Shape analysis with structural invariant checkers

Lecture Notes in Computer Science, 2013

We present a general framework for verifying programs with complex dynamic linked data structures whose correctness depends on ordering relations between stored data values. The underlying formalism of our framework is that of forest automata (FA), which has previously been developed for verification of heap-manipulating programs. We extend FA by constraints between data elements associated with nodes of the heaps represented by FA, and we present extended versions of all operations needed for using the extended FA in a fullyautomated verification approach, based on abstract interpretation. We have implemented our approach as an extension of the Forester tool and successfully applied it to a number of programs dealing with data structures such as various forms of singly-and doubly-linked lists, binary search trees, as well as skip lists.

Programming Languages and Systems, 2018

A major challenge in automated verification is to develop techniques that are able to reason about fine-grained concurrent algorithms that consist of an unbounded number of concurrent threads, which operate on an unbounded domain of data values, and use unbounded dynamically allocated memory. Existing automated techniques consider the case where shared data is organized into singly-linked lists. We present a novel shape analysis for automated verification of fine-grained concurrent algorithms that can handle heap structures which are more complex than just singly-linked lists, in particular skip lists and arrays of singly linked lists, while at the same time handling an unbounded number of concurrent threads, an unbounded domain of data values (including timestamps), and an unbounded shared heap. Our technique is based on a novel shape abstraction, which represents a set of heaps by a set of fragments. A fragment is an abstraction of a pair of heap cells that are connected by a pointer field. We have implemented our approach and applied it to automatically verify correctness, in the sense of linearizability, of most linearizable concurrent implementations of sets, stacks, and queues, which employ singly-linked lists, skip lists, or arrays of singlylinked lists with timestamps, which are known to us in the literature.

Lecture Notes in Computer Science, 2010

Detailed memory models that expose individual fields are necessary to precisely analyze code that makes use of low-level aspects such as, pointers to fields and untagged unions. Yet, higher-level representations that collect fields into records are often used because they are typically more convenient and efficient in modeling the program heap. In this paper, we present a shape graph representation of memory that exposes individual fields while largely retaining the convenience of an object-level model. This representation has a close connection to particular kinds of formulas in separation logic. Then, with this representation, we show how to extend the Xisa shape analyzer for low-level aspects, including pointers to fields, C-style nested structures and unions, malloc and free, and array values, with minimal changes to the core algorithms (e.g., materialization and summarization).

Electronic Proceedings in Theoretical Computer Science, 2013

The aim of static analysis is to infer invariants about programs that are precise enough to establish semantic properties, such as the absence of run-time errors. Broadly speaking, there are two major branches of static analysis for imperative programs. Pointer and shape analyses focus on inferring properties of pointers, dynamically-allocated memory, and recursive data structures, while numeric analyses seek to derive invariants on numeric values. Although simultaneous inference of shapenumeric invariants is often needed, this case is especially challenging and is not particularly well explored. Notably, simultaneous shape-numeric inference raises complex issues in the design of the static analyzer itself. In this paper, we study the construction of such shape-numeric, static analyzers. We set up an abstract interpretation framework that allows us to reason about simultaneous shape-numeric properties by combining shape and numeric abstractions into a modular, expressive abstract domain. Such a modular structure is highly desirable to make its formalization and implementation easier to do and get correct. To achieve this, we choose a concrete semantics that can be abstracted step-by-step, while preserving a high level of expressiveness. The structure of abstract operations (i.e., transfer, join, and comparison) follows the structure of this semantics. The advantage of this construction is to divide the analyzer in modules and functors that implement abstractions of distinct features.

Lecture Notes in Computer Science, 2013

Real-world data structures are often enhanced with additional pointers capturing alternative paths through a basic inductive skeleton (e.g., back pointers, head pointers). From the static analysis point of view, we must obtain several interlocking shape invariants. At the same time, it is well understood in abstract interpretation design that supporting a separation of concerns is critically important to designing powerful static analyses. Such a separation of concerns is often obtained via a reduced product on a case-by-case basis. In this paper, we lift this idea to abstract domains for shape analyses, introducing a domain combination operator for memory abstractions. As an example, we present simultaneous separating shape graphs, a product construction that combines instances of separation logic-based shape domains. The key enabler for this construction is a static analysis on inductive data structure definitions to derive relations between the skeleton and the alternative paths. From the engineering standpoint, this construction allows each component to reason independently about different aspects of the data structure invariant and then separately exchange information via a reduction operator. From the usability standpoint, we enable describing a data structure invariant in terms of several inductive definitions that hold simultaneously.