Automated generation of code using backtracking parsers for attribute grammars

Authors:
Wolfgang Keller
Affiliations:
Technische Universität Münehen, Institut für Informatik, Arcisstr. - D 8000 München 2 - West-Germany
Venue:
ACM SIGPLAN Notices
Year:
1991

Citing 8
Cited 0

Attributed linear intermediate representations for retargetable code generators

Software—Practice & Experience
Affix grammar driven code generation

ACM Transactions on Programming Languages and Systems (TOPLAS)
Bulldog: a compiler for VLSI architectures

Bulldog: a compiler for VLSI architectures
Compilers: principles, techniques, and tools

Compilers: principles, techniques, and tools
Integrating code generation and peephole optimization

Acta Informatica
Efficient tree pattern matching (extended abstract): an aid to code generation

POPL '85 Proceedings of the 12th ACM SIGACT-SIGPLAN symposium on Principles of programming languages
Rewrite Systems, Pattern Matching, and Code Generation

Rewrite Systems, Pattern Matching, and Code Generation
A machine independent algorithm for code generation and its use in retargetable compilers.

A machine independent algorithm for code generation and its use in retargetable compilers.

Quantified Score

Hi-index	0.00

Visualization

Abstract

Code generation with the Graham Glanville scheme can be improved, using backtracking parsers for attribute grammars. This eliminates semantic and syntactic blocking during code selection and guarantees the cheapest code possible, if desired. Multipass translation allows the use of nontrivial register allocation strategies. The code generator generator system BECC is a LALR(1) based backtracking parser generator for a derivate of S-attributed grammars which provides several interpreters that differ in their abilities and performance.