#![cfg(test)] use std::ffi::CStr; use windows::{ core::*, Win32::Foundation::*, Win32::Graphics::Direct3D::Fxc::*, Win32::Graphics::Direct3D::*, Win32::Graphics::Direct3D12::*, Win32::Graphics::Dxgi::Common::*, Win32::Graphics::Dxgi::*, Win32::System::LibraryLoader::*, Win32::System::Threading::*, Win32::UI::WindowsAndMessaging::*, }; mod descriptor_heap; static SHADER: &[u8] = b"struct PSInput { float4 position : SV_POSITION; float4 color : COLOR; }; PSInput VSMain(float4 position : POSITION, float4 color : COLOR) { PSInput result; result.position = position; result.color = color; return result; } float4 PSMain(PSInput input) : SV_TARGET { return input.color; }\0"; use std::mem::transmute; use std::path::Path; pub trait DXSample { fn new(filter: impl AsRef, command_line: &SampleCommandLine) -> Result where Self: Sized; fn bind_to_window(&mut self, hwnd: &HWND) -> Result<()>; fn update(&mut self) {} fn render(&mut self) {} fn on_key_up(&mut self, _key: u8) {} fn on_key_down(&mut self, _key: u8) {} fn title(&self) -> String { "DXSample".into() } fn window_size(&self) -> (i32, i32) { (600, 800) } } #[derive(Clone)] pub struct SampleCommandLine { pub use_warp_device: bool, } fn run_sample(mut sample: S) -> Result<()> where S: DXSample, { let instance = unsafe { GetModuleHandleA(None)? }; let wc = WNDCLASSEXA { cbSize: std::mem::size_of::() as u32, style: CS_HREDRAW | CS_VREDRAW, lpfnWndProc: Some(wndproc::), hInstance: HINSTANCE::from(instance), hCursor: unsafe { LoadCursorW(None, IDC_ARROW)? }, lpszClassName: s!("RustWindowClass"), ..Default::default() }; let size = sample.window_size(); let atom = unsafe { RegisterClassExA(&wc) }; debug_assert_ne!(atom, 0); let mut window_rect = RECT { left: 0, top: 0, right: size.0, bottom: size.1, }; unsafe { AdjustWindowRect(&mut window_rect, WS_OVERLAPPEDWINDOW, false)? }; let mut title = sample.title(); title.push('\0'); let hwnd = unsafe { CreateWindowExA( WINDOW_EX_STYLE::default(), s!("RustWindowClass"), PCSTR(title.as_ptr()), WS_OVERLAPPEDWINDOW, CW_USEDEFAULT, CW_USEDEFAULT, window_rect.right - window_rect.left, window_rect.bottom - window_rect.top, None, // no parent window None, // no menus instance, Some(&mut sample as *mut _ as _), ) }; let hwnd = hwnd?; sample.bind_to_window(&hwnd).unwrap(); unsafe { ShowWindow(hwnd, SW_SHOW) }; loop { let mut message = MSG::default(); if unsafe { PeekMessageA(&mut message, None, 0, 0, PM_REMOVE) }.into() { unsafe { TranslateMessage(&message); DispatchMessageA(&message); } if message.message == WM_QUIT { break; } } } Ok(()) } fn sample_wndproc(sample: &mut S, message: u32, wparam: WPARAM) -> bool { match message { WM_KEYDOWN => { sample.on_key_down(wparam.0 as u8); true } WM_KEYUP => { sample.on_key_up(wparam.0 as u8); true } WM_PAINT => { sample.update(); sample.render(); true } _ => false, } } extern "system" fn wndproc( window: HWND, message: u32, wparam: WPARAM, lparam: LPARAM, ) -> LRESULT { match message { WM_CREATE => { unsafe { let create_struct: &CREATESTRUCTA = transmute(lparam); SetWindowLongPtrA(window, GWLP_USERDATA, create_struct.lpCreateParams