{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,13]],"date-time":"2026-11-13T17:48:43Z","timestamp":1794592123695,"version":"4.9.1"},"reference-count":44,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T00:00:00Z","timestamp":1784764800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Archit. Code Optim."],"published-print":{"date-parts":[[2026,9,30]]},"abstract":"<jats:p>\n                    Memory safety errors continue to pose a significant threat to current computing systems, and graphics processing units (GPUs) are no exception. A prominent class of memory safety algorithms is\n                    <jats:italic toggle=\"yes\">allocation-based<\/jats:italic>\n                    solutions. The key idea is to maintain each allocation\u2019s metadata (base address and size) in a disjoint table and retrieve it at runtime to verify memory accesses. While several previous solutions have adopted allocation-based algorithms (e.g., cuCatch\u00a0[\n                    <jats:xref ref-type=\"bibr\">39<\/jats:xref>\n                    ] and GPUShield\u00a0[\n                    <jats:xref ref-type=\"bibr\">18<\/jats:xref>\n                    ]), they typically suffer from high memory overheads or scalability problems. In this work, we examine the key characteristics of real-world GPU workloads and observe several differences between GPU and CPU applications regarding memory access patterns, memory footprint, number of live allocations, and active allocation working set. Our observations motivate\u00a0GPUArmor,\n                    <jats:xref ref-type=\"fn\">\n                      <jats:sup>1<\/jats:sup>\n                    <\/jats:xref>\n                    a hardware-software co-design framework for memory safety on GPUs. We show that a simple compiler analysis combined with lightweight hardware support can help prevent spatial and temporal memory violations on modern GPU workloads with 8% average run time overheads while cuCatch incurs nearly 36% overhead. In fact,\u00a0GPUArmor achieves speed-of-light performance with negligible storage requirements. This result benefits both base and bounds solutions and memory tagging techniques, which we showcase with\u00a0GPUArmor-Binary, a variation of\u00a0GPUArmor that does not require recompilation, and achieves 2.2% slowdowns while significantly reducing storage overheads beyond traditional memory tagging approaches.\n                  <\/jats:p>","DOI":"10.1145\/3815783","type":"journal-article","created":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T10:29:53Z","timestamp":1778322593000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["GPUArmor: A Hardware-Software Co-Design for Efficient and Scalable Memory Safety on GPUs"],"prefix":"10.1145","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6971-6996","authenticated-orcid":false,"given":"Mohamed","family":"Tarek Ibn Ziad","sequence":"first","affiliation":[{"name":"NVIDIA","place":["Westford, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6976-9000","authenticated-orcid":false,"given":"Sana","family":"Damani","sequence":"additional","affiliation":[{"name":"NVIDIA","place":["Santa Clara, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1350-0165","authenticated-orcid":false,"given":"Mark","family":"Stephenson","sequence":"additional","affiliation":[{"name":"NVIDIA","place":["Austin, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6701-6099","authenticated-orcid":false,"given":"Stephen W.","family":"Keckler","sequence":"additional","affiliation":[{"name":"NVIDIA","place":["Austin, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5709-2992","authenticated-orcid":false,"given":"Aamer","family":"Jaleel","sequence":"additional","affiliation":[{"name":"NVIDIA","place":["Westford, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,7,23]]},"reference":[{"key":"e_1_3_2_2_2","article-title":"Memory Tagging Extension: Enhancing memory safety through architecture","year":"2019","unstructured":"ARM. 2019. Memory Tagging Extension: Enhancing memory safety through architecture. Retrieved June 17, 2026 from https:\/\/developer.arm.com\/community\/arm-community-blogs\/b\/architectures-and-processors-blog\/posts\/enhancing-memory-safety","journal-title":"Retrieved