3 @settitle GDB, The GNU Debugger
6 This file documents the GNU debugger GDB.
8 Copyright (C) 1988, 1989 Free Software Foundation, Inc.
10 Permission is granted to make and distribute verbatim copies of
11 this manual provided the copyright notice and this permission notice
12 are preserved on all copies.
15 Permission is granted to process this file through Tex and print the
16 results, provided the printed document carries copying permission
17 notice identical to this one except for the removal of this paragraph
18 (this paragraph not being relevant to the printed manual).
21 Permission is granted to copy and distribute modified versions of this
22 manual under the conditions for verbatim copying, provided also that the
23 section entitled ``GNU General Public License'' is included exactly as
24 in the original, and provided that the entire resulting derived work is
25 distributed under the terms of a permission notice identical to this
28 Permission is granted to copy and distribute translations of this manual
29 into another language, under the above conditions for modified versions,
30 except that the section entitled ``GNU General Public License'' may be
31 included in a translation approved by the author instead of in the
35 @setchapternewpage odd
39 @center @titlefont{GDB Manual}
41 @center The GNU Source-Level Debugger
43 @center Third Edition, GDB version 3.4
47 @center Richard M. Stallman
49 @vskip 0pt plus 1filll
50 Copyright @copyright{} 1988, 1989 Free Software Foundation, Inc.
52 Permission is granted to make and distribute verbatim copies of
53 this manual provided the copyright notice and this permission notice
54 are preserved on all copies.
56 Permission is granted to copy and distribute modified versions of this
57 manual under the conditions for verbatim copying, provided also that the
58 section entitled ``GNU General Public License'' is included exactly as
59 in the original, and provided that the entire resulting derived work is
60 distributed under the terms of a permission notice identical to this
63 Permission is granted to copy and distribute translations of this manual
64 into another language, under the above conditions for modified versions,
65 except that the section entitled ``GNU General Public License'' may be
66 included in a translation approved by the author instead of in the
71 @node Top, Top, Top, (DIR)
72 @unnumbered Summary of GDB
74 The purpose of a debugger such as GDB is to allow you to execute another
75 program while examining what is going on inside it. We call the other
76 program ``your program'' or ``the program being debugged''.
78 GDB can do four kinds of things (plus other things in support of these):
82 Start the program, specifying anything that might affect its behavior.
85 Make the program stop on specified conditions.
88 Examine what has happened, when the program has stopped, so that you
92 Change things in the program, so you can correct the effects of one bug
93 and go on to learn about another without having to recompile first.
96 GDB can be used to debug programs written in C and C++. Pascal support
97 is being implemented, and Fortran support will be added when a GNU
98 Fortran compiler is written.
101 * License:: The GNU General Public License gives you permission
102 to redistribute GDB on certain terms; and also
103 explains that there is no warranty.
104 * User Interface:: GDB command syntax and input and output conventions.
105 * Files:: Specifying files for GDB to operate on.
106 * Options:: GDB arguments and options.
107 * Compilation::Compiling your program so you can debug it.
108 * Running:: Running your program under GDB.
109 * Stopping:: Making your program stop. Why it may stop. What to do then.
110 * Stack:: Examining your program's stack.
111 * Source:: Examining your program's source files.
112 * Data:: Examining data in your program.
113 * Symbols:: Examining the debugger's symbol table.
114 * Altering:: Altering things in your program.
115 * Sequences:: Canned command sequences for repeated use.
116 * Emacs:: Using GDB through GNU Emacs.
117 * Remote:: Remote kernel debugging across a serial line.
118 * Commands:: Index of GDB commands.
119 * Concepts:: Index of GDB concepts.
122 @node License, User Interface, Top, Top
123 @unnumbered GNU GENERAL PUBLIC LICENSE
124 @center Version 1, February 1989
127 Copyright @copyright{} 1989 Free Software Foundation, Inc.
128 675 Mass Ave, Cambridge, MA 02139, USA
130 Everyone is permitted to copy and distribute verbatim copies
131 of this license document, but changing it is not allowed.
134 @unnumberedsec Preamble
136 The license agreements of most software companies try to keep users
137 at the mercy of those companies. By contrast, our General Public
138 License is intended to guarantee your freedom to share and change free
139 software---to make sure the software is free for all its users. The
140 General Public License applies to the Free Software Foundation's
141 software and to any other program whose authors commit to using it.
