arXiv subjects

Aymeric Fromherz

Publications and source records attributed to Aymeric Fromherz.

11 recordsLinked to original sources

arXiv2026

Scaling Verification of Cryptographic Software with Aeneas, Rust, and Lean

We develop a new methodology for verifying cryptographic software. We target production code written in Rust for performance and system integration, rather than verification convenience. Rust's ownership discipline enables Aeneas to extract a pure model of this code in Lean, relieving us from low-level reasoning about pointer liveness and aliasing. Lean's extensibility lets us develop tactics and libraries that greatly simplify reasoning about extracted Rust code. We design and tune our toolchain to facilitate the use of AI. Agents autonomously write formal proofs, which are independently verified by the Lean kernel. Agents also assist in the formalization of cryptographic standards and platform-specific intrinsics, which still requires expert design and review. We apply our methodology to SymCrypt, Microsoft's cryptographic provider. We verify its implementations of algorithms such as SHA-3 and ML-KEM, which were ported from C to Rust. We also extend SymCrypt with experimental optimizations and implementations of algorithms such as FrodoKEM, ML-DSA, and HPKE to explore the scalability of writing, adapting, and verifying cryptographic code. Our 237~KLOC Lean development establishes safety, panic-freedom, and functional correctness of 16.7~KLOC of Rust code supporting post-quantum cipher suites for x86-64 and ARM platforms. Our evaluation shows that verified Rust can meet SymCrypt's performance, portability, deployment, and maintainability requirements.

cs.CR↗

arXiv2026

Scylla: Translating an Applicative Subset of C to Safe Rust

The popularity of the Rust language continues to explode; yet, many critical codebases remain authored in C. Automatically translating C to Rust is thus an appealing course of action. Several works have gone down this path, handling an ever-increasing subset of C through a variety of Rust features, such as unsafe. While the prospect of automation is appealing, producing code that relies on unsafe negates the memory safety guarantees offered by Rust, and therefore the main advantages of porting existing codebases to memory-safe languages. We instead advocate for a different approach, where the programmer iterates on the original C, gradually making the code more structured until it becomes eligible for compilation to safe Rust. This means that redesigns and rewrites can be evaluated incrementally for performance and correctness against existing test suites and production environments. Compiling structured C to safe Rust relies on the following contributions: a type-directed translation from (a subset of) C to safe Rust; a novel static analysis based on "split trees" which allows expressing C's pointer arithmetic using Rust's slices and splitting operations; an analysis that infers which borrows need to be mutable; and a compilation strategy for C pointer types that is compatible with Rust's distinction between non-owned and owned allocations. We evaluate our approach on real-world cryptographic libraries, binary parsers and serializers, and a file compression library. We show that these can be rewritten to Rust with small refactors of the original C code, and that the resulting Rust code exhibits similar performance characteristics as the original C code. As part of our translation process, we also identify and report undefined behaviors in the bzip2 compression library and in Microsoft's implementation of the FrodoKEM cryptographic primitive.

cs.PL↗

arXiv2025

Charon: An Analysis Framework for Rust

With the explosion in popularity of the Rust programming language, a wealth of tools have recently been developed to analyze, verify, and test Rust programs. Alas, the Rust ecosystem remains relatively young, meaning that every one of these tools has had to re-implement difficult, time-consuming machinery to interface with the Rust compiler and its cargo build system, to hook into the Rust compiler's internal representation, and to expose an abstract syntax tree (AST) that is suitable for analysis rather than optimized for efficiency. We address this missing building block of the Rust ecosystem, and propose Charon, an analysis framework for Rust. Charon acts as a swiss-army knife for analyzing Rust programs, and deals with all of the tedium above, providing clients with a clean, stable AST that can serve as the foundation of many analyses. We demonstrate the usefulness of Charon through a series of case studies, ranging from a Rust verification framework (Aeneas), a compiler from Rust to C (Eurydice), and a novel taint-checker for cryptographic code. To drive the point home, we also re-implement a popular existing analysis (Rudra), and show that it can be replicated by leveraging the Charon framework.

