Output nondeterminism detection for programming models combining dataflow with shared memory

•We identify output nondeterminism errors in dataflow applications on shared memory.•A technique for detecting output nondeterminism based on happens-before relation is proposed.•A tool using this technique detects real and synthesized bugs in dataflow programs.•Extensive evaluation of the tool usin...

Full description

Saved in:
Bibliographic Details
Published in:Parallel computing Vol. 71; pp. 42 - 57
Main Authors: Matar, Hassan Salehe, Mutlu, Erdal, Tasiran, Serdar, Unat, Didem
Format: Journal Article
Language:English
Published: Elsevier B.V 01.01.2018
Subjects:
Data races
Dataflow models
Happens-before
LLVM
Nondeterminism detection
Shared memory
Nondeterminism detection
Data races
LLVM
Happens-before
Shared memory
Dataflow models
ISSN:0167-8191, 1872-7336
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•We identify output nondeterminism errors in dataflow applications on shared memory.•A technique for detecting output nondeterminism based on happens-before relation is proposed.•A tool using this technique detects real and synthesized bugs in dataflow programs.•Extensive evaluation of the tool using real and synthesized bugs is presented.•A method to detect commuting tasks that result in false alarms is proposed. Implementing highly concurrent programs can be challenging because programmers can easily introduce unintended nondeterminism, which has the potential to affect the program output. We propose and implement a technique for detecting unintended nondeterminism in applications developed on shared memory systems with dataflow execution model. Such nondeterminism bugs may be caused by missing or incorrect ordering of task dependencies that are used for ensuring certain ordering of tasks. The proposed method is based on the formulation of happens-before relation on tasks executions in a dataflow dependency graph. Its implementation is composed of two main phases; log recording and detection. For recording the necessary information from the execution, the tool instruments the dataflow framework and the applications, on top of the LLVM compiler infrastructure. Later it processes the collected log and reports on the found output nondeterminism in the execution. The tool can integrate well with the development cycle to provide the programmer with a testing framework against possible nondeterminism bugs. To demonstrate its effectiveness, we study a set of benchmark applications written in Atomic DataFlow programming model and report on real nondeterminism bugs in them.
ISSN:0167-8191
1872-7336
DOI:10.1016/j.parco.2017.11.008