\input texinfo @setfilename gdb.info @synindex ky cp @c FOR UPDATES LEADING TO THIS DRAFT, GDB CHANGELOG CONSULTED BETWEEN: @c Tue Feb 26 01:47:07 1991 Cygnus John Gilmore (cygnus at yuba) @c Sat Dec 22 02:51:40 1990 John Gilmore (gnu at cygint) @ifinfo This file documents the GNU debugger GDB. Copyright (C) 1988, 1989, 1990, 1991 Free Software Foundation, Inc. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. @ignore Permission is granted to process this file through Tex and print the results, provided the printed document carries copying permission notice identical to this one except for the removal of this paragraph (this paragraph not being relevant to the printed manual). @end ignore Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided also that the section entitled ``GNU General Public License'' is included exactly as in the original, and provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that the section entitled ``GNU General Public License'' may be included in a translation approved by the author instead of in the original English. @end ifinfo @c @smallbook @setchapternewpage odd @settitle Using GDB (v3.94) @iftex @finalout @end iftex @titlepage @title{Using GDB} @subtitle{A Guide to the GNU Source-Level Debugger} @sp 1 @c Maybe crank this up to "Fourth Edition" when released at FSF @c @subtitle Third Edition---GDB version 4.0 @subtitle GDB version 3.94 @subtitle January 1991 @author{Richard M. Stallman} @author{(Revised by Roland Pesch for Cygnus Support)} @page @tex \def\$#1${{#1}} % Kluge: collect RCS revision info without $...$ \xdef\manvers{\$Revision$} % For use in headers, footers too {\parskip=0pt \hfill Cygnus Support\par \hfill \manvers\par \hfill \TeX{}info \texinfoversion\par } @end tex @vskip 0pt plus 1filll Copyright @copyright{} 1988, 1989, 1990, 1991 Free Software Foundation, Inc. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided also that the section entitled ``GNU General Public License'' is included exactly as in the original, and provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that the section entitled ``GNU General Public License'' may be included in a translation approved by the author instead of in the original English. @end titlepage @page @node Top, New Features, (dir), (dir) @unnumbered Summary of GDB The purpose of a debugger such as GDB is to allow you to see what is going on ``inside'' another program while it executes---or what another program was doing at the moment it crashed. We call the other program ``your program'' or ``the program being debugged''. GDB can do four main kinds of things (plus other things in support of these) to help you catch bugs in the act: @enumerate @item Start your program, specifying anything that might affect its behavior. @item Make your program stop on specified conditions. @item Examine what has happened, when your program has stopped. @item Change things in your program, so you can experiment with correcting the effects of one bug and go on to learn about another. @end enumerate 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. @unnumberedsec Free Software GDB is Free Software, protected by the GNU General Public License (GPL). The GPL gives you the freedom to copy or adapt a licensed program---but every person getting a copy also gets with it the freedom to modify that copy (which means that they must get access to the source code), and the freedom to distribute further copies. Typical software companies use copyrights to limit your freedoms; the Free Software Foundation uses the GPL to preserve these freedoms. Fundamentally, the General Public License is a license which says that you have these freedoms and that you can't take these freedoms away from anyone else. For full details, @pxref{License}. @menu * New Features:: New Features in GDB version 3.94 * Invocation:: Starting GDB * User Interface:: GDB Commands and Displays * Files:: Specifying GDB's Files * Compilation:: Compiling Your Program for Debugging * Targets:: Specifying a Debugging Target * Running:: Running Your Program Under GDB * Stopping:: Stopping and Continuing * Stack:: Examining the Stack * Source:: Examining Source Files * Data:: Examining Data * Symbols:: Examining the Symbol Table * Altering:: Altering Execution * Sequences:: Canned Sequences of Commands * Emacs:: Using GDB under GNU Emacs * Remote:: Remote Debugging * GDB Bugs:: Reporting Bugs in GDB * Installing GDB:: Installing GDB * License:: GNU GENERAL PUBLIC LICENSE * Commands:: Command Index * Concepts:: Index --- The Detailed Node Listing --- Starting GDB * File Options:: File-specifying Options and Arguments * Mode Options:: Mode Options * Remote i960-Nindy:: Starting GDB with a Remote Intel 960 (Nindy) Specifying a Debugging Target * Active Targets:: Active Targets * Target Commands:: Commands for Managing Targets Running Your Program Under GDB * Arguments:: Specifying the arguments for your program. * Environment:: Specifying the environment for your program. * Working Directory:: Specifying the working directory for giving to your program when it is run. * Input/Output:: Specifying the program's standard input and output. * Attach:: Debugging a process started outside GDB. * Kill Process:: Getting rid of the child process running your program. Stopping and Continuing * Signals:: Fatal signals in your program just stop it; then you can use GDB to see what is going on. * Breakpoints:: Breakpoints let you stop your program when it reaches a specified point in the code. an expression changes. * Continuing:: Resuming execution until the next signal or breakpoint. * Stepping:: Stepping runs the program a short distance and then stops it wherever it has come to. Breakpoints * Set Breaks:: How to establish breakpoints. * Exception Handling:: How GDB supports exception handling for C++. * Delete Breaks:: How to remove breakpoints no longer needed. * Disabling:: How to disable breakpoints (turn them off temporarily). * Conditions:: Making extra conditions on whether to stop. * Break Commands:: Commands to be executed at a breakpoint. * Error in Breakpoints:: Examining the Stack * Frames:: Explanation of stack frames and terminology. * Backtrace:: Summarizing many frames at once. * Selection:: How to select a stack frame. * Frame Info:: Information on a Frame Examining Source Files * List:: Using the @samp{list} command to print source files. * Search:: Commands for searching source files. * Source Path:: Specifying the directories to search for source files. Examining Data * Expressions:: Expressions that can be computed and printed. * Variables:: Using your program's variables in expressions. * Arrays:: Examining part of memory as an array. * Format options:: Controlling how structures and arrays are printed. * Output formats:: Specifying formats for printing values. * Auto Display:: Printing certain expressions whenever program stops. * Value History:: Referring to values previously printed. * Convenience Vars:: Giving names to values for future reference. * Registers:: Referring to and storing in machine registers. Output formats * Memory:: Examining Memory Altering Execution * Assignment:: Altering variable values or memory contents. * Jumping:: Altering control flow. * Signaling:: Making signals happen in the program. * Returning:: Making a function return prematurely. * Calling:: Calling functions from your program Canned Sequences of Commands * Define:: User-defined commands. * Command Files:: Command files. * Output:: Controlled output commands useful in user-defined commands and command files. Remote Debugging * Remote Commands:: Commands used to start and finish remote debugging. Reporting Bugs in GDB * Bug Criteria:: Have You Found a Bug? * Bug Reporting:: How to Report Bugs @end menu @node New Features, Invocation, Top, Top @unnumbered New Features in GDB version 3.94 @itemize @bullet @item TARGETS: Using the new command @samp{target}, you can select at runtime whether you are debugging local files, local processes, standalone systems over the serial port, realtime systems over a TCP/IP connection, etc. GDB now uses a function vector to mediate access to all the different possible targets, making it much easier to add support for new remote protocols. @item WATCHPOINTS: GDB now sports watchpoints as well as breakpoints. You can use a watchpoint to stop execution whenever the value of an expression changes, without having to predict a particular place in the inferior process where this may happen. @item OBJECT CODE FORMATS: GDB uses a new scheme called Binary File Descriptors (BFD) to permit it to switch dynamically, without reconfiguration or recompilation, between different object-file formats. Formats currently supported are COFF, a.out, and the new Intel 960 b.out; files may be read as .o's, archive libraries, or core dumps. BFD is available as a subroutine library so that other programs may take advantage of it, and the other GNU binary utilities are being converted to use it. @item CONFIGURATION: You must still choose a particular machine architecture and operating system for GDB's host and target systems when GDB is built. The script @samp{config.gdb} now handles specification of separate host and target configurations. @item INTERACTION: GDB now uses the GNU @code{readline} interface to read its input; this provides inline editing of commands, using the familiar Emacs or VI keymaps, and command-history support. The user interface to GDB's control variables has been simplified and consolidated in two commands, @samp{set} and @samp{show}. Output lines are now broken at readable places, rather than overflowing onto the next line. @item SOURCE LANGUAGE: GDB now understands C++ source as well as C. Multiple inheritance is supported when used with G++ 2.0. There is also limited support for C++ exception handling: GDB can break when an exception is raised, before the stack is peeled back to the exception handler's context. You can suppress output of machine-level addresses, displaying only source language information. @item PORTS: GDB has been ported to the following new architectures: AT&T 3b1, Acorn RISC machine, HP300 running HPUX, big- and little- endian MIPS machines, Motorola 88k, Sun 386i, and Sun 3 running SunOS 4. In addition, the following are supported as targets only: AMD 29k, Intel 960, and Wind River's VxWorks. @item SHARED LIBRARIES: GDB 3.94 supports SunOS shared libraries. @item WORK IN PROGRESS: kernel debugging for BSD and Mach systems; Tahoe and HPPA architecture support. @end itemize @node Invocation, User Interface, New Features, Top @chapter Starting GDB GDB is invoked with the shell command @samp{gdb}. Once started, it reads commands from the terminal until you tell it to exit. The most usual way to start GDB is with one argument or two, specifying an executable program as the argument: @example gdb program @end example @noindent or you can start with both an executable program and a core file specified: @example gdb program core @end example You can get more detailed control over how GDB starts up using a number of command-line options. All the options and command line arguments given are processed in sequential order. The order makes a difference when the @samp{-x} option is used. @menu * File Options:: File-specifying Options and Arguments * Mode Options:: Mode Options * Remote i960-Nindy:: Starting GDB with a Remote Intel 960 (Nindy) @end menu @node File Options, Mode Options, Invocation, Invocation @section File-specifying Options and Arguments As shown in the example, any arguments other than options specify an executable file and core file; that is, the first argument encountered with no associated option flag is equivalent to a @samp{-se} option, and the second, if any, is equivalent to a @samp{-c} option. @table @code @item -s @var{file} Read symbol table from file @var{file}. @item -e @var{file} Use file @var{file} as the executable file to execute when appropriate, and for examining pure data in conjunction with a core dump. @item -se @var{file} Read symbol table from file @var{file} and use it as the executable file. @item -c @var{file} Use file @var{file} as a core dump to examine. @item -x @var{file} Execute GDB commands from file @var{file}. @xref{Command Files}. @item -d @var{directory} Add @var{directory} to the path to search for source files. @end table @node Mode Options, Remote i960-Nindy, File Options, Invocation @section Mode Options @table @code @item -nx Do not execute commands from the init files @file{.gdbinit}. Normally, the commands in these files are executed after all the command options and arguments have been processed. @xref{Command Files}. @item -q ``Quiet''. Do not print the introductory and copyright messages. These messages are also suppressed in batch mode, or if an executable file name is specified on the GDB command line. @item -batch Run in batch mode. Exit with code @code{0} after processing all the command files specified with @samp{-x} (and @file{.gdbinit}, if not inhibited). Exit with nonzero status if an error occurs in executing the GDB commands in the command files. Batch mode may be useful for running GDB as a filter, for example to download and run a program on another computer; in order to make this more useful, the message @example Program exited normally. @end example @noindent (which is ordinarily issued whenever a program running under GDB control terminates) is not issued when running in batch mode. @item -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 @samp{\032} characters, followed by the file name, line number and character position separated by colons, and a newline. The Emacs-to-GDB interface program uses the two @samp{\032} characters as a signal to display the source code for the frame. @item -b @var{bps} Set the line speed (baud rate or bps) of any serial interface used by GDB for remote debugging. @end table @node Remote i960-Nindy, , Mode Options, Invocation @section Starting GDB with a Remote Intel 960 (Nindy) ``Nindy'' is the name of a Rom Monitor program for Intel 960 target systems. When GDB is configured to control a remote Intel 960 using Nindy, you can tell GDB how to connect to the 960 in several ways: @itemize @bullet @item Through command line options specifying device, baud rate, and protocol; @item By responding to a prompt on startup; @item By using the @samp{target} command at any point during your GDB session. @end itemize The command-line options for Nindy are detailed below. If you simply start @code{gdb960} without using options to specify a serial port, you are prompted for it, @emph{before} you reach the ordinary GDB prompt: @example Attach /dev/ttyNN -- specify NN, or "quit" to quit: @end example @noindent You can, if you choose, simply start up with no Nindy connection by responding to the prompt with an empty line. If you do this, and later wish to attach to Nindy, use @samp{target} (@pxref{Target Commands}). These are the startup options for beginning your GDB session with a Nindy-960 board attached: @table @code @item -r @var{port} Specify the serial port name of a serial interface to be used to connect to the target system. This option is only available when GDB is configured for the Intel 960 target architecture. You may specify @var{port} as any of: a full pathname (e.g. @samp{-r /dev/ttya}), a device name in @samp{/dev} (e.g. @samp{-r ttya}), or simply the unique suffix for a specific @code{tty} (e.g. @samp{-r a}). @item -O (An uppercase letter ``O'', not a zero.) Specify that GDB should use the ``old'' Nindy monitor protocol to connect to the target system. This option is only available when GDB is configured for the Intel 960 target architecture. @quotation @emph{Warning:} if you specify @samp{-O}, but are actually trying to connect to a target system using the current protocol, the connection will fail appearing to be a speed mismatch, and GDB will repeatedly attempt to reconnect at several different line speeds. You can abort this process with an interrupt. @end quotation @item -brk Specify that GDB should first send a @samp{BREAK} signal to the target system, in an attempt to reset it, before connecting to a Nindy target. This option is only available when GDB is configured for the Intel 960 target architecture. @quotation @emph{Warning:} Many target systems do not have the hardware that this requires; it only works with a few boards. @end quotation @end table The standard @samp{-b} option controls the line speed used on the serial port. @node User Interface, Files, Invocation, Top @chapter GDB Commands and Displays 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. For example, the command @samp{step} accepts an argument which is the number of times to step, as in @samp{step 5}. You can also use the @samp{step} command with no arguments. Some command names do not allow any arguments. @cindex abbreviation GDB command names may always be truncated if that abbreviation is unambiguous. Other possible command abbreviations are listed in the documentation of the individual commands. Sometimes even ambiguous abbreviations are allowed; for example, @samp{s} is specially defined as equivalent to @samp{step} even though there are other commands whose names start with @samp{s}. @cindex repeating commands A blank line as input to GDB means to repeat the previous command. Certain commands will not repeat this way; these are commands for which unintentional repetition might cause trouble and which you are unlikely to want to repeat. Certain others (@samp{list} and @samp{x}) act differently when repeated because that is more useful. @kindex # @cindex comment A line of input starting with @samp{#} is a comment; it does nothing. This is useful mainly in command files (@xref{Command Files}). @cindex online documentation @kindex help @table @code @item help @itemx help @var{category} @itemx help @var{command} You can always ask GDB itself for information on its commands, using the command @samp{help}. With a command name as argument, it will display a paragraph on how to use the command. Used with no arguments, @samp{help} displays a short list of named categories of commands; you can then use @samp{help @var{category}} to list the individual commands in a category. @kindex info version @item info version As GDB evolves, new commands are introduced, and old ones may wither away. If multiple versions of GDB are in use at your site, it may occasionally be useful to make sure what version of GDB you're running. GDB announces its version whenever it starts up; but you can make it repeat this information with the @samp{info version} command. @end table @cindex prompt GDB indicates its readiness to read a command by printing a string called the @dfn{prompt}. This string is normally @samp{(gdb)}. You can change the prompt string with the @samp{set prompt} command. For instance, when debugging GDB with GDB, it is useful to change the prompt in one of the GDBs so that you tell which one you are talking to. @table @code @item set prompt @var{newprompt} @kindex set prompt Directs GDB to use @var{newprompt} as its prompt string henceforth. @kindex show prompt @item show prompt Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}} @end table @cindex exiting GDB @kindex quit To exit GDB, use the @samp{quit} command (abbreviated @samp{q}), or type an end-of-file character (usually @ctrl{d}). An interrupt (often @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 the interrupt character at any time because GDB does not allow it to take effect until a time when it is safe. @cindex readline @cindex command line editing @cindex history substitution GDB reads its input commands via the @code{readline} interface. This GNU library provides consistent behavior for programs which provide a command line interface to the user. Advantages are @samp{emacs}-style or @samp{vi}-style inline editing of commands, @samp{csh}-like history substitution, and a storage and recall of command history across debugging sessions. You may control the behavior of command line editing in GDB with the command @samp{set}. You may check the status of any of these settings with the command @samp{show}. @table @code @kindex set editing @cindex editing @item set editing @itemx set editing on Enable command line editing (enabled by default). @item set editing off Disable command line editing. @kindex show editing @item show editing Show whether command line editing is enabled. @cindex history file @kindex set history file @item set history file @var{filename} Set the name of the GDB command history file to @samp{filename}. This is the file from which GDB will read an initial command history list or to which it will write this list when it exits. This list is accessed through history expansion or through the history command editing characters listed below. This file defaults to the value of the environmental variable @code{GDBHISTFILE}, or to @code{./.gdb_history} if this variable is not set. @cindex history write @kindex set history write @item set history write @itemx set history write on Make GDB record command history in a file, whose name may be specified with the @samp{set history file} command. By default, this option is disabled. @item set history write off Make GDB stop recording command history in a file. @cindex history size @kindex set history size @item set history size @var{size} Set the number of commands which GDB will keep in its history list. This defaults to the value of the environmental variable @code{HISTSIZE}, or to 256 if this variable is not set. @end table @cindex history expansion History expansion assigns special meaning to the character @samp{!