{
D3D12_FEATURE_DATA_ROOT_SIGNATURE featureData = {};
// This is the highest version the sample supports. If CheckFeatureSupport succeeds, the HighestVersion returned will not be greater than this.
featureData.HighestVersion = D3D_ROOT_SIGNATURE_VERSION_1_1;
if (FAILED(m_device->CheckFeatureSupport(D3D12_FEATURE_ROOT_SIGNATURE, &featureData, sizeof(featureData))))
{
featureData.HighestVersion = D3D_ROOT_SIGNATURE_VERSION_1_0;
}
// Create a root signature consisting of a descriptor table with a single CBV.
{
CD3DX12_DESCRIPTOR_RANGE1 ranges[1];
CD3DX12_ROOT_PARAMETER1 rootParameters[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0, 0, D3D12_DESCRIPTOR_RANGE_FLAG_DATA_STATIC);
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_VERTEX);
// Allow input layout and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_PIXEL_SHADER_ROOT_ACCESS;
CD3DX12_VERSIONED_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init_1_1(_countof(rootParameters), rootParameters, 0, nullptr, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3DX12SerializeVersionedRootSignature(&rootSignatureDesc, featureData.HighestVersion, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_sceneRootSignature)));
NAME_D3D12_OBJECT(m_sceneRootSignature);
}
// Create a root signature consisting of a descriptor table with a SRV and a sampler.
{
CD3DX12_DESCRIPTOR_RANGE1 ranges[1];
CD3DX12_ROOT_PARAMETER1 rootParameters[1];
// We don't modify the SRV in the post-processing command list after
// SetGraphicsRootDescriptorTable is executed on the GPU so we can use the default
// range behavior: D3D12_DESCRIPTOR_RANGE_FLAG_DATA_STATIC_WHILE_SET_AT_EXECUTE
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_PIXEL);
// Allow input layout and pixel shader access and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS;
// Create a sampler.
D3D12_STATIC_SAMPLER_DESC sampler = {};
sampler.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
sampler.AddressU = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressV = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressW = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.MipLODBias = 0;
sampler.MaxAnisotropy = 0;
sampler.ComparisonFunc = D3D12_COMPARISON_FUNC_NEVER;
sampler.BorderColor = D3D12_STATIC_BORDER_COLOR_TRANSPARENT_BLACK;
sampler.MinLOD = 0.0f;
sampler.MaxLOD = D3D12_FLOAT32_MAX;
sampler.ShaderRegister = 0;
sampler.RegisterSpace = 0;
sampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_PIXEL;
CD3DX12_VERSIONED_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init_1_1(_countof(rootParameters), rootParameters, 1, &sampler, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3DX12SerializeVersionedRootSignature(&rootSignatureDesc, featureData.HighestVersion, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_postRootSignature)));
NAME_D3D12_OBJECT(m_postRootSignature);
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> sceneVertexShader;
ComPtr<ID3DBlob> scenePixelShader;
ComPtr<ID3DBlob> postVertexShader;
ComPtr<ID3DBlob> postPixelShader;
ComPtr<ID3DBlob> error;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"sceneShaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &sceneVertexShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"sceneShaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &scenePixelShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"postShaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &postVertexShader, &error));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"postShaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &postPixelShader, &error));
// Define the vertex input layouts.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
D3D12_INPUT_ELEMENT_DESC scaleInputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state objects (PSOs).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_sceneRootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(sceneVertexShader.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(scenePixelShader.Get());
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_scenePipelineState)));
NAME_D3D12_OBJECT(m_scenePipelineState);
psoDesc.InputLayout = { scaleInputElementDescs, _countof(scaleInputElementDescs) };
psoDesc.pRootSignature = m_postRootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(postVertexShader.Get());
psoDesc.PS = CD3DX12_SHADER_BYTECODE(postPixelShader.Get());
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_postPipelineState)));
NAME_D3D12_OBJECT(m_postPipelineState);
}
// Single-use command allocator and command list for creating resources.
ComPtr<ID3D12CommandAllocator> commandAllocator;
ComPtr<ID3D12GraphicsCommandList> commandList;
ThrowIfFailed(m_device->CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT, IID_PPV_ARGS(&commandAllocator)));
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, commandAllocator.Get(), nullptr, IID_PPV_ARGS(&commandList)));
// Create the command lists.
{
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_sceneCommandAllocators[m_frameIndex].Get(), m_scenePipelineState.Get(), IID_PPV_ARGS(&m_sceneCommandList)));
NAME_D3D12_OBJECT(m_sceneCommandList);
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_postCommandAllocators[m_frameIndex].Get(), m_postPipelineState.Get(), IID_PPV_ARGS(&m_postCommandList)));
NAME_D3D12_OBJECT(m_postCommandList);
// Close the command lists.
ThrowIfFailed(m_sceneCommandList->Close());
ThrowIfFailed(m_postCommandList->Close());
}
LoadSizeDependentResources();
LoadSceneResolutionDependentResources();
// Create/update the vertex buffer.
ComPtr<ID3D12Resource> sceneVertexBufferUpload;
{
// Define the geometry for a thin quad that will animate across the screen.
const float x = QuadWidth / 2.0f;
const float y = QuadHeight / 2.0f;
SceneVertex quadVertices[] =
{
{ { -x, -y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } },
{ { -x, y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } },
{ { x, -y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } },
{ { x, y, 1.0f }, { 1.0f, 1.0f, 1.0f, 1.0f } }
};
const UINT vertexBufferSize = sizeof(quadVertices);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_sceneVertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&sceneVertexBufferUpload)));
NAME_D3D12_OBJECT(m_sceneVertexBuffer);
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
UINT8* pVertexDataBegin;
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(sceneVertexBufferUpload->Map(0, &readRange, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, quadVertices, sizeof(quadVertices));
sceneVertexBufferUpload->Unmap(0, nullptr);
commandList->CopyBufferRegion(m_sceneVertexBuffer.Get(), 0, sceneVertexBufferUpload.Get(), 0, vertexBufferSize);
commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_sceneVertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer views.
m_sceneVertexBufferView.BufferLocation = m_sceneVertexBuffer->GetGPUVirtualAddress();
m_sceneVertexBufferView.StrideInBytes = sizeof(SceneVertex);
m_sceneVertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create/update the fullscreen quad vertex buffer.
ComPtr<ID3D12Resource> postVertexBufferUpload;
{
// Define the geometry for a fullscreen quad.
PostVertex quadVertices[] =
{
{ { -1.0f, -1.0f, 0.0f, 1.0f }, { 0.0f, 0.0f } }, // Bottom left.
{ { -1.0f, 1.0f, 0.0f, 1.0f }, { 0.0f, 1.0f } }, // Top left.
{ { 1.0f, -1.0f, 0.0f, 1.0f }, { 1.0f, 0.0f } }, // Bottom right.
{ { 1.0f, 1.0f, 0.0f, 1.0f }, { 1.0f, 1.0f } } // Top right.