142 You can use it for your programs, too.
144 When we speak of free software, we are referring to freedom, not
145 price. Specifically, the General Public License is designed to make
146 sure that you have the freedom to give away or sell copies of free
147 software, that you receive source code or can get it if you want it,
148 that you can change the software or use pieces of it in new free
149 programs; and that you know you can do these things.
151 To protect your rights, we need to make restrictions that forbid
152 anyone to deny you these rights or to ask you to surrender the rights.
153 These restrictions translate to certain responsibilities for you if you
154 distribute copies of the software, or if you modify it.
156 For example, if you distribute copies of a such a program, whether
157 gratis or for a fee, you must give the recipients all the rights that
158 you have. You must make sure that they, too, receive or can get the
159 source code. And you must tell them their rights.
161 We protect your rights with two steps: (1) copyright the software, and
162 (2) offer you this license which gives you legal permission to copy,
163 distribute and/or modify the software.
165 Also, for each author's protection and ours, we want to make certain
166 that everyone understands that there is no warranty for this free
167 software. If the software is modified by someone else and passed on, we
168 want its recipients to know that what they have is not the original, so
169 that any problems introduced by others will not reflect on the original
170 authors' reputations.
172 The precise terms and conditions for copying, distribution and
176 @unnumberedsec TERMS AND CONDITIONS
179 @center TERMS AND CONDITIONS
184 This License Agreement applies to any program or other work which
185 contains a notice placed by the copyright holder saying it may be
186 distributed under the terms of this General Public License. The
187 ``Program'', below, refers to any such program or work, and a ``work based
188 on the Program'' means either the Program or any work containing the
189 Program or a portion of it, either verbatim or with modifications. Each
190 licensee is addressed as ``you''.
193 You may copy and distribute verbatim copies of the Program's source
194 code as you receive it, in any medium, provided that you conspicuously and
195 appropriately publish on each copy an appropriate copyright notice and
196 disclaimer of warranty; keep intact all the notices that refer to this
197 General Public License and to the absence of any warranty; and give any
198 other recipients of the Program a copy of this General Public License
199 along with the Program. You may charge a fee for the physical act of
203 You may modify your copy or copies of the Program or any portion of
204 it, and copy and distribute such modifications under the terms of Paragraph
205 1 above, provided that you also do the following:
209 cause the modified files to carry prominent notices stating that
210 you changed the files and the date of any change; and
213 cause the whole of any work that you distribute or publish, that
214 in whole or in part contains the Program or any part thereof, either
215 with or without modifications, to be licensed at no charge to all
216 third parties under the terms of this General Public License (except
217 that you may choose to grant warranty protection to some or all
218 third parties, at your option).
221 If the modified program normally reads commands interactively when
222 run, you must cause it, when started running for such interactive use
223 in the simplest and most usual way, to print or display an
224 announcement including an appropriate copyright notice and a notice
225 that there is no warranty (or else, saying that you provide a
226 warranty) and that users may redistribute the program under these
227 conditions, and telling the user how to view a copy of this General
231 You may charge a fee for the physical act of transferring a
232 copy, and you may at your option offer warranty protection in
236 Mere aggregation of another independent work with the Program (or its
237 derivative) on a volume of a storage or distribution medium does not bring
238 the other work under the scope of these terms.
241 You may copy and distribute the Program (or a portion or derivative of
242 it, under Paragraph 2) in object code or executable form under the terms of
243 Paragraphs 1 and 2 above provided that you also do one of the following:
247 accompany it with the complete corresponding machine-readable
248 source code, which must be distributed under the terms of
249 Paragraphs 1 and 2 above; or,
252 accompany it with a written offer, valid for at least three
253 years, to give any third party free (except for a nominal charge
254 for the cost of distribution) a complete machine-readable copy of the
255 corresponding source code, to be distributed under the terms of
256 Paragraphs 1 and 2 above; or,
259 accompany it with the information you received as to where the
260 corresponding source code may be obtained. (This alternative is
261 allowed only for noncommercial distribution and only if you
262 received the program in object code or executable form alone.)
265 Source code for a work means the preferred form of the work for making
266 modifications to it. For an executable file, complete source code means
267 all the source code for all modules it contains; but, as a special
268 exception, it need not include source code for modules which are standard
269 libraries that accompany the operating system on which the executable
270 file runs, or for standard header files or definitions files that
271 accompany that operating system.
