Merge #70

70: Wholesome update for wgpu-0.3 r=everyone a=kvark

~~This update is a little more opinionated, i.e. the BGL creation just accepts a slice of things instead of a descriptor. I believe the reason for descriptors in the upstream API is mostly techincal - it's more convenient to generate bindings in Googles giant codegen. Following it verbatim doesn't appear to be strictly necessary, especially if we can extract better usability from Rust features.~~

The change also disables "gl" feature for the release. We'll re-enable it once we can provide it nicely, it will not be a breaking change.

I also noticed that barriers are not working correctly on the mipmap example. Will investigate separately - the native part is already published anyway. Edit: fixed by gfx-rs/wgpu#302

Co-authored-by: Dzmitry Malyshau <[email protected]>

Author: bors[bot]

.travis.yml

@@ -10,6 +10,7 @@ branches:
   - staging.tmp
 
 script:
-  - cargo test
+  - cargo check
   - if [[ $TRAVIS_OS_NAME == "linux" ]]; then cargo test --no-run --features vulkan; fi
-  - if [[ $TRAVIS_OS_NAME == "windows" ]]; then cargo test --no-run --features vulkan; fi
+  - if [[ $TRAVIS_OS_NAME == "osx" ]]; then cargo test --no-run --features metal; fi
+  - if [[ $TRAVIS_OS_NAME == "windows" ]]; then cargo test --no-run --features dx12; fi

Cargo.toml

@@ -4,9 +4,12 @@ version = "0.3.0"
 authors = [
 	"Dzmitry Malyshau <[email protected]>",
 	"Joshua Groves <[email protected]>",
+	"Lucas Kent <[email protected]>",
+	"kyren <[email protected]>",
+	"Cormac O'Brien <[email protected]>",
 ]
 edition = "2018"
-description = "Rusty wgpu API wrapper"
+description = "Rusty WebGPU API wrapper"
 homepage = "https://github.com/gfx-rs/wgpu-rs"
 repository = "https://github.com/gfx-rs/wgpu-rs"
 keywords = ["graphics"]
@@ -20,12 +23,13 @@ metal = ["wgn/gfx-backend-metal"]
 dx11 = ["wgn/gfx-backend-dx11"]
 dx12 = ["wgn/gfx-backend-dx12"]
 vulkan = ["wgn/gfx-backend-vulkan"]
-gl = ["wgn/gfx-backend-gl"]
+#TODO: there is no way to use glutin-0.20 with GL backend v0.3 yet
+#gl = ["wgn/gfx-backend-gl"]
 
 [dependencies]
-#TODO: only depend on the published version
-wgn = { package = "wgpu-native", version = "0.2.6", features = ["local", "window-winit"], git = "https://github.com/gfx-rs/wgpu", rev = "a47ff090bb042f1cb2ad7b13c76eca228390a311"}
+wgn = { package = "wgpu-native", version = "0.3.1", features = ["local"] }
 arrayvec = "0.4"
+raw-window-handle = "0.1"
 zerocopy = "0.2"
 
 [dev-dependencies]
@@ -34,3 +38,4 @@ env_logger = "0.6"
 glsl-to-spirv = "0.1"
 log = "0.4"
 png = "0.15"
+winit = "0.20.0-alpha3"

examples/capture/main.rs

@@ -9,16 +9,16 @@ fn main() {
 
     let instance = wgpu::Instance::new();
 
-    let adapter = instance.get_adapter(Some(&wgpu::RequestAdapterOptions {
+    let adapter = instance.request_adapter(&wgpu::RequestAdapterOptions {
         power_preference: wgpu::PowerPreference::LowPower,
-    }));
+    });
 
-    let mut device = adapter.request_device(Some(&wgpu::DeviceDescriptor {
+    let mut device = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },
         limits: wgpu::Limits::default(),
-    }));
+    });
 
