1988
DOI: 10.1109/43.3169
|View full text |Cite
|
Sign up to set email alerts
|

Sehwa: a software package for synthesis of pipelines from behavioral specifications

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
101
0
2

Year Published

1991
1991
1999
1999

Publication Types

Select...
4
4
1

Relationship

0
9

Authors

Journals

citations
Cited by 330 publications
(103 citation statements)
references
References 8 publications
0
101
0
2
Order By: Relevance
“…With very little additional hardware, the throughput of a data can be improved by pipelining the path 1143, [22]. A pipelined data path (also called functional pipelining) is a data path in which operations of different instances can be performed concurrently.…”
Section: Considerations In Data-path Synthesismentioning
confidence: 99%
See 1 more Smart Citation
“…With very little additional hardware, the throughput of a data can be improved by pipelining the path 1143, [22]. A pipelined data path (also called functional pipelining) is a data path in which operations of different instances can be performed concurrently.…”
Section: Considerations In Data-path Synthesismentioning
confidence: 99%
“…Two consecutive tasks can be initiated with a certain interval, which is called the latency of the pipelined data path. In [22], a theorem for pipelined data path is available, stated as the following.…”
Section: Functional Pipelining (Pipelined Data Path)mentioning
confidence: 99%
“…Most of this work is primarily concerned with functional pipelining. Many synthesis tools target instruction-set architectures [10,4,3,14,15,1,9,2]; our tool, on the other hand, targets the more general class of sequential circuits. Other approaches start with a C program [8,5].…”
Section: Related Workmentioning
confidence: 99%
“…The notion of closeconnection has been dened in several ways, the most com- The input specications usually have a \V" structure, implying that the specication has a large number of operations near the input and relatively smaller number of operations near the outputs. Given that among several list-scheduling algorithms, backward and forward listscheduling algorithms are popular [10], seeding can be either done at the top or the bottom of the specication. Here, we have c hosen to do the seeding at the top.…”
Section: Concurrent P Artitioner/schedulermentioning
confidence: 99%