{"title":"Speeding up SoC virtual platform simulation by data-dependency aware virtual synchronization","authors":"Kuen-Huei Lin, S. Cai, Chung-Yang Huang","doi":"10.1109/SOCDC.2008.4815621","DOIUrl":null,"url":null,"abstract":"This paper presents a novel simulation technique for SoC virtual platforms at transaction level to achieve million-instructions-per-second (MIPS) level of simulation speed, while retaining accurate and detailed cycle information for system explorations. We first apply virtual synchronization concept to our simulation algorithm and then apply a novel data-dependency aware concept to determine precise synchronization points in simulation. Our experimental result shows that comparing to the traditional cycle-accurate TLM simulation, our approach is about 44 times faster with the simulation speed around 3.66 MPIS.","PeriodicalId":405078,"journal":{"name":"2008 International SoC Design Conference","volume":"50 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 International SoC Design Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SOCDC.2008.4815621","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14
Abstract
This paper presents a novel simulation technique for SoC virtual platforms at transaction level to achieve million-instructions-per-second (MIPS) level of simulation speed, while retaining accurate and detailed cycle information for system explorations. We first apply virtual synchronization concept to our simulation algorithm and then apply a novel data-dependency aware concept to determine precise synchronization points in simulation. Our experimental result shows that comparing to the traditional cycle-accurate TLM simulation, our approach is about 44 times faster with the simulation speed around 3.66 MPIS.