     // Rendered image is 256×256 with 32-bit RGBA color
     let size = 256u32;
@@ -52,7 +52,7 @@ fn main() {
             device.create_command_encoder(&wgpu::CommandEncoderDescriptor { todo: 0 });
         encoder.begin_render_pass(&wgpu::RenderPassDescriptor {
             color_attachments: &[wgpu::RenderPassColorAttachmentDescriptor {
-                attachment: &texture.create_view(None),
+                attachment: &texture.create_default_view(),
                 resolve_target: None,
                 load_op: wgpu::LoadOp::Clear,
                 store_op: wgpu::StoreOp::Store,
@@ -78,7 +78,7 @@ fn main() {
             texture_extent,
         );
 
-        encoder.finish(None)
+        encoder.finish()
     };
 
     device.get_queue().submit(&[command_buffer]);

examples/cube/main.rs

@@ -129,26 +129,22 @@ impl framework::Example for Example {
                 wgpu::BindGroupLayoutBinding {
                     binding: 0,
                     visibility: wgpu::ShaderStage::VERTEX,
-                    ty: wgpu::BindingType::UniformBuffer,
-                    dynamic: false,
-                    multisampled: false,
-                    texture_dimension: wgpu::TextureViewDimension::D2,
+                    ty: wgpu::BindingType::UniformBuffer {
+                        dynamic: false,
+                    },
                 },
                 wgpu::BindGroupLayoutBinding {
                     binding: 1,
                     visibility: wgpu::ShaderStage::FRAGMENT,
-                    ty: wgpu::BindingType::SampledTexture,
-                    dynamic: false,
-                    multisampled: false,
-                    texture_dimension: wgpu::TextureViewDimension::D2,
+                    ty: wgpu::BindingType::SampledTexture {
+                        multisampled: false,
+                        dimension: wgpu::TextureViewDimension::D2,
+                    },
                 },
                 wgpu::BindGroupLayoutBinding {
                     binding: 2,
                     visibility: wgpu::ShaderStage::FRAGMENT,
                     ty: wgpu::BindingType::Sampler,
-                    dynamic: false,
-                    multisampled: false,
-                    texture_dimension: wgpu::TextureViewDimension::D2,
                 },
             ],
         });
@@ -173,7 +169,7 @@ impl framework::Example for Example {
             format: wgpu::TextureFormat::Rgba8UnormSrgb,
             usage: wgpu::TextureUsage::SAMPLED | wgpu::TextureUsage::COPY_DST,
         });
-        let texture_view = texture.create_view(None);
+        let texture_view = texture.create_default_view();
         let temp_buf = device
             .create_buffer_mapped(texels.len(), wgpu::BufferUsage::COPY_SRC)
             .fill_from_slice(&texels);
@@ -298,7 +294,7 @@ impl framework::Example for Example {
         });
 
         // Done
-        let init_command_buf = init_encoder.finish(None);
+        let init_command_buf = init_encoder.finish();
         device.get_queue().submit(&[init_command_buf]);
         Example {
             vertex_buf,
@@ -310,7 +306,7 @@ impl framework::Example for Example {
         }
     }
 
-    fn update(&mut self, _event: wgpu::winit::WindowEvent) {
+    fn update(&mut self, _event: winit::event::WindowEvent) {
         //empty
     }
 
@@ -325,7 +321,7 @@ impl framework::Example for Example {
         let mut encoder =
             device.create_command_encoder(&wgpu::CommandEncoderDescriptor { todo: 0 });
         encoder.copy_buffer_to_buffer(&temp_buf, 0, &self.uniform_buf, 0, 64);
-        device.get_queue().submit(&[encoder.finish(None)]);
+        device.get_queue().submit(&[encoder.finish()]);
     }
 
     fn render(&mut self, frame: &wgpu::SwapChainOutput, device: &mut wgpu::Device) {
@@ -354,7 +350,7 @@ impl framework::Example for Example {
             rpass.draw_indexed(0 .. self.index_count as u32, 0, 0 .. 1);
         }
 
-        device.get_queue().submit(&[encoder.finish(None)]);
+        device.get_queue().submit(&[encoder.finish()]);
     }
 }
 

examples/framework.rs

@@ -1,6 +1,8 @@
 use log::info;
+use winit::event::WindowEvent;
 