274 You may not copy, modify, sublicense, distribute or transfer the
275 Program except as expressly provided under this General Public License.
276 Any attempt otherwise to copy, modify, sublicense, distribute or transfer
277 the Program is void, and will automatically terminate your rights to use
278 the Program under this License. However, parties who have received
279 copies, or rights to use copies, from you under this General Public
280 License will not have their licenses terminated so long as such parties
281 remain in full compliance.
284 By copying, distributing or modifying the Program (or any work based
285 on the Program) you indicate your acceptance of this license to do so,
286 and all its terms and conditions.
289 Each time you redistribute the Program (or any work based on the
290 Program), the recipient automatically receives a license from the original
291 licensor to copy, distribute or modify the Program subject to these
292 terms and conditions. You may not impose any further restrictions on the
293 recipients' exercise of the rights granted herein.
296 The Free Software Foundation may publish revised and/or new versions
297 of the General Public License from time to time. Such new versions will
298 be similar in spirit to the present version, but may differ in detail to
299 address new problems or concerns.
301 Each version is given a distinguishing version number. If the Program
302 specifies a version number of the license which applies to it and ``any
303 later version'', you have the option of following the terms and conditions
304 either of that version or of any later version published by the Free
305 Software Foundation. If the Program does not specify a version number of
306 the license, you may choose any version ever published by the Free Software
310 If you wish to incorporate parts of the Program into other free
311 programs whose distribution conditions are different, write to the author
312 to ask for permission. For software which is copyrighted by the Free
313 Software Foundation, write to the Free Software Foundation; we sometimes
314 make exceptions for this. Our decision will be guided by the two goals
315 of preserving the free status of all derivatives of our free software and
316 of promoting the sharing and reuse of software generally.
326 BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
327 FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
328 OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
329 PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
330 OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
331 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
332 TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
333 PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
334 REPAIR OR CORRECTION.
337 IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL
338 ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
339 REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
340 INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES
341 ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT
342 LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES
343 SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE
344 WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN
345 ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
349 @heading END OF TERMS AND CONDITIONS
352 @center END OF TERMS AND CONDITIONS
356 @unnumberedsec Appendix: How to Apply These Terms to Your New Programs
358 If you develop a new program, and you want it to be of the greatest
359 possible use to humanity, the best way to achieve this is to make it
360 free software which everyone can redistribute and change under these
363 To do so, attach the following notices to the program. It is safest to
364 attach them to the start of each source file to most effectively convey
365 the exclusion of warranty; and each file should have at least the
366 ``copyright'' line and a pointer to where the full notice is found.
369 @var{one line to give the program's name and a brief idea of what it does.}
370 Copyright (C) 19@var{yy} @var{name of author}
372 This program is free software; you can redistribute it and/or modify
373 it under the terms of the GNU General Public License as published by
374 the Free Software Foundation; either version 1, or (at your option)
377 This program is distributed in the hope that it will be useful,
378 but WITHOUT ANY WARRANTY; without even the implied warranty of
379 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
380 GNU General Public License for more details.
382 You should have received a copy of the GNU General Public License
383 along with this program; if not, write to the Free Software
384 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
387 Also add information on how to contact you by electronic and paper mail.
389 If the program is interactive, make it output a short notice like this
390 when it starts in an interactive mode:
393 Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
394 Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
395 This is free software, and you are welcome to redistribute it
396 under certain conditions; type `show c' for details.
399 The hypothetical commands `show w' and `show c' should show the
400 appropriate parts of the General Public License. Of course, the
401 commands you use may be called something other than `show w' and `show
402 c'; they could even be mouse-clicks or menu items---whatever suits your
405 You should also get your employer (if you work as a programmer) or your
406 school, if any, to sign a ``copyright disclaimer'' for the program, if
407 necessary. Here a sample; alter the names:
410 Yoyodyne, Inc., hereby disclaims all copyright interest in the
411 program `Gnomovision' (a program to direct compilers to make passes
412 at assemblers) written by James Hacker.
414 @var{signature of Ty Coon}, 1 April 1989
415 Ty Coon, President of Vice
418 That's all there is to it!
420 @node User Interface, Files, License, Top
421 @chapter GDB Input and Output Conventions
423 GDB is invoked with the shell command @samp{gdb}. Once started, it reads
424 commands from the terminal until you tell it to exit.
426 A GDB command is a single line of input. There is no limit on how long
427 it can be. It starts with a command name, which is followed by arguments
428 whose meaning depends on the command name. For example, the command
429 @samp{step} accepts an argument which is the number of times to step,
430 as in @samp{step 5}. You can also use the @samp{step} command with
431 no arguments. Some command names do not allow any arguments.
