{"id":"https://openalex.org/W2021688604","doi":"https://doi.org/10.1145/2541940.2541978","title":"Speculative hardware/software co-designed floating-point multiply-add fusion","display_name":"Speculative hardware/software co-designed floating-point multiply-add fusion","publication_year":2014,"publication_date":"2014-02-24","ids":{"openalex":"https://openalex.org/W2021688604","doi":"https://doi.org/10.1145/2541940.2541978","mag":"2021688604"},"language":"en","primary_location":{"id":"doi:10.1145/2541940.2541978","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2541940.2541978","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 19th international conference on Architectural support for programming languages and operating systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5015144284","display_name":"Marc Lupon","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210136471","display_name":"FC Barcelona","ror":"https://ror.org/04bpz1v84","country_code":"ES","type":"other","lineage":["https://openalex.org/I4210136471"]}],"countries":["ES","US"],"is_corresponding":false,"raw_author_name":"Marc Lupon","raw_affiliation_strings":["Intel Barcelona Research Center, Barcelona, Spain","Intel Barcelona Research Center, Barcelona, Spain#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain","institution_ids":["https://openalex.org/I4210136471"]},{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain#TAB#","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046385142","display_name":"Enric Gibert","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210136471","display_name":"FC Barcelona","ror":"https://ror.org/04bpz1v84","country_code":"ES","type":"other","lineage":["https://openalex.org/I4210136471"]}],"countries":["ES","US"],"is_corresponding":false,"raw_author_name":"Enric Gibert","raw_affiliation_strings":["Intel Barcelona Research Center, Barcelona, Spain","Intel Barcelona Research Center, Barcelona, Spain#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain","institution_ids":["https://openalex.org/I4210136471"]},{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain#TAB#","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000686736","display_name":"Grigorios Magklis","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210136471","display_name":"FC Barcelona","ror":"https://ror.org/04bpz1v84","country_code":"ES","type":"other","lineage":["https://openalex.org/I4210136471"]}],"countries":["ES","US"],"is_corresponding":false,"raw_author_name":"Grigorios Magklis","raw_affiliation_strings":["Intel Barcelona Research Center, Barcelona, Spain","Intel Barcelona Research Center, Barcelona, Spain#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain","institution_ids":["https://openalex.org/I4210136471"]},{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain#TAB#","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001719897","display_name":"S. Samudrala","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sridhar Samudrala","raw_affiliation_strings":["work performed while at Intel Corporation, Austin, USA","work performed while at Intel Corporation, Austin, USA#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"work performed while at Intel Corporation, Austin, USA","institution_ids":["https://openalex.org/I1343180700"]},{"raw_affiliation_string":"work performed while at Intel Corporation, Austin, USA#TAB#","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069243017","display_name":"Ra\u00fal Alcaraz","orcid":"https://orcid.org/0000-0002-0942-3638"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210136471","display_name":"FC Barcelona","ror":"https://ror.org/04bpz1v84","country_code":"ES","type":"other","lineage":["https://openalex.org/I4210136471"]}],"countries":["ES","US"],"is_corresponding":false,"raw_author_name":"Ra\u00fal Mart\u00ednez","raw_affiliation_strings":["Intel Barcelona Research Center, Barcelona, Spain","Intel Barcelona Research Center, Barcelona, Spain#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain","institution_ids":["https://openalex.org/I4210136471"]},{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain#TAB#","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109218173","display_name":"Kyriakos Stavrou","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]},{"id":"https://openalex.org/I4210136471","display_name":"FC Barcelona","ror":"https://ror.org/04bpz1v84","country_code":"ES","type":"other","lineage":["https://openalex.org/I4210136471"]}],"countries":["ES","US"],"is_corresponding":false,"raw_author_name":"Kyriakos Stavrou","raw_affiliation_strings":["Intel Barcelona Research Center, Barcelona, Spain","Intel Barcelona Research Center, Barcelona, Spain#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain","institution_ids":["https://openalex.org/I4210136471"]},{"raw_affiliation_string":"Intel Barcelona Research Center, Barcelona, Spain#TAB#","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080905072","display_name":"David R. Ditzel","orcid":"https://orcid.org/0000-0003-0944-3353"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David R. Ditzel","raw_affiliation_strings":["work performed while at Intel Corporation, Santa Clara, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"work performed while at Intel Corporation, Santa Clara, USA","institution_ids":["https://openalex.org/I1343180700"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":3.1511,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.91424431,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"623","last_page":"638"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10363","display_name":"Low-power high-performance VLSI design","score":0.9984999895095825,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/rounding","display_name":"Rounding","score":0.9116206169128418},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8448889255523682},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5643416047096252},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.5515727996826172},{"id":"https://openalex.org/keywords/floating-point-unit","display_name":"Floating-point unit","score":0.5371763110160828},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5141506791114807},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.46189913153648376},{"id":"https://openalex.org/keywords/instructions-per-cycle","display_name":"Instructions per cycle","score":0.4546717405319214},{"id":"https://openalex.org/keywords/arithmetic","display_name":"Arithmetic","score":0.3533168435096741},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2925260365009308},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.2428375780582428},{"id":"https://openalex.org/keywords/central-processing-unit","display_name":"Central