} (@pxref{Event Designators}). Since @samp{!} is also the logical not operator in C, history expansion is off by default. If you decide to enable history expansion with the @samp{set history expansion on} command, you may sometimes need to follow @samp{!} (when it is used as logical not, in an expression) with a space or a tab to prevent it from being expanded. The @code{readline} history facilities will not attempt substitution on the strings @samp{!=} and @samp{!(}, even when history expansion is enabled. The commands to control history expansion are: @table @code @kindex set history expansion @item set history expansion on @itemx set history expansion Enable history expansion. History expansion is off by default. @item set history expansion off Disable history expansion. The @code{readline} code comes with more complete documentation of editing and history expansion features. Users unfamiliar with @samp{emacs} or @samp{vi} may wish to read it. @xref{Command Line Editing}. @kindex show history @item show history @itemx show history file @itemx show history write @itemx show history size @itemx show history expansion These commands display the state of the GDB history parameters. @samp{show history} by itself displays all four states. @kindex show @kindex info set @item show @itemx info set This chapter introduces a number of internal GDB variables that you can control with the @samp{set} command, and display with the @samp{show} command. A number of others are introduced throughout the manual. To display all the settable parameters and their current values, you can use @samp{show} with no arguments; you may also use @samp{info set}. Both commands produce the same display. @end table @table @code @kindex info editing @item info editing Display the last ten commands in the command history. @item info editing @var{n} Print ten commands centered on command number @var{n}. @item info editing + Print ten commands just after the commands last printed. @end table Occasionally it is useful to execute a shell command from within GDB. This can be done with the @samp{shell} command. @table @code @item shell @var{command string} @kindex shell @cindex shell escape Directs GDB to invoke an inferior shell to execute @var{command string}. The environment variable @code{SHELL} is used if it exists, otherwise GDB uses @samp{/bin/sh}. @end table The utility @samp{make} is often needed in development environments. You don't have to use the @samp{shell} command for this purpose in GDB: @table @code @item make @dots{} @kindex make @cindex calling make Causes GDB to execute an inferior @code{make} program with the specified arguments. This is equivalent to @samp{shell make @dots{}}. @end table @cindex screen size @cindex pauses in output Certain commands to GDB may produce large amounts of information output to the screen. To help you read all of it, GDB pauses and asks you for input at the end of each page of output. Type @key{RET} when you want to continue the output. Normally GDB knows the size of the screen from the termcap data base together with the value of the @code{TERM} environment variable and the @code{stty rows} and @code{stty cols} settings. If this is not correct, you can override it with the @samp{set screen-height} and @samp{set screen-width} commands: GDB also uses the screen width setting to determine when to wrap lines of output. Depending what is being printed, it tries to break the line at a readable place, rather than simply letting it overflow onto the following line. @table @code @item set screen-height @var{lpp} @itemx show screen-height @itemx set screen-width @var{cpl} @itemx show screen-width @kindex set screen-height @kindex set screen-width @kindex show screen-width @kindex show screen-height These @samp{set} commands specify a screen height of @var{lpp} lines and a screen width of @var{cpl} characters. The associated @samp{show} commands display the current settings. If you specify a height of zero lines, GDB will not pause during output no matter how long the output is. This is useful if output is to a file or to an editor buffer. @end table @cindex number representation @cindex entering numbers You can always enter numbers in octal, decimal, or hexadecimal in GDB by the usual conventions: octal numbers begin with @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}. Numbers that begin with none of these are, by default, entered in base 10; likewise, the default display for numbers---when no particular format is specified---is base 10. You can change the default base for both input and output with the @samp{set radix} command. @table @code @kindex set radix @item set radix @var{base} Set the default base for numeric input and display. Supported choices for @var{base} are decimal 8, 10, 16. @var{base} must itself be specified either unambiguously or using the current default radix; for example, any of @example set radix 012 set radix 10. set radix 0xa @end example @noindent will set the base to decimal. On the other hand, @samp{set radix 10} will leave the radix unchanged no matter what it was. @kindex show radix @item show radix Display the current default base for numeric input and display. @end table By default, GDB is silent about its inner workings. If you are running on a slow machine, you may want to use the @samp{set verbose} command. It will make GDB tell you when it does a lengthy internal operation, so you won't think it has crashed. Currently, the messages controlled by @samp{set verbose} are those which announce that the symbol table for a source file is being read (@pxref{Files}, in the description of the command @samp{symbol-file}). @c The following is the right way to do it, but emacs 18.55 doesn't support @c @ref, and neither the emacs lisp manual version of texinfmt or makeinfo @c is released. @ignore see @samp{symbol-file} in @ref{Files}). @end ignore @table @code @kindex set verbose @item set verbose on Enables GDB's output of certain informational messages. @item set verbose off Disables GDB's output of certain informational messages. @kindex show verbose @item show verbose Displays whether @samp{set verbose} is on or off. @end table By default, if GDB encounters bugs in the symbol table of an object file, it prints a single message about each type of problem it finds, then shuts up. You can suppress these messages, or allow more than one such message to be printed if you want to see how frequent the problems are. @xref{Files}. @table @code @kindex set complaints @item set complaints @var{limit} Permits GDB to output @var{limit} complaints about each type of unusual symbols before becoming silent about the problem. Set @var{limit} to zero to suppress all complaints; set it to a large number to prevent complaints from being suppressed. @kindex show complaints @item show complaints Displays how many symbol complaints GDB is permitted to produce. @end table By default, GDB is cautious, and asks what sometimes seem to be a lot of stupid questions. For example, if you try to run a program which is already running: @example (gdb) run The program being debugged has been started already. Start it from the beginning? (y or n) @end example If you're willing to unflinchingly face the consequences of your own commands, you can disable this ``feature'': @table @code @kindex set caution @cindex flinching @cindex stupid questions @item set caution off Disables cautious questions. @item set caution on Enables cautious questions (the default). @item show caution @kindex show caution Displays state of cautious questions. @end table @node Files, Compilation, User Interface, Top @chapter Specifying GDB's Files @cindex core dump file @cindex executable file @cindex symbol table GDB needs to know the file name of the program to be debugged, both in order to read its symbol table and in order to start the program. To debug a core dump of a previous run, GDB must be told the file name of the core dump. The usual way to specify the executable and core dump file names is with the command arguments given when you start GDB, as discussed in @pxref{Invocation}. 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. @table @code @item file @var{filename} @kindex file Use @var{filename} as the program to be debugged. It is read for its symbols, for getting the contents of pure memory, and it is the program executed when you use the @samp{run} command. If you do not specify a directory and the file is not found in GDB's working directory, GDB will use the environment variable @code{PATH} as a list of directories to search, just as the shell does when looking for a program to run. @samp{file} with no argument makes both executable file and symbol table become unspecified. @item exec-file @var{filename} @kindex exec-file Specify that the program to be run (but not the symbol table) is found in @var{filename}. GDB will search the environment variable @code{PATH} if necessary to locate the program. @item symbol-file @var{filename} @kindex symbol-file Read symbol table information from file @var{filename}. @code{PATH} is searched when necessary. Use the @samp{file} command to get both symbol table and program to run from the same file. @samp{symbol-file} with no argument clears out GDB's information on your program's symbol table. The @samp{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. On some kinds of object files, the @samp{symbol-file} command does not actually read the symbol table in full right away. Instead, it scans the symbol table quickly to find which source files and which symbols are present. The details are read later, one source file at a time, when they are needed. The purpose of this two-stage reading strategy is to make GDB start up faster. For the most part, it is invisible except for occasional pauses while the symbol table details for a particular source file are being read. (The @samp{set verbose} command can turn these pauses into messages if desired. @xref{User Interface}). When the symbol table is stored in COFF format, @samp{symbol-file} does read the symbol table data in full right away. We haven't implemented the two-stage strategy for COFF yet. When GDB is configured for a particular environment, it will understand debugging information in whatever format is the standard generated for that environment; you may use either the GNU compiler GCC, or other compilers that adhere to the local conventions. Best results are usually obtained from GCC; for example, using GCC you can generate debugging information for optimized code. While the symbol file is being read, GDB will occasionally encounter problems, such as symbol types it does not recognize, or known bugs in compiler output. By default, it prints one message about each such type of problem, no matter how many times the problem occurs. You can ask it to print more messages, to see how many times the problems occur, or can shut the messages off entirely, with the @samp{set complaints} command (@xref{User Interface}). The messages currently printed, and their meanings, are: @table @code @item inner block not inside outer block in @var{symbol} The symbol information shows where symbol scopes begin and end (such as at the start of a function or a block of statements). This error indicates that an inner scope block is not fully contained in its outer scope blocks. GDB circumvents the problem by treating the inner block as if it had the same scope as the outer block. @var{symbol} may be ``(don't know)'' if the outer block is not a function. @item block at @var{address} out of order The symbol information for symbol scope blocks should occur in order of increasing addresses. This error indicates that it does not do so. GDB does not circumvent this problem, and will have trouble locating symbols in the source file whose symbols being read. (You can often determine what source file is affected by turning on @samp{info verbose}. @xref{User Interface}.) @item bad block start address patched The symbol information for a symbol scope block has a start address smaller than the address of the preceding source line. This is known to occur in the SunOS 4.1.1 (and earlier) C compiler. GDB circumvents the problem by treating the symbol scope block as starting on the previous source line. @comment @item{encountered DBX-style class variable debugging information. @comment You seem to have compiled your program with "g++ -g0" instead of "g++ -g". @comment Therefore GDB will not know about your class variables} @comment @comment This error indicates that the symbol information produced for a C++ @comment program includes zero-size fields, which indicated static fields in @comment a previous release of the G++ compiler. This message is probably @comment obsolete. @comment @item bad string table offset in symbol @var{n} Symbol number @var{n} contains a pointer into the string table which is larger than the size of the string table. GDB circumvents the problem by considering the symbol to have the name @code{foo}, which may cause other problems if many symbols end up with this name. @cindex{foo} @item unknown symbol type @code{0xNN} The symbol information contains new data types that GDB does not yet know how to read. @code{0xNN} is the symbol type of the misunderstood information, in hexadecimal. GDB circumvents the error by ignoring this symbol information. This will usually allow the program to be debugged, though certain symbols will not be accessible. If you encounter such a problem and feel like debugging it, you can debug gdb with itself, breakpoint on "complain", then go "up" to read_dbx_symtab() and examine *bufp to see the symbol. @c @item stub type has NULL name @c @c FIXME, Mike Tiemann needs to write about what this means. @item const/volatile indicator missing, got 'X' The symbol information for a C++ type is missing some information that the compiler should have output for it. @item C++ type mismatch between compiler and debugger The debugger could not parse a type specification output by the compiler for some C++ object. @end table @item core-file @var{filename} @itemx core @var{filename} @kindex core @kindex core-file Specify the whereabouts of a core dump file to be used as the ``contents of memory''. Traditionally, core files contain only some parts of the address space of the process that generated them; GDB can access the executable file itself for other parts. @samp{core-file} with no argument specifies that no core file is to be used. Note that the core file is ignored when your program is actually running under GDB. So, if you have been running the program and you wish to debug a core file instead, you must kill the subprocess in which the program is running. To do this, use the @samp{kill} command (@pxref{Kill Process}). @item load @var{filename} @kindex load This command will dynamically link @var{filename} on the current target, performing any necessary downloads, then add @var{filename}'s symbol table in the same way as the @samp{add-syms} command. @item add-syms @var{filename} @var{address} @kindex add-syms @cindex dynamic linking The @samp{add-syms} command reads additional symbol table information from the file @var{filename}. You would use this command when that file has been dynamically loaded (by some other means) into the program that is running. @var{address} should be the memory address at which the file has been loaded; GDB cannot figure this out for itself. The symbol table of the file @var{filename} is added to the symbol table originally read with the @samp{symbol-file} command. You can use the @samp{add-syms} command any number of times; the new symbol data thus read keeps adding to the old. The @samp{symbol-file} command forgets all the symbol data GDB has read. @item info files @itemx info target @kindex info files @kindex info target @samp{info files} and @samp{info target} are synonymous; both print the current targets (@pxref{Targets}), including the names of the executable and core dump files currently in use by GDB, and the files from which symbols were loaded. Beware: the similar command @samp{info targets} lists all possible targets rather than current ones. @end table 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. @kindex sharedlibrary @kindex share @cindex shared libraries GDB supports the SunOS shared library format. Symbols from a shared library cannot be referenced before the shared library has been linked with the program. (That is to say, after one types @samp{run} and the function @code{main()} has been entered; or when examining core files.) Once the shared library has been linked in, you can use the following commands: @table @code @item sharedlibrary @var{regex} @itemx share @var{regex} Load shared object library symbols for files matching a UNIX regular expression. @item share @itemx sharedlibrary Load symbols for all shared libraries. @item info share @itemx info sharedlibrary @kindex info sharedlibrary @kindex info share Print the names of the shared libraries which are currently loaded. @end table @node Compilation, Targets, Files, Top @chapter Compiling Your Program for Debugging In order to debug a program effectively, you