https://github.com/sdasgup3/partial-redundancy-elimination

Partial Redundancy Elimination Pass in LLVM
https://github.com/sdasgup3/partial-redundancy-elimination

bitvector common-subexpression-elimination compiler-optimization data-flow-analysis llvm loop-invariants

Last synced: about 1 month ago
JSON representation

Partial Redundancy Elimination Pass in LLVM

• Host: GitHub
• URL: https://github.com/sdasgup3/partial-redundancy-elimination
• Owner: sdasgup3
• License: mit
• Created: 2014-03-27T09:40:09.000Z (almost 11 years ago)
• Default Branch: master
• Last Pushed: 2019-05-20T17:21:11.000Z (over 5 years ago)
• Last Synced: 2023-08-14T19:16:38.814Z (over 1 year ago)
• Topics: bitvector, common-subexpression-elimination, compiler-optimization, data-flow-analysis, llvm, loop-invariants
• Language: TeX
• Homepage:
• Size: 22.7 MB
• Stars: 11
• Watchers: 3
• Forks: 4
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
[![DOI](https://zenodo.org/badge/18170897.svg)](https://zenodo.org/badge/latestdoi/18170897)

Partial Redundancy Elimination (PRE)

====================================

PRE is a compiler optimization that eliminates

expressions that are redundant on some but not necessarily all paths through a

program. We implement a PRE optimization pass in the LLVM framework, and evaluate the benefits 

on a variety of applications. PRE is a powerful technique that subsumes 

Common Subexpression Elimination (CSE) and Loop Invariant Code Motion (LICM), and hence has a potential to 

greatly improve performance.

Commands

-------

opts_lcm = "-mem2reg -loop-rotate -reassociate -lcm  -mem2reg -simplifycfy; 

$opt -load $load_pass  $opts_lcm   $test.bc -o  $test-lcm.bc