GDB Manual

Summary of GDB

The purpose of a debugger such as GDB is to allow you to execute another program while examining what is going on inside it. We call the other program "your program" or "the program being debugged".

GDB can do four kinds of things (plus other things in support of these):

1. Start the program, specifying anything that might affect its behavior.

2. Make the program stop on specified conditions.

3. Examine what has happened, when the program has stopped, so that you can see bugs happen.

4. Change things in the program, so you can correct the effects of one bug and go on to learn about another without having to recompile first.

GDB can be used to debug programs written in C and C++. Pascal support is being implemented, and Fortran support will be added when a GNU Fortran compiler is written.

GDB General Public License		The GDB General Public License gives you permission to redistribute GDB on certain terms; and also explains that there is no warranty.
1. GDB Input Conventions		GDB command syntax and input conventions.
2. Specifying GDB's Files		Specifying files for GDB to operate on.
3. Options and Arguments for GDB		GDB arguments and options.
4. Compiling Your Program for Debugging		Compiling your program so you can debug it.
5. Running Your Program Under GDB		Running your program under GDB.
6. Stopping and Continuing		Making your program stop. Why it may stop. What to do then.
7. Examining the Stack		Examining your program's stack.
8. Examining Source Files		Examining your program's source files.
9. Examining Data		Examining data in your program.
10. Examining the Symbol Table		Examining the debugger's symbol table.
11. Altering Execution		Altering things in your program.
12. Canned Sequences of Commands		Canned command sequences for repeated use.
13. Using GDB under GNU Emacs		Using GDB through GNU Emacs.
14. Remote Kernel Debugging		Remote kernel debugging across a serial line.
Command Index		Index of GDB commands.
Concept Index		Index of GDB concepts.

GDB General Public License

The license agreements of most software companies keep you at the mercy of those companies. By contrast, our general public license is intended to give everyone the right to share GDB. To make sure that you get the rights we want you to have, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. Hence this license agreement.

Specifically, we want to make sure that you have the right to give away copies of GDB, that you receive source code or else can get it if you want it, that you can change GDB or use pieces of it in new free programs, and that you know you can do these things.

To make sure that everyone has such rights, we have to forbid you to deprive anyone else of these rights. For example, if you distribute copies of GDB, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must tell them their rights.

Also, for our own protection, we must make certain that everyone finds out that there is no warranty for GDB. If GDB is modified by someone else and passed on, we want its recipients to know that what they have is not what we distributed, so that any problems introduced by others will not reflect on our reputation.

Therefore we (Richard Stallman and the Free Software Foundation, Inc.) make the following terms which say what you must do to be allowed to distribute or change GDB.

Copying Policies

1. You may copy and distribute verbatim copies of GDB source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each file a valid copyright notice "Copyright (C) 1988 Free Software Foundation, Inc." (or with whatever year is appropriate); keep intact the notices on all files that refer to this License Agreement and to the absence of any warranty; and give any other recipients of the GDB program a copy of this License Agreement along with the program. You may charge a distribution fee for the physical act of transferring a copy.

2. You may modify your copy or copies of GDB source code or any portion of it, and copy and distribute such modifications under the terms of Paragraph 1 above, provided that you also do the following:
   • cause the modified files to carry prominent notices stating that you changed the files and the date of any change; and

   • cause the whole of any work that you distribute or publish, that in whole or in part contains or is a derivative of GDB or any part thereof, to be licensed at no charge to all third parties on terms identical to those contained in this License Agreement (except that you may choose to grant more extensive warranty protection to some or all third parties, at your option).

   • if the modified program serves as a debugger, cause it, when started running in the simplest and usual way, to print an announcement including a valid copyright notice "Copyright (C) 1988 Free Software Foundation, Inc." (or with the year that is appropriate), saying that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License Agreement.

   • You may charge a distribution fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee.

   Mere aggregation of another unrelated program with this program (or its derivative) on a volume of a storage or distribution medium does not bring the other program under the scope of these terms.

3. You may copy and distribute GDB (or a portion or derivative of it, under Paragraph 2) in object code or executable form under the terms of Paragraphs 1 and 2 above provided that you also do one of the following:
   • accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Paragraphs 1 and 2 above; or,

   • accompany it with a written offer, valid for at least three years, to give any third party free (except for a nominal shipping charge) a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Paragraphs 1 and 2 above; or,

   • accompany it with the information you received as to where the corresponding source code may be obtained. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form alone.)

   For an executable file, complete source code means all the source code for all modules it contains; but, as a special exception, it need not include source code for modules which are standard libraries that accompany the operating system on which the executable file runs.

4. You may not copy, sublicense, distribute or transfer GDB except as expressly provided under this License Agreement. Any attempt otherwise to copy, sublicense, distribute or transfer GDB is void and your rights to use GDB under this License agreement shall be automatically terminated. However, parties who have received computer software programs from you with this License Agreement will not have their licenses terminated so long as such parties remain in full compliance.

5. If you wish to incorporate parts of GDB into other free programs whose distribution conditions are different, write to the Free Software Foundation. We have not yet worked out a simple rule that can be stated here, but we will often permit this. We will be guided by the two goals of preserving the free status of all derivatives our free software and of promoting the sharing and reuse of software.

NO WARRANTY

BECAUSE GDB IS LICENSED FREE OF CHARGE, WE PROVIDE ABSOLUTELY NO WARRANTY, TO THE EXTENT PERMITTED BY APPLICABLE STATE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING, THE FREE SOFTWARE FOUNDATION, INC, RICHARD M. STALLMAN AND/OR OTHER PARTIES PROVIDE GDB "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE GDB PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.

IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW WILL FREE SOFTWARE FOUNDATION, INC., RICHARD M. STALLMAN, AND/OR ANY OTHER PARTY WHO MAY MODIFY AND REDISTRIBUTE GDB AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY LOST PROFITS, LOST MONIES, OR OTHER SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH PROGRAMS NOT DISTRIBUTED BY FREE SOFTWARE FOUNDATION, INC.) THE PROGRAM, EVEN IF YOU HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES, OR FOR ANY CLAIM BY ANY OTHER PARTY.

1. GDB Input Conventions

GDB is invoked with the shell command `gdb'. Once started, it reads commands from the terminal until you tell it to exit.

A GDB command is a single line of input. There is no limit on how long it can be. It starts with a command name, which is followed by arguments whose meaning depends on the command name. Some command names do not allow arguments.

GDB command names may always be abbreviated if the abbreviation is unambiguous. Sometimes even ambiguous abbreviations are allowed; for example, `s' is specially defined as equivalent to `step' even though there are other commands whose names start with `s'. Possible command abbreviations are often stated in the documentation of the individual commands.

A blank line as input to GDB means to repeat the previous command verbatim. Certain commands do not allow themselves to be repeated this way; these are commands for which unintentional repetition might cause trouble and which you are unlikely to want to repeat. Certain others (`list' and `x') act differently when repeated because that is more useful.

A line of input starting with `#' is a comment; it does nothing. This is useful mainly in command files (See section 12.2 Command Files).

Occasionally it is useful to execute a shell command from within gdb. This can be done with the `shell' command, or the shell escape character `!'.

shell shell command string

!shell command string

Directs GDB to invoke an inferior shell to execute `shell command string'. The environmental variable `SHELL' is used if it exists, otherwise gdb uses `/bin/sh'.

GDB prompts for commands with a string that is normally `(gdb)'. When debugging GDB with GDB, it is useful to change the prompt in one of the GDBs so that you can distinguish them. This can be done with the `set prompt' command.

set prompt newprompt

Directs GDB to use newprompt as its prompt string henceforth.

To exit GDB, use the `quit' command (abbreviated `q'). Ctrl-c will not exit from GDB, but rather will terminate the action of any GDB command that is in progress and return to GDB command level. It is safe to type Ctrl-c at any time because GDB does not allow it to take effect until a time when it is safe.

2. Specifying GDB's Files

GDB needs to know the filename of the program to be debugged. To debug a core dump of a previous run, GDB must be told the filename of the core dump.

2.1 Specifying Files with Arguments

Specifying files with arguments

                                (when you start GDB).

2.2 Specifying Files with Commands

Specifying files with GDB commands.

2.1 Specifying Files with Arguments

The usual way to specify the executable and core dump file names is with two command arguments given when you start GDB. The first argument is used as the file for execution and symbols, and the second argument (if any) is used as the core dump file name. Thus,

gdb progm core

specifies `progm' as the executable program and `core' as a core dump file to examine. (You do not need to have a core dump file if what you plan to do is debug the program interactively.)

See section 3. Options and Arguments for GDB, for full information on command options and arguments for GDB.

2.2 Specifying Files with Commands

Usually you specify the files for GDB to work with by giving arguments when you invoke GDB. But occasionally it is necessary to change to a different file during a GDB session. Or you may run GDB and forget to specify the files you want to use. In these situations the GDB commands to specify new files are useful.

exec-file filename

Specify that the program to be run is found in filename. If you do not specify a directory and the file is not found in GDB's working directory, GDB will use the environment variable `PATH' as a list of directories to search, just as the shell does when looking for a program to run.

symbol-file filename

Read symbol table information from file filename. `PATH' is searched when necessary. Most of the time you will use both the `exec-file' and `symbol-file' commands on the same file.

