Points-to analysis using BDDs

Berndl, Marc; Lhoták, Ondřej; Qian, Feng; Hendren, Laurie; Umanee, Navindra

doi:10.1145/781131.781144

Cited by 157 publications

(141 citation statements)

References 30 publications

Supporting

Mentioning

141

Contrasting

Order By: Relevance

“…The literature contains many variations of points-to analysis: context-sensitive versus context-insensitive, flow-sensitive versus flow-insensitive, etc. [3,6,12,13,15,33,35]. These variations make different trade-offs between precision and running time, but production compilers like gcc and LLVM seem to have settled on context-insensitive, flow-insensitive points-to analysis because the more precise alternatives are currently intractable for very large programs.…”

Section: Inclusion-based Points-to Analysismentioning

confidence: 99%

“…Another important difference is the use of a Binary Decision Diagram [8] data structure. The benefits of BDDs in the context of points-to analysis have been touted by many researchers [6,35]. The reference implementation uses a BDD to compactly represent the points-to edges, while all the other types of edges are internally represented using sparse bit vectors.…”

Section: Experimental Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

A GPU implementation of inclusion-based points-to analysis

Méndez-Lojo

Burtscher

Pingali

2012

Proceedings of the 17th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

Graphics Processing Units (GPUs) have emerged as powerful accelerators for many regular algorithms that operate on dense arrays and matrices. In contrast, we know relatively little about using GPUs to accelerate highly irregular algorithms that operate on pointer-based data structures such as graphs. For the most part, research has focused on GPU implementations of graph analysis algorithms that do not modify the structure of the graph, such as algorithms for breadth-first search and strongly-connected components.In this paper, we describe a high-performance GPU implementation of an important graph algorithm used in compilers such as gcc and LLVM: Andersen-style inclusion-based points-to analysis. This algorithm is challenging to parallelize effectively on GPUs because it makes extensive modifications to the structure of the underlying graph and performs relatively little computation. In spite of this, our program, when executed on a 14 Streaming Multiprocessor GPU, achieves an average speedup of 7x compared to a sequential CPU implementation and outperforms a parallel implementation of the same algorithm running on 16 CPU cores.Our implementation provides general insights into how to produce high-performance GPU implementations of graph algorithms, and it highlights key differences between optimizing parallel programs for multicore CPUs and for GPUs.

show abstract

Section: Inclusion-based Points-to Analysismentioning

confidence: 99%

Section: Experimental Evaluationmentioning

confidence: 99%

A GPU implementation of inclusion-based points-to analysis

Méndez-Lojo

Burtscher

Pingali

2012

Proceedings of the 17th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

show abstract

“…The costs of the analysis generally include time and memory consumptions. Recent advances in points-to analysis research have concentrated on the improvement of the analysis precision without the increase of memory costs [22,4]. In what follows we point out the main dimensions proposed by Ryder [69] that determine the relative precision and cost of analysis and that are relevant to our points-to analysis presentation:…”

Section: Precision and Costsmentioning

confidence: 99%

“…A similar approach is presented by Lhoták's and Hendren's Paddle framework [4] that integrates BDD-based analysis into the Java optimization framework Soot. Paddle is implemented based on the Jedd language [44], an extension of Java for expressing program-analysis in terms of relations using BDDs to store and manipulate these relations.…”

Section: Points-to Analysismentioning

confidence: 99%

Decisions: Algebra and Implementation

Danylenko

Lundberg

Löwe

2011

Machine Learning and Data Mining in Pattern Recognition

View full text Add to dashboard Cite

Processing decision information is a constitutive part in a number of applications in Computer Science fields. In general, decision information can be used to deduce the relationship between a certain context and a certain decision. Decision information is represented by a decision model that captures this information. Frequently used examples of decision models are decision tables and decision trees. The choice of an appropriate decision model has an impact on application performance in terms of memory consumption and execution time. High memory expenses can possibly occur due to redundancy in a decision model; and high execution time is often a consequence of an unsuitable decision model. Applications in different domains try to overcome these problems by introducing new data structures or algorithms for implementing decision models. These solutions are usually domain-specific and hard to transfer from one domain to another.Different application domains of Computer Science often process decision information in a similar way and, hence, have similar problems. We should thus be able to present a unifying approach that can be applicable in all application domains for capturing and manipulating decision information. Therefore, the goal of this thesis is (i) to suggest a general structure (Decision Algebra) which provides a common theoretical framework that captures decision information and defines operations (signatures) for storing, accessing, merging, approximating, and manipulating such information along with some general algebraic laws regardless of the used implementation. Our Decision Algebra allows defining different construction strategies for decision models and data structures that capture decision information as implementation variants, and it simplifies experimental comparisons between them.Additionally, this thesis presents (ii) an implementation of Decision Algebra capturing the information in a non-redundant way and performing the operations efficiently. In fact, we show that existing decision models that originated in the field of Data Mining and Machine Learning and variants thereof as exploited in special algorithms can be understood as alternative implementation variants of the Decision Algebra by varying the implementations of the Decision Algebra operations. Hence, this work (iii) will contribute to a classification of existing technology for processing decision information in different application domains of Computer Science.

show abstract

“…The experiments in [16] involved small set sharings (of at most 50 elements at a time) while in this paper we show how with ZBDDs we can scale up to thousands of sharings and still get reasonable times. There has been extensive work in recent years on the use of BDDs [24,2,22,25] to represent (abstract) points-to information. In these abstractions, information is stored in the form of (v, a) pairs, where each such pair indicates that v may point to the allocation site a.…”

Section: Related Workmentioning

confidence: 99%

Efficient Set Sharing Using ZBDDs

Méndez-Lojo

Lhoták

Hermenegildo

2008

Languages and Compilers for Parallel Computing

View full text Add to dashboard Cite

Abstract. Set sharing is an abstract domain in which each concrete object is represented by the set of local variables from which it might be reachable. It is a useful abstraction to detect parallelism opportunities, since it contains definite information about which variables do not share in memory, i.e., about when the memory regions reachable from those variables are disjoint. Set sharing is a more precise alternative to pair sharing, in which each domain element is a set of all pairs of local variables from which a common object may be reachable. However, the exponential complexity of some set sharing operations has limited its wider application. This work introduces an efficient implementation of the set sharing domain using Zero-supressed Binary Decision Diagrams (ZBDDs). Because ZBDDs were designed to represent sets of combinations (i.e., sets of sets), they naturally represent elements of the set sharing domain. We show how to synthesize the operations needed in the set sharing transfer functions from basic ZBDD operations. For some of the operations, we devise custom ZBDD algorithms that perform better in practice. We also compare our implementation of the abstract domain with an efficient, compact, bitset-based alternative, and show that the ZBDD version scales better in terms of both memory usage and running time.

show abstract

Points-to analysis using BDDs

Cited by 157 publications

References 30 publications

A GPU implementation of inclusion-based points-to analysis

A GPU implementation of inclusion-based points-to analysis

Decisions: Algebra and Implementation

Efficient Set Sharing Using ZBDDs

Contact Info

Product

Resources

About