processing unit","score":0.17741331458091736}],"concepts":[{"id":"https://openalex.org/C136625980","wikidata":"https://www.wikidata.org/wiki/Q663208","display_name":"Rounding","level":2,"score":0.9116206169128418},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8448889255523682},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5643416047096252},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.5515727996826172},{"id":"https://openalex.org/C110305270","wikidata":"https://www.wikidata.org/wiki/Q733507","display_name":"Floating-point unit","level":3,"score":0.5371763110160828},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5141506791114807},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.46189913153648376},{"id":"https://openalex.org/C156972235","wikidata":"https://www.wikidata.org/wiki/Q1443434","display_name":"Instructions per cycle","level":3,"score":0.4546717405319214},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.3533168435096741},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2925260365009308},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.2428375780582428},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.17741331458091736},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/2541940.2541978","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2541940.2541978","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 19th international conference on Architectural support for programming languages and operating systems","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W79353637","https://openalex.org/W153032839","https://openalex.org/W605824955","https://openalex.org/W1542196645","https://openalex.org/W1979072566","https://openalex.org/W1980480935","https://openalex.org/W1983778743","https://openalex.org/W2026686501","https://openalex.org/W2036482573","https://openalex.org/W2057234250","https://openalex.org/W2057543661","https://openalex.org/W2070273314","https://openalex.org/W2088020108","https://openalex.org/W2099336824","https://openalex.org/W2100974190","https://openalex.org/W2109611156","https://openalex.org/W2120371291","https://openalex.org/W2121764011","https://openalex.org/W2124813728","https://openalex.org/W2130123999","https://openalex.org/W2130766965","https://openalex.org/W2134663450","https://openalex.org/W2136854245","https://openalex.org/W2140351961","https://openalex.org/W2143194988","https://openalex.org/W2144133815","https://openalex.org/W2144726606","https://openalex.org/W2148865465","https://openalex.org/W2150023094","https://openalex.org/W2153456949","https://openalex.org/W2158942957","https://openalex.org/W2168020750","https://openalex.org/W2204310803","https://openalex.org/W2207050309","https://openalex.org/W2465756754","https://openalex.org/W2541452119","https://openalex.org/W2545221752","https://openalex.org/W3141428879","https://openalex.org/W4236971743","https://openalex.org/W4250702372","https://openalex.org/W4251914687","https://openalex.org/W6618621456","https://openalex.org/W6729167888"],"related_works":["https://openalex.org/W4376144198","https://openalex.org/W2014186618","https://openalex.org/W2064131220","https://openalex.org/W2540610482","https://openalex.org/W2158464203","https://openalex.org/W2183609869","https://openalex.org/W1520019936","https://openalex.org/W2027965104","https://openalex.org/W2073750546","https://openalex.org/W2470712352"],"abstract_inverted_index":{"A":[0],"Fused":[1],"Multiply-Add":[2,155],"(FMA)":[3],"instruction":[4,157,169],"is":[5,46,110,170],"currently":[6],"available":[7],"in":[8,222],"many":[9],"general-purpose":[10],"processors.":[11],"It":[12],"increases":[13,22],"performance":[14,237],"by":[15,24,53,121,172,238,244],"reducing":[16],"latency":[17,214],"of":[18,41,87,119,146,224],"dependent":[19],"operations":[20],"and":[21,62,135,189,207,215,240],"precision":[23],"computing":[25],"the":[26,38,123,127,143],"result":[27,130],"as":[28,67,131],"an":[29,160,164],"indivisible":[30],"operation":[31,45,162],"with":[32,75,116,151,163],"no":[33],"intermediate":[34,165],"rounding.":[35,166],"However,":[36],"since":[37],"arithmetic":[39],"behavior":[40],"a":[42,104,147,152,173,181,218,228],"single-rounding":[43,206],"FMA":[44,97,161,197],"different":[47],"than":[48],"independent":[49],"FP":[50,54,133,136,187,190],"multiply":[51,134,188],"followed":[52],"add":[55,137,191],"instructions,":[56],"some":[57],"algorithms":[58],"require":[59],"significant":[60],"revalidation":[61],"rewriting":[63],"efforts":[64],"to":[65,73,84,95,112,125,185,203,216],"work":[66],"expected":[68],"when":[69],"they":[70],"are":[71,92],"compiled":[72],"operate":[74],"FMA--a":[76],"cost":[77],"that":[78,109,158,179,233],"developers":[79],"may":[80],"not":[81,93],"be":[82],"willing":[83],"pay.":[85],"Because":[86],"that,":[88],"abundant":[89],"legacy":[90],"applications":[91],"able":[94,111],"utilize":[96],"instructions.":[98,195],"In":[99,139],"this":[100],"paper":[101],"we":[102,141],"propose":[103],"novel":[105],"HW/SW":[106,148],"collaborative":[107],"technique":[108],"efficiently":[113],"execute":[114],"workloads":[115],"increased":[117],"utilization":[118],"FMA,":[120],"adding":[122],"option":[124],"get":[126],"same":[128],"numerical":[129],"separate":[132],"pairs.":[138],"particular,":[140],"extended":[142],"host":[144],"ISA":[145],"co-designed":[149],"processor":[150],"new":[153,168],"Combined":[154],"(CMA)":[156],"performs":[159],"This":[167],"used":[171],"transparent":[174],"dynamic":[175],"translation":[176],"software":[177],"layer":[178],"uses":[180],"speculative":[182],"instruction-fusion":[183],"optimization":[184],"transform":[186],"sequences":[192],"into":[193],"CMA":[194,234],"The":[196],"unit":[198],"has":[199],"been":[200],"slightly":[201],"modified":[202],"support":[204],"both":[205],"double-rounding":[208],"fused":[209],"instructions":[210,243],"without":[211],"increasing":[212],"their":[213],"provide":[217],"conservative":[219],"fall-back":[220],"path":[221],"case":[223],"mispeculation.":[225],"Evaluation":[226],"on":[227],"cycle-accurate":[229],"timing":[230],"simulator":[231],"showed":[232],"improved":[235],"SPECfp":[236],"6.3%":[239],"reduced":[241],"executed":[242],"4.7%.":[245]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2019,"cited_by_count":1},{"year":2017,"cited_by_count":1},{"year":2016,"cited_by_count":5},{"year":2014,"cited_by_count":4}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}