Compiler Construction

Name: Compiler Construction | 21st International Conference, CC 2012, Held as Part of the European Joint Conferences on Theory and Practice of Software, ETAPS 2012, Tallinn, Estonia, March 24 -- April 1, 2012, Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

21st International Conference, CC 2012, Held as Part of the European Joint Conferences on Theory and Practice of Software, ETAPS 2012, Tallinn, Estonia, March 24 -- April 1, 2012, Proceedings

Michael O'Boyle(Herausgeber*in)

Springer (Verlag)

Erschienen am 22. März 2012

XVI, 245 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-28652-0 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Title
Foreword
Preface
Organization
Table of Contents
GPU Optimisation
Improving Performance of OpenCL on CPUs
Introduction
Contributions
Structure of This Paper
OpenCL Driver Implementation
SIMD Vectorization
Runtime Callbacks
Continuation-Based Synchronization
Exploiting Uniform Computations
Uniform Value Analysis
Divergence Analysis
Optimizations
Related Work
Experimental Evaluation
Benchmarks
Conclusion
References
Automatic Restructuring of GPU Kernels for Exploiting Inter-thread Data Locality
Introduction
Related Work
An Example
Automatic Thread Coarsening
Notation and Terminology
Dependence Analysis
Safety Analysis
Profitability Analysis
Code Transformation
CREST : CUDA Kernel Restructuring and AutoTuning
Kernel Extraction
Code Restructurer
Autotuning Support
Experimental Results
Conclusions
References
Program Analysis
Programming Paradigm Driven Heap Analysis
Introduction
Abstract Heap Domain
Concrete Heaps
Abstract Heaps
Abstraction Relation
Example Heap
Normal Form
Interprocedural Analysis
Unique Fresh Cutpoints
Context-Sensitivity Heuristics
Complete Partial Context-Sensitive Call Analysis
Experimental Evaluation
Related Work
Conclusion
References
Parallel Replication-Based Points-To Analysis
Introduction
Motivation and Background
Replication-Based Analysis
Algorithm Outline
Replication
Parallel Points-to Analysis Algorithm
Load Balancing
Parallel Online Cycle Elimination
Reducing the Number of Copies
Limited Scheduling
Context-Sensitive Analysis
Experimental Evaluation
Context-Insensitive Analysis
Context-Sensitive Analysis
Related Work
Conclusion
References
A New Method for Program Inversion
Introduction
Related Work
Problem Setup
Reversing Functions without Loops
Framework Overview
Building the Value Search Graph
The Route Graph
Searching the Value Search Graph
Costs in Route Graphs.
Search Algorithm.
Instrumentation and Code Generation
Representing CFG Path Sets.
Recording CFG Paths.
Inserting State Saving Statements.
Building a CFG for the Reverse Function.
Generating Code.
Handling Loops
Experiment Results
Conclusion and Future Work
References
Analytical Bounds for Optimal Tile Size Selection
Introduction
Motivation and Case Study
Background
DL: Distinct Lines
Parametric Tiling
ML: Minimum Working Set Lines
Operational Definition of ML
Model of Computation
Distance in Tiled Iteration Space
Temporal and Spatial Reuse Distance
Computation of ML
Example
Bounding the Search Space by Using DL and ML
Capacity Constraint for Intra-tile Reuse
Capacity Constraint for Inter-tile Reuse
Empirical Search within Bounded Search Space for Single-Level Tiling
Compiler Pass for Bounded Search Space
Experimental Results
Performance Distribution of Different Tile Sizes
Search Space Reduction by DL/ML Model
Summary of Experiments
Related Work
Conclusion
References
Objects and Components
Static Detection of Unsafe Component Loadings
Introduction
Static Detection Algorithm
Background
Detailed Analysis
Empirical Evaluation
Implementation
Evaluation Setup and Results
Related Work
Conclusion and Future Work
References
Object Model Construction for Inheritance in C++ and Its Applications to Program Analysis
Introduction
Rossie-Friedman Sub-objects
Clarifier
CHROME Model
Implementation and Experiments
Complexity of the Lowered C Programs
Effectiveness of Lowering for Software Verification
Applicability in Other Verification Tools
Related Work
Conclusions
References
GC-Safe Interprocedural Unboxing
Introduction
GC Safety
A Core Language for GC Safety
Operational Semantics
Traceability
Flow Analysis
Unboxing
The Unboxing Optimization
Acceptable Unboxings
Construction of an Acceptable Unboxing
Related Work
References
Dynamic Analysis and Runtime Support
Compiler Support for Value-Based Indirect Branch Prediction
Introduction
Value Based BTB Indexing (VBBI) Background
Compiler Analysis for VBBI
Optimizations for Improving VBBI Prediction Accuracy
Compiler Optimizations
Instruction Hoisting.
Function Inlining.
Inter-procedural Dataflow Analysis.
Hardware Optimization
Simulation Methodology
Results
VBBI versus Traditional BTB
VBBI versus Tagged Target Cache
Previous Work
Conclusion
References
Compiler Support for Fine-Grain Software-Only Checkpointing
Introduction
Basic Checkpointing
Optimizations
Buffering Implementation
Checkpoint-Enabled Applications
Checkpoint Support for Debugging
Checkpoint Support for Automated Back-Tracking
Evaluation
Checkpoint Region Selection
Comparison with Existing Checkpointing Solutions
Optimization Effectiveness
Overhead of Back-Tracking Support
Related Work
Conclusion
References
VMAD: An Advanced Dynamic Program Analysis and Instrumentation Framework
Introduction
Framework Overview
The Virtual Machine VMAD
Preparing the Code at Compile Time Using LLVM
Illustrating Applications
Analyzing Memory Accesses in Loop Nests
Dynamic Dependence Analysis
Dynamic Version Selection
Other Possible Applications
Related Work
Conclusion
References
Sambamba: A Runtime System for Online Adaptive Parallelization
Introduction
TheSambamba Framework
Adaptive Parallelization
Data Dependence Analysis
Parallel CFG Construction
Scheduling and Parallel Execution
State of the Project
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken