cc169af549
935: Properly implement compute pass usage scopes r=cwfitzgerald a=kvark **Connections** Fixes #934 Closes #933 **Description** The old barriers in a compute pass were totally wrong. They didn't combine usages across the bind groups, or with indirect buffer. This change implements the *actual* semantics as seen by the spec. At dispatch time we gather the usages across bind groups needed by the active pipeline, possibly add the indirect buffer into the mix, and that combined used is what we transition into. I think we can find a few interesting ideas on how to make this faster. In particular - how to avoid a separate `TrackerSet` per pass. But for now correctness is more important. **Testing** Tested on wgpu-rs examples Co-authored-by: Dzmitry Malyshau <kvarkus@gmail.com>
This is an active GitHub mirror of the WebGPU implementation in Rust, which now lives in "gfx/wgpu" of Mozilla-central. Issues and pull requests are accepted, but some bidirectional synchronization may be involved.
WebGPU
This is the core logic of an experimental WebGPU implementation. It's written in Rust and is based on gfx-hal with help of gfx-extras. See the upstream WebGPU specification (work in progress).
The implementation consists of the following parts:
- internal Rust API for WebGPU implementations to use
- Rust types shared between wgpu-core,wgpu-native, andwgpu-rsplayer- standalone application for replaying the API traces, useswinit
This repository contains the core of wgpu, and is not usable directly by applications.
If you are looking for the user-facing Rust API, you need wgpu-rs.
If you are looking for the native implementation or bindings to the API in other languages, you need wgpu-native.
Supported Platforms
| API | Windows 7/10 | Linux & Android | macOS & iOS |
|---|---|---|---|
| DX11 | ✅ | ||
| DX12 | ✔️ | ||
| Vulkan | ✔️ | ✔️ | |
| Metal | ✔️ | ||
| OpenGL | 🚧 | 🚧 |
✔️ = Primary support — ✅ = Secondary support — 🚧 = Unsupported, but support in progress