1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
// Copyright 2022 the Vello Authors
// SPDX-License-Identifier: Apache-2.0 OR MIT
#![warn(clippy::doc_markdown, clippy::semicolon_if_nothing_returned)]
//! Vello is an experimental 2d graphics rendering engine written in Rust, using [`wgpu`].
//! It efficiently draws large 2d scenes with interactive or near-interactive performance.
//!
//! ![image](https://github.com/linebender/vello/assets/8573618/cc2b742e-2135-4b70-8051-c49aeddb5d19)
//!
//!
//! ## Motivation
//!
//! Vello is meant to fill the same place in the graphics stack as other vector graphics renderers like [Skia](https://skia.org/), [Cairo](https://www.cairographics.org/), and its predecessor project [Piet](https://www.cairographics.org/).
//! On a basic level, that means it provides tools to render shapes, images, gradients, texts, etc, using a PostScript-inspired API, the same that powers SVG files and [the browser `<canvas>` element](https://developer.mozilla.org/en-US/docs/Web/API/CanvasRenderingContext2D).
//!
//! Vello's selling point is that it gets better performance than other renderers by better leveraging the GPU.
//! In traditional PostScript-style renderers, some steps of the render process like sorting and clipping either need to be handled in the CPU or done through the use of intermediary textures.
//! Vello avoids this by using prefix-scan algorithms to parallelize work that usually needs to happen in sequence, so that work can be offloaded to the GPU with minimal use of temporary buffers.
//!
//! This means that Vello needs a GPU with support for compute shaders to run.
//!
//!
//! ## Getting started
//!
//! Vello is meant to be integrated deep in UI render stacks.
//! While drawing in a Vello scene is easy, actually rendering that scene to a surface setting up a wgpu context, which is a non-trivial task.
//!
//! To use Vello as the renderer for your PDF reader / GUI toolkit / etc, your code will have to look roughly like this:
//!
//! ```ignore
//! // Initialize wgpu and get handles
//! let (width, height) = ...;
//! let device: wgpu::Device = ...;
//! let queue: wgpu::Queue = ...;
//! let surface: wgpu::Surface<'_> = ...;
//! let texture_format: wgpu::TextureFormat = ...;
//! let mut renderer = Renderer::new(
//! &device,
//! RendererOptions {
//! surface_format: Some(texture_format),
//! use_cpu: false,
//! antialiasing_support: vello::AaSupport::all(),
//! num_init_threads: NonZeroUsize::new(1),
//! },
//! ).expect("Failed to create renderer");
//!
//! // Create scene and draw stuff in it
//! let mut scene = vello::Scene::new();
//! scene.fill(
//! vello::peniko::Fill::NonZero,
//! vello::Affine::IDENTITY,
//! vello::Color::rgb8(242, 140, 168),
//! None,
//! &vello::Circle::new((420.0, 200.0), 120.0),
//! );
//!
//! // Draw more stuff
//! scene.push_layer(...);
//! scene.fill(...);
//! scene.stroke(...);
//! scene.pop_layer(...);
//!
//! // Render to your window/buffer/etc.
//! let surface_texture = surface.get_current_texture()
//! .expect("failed to get surface texture");
//! renderer
//! .render_to_surface(
//! &device,
//! &queue,
//! &scene,
//! &surface_texture,
//! &vello::RenderParams {
//! base_color: Color::BLACK, // Background color
//! width,
//! height,
//! antialiasing_method: AaConfig::Msaa16,
//! },
//! )
//! .expect("Failed to render to surface");
//! surface_texture.present();
//! ```
//!
//! See the [`examples/`](https://github.com/linebender/vello/tree/main/examples) folder to see how that code integrates with frameworks like winit and bevy.
mod cpu_dispatch;
mod cpu_shader;
mod recording;
mod render;
mod scene;
mod shaders;
#[cfg(feature = "wgpu")]
mod wgpu_engine;
use std::num::NonZeroUsize;
/// Styling and composition primitives.
pub use peniko;
/// 2D geometry, with a focus on curves.
