Filters








396 Hits in 4.1 sec

Section based program analysis to reduce overhead of detecting unsynchronized thread communication

Madan Das, Gabriel Southern, Jose Renau
2015 Proceedings of the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming - PPoPP 2015  
We propose Section-Based Program Analysis (SBPA), a novel way to decompose the program into disjoint code sections to identify and eliminate instrumenting such loads and stores during program compilation  ...  , so that the program runtime overhead is significantly reduced.  ...  Opinions, findings and conclusions expressed herein are those of the authors and do not reflect the views of the NSF.  ... 
doi:10.1145/2688500.2688552 dblp:conf/ppopp/DasSR15 fatcat:viiast7cpzeqbmoaefhzcie4ia

Instrumentation bias for dynamic data race detection

Benjamin P. Wood, Man Cao, Michael D. Bond, Dan Grossman
2017 Proceedings of the ACM on Programming Languages  
Overall, FIB e�ectively exploits latent analysis and program invariants to bring strong integrity guarantees to an otherwise unsynchronized data race detection algorithm at minimal cost. least one access  ...  This paper presents Fast Instrumentation Bias (FIB), a sound and complete dynamic data race detection algorithm that improves performance by reducing or eliminating the costs of analysis atomicity.  ...  We thank the anonymous reviewers for insightful and helpful feedback on earlier versions of this paper that led to several improvements.  ... 
doi:10.1145/3133893 dblp:journals/pacmpl/WoodCBG17 fatcat:oeiwdxc4abbjxhxu4zeb55gqzq

Accelerating Data Race Detection with Minimal Hardware Support [chapter]

Rodrigo Gonzalez-Alberquilla, Karin Strauss, Luis Ceze, Luis Piñuel
2011 Lecture Notes in Computer Science  
This approach is capable of detecting virtually all data races detected by a traditional happened-before data race detection approach, but at significantly lower space and performance overhead.  ...  To leverage this support, we propose a new algorithm for race detection. This detection algorithm uses StateChk to eliminate many expensive operations.  ...  Note that all state necessary to the analysis is local to a thread, so no inter-thread communication is required; HapB, in contrast, requires substantial inter-thread communication.  ... 
doi:10.1007/978-3-642-23400-2_4 fatcat:i7yl4eldibfsrhb4e7fe7wrgda

Effective pattern-driven concurrency bug detection for operating systems

Shin Hong, Moonzoo Kim
2013 Journal of Systems and Software  
However, current analysis techniques and tools for validating concurrent programs often fail to detect concurrency bugs in operating systems (OSes) due to the complex characteristics of OSes.  ...  To detect concurrency bugs in OSes in a practical manner, we have developed the COncurrency Bug dETector (COBET) framework based on composite bug patterns augmented with semantic conditions.  ...  Since COBET is very fast to apply bug patterns to large program code (seeTables 3-5), a large number of bug patterns may not cause much overhead to detect concurrency bugs.  ... 
doi:10.1016/j.jss.2012.08.063 fatcat:l36lcjqw4fferk52ut54qryohq

Static Analyses for Eliminating Unnecessary Synchronization from Java Programs [chapter]

Jonathan Aldrich, Craig Chambers, Emin Gun Sirer, Susan Eggers
1999 Lecture Notes in Computer Science  
This paper presents and evaluates a set of analyses designed to reduce synchronization overhead in Java programs.  ...  To reduce this overhead, programmers often try to eliminate unnecessary lock operations by hand.  ...  We appreciate feedback and pointers to related work from David Grove, Martin Rinard, William Chan, Satoshi Matsuoka, members of the Vortex group, and the anonymous reviewers.  ... 
doi:10.1007/3-540-48294-6_2 fatcat:d572bk4deffr7j26hipwunzyty

Predictive Data Race Detection for GPUs [article]

Sagnik Dey, Mayant Mukul, Parth Sharma, Swarnendu Biswas
2021 arXiv   pre-print
Existing predictive partial orders cannot be na\"ively applied to analyze GPU kernels because of the differences in programming models.  ...  This work proposes GWCP, a predictive partial order for data race detection of GPU kernels.  ...  This material is based upon work supported by IITK Initiation Grant and Science and Engineering Research Board under Grant SRG/2019/000384.  ... 
arXiv:2111.12478v1 fatcat:eygepkuv2jddfpiio3p65kcag4

Efficient Deterministic Replay of Multithreaded Executions in a Managed Language Virtual Machine

Michael D. Bond, Milind Kulkarni, Man Cao, Meisam Fathi Salmi, Jipeng Huang
2015 Proceedings of the Principles and Practices of Programming on The Java Platform - PPPJ '15  
Shared-memory parallel programs are inherently nondeterministic, making it difficult to diagnose rare bugs and to achieve deterministic execution.  ...  Existing multithreaded record & replay approaches have serious limitations such as relying on custom hardware, handling only data-race-free executions, or slowing programs by an order of magnitude.  ...  Thanks to the anonymous reviewers for insightful suggestions on the text.  ... 
doi:10.1145/2807426.2807434 dblp:conf/pppj/BondKCSH15 fatcat:4mqge2uucbe3vkmjdphlrx7mpu

