Computer Science > Logic in Computer Science
[Submitted on 25 Jul 2023 (v1), last revised 19 Nov 2024 (this version, v4)]
Title:Affine Disjunctive Invariant Generation with Farkas' Lemma
View PDF HTML (experimental)Abstract:In the verification of loop programs, disjunctive invariants are essential to capture complex loop dynamics such as phase and mode changes. In this work, we develop a novel approach for the automated generation of affine disjunctive invariants for affine while loops via Farkas' Lemma, a fundamental theorem on linear inequalities. Our main contributions are two-fold. First, we combine Farkas' Lemma with a succinct control flow transformation to derive disjunctive invariants from the conditional branches in the loop. Second, we propose an invariant propagation technique that minimizes the invariant computation effort by propagating previously solved invariants to yet unsolved locations as much as possible. Furthermore, we resolve the infeasibility checking in the application of Farkas' Lemma which has not been addressed previously, and extend our approach to nested loops via loop summary. Experimental evaluation over more than 100 affine while loops (mostly from SV-COMP 2023) demonstrates that our approach is promising to generate tight linear invariants over affine programs.
Submission history
From: Jingyu Ke [view email][v1] Tue, 25 Jul 2023 08:26:05 UTC (155 KB)
[v2] Wed, 2 Aug 2023 13:10:32 UTC (155 KB)
[v3] Mon, 27 Nov 2023 16:54:57 UTC (326 KB)
[v4] Tue, 19 Nov 2024 11:59:34 UTC (278 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.