Formalizing Nvidia PTX in Lean to Prove GPU Kernel Correctness
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
The paper formalizes Nvidia's PTX GPU memory model in the Lean proof assistant so that an LLM can write proofs that specific GPU kernels compute correct results, with Lean's small trusted checker verifying each proof.