{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/autophase-compiler-phase-ordering-for-high","title":"AutoPhase: Compiler Phase-Ordering for High Level Synthesis with Deep Reinforcement Learning","arxiv_id":"1901.04615","date":"2019-01-15","proceeding":null,"authors":["Ameer Haj-Ali","Qijing Huang","William Moses","John Xiang","Ion Stoica","Krste Asanovic","John Wawrzynek"],"abstract":"The performance of the code generated by a compiler depends on the order in\nwhich the optimization passes are applied. In high-level synthesis, the quality\nof the generated circuit relates directly to the code generated by the\nfront-end compiler. Choosing a good order--often referred to as the\nphase-ordering problem--is an NP-hard problem. In this paper, we evaluate a new\ntechnique to address the phase-ordering problem: deep reinforcement learning.\nWe implement a framework in the context of the LLVM compiler to optimize the\nordering for HLS programs and compare the performance of deep reinforcement\nlearning to state-of-the-art algorithms that address the phase-ordering\nproblem. Overall, our framework runs one to two orders of magnitude faster than\nthese algorithms, and achieves a 16% improvement in circuit performance over\nthe -O3 compiler flag.","url_abs":"http://arxiv.org/abs/1901.04615v2","url_pdf":"http://arxiv.org/pdf/1901.04615v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"autophase-compiler-phase-ordering-for-high","repo_url":"https://github.com/ucb-bar/autophase","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"high-level-synthesis","task_name":"High-Level Synthesis"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}