Read from a read_write storage buffer in a vertex shader without vertex_writable_storage

[mmgeorge]

Assume I have the following shader:

@binding(0) @group(0) var<storage, read_write> particles : array<f32>;

@compute @workgroup_size(1) 
fn compute_main()
  // write to particles
}

@vertex 
fn vs_main() {
  // read only from particles
}

...

If I create a BindGroupLayout with:

ty: gpu::BufferBindingType::Storage { read_only: true },
visibility: gpu::ShaderStages::VERTEX_FRAGMENT | gpu::ShaderStages::COMPUTE,

I'll expectedly run into an error since I've marked it as read_only:

In Device::create_render_pipeline, label = 'physics::render::pipeline:22'
     Error matching ShaderStages(VERTEX) shader requirements against the pipeline
       Shader global ResourceBinding { group: 0, binding: 0 } is not available in the pipeline layout
         Storage class Storage { access: StorageAccess(LOAD) } doesn't match the shader Storage { access: StorageAccess(LOAD | STORE) }

However, if I swap this to now be writable:

ty: gpu::BufferBindingType::Storage { read_only: false },
visibility: gpu::ShaderStages::VERTEX_FRAGMENT | gpu::ShaderStages::COMPUTE,

I'll get another validation error telling me that I need to enable the vertex_writable_storage feature:

Caused by:
In Device::create_bind_group_layout, label = 'physics::render::shader:18 [group0]'
     Binding 0 entry is invalid
       Features Features(VERTEX_WRITABLE_STORAGE) are required but not enabled on the device

Is it possible to re-use a ShaderModule, BindGroupLayout, and BindGroup to point to a shared writable storage buffer? Or is the only way to go about this to break this up into two separate shaders & layouts? I suppose in that case I could still reuse the same BindGroup?

Any help appreciated!