Debugging with GDB (February 2008)
Table Of Contents
- Summary of GDB
- A Sample GDB Session
- Getting In and Out of GDB
- GDB Commands
- Running Programs Under GDB
- Stopping and Continuing
- Examining the Stack
- Examining Source Files
- Examining Data
- Using GDB with Different Languages
- Examining the Symbol Table
- Altering Execution
- GDB Files
- Specifying a Debugging Target
- HP-UX Configuration-Specific Information
- Summary of HP Enhancements to GDB
- HP-UX dependencies
- Supported Platforms and Modes
- HP-UX targets
- Support for Alternate root
- Specifying object file directories
- Fix and continue debugging
- Inline Support
- Debugging Macros
- Debugging Memory Problems
- When to suspect a memory leak
- Memory debugging restrictions
- Memory Debugging Methodologies
- Debugging Memory in Interactive Mode
- Debugging Memory in Batch Mode
- Debugging Memory Interactively After Attaching to a Running Process
- Configuring memory debugging settings
- Scenarios in memory debugging
- Stop when freeing unallocated or deallocated blocks
- Stop when freeing a block if bad writes occurred outside block boundary
- Stop when a specified block address is allocated or deallocated
- Scramble previous memory contents at malloc/free calls
- Detect dangling pointers and dangling blocks
- Detect in-block corruption of freed blocks
- Specify the amount of guard bytes for every block of allocated memory
- Comparison of Memory Debugging Commands in Interactive Mode and Batch Mode
- Heap Profiling
- Memory Checking Analysis for User Defined Memory Management Routines
- Commands to track the change in data segment value
- Thread Debugging Support
- Debugging MPI Programs
- Debugging multiple processes ( programs with fork and vfork calls)
- Debugging Core Files
- Printing the Execution Path Entries for the Current Frame or Thread
- Invoking GDB Before a Program Aborts
- Aborting a Command Line Call
- Instruction Level Stepping
- Enhanced support for watchpoints and breakpoints
- Debugging support for shared libraries
- Language support
- Enhanced Java Debugging Support
- Commands for Examining Java Virtual Machine(JVM) internals
- Support for stack traces in Java, C, and C++ programs
- Support for 64-bit Java, C, aC++ stack unwinding
- Enhanced support for C++ templates
- Support for __fpreg data type on IPF
- Support for _Complex variables in HP C
- Support for debugging namespaces
- Command for evaluating the address of an expression
- Viewing Wide Character Strings
- Support for output logging
- Getting information from a non-debug executable
- Debugging optimized code
- Visual Interface for WDB
- Starting and stopping Visual Interface for WDB
- Navigating the Visual Interface for WDB display
- Specifying foreground and background colors
- Using the X-window graphical interface
- Using the TUI mode
- Changing the size of the source or debugger pane
- Using commands to browse through source files
- Loading source files
- Editing source files
- Editing the command line and command-line history
- Saving the contents of a debugging session to a file
- Support for ddd
- Support for XDB commands
- GNU GDB Logging Commands
- Support for command line calls in a stripped executable
- Displaying the current block scope information
- Linux support
- The HP-UX Terminal User Interface
- XDB to WDB Transition Guide
- By-function lists of XDB commands and HP WDB equivalents
- Overall breakpoint commands
- XDB data formats and HP WDB equivalents
- XDB location syntax and HP WDB equivalents
- XDB special language operators and HP WDB equivalents
- XDB special variables and HP WDB equivalents
- XDB variable identifiers and HP WDB equivalents
- Alphabetical lists of XDB commands and HP WDB equivalents
- Controlling GDB
- Canned Sequences of Commands
- Using GDB under gnu Emacs
- GDB Annotations
- The gdb/mi Interface
- Function and purpose
- Notation and terminology
- gdb/mi Command Syntax
- gdb/mi compatibility with CLI
- gdb/mi output records
- gdb/mi command description format
- gdb/mi breakpoint table commands
- gdb/mi Data manipulation
- gdb/mi program control
- Miscellaneous GDB commands in gdb/mi
- gdb/mi Stack Manipulation Commands
- gdb/mi Symbol query commands
- gdb/mi Target Manipulation Commands
- gdb/mi thread commands
- gdb/mi tracepoint commands
- gdb/mi variable objects
- Reporting Bugs in GDB
- Installing GDB
- Index
28 Debugging with GDB
If you exit GDB or use the run command while you have an attached process, you kill
that process. By default, GDB asks for confirmation if you try to do either of these things;
you can control whether or not you need to confirm by using the set confirm command
(see Section 17.6 [Optional warnings and messages], page 252).
NOTE: When GDB attaches to a running program you may get a message
saying "Attaching to process #nnnnn failed."
The most likely cause for this message is that you have attached to a process
that was started across an NFS mount. Versions of the HP-UX kernel before
11.x have a restriction that prevents a debugger from attaching to a process
started from an NFS mount, unless the mount was made non-interruptible
with the -nointr flag, see mount(1).
4.8 Killing the child process
Following command is used to kill the child process:
kill Kill the child process in which your program is running under GDB.
The kill command is useful if you wish to debug a core dump instead of a running
process. GDB ignores any core dump file while your program is running.
On some operating systems, a program cannot be executed outside GDB while you have
breakpoints set on it inside GDB. You can use the kill command in this situation to permit
running your program outside the debugger.
The kill command is also useful if you wish to recompile and relink your program,
since on many systems it is impossible to modify an executable file while it is running in a
process. In this case, when you next type run, GDB notices that the file has changed, and
reads the symbol table again (while trying to preserve your current breakpoint settings).
4.9 Debugging programs with multiple threads
In some operating systems, such as HP-UX and Solaris, a single program may have more
than one thread of execution. The precise semantics of threads differ from one operating
system to another, but in general the threads of a single program are akin to multiple
processes—except that they share one address space (that is, they can all examine and
modify the same variables). On the other hand, each thread has its own registers and
execution stack, and private memory.
GDB provides these facilities for debugging multi-thread programs:
• automatic notification of new threads
• thread-specific breakpoints
Warning: These facilities are not yet available on every GDB configura-
tion where the operating system supports threads. If your GDB does not
support threads, these commands have no effect. For example, a system
without thread support shows no output from ‘info threads’, and always
rejects the thread command, like this: