"float32-filterable" feature support?

[HymnsForDisco]

The WebGPU spec has an API for optional features. Likewise, wgpu has a Features API, where the docs say

... Corresponds to WebGPU GPUFeatureName.

I'm a bit confused because the set of "Features" enumerated seems to be missing some (like "float32-filterable"), and contains other "extra" features that aren't mentioned in the WebGPU spec (like SHADER_F64).

Basically, I'm just wondering what's going on here, and if there's any way to enable the "float32-filterable" optional feature with wgpu.

[kpreid]

FWIW, I wrote the “Corresponds to…” notes in the docs, and I just meant “this type serves the same role”, for cross-reference, not “it should be expected to be completely identical”.

As to the actual differences, I'd expect but don't know for sure that they are not so much intentional as wgpu not yet supporting all optional functionality defined by WebGPU — and in the other direction, supporting some of its own “extensions”; wgpu has quite a few additional things that aren't specified by WebGPU.

[HymnsForDisco]

Understood, and thanks for the prompt reply. With that in mind, I'll create an issue / feature request about this.