Effective Random Test Generation for Deep Learning Compilers

Deep learning compilers help address difficulties of deploying deep learning models on diverse types of hardware. Testing deep learning compilers is highly crucial, because they are impacting countless AI applications that use them for model optimization and deployment. To test deep learning compile...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-02
Hauptverfasser: Ren, Luyao, Wang, ZiHeng, Xiong, Yingfei, Zhang, Li, Jiang, Guoyue, Xie, Tao
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Ren, Luyao
Wang, ZiHeng
Xiong, Yingfei
Zhang, Li
Jiang, Guoyue
Xie, Tao
description Deep learning compilers help address difficulties of deploying deep learning models on diverse types of hardware. Testing deep learning compilers is highly crucial, because they are impacting countless AI applications that use them for model optimization and deployment. To test deep learning compilers, random testing, being popularly used for compiler testing practices, faces the challenge of generating semantically valid test inputs, i.e., deep learning models that satisfy the semantic model specifications (in short as semantic specifications). To tackle this challenge, in this paper, we propose a novel approach named Isra, including a domain-specific constraint solver that resolves the constraints from the semantic specifications without backtracking. We implement and apply our approach on three popular real-world deep learning compilers including TVM, Glow, and a commercial compiler. The evaluation results show that Isra is more effective than the state-of-the-art approaches and the baseline approaches on constructing valid test inputs for compiler-bug detection, and Isra successfully finds 24 previously unknown bugs in released versions of the three compilers. These results indicate effectiveness and practical value of Isra.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2772191593</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2772191593</sourcerecordid><originalsourceid>FETCH-proquest_journals_27721915933</originalsourceid><addsrcrecordid>eNqNjLEKwjAUAIMgWLT_8MC50LwYa8GtVh2cpHsJ-iIpbRKT1u-3gx_gdMMdt2AJCsGzww5xxdIYuzzPcV-glCJhx1preozmQ3BX9ukGaCiOcCFLQY3GWdAuwInIw41UsMa-oHKDNz2FuGFLrfpI6Y9rtj3XTXXNfHDvaf60nZuCnVWLRYG85LIU4r_qCw0aNz4</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2772191593</pqid></control><display><type>article</type><title>Effective Random Test Generation for Deep Learning Compilers</title><source>Free eJournals</source><creator>Ren, Luyao ; Wang, ZiHeng ; Xiong, Yingfei ; Zhang, Li ; Jiang, Guoyue ; Xie, Tao</creator><creatorcontrib>Ren, Luyao ; Wang, ZiHeng ; Xiong, Yingfei ; Zhang, Li ; Jiang, Guoyue ; Xie, Tao</creatorcontrib><description>Deep learning compilers help address difficulties of deploying deep learning models on diverse types of hardware. Testing deep learning compilers is highly crucial, because they are impacting countless AI applications that use them for model optimization and deployment. To test deep learning compilers, random testing, being popularly used for compiler testing practices, faces the challenge of generating semantically valid test inputs, i.e., deep learning models that satisfy the semantic model specifications (in short as semantic specifications). To tackle this challenge, in this paper, we propose a novel approach named Isra, including a domain-specific constraint solver that resolves the constraints from the semantic specifications without backtracking. We implement and apply our approach on three popular real-world deep learning compilers including TVM, Glow, and a commercial compiler. The evaluation results show that Isra is more effective than the state-of-the-art approaches and the baseline approaches on constructing valid test inputs for compiler-bug detection, and Isra successfully finds 24 previously unknown bugs in released versions of the three compilers. These results indicate effectiveness and practical value of Isra.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Compilers ; Deep learning ; Optimization ; Semantics ; Specifications</subject><ispartof>arXiv.org, 2023-02</ispartof><rights>2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Ren, Luyao</creatorcontrib><creatorcontrib>Wang, ZiHeng</creatorcontrib><creatorcontrib>Xiong, Yingfei</creatorcontrib><creatorcontrib>Zhang, Li</creatorcontrib><creatorcontrib>Jiang, Guoyue</creatorcontrib><creatorcontrib>Xie, Tao</creatorcontrib><title>Effective Random Test Generation for Deep Learning Compilers</title><title>arXiv.org</title><description>Deep learning compilers help address difficulties of deploying deep learning models on diverse types of hardware. Testing deep learning compilers is highly crucial, because they are impacting countless AI applications that use them