};
const UINT vertexBufferSize = sizeof(quadVertices);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_postVertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&postVertexBufferUpload)));
NAME_D3D12_OBJECT(m_postVertexBuffer);
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
UINT8* pVertexDataBegin;
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(postVertexBufferUpload->Map(0, &readRange, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, quadVertices, sizeof(quadVertices));
postVertexBufferUpload->Unmap(0, nullptr);
commandList->CopyBufferRegion(m_postVertexBuffer.Get(), 0, postVertexBufferUpload.Get(), 0, vertexBufferSize);
commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_postVertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
// Initialize the vertex buffer views.
m_postVertexBufferView.BufferLocation = m_postVertexBuffer->GetGPUVirtualAddress();
m_postVertexBufferView.StrideInBytes = sizeof(PostVertex);
m_postVertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create the constant buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(sizeof(SceneConstantBuffer) * FrameCount),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_sceneConstantBuffer)));
NAME_D3D12_OBJECT(m_sceneConstantBuffer);
// Describe and create constant buffer views.
D3D12_CONSTANT_BUFFER_VIEW_DESC cbvDesc = {};
cbvDesc.BufferLocation = m_sceneConstantBuffer->GetGPUVirtualAddress();
cbvDesc.SizeInBytes = sizeof(SceneConstantBuffer);
CD3DX12_CPU_DESCRIPTOR_HANDLE cpuHandle(m_cbvSrvHeap->GetCPUDescriptorHandleForHeapStart(), 1, m_cbvSrvDescriptorSize);
for (UINT n = 0; n < FrameCount; n++)
{
m_device->CreateConstantBufferView(&cbvDesc, cpuHandle);
cbvDesc.BufferLocation += sizeof(SceneConstantBuffer);
cpuHandle.Offset(m_cbvSrvDescriptorSize);
}
// Map and initialize the constant buffer. We don't unmap this until the
// app closes. Keeping things mapped for the lifetime of the resource is okay.
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_sceneConstantBuffer->Map(0, &readRange, reinterpret_cast<void**>(&m_pCbvDataBegin)));
memcpy(m_pCbvDataBegin, &m_sceneConstantBufferData, sizeof(m_sceneConstantBufferData));
}
// Close the resource creation command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(commandList->Close());
ID3D12CommandList* ppCommandLists[] = { commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute before continuing.
WaitForGpu();
}
}
void Triangle::init_shader(ComPtr<ID3D12Device> pD3D12Device)
{
CD3DX12_DESCRIPTOR_RANGE ranges[1];
CD3DX12_ROOT_PARAMETER rootParameters[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_VERTEX);
// Allow input layout and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_PIXEL_SHADER_ROOT_ACCESS;
//Create a root signature
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error);
pD3D12Device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_pRootSignature));
//Create the pipeline state, which includes compiling and loading shaders
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#ifdef _DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
D3DCompileFromFile(L"triangle.vsh", nullptr, nullptr, "VS_MAIN", "vs_5_0", compileFlags, 0, &vertexShader, nullptr);
D3DCompileFromFile(L"triangle.psh", nullptr, nullptr, "PS_MAIN", "ps_5_0", compileFlags, 0, &pixelShader, nullptr);
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
//Descrbe and create the graphics pipeline state object (PSO)
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = {inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_pRootSignature.Get();
psoDesc.VS = {vertexShader->GetBufferPointer(), vertexShader->GetBufferSize() };
psoDesc.PS = {pixelShader->GetBufferPointer(), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
HRESULT res = pD3D12Device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pPipelineState));
assert(res == S_OK);
}
// Load the sample assets.
void D3D12Multithreading::LoadAssets()
{
// Create the root signature.
{
CD3DX12_DESCRIPTOR_RANGE ranges[4]; // Perfomance TIP: Order from most frequent to least frequent.
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 2, 1); // 2 frequently changed diffuse + normal textures - using registers t1 and t2.
ranges[1].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0); // 1 frequently changed constant buffer.
ranges[2].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0); // 1 infrequently changed shadow texture - starting in register t0.
ranges[3].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SAMPLER, 2, 0); // 2 static samplers.
CD3DX12_ROOT_PARAMETER rootParameters[4];
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_PIXEL);
rootParameters[1].InitAsDescriptorTable(1, &ranges[1], D3D12_SHADER_VISIBILITY_ALL);
rootParameters[2].InitAsDescriptorTable(1, &ranges[2], D3D12_SHADER_VISIBILITY_PIXEL);
rootParameters[3].InitAsDescriptorTable(1, &ranges[3], D3D12_SHADER_VISIBILITY_PIXEL);
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state, which includes loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#ifdef _DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = D3DCOMPILE_OPTIMIZATION_LEVEL3;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
D3D12_INPUT_LAYOUT_DESC inputLayoutDesc;
inputLayoutDesc.pInputElementDescs = SampleAssets::StandardVertexDescription;
inputLayoutDesc.NumElements = _countof(SampleAssets::StandardVertexDescription);
CD3DX12_DEPTH_STENCIL_DESC depthStencilDesc(D3D12_DEFAULT);
depthStencilDesc.DepthEnable = true;
depthStencilDesc.DepthWriteMask = D3D12_DEPTH_WRITE_MASK_ALL;
depthStencilDesc.DepthFunc = D3D12_COMPARISON_FUNC_LESS_EQUAL;
depthStencilDesc.StencilEnable = FALSE;
// Describe and create the PSO for rendering the scene.
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = inputLayoutDesc;
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState = depthStencilDesc;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
// Alter the description and create the PSO for rendering
// the shadow map. The shadow map does not use a pixel
// shader or render targets.
psoDesc.PS.pShaderBytecode = 0;
psoDesc.PS.BytecodeLength = 0;
psoDesc.RTVFormats[0] = DXGI_FORMAT_UNKNOWN;
psoDesc.NumRenderTargets = 0;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineStateShadowMap)));
}
// Create temporary command list for initial GPU setup.
ComPtr<ID3D12GraphicsCommandList> commandList;
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocator.Get(), m_pipelineState.Get(), IID_PPV_ARGS(&commandList)));
// Create render target views (RTVs).
CD3DX12_CPU_DESCRIPTOR_HANDLE rtvHandle(m_rtvHeap->GetCPUDescriptorHandleForHeapStart());
for (UINT i = 0; i < FrameCount; i++)
{
ThrowIfFailed(m_swapChain->GetBuffer(i, IID_PPV_ARGS(&m_renderTargets[i])));
m_device->CreateRenderTargetView(m_renderTargets[i].Get(), nullptr, rtvHandle);
rtvHandle.Offset(1, m_rtvDescriptorSize);
}
// Create the depth stencil.