`symbol-file' with no argument clears out GDB's symbol table.

core-file filename

Specify the whereabouts of a core dump file to be used as the "contents of memory". Note that the core dump contains only the writable parts of memory; the read-only parts must come from the executable file.

`core-file' with no argument specifies that no core file is to be used.

add-file filename address

The `add-file' command takes two arguments, a file name, and the address at which that file has been (or should be) dynamically loaded. GDB will then treat that file as though it had always been dynamically linked, and provide the user with all the normal GDB features, including symbolic debugging.

With the `add-file' command, it is possible to debug code which was not present in the initial load image of the program under test. Suppose you have a program which can, while running, dynamically link a program fragment into its address space. One program which does this is KCL, a free common lisp implementation. The fragment will be loaded into the main program's address space at some address, and the main program can then call functions within the fragment by calculating (or otherwise obtaining) their addresses.

kill

Cancel running the program under GDB. This could be used if you wish to debug a core dump instead. GDB ignores any core dump file if it is actually running the program, so the `kill' command is the only sure way to go back to using the core dump file.

info files

Print the names of the executable and core dump files currently in use by GDB, and the file from which symbols were loaded.

While all three file-specifying commands allow both absolute and relative file names as arguments, GDB always converts the file name to an absolute one and remembers it that way.

The `symbol-file' command causes GDB to forget the contents of its convenience variables, the value history, and all breakpoints and auto-display expressions. This is because they may contain pointers to the internal data recording symbols and data types, which are part of the old symbol table data being discarded inside GDB.

3. Options and Arguments for GDB

When you invoke GDB, you can pass commands telling it what files to operate on and what other things to do.

3.1 Mode Options		Options controlling modes of operation.
3.2 File-specifying Options		Options to specify files (executable, coredump, commands)
3.3 Other Arguments		Any other arguments without options also specify files.

3.1 Mode Options

`-nx'

Do not execute commands from the init files `.gdbinit'. Normally, the commands in these files are executed after all the command options and arguments have been processed. See section 12.2 Command Files.

`-q'

"Quiet". Do not print the usual introductory messages.

`-batch'

Run in batch mode. Exit with code 1 after processing all the command files specified with `-x' (and `.gdbinit', if not inhibited). Exit also if, due to an error, GDB would otherwise attempt to read a command from the terminal.

`-fullname'

This option is used when Emacs runs GDB as a subprocess. It tells GDB to output the full file name and line number in a standard, recognizable fashion each time a stack frame is displayed (which includes each time the program stops). This recognizable format looks like two `\032' characters, followed by the filename, line number and character position separated by colons, and a newline. The Emacs-to-GDB interface program uses the two `\032' characters as a signal to display the source code for the frame.

3.2 File-specifying Options

All the options and command line arguments given are processed in sequential order. The order makes a difference when the `-x' command is used.

`-s file'

Read symbol table from file file.

`-e file'

Use file file as the executable file to execute when appropriate, and for examining pure data in conjunction with a core dump.

`-se file'

Read symbol table from file file and use it as the executable file.

`-c file'

Use file file as a core dump to examine.

`-x file'

Execute GDB commands from file file.

`-d directory'

Add directory to the path to search for source files.

3.3 Other Arguments

If there are arguments to GDB that are not options or associated with options, the first one specifies the symbol table and executable file name (as if it were preceded by `-se') and the second one specifies a core dump file name (as if it were preceded by `-c').

4. Compiling Your Program for Debugging

In order to debug a program effectively, you need to ask for debugging information when you compile it. This information in the object file describes the data type of each variable or function and the correspondence between source line numbers and addresses in the executable code.

To request debugging information, specify the `-g' option when you run the compiler.

The Unix C compiler is unable to handle the `-g' and `-O' options together. This means that you cannot ask for optimization if you ask for debugger information.

The GNU C compiler supports `-g' with or without `-O', making it possible to debug optimized code. We recommend that you always use `-g' whenever you compile a program. You may think the program is correct, but there's no sense in pushing your luck.

If you are using the GNU C compiler, the GNU assembler and the GNU linker, you can choose between two formats of debugging information: the standard Unix format, which is what you get with `-g', and GDB's own format, which you request by using `-gg' instead of `-g'. This stores debugging information in the executable file in a format much like that which is used inside GDB. This has these advantages and disadvantages:

• GDB can read `-gg' format more than twice as fast as Unix `-g' format.

• The `-gg' format uses much more disk space than Unix format.

• The Unix debuggers can understand only Unix format, so you cannot use Unix source-level debuggers if you compile with `-gg'. (The adb debugger works with either format; it does not use this information in any case.)

5. Running Your Program Under GDB

To start your program under GDB, use the `run' command. The program must already have been specified using the `exec-file' command or with an argument to GDB (see section 2. Specifying GDB's Files); what `run' does is create an inferior process, load the program into it, and set it in motion.

The execution of a program is affected by certain information it receives from its superior. GDB provides ways to specify them, which you must do before starting the program. (You can change them after starting the program, but such changes do not affect the program unless you start it over again.)

The arguments.

You specify the arguments to give the program as the arguments of the `run' command.

The environment.

The program normally inherits its environment from GDB, but you can use the GDB commands `set environment' and `unset environment' to change parts of the environment that will be given to the program.

The working directory.

The program inherits its working directory from GDB. You can set GDB's working directory with the `cd' command in GDB.

After the `run' command, the debugger does nothing but wait for your program to stop. See section 6. Stopping and Continuing.

Note that once your program has been started by the `run' command, you may evaluate expressions that involve calls to functions in the inferior. See section 9.1 Expressions. If you wish to evaluate a function simply for it's side affects, you may use the `set' command. See section 11.1 Assignment to Variables.

5.1 Your Program's Arguments		Specifying the arguments for your program.
5.2 Your Program's Environment		Specifying the environment for your program.
5.3 Your Program's Working Directory		Specifying the working directory for giving to your program when it is run.
5.4 Your Program's Input and Output		Specifying the program's standard input and output.
5.5 Debugging an Already-Running Process		Debugging a process started outside GDB.

5.1 Your Program's Arguments

You specify the arguments to give the program as the arguments of the `run' command. They are passed to a shell, which expands wildcard characters and performs redirection of I/O, and thence to the program.

`run' with no arguments uses the same arguments used by the previous `run'.

The command `set args' can be used to specify the arguments to be used the next time the program is run. If `set args' has no arguments, it means to use no arguments the next time the program is run. If you have run your program with arguments and want to run it again with no arguments, this is the only way to do so.

5.2 Your Program's Environment

The environment consists of a set of environment variables and their values. Environment variables conventionally record such things as your user name, your home directory, your terminal type, and your search path for programs to run. Usually you set up environment variables with the shell and they are inherited by all the other programs you run. When debugging, it can be useful to try running the program with different environments without having to start the debugger over again.

info environment varname

Print the value of environment variable varname to be given to your program when it is started. This command can be abbreviated `i env varname'.

info environment

Print the names and values of all environment variables to be given to your program when it is started. This command can be abbreviated `i env'.

set environment varname value

set environment varname = value

Sets environment variable varname to value, for your program only, not for GDB itself. value may be any string; the values of environment variables are just strings, and any interpretation is supplied by your program itself. The value parameter is optional; if it is eliminated, the variable is set to a null value. This command can be abbreviated as short as `set e'.

delete environment varname

unset environment varname

Remove variable varname from the environment to be passed to your program. This is different from `set env varname =' because `delete environment' makes a variable not be defined at all, which is distinguishable from an empty value. This command can be abbreviated `d e'.

5.3 Your Program's Working Directory

Each time you start your program with `run', it inherits its working directory from the current working directory of GDB. GDB's working directory is initially whatever it inherited from its superior, but you can specify the working directory for GDB with the `cd' command.

The GDB working directory also serves as a default for the commands that specify files for GDB to operate on. See section 2. Specifying GDB's Files.

cd directory

Set GDB's working directory to directory.

pwd

Print GDB's working directory.

5.4 Your Program's Input and Output

By default, the program you run under GDB does input and output to the same terminal that GDB uses.

You can redirect the program's input and/or output using `sh'-style redirection commands in the `run' command. For example,

run > outfile

starts the program, diverting its output to the file `outfile'.

Another way to specify where the program should do input and output is with the `tty' command. This command accepts a file name as argument, and causes this file to be the default for future `run' commands. It also resets the controlling terminal for future `run' commands. For example,

tty /dev/ttyb

directs that processes started with subsequent `run' commands default to do input and output on the terminal `/dev/ttyb' and sets the controlling terminal to `/dev/ttyb'. An explicit redirection in `run' overrides the `tty' command's effect on input/output redirection.

When you use the `tty' command or redirect input in the `run' command, the input for your program comes from the specified file, but the input for GDB still comes from your terminal.

5.5 Debugging an Already-Running Process

Some operating systems (in particular, Sun) allow GDB to begin debugging an already-running process that was started outside of GDB. To do this you must use the `attach' command instead of the `run' command.

The `attach' command requires one argument, which is the process-id of the process you want to debug. (The usual way to find out the process-id of the process is with the `ps' utility.)

The first thing GDB does after arranging to debug the process is to stop it. You can examine and modify an attached process with all the GDB commands that ordinarily available when you start processes with `run'. You can insert breakpoints; you can step and continue; you can modify storage. If you would rather the process continue running, use the `continue' command after attaching.