pub use peniko::kurbo;
#[doc(hidden)]
pub use skrifa;
pub mod glyph;
#[cfg(feature = "wgpu")]
pub mod util;
pub use render::Render;
pub use scene::{DrawGlyphs, Scene};
#[cfg(feature = "wgpu")]
pub use util::block_on_wgpu;
pub use recording::{
BufferProxy, Command, ImageFormat, ImageProxy, Recording, ResourceId, ResourceProxy, ShaderId,
};
pub use shaders::FullShaders;
#[cfg(feature = "wgpu")]
use wgpu_engine::{ExternalResource, WgpuEngine};
/// Temporary export, used in `with_winit` for stats
pub use vello_encoding::BumpAllocators;
#[cfg(feature = "wgpu")]
use wgpu::{Device, Queue, SurfaceTexture, TextureFormat, TextureView};
#[cfg(feature = "wgpu-profiler")]
use wgpu_profiler::{GpuProfiler, GpuProfilerSettings};
/// Catch-all error type.
pub type Error = Box<dyn std::error::Error>;
/// Specialization of `Result` for our catch-all error type.
pub type Result<T> = std::result::Result<T, Error>;
/// Represents the antialiasing method to use during a render pass.
#[derive(Copy, Clone, PartialEq, Eq)]
pub enum AaConfig {
Area,
Msaa8,
Msaa16,
}
/// Represents the set of antialiasing configurations to enable during pipeline creation.
pub struct AaSupport {
pub area: bool,
pub msaa8: bool,
pub msaa16: bool,
}
impl AaSupport {
pub fn all() -> Self {
Self {
area: true,
msaa8: true,
msaa16: true,
}
}
pub fn area_only() -> Self {
Self {
area: true,
msaa8: false,
msaa16: false,
}
}
}
/// Renders a scene into a texture or surface.
#[cfg(feature = "wgpu")]
pub struct Renderer {
#[cfg_attr(not(feature = "hot_reload"), allow(dead_code))]
options: RendererOptions,
engine: WgpuEngine,
shaders: FullShaders,
blit: Option<BlitPipeline>,
target: Option<TargetTexture>,
#[cfg(feature = "wgpu-profiler")]
pub profiler: GpuProfiler,
#[cfg(feature = "wgpu-profiler")]
pub profile_result: Option<Vec<wgpu_profiler::GpuTimerQueryResult>>,
}
/// Parameters used in a single render that are configurable by the client.
pub struct RenderParams {
/// The background color applied to the target. This value is only applicable to the full
/// pipeline.
pub base_color: peniko::Color,
/// Dimensions of the rasterization target
pub width: u32,
pub height: u32,
/// The anti-aliasing algorithm. The selected algorithm must have been initialized while
/// constructing the `Renderer`.
pub antialiasing_method: AaConfig,
}
#[cfg(feature = "wgpu")]
pub struct RendererOptions {
/// The format of the texture used for surfaces with this renderer/device
/// If None, the renderer cannot be used with surfaces
pub surface_format: Option<TextureFormat>,
/// If true, run all stages up to fine rasterization on the CPU.
// TODO: Consider evolving this so that the CPU stages can be configured dynamically via
// `RenderParams`.
pub use_cpu: bool,
/// Represents the enabled set of AA configurations. This will be used to determine which
/// pipeline permutations should be compiled at startup.
pub antialiasing_support: AaSupport,
/// How many threads to use for initialisation of shaders.
///
/// Use `Some(1)` to use a single thread. This is recommended when on macOS
/// (see <https://github.com/bevyengine/bevy/pull/10812#discussion_r1496138004>)
///
/// Set to `None` to use a heuristic which will use many but not all threads
///
/// Has no effect on WebAssembly
pub num_init_threads: Option<NonZeroUsize>,
}
#[cfg(feature = "wgpu")]
impl Renderer {
/// Creates a new renderer for the specified device.
pub fn new(device: &Device, options: RendererOptions) -> Result<Self> {
let mut engine = WgpuEngine::new(options.use_cpu);
// If we are running in parallel (i.e. the number of threads is not 1)
if options.num_init_threads != NonZeroUsize::new(1) {
#[cfg(not(target_arch = "wasm32"))]
engine.use_parallel_initialisation();
}
let shaders = shaders::full_shaders(device, &mut engine, &options)?;
#[cfg(not(target_arch = "wasm32"))]
engine.build_shaders_if_needed(device, options.num_init_threads);
let blit = options
.surface_format
.map(|surface_format| BlitPipeline::new(device, surface_format));
Ok(Self {
options,
engine,
shaders,
blit,
target: None,
// Use 3 pending frames
#[cfg(feature = "wgpu-profiler")]
profiler: GpuProfiler::new(GpuProfilerSettings {
..Default::default()
})?,
#[cfg(feature = "wgpu-profiler")]
profile_result: None,
})
}
/// Renders a scene to the target texture.