need to ask for debugging information when you compile it. This debugging information is stored in the object file; it 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 @samp{-g} option when you run the compiler. The Unix C compiler is unable to handle the @samp{-g} and @samp{-O} options together. This means that you cannot ask for optimization if you ask for debugger information. The GNU C compiler supports @samp{-g} with or without @samp{-O}, making it possible to debug optimized code. We recommend that you @emph{always} use @samp{-g} whenever you compile a program. You may think the program is correct, but there's no sense in pushing your luck. Some things do not work as well with @samp{-g -O} as with just @samp{-g}, particularly on machines with instruction scheduling. If in doubt, recompile with @samp{-g} alone, and if this fixes the problem, please report it as a bug (including a test case---@pxref{GDB Bugs}). Older versions of the GNU C compiler, GCC, permitted a variant option @samp{-gg} for debugging information. GDB no longer supports this format; if your GNU C compiler has this option, do not use it. @ignore @comment As far as I know, there are no cases in which GDB will @comment produce strange output in this case. (but no promises). If your program includes archives made with the @code{ar} program, and if the object files used as input to @code{ar} were compiled without the @samp{-g} option and have names longer than 15 characters, GDB will get confused reading the program's symbol table. No error message will be given, but GDB may behave strangely. The reason for this problem is a deficiency in the Unix archive file format, which cannot represent file names longer than 15 characters. To avoid this problem, compile the archive members with the @samp{-g} option or use shorter file names. Alternatively, use a version of GNU @code{ar} dated more recently than August 1989. @end ignore @node Targets, Running, Compilation, Top @chapter Specifying a Debugging Target @cindex debugging target @kindex target A @dfn{target} is an interface between the debugger and a particular kind of file or process. Often, you will be able to run GDB in the same host environment as the program you are debugging; in that case, the debugging target is specified as a side effect of the @samp{file} or @samp{core} commands. When you need more flexibility---for example, running GDB on a physically separate host, controlling standalone systems over a serial port, or realtime systems over a TCP/IP connection---you can use the @samp{target} command. @menu * Active Targets:: Active Targets * Target Commands:: Commands for Managing Targets @end menu @node Active Targets, Target Commands, Targets, Targets @section Active Targets @cindex stacking targets @cindex active targets @cindex multiple targets Targets are managed in three @dfn{strata} that correspond to different classes of target: processes, core files, and executable files. This allows you to (for example) start a process and inspect its activity without abandoning your work on a core file. More than one target can potentially respond to a request. In particular, when you access memory GDB will walk down the three strata of targets until it finds a target that can handle that particular address. Strata are always examined in a fixed order: first a process if there is one, then a core file if there is one, and finally an executable file if there is one of those. When you specify a new target in a given stratum, it replaces any target previously in that stratum. To get rid of a target without replacing it, use the @samp{detach} command. The related command @samp{attach} provides you with a way of choosing a particular running process as a new target. @xref{Attach}. @node Target Commands, , Active Targets, Targets @section Commands for Managing Targets @table @code @item target @var{type} @var{parameters} Connects the GDB host environment to a target machine or process. A target is typically a protocol for talking to debugging facilities. You use the argument @var{type} to specify the type or protocol of the target machine; for example, @samp{target vxworks} for a TCP/IP link to a VxWorks system. Further @var{parameters} are interpreted by the target protocol, but typically include things like device names or host names to connect with, process numbers, and baud rates. Executing @example target @var{type} @end example @noindent (without any parameters) will issue a message about what parameters are required for that target type. @item info targets @kindex info targets Displays the names of all targets available. Beware: the similar command @samp{info target} displays targets currently in use rather than all available ones. @samp{info files} gives the same information as @samp{info target} (@pxref{Files}). @end table Here are some common targets (available, or not, depending on GDB configuration): @table @code @item target remote @var{dev} @kindex target remote Remote serial target in gdb-specific protocol. The argument @var{dev} specifies what serial device to use for the connection (e.g. @code{/dev/ttya}). @item target exec @var{prog} @kindex target exec An executable file. @samp{target exec @var{prog}} is the same as @samp{exec-file @var{prog}}. @item target core @var{filename} @kindex target core A core dump file. @samp{target core @var{filename}} is the same as @samp{core-file @var{filename}}. @item target nindy @var{devicename} @kindex target nindy An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is the name of the serial device to use for the connection, e.g. @samp{/dev/ttya}. @item target vxworks @var{machinename} @kindex target vxworks A VxWorks system, attached via TCP/IP. The argument @var{machinename} is the target system's machine name or IP address. @end table Different targets are available on different configurations of GDB; your configuration may have more or fewer targets. @node Running, Stopping, Targets, Top @chapter Running Your Program Under GDB @cindex running @kindex run To start your program under GDB, use the @samp{run} command. Except on VxWorks, the program must already have been specified using the @samp{file} or @samp{exec-file} command, or with an argument to GDB (@pxref{Files}). On targets that support processes, @samp{run} creates an inferior process and makes that process run your program. On other targets, @samp{run} jumps to the location it has recorded for the start of the program. The execution of a program is affected by certain information it receives from its superior. GDB provides ways to specify this information, which you must do @i{before} starting the program. (You can change it after starting the program, but such changes do not affect the program unless you start it over again.) This information may be divided into three categories: @table @asis @item The @i{arguments.} You specify the arguments to give the program as the arguments of the @samp{run} command. If a shell is available on your target, the shell is used to pass the arguments, so that you may use normal conventions (for example regular expression expansion or variable substitution) in describing the arguments. In Unix systems, you can control which shell is used with the @code{SHELL} environment variable. @item The @i{environment.} The program normally inherits its environment from GDB, but you can use the GDB commands @samp{set environment} and @samp{unset environment} to change parts of the environment that will be given to the program.@refill @item The @i{working directory.} The program inherits its working directory from GDB. You can set GDB's working directory with the @samp{cd} command in GDB. @end table When you issue the @samp{run} command, your program begins to execute immediately. @xref{Stopping}, for discussion of how to arrange for your program to stop. Note that once your program has been started by the @samp{run} command, you may evaluate expressions that involve calls to functions in the inferior, using the @samp{print} or @samp{call} commands. @xref{Data}. If the modification time of your symbol file has changed since the last time GDB read its symbols, GDB will discard its symbol table and re-read it. In this process, it tries to retain your current breakpoints. @menu * Arguments:: Specifying the arguments for your program. * Environment:: Specifying the environment for your program. * Working Directory:: Specifying the working directory for giving to your program when it is run. * Input/Output:: Specifying the program's standard input and output. * Attach:: Debugging a process started outside GDB. * Kill Process:: Getting rid of the child process running your program. @end menu @node Arguments, Environment, Running, Running @section Your Program's Arguments @cindex arguments (to your program) The arguments to your program are specified by the arguments of the @samp{run} command. They are passed to a shell, which expands wildcard characters and performs redirection of I/O, and thence to the program. @samp{run} with no arguments uses the same arguments used by the previous @samp{run}. @kindex set args @table @code @item set args The command @samp{set args} can be used to specify the arguments to be used the next time the program is run. If @samp{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. @item show args @kindex show args Show the arguments to give your program when it is started. @end table @node Environment, Working Directory, Arguments, Running @section Your Program's Environment @cindex environment (of your program) The @dfn{environment} consists of a set of @dfn{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. @table @code @item show environment @var{varname} @kindex show environment Print the value of environment variable @var{varname} to be given to your program when it is started. @item show environment Print the names and values of all environment variables to be given to your program when it is started. @item set environment @var{varname} @var{value} @itemx set environment @var{varname} = @var{value} @kindex set environment Sets environment variable @var{varname} to @var{value}, for your program only, not for GDB itself. @var{value} may be any string; the values of environment variables are just strings, and any interpretation is supplied by your program itself. The @var{value} parameter is optional; if it is eliminated, the variable is set to a null value. For example, this command: @example set env USER = foo @end example @noindent tells the program, when subsequently run, to assume it is being run on behalf of the user named @samp{foo}. @item delete environment @var{varname} @itemx unset environment @var{varname} @kindex delete environment @kindex unset environment Remove variable @var{varname} from the environment to be passed to your program. This is different from @samp{set env @var{varname}@ =}; @samp{delete environment} removes the variable from the environment, rather than assigning it an empty value. This command can be abbreviated @samp{d e}. @end table @node Working Directory, Input/Output, Environment, Running @section Your Program's Working Directory @cindex working directory (of your program) Each time you start your program with @samp{run}, it inherits its working directory from the current working directory of GDB. GDB's working directory is initially whatever it inherited from its parent process (typically the shell), but you can specify a new working directory in GDB with the @samp{cd} command. The GDB working directory also serves as a default for the commands that specify files for GDB to operate on. @xref{Files}. @table @code @item cd @var{directory} @kindex cd Set GDB's working directory to @var{directory}. @item pwd @kindex pwd Print GDB's working directory. @end table @node Input/Output, Attach, Working Directory, Running @section Your Program's Input and Output @cindex redirection @cindex i/o @cindex terminal @cindex controlling terminal 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 @samp{sh}-style redirection commands in the @samp{run} command. For example, @example run > outfile @end example @noindent starts the program, diverting its output to the file @file{outfile}. @kindex tty Another way to specify where the program should do input and output is with the @samp{tty} command. This command accepts a file name as argument, and causes this file to be the default for future @samp{run} commands. It also resets the controlling terminal for the child process, for future @samp{run} commands. For example, @example tty /dev/ttyb @end example @noindent directs that processes started with subsequent @samp{run} commands default to do input and output on the terminal @file{/dev/ttyb} and have that as their controlling terminal. An explicit redirection in @samp{run} overrides the @samp{tty} command's effect on input/output redirection, but not its effect on the controlling terminal. When you use the @samp{tty} command or redirect input in the @samp{run} command, only the @emph{input for your program} is affected. The input for GDB still comes from your terminal. @node Attach, Kill Process, Input/Output, Running @section Debugging an Already-Running Process @kindex attach @cindex attach @table @code @item attach @var{process--id} This command attaches to a running process, if your currently selected target supports processes. (@samp{info files} will show your active targets). The command takes as argument a process ID. You specify a process ID to debug an already-running process that was started outside of GDB. (The usual way to find out the process-id of a Unix process is with the @code{ps} utility, or with the @code{jobs -l} shell command.) In this case, you must have permission to send the process a signal, and it must have the same effective user ID as the debugger. @end table When using @samp{attach}, you should first use the @samp{file} command to specify the program running in the process and load its symbol table. 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 @samp{run}. You can insert breakpoints; you can step and continue; you can modify storage. If you would rather the process continue running, you may use the @samp{continue} command after attaching GDB to the process. @kindex detach When you have finished debugging the attached process, you can use the @samp{detach} command to release it from GDB's control. Detaching the process continues its execution. After the @samp{detach} command, that process and GDB become completely independent once more, and you are ready to @samp{attach} another process or start one with @samp{run}. If you exit GDB or use the @samp{run} command while you have an attached process, you kill that process. By default, you will be asked for confirmation if you try to do either of these things; you can control whether or not this happens by using the @samp{set caution} command (@pxref{User Interface}). @node Kill Process, , Attach, Running @section Killing the Child Process @table @code @item kill @kindex kill Kill the child process in which your program is running under GDB. @end table This command is useful if you wish to debug a core dump instead. GDB ignores any core dump file if it is actually running the program. On some operating systems, you can't execute your program in another process while breakpoints are active inside GDB. The @samp{kill} command is also useful in this situation, if you wish to run the program outside the debugger. The @samp{kill} command is also useful if you wish to recompile and relink the program, since on many systems it is impossible to modify an executable file which is running in a process. In this case, when you next type @samp{run}, GDB will notice that the file has changed, and will re-read the symbol table (while trying to preserve your current breakpoint settings). @node Stopping, Stack, Running, Top @chapter Stopping and Continuing When you run a program normally, it runs until it terminates. The principal 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 investigate and find out why. @menu * Signals:: Fatal signals in your program just stop it; then you can use GDB to see what is going on. * Breakpoints:: Breakpoints let you stop your program when it reaches a specified point in the code. an expression changes. * Continuing:: Resuming execution until the next signal or breakpoint. * Stepping:: Stepping runs the program a short distance and then stops it wherever it has come to. @end menu @node Signals, Breakpoints, Stopping, Stopping @section Signals @cindex 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, @code{SIGINT} is the signal a program gets when you type @kbd{Ctrl-c}; @code{SIGSEGV} is the signal a program gets from referencing a place in memory far away from all the areas in use; @code{SIGALRM} occurs when the alarm clock timer goes off (which happens only if the program has requested an alarm). @cindex fatal signals Some signals, including @code{SIGALRM}, are a normal part of the functioning of the program. Others, such as @code{SIGSEGV}, indicate errors; these signals are @dfn{fatal} (kill the program immediately) if the program has not specified in advance some other way to handle the signal. @code{SIGINT} does not indicate an error in the program, but it is normally fatal so it can carry out the purpose of @kbd{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. @cindex handling signals Normally, GDB is set up to ignore non-erroneous signals like @code{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 @samp{handle} command. @table @code @item info signal @kindex 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. @item handle @var{signal} @var{keywords}@dots{} @kindex handle Change the way GDB handles signal @var{signal}. @var{signal} can be the number of a signal or its name (with or without the @samp{SIG} at the beginning). The @var{keywords} say what change to make. @end table The keywords allowed by the @samp{handle} command can be abbreviated. Their full names are @table @code @item 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. @item stop GDB should stop the program when this signal happens. This implies the @samp{print} keyword as well. @item print GDB should print a message when this signal happens. @item noprint GDB should not mention the occurrence of the signal at all. This implies the @samp{nostop} keyword as well. @item 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. @item nopass GDB should not allow the program to see this signal. @end table 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 @samp{pass} is in effect for the signal in question @i{at that time}. In other words, after GDB reports a signal, you can use the @samp{handle} command with @samp{pass} or @samp{nopass} to control whether that signal will be seen by the program when you later continue it. You can also use the @samp{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. @xref{Signaling}. @node Breakpoints, Continuing, Signals, Stopping @section Breakpoints @cindex breakpoints A @dfn{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 @dfn{enabled} or @dfn{disabled}; if disabled, it has no effect on the program until you enable it again. @table @code @kindex info break @kindex $_ @item info break The command @samp{info break} prints a list of all breakpoints set and not deleted, showing their numbers, where in the program they are, and any