{
CD3DX12_RESOURCE_DESC shadowTextureDesc(
D3D12_RESOURCE_DIMENSION_TEXTURE2D,
0,
static_cast<UINT>(m_viewport.Width),
static_cast<UINT>(m_viewport.Height),
1,
1,
DXGI_FORMAT_D32_FLOAT,
1,
0,
D3D12_TEXTURE_LAYOUT_UNKNOWN,
D3D12_RESOURCE_FLAG_ALLOW_DEPTH_STENCIL | D3D12_RESOURCE_FLAG_DENY_SHADER_RESOURCE);
D3D12_CLEAR_VALUE clearValue; // Performance tip: Tell the runtime at resource creation the desired clear value.
clearValue.Format = DXGI_FORMAT_D32_FLOAT;
clearValue.DepthStencil.Depth = 1.0f;
clearValue.DepthStencil.Stencil = 0;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&shadowTextureDesc,
D3D12_RESOURCE_STATE_DEPTH_WRITE,
&clearValue,
IID_PPV_ARGS(&m_depthStencil)));
// Create the depth stencil view.
m_device->CreateDepthStencilView(m_depthStencil.Get(), nullptr, m_dsvHeap->GetCPUDescriptorHandleForHeapStart());
}
// Load scene assets.
UINT fileSize = 0;
UINT8* pAssetData;
ThrowIfFailed(ReadDataFromFile(GetAssetFullPath(SampleAssets::DataFileName).c_str(), &pAssetData, &fileSize));
// Create the vertex buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::VertexDataSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::VertexDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBufferUpload)));
// Copy data to the upload heap and then schedule a copy
// from the upload heap to the vertex buffer.
D3D12_SUBRESOURCE_DATA vertexData = {};
vertexData.pData = pAssetData + SampleAssets::VertexDataOffset;
vertexData.RowPitch = SampleAssets::VertexDataSize;
vertexData.SlicePitch = vertexData.RowPitch;
PIXBeginEvent(commandList.Get(), 0, L"Copy vertex buffer data to default resource...");
UpdateSubresources<1>(commandList.Get(), m_vertexBuffer.Get(), m_vertexBufferUpload.Get(), 0, 0, 1, &vertexData);
commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_vertexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER));
PIXEndEvent(commandList.Get());
}
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.SizeInBytes = SampleAssets::VertexDataSize;
m_vertexBufferView.StrideInBytes = SampleAssets::StandardVertexStride;
}
// Create the index buffer.
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::IndexDataSize),
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_indexBuffer)));
{
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(SampleAssets::IndexDataSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_indexBufferUpload)));
// Copy data to the upload heap and then schedule a copy
// from the upload heap to the index buffer.
D3D12_SUBRESOURCE_DATA indexData = {};
indexData.pData = pAssetData + SampleAssets::IndexDataOffset;
indexData.RowPitch = SampleAssets::IndexDataSize;
indexData.SlicePitch = indexData.RowPitch;
PIXBeginEvent(commandList.Get(), 0, L"Copy index buffer data to default resource...");
UpdateSubresources<1>(commandList.Get(), m_indexBuffer.Get(), m_indexBufferUpload.Get(), 0, 0, 1, &indexData);
commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_indexBuffer.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_INDEX_BUFFER));
PIXEndEvent(commandList.Get());
}
// Initialize the index buffer view.
m_indexBufferView.BufferLocation = m_indexBuffer->GetGPUVirtualAddress();
m_indexBufferView.SizeInBytes = SampleAssets::IndexDataSize;
m_indexBufferView.Format = SampleAssets::StandardIndexFormat;
}
// Create shader resources.
{
// Get the CBV SRV descriptor size for the current device.
const UINT cbvSrvDescriptorSize = m_device->GetDescriptorHandleIncrementSize(D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV);
// Get a handle to the start of the descriptor heap.
CD3DX12_CPU_DESCRIPTOR_HANDLE cbvSrvHandle(m_cbvSrvHeap->GetCPUDescriptorHandleForHeapStart());
{
// Describe and create 2 null SRVs. Null descriptors are needed in order
// to achieve the effect of an "unbound" resource.
D3D12_SHADER_RESOURCE_VIEW_DESC nullSrvDesc = {};
nullSrvDesc.ViewDimension = D3D12_SRV_DIMENSION_TEXTURE2D;
nullSrvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
nullSrvDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
nullSrvDesc.Texture2D.MipLevels = 1;
nullSrvDesc.Texture2D.MostDetailedMip = 0;
nullSrvDesc.Texture2D.ResourceMinLODClamp = 0.0f;
m_device->CreateShaderResourceView(nullptr, &nullSrvDesc, cbvSrvHandle);
cbvSrvHandle.Offset(cbvSrvDescriptorSize);
m_device->CreateShaderResourceView(nullptr, &nullSrvDesc, cbvSrvHandle);
cbvSrvHandle.Offset(cbvSrvDescriptorSize);
}
// Create each texture and SRV descriptor.
const UINT srvCount = _countof(SampleAssets::Textures);
PIXBeginEvent(commandList.Get(), 0, L"Copy diffuse and normal texture data to default resources...");
for (int i = 0; i < srvCount; i++)
{
// Describe and create a Texture2D.
const SampleAssets::TextureResource &tex = SampleAssets::Textures[i];
CD3DX12_RESOURCE_DESC texDesc(
D3D12_RESOURCE_DIMENSION_TEXTURE2D,
0,
tex.Width,
tex.Height,
1,
static_cast<UINT16>(tex.MipLevels),
tex.Format,
1,
0,
D3D12_TEXTURE_LAYOUT_UNKNOWN,
D3D12_RESOURCE_FLAG_NONE);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&texDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_textures[i])));
{
const UINT subresourceCount = texDesc.DepthOrArraySize * texDesc.MipLevels;
UINT64 uploadBufferSize = GetRequiredIntermediateSize(m_textures[i].Get(), 0, subresourceCount);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(uploadBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_textureUploads[i])));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the Texture2D.
D3D12_SUBRESOURCE_DATA textureData = {};
textureData.pData = pAssetData + tex.Data->Offset;
textureData.RowPitch = tex.Data->Pitch;
textureData.SlicePitch = tex.Data->Size;
UpdateSubresources(commandList.Get(), m_textures[i].Get(), m_textureUploads[i].Get(), 0, 0, subresourceCount, &textureData);
commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_textures[i].Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE));
}
// Describe and create an SRV.
D3D12_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
srvDesc.ViewDimension = D3D12_SRV_DIMENSION_TEXTURE2D;
srvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
srvDesc.Format = tex.Format;
srvDesc.Texture2D.MipLevels = tex.MipLevels;
srvDesc.Texture2D.MostDetailedMip = 0;
srvDesc.Texture2D.ResourceMinLODClamp = 0.0f;
m_device->CreateShaderResourceView(m_textures[i].Get(), &srvDesc, cbvSrvHandle);
// Move to the next descriptor slot.
cbvSrvHandle.Offset(cbvSrvDescriptorSize);
}
PIXEndEvent(commandList.Get());
}
free(pAssetData);
// Create the samplers.
{
// Get the sampler descriptor size for the current device.
const UINT samplerDescriptorSize = m_device->GetDescriptorHandleIncrementSize(D3D12_DESCRIPTOR_HEAP_TYPE_SAMPLER);
// Get a handle to the start of the descriptor heap.
CD3DX12_CPU_DESCRIPTOR_HANDLE samplerHandle(m_samplerHeap->GetCPUDescriptorHandleForHeapStart());
// Describe and create the wrapping sampler, which is used for
// sampling diffuse/normal maps.