June 17, 2026 from https:\/\/developer.arm.com\/community\/arm-community-blogs\/b\/architectures-and-processors-blog\/posts\/enhancing-memory-safety"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1002\/spe.4380180902"},{"key":"e_1_3_2_4_2","first-page":"841","volume-title":"Proceedings of the 29th USENIX Security Symposium","author":"Cloosters Tobias","year":"2020","unstructured":"Tobias Cloosters, Michael Rodler, and Lucas Davi. 2020. TeeRex: Discovery and exploitation of memory corruption vulnerabilities in SGX enclaves. In Proceedings of the 29th USENIX Security Symposium. 841\u2013858. Retrieved from https:\/\/www.usenix.org\/conference\/usenixsecurity20\/presentation\/cloosters"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2019.00089"},{"key":"e_1_3_2_6_2","volume-title":"Introduction to Algorithms, Fourth Edition","author":"Cormen Thomas H.","year":"2022","unstructured":"Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, and Clifford Stein. 2022. Introduction to Algorithms, Fourth Edition. MIT. 2021037260 Retrieved from https:\/\/books.google.com\/books?id=HOJyzgEACAAJ"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-47099-3_9"},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2020.3042965"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.1145\/3243176.3243194"},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589080"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.5555\/3049832.3049840"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/3204919.3204934"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.5555\/3698900.3699126"},{"key":"e_1_3_2_14_2","article-title":"Simplifying GPU Application Development with Heterogeneous Memory Management","author":"Hubbard John","year":"2023","unstructured":"John Hubbard, Gonzalo Brito, Chirayu Garg, Nikolay Sakharnykh, and Fred Oh. 2023. Simplifying GPU Application Development with Heterogeneous Memory Management. Retrieved June 17, 2026 from https:\/\/developer.nvidia.com\/blog\/simplifying-gpu-application-development-with-heterogeneous-memory-management\/","journal-title":"Retrieved June 17, 2026 from https:\/\/developer.nvidia.com\/blog\/simplifying-gpu-application-development-with-heterogeneous-memory-management\/"},{"key":"e_1_3_2_15_2","article-title":"5-Level Paging and 5-Level EPT","author":"Corp. Intel","year":"2017","unstructured":"Intel Corp.2017. 5-Level Paging and 5-Level EPT. Retrieved June 17, 2026 from https:\/\/software.intel.com\/content\/www\/us\/en\/develop\/download\/5-level-paging-and-5-level-ept-white-paper.html","journal-title":"Retrieved June 17, 2026 from https:\/\/software.intel.com\/content\/www\/us\/en\/develop\/download\/5-level-paging-and-5-level-ept-white-paper.html"},{"key":"e_1_3_2_16_2","article-title":"Security Analysis of CHERI ISA","author":"Joly Nicolas","year":"2020","unstructured":"Nicolas Joly, Saif ElSherei, and Saar Amar. 2020. Security Analysis of CHERI ISA. Retrieved June 17, 2026 from https:\/\/github.com\/microsoft\/MSRC-Security-Research\/blob\/master\/papers\/2020\/Security%20analysis%20of%20CHERI%20ISA.pdf","journal-title":"Retrieved June 17, 2026 from https:\/\/github.com\/microsoft\/MSRC-Security-Research\/blob\/master\/papers\/2020\/Security%20analysis%20of%20CHERI%20ISA.pdf"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.1145\/2678373.2665726"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA61900.2025.00122"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527420"},{"key":"e_1_3_2_20_2","unstructured":"LLVM Project. 2023. Scudo Hardened Allocator. Retrieved June 17 2026 from https:\/\/llvm.org\/docs\/ScudoHardenedAllocator.html"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11416-015-0251-1"},{"key":"e_1_3_2_22_2","article-title":"CACTI: An analytical cache and memory access time, cycle time, area, and power model","author":"Muralimanohar N.","year":"2024","unstructured":"N. Muralimanohar, A. Shafiee, and V. Srinivas. 