When you are finished debugging the attached process, you can use the `detach' command to release it from GDB's control. Detaching the process continues its execution. After the `detach' command, that process and GDB become completely independent once more, and you are ready to `attach' another process or start one with `run'.

If you exit GDB or use the `run' command while you have an attached process, you kill that process. You will be asked for confirmation if you try to do either of these things.

6. Stopping and Continuing

When you run a program normally, it runs until exiting. The purpose of using a debugger is so that you can stop it before that point; or so that if the program runs into trouble you can find out why.

6.1 Signals		Fatal signals in your program just stop it; then you can use GDB to see what is going on.
6.2 Breakpoints		Breakpoints let you stop your program when it reaches a specified point in the code.
6.3 Continuing		Resuming execution until the next signal or breakpoint.
6.4 Stepping		Stepping runs the program a short distance and then stops it wherever it has come to.

6.1 Signals

A signal is an asynchronous event that can happen in a program. The operating system defines the possible kinds of signals, and gives each kind a name and a number. For example, SIGINT is the signal a program gets when you type Ctrl-c; SIGSEGV is the signal a program gets from referencing a place in memory far away from all the areas in use; SIGALRM occurs when the alarm clock timer goes off (which happens only if the program has requested an alarm).

Some signals, including SIGALRM, are a normal part of the functioning of the program. Others, such as SIGSEGV, indicate errors; these signals are fatal (kill the program immediately) if the program has not specified in advance some other way to handle the signal. SIGINT does not indicate an error in the program, but it is normally fatal so it can carry out the purpose of Ctrl-c: to kill the program.

GDB has the ability to detect any occurrence of a signal in the program running under GDB's control. You can tell GDB in advance what to do for each kind of signal.

Normally, GDB is set up to ignore non-erroneous signals like SIGALRM (so as not to interfere with their role in the functioning of the program) but to stop the program immediately whenever an error signal happens. You can change these settings with the `handle' command. You must specify which signal you are talking about with its number.

info signal

Print a table of all the kinds of signals and how GDB has been told to handle each one. You can use this to see the signal numbers of all the defined types of signals.

handle signalnum keywords...

Change the way GDB handles signal signalnum. The keywords say what change to make.

To use the `handle' command you must know the code number of the signal you are concerned with. To find the code number, type `info signal' which prints a table of signal names and numbers.

The keywords allowed by the handle command can be abbreviated. Their full names are

stop

GDB should stop the program when this signal happens. This implies the `print' keyword as well.

print

GDB should print a message when this signal happens.

nostop

GDB should not stop the program when this signal happens. It may still print a message telling you that the signal has come in.

noprint

GDB should not mention the occurrence of the signal at all. This implies the `nostop' keyword as well.

pass

GDB should allow the program to see this signal; the program will be able to handle the signal, or may be terminated if the signal is fatal and not handled.

nopass

GDB should not allow the program to see this signal.

When a signal has been set to stop the program, the program cannot see the signal until you continue. It will see the signal then, if `pass' is in effect for the signal in question at that time. In other words, after GDB reports a signal, you can use the `handle' command with `pass' or `nopass' to control whether that signal will be seen by the program when you later continue it.

You can also use the `signal' command to prevent the program from seeing a signal, or cause it to see a signal it normally would not see, or to give it any signal at any time. See section 11.3 Giving the Program a Signal.

6.2 Breakpoints

A breakpoint makes your program stop whenever a certain point in the program is reached. You set breakpoints explicitly with GDB commands, specifying the place where the program should stop by line number, function name or exact address in the program. You can add various other conditions to control whether the program will stop.

Each breakpoint is assigned a number when it is created; these numbers are successive integers starting with 1. In many of the commands for controlling various features of breakpoints you use the breakpoint number to say which breakpoint you want to change. Each breakpoint may be enabled or disabled; if disabled, it has no effect on the program until you enable it again.

The command `info break' prints a list of all breakpoints set and not cleared, showing their numbers, where in the program they are, and any special features in use for them. Disabled breakpoints are included in the list, but marked as disabled. `info break' with a breakpoint number as argument lists only that breakpoint. The convenience variable `$_' and the default examining-address for the `x' command are set to the address of the last breakpoint listed (see section 9.4.1 Examining Memory).

6.2.1 Setting Breakpoints		How to establish breakpoints.
6.2.2 Clearing Breakpoints		How to remove breakpoints no longer needed.
6.2.3 Disabling Breakpoints		How to disable breakpoints (turn them off temporarily).
6.2.4 Break Conditions		Making extra conditions on whether to stop.
6.2.5 Commands Executed on Breaking		Commands to be executed at a breakpoint.
6.2.6 "Cannot Insert Breakpoints" Error		"Cannot insert breakpoints" error--why, what to do.

6.2.1 Setting Breakpoints

Breakpoints are set with the `break' command (abbreviated `b'). You have several ways to say where the breakpoint should go.

break function

Set a breakpoint at entry to function function.

break linenum

Set a breakpoint at line linenum in the current source file. That file is the last file whose source text was printed. This breakpoint will stop the program just before it executes any of the code on that line.

break filename:linenum

Set a breakpoint at line linenum in source file filename.

break filename:function

Set a breakpoint at entry to function function found in file filename. Specifying a filename as well as a function name is superfluous except when multiple files contain similarly named functions.

break *address

Set a breakpoint at address address. You can use this to set breakpoints in parts of the program which do not have debugging information or source files.

break

Set a breakpoint at the next instruction to be executed in the selected stack frame (see section 7. Examining the Stack). In any selected frame but the innermost, this will cause the program to stop as soon as control returns to that frame. This is equivalent to a `finish' command in the frame inside the selected frame. If this is done in the innermost frame gdb will stop the next time it reaches the current location; this may be useful inside of loops. It does not stop at this breakpoint immediately upon continuation of the program since no code would be executed if it did.

break ... if cond

Set a breakpoint with condition cond; evaluate the expression cond each time the breakpoint is reached, and stop only if the value is nonzero. `...' stands for one of the possible arguments described above (or no argument) specifying where to break. See section 6.2.4 Break Conditions, for more information on breakpoint conditions.

tbreak args

Set a breakpoint enabled only for one stop. args are the same as in the `break' command, and the breakpoint is set in the same way, but the breakpoint is automatically disabled the first time it is hit.

GDB allows you to set any number of breakpoints at the same place in the program. There is nothing silly or meaningless about this. When the breakpoints are conditional, this is even useful (see section 6.2.4 Break Conditions).

6.2.2 Clearing Breakpoints

It is often necessary to eliminate a breakpoint once it has done its job and you no longer want the program to stop there. This is called clearing or `deleting' the breakpoint. A breakpoint that has been cleared no longer exists in any sense.

With the `clear' command you can clear breakpoints according to where they are in the program. With the `delete' command you can clear individual breakpoints by specifying their breakpoint numbers.

It is not necessary to clear a breakpoint to proceed past it. GDB automatically ignores breakpoints in the first instruction to be executed when you continue execution at the same address where the program stopped.

clear

Clear any breakpoints at the next instruction to be executed in the selected stack frame (see section 7.3 Selecting a Frame). When the innermost frame is selected, this is a good way to clear a breakpoint that the program just stopped at.

clear function

clear filename:function

Clear any breakpoints set at entry to the function function.

clear linenum

clear filename:linenum

Clear any breakpoints set at or within the code of the specified line.

delete bnums...

Delete the breakpoints of the numbers specified as arguments. A breakpoint deleted is forgotten completely.

6.2.3 Disabling Breakpoints

Rather than clearing a breakpoint, you might prefer to disable it. This makes the breakpoint inoperative as if it had been cleared, but remembers the information on the breakpoint so that you can enable it again later.

You disable and enable breakpoints with the `enable' and `disable' commands, specifying one or more breakpoint numbers as arguments. Use `info break' to print a list of breakpoints if you don't know which breakpoint numbers to use.

A breakpoint can have any of four different states of enablement:

• Enabled. The breakpoint will stop the program. A breakpoint made with the `break' command starts out in this state.
• Disabled. The breakpoint has no effect on the program.
• Enabled once. The breakpoint will stop the program, but when it does so it will become disabled. A breakpoint made with the `tbreak' command starts out in this state.
• Enabled for deletion. The breakpoint will stop the program, but immediately after it does so it will be deleted permanently.

You change the state of enablement of a breakpoint with the following commands:

disable breakpoints bnums...

disable bnums...

Disable the specified breakpoints. A disabled breakpoint has no effect but is not forgotten. All options such as ignore-counts, conditions and commands are remembered in case the breakpoint is enabled again later.

enable breakpoints bnums...

enable bnums...

Enable the specified breakpoints. They become effective once again in stopping the program, until you specify otherwise.

enable breakpoints once bnums...

enable once bnums...

Enable the specified breakpoints temporarily. Each will be disabled again the next time it stops the program (unless you have used one of these commands to specify a different state before that time comes).

enable breakpoints delete bnums...

enable delete bnums...

Enable the specified breakpoints to work once and then die. Each of the breakpoints will be deleted the next time it stops the program (unless you have used one of these commands to specify a different state before that time comes).

Aside from the automatic disablement or deletion of a breakpoint when it stops the program, which happens only in certain states, the state of enablement of a breakpoint changes only when one of the commands above is used.