Demand-driven software race detection using hardware performance counters

Joseph L. Greathouse, Zhiqiang Ma, Matthew I. Frank, Ramesh Peri, Todd Austin
2011 SIGARCH Computer Architecture News  
When little sharing takes place, this demand-driven analysis is much faster than contemporary continuous-analysis tools without a large loss of detection accuracy.  ...  While this method of bug hunting can find races that are normally difficult to observe, it also suffers from high runtime overheads.  ...  ACKNOWLEDGMENTS We wish to thank Matt Braun and the rest of the Inspector XE team for their help and insights throughout this project.  ... 
doi:10.1145/2024723.2000084 fatcat:pokfk77j3vgvvc2ywthuatdhqm

Demand-driven software race detection using hardware performance counters

Joseph L. Greathouse, Zhiqiang Ma, Matthew I. Frank, Ramesh Peri, Todd Austin
2011 Proceeding of the 38th annual international symposium on Computer architecture - ISCA '11  
When little sharing takes place, this demand-driven analysis is much faster than contemporary continuous-analysis tools without a large loss of detection accuracy.  ...  While this method of bug hunting can find races that are normally difficult to observe, it also suffers from high runtime overheads.  ...  ACKNOWLEDGMENTS We wish to thank Matt Braun and the rest of the Inspector XE team for their help and insights throughout this project.  ... 
doi:10.1145/2000064.2000084 dblp:conf/isca/GreathouseMFPA11 fatcat:ktlxd3qsqrbqhapvgvhepq7jpa

Hybrid dynamic data race detection

Robert O'Callahan, Jong-Deok Choi
2003 Proceedings of the ninth ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '03  
We present a new method for dynamically detecting potential data races in multithreaded programs. Our method improves on the state of the art in accuracy, in usability, and in overhead.  ...  We reduce overhead compared to previous detectors -particularly for large applications such as Web application servers -by not relying on happens-before detection alone, by introducing a new optimization  ...  Acknowledgements We would like to thank Dan Pelleg and the CMU zephyr community for identifying the minimum hitting set problem, Darko Marinov for helpful comments on the paper, and Mark Christiaens and  ... 
doi:10.1145/781498.781528 dblp:conf/ppopp/OCallahanC03 fatcat:aiillvjvzfazhbmbi5735k4jx4

Hybrid dynamic data race detection

Robert O'Callahan, Jong-Deok Choi
2003 SIGPLAN notices  
We present a new method for dynamically detecting potential data races in multithreaded programs. Our method improves on the state of the art in accuracy, in usability, and in overhead.  ...  We reduce overhead compared to previous detectors -particularly for large applications such as Web application servers -by not relying on happens-before detection alone, by introducing a new optimization  ...  Acknowledgements We would like to thank Dan Pelleg and the CMU zephyr community for identifying the minimum hitting set problem, Darko Marinov for helpful comments on the paper, and Mark Christiaens and  ... 
doi:10.1145/966049.781528 fatcat:kmoixegv3nfi7lt7fli35si7zq

Hybrid dynamic data race detection

Robert O'Callahan, Jong-Deok Choi
2003 Proceedings of the ninth ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '03  
We present a new method for dynamically detecting potential data races in multithreaded programs. Our method improves on the state of the art in accuracy, in usability, and in overhead.  ...  We reduce overhead compared to previous detectors -particularly for large applications such as Web application servers -by not relying on happens-before detection alone, by introducing a new optimization  ...  Acknowledgements We would like to thank Dan Pelleg and the CMU zephyr community for identifying the minimum hitting set problem, Darko Marinov for helpful comments on the paper, and Mark Christiaens and  ... 
doi:10.1145/781527.781528 fatcat:2r5riqot5zeejkpwybqd62m3zi

Comparative Analysis of Dynamic Data Race Detection Techniques [article]

Danial Entezari
2022 arXiv   pre-print
This paper also reveals the extent to which dynamic data race detection (also called dynamic analysis) can identify harmful data races, how it can be implemented, and how it compares to other forms of  ...  threads are not using any exclusive locks to control their accesses to that memory.  ...  The works cited in this paper generally provide performance results and in-depth analysis of the tools they have presented.  ... 
arXiv:2206.10338v1 fatcat:ysygp66edjaf3mxuombfrqt22m

Ready, set, Go!

Daniel Schnetzer Fava, Martin Steffen
2020 Science of Computer Programming  
different threads.  ...  In this paper, we present a data-race detector for a language with channel communication as its sole synchronization primitive, and provide a semantics directly tied to the happens-before relation, thus  ...  In order to further reduce the memory foot-print of modern race detection implementations, we are thus left with devising approaches to managing threads' and locks' memory overhead.  ... 
doi:10.1016/j.scico.2020.102473 fatcat:oeaczq5tzzfw5egmdnpworxppq

Ready, set, Go! Data-race detection and the Go language [article]

Daniel Schnetzer Fava, Martin Steffen
2020 arXiv   pre-print
from different threads.  ...  Data races are often discussed in the context of lock acquisition and release, with race-detection algorithms routinely relying on vector clocks as a means of capturing the relative ordering of events  ...  In order to further reduce the memory foot-print of modern race detection implementations, we are thus left with devising approaches to managing threads' and locks' memory overhead.  ... 
arXiv:1910.12643v3 fatcat:ski5kjuj4vhxtpeamvjgsh5jvy
« Previous Showing results 1 — 15 out of 396 results