 #[cfg_attr(rustfmt, rustfmt_skip)]
+#[allow(unused)]
 pub const OPENGL_TO_WGPU_MATRIX: cgmath::Matrix4<f32> = cgmath::Matrix4::new(
     1.0, 0.0, 0.0, 0.0,
     0.0, -1.0, 0.0, 0.0,
@@ -33,53 +35,47 @@ pub fn load_glsl(code: &str, stage: ShaderStage) -> Vec<u32> {
     wgpu::read_spirv(glsl_to_spirv::compile(&code, ty).unwrap()).unwrap()
 }
 
-pub trait Example {
+pub trait Example: 'static {
     fn init(sc_desc: &wgpu::SwapChainDescriptor, device: &mut wgpu::Device) -> Self;
     fn resize(&mut self, sc_desc: &wgpu::SwapChainDescriptor, device: &mut wgpu::Device);
-    fn update(&mut self, event: wgpu::winit::WindowEvent);
+    fn update(&mut self, event: WindowEvent);
     fn render(&mut self, frame: &wgpu::SwapChainOutput, device: &mut wgpu::Device);
 }
 
 pub fn run<E: Example>(title: &str) {
-    use wgpu::winit::{
-        ElementState,
-        Event,
-        EventsLoop,
-        KeyboardInput,
-        VirtualKeyCode,
-        WindowEvent,
+    use winit::{
+        event_loop::{ControlFlow, EventLoop},
+        event,
     };
 
     env_logger::init();
-
-    let mut events_loop = EventsLoop::new();
-
+    let event_loop = EventLoop::new();
     info!("Initializing the window...");
 
     #[cfg(not(feature = "gl"))]
     let (_window, instance, hidpi_factor, size, surface) = {
-        use wgpu::winit::Window;
+        use raw_window_handle::HasRawWindowHandle as _;
 
-        let instance = wgpu::Instance::new();
 
-        let window = Window::new(&events_loop).unwrap();
+        let window = winit::window::Window::new(&event_loop).unwrap();
         window.set_title(title);
-        let hidpi_factor = window.get_hidpi_factor();
-        let size = window.get_inner_size().unwrap().to_physical(hidpi_factor);
+        let hidpi_factor = window.hidpi_factor();
+        let size = window.inner_size().to_physical(hidpi_factor);
 
-        let surface = instance.create_surface(&window);
+        let instance = wgpu::Instance::new();
+        let surface = instance.create_surface(window.raw_window_handle());
 
         (window, instance, hidpi_factor, size, surface)
     };
 
     #[cfg(feature = "gl")]
     let (_window, instance, hidpi_factor, size, surface) = {
-        let wb = wgpu::winit::WindowBuilder::new();
+        let wb = winit::WindowBuilder::new();
         let cb = wgpu::glutin::ContextBuilder::new().with_vsync(true);
-        let context = cb.build_windowed(wb, &events_loop).unwrap();
+        let context = cb.build_windowed(wb, &event_loop).unwrap();
         context.window().set_title(title);
 
-        let hidpi_factor = context.window().get_hidpi_factor();
+        let hidpi_factor = context.window().hidpi_factor();
         let size = context
             .window()
             .get_inner_size()
@@ -94,16 +90,16 @@ pub fn run<E: Example>(title: &str) {
         (window, instance, hidpi_factor, size, surface)
     };
 
-    let adapter = instance.get_adapter(Some(&wgpu::RequestAdapterOptions {
+    let adapter = instance.request_adapter(&wgpu::RequestAdapterOptions {
         power_preference: wgpu::PowerPreference::LowPower,
-    }));
+    });
 
