Computer Sciences Dept, University of Wisconsin-Madison

Discussion Group for Compiler Optimization (Summer 1997)

[   Reading list   |    Participants  |   Resources   ]      

An informal series of discussions was held during the summer of '97 at the Computer Sciences Department, University of Wisconsin-Madison. The topic of the discussions was compiler optimizations (mainly for instruction-level parallelism).

Reading list

• Scheduling & Speculative/Predicated Execution
  
               MAY 27
  
  

  • Lam : Software Pipelining : An Effective Scheduling Technique for VLIW Machines (PLDI'88)
  • Warter et al : Enhanced Modulo Scheduling for Loops with Conditional Branches (MICRO'25)
    
    JUNE 10
    
    
  • Chang et al : Three Superblock Scheduling Models for Superscalar and Superpipelined Processors (UIUC CRHC Tech Report CRHC-91-29)
  • Mahlke et al : Effective Compiler Support for Predicated Execution Using the Hyperblock (MICRO'25)
    
    JUNE 24
    
    
  • Kerns & Eggers : Balanced Scheduling : Instruction Scheduling  When Memory Latency is Uncertain (PLDI'93)
  • Lo & Eggers : Improving Balanced Scheduling with Compiler Optimizations that increase ILP (PLDI'95)
    
    JULY 1
    
    
  • Smith, Horowitz & Lam : Efficient Superscalar Performance Through Boosting (ASPLOS'92)
  • Mahlke et al : Sentinel Scheduling for VLIW and Superscalar Processors (ASPLOS'92)
    
    

• Branch optimizations
  
               JULY 8
  
  

  • Ball & Larus : Branch Prediction for Free (PLDI'93)
  • Patterson : Accurate Static Branch Prediction by Value Propagation (PLDI'95)
    
    Another interesting paper in this genre :
    Calder, Grunwald et al : Corpus-based Static Branch Prediction (PLDI '95)
    
    JULY 15
    
    
  • Mueller & Whalley : Avoiding Conditional Branches by Code Replication (PLDI'95)
  • Young & Smith : Improving the Accuracy of Static Branch Prediction Using Branch  Correlation (ASPLOS'94)
    
    

• Prefetching
  
               JULY 22
  
  

  • Rogers & Li : Software Support for Speculative Loads (ASPLOS'92)
  • Mowry, Lam & Gupta : Design and Evaluation of a Compiler Algorithm for Prefetching (ASPLOS'92)
    
    JULY 29
    
    
  • Luk & Mowry : Compiler-based Prefetching for Recursive Data Structures (ASPLOS'96)
    
    

• Cache and memory-aware optimizations
  
  
                 AUGUST 5
  
  

  • Wolfe & Lam : A Data Locality Optimizing Algorithm (PLDI'91)
  • Grunwald, Zorn & Henderson : Improving the Cache Locality of Memory Allocation (PLDI'93)
    
     AUGUST 12
    
    
  • Chi & Dietz : Unified Management of Registers and Cache Using Liveness and Cache Bypass (PLDI'89)
  • Barrett & Zorn : Using Lifetime Predictors to Improve Memory Allocation Performance (PLDI'93)
    
    AUGUST 19 
     
  • Pettis & Hansen : Profile Guided Code Positioning (PLDI'90)
  • Calder & Grunwald : Reducing Branch Costs via Branch Alignment (ASPLOS'94)
    
    

Participants :


                         Faculty                                             Students

                   Charles Fischer        Milo Martin                              Craig Zilles

                    Jim Larus                 Manoj Plakal                           S Subramanya Sastry

                   Susan Horwitz           Amir Roth                                Ravi Rajwar

                   Tom Reps                    Chris Lukas                             Monika ten Bruggencate

                                                          Dave Melski       

Resources :

• Bibliographies :
  
  
  • PLDI ('88 - '96)
  • MICRO ('89 - '96)
  • PPoPP ('90 - '95)
  • ISCA ('83 - '96)
  • ASPLOS ('82 - '96)
  • SEL-HPC Archive (Compilers & Interpreters section)
  • ACAPS Bibliographies
  • Collection of Computer Science Bibliographies : PL/Compiler section
    
    • Optimizing compilers
    • Optimization in compiling
      
  • Rau & Fisher's ILP/Compiling for ILP bibliography
    
    
• Research groups :
  
  
  • Stanford : SUIF
  • UIUC : IMPACT, Polaris
  • Washington, Seattle
  • Rice
  • Harvard : HUBE, Morph
  • McGill : McCat
    
    
• Courses in compiler optimization elsewhere :
  
  
  • An MIT reading group on compiler optimizations (a Harvard section as well).
  • Stanford : CS343 - Topics in Compilers, Monica Lam
  • Maryland : CMSC732 - Compiling for High-Performance Architectures, C-W Tseng
  • UMass, Amherst : 791U - Compiler Dependence and Pointer Analysis, Kathryn McKinley (PostScript of readings list)

Last updated on