Valgrind User Manual

Table of Contents

1. Introduction

1.1. An Overview of Valgrind

1.2. How to navigate this manual

2. Using and understanding the Valgrind core

2.1. What Valgrind does with your program

2.2. Getting started

2.3. The Commentary

2.4. Reporting of errors

2.5. Suppressing errors

2.6. Command-line flags for the Valgrind core

2.6.1. Tool-selection option

2.6.2. Basic Options

2.6.3. Error-related options

2.6.4. malloc()-related Options

2.6.5. Uncommon Options

2.6.6. Debugging Valgrind Options

2.6.7. Setting default Options

2.7. Support for Threads

2.8. Handling of Signals

2.9. Building and Installing Valgrind

2.10. If You Have Problems

2.11. Limitations

2.12. An Example Run

2.13. Warning Messages You Might See

3. Using and understanding the Valgrind core: Advanced Topics

3.1. The Client Request mechanism

3.2. Function wrapping

3.2.1. A Simple Example

3.2.2. Wrapping Specifications

3.2.3. Wrapping Semantics

3.2.4. Debugging

3.2.5. Limitations - control flow

3.2.6. Limitations - original function signatures

3.2.7. Examples

4. Memcheck: a heavyweight memory checker

4.1. Kinds of bugs that Memcheck can find

4.2. Command-line flags specific to Memcheck

4.3. Explanation of error messages from Memcheck

4.3.1. Illegal read / Illegal write errors

4.3.2. Use of uninitialised values

4.3.3. Illegal frees

4.3.4. When a block is freed with an inappropriate deallocation function

4.3.5. Passing system call parameters with inadequate read/write permissions

4.3.6. Overlapping source and destination blocks

4.3.7. Memory leak detection

4.4. Writing suppression files

4.5. Details of Memcheck's checking machinery

4.5.1. Valid-value (V) bits

4.5.2. Valid-address (A) bits

4.5.3. Putting it all together

4.6. Client Requests

4.7. Memory Pools: describing and working with custom allocators

4.8. Debugging MPI Parallel Programs with Valgrind

4.8.1. Building and installing the wrappers

4.8.2. Getting started

4.8.3. Controlling the wrapper library

4.8.4. Abilities and limitations

4.8.5. Writing new wrappers

4.8.6. What to expect when using the wrappers

5. Cachegrind: a cache and branch profiler

5.1. Cache and branch profiling

5.1.1. Overview

5.1.2. Cache simulation specifics

5.1.3. Branch simulation specifics

5.2. Profiling programs

5.2.1. Output file

5.2.2. Cachegrind options

5.2.3. Annotating C/C++ programs

5.2.4. Annotating assembly code programs

5.2.5. Forking Programs

5.3. cg_annotate options

5.3.1. Warnings

5.3.2. Things to watch out for

5.3.3. Accuracy

5.4. Merging profiles with cg_merge

5.5. Acting on Cachegrind's information

5.6. Implementation details

5.6.1. How Cachegrind works

5.6.2. Cachegrind output file format

6. Callgrind: a call graph profiler

6.1. Overview

6.1.1. Functionality

6.1.2. Basic Usage

6.2. Advanced Usage

6.2.1. Multiple profiling dumps from one program run

6.2.2. Limiting the range of collected events

6.2.3. Avoiding cycles

6.2.4. Forking Programs

6.3. Command line option reference

6.3.1. Miscellaneous options

6.3.2. Dump creation options

6.3.3. Activity options

6.3.4. Data collection options

6.3.5. Cost entity separation options

6.3.6. Cache simulation options

6.4. Callgrind specific client requests

7. Helgrind: a thread error detector

7.1. Overview

7.2. Detected errors: Misuses of the POSIX pthreads API

7.3. Detected errors: Inconsistent Lock Orderings

7.4. Detected errors: Data Races

7.4.1. A Simple Data Race

7.4.2. Helgrind's Race Detection Algorithm

7.4.3. Interpreting Race Error Messages

7.5. Hints and Tips for Effective Use of Helgrind

7.6. Helgrind Options

7.7. A To-Do List for Helgrind

8. DRD: a thread error detector

8.1. Background

8.1.1. Multithreaded Programming Paradigms

8.1.2. POSIX Threads Programming Model

8.1.3. Multithreaded Programming Problems

8.1.4. Data Race Detection

8.2. Using DRD

8.2.1. Command Line Options

8.2.2. Detected Errors: Data Races

8.2.3. Detected Errors: Lock Contention

8.2.4. Detected Errors: Misuse of the POSIX threads API

8.2.5. Client Requests

8.2.6. Debugging GNOME Programs

8.2.7. Debugging Qt Programs

8.2.8. Debugging Boost.Thread Programs

8.2.9. Debugging OpenMP Programs

8.2.10. DRD and Custom Memory Allocators

8.2.11. DRD Versus Memcheck

8.2.12. Resource Requirements

8.2.13. Hints and Tips for Effective Use of DRD

8.3. Using the POSIX Threads API Effectively

8.3.1. Mutex types

8.3.2. Condition variables

8.3.3. pthread_cond_timedwait() and timeouts

8.3.4. Assigning names to threads

8.4. Limitations

8.5. Feedback

9. Massif: a heap profiler

9.1. Heap profiling

9.2. Using Massif

9.2.1. An Example Program

9.2.2. The Output Preamble

9.2.3. The Output Graph

9.2.4. The Snapshot Details

9.2.5. Forking Programs

9.3. Massif Options

9.4. ms_print Options

9.5. Massif's output file format

10. Ptrcheck: an (experimental) pointer checking tool

10.1. Overview

10.2. Ptrcheck Options

10.3. How Ptrcheck Works: Heap Checks

10.4. How Ptrcheck Works: Stack and Global Checks

10.5. Comparison with Memcheck

10.6. Limitations

10.7. Still To Do: User Visible Functionality

10.8. Still To Do: Implementation Tidying

11. Nulgrind: the "null" tool

12. Lackey: a simple profiler and memory tracer

12.1. Overview

12.2. Lackey Options

13. Writing a New Valgrind Tool

13.1. Introduction

13.1.1. Tools

13.2. Writing a Tool

13.2.1. How tools work

13.2.2. Getting the code

13.2.3. Getting started

13.2.4. Writing the code

13.2.5. Initialisation

13.2.6. Instrumentation

13.2.7. Finalisation

13.2.8. Other Important Information

13.2.9. Words of Advice

13.3. Advanced Topics

13.3.1. Suppressions

13.3.2. Documentation

13.3.3. Regression Tests

13.3.4. Profiling

13.3.5. Other Makefile Hackery

13.3.6. Core/tool Interface Versions

13.4. Final Words