Many-core compiler fuzzing

C Lidbury, A Lascu, N Chong, AF Donaldson - ACM SIGPLAN Notices, 2015 - dl.acm.org
… We address the compiler correctness problem for many-core systems through novel applications
of fuzz testing to OpenCL compilers. Focusing on two methods from prior work, random …

Compiler fuzzing: How much does it matter?

M Marcozzi, Q Tang, AF Donaldson… - Proceedings of the ACM on …, 2019 - dl.acm.org
… The manual analysis of a selection of the binary-level differences caused by some of our
compiler bugs, both fuzzer-found and non fuzzer-found, shows that either these binary …

Cudasmith: A fuzzer for cuda compilers

B Jiang, X Wang, WK Chan, TH Tse, N Li… - 2020 IEEE 44th …, 2020 - ieeexplore.ieee.org
… 4) Atomic mode: The atomic functions are important instruments for threads to synchronize
with one another for manycore computing. Hence, the atomic mode of CUDAsmith aims at …

Compiler fuzzing through deep learning

C Cummins, P Petoumenos, A Murray… - Proceedings of the 27th …, 2018 - dl.acm.org
… We present a novel framework for compiler fuzzing. By posing the generation of random …
We demonstrated our approach by targeting the challenging many-core domain of OpenCL. Our …

A survey of modern compiler fuzzing

H Ma - arXiv preprint arXiv:2306.06884, 2023 - arxiv.org
… bug", "compiler vulnerabilit(y/ies)", "compiler fault". … compiler fuzzing", we remove all
compiler fuzzing work for old-fashioned compilers, such as the Fortran compiler, the Ada compiler

GrayC: Greybox fuzzing of compilers and analysers for C

K Even-Mendoza, A Sharma, AF Donaldson… - Proceedings of the 32nd …, 2023 - dl.acm.org
… Lacking feedback, the fuzzers have no way of adapting their generation strategy to find …
compiler fuzzing techniques that yield valid programs capable of detecting deep compiler bugs, …

Fuzzing MLIR Compiler Infrastructure via Operation Dependency Analysis

C Suo, J Chen, S Liu, J Jiang, Y Zhao… - Proceedings of the 33rd …, 2024 - dl.acm.org
… a novel fuzzing technique for the MLIR compiler infrastructure, called MLIRod (MLIR fuzzing
… We propose a novel fuzzing technique for the MLIR compiler infrastructure, called MLIRod, …

A Generative and Mutational Approach for Synthesizing Bug-Exposing Test Cases to Guide Compiler Fuzzing

G Ye, T Hu, Z Tang, Z Fan, SH Tan, B Zhang… - Proceedings of the 31st …, 2023 - dl.acm.org
… e ective strategy for compiler fuzzing should direct fuzzing e orts toward … compiler fuzzing
framework that combines generative and mutational techniques while improving the compiler

Fuzzing loop optimizations in compilers for C++ and data-parallel languages

V Livinskii, D Babokin, J Regehr - Proceedings of the ACM on …, 2023 - dl.acm.org
Compilers are part of the foundation upon which software systems are built; they need to be
… generative compiler fuzzer. This new version has found 122 bugs, both in compilers for data-…

Fuzzing deep learning compilers with hirgen

H Ma, Q Shen, Y Tian, J Chen, SC Cheung - Proceedings of the 32nd …, 2023 - dl.acm.org
… To bridge the gap, we propose the rst DL compiler fuzzing technique that focuses on high-level
optimization, HirGen. HirGen is designed to satisfy the following four objectives: 1) the …