434 GDB command names may always be abbreviated if the abbreviation is
435 unambiguous. Sometimes even ambiguous abbreviations are allowed; for
436 example, @samp{s} is specially defined as equivalent to @samp{step}
437 even though there are other commands whose names start with @samp{s}.
438 Possible command abbreviations are often stated in the documentation
439 of the individual commands.
441 @cindex repeating commands
442 A blank line as input to GDB means to repeat the previous command verbatim.
443 Certain commands do not allow themselves to be repeated this way; these are
444 commands for which unintentional repetition might cause trouble and which
445 you are unlikely to want to repeat. Certain others (@samp{list} and
446 @samp{x}) act differently when repeated because that is more useful.
448 A line of input starting with @samp{#} is a comment; it does nothing.
449 This is useful mainly in command files (@xref{Command Files}).
452 GDB indicates its readiness to read a command by printing a string
453 called the @dfn{prompt}. This string is normally @samp{(gdb)}. You can
454 change the prompt string with the @samp{set prompt} command. For
455 instance, when debugging GDB with GDB, it is useful to change the prompt
456 in one of the GDBs so that you tell which one you are talking to.
459 @item set prompt @var{newprompt}
461 Directs GDB to use @var{newprompt} as its prompt string henceforth.
466 To exit GDB, use the @samp{quit} command (abbreviated @samp{q}).
467 @kbd{Ctrl-c} will not exit from GDB, but rather will terminate the action
468 of any GDB command that is in progress and return to GDB command level.
469 It is safe to type @kbd{Ctrl-c} at any time because GDB does not allow
470 it to take effect until a time when it is safe.
473 @cindex pauses in output
474 Certain commands to GDB may produce large amounts of information output
475 to the screen. To help you read all of it, GDB pauses and asks you for
476 input at the end of each page of output. Type @key{RET} when you want
477 to continue the output. Normally GDB knows the size of the screen from
478 on the termcap data base together with the value of the @code{TERM}
479 environment variable; if this is not correct, you can override it with
480 the @samp{set screensize} command:
483 @item set screensize @var{lpp}
484 @itemx set screensize @var{lpp} @var{cpl}
485 @kindex set screensize
486 Specify a screen height of @var{lpp} lines and (optionally) a width of
487 @var{cpl} characters. If you omit @var{cpl}, the width does not change.
489 If you specify a height of zero lines, GDB will not pause during output
490 no matter how long the output is. This is useful if output is to a file
491 or to an editor buffer.
494 Also, GDB may at times produce more information about its own workings
495 than is of interest to the user. Some of these informational messages
496 can be turned on and off with the @samp{set verbose} command:
500 @item set verbose off
501 Disables GDB's output of certain informational messages.
504 Re-enables GDB's output of certain informational messages.
507 Currently, the messages controlled by @samp{set verbose} are those which
508 announce that the symbol table for a source file is being read
509 (@pxref{File Commands}, in the description of the command
511 @c The following is the right way to do it, but emacs 18.55 doesn't support
512 @c @ref, and neither the emacs lisp manual version of texinfmt or makeinfo
515 see @samp{symbol-file} in @ref{File Commands}).
518 @node Files, Compilation, User Interface, Top
519 @chapter Specifying GDB's Files
521 @cindex core dump file
522 @cindex executable file
524 GDB needs to know the file name of the program to be debugged, both in
525 order to read its symbol table and in order to start the program. To
526 debug a core dump of a previous run, GDB must be told the file name of
530 * Arguments: File Arguments. Specifying files with arguments
531 (when you start GDB).
532 * Commands: File Commands. Specifying files with GDB commands.
535 @node File Arguments, File Commands, Files, Files
536 @section Specifying Files with Arguments
538 The usual way to specify the executable and core dump file names is with
539 two command arguments given when you start GDB. The first argument is used
540 as the file for execution and symbols, and the second argument (if any) is
541 used as the core dump file name. Thus,
548 specifies @file{progm} as the executable program and @file{core} as a core
549 dump file to examine. (You do not need to have a core dump file if what
550 you plan to do is debug the program interactively.)
