{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T09:36:45Z","timestamp":1761989805021,"version":"3.41.0"},"reference-count":61,"publisher":"Association for Computing Machinery (ACM)","issue":"FSE","funder":[{"name":"National Key Research and Development Program of China","award":["2023YFB3001804"],"award-info":[{"award-number":["2023YFB3001804"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["62372373, 62472351"],"award-info":[{"award-number":["62372373, 62472351"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Shaanxi Province Key Research and Development Project","award":["2024CY2-GJHX-60"],"award-info":[{"award-number":["2024CY2-GJHX-60"]}]},{"name":"China Postdoctoral Science Foundation Fellowship","award":["2022M712575"],"award-info":[{"award-number":["2022M712575"]}]},{"name":"Graduate Innovation Project of Northwest University","award":["CX2024199"],"award-info":[{"award-number":["CX2024199"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Softw. Eng."],"published-print":{"date-parts":[[2025,6,19]]},"abstract":"<jats:p>MLIR is a new way of creating compiler infrastructures that can be easily reused and extended. Current MLIR fuzzing methods focus primarily on test case generation or mutation using randomly selected passes. However, they often overlook the hierarchical structure of MLIR, resulting in inefficiencies in bug detection, especially for issues triggered by downstream dialects. Random testing lacks a focused approach to exploring the code space, resulting in wasted resources on normal components and overlooking bug-prone areas. To address these limitations, we introduce MLIRTracer, a top-down fuzzing approach that targets the highest level of MLIR programs (tosa IR) with a directed testing strategy. Our method systematically traverses the hierarchical code space of MLIR, from tosa IR to the lower levels, while prioritizing tests of bug-prone areas through directed exploration. MLIRTracer has successfully detected 73 bugs, with 61 already resolved by the MLIR developers.<\/jats:p>","DOI":"10.1145\/3729372","type":"journal-article","created":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T15:15:34Z","timestamp":1750346134000},"page":"2288-2310","source":"Crossref","is-referenced-by-count":1,"title":["Directed Testing in MLIR: Unleashing Its Potential by Overcoming the Limitations of Random Fuzzing"],"prefix":"10.1145","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9907-5859","authenticated-orcid":false,"given":"Weiyuan","family":"Tong","sequence":"first","affiliation":[{"name":"Northwest University, Xi'an, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2037-0127","authenticated-orcid":false,"given":"Zixu","family":"Wang","sequence":"additional","affiliation":[{"name":"Northwest University, Xi'an, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4333-2334","authenticated-orcid":false,"given":"Zhanyong","family":"Tang","sequence":"additional","affiliation":[{"name":"Northwest University, Xi'an, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3542-4869","authenticated-orcid":false,"given":"Jianbin","family":"Fang","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2239-6723","authenticated-orcid":false,"given":"Yuqun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Southern University of Science and Technology, Shenzhen, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2074-4253","authenticated-orcid":false,"given":"Guixin","family":"Ye","sequence":"additional","affiliation":[{"name":"Northwest University, Xi'an, China"}]}],"member":"320","published-online":{"date-parts":[[2025,6,19]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"d.]. ANTLR v4. https:\/\/github.com\/antlr\/antlr4 Accessed on","year":"2024","unstructured":"[n. d.]. ANTLR v4. https:\/\/github.com\/antlr\/antlr4 Accessed on September 2024"},{"key":"e_1_2_1_2_1","volume-title":"d.]. BTOR2MLIR: A Format and Toolchain for Hardware Verification. https:\/\/github.com\/jetafese\/btor2mlir Accessed on","year":"2024","unstructured":"[n. d.]. BTOR2MLIR: A Format and Toolchain for Hardware Verification. https:\/\/github.com\/jetafese\/btor2mlir Accessed on September 2024"},{"key":"e_1_2_1_3_1","volume-title":"d.]. Catalyst. https:\/\/github.com\/PennyLaneAI\/catalyst Accessed on","year":"2024","unstructured":"[n. d.]. Catalyst. https:\/\/github.com\/PennyLaneAI\/catalyst Accessed on September 2024"},{"key":"e_1_2_1_4_1","volume-title":"d.]. CIRCT:Circuit IR Compilers and Tools. https:\/\/github.com\/llvm\/circt Accessed on","year":"2024","unstructured":"[n. d.]. CIRCT:Circuit IR Compilers and Tools. https:\/\/github.com\/llvm\/circt Accessed on September 2024"},{"key":"e_1_2_1_5_1","volume-title":"d.]. Developer Guide. https:\/\/mlir.llvm.org\/getting_started\/DeveloperGuide\/ Accessed on","year":"2024","unstructured":"[n. d.]