6.2.4 Break Conditions

The simplest sort of breakpoint breaks every time the program reaches a specified place. You can also specify a condition for a breakpoint. A condition is just a boolean expression in your programming language (See section 9.1 Expressions). A breakpoint with a condition evaluates the expression each time the program reaches it, and the program stops only if the condition is true.

Break conditions may have side effects, and may even call functions in your program. These may sound like strange things to do, but their effects are completely predictable unless there is another enabled breakpoint at the same address. (In that case, GDB might see the other breakpoint first and stop the program without checking the condition of this one.) Note that breakpoint commands are usually more convenient and flexible for the purpose of performing side effects when a breakpoint is reached (see section 6.2.5 Commands Executed on Breaking).

Break conditions can be specified when a breakpoint is set, by using `if' in the arguments to the `break' command. See section 6.2.1 Setting Breakpoints. They can also be changed at any time with the `condition' command:

condition bnum expression

Specify expression as the break condition for breakpoint number bnum. From now on, this breakpoint will stop the program only if the value of expression is true (nonzero, in C). expression is not evaluated at the time the `condition' command is given. See section 9.1 Expressions.

condition bnum

Remove the condition from breakpoint number bnum. It becomes an ordinary unconditional breakpoint.

A special feature is provided for one kind of condition: to prevent the breakpoint from doing anything until it has been reached a certain number of times. This is done with the ignore count of the breakpoint. When the program reaches a breakpoint whose ignore count is positive, then instead of stopping, it just decrements the ignore count by one and continues.

ignore bnum count

Set the ignore count of breakpoint number bnum to count. The next count times the breakpoint is reached, it will not stop.

To make the breakpoint stop the next time it is reached, specify a count of zero.

cont count

Continue execution of the program, setting the ignore count of the breakpoint that the program stopped at to count minus one. Continuing through the breakpoint does not itself count as one of count. Thus, the program will not stop at this breakpoint until the count'th time it is hit.

This command is allowed only when the program stopped due to a breakpoint. At other times, the argument to `cont' is ignored.

If a breakpoint has a positive ignore count and a condition, the condition is not checked. Once the ignore count reaches zero, the condition will start to be checked.

Note that you could achieve the effect of the ignore count with a condition such as `$foo-- <= 0' using a debugger convenience variable that is decremented each time. That is why the ignore count is considered a special case of a condition. See section 9.7 Convenience Variables.

6.2.5 Commands Executed on Breaking

You can give any breakpoint a series of commands to execute when the program stops due to that breakpoint. For example, you might want to print the values of certain expressions, or enable other breakpoints.

commands bnum

Specify commands for breakpoint number bnum. The commands themselves appear on the following lines. Type a line containing just `end' to terminate the commands.

To remove all commands from a breakpoint, use the command `commands' and follow it immediately by `end'; that is, give no commands.

With no arguments, `commands' refers to the last breakpoint set.

It is possible for breakpoint commands to start the program up again. Simply use the `cont' command, or `step', or any other command to resume execution. However, any remaining breakpoint commands are ignored. When the program stops again, GDB will act according to why that stop took place.

If the first command specified is `silent', the usual message about stopping at a breakpoint is not printed. This may be desirable for breakpoints that are to print a specific message and then continue. If the remaining commands too print nothing, you will see no sign that the breakpoint was reached at all. `silent' is not really a command; it is meaningful only at the beginning of the commands for a breakpoint.

The commands `echo' and `output' that allow you to print precisely controlled output are often useful in silent breakpoints. See section 12.3 Commands for Controlled Output.

For example, here is how you could use breakpoint commands to print the value of x at entry to foo whenever it is positive. We assume that the newly created breakpoint is number 4; `break' will print the number that is assigned.

break foo if x>0 commands 4 silent echo x is\040 output x echo \n cont end

One application for breakpoint commands is to correct one bug so you can test another. Put a breakpoint just after the erroneous line of code, give it a condition to detect the case in which something erroneous has been done, and give it commands to assign correct values to any variables that need them. End with the `cont' command so that the program does not stop, and start with the `silent' command so that no output is produced. Here is an example:

break 403 commands 5 silent set x = y + 4 cont end

One deficiency in the operation of automatically continuing breakpoints under Unix appears when your program uses raw mode for the terminal. GDB switches back to its own terminal modes (not raw) before executing commands, and then must switch back to raw mode when your program is continued. This causes any pending terminal input to be lost.

In the GNU system, this will be fixed by changing the behavior of terminal modes.

Under Unix, when you have this problem, you might be able to get around it by putting your actions into the breakpoint condition instead of commands. For example

condition 5 (x = y + 4), 0

is a condition expression (See section 9.1 Expressions) that will change x as needed, then always have the value 0 so the program will not stop. Loss of input is avoided here because break conditions are evaluated without changing the terminal modes. When you want to have nontrivial conditions for performing the side effects, the operators `&&', `||' and `? ... :' may be useful.

6.2.6 "Cannot Insert Breakpoints" Error

Under some Unix systems, breakpoints cannot be used in a program if any other process is running that program. Attempting to run or continue the program with a breakpoint in this case will cause GDB to stop it.

When this happens, you have three ways to proceed:

1. Remove or disable the breakpoints, then continue.

2. Suspend GDB, and copy the file containing the program to a new name. Resume GDB and use the `exec-file' command to specify that GDB should run the program under that name. Then start the program again.

3. Recompile the program so that the text is non-sharable (a.out format OMAGIC).

6.3 Continuing

After your program stops, most likely you will want it to run some more if the bug you are looking for has not happened yet.

cont

Continue running the program at the place where it stopped.

If the program stopped at a breakpoint, the place to continue running is the address of the breakpoint. You might expect that continuing would just stop at the same breakpoint immediately. In fact, `cont' takes special care to prevent that from happening. You do not need to clear the breakpoint to proceed through it after stopping at it.

You can, however, specify an ignore-count for the breakpoint that the program stopped at, by means of an argument to the `cont' command. See section 6.2.4 Break Conditions.

If the program stopped because of a signal other than SIGINT or SIGTRAP, continuing will cause the program to see that signal. You may not want this to happen. For example, if the program stopped due to some sort of memory reference error, you might store correct values into the erroneous variables and continue, hoping to see more execution; but the program would probably terminate immediately as a result of the fatal signal once it sees the signal. To prevent this, you can continue with `signal 0'. See section 11.3 Giving the Program a Signal. You can also act in advance to prevent the program from seeing certain kinds of signals, using the `handle' command (see section 6.1 Signals).

6.4 Stepping

Stepping means setting your program in motion for a limited time, so that control will return automatically to the debugger after one line of code or one machine instruction. Breakpoints are active during stepping and the program will stop for them even if it has not gone as far as the stepping command specifies.

step

Proceed the program until control reaches a different line, then stop it and return to the debugger. This command is abbreviated `s'.

step count

Proceed as in `step', but do so count times. If a breakpoint or a signal not related to stepping is reached before count steps, stepping stops right away.

This command may be given when control is within a routine for which there is no debugging information. In that case, execution will proceed until control reaches a different routine, or is about to return from this routine. An argument repeats this action.

next

Similar to `step', but any function calls appearing within the line of code are executed without stopping. Execution stops when control reaches a different line of code at the stack level which was executing when the `next' command was given. This command is abbreviated `n'.

An argument is a repeat count, as in `step'.

`next' within a routine without debugging information acts as does `step', but any function calls appearing within the code of the routine are executed without stopping.

finish

Continue running until just after the selected stack frame returns (or until there is some other reason to stop, such as a fatal signal or a breakpoint). Print value returned by the selected stack frame (if any).

Contrast this with the `return' command (see section 11.4 Returning from a Function).

until

Proceed the program until control reaches a line greater than the current line, then stop is and return to the debugger. Control is also returned to the debugger if the program exits the current stack frame. Note that this form of the command uses single stepping, and hence is slower than `until' with an argument. This command is abbreviated `u'.

until location

Proceed the program until either the specified location is reached, or the current (innermost) stack frame returns. This form of the command uses breakpoints, and hence is quicker than `until' without an argument.

stepi

si

Proceed one machine instruction, then stop and return to the debugger.

It is often useful to do `display/i $pc' when stepping by machine instructions. This will cause the next instruction to be executed to be displayed automatically at each stop. See section 9.5 Automatic Display.

An argument is a repeat count, as in `step'.

nexti

ni

Proceed one machine instruction, but if it is a subroutine call, proceed until the subroutine returns.

An argument is a repeat count, as in `next'.

A typical technique for using stepping is to put a breakpoint (see section 6.2 Breakpoints) at the beginning of the function or the section of the program in which a problem is believed to lie, and then step through the suspect area, examining the variables that are interesting, until the problem happens.

The `cont' command can be used after stepping to resume execution until the next breakpoint or signal.

7. Examining the Stack

When your program has stopped, the first thing you need to know is where it stopped and how it got there.

Each time your program performs a function call, the information about where in the program the call was made from is saved in a block of data called a stack frame. The frame also contains the arguments of the call and the local variables of the function that was called. All the stack frames are allocated in a region of memory called the call stack.

When your program stops, the GDB commands for examining the stack allow you to see all of this information.

