Browse skills

Compile hand-written native compute shaders and dispatch them from a Stream through NativeShaderExt — HLSL on dx/vk (DXC), GLSL on vk (glslang), CUDA C++ on cuda (NVRTC) — covering compile/reflect/load, NativeShaderLauncher binding by index/register+space/positional, uniforms (root constants / push constants / scalar kernel params), the per-argument Usage contract, include dirs, and lifetime.

The English summary is being prepared. Showing the original description.

RenderDoc GPU frame-capture analysis — headless .rdc capture, batch deserialization, frame reports and full-text search scripts, plus the manual Event Browser / Pipeline State / shader-debug workflow. Use when an image is wrong (missing, black, NaN, wrong color, flickering), when pipeline state or resource contents need ground truth, or when a frame looks slow.

The English summary is being prepared. Showing the original description.

LuisaCompute API guide covering the DSL (kernels, callables, structs, buffers, atomics, control flow, dispatch), the runtime API (Context, Device, Stream, buffers, images, ray tracing, rasterization), AST usage markers and AST-to-XIR lowering, the core library (traits, types, logging, math, fiber, allocators, STL containers), and vstd containers (HashMap, queues, pools, variant, smart pointers, utilities).

The English summary is being prepared. Showing the original description.

Optimize LuisaCompute DSL kernels using warp/wave primitives, shared-memory aggregation, block-level collectives, thread-group (block) size / occupancy tuning, and C++ branch-prediction hints (`[[likely]]`/`[[unlikely]]`). Use when kernels bottleneck on atomics, reductions, inter-thread communication, or poor occupancy.

The English summary is being prepared. Showing the original description.

Project layout, module architecture, compiler pipeline, and design patterns.

The English summary is being prepared. Showing the original description.

XMake build configuration, options, commands, and patterns for LuisaCompute.

The English summary is being prepared. Showing the original description.

High-performance C JSON parsing, creation, and modification with yyjson. Use when working with yyjson documents, string copy-vs-borrow ownership, dangling c_str() bugs, yyjson_alc allocators, or writer output freeing.

The English summary is being prepared. Showing the original description.

Backend plugin architecture, DeviceInterface, dynamic loading, and command encoding.

The English summary is being prepared. Showing the original description.

Native Vulkan XIR-to-SPIR-V codegen, legalization, validation, bindings, control flow, and target features.

The English summary is being prepared. Showing the original description.

Show uncommitted changes and commit history via git. Use when: (1) inspecting what changed in a file before committing, (2) checking diff of tracked/new/deleted files, (3) reviewing a specific commit's changes, (4) listing files changed in a commit.

The English summary is being prepared. Showing the original description.

SPIRV-Tools optimizer pass development, IR manipulation, PassTest registration, and LuisaCompute spvtools::Optimizer integration (presets, LUISA_SPIRV_OPT_* env knobs).

The English summary is being prepared. Showing the original description.

Debug crashes and test failures via stack-traces, host/device logging, DSL buffer inspection, and add_debug_checks kernel restrict-mode guards.

The English summary is being prepared. Showing the original description.

Navigate and modify LuisaCompute AST-to-XIR lowering, CFG/SSA representation changes, pass composition, and Metal XIR-to-LLVM/AIR normalization. Use when tracing AST-to-XIR calls, changing pass order, debugging structured versus plain CFG, or extending AIR ABI features such as bool/byte layout, PACK, acceleration, raster stages, runtime builtins, stateful ray queries, external linkage, and fail-closed boundaries.

The English summary is being prepared. Showing the original description.

Experimental Vulkan AST-to-LLVM-to-SPIR-V backend, its fail-closed runtime-interface boundary, LLVM build integration, and validation path.

The English summary is being prepared. Showing the original description.

Implement, compose, test, and debug LuisaCompute XIR transformation and analysis passes. Use when modifying files under src/xir/passes or include/luisa/xir/passes, changing CFG or SSA invariants, authoring a PassPipeline, or diagnosing pass-order, reachability, PHI, dominance, aliasing, or memory-effect bugs.

The English summary is being prepared. Showing the original description.

glslang SPIR-V Builder API for types, instructions, control flow, and decorations.

The English summary is being prepared. Showing the original description.

clangd LSP over HTTP for C++ syntax checks and symbol navigation in LuisaCompute sources.

The English summary is being prepared. Showing the original description.

CMake build options, custom functions, and backend patterns for LuisaCompute.

The English summary is being prepared. Showing the original description.

HLSL code generation, StringBuilder patterns, builtin headers, and DXIL embedding.

The English summary is being prepared. Showing the original description.

Boost.UT test layout, device helpers, adding tests, and running them.

The English summary is being prepared. Showing the original description.

Manual AST construction with FunctionBuilder for kernels and callables without DSL sugar.

The English summary is being prepared. Showing the original description.

C++ naming, formatting, static analysis, and RTTI rules for LuisaCompute.

The English summary is being prepared. Showing the original description.