special features in use for them. Disabled breakpoints are included in the list, but marked as disabled. @samp{info break} with a breakpoint number as argument lists only that breakpoint. The convenience variable @code{$_} and the default examining-address for the @samp{x} command are set to the address of the last breakpoint listed (@pxref{Memory}). @kindex info watch @item info watch This command prints a list of watchpoints. @cindex watchpoints A @dfn{watchpoint} is a special breakpoint that stops your program when the value of an expression changes. You can use a watchpoint to stop execution whenever the value of an expression changes, without having to predict a particular place in the inferior process where this may happen. Aside from the different syntax in setting a watchpoint, it is managed exactly like any other breakpoint and is enabled, disabled, and deleted using exactly the same commands. Watchpoints currently execute two orders of magnitude more slowly than other breakpoints, but this can well be worth it to catch errors where you have no clue what part of your program is the culprit. Some processors provide special hardware to implement this feature; future releases of GDB will use such hardware if it is available. @end table @menu * Set Breaks:: How to establish breakpoints. * Exception Handling:: How GDB supports exception handling for C++. * Delete Breaks:: How to remove breakpoints no longer needed. * Disabling:: How to disable breakpoints (turn them off temporarily). * Conditions:: Making extra conditions on whether to stop. * Break Commands:: Commands to be executed at a breakpoint. * Error in Breakpoints:: @end menu @node Set Breaks, Exception Handling, Breakpoints, Breakpoints @subsection Setting Breakpoints @kindex break @kindex watch Breakpoints are set with the @samp{break} command (abbreviated @samp{b}). Watchpoints are set with the @samp{watch} command. You have several ways to say where the breakpoint should go. @table @code @item break @var{function} Set a breakpoint at entry to function @var{function}. @item break @var{+offset} @itemx break @var{-offset} Set a breakpoint some number of lines forward or back from the position at which execution stopped in the currently selected frame. @item break @var{linenum} Set a breakpoint at line @var{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. @item break @var{filename}:@var{linenum} Set a breakpoint at line @var{linenum} in source file @var{filename}. @item break @var{filename}:@var{function} Set a breakpoint at entry to function @var{function} found in file @var{filename}. Specifying a file name as well as a function name is superfluous except when multiple files contain similarly named functions. @item break *@var{address} Set a breakpoint at address @var{address}. You can use this to set breakpoints in parts of the program which do not have debugging information or source files. @item break Set a breakpoint at the next instruction to be executed in the selected stack frame (@pxref{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 @samp{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. GDB normally ignores breakpoints when it resumes execution, until at least one instruction has been executed. If it did not do this, you would be unable to proceed past a breakpoint without first disabling the breakpoint. This rule applies whether or not the breakpoint already existed when the program stopped. @item break @dots{} if @var{cond} Set a breakpoint with condition @var{cond}; evaluate the expression @var{cond} each time the breakpoint is reached, and stop only if the value is nonzero. @samp{@dots{}} stands for one of the possible arguments described above (or no argument) specifying where to break. @xref{Conditions}, for more information on breakpoint conditions. @item tbreak @var{args} @kindex tbreak Set a breakpoint enabled only for one stop. @var{args} are the same as in the @samp{break} command, and the breakpoint is set in the same way, but the breakpoint is automatically disabled the first time it is hit. @xref{Disabling}. @item rbreak @var{regex} @kindex rbreak Set a breakpoint on all functions matching @var{regex}. This is useful for setting breakpoints on overloaded functions that are not members of any special classes. This command sets an unconditional breakpoint on all matches, printing a list of all breakpoints it set. Once these breakpoints are set, they are treated just like the breakpoints set with the @samp{break} command. They can be deleted, disabled, made conditional, etc., in the standard ways. @kindex watch @item watch @var{expr} Set a watchpoint for an expression. @end table 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 (@pxref{Conditions}). @node Exception Handling, Delete Breaks, Set Breaks, Breakpoints @subsection Breakpoints and Exception Handling @cindex exception handlers Some languages, such as GNU C++, implement exception handling. GDB can be used to examine what caused the program to raise an exception and to list the exceptions the program is prepared to handle at a given point in time. @cindex raise exceptions GNU C++ raises an exception by calling a library function named @code{__raise_exception} which has the following ANSI C interface: @example /* ADDR is where the exception identifier is stored. ID is the exception identifier. */ void __raise_exception (void **addr, void *id); @end example @noindent You can make the debugger catch all exceptions @emph{before} any stack unwinding takes place: set a breakpoint on @code{__raise_exception} (@pxref{Breakpoints}). If you set a breakpoint in an exception handler instead, it may not be easy to find out where the exception was raised. By using a conditional breakpoint (@xref{Conditions}), you can cause the debugger to stop only when a specific exception is raised. Multiple conditional breakpoints can be used to stop the program when any of a number of exceptions are raised. @table @code @item catch @var{exceptions} @kindex catch Breakpoints can be set at active exception handlers by using the @samp{catch} command. @var{exceptions} is a list of names of exceptions to catch. @end table You can use @samp{info catch} to list active exception handlers; @pxref{Frame Info}. There are currently some limitations to exception handling in GDB. These will be corrected in a future release. @itemize @bullet @item If you call a function interactively, GDB will normally return control to you when the function has finished executing. If the call raises an exception, however, the call may bypass the mechanism that returns control to the user and cause the program to simply continue running until it hits a breakpoint, catches a signal that GDB is listening for, or exits. @item You cannot raise an exception interactively. @item You cannot interactively install an exception handler. @end itemize @node Delete Breaks, Disabling, Exception Handling, Breakpoints @subsection Deleting Breakpoints @cindex clearing breakpoints and watchpoints @cindex deleting breakpoints and watchpoints 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 @dfn{deleting} the breakpoint. A breakpoint that has been deleted no longer exists in any sense; it is forgotten. With the @samp{clear} command you can delete breakpoints according to where they are in the program. With the @samp{delete} command you can delete individual breakpoints by specifying their breakpoint numbers. It is not necessary to delete a breakpoint to proceed past it. GDB automatically ignores breakpoints on the first instruction to be executed when you continue execution without changing the execution address. @table @code @item clear @kindex clear Delete any breakpoints at the next instruction to be executed in the selected stack frame (@pxref{Selection}). When the innermost frame is selected, this is a good way to delete a breakpoint that the program just stopped at. @item clear @var{function} @itemx clear @var{filename}:@var{function} Delete any breakpoints set at entry to the function @var{function}. @item clear @var{linenum} @itemx clear @var{filename}:@var{linenum} Delete any breakpoints set at or within the code of the specified line. @item delete breakpoints @var{bnums}@dots{} @itemx delete @var{bnums}@dots{} @itemx delete @kindex delete breakpoints @kindex delete Delete the breakpoints of the numbers specified as arguments. If no argument is specified, delete all breakpoints. @end table @node Disabling, Conditions, Delete Breaks, Breakpoints @subsection Disabling Breakpoints @cindex disabled breakpoints @cindex enabled breakpoints Rather than deleting a breakpoint, you might prefer to @dfn{disable} it. This makes the breakpoint inoperative as if it had been deleted, but remembers the information on the breakpoint so that you can @dfn{enable} it again later. You disable and enable breakpoints with the @samp{enable} and @samp{disable} commands, optionally specifying one or more breakpoint numbers as arguments. Use @samp{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: @itemize @bullet @item Enabled. The breakpoint will stop the program. A breakpoint made with the @samp{break} command starts out in this state. @item Disabled. The breakpoint has no effect on the program. @item Enabled once. The breakpoint will stop the program, but when it does so it will become disabled. A breakpoint made with the @samp{tbreak} command starts out in this state. @item Enabled for deletion. The breakpoint will stop the program, but immediately after it does so it will be deleted permanently. @end itemize You can use the following commands to enable or disable a breakpoint: @table @code @item disable breakpoints @var{bnums}@dots{} @itemx disable @var{bnums}@dots{} @itemx disable @kindex disable breakpoints @kindex disable Disable the specified breakpoints---or all breakpoints, if none are listed. 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. @item enable breakpoints @var{bnums}@dots{} @itemx enable @var{bnums}@dots{} @itemx enable @kindex enable breakpoints @kindex enable Enable the specified breakpoints (or all defined breakpoints). They become effective once again in stopping the program, until you specify otherwise. @item enable breakpoints once @var{bnums}@dots{} @itemx enable once @var{bnums}@dots{} 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). @item enable breakpoints delete @var{bnums}@dots{} @itemx enable delete @var{bnums}@dots{} 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). @end table Save for a breakpoint set with @samp{tbreak} (@pxref{Set Breaks}), breakpoints that you set are enabled or disabled only when you use one of the commands above. (The command @samp{until} can set and delete a breakpoint on its own, but it will not change the state of your breakpoints). @node Conditions, Break Commands, Disabling, Breakpoints @subsection Break Conditions @cindex conditional breakpoints @cindex breakpoint conditions The simplest sort of breakpoint breaks every time the program reaches a specified place. You can also specify a @dfn{condition} for a breakpoint. A condition is just a boolean expression in your programming language. (@xref{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 (@pxref{Break Commands}). Break conditions can be specified when a breakpoint is set, by using @samp{if} in the arguments to the @samp{break} command. @xref{Set Breaks}. They can also be changed at any time with the @samp{condition} command: @table @code @item condition @var{bnum} @var{expression} @kindex condition Specify @var{expression} as the break condition for breakpoint number @var{bnum}. From now on, this breakpoint will stop the program only if the value of @var{expression} is true (nonzero, in C). @var{expression} is not evaluated at the time the @samp{condition} command is given. When you call @samp{condition}, the expression you specify is checked immediately for syntactic correctness, and to determine whether symbols in it have referents in the context of your breakpoint. @xref{Expressions}. @item condition @var{bnum} Remove the condition from breakpoint number @var{bnum}. It becomes an ordinary unconditional breakpoint. @end table @cindex ignore count (of breakpoint) A special case of a breakpoint condition is to stop only when the breakpoint has been reached a certain number of times. This is so useful that there is a special way to do it, using the @dfn{ignore count} of the breakpoint. Every breakpoint has an ignore count, which is an integer. Most of the time, the ignore count is zero, and therefore has no effect. But if the program reaches a breakpoint whose ignore count is positive, then instead of stopping, it just decrements the ignore count by one and continues. As a result, if the ignore count value is @var{n}, the breakpoint will not stop the next @var{n} times it is reached. @table @code @item ignore @var{bnum} @var{count} @kindex ignore Set the ignore count of breakpoint number @var{bnum} to @var{count}. The next @var{count} times the breakpoint is reached, your program's execution will not stop; other than to decrement the ignore count, GDB takes no action. To make the breakpoint stop the next time it is reached, specify a count of zero. @item continue @var{count} @itemx cont @var{count} @itemx c @var{count} @itemx fg @var{count} @kindex cont @var{count} @kindex continue @var{count} Continue execution of the program, setting the ignore count of the breakpoint that the program stopped at to @var{count} minus one. Thus, the program will not stop at this breakpoint until the @var{count}'th time it is reached. This command is allowed only when the program stopped due to a breakpoint. At other times, the argument to @samp{cont} is ignored. The synonym @samp{fg} is provided purely for convenience, and has exactly the same behavior as other forms of the command. @end table 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 be checked. Note that you could achieve the effect of the ignore count with a condition such as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that is decremented each time. @xref{Convenience Vars}. @node Break Commands, Error in Breakpoints, Conditions, Breakpoints @subsection Commands Executed on Breaking @cindex breakpoint commands 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. @table @code @item commands @var{bnum} @kindex commands Specify a list of commands for breakpoint number @var{bnum}. The commands themselves appear on the following lines. Type a line containing just @samp{end} to terminate the commands. To remove all commands from a breakpoint, use the command @samp{commands} and follow it immediately by @samp{end}; that is, give no commands. With no arguments, @samp{commands} refers to the last breakpoint set (not to the breakpoint most recently encountered). @end table You can use breakpoint commands to start the program up again. Simply use the @samp{cont} command, or @samp{step}, or any other command to resume execution. However, if you do this, any further commands in the same breakpoint's command list are ignored. When the program stops again, GDB will act according to the cause of that stop. @kindex silent If the first command specified is @samp{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. @samp{silent} is not really a command; it is meaningful only at the beginning of the commands for a breakpoint. The commands @samp{echo} and @samp{output} that allow you to print precisely controlled output are often useful in silent breakpoints. @xref{Output}. For example, here is how you could use breakpoint commands to print the value of @code{x} at entry to @code{foo} whenever it is positive. @example break foo if x>0 commands silent echo x is\040 output x echo \n cont end @end example 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 @samp{cont} command so that the program does not stop, and start with the @samp{silent} command so that no output is produced. Here is an example: @example break 403 commands silent set x = y + 4 cont end @end example 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 @example condition 5 (x = y + 4), 0 @end example @noindent specifies a condition expression (@xref{Expressions}) that will change @code{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 @samp{&&}, @samp{||} and @samp{?@dots{}:} may be useful. @node Error in Breakpoints, , Break Commands, Breakpoints @subsection ``Cannot Insert Breakpoints'' Error Under some operating systems, breakpoints cannot be used in a program if any other process is running that program. In this situation, attempting to run or continue a program with a breakpoint will cause GDB to stop the other process. When this happens, you have three ways to proceed: @enumerate @item Remove or disable the breakpoints, then continue. @item Suspend GDB, and copy the file containing the program to a new name. Resume GDB and use the @samp{exec-file} command to specify that GDB should run the program under that name. Then start the program again. @item Relink the program so that the text segment is nonsharable, using the linker option @samp{-N}. The operating system limitation may not apply to nonsharable executables. @end enumerate @node Continuing, Stepping, Breakpoints, Stopping @section 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. @table @code @item continue @item cont @kindex cont @kindex continue Continue running the program at the place where it stopped. @end table 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, @samp{cont} takes special care to prevent that from happening. You do not need to delete 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 @samp{cont} command. @xref{Conditions}. If the program stopped because of a signal other than @code{SIGINT} or @code{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 @samp{signal 0}. @xref{Signaling}. You can also act in advance to control what signals your program will see, using the @samp{handle} command (@pxref{Signals}). @node Stepping, , Continuing, Stopping @section Stepping @cindex stepping @dfn{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. @table @code @item step @kindex step Continue running the program until control reaches a different line, then stop it and return control to the debugger. This command is abbreviated @samp{s}. This command may be given when control is within a function for which there is no debugging information. In that case, execution will proceed until control reaches a different function, or is about to return from this function. An argument repeats this action. @item step @var{count} Continue running as in @samp{step}, but do so @var{count} times. If a breakpoint is reached or a signal not related to stepping occurs before @var{count} steps, stepping stops right away. @item next @kindex next Similar to @samp{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 @samp{next} command was given. This command is abbreviated @samp{n}. An argument is a repeat count, as in @samp{step}. @samp{next} within a function without debugging information acts as does @samp{step}, but any function calls appearing within the code of the function are executed without stopping. @item finish @kindex 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 @samp{return} command (@pxref{Returning}). @item until @kindex until This command is used to avoid single stepping through a loop more than once. It is like the @samp{next} command, except that