add-sgl-kernel

What problem does it solve?

This Skill provides a comprehensive guide to integrating custom, heavyweight CUDA/C++ kernels into the sgl-kernel library, streamlining the development of high-performance GPU operations.

Core Features & Use Cases

• Kernel Implementation: Step-by-step instructions for writing CUDA kernels for element-wise operations.
• Integration Workflow: Details on updating C++ declarations, Torch extension registration, CMake build files, and Python APIs.
• Testing & Benchmarking: Guidance on creating robust tests and performance benchmarks for new kernels.
• Use Case: A developer needs to add a custom tensor scaling operation to sgl-kernel that leverages CUTLASS for maximum performance. This Skill guides them through the entire process from kernel implementation to testing.

Quick Start

Follow the tutorial to implement a new scale kernel in sgl-kernel, including tests and benchmarks.