as _); } LRESULT::default() } WM_DESTROY => { unsafe { PostQuitMessage(0) }; LRESULT::default() } _ => { let user_data = unsafe { GetWindowLongPtrA(window, GWLP_USERDATA) }; let sample = std::ptr::NonNull::::new(user_data as _); let handled = sample.map_or(false, |mut s| { sample_wndproc(unsafe { s.as_mut() }, message, wparam) }); if handled { LRESULT::default() } else { unsafe { DefWindowProcA(window, message, wparam, lparam) } } } } } fn get_hardware_adapter(factory: &IDXGIFactory4) -> Result { for i in 0.. { let adapter = unsafe { factory.EnumAdapters1(i)? }; let mut desc = unsafe { adapter.GetDesc1()? }; if (desc.Flags & DXGI_ADAPTER_FLAG_SOFTWARE.0 as u32) != DXGI_ADAPTER_FLAG_NONE.0 as u32 { // Don't select the Basic Render Driver adapter. If you want a // software adapter, pass in "/warp" on the command line. continue; } // Check to see whether the adapter supports Direct3D 12, but don't // create the actual device yet. if unsafe { D3D12CreateDevice( &adapter, D3D_FEATURE_LEVEL_11_0, std::ptr::null_mut::>(), ) } .is_ok() { return Ok(adapter); } } unreachable!() } unsafe extern "system" fn debug_log( category: D3D12_MESSAGE_CATEGORY, severity: D3D12_MESSAGE_SEVERITY, _id: D3D12_MESSAGE_ID, pdescription: ::windows::core::PCSTR, _pcontext: *mut ::core::ffi::c_void, ) { unsafe { let desc = CStr::from_ptr(pdescription.as_ptr().cast()); eprintln!("[{severity:?}-{category:?}] {desc:?}") } } pub mod d3d12_hello_triangle { use super::*; use crate::hello_triangle::descriptor_heap::CpuStagingHeap; use d3d12_descriptor_heap::D3D12DescriptorHeap; use librashader_common::{Size, Viewport}; use librashader_runtime_d3d12::{D3D12InputImage, D3D12OutputView, FilterChainD3D12}; use std::mem::ManuallyDrop; use std::ops::Deref; use std::path::Path; const FRAME_COUNT: u32 = 2; pub struct Sample { dxgi_factory: IDXGIFactory4, device: ID3D12Device, resources: Option, pub filter: FilterChainD3D12, framecount: usize, } struct Resources { command_queue: ID3D12CommandQueue, swap_chain: IDXGISwapChain3, frame_index: u32, render_targets: [ID3D12Resource; FRAME_COUNT as usize], rtv_heap: ID3D12DescriptorHeap, rtv_descriptor_size: usize, viewport: D3D12_VIEWPORT, scissor_rect: RECT, command_allocator: ID3D12CommandAllocator, root_signature: ID3D12RootSignature, pso: ID3D12PipelineState, command_list: ID3D12GraphicsCommandList, framebuffer: ID3D12Resource, // we need to keep this around to keep the reference alive, even though // nothing reads from it #[allow(dead_code)] vertex_buffer: ID3D12Resource, vbv: D3D12_VERTEX_BUFFER_VIEW, fence: ID3D12Fence, fence_value: u64, fence_event: HANDLE, frambuffer_heap: D3D12DescriptorHeap, } impl DXSample for Sample { fn new(filter: impl AsRef, command_line: &SampleCommandLine) -> Result { let (dxgi_factory, device) = create_device(command_line)?; // if let Ok(queue) = device.cast::() { unsafe { queue .RegisterMessageCallback( Some(debug_log), D3D12_MESSAGE_CALLBACK_FLAG_NONE, std::ptr::null_mut(), &mut 0, ) .expect("could not register message callback"); } } let filter = unsafe { FilterChainD3D12::load_from_path( filter, &device, Some( &librashader_runtime_d3d12::options::FilterChainOptionsD3D12 { disable_cache: true, force_hlsl_pipeline: false, force_no_mipmaps: false, ..Default::default() }, ), ) .unwrap() }; Ok(Sample { dxgi_factory, device, resources: None, filter, framecount: 0, }) } fn bind_to_window(&mut self, hwnd: &HWND) -> Result<()> { let command_queue: ID3D12CommandQueue = unsafe { self.device.CreateCommandQueue(&D3D12_COMMAND_QUEUE_DESC { Type: D3D12_COMMAND_LIST_TYPE_DIRECT, ..Default::default() })? }; unsafe { command_queue.SetName(w!("MainLoopQueue"))?; } let (width, height) = self.window_size(); let swap_chain_desc = DXGI_SWAP_CHAIN_DESC1 { BufferCount: FRAME_COUNT, Width: width as u32, Height: height as u32, Format: DXGI_FORMAT_R8G8B8A8_UNORM, BufferUsage: DXGI_USAGE_RENDER_TARGET_OUTPUT, SwapEffect: DXGI_SWAP_EFFECT_FLIP_DISCARD, SampleDesc: DXGI_SAMPLE_DESC { Count: 1, ..Default::default() }, ..Default::default() }; let swap_chain: IDXGISwapChain3 = unsafe { self.dxgi_factory.CreateSwapChainForHwnd( &command_queue, *hwnd, &swap_chain_desc, None, None, )? } .cast()?; // This sample does not support fullscreen transitions unsafe { self.dxgi_factory .MakeWindowAssociation(*hwnd, DXGI_MWA_NO_ALT_ENTER)?; } let frame_index = unsafe { swap_chain.GetCurrentBackBufferIndex() }; let rtv_heap: ID3D12DescriptorHeap = unsafe { self.device .CreateDescriptorHeap(&D3D12_DESCRIPTOR_HEAP_DESC { NumDescriptors: FRAME_COUNT + 1, Type: D3D12_DESCRIPTOR_HEAP_TYPE_RTV, ..Default::default() }) }?; let rtv_descriptor_size = unsafe { self.device .GetDescriptorHandleIncrementSize(D3D12_DESCRIPTOR_HEAP_TYPE_RTV) } as usize; let rtv_handle = unsafe { rtv_heap.GetCPUDescriptorHandleForHeapStart() }; let render_targets: [ID3D12Resource; FRAME_COUNT as usize] = array_init::try_array_init(|i: usize| -> Result { let render_target: ID3D12Resource = unsafe { swap_chain.GetBuffer(i as u32) }?; unsafe { self.device.CreateRenderTargetView( &render_target, None, D3D12_CPU_DESCRIPTOR_HANDLE { ptr: rtv_handle.ptr + i * rtv_descriptor_size, }, ) }; Ok(render_target) })?; let framebuffer: ID3D12Resource = unsafe { let render_target: ID3D12Resource = swap_chain.GetBuffer(0)?; let mut desc = render_target.GetDesc(); let mut heapprops = D3D12_HEAP_PROPERTIES::default(); let mut heappflags = D3D12_HEAP_FLAGS::default(); render_target.GetHeapProperties(Some(&mut heapprops), Some(&mut heappflags))?; desc.Flags &= !D3D12_RESOURCE_FLAG_DENY_SHADER_RESOURCE; let mut fb = None; self.device.CreateCommittedResource( &heapprops, D3D12_HEAP_FLAG_NONE, &desc, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE, None, &mut fb, )?; fb.unwrap() }; unsafe { framebuffer.SetName(w!