D3D12_SAMPLER_DESC wrapSamplerDesc = {};
wrapSamplerDesc.Filter = D3D12_FILTER_MIN_MAG_MIP_LINEAR;
wrapSamplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
wrapSamplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
wrapSamplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_WRAP;
wrapSamplerDesc.MinLOD = 0;
wrapSamplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
wrapSamplerDesc.MipLODBias = 0.0f;
wrapSamplerDesc.MaxAnisotropy = 1;
wrapSamplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
wrapSamplerDesc.BorderColor[0] = wrapSamplerDesc.BorderColor[1] = wrapSamplerDesc.BorderColor[2] = wrapSamplerDesc.BorderColor[3] = 0;
m_device->CreateSampler(&wrapSamplerDesc, samplerHandle);
// Move the handle to the next slot in the descriptor heap.
samplerHandle.Offset(samplerDescriptorSize);
// Describe and create the point clamping sampler, which is
// used for the shadow map.
D3D12_SAMPLER_DESC clampSamplerDesc = {};
clampSamplerDesc.Filter = D3D12_FILTER_MIN_MAG_MIP_POINT;
clampSamplerDesc.AddressU = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
clampSamplerDesc.AddressV = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
clampSamplerDesc.AddressW = D3D12_TEXTURE_ADDRESS_MODE_CLAMP;
clampSamplerDesc.MipLODBias = 0.0f;
clampSamplerDesc.MaxAnisotropy = 1;
clampSamplerDesc.ComparisonFunc = D3D12_COMPARISON_FUNC_ALWAYS;
clampSamplerDesc.BorderColor[0] = clampSamplerDesc.BorderColor[1] = clampSamplerDesc.BorderColor[2] = clampSamplerDesc.BorderColor[3] = 0;
clampSamplerDesc.MinLOD = 0;
clampSamplerDesc.MaxLOD = D3D12_FLOAT32_MAX;
m_device->CreateSampler(&clampSamplerDesc, samplerHandle);
}
// Create lights.
for (int i = 0; i < NumLights; i++)
{
// Set up each of the light positions and directions (they all start
// in the same place).
m_lights[i].position = { 0.0f, 15.0f, -30.0f, 1.0f };
m_lights[i].direction = { 0.0, 0.0f, 1.0f, 0.0f };
m_lights[i].falloff = { 800.0f, 1.0f, 0.0f, 1.0f };
m_lights[i].color = { 0.7f, 0.7f, 0.7f, 1.0f };
XMVECTOR eye = XMLoadFloat4(&m_lights[i].position);
XMVECTOR at = XMVectorAdd(eye, XMLoadFloat4(&m_lights[i].direction));
XMVECTOR up = { 0, 1, 0 };
m_lightCameras[i].Set(eye, at, up);
}
// Close the command list and use it to execute the initial GPU setup.
ThrowIfFailed(commandList->Close());
ID3D12CommandList* ppCommandLists[] = { commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create frame resources.
for (int i = 0; i < FrameCount; i++)
{
m_frameResources[i] = new FrameResource(m_device.Get(), m_pipelineState.Get(), m_pipelineStateShadowMap.Get(), m_dsvHeap.Get(), m_cbvSrvHeap.Get(), &m_viewport, i);
m_frameResources[i]->WriteConstantBuffers(&m_viewport, &m_camera, m_lightCameras, m_lights);
}
m_currentFrameResourceIndex = 0;
m_pCurrentFrameResource = m_frameResources[m_currentFrameResourceIndex];
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValue, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValue++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
// Signal and increment the fence value.
const UINT64 fenceToWaitFor = m_fenceValue;
ThrowIfFailed(m_commandQueue->Signal(m_fence.Get(), fenceToWaitFor));
m_fenceValue++;
// Wait until the fence is completed.
ThrowIfFailed(m_fence->SetEventOnCompletion(fenceToWaitFor, m_fenceEvent));
WaitForSingleObject(m_fenceEvent, INFINITE);
}
}
// Load the sample assets.
void D3D12HelloTriangle::LoadAssets()
{
// Create an empty root signature.
{
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(0, nullptr, 0, nullptr, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#if defined(_DEBUG)
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocator.Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
// Command lists are created in the recording state, but there is nothing
// to record yet. The main loop expects it to be closed, so close it now.
ThrowIfFailed(m_commandList->Close());
// Create the vertex buffer.
{
// Define the geometry for a triangle.
Vertex triangleVertices[] =
{
{ { 0.0f, 0.25f * m_aspectRatio, 0.0f }, { 1.0f, 0.0f, 0.0f, 1.0f } },
{ { 0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 1.0f, 0.0f, 1.0f } },
{ { -0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 0.0f, 1.0f, 1.0f } }
};
const UINT vertexBufferSize = sizeof(triangleVertices);
// Note: using upload heaps to transfer static data like vert buffers is not
// recommended. Every time the GPU needs it, the upload heap will be marshalled
// over. Please read up on Default Heap usage. An upload heap is used here for
// code simplicity and because there are very few verts to actually transfer.
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
// Copy the triangle data to the vertex buffer.
UINT8* pVertexDataBegin;
CD3DX12_RANGE readRange(0, 0); // We do not intend to read from this resource on the CPU.
ThrowIfFailed(m_vertexBuffer->Map(0, &readRange, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, triangleVertices, sizeof(triangleVertices));
m_vertexBuffer->Unmap(0, nullptr);
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValue = 1;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForPreviousFrame();
}
}
/**
***************************************************************************************************
* DX12ImageRenderer::Init
*
* @brief
* Initialize all members needed by this rendering class.
*
* @return
* S_OK if successful.
***************************************************************************************************
*/
HRESULT DX12ImageRenderer::Init(const DX12ImageRendererConfig& config)
{
memcpy(&m_config, &config, sizeof(m_config));
m_srvUavCbDescriptorSize = m_config.pDevice->GetDescriptorHandleIncrementSize(D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV);
// Create command allocator
HRESULT result = m_config.pDevice->CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT, IID_PPV_ARGS(&m_pCmdAllocator));
// Create command list
if (result == S_OK)
{
result = m_config.pDevice->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_pCmdAllocator, nullptr, IID_PPV_ARGS(&m_pCmdList));
}
// Create descriptor heap for RTV
if (result == S_OK)
{
D3D12_DESCRIPTOR_HEAP_DESC rtvHeapDesc = {};
rtvHeapDesc.NumDescriptors = 1;
rtvHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_RTV;
rtvHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_NONE;
result = m_config.pDevice->CreateDescriptorHeap(&rtvHeapDesc, IID_PPV_ARGS(&m_pInternalRtvHeap));
}
// Create descriptor heap for SRV, UAV, and CBV
if (result == S_OK)
{
// Describe and create a shader resource view (SRV) heap for the texture.