-    let mut device = adapter.request_device(Some(&wgpu::DeviceDescriptor {
+    let mut device = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },
         limits: wgpu::Limits::default(),
-    }));
+    });
 
     let mut sc_desc = wgpu::SwapChainDescriptor {
         usage: wgpu::TextureUsage::OUTPUT_ATTACHMENT,
@@ -118,10 +114,14 @@ pub fn run<E: Example>(title: &str) {
     let mut example = E::init(&sc_desc, &mut device);
 
     info!("Entering render loop...");
-    let mut running = true;
-    while running {
-        events_loop.poll_events(|event| match event {
-            Event::WindowEvent {
+    event_loop.run(move |event, _, control_flow| {
+        *control_flow = if cfg!(feature = "metal-auto-capture") {
+            ControlFlow::Exit
+        } else {
+            ControlFlow::Poll
+        };
+        match event {
+            event::Event::WindowEvent {
                 event: WindowEvent::Resized(size),
                 ..
             } => {
@@ -132,30 +132,30 @@ pub fn run<E: Example>(title: &str) {
                 swap_chain = device.create_swap_chain(&surface, &sc_desc);
                 example.resize(&sc_desc, &mut device);
             }
-            Event::WindowEvent { event, .. } => match event {
+            event::Event::WindowEvent { event, .. } => match event {
                 WindowEvent::KeyboardInput {
                     input:
-                        KeyboardInput {
-                            virtual_keycode: Some(VirtualKeyCode::Escape),
-                            state: ElementState::Pressed,
+                        event::KeyboardInput {
+                            virtual_keycode: Some(event::VirtualKeyCode::Escape),
+                            state: event::ElementState::Pressed,
                             ..
                         },
                     ..
                 }
                 | WindowEvent::CloseRequested => {
-                    running = false;
+                    *control_flow = ControlFlow::Exit;
                 }
                 _ => {
                     example.update(event);
                 }
             },
+            event::Event::EventsCleared => {
+                let frame = swap_chain.get_next_texture();
+                example.render(&frame, &mut device);
+            }
             _ => (),
-        });
-
-        let frame = swap_chain.get_next_texture();
-        example.render(&frame, &mut device);
-        running &= !cfg!(feature = "metal-auto-capture");
-    }
+        }
+    });
 }
 
 // This allows treating the framework as a standalone example,

examples/hello-compute/main.rs

@@ -15,16 +15,16 @@ fn main() {
     let size = (numbers.len() * std::mem::size_of::<u32>()) as wgpu::BufferAddress;
 
     let instance = wgpu::Instance::new();
-    let adapter = instance.get_adapter(Some(&wgpu::RequestAdapterOptions {
+    let adapter = instance.request_adapter(&wgpu::RequestAdapterOptions {
         power_preference: wgpu::PowerPreference::Default,
-    }));
+    });
 
-    let mut device = adapter.request_device(Some(&wgpu::DeviceDescriptor {
+    let mut device = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },
         limits: wgpu::Limits::default(),
-    }));
+    });
 
     let cs = include_bytes!("shader.comp.spv");
     let cs_module = device.create_shader_module(&wgpu::read_spirv(std::io::Cursor::new(&cs[..])).unwrap());
@@ -46,14 +46,13 @@ fn main() {
     });
 
     let bind_group_layout = device.create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
-        bindings: &[wgpu::BindGroupLayoutBinding {
-            binding: 0,
-            visibility: wgpu::ShaderStage::COMPUTE,
-            ty: wgpu::BindingType::StorageBuffer,
-            dynamic: false,
-            multisampled: false,
-            texture_dimension: wgpu::TextureViewDimension::D2,
-        }],
+        bindings: &[
+            wgpu::BindGroupLayoutBinding {
+                binding: 0,
+                visibility: wgpu::ShaderStage::COMPUTE,
+                ty: wgpu::BindingType::StorageBuffer { dynamic: false, readonly: false },
+            },
+        ],
     });
 
     let bind_group = device.create_bind_group(&wgpu::BindGroupDescriptor {
@@ -89,7 +88,7 @@ fn main() {
     }
     encoder.copy_buffer_to_buffer(&storage_buffer, 0, &staging_buffer, 0, size);
 
-    device.get_queue().submit(&[encoder.finish(None)]);
+    device.get_queue().submit(&[encoder.finish()]);
 
     staging_buffer.map_read_async(0, size, |result: wgpu::BufferMapAsyncResult<&[u32]>| {
         if let Ok(mapping) = result {