("framebuffer"))?; } let viewport = D3D12_VIEWPORT { TopLeftX: 0.0, TopLeftY: 0.0, Width: width as f32, Height: height as f32, MinDepth: D3D12_MIN_DEPTH, MaxDepth: D3D12_MAX_DEPTH, }; let scissor_rect = RECT { left: 0, top: 0, right: width, bottom: height, }; let command_allocator = unsafe { self.device .CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT) }?; let root_signature = create_root_signature(&self.device)?; let pso = create_pipeline_state(&self.device, &root_signature)?; let command_list: ID3D12GraphicsCommandList = unsafe { self.device.CreateCommandList( 0, D3D12_COMMAND_LIST_TYPE_DIRECT, &command_allocator, &pso, ) }?; unsafe { command_list.Close()?; }; let aspect_ratio = width as f32 / height as f32; let (vertex_buffer, vbv) = create_vertex_buffer(&self.device, aspect_ratio)?; let fence = unsafe { self.device.CreateFence(0, D3D12_FENCE_FLAG_NONE) }?; let fence_value = 1; let fence_event = unsafe { CreateEventA(None, false, false, None)? }; self.resources = Some(Resources { command_queue, swap_chain, frame_index, render_targets, rtv_heap, rtv_descriptor_size, viewport, scissor_rect, command_allocator, root_signature, pso, command_list, framebuffer, vertex_buffer, vbv, fence, fence_value, fence_event, frambuffer_heap: unsafe { D3D12DescriptorHeap::new(&self.device, 1024).unwrap() }, }); Ok(()) } fn title(&self) -> String { "librashader DirectX 12".into() } fn window_size(&self) -> (i32, i32) { (800, 600) } fn render(&mut self) { if let Some(resources) = &mut self.resources { let srv = resources.frambuffer_heap.allocate_descriptor().unwrap(); unsafe { self.device.CreateShaderResourceView( &resources.framebuffer, Some(&D3D12_SHADER_RESOURCE_VIEW_DESC { Format: DXGI_FORMAT_R8G8B8A8_UNORM, ViewDimension: D3D12_SRV_DIMENSION_TEXTURE2D, Shader4ComponentMapping: D3D12_DEFAULT_SHADER_4_COMPONENT_MAPPING, Anonymous: D3D12_SHADER_RESOURCE_VIEW_DESC_0 { Texture2D: D3D12_TEX2D_SRV { MipLevels: u32::MAX, ..Default::default() }, }, }), *srv.as_ref(), ) } populate_command_list(resources, &mut self.filter, self.framecount, *srv.as_ref()) .unwrap(); // Execute the command list. let command_list: Option = resources.command_list.cast().ok(); unsafe { resources.command_queue.ExecuteCommandLists(&[command_list]) }; // Present the frame. unsafe { resources.swap_chain.Present(1, DXGI_PRESENT::default()) } .ok() .unwrap(); wait_for_previous_frame(resources); self.framecount += 1; } } } fn populate_command_list( resources: &mut Resources, filter: &mut FilterChainD3D12, frame_count: usize, framebuffer: D3D12_CPU_DESCRIPTOR_HANDLE, ) -> Result<()> { // Command list allocators can only be reset when the associated // command lists have finished execution on the GPU; apps should use // fences to determine GPU execution progress. unsafe { resources.command_allocator.Reset()?; } let command_list = &resources.command_list; // However, when ExecuteCommandList() is called on a particular // command list, that command list can then be reset at any time and // must be before re-recording. unsafe { command_list.Reset(&resources.command_allocator, &resources.pso)?; } // Set necessary state. unsafe { command_list.SetGraphicsRootSignature(&resources.root_signature); command_list.RSSetViewports(&[resources.viewport]); command_list.RSSetScissorRects(&[resources.scissor_rect]); } // Indicate that the back buffer will be used as a render target. let barrier = transition_barrier( &resources.render_targets[resources.frame_index as usize], D3D12_RESOURCE_STATE_PRESENT, D3D12_RESOURCE_STATE_RENDER_TARGET, ); unsafe { command_list.ResourceBarrier(&[barrier]) }; let rtv_handle = D3D12_CPU_DESCRIPTOR_HANDLE { ptr: unsafe { resources.rtv_heap.GetCPUDescriptorHandleForHeapStart() }.ptr + resources.frame_index as usize * resources.rtv_descriptor_size, }; unsafe { command_list.OMSetRenderTargets(1, Some(&rtv_handle), false, None) }; // Record commands. unsafe { command_list.ClearRenderTargetView(rtv_handle, &[0.3, 0.4, 0.6, 1.0], Some(&[])); command_list.IASetPrimitiveTopology(D3D_PRIMITIVE_TOPOLOGY_TRIANGLELIST); command_list.IASetVertexBuffers(0, Some(&[resources.vbv])); command_list.DrawInstanced(3, 1, 0, 0); command_list.ResourceBarrier(&[transition_barrier( &resources.render_targets[resources.frame_index as usize], D3D12_RESOURCE_STATE_RENDER_TARGET, D3D12_RESOURCE_STATE_COPY_SOURCE, )]); command_list.ResourceBarrier(&[transition_barrier( &resources.framebuffer, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE, D3D12_RESOURCE_STATE_COPY_DEST, )]); } unsafe { command_list.CopyResource( &resources.framebuffer, &resources.render_targets[resources.frame_index as usize], ); command_list.ResourceBarrier(&[transition_barrier( &resources.framebuffer, D3D12_RESOURCE_STATE_COPY_DEST, D3D12_RESOURCE_STATE_PIXEL_SHADER_RESOURCE, )]); command_list.ResourceBarrier(&[transition_barrier( &resources.render_targets[resources.frame_index as usize], D3D12_RESOURCE_STATE_COPY_SOURCE, D3D12_RESOURCE_STATE_RENDER_TARGET, )]); filter .frame( command_list, D3D12InputImage { resource: resources.framebuffer.clone(), descriptor: framebuffer, }, &Viewport { x: 0.0, y: 0.0, mvp: None, output: D3D12OutputView::new_from_raw( rtv_handle, Size::new( resources.viewport.Width as u32, resources.viewport.Height as u32, ), DXGI_FORMAT_R8G8B8A8_UNORM, ), }, frame_count, None, ) .unwrap(); command_list.ResourceBarrier(&[transition_barrier( &resources.render_targets[resources.frame_index as usize], D3D12_RESOURCE_STATE_RENDER_TARGET, D3D12_RESOURCE_STATE_PRESENT, )]); } unsafe { command_list.Close() } } fn transition_barrier( resource: &ID3D12Resource, state_before: D3D12_RESOURCE_STATES, state_after: D3D12_RESOURCE_STATES, ) -> D3D12_RESOURCE_BARRIER { D3D12_RESOURCE_BARRIER { Type: D3D12_RESOURCE_BARRIER_TYPE_TRANSITION, Flags: D3D12_RESOURCE_BARRIER_FLAG_NONE, Anonymous: D3D12_RESOURCE_BARRIER_0 { Transition: ManuallyDrop::new(D3D12_RESOURCE_TRANSITION_BARRIER { pResource: ManuallyDrop::new(Some(resource.clone())), StateBefore: state_before, StateAfter: state_after, Subresource: D3D12_RESOURCE_BARRIER_ALL_SUBRESOURCES, }), }, } } fn create_device(command_line: &SampleCommandLine) -> Result<(IDXGIFactory4, ID3D12Device)> { unsafe { let mut debug: Option = None; if let Some(debug) = D3D12GetDebugInterface(&mut debug).ok().and(debug) { eprintln!("enabling debug"); debug.EnableDebugLayer(); } } let dxgi_factory_flags = DXGI_CREATE_FACTORY_DEBUG; let dxgi_factory: IDXGIFactory4 = unsafe { CreateDXGIFactory2(dxgi_factory_flags) }?; let adapter = if command_line.use_warp_device { unsafe { dxgi_factory.EnumWarpAdapter() } } else { get_hardware_adapter(&dxgi_factory) }?; let mut device: Option = None; unsafe { D3D12CreateDevice(&adapter, D3D_FEATURE_LEVEL_12_2, &mut device) }?; Ok((dxgi_factory, device.unwrap())) } fn create_root_signature(device: &ID3D12Device) -> Result { let desc = D3D12_ROOT_SIGNATURE_DESC { Flags: D3D12_ROOT_SIGNATURE_FLAG_ALLOW_INPUT_ASSEMBLER_INPUT_LAYOUT, ..Default::default() }; let mut signature = None; let signature = unsafe { D3D12SerializeRootSignature(&desc, D3D_ROOT_SIGNATURE_VERSION_1, &mut signature, None) } .map(|()| signature.unwrap())?; unsafe { device.CreateRootSignature( 0, std::slice::from_raw_parts( signature.GetBufferPointer() as _, signature.GetBufferSize(), ), ) } } fn compile_shader(source: &[u8], entry: &[u8], version: &[u8]) -> Result { unsafe { let mut blob = None; D3DCompile( source.as_ptr().cast(), source.len(), None, None, None, PCSTR(entry.as_ptr()), PCSTR(version.as_ptr()), D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION, 0, &mut blob, None, )?; Ok(blob.unwrap()) } } fn create_pipeline_state( device: &ID3D12Device, root_signature: &ID3D12RootSignature, ) -> Result { let _compile_flags = D3DCOMPILE_DEBUG | D3DCOMPILE_SKIP_OPTIMIZATION; let vertex_shader = compile_shader(SHADER, b"VSMain\0", b"vs_5_0\0")?; let pixel_shader = compile_shader(SHADER, b"PSMain\0", b"ps_5_0\0")?; let mut input_element_descs: [D3D12_INPUT_ELEMENT_DESC; 2] = [ D3D12_INPUT_ELEMENT_DESC { SemanticName: s!("POSITION"), SemanticIndex: 0, Format: DXGI_FORMAT_R32G32B32_FLOAT, InputSlot: 0, AlignedByteOffset: 0, InputSlotClass: D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, InstanceDataStepRate: 0, }, D3D12_INPUT_ELEMENT_DESC { SemanticName: s!("COLOR"), SemanticIndex: 0, Format: DXGI_FORMAT_R32G32B32A32_FLOAT, InputSlot: 0, AlignedByteOffset: 12, InputSlotClass: D3D12_INPUT_CLASSIFICATION_PER_VERTEX_DATA, InstanceDataStepRate: 0, }, ]; let mut desc = D3D12_GRAPHICS_PIPELINE_STATE_DESC { InputLayout: D3D12_INPUT_LAYOUT_DESC { pInputElementDescs: input_element_descs.as_mut_ptr(), NumElements: input_element_descs.len() as u32, }, pRootSignature: ManuallyDrop::new(Some(root_signature.clone())), VS: D3D12_SHADER_BYTECODE { pShaderBytecode: unsafe { vertex_shader.GetBufferPointer() }, BytecodeLength: unsafe { vertex_shader.GetBufferSize() }, }, PS: D3D12_SHADER_BYTECODE { pShaderBytecode: unsafe { pixel_shader.GetBufferPointer() }, BytecodeLength: unsafe { pixel_shader.GetBufferSize() }, }, RasterizerState: D3D12_RASTERIZER_DESC { FillMode: D3D12_FILL_MODE_SOLID, CullMode: D3D12_CULL_MODE_NONE, ..Default::default() }, BlendState: D3D12_BLEND_DESC { AlphaToCoverageEnable: false.into(), IndependentBlendEnable: false.into(), RenderTarget: [ D3D12_RENDER_TARGET_BLEND_DESC { BlendEnable: false.into(), LogicOpEnable: false.into(), SrcBlend: D3D12_BLEND_ONE, DestBlend: D3D12_BLEND_ZERO, BlendOp: D3D12_BLEND_OP_ADD, SrcBlendAlpha: D3D12_BLEND_ONE, DestBlendAlpha: D3D12_BLEND_ZERO, BlendOpAlpha: D3D12_BLEND_OP_ADD, LogicOp: D3D12_LOGIC_OP_NOOP, RenderTargetWriteMask: D3D12_COLOR_WRITE_ENABLE_ALL.0 as