Publication Type

Journal Article

Version

publishedVersion

Publication Date

4-2025

Abstract

The concept of Deep Learning (DL) compiler was proposed to deploy DL models more efficiently on diverse hardware through optimization techniques. As one of the most popular DL compilers, TVM incorporates three levels (high-level, schedule, and low-level) of optimizations, which can inadvertently introduce code logic bugs and build failure bugs. Among these optimizations, scheduling optimization is the core component of DL compilers, which ensures the acceleration of models on all devices. However, the existing works only focus on the testing of high-level and low-level optimizations in TVM, fail to take the most important and challenging intermediate scheduling optimization layer into consideration.To fill the gap, we propose a Scheduling Optimization Oriented Fuzzer (Scuzer) for TVM, which is specially designed to effectively detect bugs introduced by the scheduling optimization. In particular, Scuzer first proposes a set of schedule-triggering mutators to actively trigger many scheduling optimizations. Meanwhile, observing that scheduling optimization is closely coupled with program dataflow and operator type, Scuzer additionally proposes a set of structure-enriching mutators to enrich the structure of dataflows and operators. Based on these carefully designed mutators, Scuzer then devises a multi-objective algorithm that can adaptively select different combinations of objectives at each period to guide the selection of seeds and mutators during fuzzing. We conduct extensive experiments comparing with three state-of-the-art fuzzers that can be applied in testing scheduling optimization to evaluate the effectiveness of Scuzer. The experimental results demonstrate that Scuzer outperforms the 2nd-best state-of-the-art fuzzer by 7.4% in edge coverage and achieves 7 improvement in rule-operator coverage. Scuzer has successfully detected 17 previously unknown bugs (9 are inconsistent results and 5 are inconsistent compilations) in TVM, out of which 10 have been confirmed and 5 been fixed.

Keywords

Deep Learning Compiler, Fuzzing, Scheduling Optimization, TVM

Discipline

Software Engineering

Research Areas

Software and Cyber-Physical Systems

Publication

ACM Transactions on Software Engineering and Methodology

Volume

34

Issue

4

First Page

1

Last Page

28

ISSN

1049-331X

Identifier

10.1145/3705308

Publisher

Association for Computing Machinery (ACM)

Copyright Owner and License

Authors-CC-BY

Creative Commons License

Creative Commons Attribution 3.0 License
This work is licensed under a Creative Commons Attribution 3.0 License.

Additional URL

https://doi.org/10.1145/370530

Download

Share

COinS