One of the stack frames is selected by GDB and many GDB commands refer implicitly to the selected frame. In particular, whenever you ask GDB for the value of a variable in the program, the value is found in the selected frame. There are special GDB commands to select whichever frame you are interested in.

When the program stops, GDB automatically selects the currently executing frame and describes it briefly as the `frame' command does (see section Info).

7.1 Stack Frames		Explanation of stack frames and terminology.
7.2 Backtraces		Summarizing many frames at once.
7.3 Selecting a Frame		How to select a stack frame.
7.4 Information on a Frame		Commands to print information on stack frames.

7.1 Stack Frames

The call stack is divided up into contiguous pieces called frames; each frame is the data associated with one call to one function. The frame contains the arguments given to the function, the function's local variables, and the address at which the function is executing.

When your program is started, the stack has only one frame, that of the function main. This is called the initial frame or the outermost frame. Each time a function is called, a new frame is made. Each time a function returns, the frame for that function invocation is eliminated. If a function is recursive, there can be many frames for the same function. The frame for the function in which execution is actually occurring is called the innermost frame. This is the most recently created of all the stack frames that still exist.

Inside your program, stack frames are identified by their addresses. A stack frame consists of many bytes, each of which has its own address; each kind of computer has a convention for choosing one of those bytes whose address serves as the address of the frame. Usually this address is kept in a register called the frame pointer register while execution is going on in that frame.

GDB assigns numbers to all existing stack frames, starting with zero for the innermost frame, one for the frame that called it, and so on upward. These numbers do not really exist in your program; they are to give you a way of talking about stack frames in GDB commands.

Many GDB commands refer implicitly to one stack frame. GDB records a stack frame that is called the selected stack frame; you can select any frame using one set of GDB commands, and then other commands will operate on that frame. When your program stops, GDB automatically selects the innermost frame.

7.2 Backtraces

A backtrace is a summary of how the program got where it is. It shows one line per frame, for many frames, starting with the currently executing frame (frame zero), followed by its caller (frame one), and on up the stack.

backtrace

bt

Print a backtrace of the entire stack: one line per frame for all frames in the stack.

You can stop the backtrace at any time by typing the system interrupt character, normally Control-C.

backtrace n

bt n

Similar, but stop after n frames.

backtrace -n

bt -n

Similar, but print the outermost n frames instead of the innermost.

Each line in a backtrace shows the frame number, the program counter, the function and its arguments, and the source file name and line number (if known). The program counter is omitted if is the beginning of the code for the source line. This is the same as the first of the two lines printed when you select a frame.

7.3 Selecting a Frame

Most commands for examining the stack and other data in the program work on whichever stack frame is selected at the moment. Here are the commands for selecting a stack frame; all of them finish by printing a brief description of the stack frame just selected.

frame n

Select frame number n. Recall that frame zero is the innermost (currently executing) frame, frame one is the frame that called the innermost one, and so on. The highest-numbered frame is main's frame.

frame addr

Select the frame at address addr. This is useful mainly if the chaining of stack frames has been damaged by a bug, making it impossible for GDB to assign numbers properly to all frames. In addition, this can be useful when the program has multiple stacks and switches between them.

up n

Select the frame n frames up from the frame previously selected. For positive numbers n, this advances toward the outermost frame, to higher frame numbers, to frames that have existed longer. n defaults to one.

down n

Select the frame n frames down from the frame previously selected. For positive numbers n, this advances toward the innermost frame, to lower frame numbers, to frames that were created more recently. n defaults to one.

All of these commands end by printing some information on the frame that has been selected: the frame number, the function name, the arguments, the source file and line number of execution in that frame, and the text of that source line. For example:

#3 main (argc=3, argv=??, env=??) at main.c, line 67 67 read_input_file (argv[i]);

After such a printout, the `list' command with no arguments will print ten lines centered on the point of execution in the frame. See section 8.1 Printing Source Lines.

7.4 Information on a Frame

There are several other commands to print information about the selected stack frame.

frame

This command prints a brief description of the selected stack frame. It can be abbreviated `f'. With an argument, this command is used to select a stack frame; with no argument, it does not change which frame is selected, but still prints the same information.

info frame

This command prints a verbose description of the selected stack frame, including the address of the frame, the addresses of the next frame in (called by this frame) and the next frame out (caller of this frame), the address of the frame's arguments, the program counter saved in it (the address of execution in the caller frame), and which registers were saved in the frame. The verbose description is useful when something has gone wrong that has made the stack format fail to fit the usual conventions.

info frame addr

Print a verbose description of the frame at address addr, without selecting that frame. The selected frame remains unchanged by this command.

info args

Print the arguments of the selected frame, each on a separate line.

info locals

Print the local variables of the selected frame, each on a separate line. These are all variables declared static or automatic within all program blocks that execution in this frame is currently inside of.

8. Examining Source Files

GDB knows which source files your program was compiled from, and can print parts of their text. When your program stops, GDB spontaneously prints the line it stopped in. Likewise, when you select a stack frame (see section 7.3 Selecting a Frame), GDB prints the line which execution in that frame has stopped in. You can also print parts of source files by explicit command.

8.1 Printing Source Lines		Using the `list' command to print source files.
8.2 Searching Source Files		Commands for searching source files.
8.3 Specifying Source Directories		Specifying the directories to search for source files.

8.1 Printing Source Lines

To print lines from a source file, use the `list' command (abbreviated `l'). There are several ways to specify what part of the file you want to print.

Here are the forms of the `list' command most commonly used:

list linenum

Print ten lines centered around line number linenum in the current source file.

list function

Print ten lines centered around the beginning of function function.

list

Print ten more lines. If the last lines printed were printed with a `list' command, this prints ten lines following the last lines printed; however, if the last line printed was a solitary line printed as part of displaying a stack frame (see section 7. Examining the Stack), this prints ten lines centered around that line.

list -

Print ten lines just before the lines last printed.

Repeating a `list' command with RET discards the argument, so it is equivalent to typing just `list'. This is more useful than listing the same lines again. An exception is made for an argument of `-'; that argument is preserved in repetition so that each repetition moves up in the file.

In general, the `list' command expects you to supply zero, one or two linespecs. Linespecs specify source lines; there are several ways of writing them but the effect is always to specify some source line. Here is a complete description of the possible arguments for `list':

list linespec

Print ten lines centered around the line specified by linespec.

list first,last

Print lines from first to last. Both arguments are linespecs.

list ,last

Print ten lines ending with last.

list first,

Print ten lines starting with first.

list +

Print ten lines just after the lines last printed.

list -

Print ten lines just before the lines last printed.

list

As described in the preceding table.

Here are the ways of specifying a single source line--all the kinds of linespec.

linenum

Specifies line linenum of the current source file. When a `list' command has two linespecs, this refers to the same source file as the first linespec.

+offset

Specifies the line offset lines after the last line printed. When used as the second linespec in a `list' command that has two, this specifies the line offset lines down from the first linespec.

-offset

Specifies the line offset lines before the last line printed.

filename:linenum

Specifies line linenum in the source file filename.

function

Specifies the line of the open-brace that begins the body of the function function.

filename:function

Specifies the line of the open-brace that begins the body of the function function in the file filename. The file name is needed with a function name only for disambiguation of identically named functions in different source files.

*address

Specifies the line containing the program address address. address may be any expression.

One other command is used to map source lines to program addresses.

info line linenum

Print the starting and ending addresses of the compiled code for source line linenum.

The default examine address for the `x' command is changed to the starting address of the line, so that `x/i' is sufficient to begin examining the machine code (see section 9.4.1 Examining Memory). Also, this address is saved as the value of the convenience variable `$_' (see section 9.7 Convenience Variables).

8.2 Searching Source Files

There are two commands for searching through the current source file for a regular expression.

The command `forward-search regexp' checks each line, starting with the one following the last line listed, for a match for regexp. It lists the line that is found. You can abbreviate the command name as `fo'.

The command `reverse-search regexp' checks each line, starting with the one before the last line listed and going backward, for a match for regexp. It lists the line that is found. You can abbreviate this command with as little as `rev'.

8.3 Specifying Source Directories

Executable programs do not record the directories of the source files they were compiled from, just the names. GDB remembers a list of directories to search for source files; this is called the source path. Each time GDB wants a source file, it tries all the directories in the list, in the order they are present in the list, until it finds a file with the desired name.

When you start GDB, its source path contains just the current working directory. To add other directories, use the `directory' command. Note that the search path for executable files and the working directory are not used for finding source files.

directory dirname

Add directory dirname to the end of the source path.

directory

Reset the source path to just the current working directory of GDB. This requires confirmation.

`directory' with no argument can cause source files previously found by GDB to be found in a different directory. To make this work correctly, this command also clears out the tables GDB maintains about the source files it has already found.

info directories

Print the source path: show which directories it contains.

Because the `directory' command adds to the end of the source path, it does not affect any file that GDB has already found. If the source path contains directories that you do not want, and these directories contain misleading files with names matching your source files, the way to correct the situation is as follows:

1. Choose the directory you want at the beginning of the source path. Use the `cd' command to make that the current working directory.

2. Use `directory' with no argument to reset the source path to just that directory.

3. Use `directory' with suitable arguments to add any other directories you want in the source path.

9. Examining Data

The usual way of examining data in your program is with the `print' command (abbreviated `p'). It evaluates and prints the value of any valid expression of the language the program is written in (for now, C). You type

print exp

where exp is any valid expression, and the value of exp is printed in a format appropriate to its data type.