552 @xref{Options}, for full information on options and arguments for
555 @node File Commands,, File Arguments, Files
556 @section Specifying Files with Commands
558 Usually you specify the files for GDB to work with by giving arguments when
559 you invoke GDB. But occasionally it is necessary to change to a different
560 file during a GDB session. Or you may run GDB and forget to specify the
561 files you want to use. In these situations the GDB commands to specify new
565 @item exec-file @var{filename}
567 Specify that the program to be run is found in @var{filename}. If you
568 do not specify a directory and the file is not found in GDB's working
569 directory, GDB will use the environment variable @code{PATH} as a list
570 of directories to search, just as the shell does when looking for a
573 @item symbol-file @var{filename}
575 Read symbol table information from file @var{filename}. @code{PATH}
576 is searched when necessary. Most of the time you will use both the
577 @samp{exec-file} and @samp{symbol-file} commands on the same file.
579 @samp{symbol-file} with no argument clears out GDB's symbol table.
581 The @samp{symbol-file} command does not actually read the symbol table in
582 full right away. Instead, it scans the symbol table quickly to find
583 which source files and which symbols are present. The details are read
584 later, one source file at a time, when they are needed.
586 The purpose of this two-stage reading strategy is to make GDB start up
587 faster. For the most part, it is invisible except for occasional
588 messages telling you that the symbol table details for a particular
589 source file are being read. (The @samp{set verbose} command controls
590 whether these messages are printed; @pxref{User Interface}).
592 However, you will sometimes see in backtraces lines for functions in
593 source files whose data has not been read in; these lines omit some of
594 the information, such as argument values, which cannot be printed
595 without full details of the symbol table.
597 When the symbol table is stored in COFF format, @samp{symbol-file} does
598 read the symbol table data in full right away. We haven't bothered to
599 implement the two-stage strategy for COFF.
601 @item core-file @var{filename}
603 Specify the whereabouts of a core dump file to be used as the
604 ``contents of memory''. Note that the core dump contains only the
605 writable parts of memory; the read-only parts must come from the
608 @samp{core-file} with no argument specifies that no core file is
611 Note that the core file is ignored when your program is actually running
612 under GDB. So, if you have been running the program and you wish to
613 debug a core file instead, you must kill the subprocess in which the
614 program is running. To do this, use the @samp{kill} command
615 (@pxref{Kill Process}).
617 @item add-file @var{filename} @var{address}
619 @cindex dynamic linking
620 The @samp{add-file} command reads additional symbol table information
621 from the file @var{filename}. You would use this when that file has
622 been dynamically loaded into the program that is running. @var{address}
623 should be the memory address at which the file has been loaded; GDB cannot
624 figure this out for itself.
626 The symbol table of the file @var{filename} is added to the symbol table
627 originally read with the @samp{symbol-file} command. You can use the
628 @samp{add-file} command any number of times; the new symbol data thus
629 read keeps adding to the old. The @samp{symbol-file} command forgets
630 all the symbol data GDB has read; that is the only time symbol data is
635 Print the names of the executable and core dump files currently in
636 use by GDB, and the file from which symbols were loaded.
639 While all three file-specifying commands allow both absolute and relative
640 file names as arguments, GDB always converts the file name to an absolute
641 one and remembers it that way.
643 The @samp{symbol-file} command causes GDB to forget the contents of its
644 convenience variables, the value history, and all breakpoints and
645 auto-display expressions. This is because they may contain pointers to the
646 internal data recording symbols and data types, which are part of the old
647 symbol table data being discarded inside GDB.
649 @node Compilation, Running, Files, Top
650 @chapter Compiling Your Program for Debugging
652 In order to debug a program effectively, you need to ask for debugging
653 information when you compile it. This information in the object file
654 describes the data type of each variable or function and the correspondence
655 between source line numbers and addresses in the executable code.
657 To request debugging information, specify the @samp{-g} option when you run
660 The Unix C compiler is unable to handle the @samp{-g} and @samp{-O} options
661 together. This means that you cannot ask for optimization if you ask for
662 debugger information.
664 The GNU C compiler supports @samp{-g} with or without @samp{-O}, making it
665 possible to debug optimized code. We recommend that you @emph{always} use
666 @samp{-g} whenever you compile a program. You may think the program is
667 correct, but there's no sense in pushing your luck.
669 GDB no longer supports the debugging information produced by giving the
670 GNU C compiler the @samp{-gg} option, so do not use this option.
673 @comment As far as I know, there are no cases in which GDB will
674 @comment produce strange output in this case. (but no promises).
675 If your program includes archives made with the @code{ar} program, and
676 if the object files used as input to @code{ar} were compiled without the
677 @samp{-g} option and have names longer than 15 characters, GDB will get
678 confused reading the program's symbol table. No error message will be
679 given, but GDB may behave strangely. The reason for this problem is a
680 deficiency in the Unix archive file format, which cannot represent file
681 names longer than 15 characters.