u8, }, D3D12_RENDER_TARGET_BLEND_DESC::default(), D3D12_RENDER_TARGET_BLEND_DESC::default(), D3D12_RENDER_TARGET_BLEND_DESC::default(), D3D12_RENDER_TARGET_BLEND_DESC::default(), D3D12_RENDER_TARGET_BLEND_DESC::default(), D3D12_RENDER_TARGET_BLEND_DESC::default(), D3D12_RENDER_TARGET_BLEND_DESC::default(), ], }, DepthStencilState: D3D12_DEPTH_STENCIL_DESC::default(), SampleMask: u32::max_value(), PrimitiveTopologyType: D3D12_PRIMITIVE_TOPOLOGY_TYPE_TRIANGLE, NumRenderTargets: 1, SampleDesc: DXGI_SAMPLE_DESC { Count: 1, ..Default::default() }, ..Default::default() }; desc.RTVFormats[0] = DXGI_FORMAT_R8G8B8A8_UNORM; unsafe { device.CreateGraphicsPipelineState(&desc) } } fn create_vertex_buffer( device: &ID3D12Device, _aspect_ratio: f32, ) -> Result<(ID3D12Resource, D3D12_VERTEX_BUFFER_VIEW)> { let vertices = [ Vertex { position: [0.5f32, -0.5, 0.0], color: [1.0, 0.0, 0.0, 1.0], }, Vertex { position: [-0.5, -0.5, 0.0], color: [0.0, 1.0, 0.0, 1.0], }, Vertex { position: [0.0, 0.5, 0.0], color: [0.0, 0.0, 1.0, 1.0], }, ]; // 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. let mut vertex_buffer: Option = None; unsafe { device.CreateCommittedResource( &D3D12_HEAP_PROPERTIES { Type: D3D12_HEAP_TYPE_UPLOAD, ..Default::default() }, D3D12_HEAP_FLAG_NONE, &D3D12_RESOURCE_DESC { Dimension: D3D12_RESOURCE_DIMENSION_BUFFER, Width: std::mem::size_of_val(&vertices) as u64, Height: 1, DepthOrArraySize: 1, MipLevels: 1, SampleDesc: DXGI_SAMPLE_DESC { Count: 1, Quality: 0, }, Layout: D3D12_TEXTURE_LAYOUT_ROW_MAJOR, ..Default::default() }, D3D12_RESOURCE_STATE_GENERIC_READ, None, &mut vertex_buffer, )? }; let vertex_buffer = vertex_buffer.unwrap(); // Copy the triangle data to the vertex buffer. unsafe { let mut data = std::ptr::null_mut(); vertex_buffer.Map(0, None, Some(&mut data))?; std::ptr::copy_nonoverlapping(vertices.as_ptr(), data as *mut Vertex, vertices.len()); vertex_buffer.Unmap(0, None); } let vbv = D3D12_VERTEX_BUFFER_VIEW { BufferLocation: unsafe { vertex_buffer.GetGPUVirtualAddress() }, StrideInBytes: std::mem::size_of::() as u32, SizeInBytes: std::mem::size_of_val(&vertices) as u32, }; Ok((vertex_buffer, vbv)) } #[repr(C)] struct Vertex { position: [f32; 3], color: [f32; 4], } fn wait_for_previous_frame(resources: &mut Resources) { // WAITING FOR THE FRAME TO COMPLETE BEFORE CONTINUING IS NOT BEST // PRACTICE. This is code implemented as such for simplicity. The // D3D12HelloFrameBuffering sample illustrates how to use fences for // efficient resource usage and to maximize GPU utilization. // Signal and increment the fence value. let fence = resources.fence_value; unsafe { resources.command_queue.Signal(&resources.fence, fence) } .ok() .unwrap(); resources.fence_value += 1; // Wait until the previous frame is finished. if unsafe { resources.fence.GetCompletedValue() } < fence { unsafe { resources .fence .SetEventOnCompletion(fence, resources.fence_event) } .ok() .unwrap(); unsafe { WaitForSingleObject(resources.fence_event, INFINITE) }; } resources.frame_index = unsafe { resources.swap_chain.GetCurrentBackBufferIndex() }; } } pub(crate) fn main(sample: S) -> Result<()> { run_sample(sample)?; Ok(()) }