D3D12_DESCRIPTOR_HEAP_DESC srvUavHeapDesc = {};
srvUavHeapDesc.NumDescriptors = RootParametersCount;
srvUavHeapDesc.Type = D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV;
srvUavHeapDesc.Flags = D3D12_DESCRIPTOR_HEAP_FLAG_SHADER_VISIBLE;
result = m_config.pDevice->CreateDescriptorHeap(&srvUavHeapDesc, IID_PPV_ARGS(&m_pSrvUavCbHeap));
}
// Create synchronization objects
if (result == S_OK)
{
result = m_config.pDevice->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_pFence));
m_fenceValue = 1;
// Create an event handle to use for frame synchronization
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
}
// Create root signature
if (result == S_OK)
{
CD3DX12_DESCRIPTOR_RANGE rangesGfx[RootParametersCount];
rangesGfx[RootParameterSRV].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
rangesGfx[RootParameterUAV].Init(D3D12_DESCRIPTOR_RANGE_TYPE_UAV, 1, 1);
rangesGfx[RootParameterCBV].Init(D3D12_DESCRIPTOR_RANGE_TYPE_CBV, 1, 0);
CD3DX12_ROOT_PARAMETER rootParamsGfx[RootParametersCount];
rootParamsGfx[RootParameterSRV].InitAsDescriptorTable(1, &rangesGfx[RootParameterSRV], D3D12_SHADER_VISIBILITY_PIXEL);
rootParamsGfx[RootParameterUAV].InitAsDescriptorTable(1, &rangesGfx[RootParameterUAV], D3D12_SHADER_VISIBILITY_PIXEL);
rootParamsGfx[RootParameterCBV].InitAsDescriptorTable(1, &rangesGfx[RootParameterCBV], D3D12_SHADER_VISIBILITY_ALL);
D3D12_STATIC_SAMPLER_DESC sampler = CD3DX12_STATIC_SAMPLER_DESC(0, D3D12_FILTER_MIN_MAG_MIP_LINEAR);
CD3DX12_ROOT_SIGNATURE_DESC rootSigDescGfx;
rootSigDescGfx.Init(3, rootParamsGfx, 1, &sampler, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ID3DBlob* pSigBlobGfx = nullptr;
ID3DBlob* pSigBlobGfxErr = nullptr;
result = D3D12SerializeRootSignature(&rootSigDescGfx, D3D_ROOT_SIGNATURE_VERSION_1, &pSigBlobGfx, &pSigBlobGfxErr);
result = m_config.pDevice->CreateRootSignature(0, pSigBlobGfx->GetBufferPointer(), pSigBlobGfx->GetBufferSize(), IID_PPV_ARGS(&m_pRootSignatureGraphics));
}
// Create pipeline state
if (result == S_OK)
{
const std::string shaderSrc =
#include "FsQuadToBuffer.hlsl"
;
size_t fileSize = strlen(shaderSrc.c_str());
ID3DBlob* pVertexShader = nullptr;
ID3DBlob* pVsError = nullptr;
result = D3DCompile(shaderSrc.c_str(), fileSize, NULL, NULL, NULL, "VsMain", "vs_5_0", 0, 0, &pVertexShader, &pVsError);
if (pVsError)
{
OutputDebugStringA((char*)pVsError->GetBufferPointer());
pVsError->Release();
}
ID3DBlob* pPixelShader = nullptr;
ID3DBlob* pPsError = nullptr;
result = D3DCompile(shaderSrc.c_str(), fileSize, NULL, NULL, NULL, "PsMain", "ps_5_0", 0, 0, &pPixelShader, &pPsError);
if (pPsError)
{
OutputDebugStringA((char*)pPsError->GetBufferPointer());
pPsError->Release();
}
if (result == S_OK)
{
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.VS = { reinterpret_cast<UINT8*>(pVertexShader->GetBufferPointer()), pVertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pPixelShader->GetBufferPointer()), pPixelShader->GetBufferSize() };
psoDesc.pRootSignature = m_pRootSignatureGraphics;
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
result = m_config.pDevice->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pPipelineStateGraphics));
}
}
// Create constant buffer
if (result == S_OK)
{
D3D12_HEAP_PROPERTIES heapProps = CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD);
D3D12_RESOURCE_DESC resDesc = CD3DX12_RESOURCE_DESC::Buffer(ConstBufSize);
result = m_config.pDevice->CreateCommittedResource(
&heapProps,
D3D12_HEAP_FLAG_NONE,
&resDesc,
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_pConstantBuffer));
}
if (result == S_OK)
{
// Command lists are created in the recording state, but there is nothing
// to record yet. The main loop expects it to be closed, so close it now.
result = m_pCmdList->Close();
}
return result;
}
// Load the sample assets.
void D3D12HelloTexture::LoadAssets()
{
// Create the root signature.
{
CD3DX12_DESCRIPTOR_RANGE ranges[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 1, 0);
CD3DX12_ROOT_PARAMETER rootParameters[1];
rootParameters[0].InitAsDescriptorTable(1, &ranges[0], D3D12_SHADER_VISIBILITY_PIXEL);
D3D12_STATIC_SAMPLER_DESC sampler = {};
sampler.Filter = D3D12_FILTER_MIN_MAG_MIP_POINT;
sampler.AddressU = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressV = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressW = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.MipLODBias = 0;
sampler.MaxAnisotropy = 0;
sampler.ComparisonFunc = D3D12_COMPARISON_FUNC_NEVER;
sampler.BorderColor = D3D12_STATIC_BORDER_COLOR_TRANSPARENT_BLACK;
sampler.MinLOD = 0.0f;
sampler.MaxLOD = D3D12_FLOAT32_MAX;
sampler.ShaderRegister = 0;
sampler.RegisterSpace = 0;
sampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_PIXEL;
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 1, &sampler, D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state, which includes compiling and loading shaders.
{
ComPtr<ID3DBlob> vertexShader;
ComPtr<ID3DBlob> pixelShader;
#ifdef _DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, nullptr));
ThrowIfFailed(D3DCompileFromFile(GetAssetFullPath(L"shaders.hlsl").c_str(), nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, nullptr));
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, 12, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineState)));
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocator.Get(), m_pipelineState.Get(), IID_PPV_ARGS(&m_commandList)));
// Create the vertex buffer.
{
// Define the geometry for a triangle.
Vertex triangleVertices[] =
{
{ { 0.0f, 0.25f * m_aspectRatio, 0.0f }, { 0.5f, 0.0f } },
{ { 0.25f, -0.25f * m_aspectRatio, 0.0f }, { 1.0f, 1.0f } },
{ { -0.25f, -0.25f * m_aspectRatio, 0.0f }, { 0.0f, 1.0f } }
};
const UINT vertexBufferSize = sizeof(triangleVertices);
// Note: using upload heaps to transfer static data like vert buffers is not
// recommended. Every time the GPU needs it, the upload heap will be marshalled
// over. Please read up on Default Heap usage. An upload heap is used here for
// code simplicity and because there are very few verts to actually transfer.
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
// Copy the triangle data to the vertex buffer.
UINT8* pVertexDataBegin;
ThrowIfFailed(m_vertexBuffer->Map(0, nullptr, reinterpret_cast<void**>(&pVertexDataBegin)));
memcpy(pVertexDataBegin, triangleVertices, sizeof(triangleVertices));
m_vertexBuffer->Unmap(0, nullptr);
// Initialize the vertex buffer view.
m_vertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_vertexBufferView.StrideInBytes = sizeof(Vertex);
m_vertexBufferView.SizeInBytes = vertexBufferSize;
}
// Create an upload heap to load the texture onto the GPU. ComPtr's are CPU objects
// but this heap needs to stay in scope until the GPU work is complete. We will
// synchronize with the GPU at the end of this method before the ComPtr is destroyed.
ComPtr<ID3D12Resource> textureUploadHeap;
// Create the texture.
{
// Describe and create a Texture2D.
D3D12_RESOURCE_DESC textureDesc = {};
textureDesc.MipLevels = 1;
textureDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
textureDesc.Width = TextureWidth;
textureDesc.Height = TextureHeight;
textureDesc.Flags = D3D12_RESOURCE_FLAG_NONE;
textureDesc.DepthOrArraySize = 1;
textureDesc.SampleDesc.Count = 1;
textureDesc.SampleDesc.Quality = 0;
textureDesc.Dimension = D3D12_RESOURCE_DIMENSION_TEXTURE2D;
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&textureDesc,
D3D12_RESOURCE_STATE_COPY_DEST,
nullptr,
IID_PPV_ARGS(&m_texture)));
const UINT64 uploadBufferSize = GetRequiredIntermediateSize(m_texture.Get(), 0, 1);
// Create the GPU upload buffer.
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(uploadBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&textureUploadHeap)));
// Copy data to the intermediate upload heap and then schedule a copy
// from the upload heap to the Texture2D.
std::vector<UINT8> texture = GenerateTextureData();
D3D12_SUBRESOURCE_DATA textureData = {};
textureData.pData = &texture[0];
textureData.RowPitch = TextureWidth * TexturePixelSize;
textureData.SlicePitch = textureData.RowPitch * TextureHeight;
UpdateSubresources(m_commandList.Get(), m_texture.Get(), textureUploadHeap.Get(), 0, 0, 1, &textureData);
m_commandList->ResourceBarrier(1, &CD3DX12_RESOURCE_BARRIER::Transition(m_texture.Get(), D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE));
// Describe and create a SRV for the texture.
D3D12_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
srvDesc.Shader4ComponentMapping = D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING;
srvDesc.Format = textureDesc.Format;
srvDesc.ViewDimension = D3D12_SRV_DIMENSION_TEXTURE2D;
srvDesc.Texture2D.MipLevels = 1;
m_device->CreateShaderResourceView(m_texture.Get(), &srvDesc, m_srvHeap->GetCPUDescriptorHandleForHeapStart());
}
// Close the command list and execute it to begin the initial GPU setup.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(0, D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValue = 1;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEventEx(nullptr, FALSE, FALSE, EVENT_ALL_ACCESS);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForPreviousFrame();
}
}
void Application::CreatePSO()
{
ComPtr<ID3DBlob> vertexShader, pixelShader;
#ifdef _DEBUG
// Enable better shader debugging with the graphics debugging tools.
UINT compileFlags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION;
#else
UINT compileFlags = 0;
#endif
ID3D10Blob* errorMessages;
if (FAILED(D3DCompileFromFile(L"shaders.hlsl", nullptr, nullptr, "VSMain", "vs_5_0", compileFlags, 0, &vertexShader, &errorMessages)))
{
if (errorMessages)
{
std::cout << static_cast<char*>(errorMessages->GetBufferPointer()) << std::endl;
errorMessages->Release();
}
if (errorMessages)
errorMessages->Release();
CRITICAL_ERROR("Failed to compile vertex shader.");
}
if (FAILED(D3DCompileFromFile(L"shaders.hlsl", nullptr, nullptr, "PSMain", "ps_5_0", compileFlags, 0, &pixelShader, &errorMessages)))
{
if (errorMessages)
{
std::cout << static_cast<char*>(errorMessages->GetBufferPointer()) << std::endl;
errorMessages->Release();
}
if (errorMessages)
errorMessages->Release();
CRITICAL_ERROR("Failed to compile pixel shader.");
}
// Define the vertex input layout.
D3D12_INPUT_ELEMENT_DESC inputElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, 8, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 }
};
// Describe and create the graphics pipeline state object (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.pRootSignature = rootSignature.Get();
psoDesc.VS = { reinterpret_cast<UINT8*>(vertexShader->GetBufferPointer()), vertexShader->GetBufferSize() };
psoDesc.PS = { reinterpret_cast<UINT8*>(pixelShader->GetBufferPointer()), pixelShader->GetBufferSize() };
psoDesc.DS;
psoDesc.HS;
psoDesc.GS;
psoDesc.StreamOutput;
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.SampleMask = UINT_MAX;
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.RasterizerState.CullMode = D3D12_CULL_MODE_NONE;
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.InputLayout = { inputElementDescs, _countof(inputElementDescs) };
psoDesc.IBStripCutValue = D3D12_INDEX_BUFFER_STRIP_CUT_VALUE_DISABLED;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM;
psoDesc.DSVFormat = DXGI_FORMAT_D32_FLOAT;
psoDesc.SampleDesc.Count = 1;
psoDesc.CachedPSO;
if(FAILED(device->GetD3D12Device()->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&pso))))
CRITICAL_ERROR("Failed to PSO.");
}
// Load the sample assets.
void D3D12HDR::LoadAssets()
{
// Create a root signature containing root constants for brightness information
// and the desired output curve as well as a SRV descriptor table pointing to the
// intermediate render targets.
{
CD3DX12_DESCRIPTOR_RANGE ranges[1];
ranges[0].Init(D3D12_DESCRIPTOR_RANGE_TYPE_SRV, 2, 0);
CD3DX12_ROOT_PARAMETER rootParameters[2];
rootParameters[0].InitAsConstants(4, 0);
rootParameters[1].InitAsDescriptorTable(1, &ranges[0]);
D3D12_STATIC_SAMPLER_DESC sampler = {};
sampler.Filter = D3D12_FILTER_MIN_MAG_MIP_POINT;
sampler.AddressU = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressV = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.AddressW = D3D12_TEXTURE_ADDRESS_MODE_BORDER;
sampler.ComparisonFunc = D3D12_COMPARISON_FUNC_NEVER;
sampler.MaxLOD = D3D12_FLOAT32_MAX;
sampler.ShaderVisibility = D3D12_SHADER_VISIBILITY_ALL;
// Allow input layout and deny uneccessary access to certain pipeline stages.
D3D12_ROOT_SIGNATURE_FLAGS rootSignatureFlags =
D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT |
D3D12_ROOT_SIGNATURE_FLAG_DENY_HULL_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_DOMAIN_SHADER_ROOT_ACCESS |
D3D12_ROOT_SIGNATURE_FLAG_DENY_GEOMETRY_SHADER_ROOT_ACCESS;
CD3DX12_ROOT_SIGNATURE_DESC rootSignatureDesc;
rootSignatureDesc.Init(_countof(rootParameters), rootParameters, 1, &sampler, rootSignatureFlags);
ComPtr<ID3DBlob> signature;
ComPtr<ID3DBlob> error;
ThrowIfFailed(D3D12SerializeRootSignature(&rootSignatureDesc, D3D_ROOT_SIGNATURE_VERSION_1, &signature, &error));
ThrowIfFailed(m_device->CreateRootSignature(0, signature->GetBufferPointer(), signature->GetBufferSize(), IID_PPV_ARGS(&m_rootSignature)));
}
// Create the pipeline state objects for the different views and render target formats
// as well as the intermediate blend step.