A more low-level way of examining data is with the `x' command. It examines data in memory at a specified address and prints it in a specified format.

GDB supports one command to modify the default format of displayed data:

`set array-max'

`set array-max' sets the maximum number of elements of an array which will be printed. This limit also applies to the display of strings.

9.1 Expressions		Expressions that can be computed and printed.
9.2 Program Variables		Using your program's variables in expressions.
11.1 Assignment to Variables		Setting your program's variables.
9.3 Artificial Arrays		Examining part of memory as an array.
9.4 Formats		Specifying formats for printing values.
9.4.1 Examining Memory		Examining memory explicitly.
9.5 Automatic Display		Printing certain expressions whenever program stops.
9.6 Value History		Referring to values previously printed.
9.7 Convenience Variables		Giving names to values for future reference.
9.8 Registers		Referring to and storing in machine registers.

9.1 Expressions

Many different GDB commands accept an expression and compute its value. Any kind of constant, variable or operator defined by the programming language you are using is legal in an expression in GDB. This includes conditional expressions, function calls, casts and string constants. It unfortunately does not include symbols defined by preprocessor #define commands.

Casts are supported in all languages, not just in C, because it is so useful to cast a number into a pointer so as to examine a structure at that address in memory.

GDB supports three kinds of operator in addition to those of programming languages:

@

`@' is a binary operator for treating parts of memory as arrays. See section 9.3 Artificial Arrays, for more information.

::

`::' allows you to specify a variable in terms of the file or function it is defined in. See section 9.2 Program Variables.

{type} addr

Refers to an object of type type stored at address addr in memory. addr may be any expression whose value is an integer or pointer (but parentheses are required around nonunary operators, just as in a cast). This construct is allowed regardless of what kind of data is officially supposed to reside at addr.

9.2 Program Variables

The most common kind of expression to use is the name of a variable in your program.

Variables in expressions are understood in the selected stack frame (see section 7.3 Selecting a Frame); they must either be global (or static) or be visible according to the scope rules of the programming language from the point of execution in that frame. This means that in the function

foo (a) int a; { bar (a); { int b = test (); bar (b); } }

the variable a is usable whenever the program is executing within the function foo, but the variable b is visible only while the program is executing inside the block in which b is declared.

As a special exception, you can refer to a variable or function whose scope is a single source file even if the current execution point is not in this file. But it is possible to have more than one such variable or function with the same name (if they are in different source files). In such a case, it is not defined which one you will get. If you wish, you can specify any one of them using the colon-colon construct:

block::variable

Here block is the name of the source file whose variable you want.

9.3 Artificial Arrays

It is often useful to print out several successive objects of the same type in memory; a section of an array, or an array of dynamically determined size for which only a pointer exists in the program.

This can be done by constructing an artificial array with the binary operator `@'. The left operand of `@' should be the first element of the desired array, as an individual object. The right operand should be the length of the array. The result is an array value whose elements are all of the type of the left argument. The first element is actually the left argument; the second element comes from bytes of memory immediately following those that hold the first element, and so on. Here is an example. If a program says

int *array = (int *) malloc (len * sizeof (int));

you can print the contents of array with

p *array@len

The left operand of `@' must reside in memory. Array values made with `@' in this way behave just like other arrays in terms of subscripting, and are coerced to pointers when used in expressions. (It would probably appear in an expression via the value history, after you had printed it out.)

9.4 Formats

GDB normally prints all values according to their data types. Sometimes this is not what you want. For example, you might want to print a number in hex, or a pointer in decimal. Or you might want to view data in memory at a certain address as a character string or an instruction. These things can be done with output formats.

The simplest use of output formats is to say how to print a value already computed. This is done by starting the arguments of the `print' command with a slash and a format letter. The format letters supported are:

`x'

Regard the bits of the value as an integer, and print the integer in hexadecimal.

`d'

Print as integer in signed decimal.

`u'

Print as integer in unsigned decimal.

`o'

Print as integer in octal.

`a'

Print as an address, both absolute in hex and then relative to a symbol defined as an address below it.

`c'

Regard as an integer and print it as a character constant.

`f'

Regard the bits of the value as a floating point number and print using typical floating point syntax.

For example, to print the program counter in hex (see section 9.8 Registers), type

p/x $pc

Note that no space is required before the slash; this is because command names in GDB cannot contain a slash.

To reprint the last value in the value history with a different format, you can use the `print' command with just a format and no expression. For example, `p/x' reprints the last value in hex.

9.4.1 Examining Memory

The command `x' (for `examine') can be used to examine memory under explicit control of formats, without reference to the program's data types.

`x' is followed by a slash and an output format specification, followed by an expression for an address. The expression need not have a pointer value (though it may); it is used as an integer, as the address of a byte of memory. See section 9.1 Expressions for more information on expressions.

The output format in this case specifies both how big a unit of memory to examine and how to print the contents of that unit. It is done with one or two of the following letters:

These letters specify just the size of unit to examine:

`b'

Examine individual bytes.

`h'

Examine halfwords (two bytes each).

`w'

Examine words (four bytes each).

Many assemblers and cpu designers still use `word' for a 16-bit quantity, as a holdover from specific predecessor machines of the 1970's that really did use two-byte words. But more generally the term `word' has always referred to the size of quantity that a machine normally operates on and stores in its registers. This is 32 bits for all the machines that GNU runs on.

`g'

Examine giant words (8 bytes).

These letters specify just the way to print the contents:

`x'

Print as integers in unsigned hexadecimal.

`d'

Print as integers in signed decimal.

`u'

Print as integers in unsigned decimal.

`o'

Print as integers in unsigned octal.

`a'

Print as an address, both absolute in hex and then relative to a symbol defined as an address below it.

`c'

Print as character constants.

`f'

Print as floating point. This works only with sizes `w' and `g'.

`s'

Print a null-terminated string of characters. The specified unit size is ignored; instead, the unit is however many bytes it takes to reach a null character (including the null character).

`i'

Print a machine instruction in assembler syntax (or nearly). The specified unit size is ignored; the number of bytes in an instruction varies depending on the type of machine, the opcode and the addressing modes used.

If either the manner of printing or the size of unit fails to be specified, the default is to use the same one that was used last. If you don't want to use any letters after the slash, you can omit the slash as well.

You can also omit the address to examine. Then the address used is just after the last unit examined. This is why string and instruction formats actually compute a unit-size based on the data: so that the next string or instruction examined will start in the right place. The `print' command sometimes sets the default address for the `x' command; when the value printed resides in memory, the default is set to examine the same location. `info line' also sets the default for `x', to the address of the start of the machine code for the specified line and `info breakpoints' sets it to the address of the last breakpoint listed.

When you use RET to repeat an `x' command, it does not repeat exactly the same: the address specified previously (if any) is ignored, so that the repeated command examines the successive locations in memory rather than the same ones.

You can examine several consecutive units of memory with one command by writing a repeat-count after the slash (before the format letters, if any). The repeat count must be a decimal integer. It has the same effect as repeating the `x' command that many times except that the output may be more compact with several units per line.

x/10i $pc

Prints ten instructions starting with the one to be executed next in the selected frame. After doing this, you could print another ten following instructions with

x/10

in which the format and address are allowed to default.

The addresses and contents printed by the `x' command are not put in the value history because there is often too much of them and they would get in the way. Instead, GDB makes these values available for subsequent use in expressions as values of the convenience variables `$_' and `$__'.

After an `x' command, the last address examined is available for use in expressions in the convenience variable `$_'. The contents of that address, as examined, are available in the convenience variable `$__'.

If the `x' command has a repeat count, the address and contents saved are from the last memory unit printed; this is not the same as the last address printed if several units were printed on the last line of output.

9.5 Automatic Display

If you find that you want to print the value of an expression frequently (to see how it changes), you might want to add it to the automatic display list so that GDB will print its value each time the program stops. Each expression added to the list is given a number to identify it; to remove an expression from the list, you specify that number. The automatic display looks like this:

2: foo = 38 3: bar[5] = (struct hack *) 0x3804

showing item numbers, expressions and their current values.

display exp

Add the expression exp to the list of expressions to display each time the program stops. See section 9.1 Expressions.

display/fmt exp

For fmt specifying only a display format and not a size or count, add the expression exp to the auto-display list but arranges to display it each time in the specified format fmt.

display/fmt addr

For fmt `i' or `s', or including a unit-size or a number of units, add the expression addr as a memory address to be examined each time the program stops. Examining means in effect doing `x/fmt addr'. See section 9.4.1 Examining Memory.

undisplay dnums...

delete display dnums...

Remove item numbers dnums from the list of expressions to display.

disable display dnums...

Disable the display of item numbers dnums. A disabled display item has no effect but is not forgotten. It may be later enabled.

enable display dnums...

Enable display of item numbers dnums. It becomes effective once again in auto display of its expression, until you specify otherwise.

display

Display the current values of the expressions on the list, just as is done when the program stops.

info display

Print the list of expressions to display automatically, each one with its item number, but without showing the values.

9.6 Value History

Every value printed by the `print' command is saved for the entire session in GDB's value history so that you can refer to it in other expressions.

The values printed are given history numbers for you to refer to them by. These are successive integers starting with 1. `print' shows you the history number assigned to a value by printing `$n = ' before the value; here n is the history number.

To refer to any previous value, use `$' followed by the value's history number. The output printed by `print' is designed to remind you of this. Just `$' refers to the most recent value in the history, and `$$' refers to the value before that.

For example, suppose you have just printed a pointer to a structure and want to see the contents of the structure. It suffices to type

p *$

If you have a chain of structures where the component `next' points to the next one, you can print the contents of the next one with

p *$.next

It might be useful to repeat this command many times by typing RET.

Note that the history records values, not expressions. If the value of x is 4 and you type

print x set x=5

then the value recorded in the value history by the `print' command remains 4 even though x's value has changed.

info history

Print the last ten values in the value history, with their item numbers. This is like `p $$9' repeated ten times, except that `info history' does not change the history.

info history n

Print ten history values centered on history item number n.

9.7 Convenience Variables

GDB provides convenience variables that you can use within GDB to hold on to a value and refer to it later. These variables exist entirely within GDB; they are not part of your program, and setting a convenience variable has no effect on further execution of your program. That's why you can use them freely.

Convenience variables have names starting with `$'. Any name starting with `$' can be used for a convenience variable, unless it is one of the predefined set of register names (see section 9.8 Registers).

You can save a value in a convenience variable with an assignment expression, just as you would set a variable in your program. Example:

set $foo = *object_ptr

would save in `$foo' the value contained in the object pointed to by object_ptr.

Using a convenience variable for the first time creates it; but its value is void until you assign a new value. You can alter the value with another assignment at any time.

Convenience variables have no fixed types. You can assign a convenience variable any type of value, even if it already has a value of a different type. The convenience variable as an expression has whatever type its current value has.

info convenience

Print a list of convenience variables used so far, and their values. Abbreviated `i con'.

One of the ways to use a convenience variable is as a counter to be incremented or a pointer to be advanced. For example:

set $i = 0 print bar[$i++]->contents ...repeat that command by typing RET.

Some convenience variables are created automatically by GDB and given values likely to be useful.

`$_'

The variable `$_' is automatically set by the `x' command to the last address examined (see section 9.4.1 Examining Memory). Other commands which provide a default address for `x' to examine also set `$_' to that address; these commands include `info line' and `info breakpoint'.

`$__'

The variable `$__' is automatically set by the `x' command to the value found in the last address examined.

9.8 Registers

Machine register contents can be referred to in expressions as variables with names starting with `$'. The names of registers are different for each machine; use `info registers' to see the names used on your machine. The names `$pc' and `$sp' are used on all machines for the program counter register and the stack pointer. Often `$fp' is used for a register that contains a pointer to the current stack frame.

GDB always considers the contents of an ordinary register as an integer when the register is examined in this way. Some machines have special registers which can hold nothing but floating point; these registers are considered floating point. There is no way to refer to the contents of an ordinary register as floating point value (although you can print it as a floating point value with `print/f $regname').

Some registers have distinct "raw" and "virtual" data formats. This means that the data format in which the register contents are saved by the operating system is not the same one that your program normally sees. For example, the registers of the 68881 floating point coprocessor are always saved in "extended" format, but virtually all C programs expect to work with "double" format. In such cases, GDB normally works with the virtual format only (the format that makes sense for your program), but the `info registers' command prints the data in both formats.

Register values are relative to the selected stack frame (see section 7.3 Selecting a Frame). This means that you get the value that the register would contain if all stack frames farther in were exited and their saved registers restored. In order to see the real contents of all registers, you must select the innermost frame (with `frame 0').

Some registers are never saved (typically those numbered zero or one) because they are used for returning function values; for these registers, relativization makes no difference.

info registers

Print the names and relativized values of all registers.

info registers regname

Print the relativized value of register regname. regname may be any register name valid on the machine you are using, with or without the initial `$'.

9.8.1 Examples

You could print the program counter in hex with

p/x $pc

or print the instruction to be executed next with

x/i $pc

or add four to the stack pointer with

set $sp += 4

The last is a way of removing one word from the stack, on machines where stacks grow downward in memory (most machines, nowadays). This assumes that the innermost stack frame is selected. Setting `$sp' is not allowed when other stack frames are selected.

10. Examining the Symbol Table

The commands described in this section allow you to make inquiries for information about the symbols (names of variables, functions and types) defined in your program. This information is found by GDB in the symbol table loaded by the `symbol-file' command; it is inherent in the text of your program and does not change as the program executes.

whatis exp

Print the data type of expression exp. exp is not actually evaluated, and any side-effecting operations (such as assignments or function calls) inside it do not take place. See section 9.1 Expressions.

whatis

Print the data type of `$', the last value in the value history.

info address symbol

Describe where the data for symbol is stored. For register variables, this says which register. For other automatic variables, this prints the stack-frame offset at which the variable is always stored. Note the contrast with `print &symbol', which does not work at all for register variables and for automatic variables prints the exact address of the current instantiation of the variable.

ptype typename

Print a description of data type typename. typename may be the name of a type, or for C code it may have the form `struct struct-tag', `union union-tag' or `enum enum-tag'.

info sources

Print the names of all source files in the program for which there is debugging information.

info functions

Print the names and data types of all defined functions.

info functions regexp

Print the names and data types of all defined functions whose names contain a match for regular expression regexp. Thus, `info fun step' finds all functions whose names include `step'; `info fun ^step' finds those whose names start with `step'.

info variables

Print the names and data types of all variables that are declared outside of functions.

info variables regexp

Print the names and data types of all variables, declared outside of functions, whose names contain a match for regular expression regexp.

info types

Print all data types that are defined in the program.

info types regexp

Print all data types that are defined in the program whose names contain a match for regular expression regexp.

info methods

info methods regexp

The `info-methods' command permits the user to examine all defined methods within C++ program, or (with the regexp argument) a specific set of methods found in the various C++ classes. Many C++ classes which implement a large number of differently typed methods implement a large number of methods as well. Thus, the `ptype' command can give the user a tremendous overdose of information about what methods are associated with a given class. The `info-methods' command filters these methods do to only those methods which match the regular-expression search key.

printsyms filename

Write a complete dump of the debugger's symbol data into the file filename.

11. Altering Execution

There are several ways to alter the execution of your program with GDB commands.

11.1 Assignment to Variables		Altering variable values or memory contents.
11.2 Continuing at a Different Address		Altering control flow.
11.3 Giving the Program a Signal		Making signals happen in the program.
11.4 Returning from a Function		Making a function return prematurely.

11.1 Assignment to Variables

To alter the value of a variable, evaluate an assignment expression. See section 9.1 Expressions. For example,

print x=4

would store the value 4 into the variable x, and then print the value of the assignment expression (which is 4).

If you are not interested in seeing the value of the assignment, use the `set' command instead of the `print' command. `set' is really the same as `print' except that the expression's value is not printed and is not put in the value history (see section 9.6 Value History). The expression is evaluated only for side effects.

Note that if the beginning of the argument string of the `set' command appears identical to a `set' subcommand, it may be necessary to use the `set variable' command. This command is identical to `set' except for its lack of subcommands.

GDB allows more implicit conversions in assignments than C does; you can freely store an integer value into a pointer variable or vice versa, and any structure can be converted to any other structure that is the same length or shorter.

In C, all the other assignment operators such as `+=' and `++' are supported as well.

To store into arbitrary places in memory, use the `{...}' construct to generate a value of specified type at a specified address (see section 9.1 Expressions). For example,

set {int}0x83040 = 4

11.2 Continuing at a Different Address

jump linenum

Resume execution at line number linenum. Execution may stop immediately if there is a breakpoint there.

The `jump' command does not change the current stack frame, or the stack pointer, or the contents of any memory location or any register other than the program counter. If line linenum is in a different function from the one currently executing, the results may be wild if the two functions expect different patterns of arguments or of local variables. For this reason, the `jump' command requests confirmation if the specified line is not in the function currently executing. However, even wild results are predictable based on changing the program counter.

jump *address

Resume execution at the instruction at address address.

A similar effect can be obtained by storing a new value into the register `$pc', but not exactly the same.

set $pc = 0x485

specifies the address at which execution will resume, but does not resume execution. That does not happen until you use the `cont' command or a stepping command (see section 6.4 Stepping).

11.3 Giving the Program a Signal

signal signalnum

Resume execution where the program stopped, but give it immediately the signal number signalnum.

Alternatively, if signalnum is zero, continue execution and give no signal. This is useful when the program has received a signal but you don't want the program to see that signal; the `cont' command would signal the program.

11.4 Returning from a Function

You can make any function call return immediately, using the `return' command.

First select the stack frame that you wish to return from (see section 7.3 Selecting a Frame). Then type the `return' command. If you wish to specify the value to be returned, give that as an argument.

This pops the selected stack frame (and any other frames inside of it), leaving its caller as the innermost remaining frame. That frame becomes selected. The specified value is stored in the registers used for returning values of functions.

The `return' command does not resume execution; it leaves the program stopped in the state that would exist if the function had just returned. Contrast this with the `finish' command (see section 6.4 Stepping), which resumes execution until the selected stack frame returns naturally.

12. Canned Sequences of Commands

GDB provides two ways to store sequences of commands for execution as a unit: user-defined commands and command files.