///
/// The texture is assumed to be of the specified dimensions and have been created with
/// the [`wgpu::TextureFormat::Rgba8Unorm`] format and the [`wgpu::TextureUsages::STORAGE_BINDING`]
/// flag set.
pub fn render_to_texture(
&mut self,
device: &Device,
queue: &Queue,
scene: &Scene,
texture: &TextureView,
params: &RenderParams,
) -> Result<()> {
let (recording, target) = render::render_full(scene, &self.shaders, params);
let external_resources = [ExternalResource::Image(
*target.as_image().unwrap(),
texture,
)];
self.engine.run_recording(
device,
queue,
&recording,
&external_resources,
"render_to_texture",
#[cfg(feature = "wgpu-profiler")]
&mut self.profiler,
)?;
Ok(())
}
/// Renders a scene to the target surface.
///
/// This renders to an intermediate texture and then runs a render pass to blit to the
/// specified surface texture.
///
/// The surface is assumed to be of the specified dimensions and have been configured with
/// the same format passed in the constructing [`RendererOptions`]' `surface_format`.
/// Panics if `surface_format` was `None`
pub fn render_to_surface(
&mut self,
device: &Device,
queue: &Queue,
scene: &Scene,
surface: &SurfaceTexture,
params: &RenderParams,
) -> Result<()> {
let width = params.width;
let height = params.height;
let mut target = self
.target
.take()
.unwrap_or_else(|| TargetTexture::new(device, width, height));
// TODO: implement clever resizing semantics here to avoid thrashing the memory allocator
// during resize, specifically on metal.
if target.width != width || target.height != height {
target = TargetTexture::new(device, width, height);
}
self.render_to_texture(device, queue, scene, &target.view, params)?;
let blit = self
.blit
.as_ref()
.expect("renderer should have configured surface_format to use on a surface");
let mut encoder =
device.create_command_encoder(&wgpu::CommandEncoderDescriptor { label: None });
{
let surface_view = surface
.texture
.create_view(&wgpu::TextureViewDescriptor::default());
let bind_group = device.create_bind_group(&wgpu::BindGroupDescriptor {
label: None,
layout: &blit.bind_layout,
entries: &[wgpu::BindGroupEntry {
binding: 0,
resource: wgpu::BindingResource::TextureView(&target.view),
}],
});
let mut render_pass = encoder.begin_render_pass(&wgpu::RenderPassDescriptor {
label: None,
color_attachments: &[Some(wgpu::RenderPassColorAttachment {
view: &surface_view,
resolve_target: None,
ops: wgpu::Operations {
load: wgpu::LoadOp::Clear(wgpu::Color::default()),
store: wgpu::StoreOp::Store,
},
})],
depth_stencil_attachment: None,
occlusion_query_set: None,
timestamp_writes: None,
});
render_pass.set_pipeline(&blit.pipeline);
render_pass.set_bind_group(0, &bind_group, &[]);
render_pass.draw(0..6, 0..1);
}
queue.submit(Some(encoder.finish()));
self.target = Some(target);
Ok(())
}
/// Reload the shaders. This should only be used during `vello` development
#[cfg(feature = "hot_reload")]
pub async fn reload_shaders(&mut self, device: &Device) -> Result<()> {
device.push_error_scope(wgpu::ErrorFilter::Validation);
let mut engine = WgpuEngine::new(self.options.use_cpu);
// We choose not to initialise these shaders in parallel, to ensure the error scope works correctly
let shaders = shaders::full_shaders(device, &mut engine, &self.options)?;
let error = device.pop_error_scope().await;
if let Some(error) = error {
return Err(error.into());
}
self.engine = engine;
self.shaders = shaders;
Ok(())
}
/// Renders a scene to the target texture.
///
/// The texture is assumed to be of the specified dimensions and have been created with
/// the [`wgpu::TextureFormat::Rgba8Unorm`] format and the [`wgpu::TextureUsages::STORAGE_BINDING`]
/// flag set.
