Post-compilation analysis and power reduction

Simple item page
dc.contributor.advisorCooper, Keith D.
dc.creatorWaterman, Todd Ryan
dc.date.accessioned2009-06-04T07:08:16Z
dc.date.available2009-06-04T07:08:16Z
dc.date.issued2002
dc.description.abstractOptimization outside of traditional frameworks is emerging as a new research focus in the compiler construction community. Scheduled assembly code is one area of increased interest. Optimization cannot be performed without a control-flow graph (CFG), and current CFG construction algorithms can fail on scheduled code. We present a new construction algorithm that correctly constructs CFGs and permits meaningful optimization for scheduled code. One potential post-compilation optimization is reducing power consumption by minimizing switching activity on the instruction bus. We designed and implemented an algorithm that attempts to minimize switching activity by renaming registers for Texas Instruments' TMS320C6200 processor. We gathered results using a power simulator developed inside Texas Instruments. We determine that reducing bit transitions on the instruction bus is not a profitable technique for reducing the power consumption of this particular microprocessor.
dc.format.extent37 p.en_US
dc.format.mimetypeapplication/pdf
dc.identifier.callnoTHESIS COMP.SCI. 2002 WATERMAN
dc.identifier.citationWaterman, Todd Ryan. "Post-compilation analysis and power reduction." (2002) Master’s Thesis, Rice University. <a href="https://hdl.handle.net/1911/17560">https://hdl.handle.net/1911/17560</a>.
dc.identifier.urihttps://hdl.handle.net/1911/17560
dc.language.isoeng
dc.rightsCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.
dc.subjectElectronics
dc.subjectElectrical engineering
dc.subjectComputer science
dc.titlePost-compilation analysis and power reduction
dc.typeThesis
dc.type.materialText
thesis.degree.departmentComputer Science
thesis.degree.disciplineEngineering
thesis.degree.grantorRice University
thesis.degree.levelMasters
thesis.degree.nameMaster of Science
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
1408719.PDF
Size:
1.34 MB
Format:
Adobe Portable Document Format
Download
Collections
Rice University Electronic Theses and Dissertations