. Developer Guide. https:\/\/mlir.llvm.org\/getting_started\/DeveloperGuide\/ Accessed on September 2024"},{"key":"e_1_2_1_6_1","volume-title":"d.]. Dialects. https:\/\/mlir.llvm.org\/docs\/Dialects\/ Accessed on","year":"2024","unstructured":"[n. d.]. Dialects. https:\/\/mlir.llvm.org\/docs\/Dialects\/ Accessed on September 2024"},{"key":"e_1_2_1_7_1","volume-title":"d.]. IREE:Intermediate Representation Execution Environment. https:\/\/openxla.github.io\/iree\/ Accessed on","year":"2024","unstructured":"[n. d.]. IREE:Intermediate Representation Execution Environment. https:\/\/openxla.github.io\/iree\/ Accessed on September 2024"},{"key":"e_1_2_1_8_1","volume-title":"d.]. The LLVM Compiler Infrastructure. https:\/\/github.com\/llvm\/llvm-project Accessed on","year":"2024","unstructured":"[n. d.]. The LLVM Compiler Infrastructure. https:\/\/github.com\/llvm\/llvm-project Accessed on September 2024"},{"key":"e_1_2_1_9_1","volume-title":"d.]. llvm-cov. https:\/\/www.llvm.org\/docs\/CommandGuide\/llvm-cov.html Accessed on","year":"2024","unstructured":"[n. d.]. llvm-cov. https:\/\/www.llvm.org\/docs\/CommandGuide\/llvm-cov.html Accessed on September 2024"},{"key":"e_1_2_1_10_1","volume-title":"d.]. MLIR Code Repository. https:\/\/github.com\/llvm\/llvm-project\/tree\/main\/mlir Accessed on","year":"2024","unstructured":"[n. d.]. MLIR Code Repository. https:\/\/github.com\/llvm\/llvm-project\/tree\/main\/mlir Accessed on September 2024"},{"key":"e_1_2_1_11_1","volume-title":"d.]. MLIR Language Reference. https:\/\/mlir.llvm.org\/docs\/LangRef\/ Accessed on","year":"2024","unstructured":"[n. d.]. MLIR Language Reference. https:\/\/mlir.llvm.org\/docs\/LangRef\/ Accessed on September 2024"},{"key":"e_1_2_1_12_1","volume-title":"d.]. MLIR passes. https:\/\/mlir.llvm.org\/docs\/Passes\/ Accessed on","year":"2024","unstructured":"[n. d.]. MLIR passes. https:\/\/mlir.llvm.org\/docs\/Passes\/ Accessed on September 2024"},{"key":"e_1_2_1_13_1","volume-title":"d.]. ONNX-MLIR. https:\/\/onnx.ai\/onnx-mlir\/ Accessed on","year":"2024","unstructured":"[n. d.]. ONNX-MLIR. https:\/\/onnx.ai\/onnx-mlir\/ Accessed on September 2024"},{"key":"e_1_2_1_14_1","volume-title":"d.]. Pass Infrastructure. https:\/\/mlir.llvm.org\/docs\/PassManagement\/ Accessed on","year":"2024","unstructured":"[n. d.]. Pass Infrastructure. https:\/\/mlir.llvm.org\/docs\/PassManagement\/ Accessed on September 2024"},{"key":"e_1_2_1_15_1","volume-title":"d.]. Polyblocks. https:\/\/www.polymagelabs.com\/technology\/#polyblocks Accessed on","year":"2024","unstructured":"[n. d.]. Polyblocks. https:\/\/www.polymagelabs.com\/technology\/#polyblocks Accessed on September 2024"},{"key":"e_1_2_1_16_1","volume-title":"d.]. TableGen Overview. https:\/\/llvm.org\/docs\/TableGen\/ Accessed on","year":"2024","unstructured":"[n. d.]. TableGen Overview. https:\/\/llvm.org\/docs\/TableGen\/ Accessed on September 2024"},{"key":"e_1_2_1_17_1","volume-title":"d.]. TOSA specfication. https:\/\/github.com\/llvm\/llvm-project\/blob\/main\/mlir\/include\/mlir\/Dialect\/Tosa\/IR\/TosaOps.td Accessed on","year":"2024","unstructured":"[n. d.]. TOSA specfication. https:\/\/github.com\/llvm\/llvm-project\/blob\/main\/mlir\/include\/mlir\/Dialect\/Tosa\/IR\/TosaOps.td Accessed on September 2024"},{"key":"e_1_2_1_18_1","volume-title":"d.]. Triton. https:\/\/github.com\/triton-lang\/triton Accessed on","year":"2024","unstructured":"[n. d.]. Triton. https:\/\/github.com\/triton-lang\/triton Accessed on September 2024"},{"key":"e_1_2_1_19_1","volume-title":"d.]. ULingoDB - Revolutionizing Data Processing with Compiler Technology. https:\/\/www.lingo-db.com\/ Accessed on","year":"2024","unstructured":"[n. d.]. ULingoDB - Revolutionizing Data Processing with Compiler Technology. https:\/\/www.lingo-db.com\/ Accessed on September 2024"},{"key":"e_1_2_1_20_1","volume-title":"d.]. Users of MLIR. https:\/\/mlir.llvm.org\/users\/ Accessed on","year":"2024","unstructured":"[n. d.]. Users of MLIR. https:\/\/mlir.llvm.org\/users\/ Accessed on September 2024"},{"key":"e_1_2_1_21_1","unstructured":"2025. MLIRTracer. Online. https:\/\/github.com\/compiler-testing\/MLIRTracer"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2868427"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/3544559"},{"key":"e_1_2_1_24_1","unstructured":"Uday Bondhugula. 2020. High performance code generation in mlir: An early case study with gemm. arXiv preprint arXiv:2003.00532."},{"key":"e_1_2_1_25_1","volume-title":"TVM: An Automated End-to-End Optimizing Compiler for Deep Learning. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18)","author":"Chen Tianqi","year":"2018","unstructured":"Tianqi Chen, Thierry Moreau, Ziheng Jiang, Lianmin Zheng, Eddie Yan, Haichen Shen, Meghan Cowan, Leyuan Wang, Yuwei Hu, Luis Ceze, Carlos Guestrin, and Arvind Krishnamurthy. 2018. TVM: An Automated End-to-End Optimizing Compiler for Deep Learning. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18). USENIX Association, Carlsbad, CA. 578\u2013594. isbn:978-1-939133-08-3 https:\/\/www.usenix.org\/conference\/osdi18\/presentation\/chen"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/3597926.3598067"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3540250.3549085"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.5555\/3322706.3361996"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3183440.3195030"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/2372251.2372285"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/3510003.3510092"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICSE.2012.6227193"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1155\/2013"},{"key":"e_1_2_1_34_1","unstructured":"Pengchao Hu Man Lu Lei Wang and Guoyue Jiang. 2022. TPU-MLIR: A Compiler For TPU Using MLIR. arXiv preprint arXiv:2210.15016."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSE.2021.3058671"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/3497776.3517770"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSE.2007.70773"},{"key":"e_1_2_1_38_1","volume-title":"MLIR: A compiler infrastructure for the end of Moore\u2019s law. arXiv preprint arXiv:2002.11054.","author":"Lattner Chris","year":"2020","unstructured":"Chris Lattner, Mehdi Amini, Uday Bondhugula, Albert Cohen, Andy Davis, Jacques Pienaar, River Riddle, Tatiana Shpeisman, Nicolas Vasilache, and Oleksandr Zinenko. 2020. MLIR: A compiler infrastructure for the end of Moore\u2019s law. arXiv preprint arXiv:2002.11054."},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1145\/3377811.3380923"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3575707"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/3611643.3616337"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/3527317"},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICSE43902.2021.00037"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","unstructured":"William S Moses Ivan R Ivanov Jens Domke Toshio Endo Johannes Doerfert and Oleksandr Zinenko. 2023. High-performance gpu-to-cpu transpilation and optimization via high-level parallel constructs. 119\u2013134. https:\/\/doi.org\/10.1145\/3572848.3577475 10.1145\/3572848.3577475","DOI":"10.1145\/3572848.3577475"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3324884.3418930"},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.6084\/m9.figshare.4542709.v6"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1145\/1453101.1453105"},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.1145\/3447582"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/3597926.3598088"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1145\/3650212.3680360"},{"key":"e_1_2_1_51_1","volume-title":"Mahesh Ravishankar, Thomas Raoux, Alexander Belyaev","author":"Vasilache Nicolas","year":"2022","unstructured":"Nicolas Vasilache, Oleksandr Zinenko, Aart JC Bik, Mahesh Ravishankar, Thomas Raoux, Alexander Belyaev, Matthias Springer, Tobias Gysi, Diego Caballero, and Stephan Herhut. 2022. Composable and modular code generation in MLIR: A structured and retargetable approach to tensor compiler construction. arXiv preprint arXiv:2202.03293."},{"key":"e_1_2_1_52_1","unstructured":"Nicolas Vasilache Oleksandr Zinenko Theodoros Theodoridis Priya Goyal Zachary DeVito William S Moses Sven Verdoolaege Andrew Adams and Albert Cohen. 2018. Tensor comprehensions: Framework-agnostic high-performance machine learning abstractions. arXiv preprint arXiv:1802.04730."},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1109\/ASE56229.2023.00120"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1145\/3510003.3510165"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1145\/3368089.3409761"},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.1145\/3510003.3510041"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.1145\/3489048.3522655"},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.1145\/3533767.3534220"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.1109\/ASE.2019.00127"},{"key":"e_1_2_1_60_1","doi-asserted-by":"publisher","DOI":"10.1145\/3293882.3330579"},{"key":"e_1_2_1_61_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSE.2021.3124006"}],"container-title":["Proceedings of the ACM on Software Engineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3729372","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T15:18:58Z","timestamp":1750346338000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3729372"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,19]]},"references-count":61,"journal-issue":{"issue":"FSE","published-print":{"date-parts":[[2025,6,19]]}},"alternative-id":["10.1145\/3729372"],"URL":"https:\/\/doi.org\/10.1145\/3729372","relation":{},"ISSN":["2994-970X"],"issn-type":[{"value":"2994-970X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,19]]}}}