///
/// The return value is the value of the `BumpAllocators` in this rendering, which is currently used
/// for debug output.
///
/// This return type is not stable, and will likely be changed when a more principled way to access
/// relevant statistics is implemented
pub async fn render_to_texture_async(
&mut self,
device: &Device,
queue: &Queue,
scene: &Scene,
texture: &TextureView,
params: &RenderParams,
) -> Result<Option<BumpAllocators>> {
let mut render = Render::new();
let encoding = scene.encoding();
// TODO: turn this on; the download feature interacts with CPU dispatch
let robust = false;
let recording = render.render_encoding_coarse(encoding, &self.shaders, params, robust);
let target = render.out_image();
let bump_buf = render.bump_buf();
self.engine.run_recording(
device,
queue,
&recording,
&[],
"t_async_coarse",
#[cfg(feature = "wgpu-profiler")]
&mut self.profiler,
)?;
let mut bump: Option<BumpAllocators> = None;
if let Some(bump_buf) = self.engine.get_download(bump_buf) {
let buf_slice = bump_buf.slice(..);
let (sender, receiver) = futures_intrusive::channel::shared::oneshot_channel();
buf_slice.map_async(wgpu::MapMode::Read, move |v| sender.send(v).unwrap());
if let Some(recv_result) = receiver.receive().await {
recv_result?;
} else {
return Err("channel was closed".into());
}
let mapped = buf_slice.get_mapped_range();
bump = Some(bytemuck::pod_read_unaligned(&mapped));
}
// TODO: apply logic to determine whether we need to rerun coarse, and also
// allocate the blend stack as needed.
self.engine.free_download(bump_buf);
// Maybe clear to reuse allocation?
let mut recording = Recording::default();
render.record_fine(&self.shaders, &mut recording);
let external_resources = [ExternalResource::Image(target, texture)];
self.engine.run_recording(
device,
queue,
&recording,
&external_resources,
"t_async_fine",
#[cfg(feature = "wgpu-profiler")]
&mut self.profiler,
)?;
Ok(bump)
}
/// See [`Self::render_to_surface`]
pub async fn render_to_surface_async(
&mut self,
device: &Device,
queue: &Queue,
scene: &Scene,
surface: &SurfaceTexture,
params: &RenderParams,
) -> Result<Option<BumpAllocators>> {
let width = params.width;
let height = params.height;
let mut target = self
.target
.take()
.unwrap_or_else(|| TargetTexture::new(device, width, height));
// TODO: implement clever resizing semantics here to avoid thrashing the memory allocator
// during resize, specifically on metal.
if target.width != width || target.height != height {
target = TargetTexture::new(device, width, height);
}
let bump = self
.render_to_texture_async(device, queue, scene, &target.view, params)
.await?;
let blit = self
.blit
.as_ref()
.expect("renderer should have configured surface_format to use on a surface");
let mut encoder =
device.create_command_encoder(&wgpu::CommandEncoderDescriptor { label: None });
{
let surface_view = surface
.texture
.create_view(&wgpu::TextureViewDescriptor::default());
let bind_group = device.create_bind_group(&wgpu::BindGroupDescriptor {
label: None,
layout: &blit.bind_layout,
entries: &[wgpu::BindGroupEntry {
binding: 0,
resource: wgpu::BindingResource::TextureView(&target.view),
}],
});
let mut render_pass = encoder.begin_render_pass(&wgpu::RenderPassDescriptor {
label: None,
color_attachments: &[Some(wgpu::RenderPassColorAttachment {
view: &surface_view,
resolve_target: None,
ops: wgpu::Operations {
load: wgpu::LoadOp::Clear(wgpu::Color::default()),
store: wgpu::StoreOp::Store,
},
})],
depth_stencil_attachment: None,
timestamp_writes: None,
occlusion_query_set: None,
});
render_pass.set_pipeline(&blit.pipeline);
