Warpline: Binary-Level Tracing and Unified Evidence for Agentic GPU Kernel Optimization
cs.PF, cs.PL
Submitted: 2026-11-03
Updated: 2026-11-17
Code: https://github.com/facebookexperimental/CU
Terminology
Sources
- CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
- KEET: Explaining Performance of GPU Kernels Using LLM Agents
- Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization
- TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization
- SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits
- KernelClaw: A Multi-Agent Framework for GPU Kernel Optimization with Dual-Memory Mechanisms
- Prism: Symbolic Superoptimization of Tensor Programs
- LEO: Tracing GPU Stall Root Causes via Cross-Vendor Backward Slicing
Related papers
- Instruction-Aware Transfer of GPU Implementation Crossovers: a Tensor-Product Discontinuous Galerkin Operator on GB200, H100, and A100
- denet, A lightweight command-line tool for process monitoring in benchmarking and beyond
- HybridQC: Hardware-Grounded Simulation of Tightly Integrated Hybrid Quantum-Classical Systems
- TileBench: A Controlled Benchmark for Performance Evaluation and Bottleneck Diagnosis of Tile-Based Programming Models
- The M/PH 2/1 Queue: Explicit Distributions, Sojourn Times and Numerical Benchmarks
- Rolling Round-Robin Rate: Standard Heterogeneous Throughput for SPEC CPU