when @samp{until} encounters a jump, it automatically continues execution until the program counter is greater than the address of the jump. This means that when you reach the end of a loop after single stepping though it, @samp{until} will cause the program to continue execution until the loop is exited. In contrast, a @samp{next} command at the end of a loop will simply step back to the beginning of the loop, which would force you to step through the next iteration. @samp{until} always stops the program if it attempts to exit the current stack frame. @samp{until} may produce somewhat counterintuitive results if the order of the source lines does not match the actual order of execution. For example, in a typical C @code{for}-loop, the third expression in the @code{for}-statement (the loop-step expression) is executed after the statements in the body of the loop, but is written before them. Therefore, the @samp{until} command would appear to step back to the beginning of the loop when it advances to this expression. However, it has not really done so, not in terms of the actual machine code. Note that @samp{until} with no argument works by means of single instruction stepping, and hence is slower than @samp{until} with an argument. @item until @var{location} Continue running the program until either the specified location is reached, or the current (innermost) stack frame returns. @var{location} is any of the forms of argument acceptable to @samp{break} (@pxref{Set Breaks}). This form of the command uses breakpoints, and hence is quicker than @samp{until} without an argument. @item stepi @itemx si @kindex stepi @kindex si Execute one machine instruction, then stop and return to the debugger. It is often useful to do @samp{display/i $pc} when stepping by machine instructions. This will cause the next instruction to be executed to be displayed automatically at each stop. @xref{Auto Display}. An argument is a repeat count, as in @samp{step}. @item nexti @itemx ni @kindex nexti @kindex ni Execute one machine instruction, but if it is a subroutine call, proceed until the subroutine returns. An argument is a repeat count, as in @samp{next}. @end table A typical technique for using stepping is to put a breakpoint (@pxref{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 @samp{cont} command can be used after stepping to resume execution until the next breakpoint or signal. @node Stack, Source, Stopping, Top @chapter Examining the Stack When your program has stopped, the first thing you need to know is where it stopped and how it got there. @cindex call stack 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 @dfn{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 @dfn{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 @dfn{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 @samp{frame} command does (@pxref{Frame Info, Info}). @menu * Frames:: Explanation of stack frames and terminology. * Backtrace:: Summarizing many frames at once. * Selection:: How to select a stack frame. * Frame Info:: Information on a Frame @end menu @node Frames, Backtrace, Stack, Stack @section Stack Frames @cindex frame @cindex stack frame The call stack is divided up into contiguous pieces called @dfn{stack frames}, or @dfn{frames} for short; 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. @cindex initial frame @cindex outermost frame @cindex innermost frame When your program is started, the stack has only one frame, that of the function @code{main}. This is called the @dfn{initial} frame or the @dfn{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 @dfn{innermost} frame. This is the most recently created of all the stack frames that still exist. @cindex frame pointer 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 @dfn{frame pointer register} while execution is going on in that frame. @cindex frame number 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. @cindex selected frame Many GDB commands refer implicitly to one stack frame, called the @dfn{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. @cindex frameless execution Some compilers allow functions to be compiled to run without a frame reserved for them on the stack. (For example, the GCC option @samp{-fomit-frame-pointer} will generate functions without a frame.) This is occasionally done with heavily used library functions to save the frame setup time. GDB has limited facilities for dealing with these function invocations; if the innermost function invocation has no stack frame, GDB will give it a virtual stack frame of 0 and correctly allow tracing of the function call chain. Results are undefined if a function invocation besides the innermost one is frameless. @node Backtrace, Selection, Frames, Stack @section 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. @table @code @item backtrace @itemx bt @kindex backtrace @kindex 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 @kbd{Control-C}. @item backtrace @var{n} @itemx bt @var{n} Similar, but print only the innermost @var{n} frames. @item backtrace @var{-n} @itemx bt @var{-n} Similar, but print only the outermost @var{n} frames. @end table @kindex where @kindex info stack The names @samp{where} and @samp{info stack} are additional aliases for @samp{backtrace}. Every line in the backtrace shows the frame number and the function name. The program counter value is also shown---unless you use @samp{set addressprint off}. If the function is in a source file whose symbol table data has been fully read, the backtrace shows the source file name and line number, as well as the arguments to the function. When the line number is shown, the program counter value is omitted if it is at the beginning of the code for that line number. Here is an example of a backtrace. It was made with the command @samp{bt 3}, so it shows the innermost three frames. @example #0 rtx_equal_p (x=(rtx) 0x8e58c, y=(rtx) 0x1086c4) \ (/gp/rms/cc/rtlanal.c line 337) #1 0x246b0 in expand_call (...) (...) #2 0x21cfc in expand_expr (...) (...) (More stack frames follow...) @end example @noindent The functions @code{expand_call} and @code{expand_expr} are in a file whose symbol details have not been fully read. Full detail is available for the function @code{rtx_equal_p}, which is in the file @file{rtlanal.c}. Its arguments, named @code{x} and @code{y}, are shown with their typed values. @node Selection, Frame Info, Backtrace, Stack @section 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. @table @code @item frame @var{n} @kindex frame Select frame number @var{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 @code{main}'s frame. @item frame @var{addr} Select the frame at address @var{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. @item up @var{n} @kindex up Select the frame @var{n} frames up from the frame previously selected. For positive numbers @var{n}, this advances toward the outermost frame, to higher frame numbers, to frames that have existed longer. @var{n} defaults to one. @item down @var{n} @kindex down Select the frame @var{n} frames down from the frame previously selected. For positive numbers @var{n}, this advances toward the innermost frame, to lower frame numbers, to frames that were created more recently. @var{n} defaults to one. @end table 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: @example #3 main (argc=3, argv=??, env=??) at main.c:67 67 read_input_file (argv[i]); @end example After such a printout, the @samp{list} command with no arguments will print ten lines centered on the point of execution in the frame. @xref{List}. @table @code @item up-silently @var{n} @itemx down-silently @var{n} @kindex down-silently @kindex up-silently These two commands are variants of @samp{up} and @samp{down}, respectively; they differ in that they do their work silently, without causing display of the new frame. They are intended primarily for use in GDB command scripts, where the output might be unnecessary and distracting. @end table @node Frame Info, , Selection, Stack @section Information on a Frame There are several other commands to print information about the selected stack frame. @table @code @item frame When used without any argument, this command does not change which frame is selected, but still prints a brief description of the currently selected stack frame. It can be abbreviated @samp{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 kind of information. @item info frame @kindex 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. @item info frame @var{addr} Print a verbose description of the frame at address @var{addr}, without selecting that frame. The selected frame remains unchanged by this command. @item info args @kindex info args Print the arguments of the selected frame, each on a separate line. @item info locals @kindex 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. @item info catch @kindex info catch @cindex catch exceptions @cindex exception handlers Print a list of all the exception handlers that are active in the current stack frame given the current value of @code{pc}. To see other exception handlers, visit the associated frame (using the @samp{up}, @samp{down}, or @samp{frame} commands); then type @samp{info catch}. @xref{Exception Handling}. @end table @node Source, Data, Stack, Top @chapter 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 (@pxref{Selection}), GDB prints the line which execution in that frame has stopped in. You can also print parts of source files by explicit command. @menu * List:: Using the @samp{list} command to print source files. * Search:: Commands for searching source files. * Source Path:: Specifying the directories to search for source files. @end menu @node List, Search, Source, Source @section Printing Source Lines @kindex list @kindex l To print lines from a source file, use the @samp{list} command (abbreviated @samp{l}). There are several ways to specify what part of the file you want to print. Here are the forms of the @samp{list} command most commonly used: @table @code @item list @var{linenum} Print ten lines centered around line number @var{linenum} in the current source file. @item list @var{function} Print ten lines centered around the beginning of function @var{function}. @item list Print ten more lines. If the last lines printed were printed with a @samp{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 (@pxref{Stack}), this prints ten lines centered around that line. @item list - Print ten lines just before the lines last printed. @end table Repeating a @samp{list} command with @key{RET} discards the argument, so it is equivalent to typing just @samp{list}. This is more useful than listing the same lines again. An exception is made for an argument of @samp{-}; that argument is preserved in repetition so that each repetition moves up in the file. @cindex linespec In general, the @samp{list} command expects you to supply zero, one or two @dfn{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 @samp{list}: @table @code @item list @var{linespec} Print ten lines centered around the line specified by @var{linespec}. @item list @var{first},@var{last} Print lines from @var{first} to @var{last}. Both arguments are linespecs. @item list ,@var{last} Print ten lines ending with @var{last}. @item list @var{first}, Print ten lines starting with @var{first}. @item list + Print ten lines just after the lines last printed. @item list - Print ten lines just before the lines last printed. @item list As described in the preceding table. @end table Here are the ways of specifying a single source line---all the kinds of linespec. @table @code @item @var{linenum} Specifies line @var{linenum} of the current source file. When a @samp{list} command has two linespecs, this refers to the same source file as the first linespec. @item +@var{offset} Specifies the line @var{offset} lines after the last line printed. When used as the second linespec in a @samp{list} command that has two, this specifies the line @var{offset} lines down from the first linespec. @item -@var{offset} Specifies the line @var{offset} lines before the last line printed. @item @var{filename}:@var{linenum} Specifies line @var{linenum} in the source file @var{filename}. @item @var{function} Specifies the line of the open-brace that begins the body of the function @var{function}. @item @var{filename}:@var{function} Specifies the line of the open-brace that begins the body of the function @var{function} in the file @var{filename}. The file name is needed with a function name only for disambiguation of identically named functions in different source files. @item *@var{address} Specifies the line containing the program address @var{address}. @var{address} may be any expression. @end table One other command is used to map source lines to program addresses. @table @code @item info line @var{linenum} @kindex info line Print the starting and ending addresses of the compiled code for source line @var{linenum}. @kindex $_ The default examine address for the @samp{x} command is changed to the starting address of the line, so that @samp{x/i} is sufficient to begin examining the machine code (@pxref{Memory}). Also, this address is saved as the value of the convenience variable @code{$_} (@pxref{Convenience Vars}). @end table @node Search, Source Path, List, Source @section Searching Source Files @cindex searching @kindex search @kindex forward-search @kindex reverse-search There are two commands for searching through the current source file for a regular expression. The command @samp{forward-search @var{regexp}} checks each line, starting with the one following the last line listed, for a match for @var{regexp}. It lists the line that is found. You can abbreviate the command name as @samp{fo}. The synonym @samp{search @var{regexp}} is also supported. The command @samp{reverse-search @var{regexp}} checks each line, starting with the one before the last line listed and going backward, for a match for @var{regexp}. It lists the line that is found. You can abbreviate this command with as little as @samp{rev}. @node Source Path, , Search, Source @section Specifying Source Directories @cindex source path @cindex directories for source files Executable programs sometimes do not record the directories of the source files from which they were compiled, just the names. Even when they do, the directories could be moved between the compilation and your debugging session. GDB remembers a list of directories to search for source files; this is called the @dfn{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. Note that the executable search path is @emph{not} used for this purpose. Neither is the current working directory, unless it happens to be in the source path. If it can't find a source file in the source path, and the object program records what directory it was compiled in, GDB tries that directory too. If the source path is empty, and there is no record of the compilation directory, GDB will, as a last resort, look in the current directory. Whenever you reset or rearrange the source path, GDB will clear out any information it has cached about where source files are found, where each line is in the file, etc. @kindex directory When you start GDB, its source path is empty. To add other directories, use the @samp{directory} command. @table @code @item directory @var{dirnames...} Add directory @var{dirname} to the front of the source path. Several directory names may be given to this command, separated by whitespace or @samp{:}. If a name is already in the source path, it is moved to the front of the path, so it will be searched sooner. @item directory Reset the source path to empty again. This requires confirmation. @item info directories @kindex info directories Print the source path: show which directories it contains. @end table Because the @samp{directory} command, when used with arguments, adds to the front of the source path, it can affect files 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: @enumerate @item Use @samp{directory} with no argument to reset the source path to empty. @item Use @samp{directory} with suitable arguments to add any other directories you want in the source path. You can add all the directories in one command. @end enumerate @node Data, Symbols, Source, Top @chapter Examining Data @cindex printing data @cindex examining data @kindex print @kindex inspect The usual way to examine data in your program is with the @samp{print} command (abbreviated @samp{p}), or its synonym @samp{inspect}. It evaluates and prints the value of any valid expression of the language the program is written in (for now, C or C++). You type @example print @var{exp} @end example @noindent where @var{exp} is any valid expression (in the source language), and the value of @var{exp} is printed in a format appropriate to its data type. A more low-level way of examining data is with the @samp{x} command. It examines data in memory at a specified address and prints it in a specified format. @menu * Expressions:: Expressions that can be computed and printed. * Variables:: Using your program's variables in expressions. * Arrays:: Examining part of memory as an array. * Format options:: Controlling how structures and arrays are printed. * Output formats:: Specifying formats for printing values. * Auto Display:: Printing certain expressions whenever program stops. * Value History:: Referring to values previously printed. * Convenience Vars:: Giving names to values for future reference. * Registers:: Referring to and storing in machine registers. @end menu @node Expressions, Variables, Data, Data @section Expressions @cindex 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 @code{#define} commands, or C++ expressions involving @samp{::}, the name resolution operator. 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: @table @code @item @@ @samp{@@} is a binary operator for treating parts of memory as arrays. @xref{Arrays}, for more information. @item :: @samp{::} allows you to specify a variable in terms of the file or function it is defined in. @xref{Variables}. @item @{@var{type}@} @var{addr} Refers to an object of type @var{type} stored at address @var{addr} in memory. @var{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 @var{addr}.