Page 45

Update item information
Publication Type technical report
School or College College of Engineering
Department Computing, School of
Creator Duggan, Gerald Patrick
Title Optimization as Transformations: A Compiler Paradigm
Date 1985-08
Description This thesis describes the high-level transformation phase of a prototype PSL compiler. EPIC The transformations are separated into two classes: machine dependent code generation, and machine independent code optimization. The code generation phase transforms an intermediate representation of a program into a form that is easy to translate into object code. The optimizing transformations use the results of the generation transformations, along with predefined knowledge of machine independent optimization, to improve the efficiency of the tailored code. A facility for specifying machine dependent optimization also exists. The entire transformation phase is viewed as two interrelated mechanisms operations on the code, which specify a syntactic transformation, and operations on attributes of various elements of the code, which specify semantic transformations. A method for the implementation of these mechanisms is investigated.
Type Text
Subject PSL compiler; EPIC; machine dependent code generator ; machine independent code optimization
Language eng
Bibliographic Citation Duggan, P. G. (1985). Optimization as transformations: a compiler paradigm.
Series University of Utah Computer Science Technical Report
Relation is Part of ARPANET
Format Medium application/pdf
Format Extent 29,285,271 bytes
File Name Duggan-Optimization_As_Transformations.pdf
Conversion Specifications Original scanned with Kirtas 2400 and saved as 400 ppi uncompressed TIFF. PDF generated by Adobe Acrobat Pro X for CONTENTdm display
ARK ark:/87278/s63799x3
Setname ir_computersa
Date Created 2016-04-27
Date Modified 2016-05-17
ID 98082
Reference URL

Page Metadata

Title Page 45
File Name Duggan-Optimization_As_Transformations.pdf
Setname ir_computersa
Date Created 2016-04-27
Date Modified 2016-04-27
ID 98029
Reference URL