1 repository
Generates textual descriptions and color-coded diagrams showing how logical indices map to thread IDs and register files.
Distinct from Memory Layout Visualizers: Distinct from Memory Layout Visualizers: focuses on visualizing thread-to-data mappings, not general memory layouts.
Explore 1 awesome GitHub repository matching operating systems & systems programming · Thread-Data Layout Visualizers. 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
Generates diagrams showing how logical indices map to thread IDs and register files in GPU kernels.