@refill @end table @node Variables, Arrays, Expressions, Data @section 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 (@pxref{Selection}); 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 @example foo (a) int a; @{ bar (a); @{ int b = test (); bar (b); @} @} @end example @noindent the variable @code{a} is usable whenever the program is executing within the function @code{foo}, but the variable @code{b} is visible only while the program is executing inside the block in which @code{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: @cindex colon-colon @cindex scope @kindex :: @example @var{block}::@var{variable} @end example @noindent Here @var{block} is the name of the source file whose variable you want. @cindex name resolution (C++) Unfortunately, this use of @samp{::} conflicts with the very similar use of the same notation in C++; accordingly, GDB does not support use of the C++ name resolution operator in GDB expressions. @node Arrays, Format options, Variables, Data @section Artificial Arrays @cindex artificial array @kindex @@ 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 @dfn{artificial array} with the binary operator @samp{@@}. The left operand of @samp{@@} 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 @example int *array = (int *) malloc (len * sizeof (int)); @end example @noindent you can print the contents of @code{array} with @example p *array@@len @end example The left operand of @samp{@@} must reside in memory. Array values made with @samp{@@} 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.) @node Format options, Output formats, Arrays, Data @section Format options @cindex format options GDB provides a few ways to control how arrays, structures, and symbols are printed. @table @code @item set array-max @var{number-of-elements} @kindex set array-max If GDB is printing a large array, it will stop printing after it has printed the number of elements set by the @samp{set array-max} command. This limit also applies to the display of strings. @item show array-max @kindex show array-max Display the number of elements of a large array that GDB will print before losing patience. @item set arrayprint @itemx set arrayprint on @kindex set arrayprint GDB will pretty print arrays. This format is more convenient to read, but uses more space. The default is off. @item set arrayprint off. Return to compressed format for arrays. @item show arrayprint @kindex show arrayprint Show whether compressed or pretty format is selected for displaying arrays. @item set demangle @itemx set demangle on @kindex set demangle Print C++ names in their source form rather than in the mangled form in which they are passed to the assembler and linker for type-safe linkage. The default is on. @item show demangle @kindex show demangle Show whether C++ names will be printed in mangled or demangled form. @item set asm-demangle @itemx set asm-demangle on @kindex set asm-demangle Print C++ names in their source form rather than their mangled form, even in assembler code printouts such as instruction disassemblies. The default is off. @item show asm-demangle @kindex show asm-demangle Show whether C++ names in assembly listings will be printed in mangled or demangled form. @item set vtblprint @itemx set vtblprint on @kindex set vtblprint Pretty print C++ virtual function tables. The default is off. @item set vtblprint off Do not pretty print C++ virtual function tables. @item show vtblprint @kindex show vtblprint Show whether C++ virtual function tables are pretty printed, or not. @item set addressprint @item set addressprint on @kindex set addressprint GDB will print memory addresses in stack traces, structure values, pointer values, breakpoints, etc. The default is on. @item set addressprint off Do not print addresses. @item show addressprint @kindex show addressprint Show whether or not addresses are to be printed. @item set prettyprint on @kindex set prettyprint Cause GDB to print structures in an indented format with one member per line, like this: @example $1 = @{ next = 0x0, flags = @{ sweet = 1, sour = 1 @}, meat = 0x54 "Pork" @} @end example @item set prettyprint off Cause GDB to print structures in a compact format, like this: @smallexample $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, meat \ = 0x54 "Pork"@} @end smallexample @noindent This is the default format. @item show prettyprint @kindex show prettyprint Show which format GDB will use to print structures. @item set unionprint on @kindex set unionprint Tell GDB to print unions which are contained in structures. This is the default setting. @item set unionprint off Tell GDB not to print unions which are contained in structures. @item show unionprint @kindex show unionprint Ask GDB whether or not it will print unions which are contained in structures. For example, given the declarations @smallexample typedef enum @{Tree, Bug@} Species; typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms; typedef enum @{Caterpillar, Cocoon, Butterfly@} Bug_forms; struct thing @{ Species it; union @{ Tree_forms tree; Bug_forms bug; @} form; @}; struct thing foo = @{Tree, @{Acorn@}@}; @end smallexample @noindent with @samp{set unionprint on} in effect @samp{p foo} would print @smallexample $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@} @end smallexample @noindent and with @samp{set unionprint off} in effect it would print @example $1 = @{it = Tree, form = @{...@}@} @end example @end table @node Output formats, Auto Display, Format options, Data @section Output formats @cindex formatted output @cindex output 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 @dfn{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 @samp{print} command with a slash and a format letter. The format letters supported are: @table @samp @item x Regard the bits of the value as an integer, and print the integer in hexadecimal. @item d Print as integer in signed decimal. @item u Print as integer in unsigned decimal. @item o Print as integer in octal. @item a Print as an address, both absolute in hex and as an offset from the nearest preceding symbol. This format can be used to discover where (in what function) an unknown address is located: @example (gdb) p/a 0x54320 $3 = 0x54320 <_initialize_vx+396> @end example @item c Regard as an integer and print it as a character constant. @item f Regard the bits of the value as a floating point number and print using typical floating point syntax. @end table For example, to print the program counter in hex (@pxref{Registers}), type @example p/x $pc @end example @noindent 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 @samp{print} command with just a format and no expression. For example, @samp{p/x} reprints the last value in hex. @menu * Memory:: Examining Memory @end menu @node Memory, , Output formats, Output formats @subsection Examining Memory @cindex examining memory @table @code @kindex disassemble @item disassemble This specialized command is provided to dump a range of memory as machine instructions. The default memory range is the function surrounding the program counter of the selected frame. A single argument to this command is a program counter value; the function surrounding this value will be dumped. Two arguments (separated by one or more spaces) specify a range of addresses (first inclusive, second exclusive) to be dumped. @kindex x @item x The command @samp{x} (for `examine') can be used to examine memory without reference to the program's data types. The format in which you wish to examine memory is instead explicitly specified. The allowable formats are a superset of the formats described in the previous section. @end table @samp{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. @xref{Expressions} for more information on expressions. For example, @samp{x/4xw $sp} prints the four words of memory above the stack pointer in hexadecimal. 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: @table @samp @item b Examine individual bytes. @item h Examine halfwords (two bytes each). @item w Examine words (four bytes each). @cindex word 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 GDB runs on. @item g Examine giant words (8 bytes). @end table These letters specify just the way to print the contents: @table @samp @item x Print as integers in unsigned hexadecimal. @item d Print as integers in signed decimal. @item u Print as integers in unsigned decimal. @item o Print as integers in unsigned octal. @item a Print as an address, both absolute in hex and then relative to a symbol defined as an address below it. @item c Print as character constants. @item f Print as floating point. This works only with sizes @samp{w} and @samp{g}. @item 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). @item 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. The command @samp{disassemble} gives an alternative way of inspecting machine instructions. @end table 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. When the @samp{print} command shows a value that resides in memory, @samp{print} also sets the default address for the @samp{x} command. @samp{info line} also sets the default for @samp{x}, to the address of the start of the machine code for the specified line and @samp{info breakpoints} sets it to the address of the last breakpoint listed. When you use @key{RET} to repeat an @samp{x} command, 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 @samp{x} command that many times except that the output may be more compact with several units per line. For example, @example x/10i $pc @end example @noindent prints ten instructions starting with the one to be executed next in the selected frame. After doing this, you could print another seven following instructions with @example x/7 @end example @noindent in which the format and address are allowed to default. @kindex $_ @kindex $__ The addresses and contents printed by the @samp{x} command are not put in the value history because there is often too much of them and they would get in the way. Instead, GDB makes these values available for subsequent use in expressions as values of the convenience variables @code{$_} and @code{$__}. After an @samp{x} command, the last address examined is available for use in expressions in the convenience variable @code{$_}. The contents of that address, as examined, are available in the convenience variable @code{$__}. If the @samp{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. @node Auto Display, Value History, Output formats, Data @section Automatic Display @cindex automatic display @cindex display of expressions 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 @dfn{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: @example 2: foo = 38 3: bar[5] = (struct hack *) 0x3804 @end example @noindent showing item numbers, expressions and their current values. If the expression refers to local variables, then it does not make sense outside the lexical context for which it was set up. Such an expression is disabled when execution enters a context where one of its variables is not defined. For example, if you give the command @samp{display name} while inside a function with an argument @code{name}, then this argument will be displayed while the program continues to stop inside that function. When it stops elsewhere---where there is no variable @samp{name}---display is disabled. The next time your program stops where @samp{name} is meaningful, you can enable the display expression once again. @table @code @item display @var{exp} @kindex display Add the expression @var{exp} to the list of expressions to display each time the program stops. @xref{Expressions}. @item display/@var{fmt} @var{exp} For @var{fmt} specifying only a display format and not a size or count, add the expression @var{exp} to the auto-display list but arranges to display it each time in the specified format @var{fmt}. @item display/@var{fmt} @var{addr} For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a number of units, add the expression @var{addr} as a memory address to be examined each time the program stops. Examining means in effect doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory}. @item undisplay @var{dnums}@dots{} @itemx delete display @var{dnums}@dots{} @kindex delete display @kindex undisplay Remove item numbers @var{dnums} from the list of expressions to display. @item disable display @var{dnums}@dots{} @kindex disable display Disable the display of item numbers @var{dnums}. A disabled display item is not printed automatically, but is not forgotten. It may be enabled again later. @item enable display @var{dnums}@dots{} @kindex enable display Enable display of item numbers @var{dnums}. It becomes effective once again in auto display of its expression, until you specify otherwise. @item display Display the current values of the expressions on the list, just as is done when the program stops. @item info display @kindex info display Print the list of expressions previously set up to display automatically, each one with its item number, but without showing the values. This includes disabled expressions, which are marked as such. It also includes expressions which would not be displayed right now because they refer to automatic variables not currently available. @end table @node Value History, Convenience Vars, Auto Display, Data @section Value History @cindex value history Values printed by the @samp{print} command are saved in GDB's @dfn{value history} so that you can refer to them in other expressions. Values are kept until the symbol table is re-read or discarded (for example with the @samp{file} or @samp{symbol-file} commands). When the symbol table changes, the value history is discarded, since the values may contain pointers back to the types defined in the symbol table. @cindex @code{$} @cindex @code{$$} @cindex history number The values printed are given @dfn{history numbers} for you to refer to them by. These are successive integers starting with 1. @samp{print} shows you the history number assigned to a value by printing @samp{$@var{num} = } before the value; here @var{num} is the history number. To refer to any previous value, use @samp{$} followed by the value's history number. The output printed by @samp{print} is designed to remind you of this. Just @code{$} refers to the most recent value in the history, and @code{$$} refers to the value before that. @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2} is the value just prior to @code{$$}, @code{$$1} is equivalent to @code{$$}, and @code{$$0} is equivalent to @code{$}. For example, suppose you have just printed a pointer to a structure and want to see the contents of the structure. It suffices to type @example p *$ @end example If you have a chain of structures where the component @samp{next} points to the next one, you can print the contents of the next one with this: @example p *$.next @end example @noindent It might be useful to repeat this command many times by typing @key{RET}. Note that the history records values, not expressions. If the value of @code{x} is 4 and you type this command: @example print x set x=5 @end example @noindent then the value recorded in the value history by the @samp{print} command remains 4 even though the value of @code{x} has changed. @table @code @kindex info values @item info values @itemx info history @kindex info history These two commands are synonymous. Either form will print the last ten values in the value history, with their item numbers. This is like @samp{p@ $$9} repeated ten times, except that @samp{info values} does not change the history. @item info values @var{n} Print ten history values centered on history item number @var{n}. @item info values + Print ten history values just after the values last printed. @end table @node Convenience Vars, Registers, Value History, Data @section Convenience Variables @cindex convenience variables GDB provides @dfn{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 @samp{$}. Any name starting with @samp{$} can be used for a convenience variable, unless it is one of the predefined set of register names (@pxref{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: @example set $foo = *object_ptr @end example @noindent would save in @code{$foo} the value contained in the object pointed to by @code{object_ptr}. Using a convenience variable for the first time creates it; but its value is @code{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, including structures and arrays, even if that variable already has a value of a different type. The convenience variable as an expression has whatever type its current value has. @table @code @item info convenience @kindex info convenience Print a list of convenience variables used so far, and their values. Abbreviated @samp{i con}. @end table One of the ways to use a convenience variable is as a counter to be incremented or a pointer to be advanced. For example: @example set $i = 0 print bar[$i++]->contents @i{@dots{}repeat that command by typing @key{RET}.} @end example Some convenience variables are created automatically by GDB and given values likely to be useful. @table @code @item $_ The variable @code{$_} is automatically set by the @samp{x} command to the last address examined (@pxref{Memory}). Other commands which provide a default address for @samp{x} to examine also set @code{$_} to that address; these commands include @samp{info line} and @samp{info breakpoint}. @item $__ The variable @code{$__} is automatically set by the @samp{x} command to the value found in the last address examined. @end table @node Registers, , Convenience Vars, Data @section Registers @cindex registers Machine register contents can be referred to in expressions as variables with names starting with @samp{$}. The names of registers are different for each machine; use @samp{info registers} to see the names used on your machine. The names @code{$pc} and @code{$sp} are used on most machines for the program counter register and the stack pointer. Often @code{$fp} is used for a register that contains a pointer to the current stack frame, and @code{$ps} is used for a register that contains the processor status. These standard register names may be available on your machine even though the @code{info registers} command displays them with a different name. For example, on the SPARC, @code{info registers} displays the processor status register as @code{$psr} but you can also refer to it as @code{$ps}. 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 @emph{print} it as a floating point value with @samp{print/f $@var{regname}}). Some registers have distinct ``raw'' and ``virtual'' data formats. This means that the data format in which the register contents are saved by the operating system is not the same one that your program normally sees. For example, the registers of the 68881 floating point coprocessor are always saved in ``extended'' (raw) format, but all C programs expect to work with ``double'' (virtual) format. In such cases, GDB normally works with the virtual format only (the format that makes sense for your program), but the @samp{info registers} command prints the data in both formats. Register values are relative to the selected stack frame (@pxref{Selection}). This means that you get the value that the register would contain if all stack frames farther in were exited and their saved registers restored. In order to see the real contents of all registers, you must select the innermost frame (with @samp{frame 0}). Some registers are never saved (typically those numbered zero or one) because they are used for returning function values. In some operating systems (those using the ``caller saves'' convention), there are other registers intended for free alteration by a called routine. For these registers, relativization makes no difference. @table @code @item info registers @kindex info registers Print the names and relativized values of all registers. @item info registers @var{regname} Print the relativized value of register @var{regname}. @var{regname} may be any register name valid on the machine you are using, with or without the initial @samp{$}. @end table @subsection Examples You could print the program counter in hex with @example p/x $pc @end example @noindent or print the instruction to be executed next with @example x/i $pc @end example @noindent or add four to the stack pointer with @example set $sp += 4 @end example @noindent The last is a way of removing one word from the stack, on machines where stacks grow downward in memory (most machines, nowadays). This assumes that the innermost stack frame is selected. Setting @code{$sp} is not allowed when other stack frames are selected. (To pop entire frames off the stack, regardless of machine architecture, use @samp{return}; @pxref{Returning}.) @node Symbols, Altering, Data, Top @chapter Examining the Symbol Table The commands described in this section allow you to inquire 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 @samp{symbol-file} command; it is inherent in the text of your program and does not change as the program executes. @table @code @item info address @var{symbol} @kindex info address Describe where the data for @var{symbol} is stored. For a register variable, this says which register it is kept in. For a non-register local variable, this prints the stack-frame offset at which the variable is always stored. Note the contrast with @samp{print &@var{symbol}}, which does not work at all for a register variables, and for a stack local variable prints the exact address of the current instantiation of the variable. @item whatis @var{exp} @kindex whatis Print the data type of expression @var{exp}. @var{exp} is not actually evaluated, and any side-effecting operations (such as assignments or function calls) inside it do not take place. @xref{Expressions}. @item whatis Print the data type of @code{$}, the last value in the value history. @item ptype @var{typename} @kindex ptype Print a description of data type @var{typename}. @var{typename} may be the name of a type, or for C code it may have the form @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or @samp{enum @var{enum-tag}}.