The exploitation of pipeline parallelism by compile time dataflow analysis

The automatic and implicit transformation of sequential instruction streams, which execute efficiently for pipelined architectures is the subject of this paper. This paper proposes a method which maximizes the parallel performance of an instruction pipeline by detecting and eliminating specific pipeline hazards known as resource conflicts. The detection of resource conflicts is accomplished with data dependence analysis, while the elimination of resource conflicts is accomplished by instruction stream code transformation. The transformation of instruction streams is guided by data dependence analysis, and dependence graphs. This thesis is based on the premise that the elimination of resource conflicts is synonymous with the elimination of specific arcs in the dependence graph. Examples will be given showing how detection and elimination of resource conflicts is possible through compiler optimization.

Name thesis:degree_name

Master of Science (MS)

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor dc:publisher

University of Nevada, Las Vegas

Year

1991

Author dc:creator

• Lombardo, Joseph Michael

Contributors dc:contributor

• Evangelos Yfantis

Statement dc:rights

• IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/

Language dc:language

English

Identifier

https://oasis.library.unlv.edu/rtds/149

OAI identifier oai:identifier

oai:oasis.library.unlv.edu:rtds-1148