for model optimization and deployment. To test deep learning compilers, random testing, being popularly used for compiler testing practices, faces the challenge of generating semantically valid test inputs, i.e., deep learning models that satisfy the semantic model specifications (in short as semantic specifications). To tackle this challenge, in this paper, we propose a novel approach named Isra, including a domain-specific constraint solver that resolves the constraints from the semantic specifications without backtracking. We implement and apply our approach on three popular real-world deep learning compilers including TVM, Glow, and a commercial compiler. The evaluation results show that Isra is more effective than the state-of-the-art approaches and the baseline approaches on constructing valid test inputs for compiler-bug detection, and Isra successfully finds 24 previously unknown bugs in released versions of the three compilers. These results indicate effectiveness and practical value of Isra.</description><subject>Compilers</subject><subject>Deep learning</subject><subject>Optimization</subject><subject>Semantics</subject><subject>Specifications</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNjLEKwjAUAIMgWLT_8MC50LwYa8GtVh2cpHsJ-iIpbRKT1u-3gx_gdMMdt2AJCsGzww5xxdIYuzzPcV-glCJhx1preozmQ3BX9ukGaCiOcCFLQY3GWdAuwInIw41UsMa-oHKDNz2FuGFLrfpI6Y9rtj3XTXXNfHDvaf60nZuCnVWLRYG85LIU4r_qCw0aNz4</recordid><startdate>20230204</startdate><enddate>20230204</enddate><creator>Ren, Luyao</creator><creator>Wang, ZiHeng</creator><creator>Xiong, Yingfei</creator><creator>Zhang, Li</creator><creator>Jiang, Guoyue</creator><creator>Xie, Tao</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20230204</creationdate><title>Effective Random Test Generation for Deep Learning Compilers</title><author>Ren, Luyao ; Wang, ZiHeng ; Xiong, Yingfei ; Zhang, Li ; Jiang, Guoyue ; Xie, Tao</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_27721915933</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Compilers</topic><topic>Deep learning</topic><topic>Optimization</topic><topic>Semantics</topic><topic>Specifications</topic><toplevel>online_resources</toplevel><creatorcontrib>Ren, Luyao</creatorcontrib><creatorcontrib>Wang, ZiHeng</creatorcontrib><creatorcontrib>Xiong, Yingfei</creatorcontrib><creatorcontrib>Zhang, Li</creatorcontrib><creatorcontrib>Jiang, Guoyue</creatorcontrib><creatorcontrib>Xie, Tao</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection (ProQuest)</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ren, Luyao</au><au>Wang, ZiHeng</au><au>Xiong, Yingfei</au><au>Zhang, Li</au><au>Jiang, Guoyue</au><au>Xie, Tao</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Effective Random Test Generation for Deep Learning Compilers</atitle><jtitle>arXiv.org</jtitle><date>2023-02-04</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Deep learning compilers help address difficulties of deploying deep learning models on diverse types of hardware. Testing deep learning compilers is highly crucial, because they are impacting countless AI applications that use them for model optimization and deployment. To test deep learning compilers, random testing, being popularly used for compiler testing practices, faces the challenge of generating semantically valid test inputs, i.e., deep learning models that satisfy the semantic model specifications (in short as semantic specifications). To tackle this challenge, in this paper, we propose a novel approach named Isra, including a domain-specific constraint solver that resolves the constraints from the semantic specifications without backtracking. We implement and apply our approach on three popular real-world deep learning compilers including TVM, Glow, and a commercial compiler. The evaluation results show that Isra is more effective than the state-of-the-art approaches and the baseline approaches on constructing valid test inputs for compiler-bug detection, and Isra successfully finds 24 previously unknown bugs in released versions of the three compilers. These results indicate effectiveness and practical value of Isra.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-02
issn 2331-8422
language eng
recordid cdi_proquest_journals_2772191593
source Free eJournals
subjects Compilers
Deep learning
Optimization
Semantics
Specifications
title Effective Random Test Generation for Deep Learning Compilers
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-13T13%3A58%3A36IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Effective%20Random%20Test%20Generation%20for%20Deep%20Learning%20Compilers&rft.jtitle=arXiv.org&rft.au=Ren,%20Luyao&rft.date=2023-02-04&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2772191593%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2772191593&rft_id=info:pmid/&rfr_iscdi=true