textureStore causes a panic on GL backend with compute shaders.

[KiranWells]

Description
When using the textureStore function in a compute shader, naga validates the shader but the gles hal layer panics with "Unsupported uniform type!". I get no errors when using the Vulkan backend, and the program works as expected. As far as I can tell from other references online (such as this post), this operation should be supported. However, I have very little experience with wgpu and compute shaders, so please correct me if I am wrong.

Repro steps
Create a compute pipeline with any shader that uses the textureStore function with the gles backend. For example:

// texture
@group(0) @binding(0) var my_tex: texture_storage_2d<rgba8unorm, write>;

@compute @workgroup_size(16, 16, 1)
fn main_color(@builtin(global_invocation_id) global_id: vec3<u32>) {
    textureStore(my_tex, vec2<i32>(i32(global_id.x), i32(global_id.y)), vec4<f32>(0.5));
}

Expected vs observed behavior
I expect this to fill the entire texture with a half-transparent gray. However, it panics when creating the compute pipeline.

Extra materials

Error log:

Adapter info: AdapterInfo { name: "Mesa Intel(R) UHD Graphics 620 (KBL GT2)", vendor: 32902, device: 0, device_type: IntegratedGpu, driver: "", driver_info: "", backend: Gl }
thread 'main' panicked at 'Unsupported uniform datatype!', /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-hal-0.14.1/src/gles/conv.rs:447:14
stack backtrace:
   0: rust_begin_unwind
             at /rustc/a55dd71d5fb0ec5a6a3a9e8c27b2127ba491ce52/library/std/src/panicking.rs:584:5
   1: core::panicking::panic_fmt
             at /rustc/a55dd71d5fb0ec5a6a3a9e8c27b2127ba491ce52/library/core/src/panicking.rs:142:14
   2: wgpu_hal::gles::conv::uniform_byte_size
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-hal-0.14.1/src/gles/conv.rs:447:14
   3: wgpu_hal::gles::device::<impl wgpu_hal::gles::Device>::create_pipeline
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-hal-0.14.1/src/gles/device.rs:316:36
   4: wgpu_hal::gles::device::<impl wgpu_hal::Device<wgpu_hal::gles::Api> for wgpu_hal::gles::Device>::create_compute_pipeline
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-hal-0.14.1/src/gles/device.rs:1049:21
   5: wgpu_core::device::Device<A>::create_compute_pipeline
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-core-0.14.2/src/device/mod.rs:2427:22
   6: wgpu_core::device::<impl wgpu_core::hub::Global<G>>::device_create_compute_pipeline
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-core-0.14.2/src/device/mod.rs:4943:34
   7: <wgpu::backend::direct::Context as wgpu::Context>::device_create_compute_pipeline
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-0.14.2/src/backend/direct.rs:1447:27
   8: wgpu::Device::create_compute_pipeline
             at /home/kiran/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/wgpu-0.14.2/src/lib.rs:2202:17
   9: accelerated_mandel::test
             at ./src/main.rs:69:27
  10: accelerated_mandel::main
             at ./src/main.rs:118:5
  11: core::ops::function::FnOnce::call_once
             at /rustc/a55dd71d5fb0ec5a6a3a9e8c27b2127ba491ce52/library/core/src/ops/function.rs:248:5
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.

Platform
Information about your OS, version of wgpu, your tech stack, etc.
OS: Arch Linux (rolling)
wgpu version: 0.14.2 (also having the same issue on the latest git version)
Adapter info: AdapterInfo { name: "Mesa Intel(R) UHD Graphics 620 (KBL GT2)", vendor: 32902, device: 0, device_type: IntegratedGpu, driver: "", driver_info: "", backend: Gl }

[James2022-rgb]

This should work now.

[teoxoy]

Fixed by #3361