@refill @item ptype @var{exp} Print a description of the type of expression @var{exp}. This is like @samp{whatis} except it prints a detailed description, instead of just the name of the type. For example, if the type of a variable is @samp{struct complex @{double real; double imag;@}}, @samp{whatis} will print @samp{struct complex} and @samp{ptype} will print @samp{struct complex @{double real; double imag;@}} @item info sources @kindex info sources Print the names of all source files in the program for which there is debugging information. @item info functions @kindex info functions Print the names and data types of all defined functions. @item info functions @var{regexp} Print the names and data types of all defined functions whose names contain a match for regular expression @var{regexp}. Thus, @samp{info fun step} finds all functions whose names include @samp{step}; @samp{info fun ^step} finds those whose names start with @samp{step}. @item info variables @kindex info variables Print the names and data types of all variables that are declared outside of functions (i.e., except for local variables). @item info variables @var{regexp} Print the names and data types of all variables (except for local variables) whose names contain a match for regular expression @var{regexp}. @ignore This was never implemented. @item info methods @itemx info methods @var{regexp} @kindex info methods The @samp{info-methods} command permits the user to examine all defined methods within C++ program, or (with the @var{regexp} argument) a specific set of methods found in the various C++ classes. Many C++ classes provide a large number of methods. Thus, the output from the @samp{ptype} command can be overwhelming and hard to use. The @samp{info-methods} command filters the methods, printing only those which match the regular-expression @var{regexp}. @end ignore @item printsyms @var{filename} @kindex printsyms Write a complete dump of the debugger's symbol data into the file @var{filename}. @end table @node Altering, Sequences, Symbols, Top @chapter Altering Execution Once you think you have found an error in the program, you might want to find out for certain whether correcting the apparent error would lead to correct results in the rest of the run. You can find the answer by experiment, using the GDB features for altering execution of the program. For example, you can store new values into variables or memory locations, give the program a signal, restart it at a different address, or even return prematurely from a function to its caller. @menu * Assignment:: Altering variable values or memory contents. * Jumping:: Altering control flow. * Signaling:: Making signals happen in the program. * Returning:: Making a function return prematurely. * Calling:: Calling functions from your program @end menu @node Assignment, Jumping, Altering, Altering @section Assignment to Variables @cindex assignment @cindex setting variables To alter the value of a variable, evaluate an assignment expression. @xref{Expressions}. For example, @example print x=4 @end example @noindent would store the value 4 into the variable @code{x}, and then print the value of the assignment expression (which is 4). All the assignment operators of C are supported, including the increment operators @samp{++} and @samp{--}, and combining assignments such as @samp{+=} and @samp{<<=}. @kindex set @kindex set variable @cindex variables, setting If you are not interested in seeing the value of the assignment, use the @samp{set} command instead of the @samp{print} command. @samp{set} is really the same as @samp{print} except that the expression's value is not printed and is not put in the value history (@pxref{Value History}). The expression is evaluated only for side effects. Note that if the beginning of the argument string of the @samp{set} command appears identical to a @samp{set} subcommand, it may be necessary to use the @samp{set variable} command. This command is identical to @samp{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. @comment FIXME: how do structs align/pad in these conversions? @comment /pesch@cygnus.com 18dec1990 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}} construct to generate a value of specified type at a specified address (@pxref{Expressions}). For example, @code{@{int@}0x83040} would refer to memory location 0x83040 as an integer (which implies a certain size and representation in memory), and @example set @{int@}0x83040 = 4 @end example would store the value 4 into that memory location. @node Jumping, Signaling, Assignment, Altering @section Continuing at a Different Address Ordinarily, when you continue the program, you do so at the place where it stopped, with the @samp{cont} command. You can instead continue at an address of your own choosing, with the following commands: @table @code @item jump @var{linenum} @kindex jump Resume execution at line number @var{linenum}. Execution may stop immediately if there is a breakpoint there. The @samp{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 @var{linenum} is in a different function from the one currently executing, the results may be bizarre if the two functions expect different patterns of arguments or of local variables. For this reason, the @samp{jump} command requests confirmation if the specified line is not in the function currently executing. However, even bizarre results are predictable based on careful study of the machine-language code of the program. @item jump *@var{address} Resume execution at the instruction at address @var{address}. @end table You can get much the same effect as the @code{jump} command by storing a new value into the register @code{$pc}. The difference is that this does not start the program running; it only changes the address where it @emph{will} run when it is continued. For example, @example set $pc = 0x485 @end example @noindent causes the next @samp{cont} command or stepping command to execute at address 0x485, rather than at the address where the program stopped. @xref{Stepping}. The most common occasion to use the @samp{jump} command is to back up, perhaps with more breakpoints set, over a portion of a program that has already executed. @node Signaling, Returning, Jumping, Altering @section Giving the Program a Signal @table @code @item signal @var{signalnum} @kindex signal Resume execution where the program stopped, but give it immediately the signal number @var{signalnum}. Alternatively, if @var{signalnum} is zero, continue execution without giving a signal. This is useful when the program stopped on account of a signal and would ordinary see the signal when resumed with the @samp{cont} command; @samp{signal 0} causes it to resume without a signal. @end table @node Returning, Calling, Signaling, Altering @section Returning from a Function @table @code @item return @cindex returning from a function @kindex return You can cancel execution of a function call with the @samp{return} command. @end table This command has the effect of discarding the selected stack frame (and all frames within it), so that control moves to the caller of that function. You can think of this as making the discarded frame return prematurely. First select the stack frame that you wish to return from (@pxref{Selection}). Then type the @samp{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 @samp{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 @samp{finish} command (@pxref{Stepping}), which resumes execution until the selected stack frame returns @emph{naturally}. @node Calling, , Returning, Altering @comment node-name, next, previous, up @section Calling your Program's Functions @cindex calling functions @kindex call @table @code @item call @var{expr} Evaluate the expression @var{expr} without displaying @code{void} returned values. @end table You can use this variant of the @samp{print} command if you want to execute some piece of your program, but without cluttering the output with @code{void} returned values. The result is printed and saved in the value history, if it is not void. @node Sequences, Emacs, Altering, Top @chapter Canned Sequences of Commands Aside from breakpoint commands (@pxref{Break Commands}),GDB provides two ways to store sequences of commands for execution as a unit: user-defined commands and command files. @menu * Define:: User-defined commands. * Command Files:: Command files. * Output:: Controlled output commands useful in user-defined commands and command files. @end menu @node Define, Command Files, Sequences, Sequences @section User-Defined Commands @cindex user-defined command A @dfn{user-defined command} is a sequence of GDB commands to which you assign a new name as a command. This is done with the @samp{define} command. @table @code @item define @var{commandname} @kindex define Define a command named @var{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 @samp{define} command. The end of these commands is marked by a line containing @samp{end}. @item document @var{commandname} @kindex document Give documentation to the user-defined command @var{commandname}. The command @var{commandname} must already be defined. This command reads lines of documentation just as @samp{define} reads the lines of the command definition, ending with @samp{end}. After the @samp{document} command is finished, @samp{help} on command @var{commandname} will print the documentation you have specified. You may use the @samp{document} command again to change the documentation of a command. Redefining the command with @samp{define} does not change the documentation. @end table 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. @node Command Files, Output, Define, Sequences @section Command Files @cindex command files A command file for GDB is a file of lines that are GDB commands. Comments (lines starting with @samp{#}) 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. @cindex init file @cindex @file{.gdbinit} When you start GDB, it first executes commands from its @dfn{init files}. These are files named @file{.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 @samp{-nx} option is given.) You can also request the execution of a command file with the @samp{source} command: @table @code @item source @var{filename} @kindex source Execute the command file @var{filename}. @end table 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. @node Output, , Command Files, Sequences @section 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. @table @code @item echo @var{text} @kindex echo @comment I don't consider backslash-space a standard C escape sequence @comment because it's not in ANSI. Print @var{text}. Nonprinting characters can be included in @var{text} using C escape sequences, such as @samp{\n} to print a newline. @b{No newline will be printed unless you specify one.} In addition to the standard C escape sequences a backslash followed by a space stands for a space. This is useful for outputting a string with spaces at the beginning or the end, since leading and trailing spaces are otherwise trimmed from all arguments. Thus, to print @samp{@ and foo =@ }, use the command @samp{echo \@ and foo = \@ }. @comment FIXME: verify hard copy actually issues enspaces for '@ '! Will this @comment confuse texinfo? A backslash at the end of @var{text} can be used, as in C, to continue the command onto subsequent lines. For example, @example echo This is some text\n\ which is continued\n\ onto several lines.\n @end example produces the same output as @example echo This is some text\n echo which is continued\n echo onto several lines.\n @end example @item output @var{expression} @kindex output Print the value of @var{expression} and nothing but that value: no newlines, no @samp{$@var{nn} = }. The value is not entered in the value history either. @xref{Expressions} for more information on expressions. @item output/@var{fmt} @var{expression} Print the value of @var{expression} in format @var{fmt}. @xref{Output formats}, for more information. @item printf @var{string}, @var{expressions}@dots{} @kindex printf Print the values of the @var{expressions} under the control of @var{string}. The @var{expressions} are separated by commas and may be either numbers or pointers. Their values are printed as specified by @var{string}, exactly as if the program were to execute @example printf (@var{string}, @var{expressions}@dots{}); @end example For example, you can print two values in hex like this: @example printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo @end example The only backslash-escape sequences that you can use in the format string are the simple ones that consist of backslash followed by a letter. @end table @node Emacs, Remote, Sequences, Top @chapter Using GDB under GNU Emacs @cindex 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 @kbd{M-x gdb} in Emacs. Give the executable file you want to debug as an argument. This command starts GDB as a subprocess of Emacs, with input and output through a newly created Emacs buffer. Using GDB under Emacs is just like using GDB normally except for two things: @itemize @bullet @item 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' Shell mode are available for this purpose. @item 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 (@samp{=>}) at the left margin of the current line. Emacs uses a separate buffer for source display, and splits the window to show both your GDB session and the source. Explicit GDB @samp{list} or search commands still produce output as usual, but you probably will have no reason to use them. @end itemize @quotation @emph{Warning:} If the directory where your program resides is not your current directory, it can be easy to confuse Emacs about the location of the source files, in which case the auxiliary display buffer will not appear to show your source. GDB can find programs by searching your environment's @samp{PATH} variable, so the GDB I/O session will proceed normally; but Emacs doesn't get enough information back from GDB to locate the source files in this situation. To avoid this problem, either start GDB mode from the directory where your program resides, or specify a full path name when prompted for the @kbd{M-x gdb} argument. A similar confusion can result if you use the GDB @samp{file} command to switch to debugging a program in some other location, from an existing GDB I/O buffer in Emacs. @end quotation By default, @kbd{M-x gdb} calls the program called ``@code{gdb}''. If you need to call GDB by a different name (for example, if you keep several configurations around, with different names) you can set the Emacs variable @code{gdb-command-name}; for example, @example (setq gdb-command-name "mygdb") @end example @noindent (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or in your @samp{.emacs} file) will make Emacs call the program named ``@code{mygdb}'' instead. In the GDB I/O buffer, you can use these special Emacs commands in addition to the standard Shell mode commands: @table @kbd @item C-h m Describe the features of Emacs' GDB Mode. @item M-s Execute to another source line, like the GDB @samp{step} command; also update the display window to show the current file and location. @item M-n Execute to next source line in this function, skipping all function calls, like the GDB @samp{next} command. Then update the display window to show the current file and location. @item M-i Execute one instruction, like the GDB @samp{stepi} command; update display window accordingly. @item M-x gdb-nexti Execute to next instruction, using the GDB @samp{nexti} command; update display window accordingly. @item C-c C-f Execute until exit from the selected stack frame, like the GDB @samp{finish} command. @item M-c @comment C-c C-p in emacs 19 Continue execution of the program, like the GDB @samp{cont} command. @item M-u @comment C-c C-u in emacs 19 Go up the number of frames indicated by the numeric argument (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}), like the GDB @samp{up} command.@refill @item M-d @comment C-c C-d in emacs 19 Go down the number of frames indicated by the numeric argument, like the GDB @samp{down} command. @item C-x & Read the number where the cursor is positioned, and insert it at the end of the GDB I/O buffer. For example, if you wish to disassemble code around an address that was displayed earlier, type @kbd{disassemble}; then move the cursor to the address display, and pick up the argument for @samp{disassemble} by typing @kbd{C-x &}. You can customize this further on the fly by defining elements of the list @samp{gdb-print-command}; once it is defined, you can format or otherwise process numbers picked up by @kbd{C-x &} before they are inserted. A numeric argument to @kbd{C-x &} will both flag that you wish special formatting, and act as an index to pick an element of the list. If the list element is a string, the number to be inserted is formatted using the Emacs function @samp{format}; otherwise the number is passed as an argument to the corresponding list element. @item M-x gdb-display-frame Explicitly request display of the source code surrounding the current frame location, in another window. GDB does this display automatically; but if, for example, you accidentally kill the buffer where it is displayed, this command is a way of getting it back. @end table In any source file, the Emacs command @kbd{C-x SPC} (@code{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. @comment The following dropped because Epoch is nonstandard. Reactivate @comment if/when v19 does something similar. ---pesch@cygnus.com 19dec1990 @ignore @kindex emacs epoch environment @kindex epoch @kindex inspect Version 18 of Emacs has a built-in window system called the @samp{epoch} environment. Users of this environment can use a new command, @samp{inspect} which performs identically to @samp{print} except that each value is printed in its own window. @end ignore @node Remote, GDB Bugs, Emacs, Top @chapter Remote Debugging @cindex remote debugging If you are trying to debug a program running on a machine that can't run GDB in the usual way, it is often useful to use remote debugging. For example, you might be debugging an operating system kernel, or debugging a small system which does not have a general purpose operating system powerful enough to run a full-featured debugger. Currently GDB supports remote debugging over a serial connection, and (using Sun RPC) over a TCP/IP connection. The program to be debugged on the remote machine needs to contain a debugging device driver which talks to GDB over the serial line. The same version of GDB that is used ordinarily can be used for this. Several sample remote debugging drivers are distributed with GDB; see the @file{README} file in the GDB distribution for more information. @menu * Remote Commands:: Commands used to start and finish remote debugging. @end menu For details of the communication protocol, see the comments in the GDB source file @file{remote.c}. @node Remote Commands, , Remote, Remote @section 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 @samp{target remote} command with a device name as an argument. For example: @example target remote /dev/ttyb @end example @noindent if the serial line is connected to the device named @file{/dev/ttyb}. 