{
// Create the pipeline state for the scene geometry.
D3D12_INPUT_ELEMENT_DESC gradientElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
// Describe and create the graphics pipeline state objects (PSO).
D3D12_GRAPHICS_PIPELINE_STATE_DESC psoDesc = {};
psoDesc.InputLayout = { gradientElementDescs, _countof(gradientElementDescs) };
psoDesc.pRootSignature = m_rootSignature.Get();
psoDesc.VS = CD3DX12_SHADER_BYTECODE(g_gradientVS, sizeof(g_gradientVS));
psoDesc.PS = CD3DX12_SHADER_BYTECODE(g_gradientPS, sizeof(g_gradientPS));
psoDesc.RasterizerState = CD3DX12_RASTERIZER_DESC(D3D12_DEFAULT);
psoDesc.BlendState = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
psoDesc.DepthStencilState.DepthEnable = FALSE;
psoDesc.DepthStencilState.StencilEnable = FALSE;
psoDesc.SampleMask = UINT_MAX;
psoDesc.PrimitiveTopologyType = D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE;
psoDesc.NumRenderTargets = 1;
psoDesc.RTVFormats[0] = m_intermediateRenderTargetFormat;
psoDesc.SampleDesc.Count = 1;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineStates[GradientPSO])));
// Create pipeline state for the color space triangles.
D3D12_INPUT_ELEMENT_DESC colorElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
psoDesc.InputLayout = { colorElementDescs, _countof(colorElementDescs) };
psoDesc.VS = CD3DX12_SHADER_BYTECODE(g_paletteVS, sizeof(g_paletteVS));
psoDesc.PS = CD3DX12_SHADER_BYTECODE(g_palettePS, sizeof(g_palettePS));
psoDesc.RTVFormats[0] = m_intermediateRenderTargetFormat;
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineStates[PalettePSO])));
// Create pipeline states for the final blend step.
// There will be one for each swap chain format the sample supports.
D3D12_INPUT_ELEMENT_DESC quadElementDescs[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
{ "TEXCOORD", 0, DXGI_FORMAT_R32G32_FLOAT, 0, D3D12_APPEND_ALIGNED_ELEMENT, D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, 0 },
};
psoDesc.InputLayout = { quadElementDescs, _countof(quadElementDescs) };
psoDesc.VS = CD3DX12_SHADER_BYTECODE(g_presentVS, sizeof(g_presentVS));
psoDesc.PS = CD3DX12_SHADER_BYTECODE(g_presentPS, sizeof(g_presentPS));
psoDesc.RTVFormats[0] = m_swapChainFormats[_8];
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineStates[Present8bitPSO])));
psoDesc.RTVFormats[0] = m_swapChainFormats[_10];
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineStates[Present10bitPSO])));
psoDesc.RTVFormats[0] = m_swapChainFormats[_16];
ThrowIfFailed(m_device->CreateGraphicsPipelineState(&psoDesc, IID_PPV_ARGS(&m_pipelineStates[Present16bitPSO])));
}
// Create the command list.
ThrowIfFailed(m_device->CreateCommandList(0, D3D12_COMMAND_LIST_TYPE_DIRECT, m_commandAllocators[m_frameIndex].Get(), nullptr, IID_PPV_ARGS(&m_commandList)));
// Create the vertex buffer.
{
// Create geometry for the different sections of the render target.
GradientVertex gradientVertices[] =
{
// Upper strip. SDR Gradient from [0,1].
{ { -1.0f, 0.45f, 0.0f }, { 0.0f, 0.0f, 0.0f } },
{ { -1.0f, 0.55f, 0.0f }, { 0.0f, 0.0f, 0.0f } },
{ { 0.0f, 0.45f, 0.0f }, { 1.0f, 1.0f, 1.0f } },
{ { 0.0f, 0.55f, 0.0f }, { 1.0f, 1.0f, 1.0f } },
// Lower strip. HDR Gradient from [0,9]. Perceptually, 9.0 is about 3 times as bright as 1.0. (See gradientPS.hlsl.)
{ { -1.0f, -0.55f, 0.0f }, { 0.0f, 0.0f, 0.0f } },
{ { -1.0f, -0.45f, 0.0f }, { 0.0f, 0.0f, 0.0f } },
{ { 0.0f, -0.55f, 0.0f }, { 3.0f, 3.0f, 3.0f } },
{ { 0.0f, -0.45f, 0.0f }, { 3.0f, 3.0f, 3.0f } },
};
// The vertices for the color space triangles are dependent on the size of the
// render target and will not be loaded at this time. We'll leave a gap in the
// buffer that will be filled in later.
const UINT triangleVerticesSize = sizeof(TrianglesVertex) * TrianglesVertexCount;
m_updateVertexBuffer = true;
PresentVertex presentVertices[] =
{
// 1 triangle that fills the entire render target.
{ { -1.0f, -3.0f, 0.0f }, { 0.0f, 2.0f } }, // Bottom left
{ { -1.0f, 1.0f, 0.0f }, { 0.0f, 0.0f } }, // Top left
{ { 3.0f, 1.0f, 0.0f }, { 2.0f, 0.0f } }, // Top right
};
const UINT vertexBufferSize = sizeof(gradientVertices) + triangleVerticesSize + sizeof(presentVertices);
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_DEFAULT),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_VERTEX_AND_CONSTANT_BUFFER,
nullptr,
IID_PPV_ARGS(&m_vertexBuffer)));
ThrowIfFailed(m_device->CreateCommittedResource(
&CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD),
D3D12_HEAP_FLAG_NONE,
&CD3DX12_RESOURCE_DESC::Buffer(vertexBufferSize),
D3D12_RESOURCE_STATE_GENERIC_READ,
nullptr,
IID_PPV_ARGS(&m_vertexBufferUpload)));
// Copy data to the intermediate upload heap. It will be uploaded to the
// DEFAULT buffer when the color space triangle vertices are updated.
UINT8* mappedUploadHeap = nullptr;
ThrowIfFailed(m_vertexBufferUpload->Map(0, &CD3DX12_RANGE(0, 0), reinterpret_cast<void**>(&mappedUploadHeap)));
memcpy(mappedUploadHeap, gradientVertices, sizeof(gradientVertices));
mappedUploadHeap += sizeof(gradientVertices) + triangleVerticesSize;
memcpy(mappedUploadHeap, presentVertices, sizeof(presentVertices));
m_vertexBufferUpload->Unmap(0, &CD3DX12_RANGE(0, 0));
// Initialize the vertex buffer views.
m_gradientVertexBufferView.BufferLocation = m_vertexBuffer->GetGPUVirtualAddress();
m_gradientVertexBufferView.StrideInBytes = sizeof(GradientVertex);
m_gradientVertexBufferView.SizeInBytes = sizeof(gradientVertices);
m_trianglesVertexBufferView.BufferLocation = m_gradientVertexBufferView.BufferLocation + m_gradientVertexBufferView.SizeInBytes;
m_trianglesVertexBufferView.StrideInBytes = sizeof(TrianglesVertex);
m_trianglesVertexBufferView.SizeInBytes = triangleVerticesSize;
m_presentVertexBufferView.BufferLocation = m_trianglesVertexBufferView.BufferLocation + m_trianglesVertexBufferView.SizeInBytes;
m_presentVertexBufferView.StrideInBytes = sizeof(PresentVertex);
m_presentVertexBufferView.SizeInBytes = sizeof(presentVertices);
}
LoadSizeDependentResources();
// Close the command list and execute it to begin the vertex buffer copy into
// the default heap.