render_pass.set_bind_group(0, &bind_group, &[]);
render_pass.draw(0..6, 0..1);
}
#[cfg(feature = "wgpu-profiler")]
self.profiler.resolve_queries(&mut encoder);
queue.submit(Some(encoder.finish()));
self.target = Some(target);
#[cfg(feature = "wgpu-profiler")]
self.profiler.end_frame().unwrap();
#[cfg(feature = "wgpu-profiler")]
if let Some(result) = self
.profiler
.process_finished_frame(queue.get_timestamp_period())
{
self.profile_result = Some(result);
}
Ok(bump)
}
}
#[cfg(feature = "wgpu")]
struct TargetTexture {
view: TextureView,
width: u32,
height: u32,
}
#[cfg(feature = "wgpu")]
impl TargetTexture {
pub fn new(device: &Device, width: u32, height: u32) -> Self {
let texture = device.create_texture(&wgpu::TextureDescriptor {
label: None,
size: wgpu::Extent3d {
width,
height,
depth_or_array_layers: 1,
},
mip_level_count: 1,
sample_count: 1,
dimension: wgpu::TextureDimension::D2,
usage: wgpu::TextureUsages::STORAGE_BINDING | wgpu::TextureUsages::TEXTURE_BINDING,
format: wgpu::TextureFormat::Rgba8Unorm,
view_formats: &[],
});
let view = texture.create_view(&wgpu::TextureViewDescriptor::default());
Self {
view,
width,
height,
}
}
}
#[cfg(feature = "wgpu")]
struct BlitPipeline {
bind_layout: wgpu::BindGroupLayout,
pipeline: wgpu::RenderPipeline,
}
#[cfg(feature = "wgpu")]
impl BlitPipeline {
fn new(device: &Device, format: TextureFormat) -> Self {
const SHADERS: &str = r#"
@vertex
fn vs_main(@builtin(vertex_index) ix: u32) -> @builtin(position) vec4<f32> {
// Generate a full screen quad in normalized device coordinates
var vertex = vec2(-1.0, 1.0);
switch ix {
case 1u: {
vertex = vec2(-1.0, -1.0);
}
case 2u, 4u: {
vertex = vec2(1.0, -1.0);
}
case 5u: {
vertex = vec2(1.0, 1.0);
}
default: {}
}
return vec4(vertex, 0.0, 1.0);
}
@group(0) @binding(0)
var fine_output: texture_2d<f32>;
@fragment
fn fs_main(@builtin(position) pos: vec4<f32>) -> @location(0) vec4<f32> {
let rgba_sep = textureLoad(fine_output, vec2<i32>(pos.xy), 0);
return vec4(rgba_sep.rgb * rgba_sep.a, rgba_sep.a);
}
"#;
let shader = device.create_shader_module(wgpu::ShaderModuleDescriptor {
label: Some("blit shaders"),
source: wgpu::ShaderSource::Wgsl(SHADERS.into()),
});
let bind_layout = device.create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
label: None,
entries: &[wgpu::BindGroupLayoutEntry {
visibility: wgpu::ShaderStages::FRAGMENT,
binding: 0,
ty: wgpu::BindingType::Texture {
sample_type: wgpu::TextureSampleType::Float { filterable: true },
view_dimension: wgpu::TextureViewDimension::D2,
multisampled: false,
},
count: None,
}],
});
let pipeline_layout = device.create_pipeline_layout(&wgpu::PipelineLayoutDescriptor {
label: None,
bind_group_layouts: &[&bind_layout],
push_constant_ranges: &[],
});
let pipeline = device.create_render_pipeline(&wgpu::RenderPipelineDescriptor {
label: None,
layout: Some(&pipeline_layout),
vertex: wgpu::VertexState {
module: &shader,
entry_point: "vs_main",
buffers: &[],
},
fragment: Some(wgpu::FragmentState {
module: &shader,
entry_point: "fs_main",
targets: &[Some(wgpu::ColorTargetState {
format,
blend: None,
write_mask: wgpu::ColorWrites::ALL,
})],
}),
primitive: wgpu::PrimitiveState {
topology: wgpu::PrimitiveTopology::TriangleList,
strip_index_format: None,
front_face: wgpu::FrontFace::Ccw,
cull_mode: Some(wgpu::Face::Back),
polygon_mode: wgpu::PolygonMode::Fill,
unclipped_depth: false,
conservative: false,
},
depth_stencil: None,
multisample: wgpu::MultisampleState {
count: 1,
mask: !0,
alpha_to_coverage_enabled: false,
},
multiview: None,
});
Self {
bind_layout,
pipeline,
}
}
}