2024. CACTI: An analytical cache and memory access time, cycle time, area, and power model. Retrieved June 17, 2026 from https:\/\/github.com\/HewlettPackard\/cacti","journal-title":"Retrieved June 17, 2026 from https:\/\/github.com\/HewlettPackard\/cacti"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1145\/1542476.1542504"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1145\/3433210.3453077"},{"key":"e_1_3_2_25_2","article-title":"Compute Sanitizer v2024.2.0","year":"2024","unstructured":"NVIDIA. 2024. Compute Sanitizer v2024.2.0. Retrieved June 17, 2026 from https:\/\/docs.nvidia.com\/compute-sanitizer\/ReleaseNotes\/index.html#updates-in-2024-2","journal-title":"Retrieved June 17, 2026 from https:\/\/docs.nvidia.com\/compute-sanitizer\/ReleaseNotes\/index.html#updates-in-2024-2"},{"key":"e_1_3_2_26_2","unstructured":"NVIDIA. 2024. CUDA C++ Programming Guide. Retrieved June 17 2026 from https:\/\/docs.nvidia.com\/cuda\/cuda-c-programming-guide\/contents.html"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1145\/3224423"},{"key":"e_1_3_2_28_2","article-title":"Hardware-assisted checking using Silicon Secured Memory (SSM)","year":"2015","unstructured":"Oracle. 2015. Hardware-assisted checking using Silicon Secured Memory (SSM). Retrieved June 17, 2026 from https:\/\/docs.oracle.com\/cd\/E37069_01\/html\/E37085\/gphwb.html","journal-title":"Retrieved June 17, 2026 from https:\/\/docs.oracle.com\/cd\/E37069_01\/html\/E37085\/gphwb.html"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cose.2020.102115"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1145\/1778765.1778803"},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1145\/3722041.3723099"},{"key":"e_1_3_2_32_2","volume-title":"Algorithms in Java","author":"Sedgewick R.","year":"2003","unstructured":"R. Sedgewick. 2003. Algorithms in Java. Addison-Wesley. 98025050 Retrieved from https:\/\/books.google.com\/books?id=TKpQAAAAMAAJ"},{"key":"e_1_3_2_33_2","volume-title":"Proceedings of the 2012 USENIX Annual Technical Conference","author":"Serebryany Konstantin","year":"2012","unstructured":"Konstantin Serebryany, Derek Bruening, Alexander Potapenko, and Dmitry Vyukov. 2012. AddressSanitizer: A fast address sanity checker. In Proceedings of the 2012 USENIX Annual Technical Conference. Boston, MA, USA."},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/EuroSP53844.2022.00025"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1145\/3297663.3310311"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2019.00010"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589102"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2013.13"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00076"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.1145\/3591225"},{"key":"e_1_3_2_41_2","first-page":"405","volume-title":"Proceedings of the 12th European Conference on Computer Systems","author":"Kouwe Erik van der","year":"2017","unstructured":"Erik van der Kouwe, Vinod Nigade, and Cristiano Giuffrida. 2017. DangSan: Scalable use-after-free detection. In Proceedings of the 12th European Conference on Computer Systems. Belgrade, Serbia, 405\u2013419."},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33338-5_5"},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00077"},{"key":"e_1_3_2_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD56317.2022.00108"},{"key":"e_1_3_2_45_2","doi-asserted-by":"publisher","DOI":"10.1145\/3576915.3616672"}],"container-title":["ACM Transactions on Architecture and Code Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3815783","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T15:53:02Z","timestamp":1784821982000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3815783"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,23]]},"references-count":44,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,9,30]]}},"alternative-id":["10.1145\/3815783"],"URL":"https:\/\/doi.org\/10.1145\/3815783","relation":{},"ISSN":["1544-3566","1544-3973"],"issn-type":[{"value":"1544-3566","type":"print"},{"value":"1544-3973","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,23]]},"assertion":[{"value":"2025-08-20","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-21","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-23","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}