cs.PL↗

arXiv2025

CUTECat: Concolic Execution for Computational Law

Many legal computations, including the amount of tax owed by a citizen, whether they are eligible to social benefits, or the wages due to civil state servants, are specified by computational laws. Their application, however, is performed by expert computer programs intended to faithfully transcribe the law into computer code. Bugs in these programs can lead to dramatic societal impact, e.g., paying employees incorrect amounts, or not awarding benefits to families in need. To address this issue, we consider concolic unit testing, a combination of concrete execution with SMT-based symbolic execution, and propose CUTECat, a concolic execution tool targeting implementations of computational laws. Such laws typically follow a pattern where a base case is later refined by many exceptions in following law articles, a pattern that can be formally modeled using default logic. We show how to handle default logic inside a concolic execution tool, and implement our approach in the context of Catala, a recent domain-specific language tailored to implement computational laws. We evaluate CUTECat on several programs, including the Catala implementation of the French housing benefits and Section 132 of the US tax code. We show that CUTECat can successfully generate hundreds of thousands of testcases covering all branches of these bodies of law. Through several heuristics, we improve CUTECat's scalability and usability, making the testcases understandable by lawyers and programmers alike. We believe CUTECat thus paves the way for the use of formal methods during legislative processes.

cs.PL↗

arXiv2024

Sound Borrow-Checking for Rust via Symbolic Semantics (Long Version)

The Rust programming language continues to rise in popularity, and as such, warrants the close attention of the programming languages community. In this work, we present a new foundational contribution towards the theoretical understanding of Rust's semantics. We prove that LLBC, a high-level, borrow-centric model previously proposed for Rust's semantics and execution, is sound with regards to a low-level pointer-based language à la CompCert. Specifically, we prove the following: that LLBC is a correct view over a traditional model of execution; that LLBC's symbolic semantics are a correct abstraction of LLBC programs; and that LLBC's symbolic semantics act as a borrow-checker for LLBC, i.e. that symbolically-checked LLBC programs do not get stuck when executed on a heap-and-addresses model of execution. To prove these results, we introduce a new proof style that considerably simplifies our proofs of simulation, which relies on a notion of hybrid states. Equipped with this reasoning framework, we show that a new addition to LLBC's symbolic semantics, namely a join operation, preserves the abstraction and borrow-checking properties. This in turn allows us to add support for loops to the Aeneas framework; we show, using a series of examples and case studies, that this unlocks new expressive power for Aeneas.

cs.PL↗

arXiv2024

StarMalloc: A Formally Verified, Concurrent, Performant, and Security-Oriented Memory Allocator

In this work, we present StarMalloc, a verified, security-oriented, concurrent memory allocator that can be used as a drop-in replacement in real-world projects. Using the Steel separation logic framework, we show how to specify and verify StarMalloc, relying on dependent types and modular abstractions to enable efficient verification. As part of StarMalloc, we also develop several generic datastructures and proof libraries directly reusable in future systems verification projects. We finally show that StarMalloc can be used with real-world projects, including the Firefox browser, and evaluate it against 10 state-of-the-art memory allocators, demonstrating its competitiveness.

cs.PL↗

arXiv2024

Formalizing Date Arithmetic and Statically Detecting Ambiguities for the Law

Legal expert systems routinely rely on date computations to determine the eligibility of a citizen to social benefits or whether an application has been filed on time. Unfortunately, date arithmetic exhibits many corner cases, which are handled differently from one library to the other, making faithfully transcribing the law into code error-prone, and possibly leading to heavy financial and legal consequences for users. In this work, we aim to provide a solid foundation for date arithmetic working on days, months and years. We first present a novel, formal semantics for date computations, and formally establish several semantic properties through a mechanization in the F* proof assistant. Building upon this semantics, we then propose a static analysis by abstract interpretation to automatically detect ambiguities in date computations. We finally integrate our approach in the Catala language, a recent domain-specific language for formalizing computational law, and use it to analyze the Catala implementation of the French housing benefits, leading to the discovery of several date-related ambiguities.

cs.PL↗

arXiv2023

Modularity, Code Specialization, and Zero-Cost Abstractions for Program Verification

