Module wgpu_backend 

WebGPU-style backend abstraction layer for Proof Engine.

This module provides a GPU-API-agnostic abstraction that could be backed by OpenGL, Vulkan, Metal, WebGPU, or a pure-software fallback. No actual wgpu crate dependency is used — everything is expressed as engine-level abstractions.

Sub-modules

• backend — Core GPU backend enum, capabilities, handle types, and context trait
• renderer — Multi-backend renderer with render passes and draw calls
• compute — Compute shader dispatch, storage buffers, and CPU fallback
• shader_translate — Cross-compile shaders between GLSL, WGSL, SPIRV, HLSL, MSL
• abstraction — Unified GpuDevice / GpuQueue traits
• quality — Adaptive quality management and benchmarking
• headless — Off-screen rendering, thumbnails, and server-side rendering

Re-exports

pub use backend::GpuBackend;

pub use backend::BackendCapabilities;

pub use backend::BackendContext;

pub use backend::BufferHandle;

pub use backend::TextureHandle;

pub use backend::ShaderHandle;

pub use backend::PipelineHandle;

pub use backend::ComputePipelineHandle;

pub use backend::BufferUsage;

pub use backend::TextureFormat;

pub use backend::ShaderStage;

pub use renderer::MultiBackendRenderer;

pub use compute::ComputeContext;

pub use shader_translate::ShaderLanguage;

pub use abstraction::GpuDevice;

pub use abstraction::GpuQueue;

pub use quality::QualityLevel;

pub use quality::QualityManager;

pub use headless::HeadlessRenderer;

Modules

abstraction

Unified GPU abstraction layer: GpuDevice and GpuQueue traits that provide a single API surface regardless of the underlying graphics backend.

backend

Core GPU backend abstraction: enums, handle types, capability queries, and the BackendContext trait that every concrete backend implements.

compute

Compute shader support: storage buffers, dispatch, CPU fallback, profiling.

headless

Headless (off-screen) rendering: render scenes to pixel buffers without a window, generate thumbnails, batch-render, and server-side rendering.

quality

Adaptive quality management: auto-detect hardware capability, dynamically adjust rendering quality to maintain a target frame rate, and benchmarking.

renderer

Multi-backend renderer: render passes, draw calls, compute dispatch, buffer upload / readback, and vertex layout descriptions.

shader_translate

Shader translation between GLSL, WGSL, SPIRV (text), HLSL, and MSL. Includes reflection (extracting bindings, inputs, outputs) and validation.