POPL 2025 (series) / Dafny 2025 (series) / Dafny 2025 /
Performant, Readable and Interoperable Rust from Dafny
This program is tentative and subject to change.
This paper introduces a novel Dafny-to-Rust compiler, bridging the gap between Dafny’s general-purpose formal verification capabilities and Rust’s high performance. We present an advanced translation architecture that generates efficient and largely readable Rust code from Dafny source. Our approach tackles critical challenges in type system disparities, memory management, and cross-language interoperability. We validate our compiler’s effectiveness through comprehensive performance evaluations, demonstrating competitive results: in 9 of our 10 benchmarks, the generated code was at most 22% slower than native Rust implementations, with 2 achieving performance parity.
This program is tentative and subject to change.
Sun 19 JanDisplayed time zone: Mountain Time (US & Canada) change
Sun 19 Jan
Displayed time zone: Mountain Time (US & Canada) change
14:00 - 15:30 | |||
14:00 18mTalk | Baking for Dafny: A CakeML Backend for Dafny Dafny Daniel Nezamabadi Chalmers University of Technology and University of Gothenburg, Magnus O. Myreen Chalmers University of Technology Pre-print | ||
14:18 18mTalk | Lean on Dafny: Exploring Interactive Verification of Dafny Programs in Lean Dafny Wojciech Różowski University College London, Georges-Axel Jaloyan Amazon Web Services, Sean McLaughlin Amazon Web Services | ||
14:36 18mTalk | Performant, Readable and Interoperable Rust from Dafny Dafny Mikael Mayer Automated Reasoning Group, Amazon Web Services, Shadaj Laddad University of California at Berkeley, Fabio Madge Automated Reasoning Group, Amazon Web Services, Siva Somayyajula Amazon Web Services, Jean-Baptiste Tristan Amazon Web Services | ||
14:54 18mTalk | Randomised Testing of the Dafny Compiler: Into the CI Dafny Karnbongkot Boonriong Imperial College London, Alastair F. Donaldson Imperial College London, Stefan Zetzsche Amazon Web Services | ||
15:12 18mTalk | Teaching Types and Non-Interference in Dafny Dafny Bryan Parno Carnegie Mellon University |