Coverage-Guided Tensor Compiler Fuzzing with Joint IR-Pass Mutation

Title: Coverage-Guided Tensor Compiler Fuzzing with Joint IR-Pass Mutation
Presenter: Jiawei Liu (UIUC)

Abstract: In the past decade, Deep Learning (DL) systems have been widely deployed in various application domains to facilitate our daily life, e.g., natural language processing, healthcare, activity recognition, and autonomous driving. Meanwhile, it is extremely challenging to ensure the correctness of DL systems (e.g., due to their intrinsic nondeterminism), and bugs in DL systems can cause serious consequences and may even threaten human lives. In the literature, researchers have explored various techniques to test, analyze, and verify DL models, since their quality directly affects the corresponding system behaviors. Recently, researchers have also proposed novel techniques for testing the underlying operator-level DL libraries, which provide general binary implementations for each high-level DL operator and are the foundation for running DL models on different hardware platforms. However, there is still limited work targeting the reliability of the emerging tensor compilers (also known as DL compilers), which aim to automatically compile high-level tensor computation graphs directly into high-performance binaries for better efficiency, portability, and scalability than traditional operator-level libraries.
In this talk, I'll introduce Tzer, a practical fuzzing technique for the widely used TVM tensor compiler. Tzer focuses on mutating the low-level Intermediate Representation (IR) for TVM due to the limited mutation space for the high-level IR. Our experimental results show that Tzer substantially outperforms existing fuzzing techniques on tensor compiler testing. To date, Tzer has detected 49 previously unknown bugs for TVM, with 37 bugs confirmed and 25 bugs fixed (PR merged).

Bio: Jiawei is a first-year CS PhD student at UIUC advised by Lingming Zhang. His primary research goal is to make future software infrastructures: easy-to-use, high-performance and reliable. At present, He is developing PLSE techniques to make ML Systems reliable and efficient.

-
SAMPL is an interdisciplinary machine learning research group exploring problems spanning multiple layers of the system stack including deep learning frameworks, specialized hardware for training and inference, new intermediate representations, differentiable programming, and various applications. We are part of the Paul G. Allen School of Computer Science & Engineering at the University of Washington. Our group is a collaboration between researchers from Sampa, Syslab, MODE, and PLSE.

Twitter:   / ai_sampl  
SAMPL website: https://sampl.cs.washington.edu/