Upgrade to wgpu v24

[JMS55]

Didn't remove WgpuWrapper. Not sure if it's needed or not still.

Testing

• Did you test these changes? If so, how? Example runner
• Are there any parts that need more testing? Web (portable atomics thingy?), DXC.

Migration Guide

• Bevy has upgraded to wgpu v24.
• When using the DirectX 12 rendering backend, the new priority system for choosing a shader compiler is as follows:
  • If the WGPU_DX12_COMPILER environment variable is set at runtime, it is used
  • Else if the new statically-linked-dxc feature is enabled, a custom version of DXC will be statically linked into your app at compile time.
  • Else Bevy will look in the app's working directory for dxcompiler.dll and dxil.dll at runtime.
  • Else if they are missing, Bevy will fall back to FXC (not recommended)

[github-actions]

It looks like your PR is a breaking change, but you didn't provide a migration guide.

Could you add some context on what users should update when this change get released in a new version of Bevy?
It will be used to help writing the migration guide for the version. Putting it after a ## Migration Guide will help it get automatically picked up by our tooling.

[JMS55]

I've changed how wgpu works with DX12 so that it no longer uses FXC.

If wgpu ever defaults to DX12 (I think I've seen people have that happen, despite the fact that I thought VK was supposed to be the default) then it's going to try and use a dynamic DXC and crash since it'll be missing. Users can enable the static DXC feature to fix that, or switch to VK, but it might be a little confusing.

Eventually once wgpu's DX12 backend is more stable we should consider making static-dxc on by default, and removing vulkan (and OpenGL) by default for smaller binary sizes.

[alice-i-cecile]

Kicking off an example run: bevyengine/bevy-example-runner#108

[IceSentry]

Assuming the example runner passes LGTM

[JMS55]

@alice-i-cecile good to merge?

[JMS55]

@pcwalton please review the previous two commits for gpu preprocessing changes.

[pcwalton]

LGTM!

[JMS55]

The only issue remaining is texture_binding_array fails on macOS on the example runner. Not sure if there's a way to check logs or anything...

The fact that it only fails on macOS is probably a wgpu bug though. I'd like to merge this PR, we can look into it later.

[ChristopherBiscardi]

on m1 mac

2025-02-08T21:47:30.165803Z  INFO bevy_diagnostic::system_information_diagnostics_plugin::internal: SystemInfo { os: "macOS 14.2.1 Sonoma", kernel: "23.2.0", cpu: "Apple M1 Max", core_count: "10", memory: "64.0 GiB" }
2025-02-08T21:47:30.260362Z  INFO bevy_render::renderer: AdapterInfo { name: "Apple M1 Max", vendor: 0, device: 0, device_type: IntegratedGpu, driver: "", driver_info: "", backend: Metal }

texture_binding_array

pcss

ssr

❯ cargo run --example ssr
   Compiling bevy v0.16.0-dev (/Users/chris/github/bevyengine/bevy)
    Finished `dev` profile [unoptimized + debuginfo] target(s) in 5.93s
     Running `target/debug/examples/ssr`
2025-02-08T21:48:00.292415Z  INFO bevy_diagnostic::system_information_diagnostics_plugin::internal: SystemInfo { os: "macOS 14.2.1 Sonoma", kernel: "23.2.0", cpu: "Apple M1 Max", core_count: "10", memory: "64.0 GiB" }
2025-02-08T21:48:00.386114Z  INFO bevy_render::renderer: AdapterInfo { name: "Apple M1 Max", vendor: 0, device: 0, device_type: IntegratedGpu, driver: "", driver_info: "", backend: Metal }
2025-02-08T21:48:01.014102Z  INFO bevy_winit::system: Creating new window Bevy Screen Space Reflections Example (0v1)
2025-02-08T21:48:06.615888Z ERROR wgpu::backend::wgpu_core: Shader translation error for stage ShaderStages(FRAGMENT): mapping of ResourceBinding { group: 0, binding: 17 } is missing
2025-02-08T21:48:06.615916Z ERROR wgpu::backend::wgpu_core: Please report it to https://github.com/gfx-rs/wgpu
2025-02-08T21:48:06.615937Z ERROR wgpu::backend::wgpu_core: Handling wgpu errors as fatal by default
thread 'Compute Task Pool (2)' panicked at /Users/chris/.cargo/registry/src/index.crates.io-6f17d22bba15001f/wgpu-24.0.1/src/backend/wgpu_core.rs:1303:26:
wgpu error: Validation Error

Caused by:
  In Device::create_render_pipeline, label = 'deferred_lighting_pipeline'
    Internal error in ShaderStages(FRAGMENT) shader: mapping of ResourceBinding { group: 0, binding: 17 } is missing


note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Encountered a panic in system `bevy_render::render_resource::pipeline_cache::PipelineCache::process_pipeline_queue_system`!
thread '<unnamed>' panicked at /Users/chris/github/bevyengine/bevy/crates/bevy_render/src/render_resource/pipeline_cache.rs:577:28:
index out of bounds: the len is 0 but the index is 14
Encountered a panic in system `bevy_render::renderer::render_system`!

[ickshonpe]

Doesn't break anything with UI or text rendering, I don't know anything about anything else at all.

[JMS55]

2025-02-08T21:48:06.615888Z ERROR wgpu::backend::wgpu_core: Shader translation error for stage ShaderStages(FRAGMENT): mapping of ResourceBinding { group: 0, binding: 17 } is missing
2025-02-08T21:48:06.615916Z ERROR wgpu::backend::wgpu_core: Please report it to https://github.com/gfx-rs/wgpu
2025-02-08T21:48:06.615937Z ERROR wgpu::backend::wgpu_core: Handling wgpu errors as fatal by default

Seems like a wgpu bug

[superdump]

CI broke on this in the windows dx12 example runner:

2025-02-08T23:12:40.192270Z  INFO bevy_diagnostic::system_information_diagnostics_plugin::internal: SystemInfo { os: "Windows Server 2022 Datacenter", kernel: "20348", cpu: "AMD EPYC 7763 64-Core Processor", core_count: "2", memory: "16.0 GiB" }    

thread 'main' panicked at crates\bevy_render\src\renderer\mod.rs:197:10:
Unable to find a GPU! Make sure you have installed required drivers!
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
error: process didn't exit successfully: `target\debug\examples\ambiguity_detection.exe` (exit code: 101)

[mockersf]

JMS55#32 should fix windows dx12 CI