Upgrade to wgpu 28

[ChristopherBiscardi]

Upgrade to wgpu 28

Important

This can't merge until bevyengine/naga_oil#132 does, and the dependency is updated from my fork to the release.

Also requires wgpu 28 in dlss_wgpu: bevyengine/dlss_wgpu#17

Note

This does not enable mesh shaders, and neither does the naga_oil PR. I chose to do an upgrade first, then go back and see about mesh shaders.

Here's a general list of changes and what I did. Commits are grouped by feature except for the last one, which enabled solari when I ran the solari example.

MipmapFilterMode is split from FilterMode

• Split MipmapFilterMode from FilterMode improve bevy_ui Direction docs  #8314: Split MipmapFilterMode from FilterMode gfx-rs/wgpu#8314

solution: implement From for MipmapFilterMode/ImageFilterMode since the values are the same. The split was because the spec indicates they are different types, even though they have the same values.

Push Constants are now Immediates

• Bring immediate api in line with spec gfx-rs/wgpu#8724

immediate_size is a u32 so use that instead of PushConstantRange

Capabilities name changes

• [naga] Add capabilities governing binding array types gfx-rs/wgpu#8671

Got new list from https://github.com/gfx-rs/wgpu/blob/trunk/wgpu-core/src/device/mod.rs#L449 and copied it in.

subgroup_{min,max}_size moved from Limits to AdapterInfo

• Move subgroup size handling in line with spec gfx-rs/wgpu#8609

Update the limits to have the fields they have now, mirror the logic from the other limits calls.

multiview_mask

• Add multiview limits and tests gfx-rs/wgpu#8206

set to None because we don't use it currently. Its vaguely for VR.

error scope is now a guard

• docs(CHANGELOG): Update PR link gfx-rs/wgpu#8685

retain guard and then pop() it later

---

I made one mistake during the PR, thinking set_immediates was going to be the size of the immediates and not the offset. I'd like reviewers to take a look at immediates offset and size sites specifically just in case I missed something.

Here's a bunch of examples running

Known Issues

Note

There are no current known issues. Everything previous has been solved.

• enable extensions can not be written in naga oil in a way that allows them to be used in composed modules.

Update: this was fixed by introducing a flag in naga-oil to force shaders to allow ray queries in composed modules when using bevy_solari. This is temporary and will be different in WESL-future.

old explanation

This is blocking solari from working on nvidia (solari runs successfully on macos m1) because it needs `enable wgpu_ray_query;`. Putting the declaration before `#define_import_path` means it gets stripped out (afaict), and putting it after results in

error: expected global declaration, but found a global directive
  ┌─ embedded://bevy_solari/scene/raytracing_scene_bindings.wgsl:3:1
  │
3 │ enable wgpu_ray_query;
  │ ^^^^^^ written after first global declaration
  │
  = expected global declaration, but found a global directive

• dlss_wgpu mixes apis which causes panics now.

Previous notes as dlss_wgpu was being fixed here

The wgpu release notes don't mention which PR this was introduced in, only saying:

Using both the wgpu command encoding APIs and CommandEncoder::as_hal_mut on the same encoder will now result in a panic.

It was caused by gfx-rs/wgpu#8373 which claimed to not know of any use cases

With record on finish, the actual ordering on the command buffer is deeply counter intuitive (all as_hal would come first) and I think it additionally was just flat out broken in some ways

• https://discord.com/channels/691052431525675048/743663924229963868/1453786307099758683

Possible path forward is using multiple command buffers: https://discord.com/channels/691052431525675048/743663924229963868/1453795633503670415

[ChristopherBiscardi]

telemetry seems to be very firefox-oriented, so I've set it to None.

from: gfx-rs/wgpu#8576

[ChristopherBiscardi]

This is due to enumerate_adapters being async now. Commenting because its probably worth a look for reviewers above the mass of other changes.

[tychedelia]

Okay, yeah, interesting. We want to revisit this eventually but since it's for Vulkan blocking here should be fine?

[github-actions]

Your PR increases Bevy Minimum Supported Rust Version. Please update the rust-version field in the root Cargo.toml file.

[superdump]

Just noting the git repo dependencies here for visibility as something that needs to be fixed before merging this PR.

[superdump]

I'm going to approve, but there is one missing wgpu capability (CULL_DISTANCE) and there are git dependencies (naga_oil and dlss_wgpu) that need to be resolved with updates and releases in those crates before merging this.

[ChristopherBiscardi]

CULL_DISTANCE is not in Features, so doesn't need to be mapped.

[jasmine-nominal]

I think we need to move these to a new command encoder after add_command_buffer.

We basically need 3 command encoders:

1. Regular bevy encoder, puts start timestamp before dlss, and records the barriers needed before dlss
2. Dlss encoder
3. Regular bevy encoder, puts end timestamp after dlss

However I think we'll need some changes to bevy's diagnostics API in order to make this work, as currently it expects you to use the same encoder to start/end a timestamp query iirc.

I can work on this after this PR, but just wanted to note it.

[ChristopherBiscardi]

Solari (and superresolution) with the now released dlss_wgpu v3 in this pr.

[alice-i-cecile]

@ChristopherBiscardi ready to merge once conflicts are resolved: naga_oil is published!

[atlv24]

Closes #21512

[mockersf]

Example occlusion_culling crashes since this PR was merged #22655