• Home
  • Search
  • Resource-sensitive profile-directed data flow analysis for code optimization
  • Cite Icon18
  • https://doi.org/10.1109/micro.1997.645834Copy DOI Icon

Resource-sensitive profile-directed data flow analysis for code optimization

  • Nov 22, 2002
  • Rajiv Gupta +2 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Instruction schedulers employ code motion as a means of instruction reordering to enable scheduling of instructions at points where the resources required for their execution are available. In addition, driven by the profiling data, schedulers take advantage of predication and speculation for aggressive code motion across conditional branches. Optimization algorithms for partial dead code elimination (PDE) and partial redundancy elimination (PRE) employ code sinking and hoisting to enable optimization. However, unlike instruction scheduling, these optimization algorithms are unaware of resource availability and are incapable of exploiting profiling information, speculation, and predication. In this paper we develop data flow algorithms for performing the above optimizations with the following characteristics: (i) opportunities for PRE and PDE enabled by hoisting and sinking are exploited; (ii) hoisting and sinking of a code statement is driven by availability of functional unit resources; (iii) predication and speculation is incorporated to allow aggressive hoisting and sinking; and (iv) path profile information guides predication and speculation to enable optimization.

Similar Papers
  • Research Article
  • Citations10

Partial method compilation using dynamic profile information

  • Oct 01, 2001
  • ACM SIGPLAN Notices
  • John Whaley
  • Research Article
  • Citations10

Bidirectional data flow analysis

  • Jun 01, 1999
  • ACM SIGPLAN Notices
  • Uday P Khedker +1
  • Research Article
  • Citations5

A Practical Improvement to the Partial Redundancy Elimination in SSA Form

  • Sep 30, 2008
  • Journal of Computing Science and Engineering
  • Jong-Soo Park +1
  • Book Chapter
  • Citations5

Partial Redundancy Elimination on Predicated Code

  • Jan 01, 2000
  • Jens Knoop +2
  • Conference Article
  • Citations49

Interprocedural partial redundancy elimination and its application to distributed memory compilation

  • Jun 01, 1995
  • Gagan Agrawal +2
  • Research Article
  • Citations5

Cobalt: A Language for Writing Provably-Sound Compiler Optimizations

  • May 01, 2005
  • Electronic Notes in Theoretical Computer Science
  • Sorin Lerner +2
  • Research Article
  • Citations2

Predicated Partial Redundancy Elimination using a Cost Analysis

  • Dec 01, 2003
  • Parallel Processing Letters
  • Bernhard Scholz +2
  • Conference Article
  • Citations84

Path-sensitive value-flow analysis

  • Jan 01, 1998
  • Rastisalv Bodík +1
  • Conference Article
  • Citations47

Better global scheduling using path profiles

  • Nov 01, 1998
  • Cliff Young +1
  • Research Article
  • Citations13

The El'brus-3 and MARS-M: Recent advances in Russian high-performance computing

  • Mar 01, 1992
  • The Journal of Supercomputing
  • Mikhail N Dorozhevets +1
  • Book Chapter
  • Citations13

Expansion-Based Removal of Semantic Partial Redundancies

  • Jan 01, 1999
  • Jens Knoop +2
  • Conference Article

LLVM back-end instruction scheduling optimization based on improved genetic algorithm

  • Sep 20, 2024
  • Zhening Wang +1
  • Book Chapter
  • Citations26

Compiler algorithms on if-conversion, speculative predicates assignment and predicated code optimizations

  • Jan 01, 1997
  • Jesse Z Fang
  • Conference Article
  • Citations3

Communication Optimization Algorithms based on Extend Data Flow Graph

  • Jul 01, 2007
  • Xue-Rong Gong +2
  • Research Article
  • Citations40

A Fortran 90 environment for research and prototyping of enclosure algorithms for nonlinear equations and global optimization

  • Mar 01, 1995
  • ACM Transactions on Mathematical Software
  • R Baker Kearfott
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.