For all the successes in verifying low-level, efficient, security-critical code, little has been said or studied about the structure, architecture and engineering of such large-scale proof developments. We present the design, implementation and evaluation of a set of language-based techniques that allow the programmer to modularly write and verify code at a high level of abstraction, while retaining control over the compilation process and producing high-quality, zero-overhead, low-level code suitable for integration into mainstream software. We implement our techniques within the F* proof assistant, and specifically its shallowly-embedded Low* toolchain that compiles to C. Through our evaluation, we establish that our techniques were critical in scaling the popular HACL* library past 100,000 lines of verified source code, and brought about significant gains in proof engineer productivity. The exposition of our methodology converges on one final, novel case study: the streaming API, a finicky API that has historically caused many bugs in high-profile software. Using our approach, we manage to capture the streaming semantics in a generic way, and apply it ``for free'' to over a dozen use-cases. Six of those have made it into the reference implementation of the Python programming language, replacing the previous CVE-ridden code.

cs.PL↗

arXiv2022

Self-Correcting Neural Networks For Safe Classification

Classifiers learnt from data are increasingly being used as components in systems where safety is a critical concern. In this work, we present a formal notion of safety for classifiers via constraints called safe-ordering constraints. These constraints relate requirements on the order of the classes output by a classifier to conditions on its input, and are expressive enough to encode various interesting examples of classifier safety specifications from the literature. For classifiers implemented using neural networks, we also present a run-time mechanism for the enforcement of safe-ordering constraints. Our approach is based on a self-correcting layer, which provably yields safe outputs regardless of the characteristics of the classifier input. We compose this layer with an existing neural network classifier to construct a self-correcting network (SC-Net), and show that in addition to providing safe outputs, the SC-Net is guaranteed to preserve the classification accuracy of the original network whenever possible. Our approach is independent of the size and architecture of the neural network used for classification, depending only on the specified property and the dimension of the network's output; thus it is scalable to large state-of-the-art networks. We show that our approach can be optimized for a GPU, introducing run-time overhead of less than 1ms on current hardware -- even on large, widely-used networks containing hundreds of thousands of neurons and millions of parameters.

cs.LG↗

arXiv2021

SteelCore: An Extensible Concurrent Separation Logic for Effectful Dependently Typed Programs

Much recent research has been devoted to modeling effects within type theory. Building on this work, we observe that effectful type theories can provide a foundation on which to build semantics for more complex programming constructs and program logics, extending the reasoning principles that apply within the host effectful type theory itself. Concretely, our main contribution is a semantics for concurrent separation logic (CSL) within the F* proof assistant in a manner that enables dependently typed, effectful F* programs to make use of concurrency and to be specified and verified using a full-featured, extensible CSL. In contrast to prior approaches, we directly derive the partial-correctness Hoare rules for CSL from the denotation of computations in the effectful semantics of non-deterministically interleaved atomic actions. Demonstrating the flexibility of our semantics, we build generic, verified libraries that support various concurrency constructs, ranging from dynamically allocated, storable spin locks, to protocol-indexed channels. We conclude that our effectful semantics provides a simple yet expressive basis on which to layer domain-specific languages and logics for verified, concurrent programming.

cs.PL↗

arXiv2021

Fast Geometric Projections for Local Robustness Certification

Local robustness ensures that a model classifies all inputs within an $\ell_2$-ball consistently, which precludes various forms of adversarial inputs. In this paper, we present a fast procedure for checking local robustness in feed-forward neural networks with piecewise-linear activation functions. Such networks partition the input space into a set of convex polyhedral regions in which the network's behavior is linear; hence, a systematic search for decision boundaries within the regions around a given input is sufficient for assessing robustness. Crucially, we show how the regions around a point can be analyzed using simple geometric projections, thus admitting an efficient, highly-parallel GPU implementation that excels particularly for the $\ell_2$ norm, where previous work has been less effective. Empirically we find this approach to be far more precise than many approximate verification approaches, while at the same time performing multiple orders of magnitude faster than complete verifiers, and scaling to much deeper networks.

cs.LG↗