1 个仓库
Awesome GitHub RepositoriesCompile-Time Unrolled Loops
Iterates over a fixed range sequentially within a kernel with optional compile-time unrolling for small trip counts.
Distinct from Range-Based Loop Iteration: Distinct from Range-Based Loop Iteration: focuses on compile-time unrolling for GPU kernels, not C++ range-based for syntax.
Explore 1 awesome GitHub repository matching software engineering & architecture · Compile-Time Unrolled Loops. Refine with filters or upvote what's useful.
Awesome Compile-Time Unrolled Loops GitHub Repositories
用 AI 发现最棒的仓库。我们将通过 AI 为您搜索最匹配的仓库。
- tile-ai/tilelang
tile-ai/tilelang
5,226在 GitHub 上查看TileLang is a Python-embedded domain-specific language compiler that JIT-compiles and autotunes GPU kernels. It uses a tile-based DSL, automatic software pipelining, and parallel autotuning to generate optimized GPU kernels at runtime. It supports tensor core operations with Pythonic syntax, automatic memory management, and thread mapping. The compiler searches over tile sizes, thread counts, and scheduling policies, compiling and benchmarking candidates in parallel to find the fastest kernel. It also caches compiled binaries and tuning results to disk for reuse across sessions. TileLang inc
Provides sequential loop iteration with optional compile-time unrolling for small trip counts in GPU kernels.
Python
