Abstract
Shape analysis algorithms statically infer deep properties of the runtime heap, such as whether a variable points to a cyclic or acyclic linked list. Previous shape analyses have tended to avoid features of low-level programming languages, such as memory disposal and pointer arithmetic. Yet, these features are used in many important programs, particularly systems programs.
In this talk I will describe how shape analysis for low-level code can be done with separation logic. A crucial element of the approach is the way it negotiates a transit between a low-level RAM view of memory and a higher, fictional, view that abstracts from the representation of nodes and linked structures as certain configurations of the RAM. The analysis algorithm can be seen as conducting a proof search in separation logic, and I will show how this provides a flexible way of exploring non-standard optimizations, while maintaining soundness.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yang, H. (2006). Shape Analysis for Low-Level Code. In: Yi, K. (eds) Static Analysis. SAS 2006. Lecture Notes in Computer Science, vol 4134. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11823230_18
Download citation
DOI: https://doi.org/10.1007/11823230_18
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-37756-6
Online ISBN: 978-3-540-37758-0
eBook Packages: Computer ScienceComputer Science (R0)