{"title":"利用时序诱导状态寄存器等效的性能驱动重合成","authors":"P. Kalla, M. Ciesielski","doi":"10.1145/307418.307579","DOIUrl":null,"url":null,"abstract":"This paper presents a retiming and resynthesis technique for cycle-time minimization of sequential circuits with feedback (finite state machines). Operating on the delay critical paths of the circuit, we perform a set of controlled local retimings of registers across fanout stems and logic gates, followed by local node simplifications. We guide the retiming of registers across fanout stems to induce equivalence relations among them, which are exploited for subsequent logic simplification. Our technique is able to analyze correlation of logic across register boundaries during simplification. We strive to minimize the increase in number of registers without sacrificing the cycle-time performance. The results demonstrate a favourable performance/area trade-off when compared with optimally retimed circuits.","PeriodicalId":442382,"journal":{"name":"Design, Automation and Test in Europe Conference and Exhibition, 1999. Proceedings (Cat. No. PR00078)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1999-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Performance driven resynthesis by exploiting retiming-induced state register equivalence\",\"authors\":\"P. Kalla, M. Ciesielski\",\"doi\":\"10.1145/307418.307579\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper presents a retiming and resynthesis technique for cycle-time minimization of sequential circuits with feedback (finite state machines). Operating on the delay critical paths of the circuit, we perform a set of controlled local retimings of registers across fanout stems and logic gates, followed by local node simplifications. We guide the retiming of registers across fanout stems to induce equivalence relations among them, which are exploited for subsequent logic simplification. Our technique is able to analyze correlation of logic across register boundaries during simplification. We strive to minimize the increase in number of registers without sacrificing the cycle-time performance. The results demonstrate a favourable performance/area trade-off when compared with optimally retimed circuits.\",\"PeriodicalId\":442382,\"journal\":{\"name\":\"Design, Automation and Test in Europe Conference and Exhibition, 1999. Proceedings (Cat. No. PR00078)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1999-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Design, Automation and Test in Europe Conference and Exhibition, 1999. Proceedings (Cat. No. PR00078)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/307418.307579\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Design, Automation and Test in Europe Conference and Exhibition, 1999. Proceedings (Cat. No. PR00078)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/307418.307579","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Performance driven resynthesis by exploiting retiming-induced state register equivalence
This paper presents a retiming and resynthesis technique for cycle-time minimization of sequential circuits with feedback (finite state machines). Operating on the delay critical paths of the circuit, we perform a set of controlled local retimings of registers across fanout stems and logic gates, followed by local node simplifications. We guide the retiming of registers across fanout stems to induce equivalence relations among them, which are exploited for subsequent logic simplification. Our technique is able to analyze correlation of logic across register boundaries during simplification. We strive to minimize the increase in number of registers without sacrificing the cycle-time performance. The results demonstrate a favourable performance/area trade-off when compared with optimally retimed circuits.