About the journal
Browse
Collections
Multimedia collections
Authors & reviewers
PinChecker++: Detection of Unsound Pin API Abstractions in Rust via Efficient Constraint-Based Program Synthesis
Yuxuan DAI , Yang FENG
Rust’s pinning APIs ensure memory location stability for self-referential and asynchronous constructs, as long as the pinning contract is followed. While Rust prevents contract violations in safe code, library authors may encapsulate unsafe pinning operations within ostensibly safe functions, potentially introducing soundness bugs. Detecting such unsound abstractions requires synthesizing programs that trigger violations by chaining library calls, but existing generation tools lack understanding of pinning semantics. We present PinChecker++, a constraint-based program synthesis tool for detecting unsound safe abstractions of Rust pinning APIs. PinChecker++ leverages RPIL, a lightweight intermediate representation modeling reference provenance and pinning state transitions, and encodes the synthesis problem as Boolean satisfiability constraints. Relative to our prior Prolog-based implementation, the journal extension reformulates synthesis as SAT constraints and adds wildcard types and function-level effect summaries to control constraint scale. The evaluation reproduces two Pin-contract violations in 13 Rust libraries and separately measures performance and optimization effects at target length 8; these bug-finding and performance results are reported with distinct evidence boundaries.
rust / static analysis / program synthesis / constraint solving / software reliability
Higher Education Press 2026
/
| 〈 |
|
〉 |