683 To avoid this problem, compile the archive members with the @samp{-g}
684 option or use shorter file names. Alternatively, use a version of GNU
685 @code{ar} dated more recently than August 1989.
688 @node Running, Stopping, Compilation, Top
689 @chapter Running Your Program Under GDB
693 To start your program under GDB, use the @samp{run} command. The program
694 must already have been specified using the @samp{exec-file} command or with
695 an argument to GDB (@pxref{Files}); what @samp{run} does is create an
696 inferior process, load the program into it, and set it in motion.
698 The execution of a program is affected by certain information it
699 receives from its superior. GDB provides ways to specify this
700 information, which you must do @i{before} starting the program. (You
701 can change it after starting the program, but such changes do not affect
702 the program unless you start it over again.) This information may be
703 divided into three categories:
706 @item The @i{arguments.}
707 You specify the arguments to give the program as the arguments of the
710 @item The @i{environment.}
711 The program normally inherits its environment from GDB, but you can
712 use the GDB commands @samp{set environment} and
713 @samp{unset environment} to change parts of the environment that will
714 be given to the program.@refill
716 @item The @i{working directory.}
717 The program inherits its working directory from GDB. You can set GDB's
718 working directory with the @samp{cd} command in GDB.
721 After the @samp{run} command, the debugger does nothing but wait for your
722 program to stop. @xref{Stopping}.
724 Note that once your program has been started by the @samp{run} command,
725 you may evaluate expressions that involve calls to functions in the
726 inferior. @xref{Expressions}. If you wish to evaluate a function
727 simply for its side affects, you may use the @samp{set} command.
731 * Arguments:: Specifying the arguments for your program.
732 * Environment:: Specifying the environment for your program.
733 * Working Directory:: Specifying the working directory for giving
734 to your program when it is run.
735 * Input/Output:: Specifying the program's standard input and output.
736 * Attach:: Debugging a process started outside GDB.
737 * Kill Process:: Getting rid of the child process running your program.
740 @node Arguments, Environment, Running, Running
741 @section Your Program's Arguments
743 @cindex arguments (to your program)
744 The arguments to your program are specified by the arguments of the
745 @samp{run} command. They are passed to a shell, which expands wildcard
746 characters and performs redirection of I/O, and thence to the program.
748 @samp{run} with no arguments uses the same arguments used by the previous
752 The command @samp{set args} can be used to specify the arguments to be used
753 the next time the program is run. If @samp{set args} has no arguments, it
754 means to use no arguments the next time the program is run. If you have
755 run your program with arguments and want to run it again with no arguments,
756 this is the only way to do so.
758 @node Environment, Working Directory, Arguments, Running
759 @section Your Program's Environment
761 @cindex environment (of your program)
762 The @dfn{environment} consists of a set of @dfn{environment variables} and
763 their values. Environment variables conventionally record such things as
764 your user name, your home directory, your terminal type, and your search
765 path for programs to run. Usually you set up environment variables with
766 the shell and they are inherited by all the other programs you run. When
767 debugging, it can be useful to try running the program with different
768 environments without having to start the debugger over again.
771 @item info environment @var{varname}
772 @kindex info environment
773 Print the value of environment variable @var{varname} to be given to
774 your program when it is started. This command can be abbreviated
775 @samp{i env @var{varname}}.
777 @item info environment
778 Print the names and values of all environment variables to be given to
779 your program when it is started. This command can be abbreviated
782 @item set environment @var{varname} @var{value}
783 @itemx set environment @var{varname} = @var{value}
784 @kindex set environment
785 Sets environment variable @var{varname} to @var{value}, for your program
786 only, not for GDB itself. @var{value} may be any string; the values of
787 environment variables are just strings, and any interpretation is
788 supplied by your program itself. The @var{value} parameter is optional;
789 if it is eliminated, the variable is set to a null value. This command
790 can be abbreviated as short as @samp{set e}.
792 For example, this command:
799 tells the program, when subsequently run, to assume it is being run
800 on behalf of the user named @samp{foo}.
802 @item delete environment @var{varname}
803 @itemx unset environment @var{varname}
804 @kindex delete environment
805 @kindex unset environment
806 Remove variable @var{varname} from the environment to be passed to your
807 program. This is different from @samp{set env @var{varname}@ =} because
808 @samp{delete environment} leaves the variable with no value, which is
809 distinguishable from an empty value. This command can be abbreviated
813 @node Working Directory, Input/Output, Environment, Running
814 @section Your Program's Working Directory
816 @cindex working directory (of your program)
817 Each time you start your program with @samp{run}, it inherits its
818 working directory from the current working directory of GDB. GDB's
819 working directory is initially whatever it inherited from its parent
820 process (typically the shell), but you can specify a new working
821 directory in GDB with the @samp{cd} command.
823 The GDB working directory also serves as a default for the commands
824 that specify files for GDB to operate on. @xref{Files}.
827 @item cd @var{directory}
829 Set GDB's working directory to @var{directory}.
833 Print GDB's working directory.
836 @node Input/Output, Attach, Working Directory, Running
837 @section Your Program's Input and Output
840 @cindex controlling terminal
841 By default, the program you run under GDB does input and output to the same
842 terminal that GDB uses.
844 You can redirect the program's input and/or output using @samp{sh}-style
845 redirection commands in the @samp{run} command. For example,
852 starts the program, diverting its output to the file @file{outfile}.
855 Another way to specify where the program should do input and output is
856 with the @samp{tty} command. This command accepts a file name as
857 argument, and causes this file to be the default for future @samp{run}
858 commands. It also resets the controlling terminal for the child
859 process, for future @samp{run} commands. For example,
866 directs that processes started with subsequent @samp{run} commands
867 default to do input and output on the terminal @file{/dev/ttyb} and have
868 that as their controlling terminal.
870 An explicit redirection in @samp{run} overrides the @samp{tty} command's
871 effect on input/output redirection, but not its effect on the
872 controlling terminal.
874 When you use the @samp{tty} command or redirect input in the @samp{run}
875 command, only the @emph{input for your program} is affected. The input
876 for GDB still comes from your terminal.
878 @node Attach, Kill Process, Input/Output, Running
879 @section Debugging an Already-Running Process
884 Some operating systems allow GDB to debug an already-running process
885 that was started outside of GDB. To do this, you use the @samp{attach}
886 command instead of the @samp{run} command.
888 The @samp{attach} command requires one argument, which is the process-id
889 of the process you want to debug. (The usual way to find out the
890 process-id of the process is with the @code{ps} utility.)
892 The first thing GDB does after arranging to debug the process is to stop
893 it. You can examine and modify an attached process with all the GDB
894 commands that ordinarily available when you start processes with
895 @samp{run}. You can insert breakpoints; you can step and continue; you
896 can modify storage. If you would rather the process continue running,
897 you may use the @samp{continue} command after attaching GDB to the
900 When you have finished debugging the attached process, you can use the
901 @samp{detach} command to release it from GDB's control. Detaching
902 the process continues its execution. After the @samp{detach} command,
903 that process and GDB become completely independent once more, and you
904 are ready to @samp{attach} another process or start one with @samp{run}.
906 If you exit GDB or use the @samp{run} command while you have an attached
907 process, you kill that process. You will be asked for confirmation if you
908 try to do either of these things.
910 The @samp{attach} command is also used to debug a remote machine via a
911 serial connection. @xref{Attach}, for more info.
913 @node Kill Process,, Attach, Running
914 @section Killing the Child Process
919 Kill the child process in which the program being debugged is running
922 This command is useful if you wish to debug a core dump instead. GDB
923 ignores any core dump file if it is actually running the program, so the
924 @samp{kill} command is the only sure way to make sure the core dump file
927 It is also useful if you wish to run the program outside the debugger
928 for once and then go back to debugging it.
930 The @samp{kill} command is also useful if you wish to recompile and
931 relink the program, since on many systems it is impossible to modify an
932 executable file which is running in a process. But, in this case, it is
933 just as good to exit GDB, since you will need to read a new symbol table
934 after the program is recompiled if you wish to debug the new version,
935 and restarting GDB is the easiest way to do that.
938 @node Stopping, Stack, Running, Top
939 @chapter Stopping and Continuing
941 When you run a program normally, it runs until it terminates. The
942 principal purpose of using a debugger is so that you can stop it before
943 that point; or so that if the program runs into trouble you can
944 investigate and find out why.
947 * Signals:: Fatal signals in your program just stop it;
948 then you can use GDB to see what is going on.
949 * Breakpoints:: Breakpoints let you stop your program when it
950 reaches a specified point in the code.
951 * Continuing:: Resuming execution until the next signal or breakpoint.
