scispace - formally typeset
Journal ArticleDOI

An API for Runtime Code Patching

Reads0
Chats0
TLDR
The authors present a postcompiler program manipulation tool called Dyninst, which provides a C++ class library for program instrumentation that permits machine-independent binary instrumentation programs to be written.
Abstract
The authors present a postcompiler program manipulation tool called Dyninst, which provides a C++ class library for program instrumentation. Using this library, it is possible to instrument and modify application programs during execution. A unique feature of this library is that it permits machine-independent binary instrumentation programs to be written. The authors describe the interface that a tool sees when using this library. They also discuss three simple tools built using this interface: a utility to count the number of times a function is called, a program to capture the output of an already running program to a file, and an implementation of conditional breakpoints. For the conditional breakpoint example, the authors show that by using their interface compared with gdb, they are able to execute a program with conditional breakpoints up to 900 times faster.

read more

Content maybe subject to copyright    Report

Citations
More filters
Dissertation

Application Specific, Automatic Distributed Evaluation of Performance Data on the Grid

TL;DR: This thesis addresses a novel method of automatic, distributed evaluation of performance data in such environments which can also be used in different tools and an augmented dataflow model based on the classical dataflow technology is developed.
Proceedings ArticleDOI

Propeller: A Profile Guided, Relinking Optimizer for Warehouse-Scale Applications

TL;DR: Propeller as mentioned in this paper proposes a relinking optimizer for warehouse scale work-loads, which uses basic block sections to enable a new approach to post link optimizers without disassembly.

Performance Monitoring, Analysis, and Real-Time Introspection on Large-Scale Parallel Systems

TL;DR: High-Performance Computing has become an important scientific driver and a wide variety of research ranging for example from drug design to climate modelling is nowadays performed in HPC systems.
Dissertation

Foo's To Blame: Techniques For Mapping Performance Data To Program Variables

TL;DR: The first application of this analysis is targeted at mapping performance data to high level data structures with multiple levels of abstraction, and the second application is a method for mapping cache miss information to variables.
Journal Article

A Parallel Debugger with Fast Conditional Breakpoint

TL;DR: The fundamental principles for finding and implementing fast conditional breakpoints are introduced and the design problems associated with their implementations such as code generation, code instrumentation and source program mapping are discussed.
References
More filters
Proceedings ArticleDOI

ATOM: a system for building customized program analysis tools

TL;DR: ATOM as mentioned in this paper is a single framework for building a wide range of customized program analysis tools, including block counting, profiling, dynamic memory recording, instruction and data cache simulation, pipeline simulation, evaluating branch prediction, and instruction scheduling.
Journal ArticleDOI

The Paradyn parallel performance measurement tool

TL;DR: Dynamic instrumentation lets us defer insertion until the moment it is needed (and remove it when it is no longer needed); Paradyn's Performance Consultant decides when and where to insert instrumentation.
Proceedings ArticleDOI

EEL: machine-independent executable editing

TL;DR: EEL supports a machine- and system-independent editing model that enables tool builders to modify an executable without being aware of the details of the underlying architecture or operating system or being concerned with the consequences of deleting instructions or adding foreign code.
Proceedings Article

Instrumentation and optimization of Win32/intel executables using Etch

TL;DR: Etch is a general-purpose tool for rewriting arbitrary Win32/x86 binaries without requiring source code and some of the tools that are built using it are described, including a hierarchical call graph profiler and an instruction layout optimization tool.
Proceedings ArticleDOI

Dynamic program instrumentation for scalable performance tools

TL;DR: This work presents a new technique called 'dynamic instrumentation' that provides efficient, scalable, yet detailed data collection for large-scale parallel applications and includes recommendations to operating system designers, compiler writers, and computer architects about the features necessary to permit efficient monitoring of large- scale parallel systems.
Related Papers (5)