12.1 User-Defined Commands		User-defined commands.
12.2 Command Files		Command files.
12.3 Commands for Controlled Output		Controlled output commands useful in user-defined commands and command files.

12.1 User-Defined Commands

A user-defined command is a sequence of GDB commands to which you assign a new name as a command. This is done with the `define' command.

define commandname

Define a command named commandname. If there is already a command by that name, you are asked to confirm that you want to redefine it.

The definition of the command is made up of other GDB command lines, which are given following the `define' command. The end of these commands is marked by a line containing `end'.

document commandname

Give documentation to the user-defined command commandname. The command commandname must already be defined. This command reads lines of documentation just as `define' reads the lines of the command definition, ending with `end'. After the `document' command is finished, `help' on command commandname will print the documentation you have specified.

You may use the `document' command again to change the documentation of a command. Redefining the command with `define' does not change the documentation.

User-defined commands do not take arguments. When they are executed, the commands of the definition are not printed. An error in any command stops execution of the user-defined command.

Commands that would ask for confirmation if used interactively proceed without asking when used inside a user-defined command. Many GDB commands that normally print messages to say what they are doing omit the messages when used in user-defined command.

12.2 Command Files

A command file for GDB is a file of lines that are GDB commands. Comments (lines starting with `#') may also be included. An empty line in a command file does nothing; it does not mean to repeat the last command, as it would from the terminal.

When GDB starts, it automatically executes its init files, command files named `.gdbinit'. GDB reads the init file (if any) in your home directory and then the init file (if any) in the current working directory. (The init files are not executed if the `-nx' option is given.) You can also request the execution of a command file with the `source' command:

source filename

Execute the command file filename.

The lines in a command file are executed sequentially. They are not printed as they are executed. An error in any command terminates execution of the command file.

Commands that would ask for confirmation if used interactively proceed without asking when used in a command file. Many GDB commands that normally print messages to say what they are doing omit the messages when used in a command file.

12.3 Commands for Controlled Output

During the execution of a command file or a user-defined command, the only output that appears is what is explicitly printed by the commands of the definition. This section describes three commands useful for generating exactly the output you want.

echo text

Print text. Nonprinting characters can be included in text using C escape sequences, such as `\n' to print a newline. No newline will be printed unless you specify one.

A backslash at the end of text is ignored. It is useful for outputting a string ending in spaces, since trailing spaces are trimmed from all arguments. A backslash at the beginning preserves leading spaces in the same way, because `\ ' as an escape sequence stands for a space. Thus, to print ` and foo = ', do

echo \ and foo = \

output expression

Print the value of expression and nothing but that value: no newlines, no `$nn = '. The value is not entered in the value history either. See section 9.1 Expressions for more information on expressions.

output/fmt expression

Print the value of expression in format fmt. See section 9.4 Formats, for more information.

printf string, expressions...

Print the values of the expressions under the control of string. The expressions are separated by commas and may be either numbers or pointers. Their values are printed as specified by string, exactly as if the program were to execute

printf (string, expressions...);

For example, you can print two values in hex like this:

printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo

The only backslash-escape sequences that you can use in the string are the simple ones that consist of backslash followed by a letter.

13. Using GDB under GNU Emacs

A special interface allows you to use GNU Emacs to view (and edit) the source files for the program you are debugging with GDB.

To use this interface, use the command M-x gdb in Emacs. Give the executable file you want to debug as an argument. This command starts a GDB process as a subprocess of Emacs, with input and output through a newly created Emacs buffer.

Using this GDB process is just like using GDB normally except for two things:

• All "terminal" input and output goes through the Emacs buffer. This applies both to GDB commands and their output, and to the input and output done by the program you are debugging.

  This is useful because it means that you can copy the text of previous commands and input them again; you can even use parts of the output in this way.

  All the facilities of Emacs's Shell mode are available for this purpose.

• GDB displays source code through Emacs. Each time GDB displays a stack frame, Emacs automatically finds the source file for that frame and puts an arrow (`=>') at the left margin of the current line.

  Explicit GDB `list' or search commands still produce output as usual, but you probably will have no reason to use them.

In the GDB I/O buffer, you can use these special Emacs commands:

M-s

Execute to another source line, like the GDB `step' command.

M-n

Execute to next source line in this function, skipping all function calls, like the GDB `next' command.

M-i

Execute one instruction, like the GDB `stepi' command.

M-u

Move up one stack frame (and display that frame's source file in Emacs), like the GDB `up' command.

M-d

Move down one stack frame (and display that frame's source file in Emacs), like the GDB `down' command. (This means that you cannot delete words in the usual fashion in the GDB buffer; I am guessing you won't often want to do that.)

C-c C-f

Execute until exit from the selected stack frame, like the GDB `finish' command.

In any source file, the Emacs command C-x SPC (gdb-break) tells GDB to set a breakpoint on the source line point is on.

The source files displayed in Emacs are in ordinary Emacs buffers which are visiting the source files in the usual way. You can edit the files with these buffers if you wish; but keep in mind that GDB communicates with Emacs in terms of line numbers. If you add or delete lines from the text, the line numbers that GDB knows will cease to correspond properly to the code.

14. Remote Kernel Debugging

GDB has a special facility for debugging a remote machine via a serial connection. This can be used for kernel debugging.

The program to be debugged on the remote machine needs to contain a debugging device driver which talks to GDB over the serial line using the protocol described below. The same version of GDB that is used ordinarily can be used for this.

14.1 Commands for Remote Debugging

Commands used to start and finish remote debugging.

For details of the communication protocol, see the comments in the GDB source file `remote.c'.

14.1 Commands for Remote Debugging

To start remote debugging, first run GDB and specify as an executable file the program that is running in the remote machine. This tells GDB how to find the program's symbols and the contents of its pure text. Then establish communication using the `attach' command with a device name rather than a pid as an argument. For example:

attach /dev/ttyd

if the serial line is connected to the device named `/dev/ttyd'. This will stop the remote machine if it is not already stopped.

Now you can use all the usual commands to examine and change data and to step and continue the remote program.

To resume the remote program and stop debugging it, use the `detach' command.

Command Index

Jump to:	!   $   A   B   C   D   E   F   H   I   J   K   L   N   O   P   Q   R   S   T   U   W   X

Concept Index

Jump to:	$   .   A   B   C   D   E   F   I   O   P   R   S   U   V   W

Table of Contents

Summary of GDB
GDB General Public License

Copying Policies
NO WARRANTY

1. GDB Input Conventions
2. Specifying GDB's Files

2.1 Specifying Files with Arguments
2.2 Specifying Files with Commands

3. Options and Arguments for GDB

3.1 Mode Options
3.2 File-specifying Options
3.3 Other Arguments

4. Compiling Your Program for Debugging
5. Running Your Program Under GDB

5.1 Your Program's Arguments
5.2 Your Program's Environment
5.3 Your Program's Working Directory
5.4 Your Program's Input and Output
5.5 Debugging an Already-Running Process

6. Stopping and Continuing

6.1 Signals
6.2 Breakpoints

6.2.1 Setting Breakpoints
6.2.2 Clearing Breakpoints
6.2.3 Disabling Breakpoints
6.2.4 Break Conditions
6.2.5 Commands Executed on Breaking
6.2.6 "Cannot Insert Breakpoints" Error

6.3 Continuing
6.4 Stepping

7. Examining the Stack

7.1 Stack Frames
7.2 Backtraces
7.3 Selecting a Frame
7.4 Information on a Frame

8. Examining Source Files

8.1 Printing Source Lines
8.2 Searching Source Files
8.3 Specifying Source Directories

9. Examining Data

9.1 Expressions
9.2 Program Variables
9.3 Artificial Arrays
9.4 Formats

9.4.1 Examining Memory

9.5 Automatic Display
9.6 Value History
9.7 Convenience Variables
9.8 Registers

9.8.1 Examples

10. Examining the Symbol Table
11. Altering Execution

11.1 Assignment to Variables
11.2 Continuing at a Different Address
11.3 Giving the Program a Signal
11.4 Returning from a Function

12. Canned Sequences of Commands

12.1 User-Defined Commands
12.2 Command Files
12.3 Commands for Controlled Output

13. Using GDB under GNU Emacs
14. Remote Kernel Debugging

14.1 Commands for Remote Debugging

Command Index
Concept Index

Short Table of Contents

Summary of GDB
GDB General Public License
1. GDB Input Conventions
2. Specifying GDB's Files
3. Options and Arguments for GDB
4. Compiling Your Program for Debugging
5. Running Your Program Under GDB
6. Stopping and Continuing
7. Examining the Stack
8. Examining Source Files
9. Examining Data
10. Examining the Symbol Table
11. Altering Execution
12. Canned Sequences of Commands
13. Using GDB under GNU Emacs
14. Remote Kernel Debugging
Command Index
Concept Index

About this document

• 1. Section One
• 1.1 Subsection One-One
• ...
• 1.2 Subsection One-Two
• 1.2.1 Subsubsection One-Two-One
• 1.2.2 Subsubsection One-Two-Two
• 1.2.3 Subsubsection One-Two-Three     <== Current Position
• 1.2.4 Subsubsection One-Two-Four
• 1.3 Subsection One-Three
• ...
• 1.4 Subsection One-Four