ThrowIfFailed(m_commandList->Close());
ID3D12CommandList* ppCommandLists[] = { m_commandList.Get() };
m_commandQueue->ExecuteCommandLists(_countof(ppCommandLists), ppCommandLists);
// Create synchronization objects and wait until assets have been uploaded to the GPU.
{
ThrowIfFailed(m_device->CreateFence(m_fenceValues[m_frameIndex], D3D12_FENCE_FLAG_NONE, IID_PPV_ARGS(&m_fence)));
m_fenceValues[m_frameIndex]++;
// Create an event handle to use for frame synchronization.
m_fenceEvent = CreateEvent(nullptr, FALSE, FALSE, nullptr);
if (m_fenceEvent == nullptr)
{
ThrowIfFailed(HRESULT_FROM_WIN32(GetLastError()));
}
// Wait for the command list to execute; we are reusing the same command
// list in our main loop but for now, we just want to wait for setup to
// complete before continuing.
WaitForGpu();
}
}
text_2d::text_2d(device const& dev, std::size_t max_str_size, std::size_t max_num_str, std::size_t max_buffered_frame, bool enable_filtering)
: max_word_sz(max_str_size), max_num_word(max_num_str)
{
assert(max_str_size > 0 && max_num_str > 0 && max_buffered_frame > 0);
if (dev.Get() == nullptr)
{
throw error{ "null device" };
}
words.resize(max_buffered_frame);
size_t const num_cb = max_num_str * max_buffered_frame;
size_t const num_tex = 1; // 1 text texture
srd_heap = dx12u::descriptor_heap{ dev.Get(), D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV, num_cb + num_tex };
auto cb_sz = get_cb_size(max_str_size);
auto cb_total_sz = get_cb_size(max_str_size) * num_cb;
// alloc and upload texture
{
auto font_data = load_font_texture();
gu::mip_lvl_texture_view mip_view{};
mip_view.width = global::font_texture_width;
mip_view.height = global::font_texture_height;
mip_view.pitch = mip_view.width;
mip_view.slice_sz = font_data.size();
mip_view.data = font_data.data();
gu::texture_view tex_view{};
tex_view.bpp = 1;
tex_view.num_lvl = 1;
tex_view.mip = &mip_view;
dx12u::cmd_queue q{ dev };
dx12u::cmd_allocator allocator{ dev };
dx12u::gfx_cmd_list cl = allocator.alloc();
auto tex = dx12u::make_texture(dev, cl, tex_view, false, false, DXGI_FORMAT_R8_UNORM);
cl->Close();
q.push(cl);
q.sync();
font_texture = tex.texture_rsrc;
dev->CreateShaderResourceView(font_texture.Get(), &tex.srv_desc, srd_heap.get_cpu_handle(0));
}
// alloc cb_mem
{
auto upload_heap = CD3DX12_HEAP_PROPERTIES(D3D12_HEAP_TYPE_UPLOAD);
auto buffer_desc = CD3DX12_RESOURCE_DESC::Buffer(cb_total_sz);
auto r = dev->CreateCommittedResource(&upload_heap, D3D12_HEAP_FLAG_NONE,
&buffer_desc, D3D12_RESOURCE_STATE_GENERIC_READ, nullptr, IID_PPV_ARGS(&cb_mem));
dx12u::throw_if_error(r);
for (size_t i = 0; i < num_cb; ++i)
{
// sh_mask_cb views
D3D12_CONSTANT_BUFFER_VIEW_DESC cbv_desc = {};
cbv_desc.SizeInBytes = static_cast<uint32_t>(cb_sz);
cbv_desc.BufferLocation = cb_mem->GetGPUVirtualAddress() + i * cb_sz;
dev->CreateConstantBufferView(&cbv_desc, srd_heap.get_cpu_handle(num_tex + i));
}
r = cb_mem->Map(0, nullptr, reinterpret_cast<void**>(&cb_ptr));
dx12u::throw_if_error(r);
assert(!!cb_ptr);
memset(cb_ptr, 0, cb_total_sz);
}
// pso
{
// root signature
auto filter = enable_filtering ? D3D12_FILTER_MIN_MAG_MIP_LINEAR : D3D12_FILTER_MIN_MAG_MIP_POINT;
dx12u::descriptor_sig_list descriptor_table0{};
dx12u::descriptor_sig_list descriptor_table1{};
descriptor_table0.append(dx12u::descriptor_sig{ dx12u::descriptor_type::srv, 0, 0, dx12u::shader_mask::ps });
descriptor_table1.append(dx12u::descriptor_sig{ dx12u::descriptor_type::cbv, 0, 0, dx12u::shader_mask::vs });
dx12u::descriptor_sig_list dsl{};
dsl.append(descriptor_table0); // TODO move to root once the number of sgpr is definite
dsl.append(descriptor_table1);
dx12u::static_sampler_list sl =
{
dx12u::make_default_static_sampler(0, filter),
};
rs = dx12u::make_root_signature(dev.Get(), dsl, sl);
// shader preprocessor definitions
auto const texture_char_w = static_cast<double>(global::font_texture_char_w) / static_cast<double>(global::font_texture_width);
using macro = std::pair<std::string, std::string>;
macro max_str_macro = macro{ "MAX_STR_SZ", std::to_string(max_str_size) };
macro char_w_macro = macro{ "CHAR_WIDTH", std::to_string(texture_char_w) };
D3D_SHADER_MACRO def[] =
{
max_str_macro.first.c_str(), max_str_macro.second.c_str(),
char_w_macro.first.c_str(), char_w_macro.second.c_str(),
nullptr, nullptr
};
// shader blobs
auto vs = dx12u::compile_from_file("../../framework/d3d12/shader/ui/text2d.hlsl", def, "vs_main", "vs_5_0");
auto ps = dx12u::compile_from_file("../../framework/d3d12/shader/ui/text2d.hlsl", def, "ps_main", "ps_5_0");
// pso
pso = dx12u::pipeline_state_object{ dev, rs, vs, ps };
D3D12_BLEND_DESC bd = CD3DX12_BLEND_DESC(D3D12_DEFAULT);
bd.RenderTarget[0].BlendEnable = true;
bd.RenderTarget[0].SrcBlend = D3D12_BLEND_SRC_ALPHA;
bd.RenderTarget[0].DestBlend = D3D12_BLEND_INV_SRC_ALPHA;
pso.set_blend_states(bd);
pso.set_depth_test(false);
pso.commit();
}
}
