1 مستودع
Creates persistent thread-block looping patterns for dynamic work distribution across thread blocks in GPU kernels.
Distinct from Educational Kernel Thread Implementations: Distinct from Educational Kernel Thread Implementations: focuses on GPU thread-block persistence patterns, not OS kernel thread creation.
Explore 1 awesome GitHub repository matching operating systems & systems programming · Persistent Thread Block Patterns. Refine with filters or upvote what's useful.
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
Implements persistent thread-block patterns for dynamic work distribution across GPU thread blocks.