What causes "ray tracing pipelines not yet implemented" in gfx-rs/wgpu?

Use the Vulkan backend (or another backend with ray tracing implemented) for ray tracing workloads

ray tracing pipelines not yet implemented

Error message

ray tracing pipelines not yet implemented

What it means

The DX12 HAL does not implement ray tracing pipelines; get_raytracing_pipeline_group_data is a hard stub that panics with 'ray tracing pipelines not yet implemented'. Any call into the ray-tracing pipeline API on this backend is guaranteed to panic.

Source

Thrown at wgpu-hal/src/dx12/device.rs:2314

        _desc: &crate::RayTracingPipelineDescriptor<
            super::PipelineLayout,
            super::ShaderModule,
            super::PipelineCache,
        >,
    ) -> Result<<Self::A as crate::Api>::RayTracingPipeline, crate::PipelineError> {
        unreachable!("ray tracing pipelines not yet implemented")
    }

    unsafe fn destroy_ray_tracing_pipeline(&self, _pipeline: super::RayTracingPipeline) {
        unreachable!("ray tracing pipelines not yet implemented")
    }

    unsafe fn get_raytracing_pipeline_group_data(
        &self,
        _pipeline: &super::RayTracingPipeline,
        _groups: core::ops::Range<u32>,
    ) -> Result<Vec<u8>, crate::DeviceError> {
        unimplemented!("ray tracing pipelines not yet implemented")
    }

    unsafe fn create_pipeline_cache(
        &self,
        _desc: &crate::PipelineCacheDescriptor<'_>,
    ) -> Result<super::PipelineCache, crate::PipelineCacheError> {
        Ok(super::PipelineCache)
    }
    unsafe fn destroy_pipeline_cache(&self, _: super::PipelineCache) {}

    unsafe fn create_query_set(
        &self,
        desc: &wgt::QuerySetDescriptor<crate::Label>,
    ) -> Result<super::QuerySet, crate::DeviceError> {
        let (heap_ty, raw_ty) = match desc.ty {
            wgt::QueryType::Occlusion => (
                Direct3D12::D3D12_QUERY_HEAP_TYPE_OCCLUSION,
                Direct3D12::D3D12_QUERY_TYPE_BINARY_OCCLUSION,

View on GitHub (pinned to 3e11ff59bf)

Solutions

Use the Vulkan backend (or another backend with ray tracing implemented) for ray tracing workloads
Check backend ray tracing support before calling any ray-tracing API (feature/back capability probe)
Implement the DX12 path with ID3D12Device5 RayTracingPipeline subobjects if you control the fork
Downgrade to compute-based ray tracing (BVH traversal in compute shaders) when targeting DX12

Example fix

// before
let groups = device.get_raytracing_pipeline_group_data(&pipeline, 0..n); // panics on dx12
// after
assert!(matches!(backend, Backend::Vulkan), "ray tracing only supported on vulkan");
let groups = device.get_raytracing_pipeline_group_data(&pipeline, 0..n);

Defensive patterns

Strategy: fallback

Validate before calling

// before using any ray tracing API, confirm backend support:
let rt_supported = matches!(backend, wgpu::Backend::Vulkan); // dx12/metal/gles lack it

Type guard

fn supports_ray_tracing(b: wgpu::Backend) -> bool { matches!(b, wgpu::Backend::Vulkan) }

Try / catch

// hal methods panic, not Result; wrap usage behind a capability check:
if !supports_ray_tracing(backend) { return Err("ray tracing unavailable on this backend"); }

Prevention

Never call ray-tracing hal APIs on the DX12 backend
Probe backend/feature support at init and fail fast
Track wgpu-hal releases for DX12 ray tracing implementation
Provide compute-shader ray tracing fallback

When it happens

Trigger: Calling get_raytracing_pipeline_group_data (part of the hal ray-tracing pipeline surface) on a Dx12 device — e.g. while creating/inspecting shader binding table group data for a ray tracing pipeline.

Common situations: Trying to use ray tracing through wgpu-hal's experimental API on DX12; assuming feature parity with the Vulkan backend; enabling ray tracing feature flags without checking backend support.

Related errors

AI-assisted analysis of gfx-rs/wgpu@3e11ff59bf (2026-09-03). Data as JSON: /api/errors/327bff9a3283cb58. Report an issue: GitHub.