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 @samp{detach} command. Other remote targets be available in your configuration of GDB; use @samp{info targets} to list them. @table @code @item reset @kindex reset For a target attached through a serial line, this command sends a ``break'' to the remote target system; this is only useful if the target has been equipped with a circuit to perform a hard reset (or some other interesting action) when a break is detected. @end table @node GDB Bugs, Installing GDB, Remote, Top @comment node-name, next, previous, up @chapter Reporting Bugs in GDB @cindex Bugs in GDB @cindex Reporting Bugs in GDB Your bug reports play an essential role in making GDB reliable. Reporting a bug may help you by bringing a solution to your problem, or it may not. But in any case the important function of a bug report is to help the entire community by making the next version of GDB work better. Bug reports are your contribution to the maintenance of GDB. In order for a bug report to serve its purpose, you must include the information that makes for fixing the bug. @menu * Bug Criteria:: Have You Found a Bug? * Bug Reporting:: How to Report Bugs @end menu @node Bug Criteria, Bug Reporting, GDB Bugs, GDB Bugs @section Have You Found a Bug? @cindex Bug Criteria If you are not sure whether you have found a bug, here are some guidelines: @itemize @bullet @item @cindex Fatal Signal @cindex Core Dump If the debugger gets a fatal signal, for any input whatever, that is a GDB bug. Reliable debuggers never crash. @item @cindex error on Valid Input If GDB produces an error message for valid input, that is a bug. @item @cindex Invalid Input If GDB does not produce an error message for invalid input, that is a bug. However, you should note that your idea of ``invalid input'' might be our idea of ``an extension'' or ``support for traditional practice''. @item If you are an experienced user of debugging tools, your suggestions for improvement of GDB are welcome in any case. @end itemize @node Bug Reporting, , Bug Criteria, GDB Bugs @section How to Report Bugs @cindex Bug Reports @cindex Compiler Bugs, Reporting @comment The following is meant to be neutral and helpful, not just a plug for @comment Cygnus; feedback on the issue (to "pesch@cygnus.com" or @comment "info@cygnus.com"---the latter will reach all of Cygnus) @comment is welcome. A number of companies and individuals offer support for GNU products. If you obtained GDB from a support organization, we recommend you contact that organization first. Among these organizations are Cygnus Support (Palo Alto CA, USA); C2V (Paris, France); Dynamix Corporation (King of Prussia PA, USA); The Nice Computer Company (Perth, Australia); Optimal Solutions (Seattle WA, USA); and The Pharos Group (Las Cruces NM, USA). Full contact information is in the file @samp{etc/SERVICE} in the GNU Emacs distribution. Numerous individual consultants are also listed there. @comment END NEUTRAL+HELPFUL section In any event, we also recommend that you send bug reports for GDB to one of these addresses: @example bug-gdb@@prep.ai.mit.edu @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb @end example @strong{Do not send bug reports to @samp{info-gdb}, or to @samp{help-gdb}, or to any newsgroups.} Most users of GDB do not want to receive bug reports. Those that do, have asked to be on @samp{bug-gdb}. The mailing list @samp{bug-gdb} has a newsgroup which serves as a repeater. The mailing list and the newsgroup carry exactly the same messages. Often people think of posting bug reports to the newsgroup instead of mailing them. This appears to work, but it has one problem which can be crucial: a newsgroup posting does not contain a mail path back to the sender. Thus, if we need to ask for more information, we may be unable to reach you. For this reason, it is better to send bug reports to the mailing list. As a last resort, send bug reports on paper to: @example GNU Debugger Bugs 545 Tech Square Cambridge, MA 02139 @end example The fundamental principle of reporting bugs usefully is this: @strong{report all the facts}. If you are not sure whether to state a fact or leave it out, state it! Often people omit facts because they think they know what causes the problem and they conclude that some details don't matter. Thus, you might assume that the name of the variable you use in an example does not matter. Well, probably it doesn't, but one cannot be sure. Perhaps the bug is a stray memory reference which happens to fetch from the location where that name is stored in memory; perhaps, if the name were different, the contents of that location would fool the debugger into doing the right thing despite the bug. Play it safe and give a specific, complete example. That is the easiest thing for you to do, and the most helpful. Keep in mind that the purpose of a bug report is to enable us to fix the bug if it is not known. It isn't very important what happens if the bug is already known. Therefore, always write your bug reports on the assumption that the bug is not known. Sometimes people give a few sketchy facts and ask, ``Does this ring a bell?'' Those bug reports are useless, and we urge everyone to @emph{refuse to respond to them} except to chide the sender to report bugs properly. To enable us to fix the bug, you should include all these things: @itemize @bullet @item The version of GDB. GDB announces it on startup; you can also print it at any time using @samp{info version}. Without this, we won't know whether there is any point in looking for the bug in the current version of GDB. @item A complete input script, and all necessary source files, that will reproduce the bug. @item What compiler (and its version) was used to compile GDB---e.g. ``GCC-1.37.1''. @item What compiler (and its version) was used to compile GDB. @item The command arguments you gave the compiler to compile that example and observe the bug. For example, did you use @samp{-O}? To guarantee you won't omit something important, list them all. If we were to try to guess the arguments, we would probably guess wrong and then we would not encounter the bug. @item The type of machine you are using, and the operating system name and version number. @item A description of what behavior you observe that you believe is incorrect. For example, ``It gets a fatal signal.'' Of course, if the bug is that GDB gets a fatal signal, then we will certainly notice it. But if the bug is incorrect output, we might not notice unless it is glaringly wrong. Even if the problem you experience is a fatal signal, you should still say so explicitly. Suppose something strange is going on, such as, your copy of GDB is out of synch, or you have encountered a bug in the C library on your system. (This has happened!) Your copy might crash and ours would not. If you @i{told} us to expect a crash, then when ours fails to crash, we would know that the bug was not happening for us. If you had not told us to expect a crash, then we would not be able to draw any conclusion from our observations. @item If you wish to suggest changes to the GDB source, send us context diffs. If you even discuss something in the GDB source, refer to it by context, not by line number. The line numbers in our development sources don't match those in your sources. Your line numbers would convey no useful information to us. @end itemize Here are some things that are not necessary: @itemize @bullet @item A description of the envelope of the bug. Often people who encounter a bug spend a lot of time investigating which changes to the input file will make the bug go away and which changes will not affect it. This is often time consuming and not very useful, because the way we will find the bug is by running a single example under the debugger with breakpoints, not by pure deduction from a series of examples. We recommend that you save your time for something else. Of course, if you can find a simpler example to report @emph{instead} of the original one, that is a convenience for us. Errors in the output will be easier to spot, running under the debugger will take less time, etc. However, simplification is not vital; if you don't want to do this, report the bug anyway and send us the entire test case you used. @item A patch for the bug. A patch for the bug does help us if it is a good one. But don't omit the necessary information, such as the test case, on the assumption that a patch is all we need. We might see problems with your patch and decide to fix the problem another way, or we might not understand it at all. Sometimes with a program as complicated as GDB it is very hard to construct an example that will make the program follow a certain path through the code. If you don't send us the example, we won't be able to construct one, so we won't be able to verify that the bug is fixed. And if we can't understand what bug you are trying to fix, or why your patch should be an improvement, we won't install it. A test case will help us to understand. @item A guess about what the bug is or what it depends on. Such guesses are usually wrong. Even we can't guess right about such things without first using the debugger to find the facts. @end itemize @iftex @include readline/inc-readline.texinfo @include readline/inc-history.texinfo @end iftex @node Installing GDB, License, GDB Bugs, Top @appendix Installing GDB @cindex configuring GDB @cindex installation The script @samp{config.gdb} automates the process of preparing GDB for installation; you can then use @samp{make} to actually build it. For example, @example config.gdb sun3os4 make @end example @noindent is all that's required to install GDB on a Sun 3 running SunOS 4. @table @code @kindex config.gdb @item config.gdb @var{machine} @itemx config.gdb -srcdir=@var{dir} @var{machine} This is the most usual way of configuring GDB; to debug programs running on the same machine as GDB itself. If you wish to build the GDB binaries in a completely different directory from the sources, specify a path to the source directory using the @samp{-srcdir} option. @item config.gdb -host @cindex host environments Display a list of supported host environments for GDB. @item config.gdb @var{host} @var{target} @itemx config.gdb -srcdir=@var{dir} @var{host} @var{target} @cindex cross-debugging GDB can also be used as a cross-debugger, running on a machine of one type while debugging a program running on a machine of another type. You configure it this way by specifying first the @var{host}, then the @var{target} environment on the @code{config.gdb} argument list; the @var{host} is where GDB runs, and the @var{target} is where your program runs. @xref{Remote}. Again, you can use @samp{-srcdir} to specify a path to the GDB source. @item config.gdb -target @cindex target environments Display a list of supported target environments for GDB. @end table @node License, Commands, Installing GDB, Top @unnumbered GNU GENERAL PUBLIC LICENSE @center Version 1, February 1989 @display Copyright @copyright{} 1989 Free Software Foundation, Inc. 675 Mass Ave, Cambridge, MA 02139, USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. @end display @unnumberedsec Preamble The license agreements of most software companies try to keep users at the mercy of those companies. By contrast, our General Public License is intended to guarantee your freedom to share and change free software---to make sure the software is free for all its users. The General Public License applies to the Free Software Foundation's software and to any other program whose authors commit to using it. You can use it for your programs, too. When we speak of free software, we are referring to freedom, not price. Specifically, the General Public License is designed to make sure that you have the freedom to give away or sell copies of free software, that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of a such a program, whether gratis or for a fee, 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. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. The precise terms and conditions for copying, distribution and modification follow. @iftex @unnumberedsec TERMS AND CONDITIONS @end iftex @ifinfo @center TERMS AND CONDITIONS @end ifinfo @enumerate @item This License Agreement applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The ``Program'', below, refers to any such program or work, and a ``work based on the Program'' means either the Program or any work containing the Program or a portion of it, either verbatim or with modifications. Each licensee is addressed as ``you''. @item You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this General Public License and to the absence of any warranty; and give any other recipients of the Program a copy of this General Public License along with the Program. You may charge a fee for the physical act of transferring a copy. @item You may modify your copy or copies of the Program 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: @itemize @bullet @item cause the modified files to carry prominent notices stating that you changed the files and the date of any change; and @item cause the whole of any work that you distribute or publish, that in whole or in part contains the Program or any part thereof, either with or without modifications, to be licensed at no charge to all third parties under the terms of this General Public License (except that you may choose to grant warranty protection to some or all third parties, at your option). @item If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the simplest and most usual way, to print or display an announcement including an appropriate copyright notice and a notice 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 General Public License. @item You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. @end itemize Mere aggregation of another independent work with the Program (or its derivative) on a volume of a storage or distribution medium does not bring the other work under the scope of these terms. @item You may copy and distribute the Program (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: @itemize @bullet @item accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Paragraphs 1 and 2 above; or, @item accompany it with a written offer, valid for at least three years, to give any third party free (except for a nominal charge for the cost of distribution) a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Paragraphs 1 and 2 above; or, @item 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.) @end itemize Source code for a work means the preferred form of the work for making modifications to it. 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, or for standard header files or definitions files that accompany that operating system. @item You may not copy, modify, sublicense, distribute or transfer the Program except as expressly provided under this General Public License. Any attempt otherwise to copy, modify, sublicense, distribute or transfer the Program is void, and will automatically terminate your rights to use the Program under this License. However, parties who have received copies, or rights to use copies, from you under this General Public License will not have their licenses terminated so long as such parties remain in full compliance. @item By copying, distributing or modifying the Program (or any work based on the Program) you indicate your acceptance of this license to do so, and all its terms and conditions. @item Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. @item The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of the license which applies to it and ``any later version'', you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of the license, you may choose any version ever published by the Free Software Foundation. @item If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. @iftex @heading NO WARRANTY @end iftex @ifinfo @center NO WARRANTY @end ifinfo @item BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM ``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 PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. @item IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. @end enumerate @iftex @heading END OF TERMS AND CONDITIONS @end iftex @ifinfo @center END OF TERMS AND CONDITIONS @end ifinfo @page @unnumberedsec How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to humanity, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the ``copyright'' line and a pointer to where the full notice is found. @smallexample @var{one line to give the program's name and a brief idea of what it does.} Copyright (C) 19@var{yy} @var{name of author} This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 1, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. @end smallexample Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: @smallexample Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author} Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. @end smallexample The hypothetical commands `show w' and `show c' should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than `show w' and `show c'; they could even be mouse-clicks or menu items---whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a ``copyright disclaimer'' for the program, if necessary. Here is a sample; alter the names: @smallexample Yoyodyne, Inc., hereby disclaims all copyright interest in the program `Gnomovision' (a program to direct compilers to make passes at assemblers) written by James Hacker. @var{signature of Ty Coon}, 1 April 1989 Ty Coon, President of Vice @end smallexample That's all there is to it! @node Commands, Concepts, License, Top @unnumbered Command Index @printindex ky @node Concepts, , Commands, Top @unnumbered Index @printindex cp @contents @bye