952 * Stepping:: Stepping runs the program a short distance and
953 then stops it wherever it has come to.
956 @node Signals, Breakpoints, Stopping, Stopping
960 A signal is an asynchronous event that can happen in a program. The
961 operating system defines the possible kinds of signals, and gives each kind
962 a name and a number. For example, @code{SIGINT} is the signal a program
963 gets when you type @kbd{Ctrl-c}; @code{SIGSEGV} is the signal a program
964 gets from referencing a place in memory far away from all the areas in use;
965 @code{SIGALRM} occurs when the alarm clock timer goes off (which happens
966 only if the program has requested an alarm).
968 @cindex fatal signals
969 Some signals, including @code{SIGALRM}, are a normal part of the
970 functioning of the program. Others, such as @code{SIGSEGV}, indicate
971 errors; these signals are @dfn{fatal} (kill the program immediately) if the
972 program has not specified in advance some other way to handle the signal.
973 @code{SIGINT} does not indicate an error in the program, but it is normally
974 fatal so it can carry out the purpose of @kbd{Ctrl-c}: to kill the program.
976 GDB has the ability to detect any occurrence of a signal in the program
977 running under GDB's control. You can tell GDB in advance what to do for
980 @cindex handling signals
981 Normally, GDB is set up to ignore non-erroneous signals like @code{SIGALRM}
982 (so as not to interfere with their role in the functioning of the program)
983 but to stop the program immediately whenever an error signal happens.
984 You can change these settings with the @samp{handle} command. You must
985 specify which signal you are talking about with its number.
990 Print a table of all the kinds of signals and how GDB has been told to
991 handle each one. You can use this to see the signal numbers of all
992 the defined types of signals.
994 @item handle @var{signalnum} @var{keywords}@dots{}
996 Change the way GDB handles signal @var{signalnum}. The @var{keywords}
997 say what change to make.
1000 To use the @samp{handle} command you must know the code number of the
1001 signal you are concerned with. To find the code number, type @samp{info
1002 signal} which prints a table of signal names and numbers.
1004 The keywords allowed by the handle command can be abbreviated. Their full
1009 GDB should stop the program when this signal happens. This implies
1010 the @samp{print} keyword as well.
1013 GDB should print a message when this signal happens.
1016 GDB should not stop the program when this signal happens. It may
1017 still print a message telling you that the signal has come in.
1020 GDB should not mention the occurrence of the signal at all. This
1021 implies the @samp{nostop} keyword as well.
1024 GDB should allow the program to see this signal; the program will be
1025 able to handle the signal, or may be terminated if the signal is fatal
1029 GDB should not allow the program to see this signal.
1032 When a signal has been set to stop the program, the program cannot see the
1033 signal until you continue. It will see the signal then, if @samp{pass} is
1034 in effect for the signal in question @i{at that time}. In other words,
1035 after GDB reports a signal, you can use the @samp{handle} command with
1036 @samp{pass} or @samp{nopass} to control whether that signal will be seen by
1037 the program when you later continue it.
1039 You can also use the @samp{signal} command to prevent the program from
1040 seeing a signal, or cause it to see a signal it normally would not see,
1041 or to give it any signal at any time. @xref{Signaling}.
1043 @node Breakpoints, Continuing, Signals, Stopping
1044 @section Breakpoints
1047 A @dfn{breakpoint} makes your program stop whenever a certain point in the
1048 program is reached. You set breakpoints explicitly with GDB commands,
1049 specifying the place where the program should stop by line number, function
1050 name or exact address in the program. You can add various other conditions
1051 to control whether the program will stop.
1053 Each breakpoint is assigned a number when it is created; these numbers are
1054 successive integers starting with 1. In many of the commands for controlling
1055 various features of breakpoints you use the breakpoint number to say which
1056 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
1057 @dfn{disabled}; if disabled, it has no effect on the program until you
1062 The command @samp{info break} prints a list of all breakpoints set and not
1063 deleted, showing their numbers, where in the program they are, and any
1064 special features in use for them. Disabled breakpoints are included in the
1065 list, but marked as disabled. @samp{info break} with a breakpoint number
1066 as argument lists only that breakpoint. The convenience variable @code{$_}
1067 and the default examining-address for the @samp{x} command are set to the
1068 address of the last breakpoint listed (@pxref{Memory}).
1071 * Set Breaks:: How to establish breakpoints.
1072 * Delete Breaks:: How to remove breakpoints no longer needed.
1073 * Disabling:: How to disable breakpoints (turn them off temporarily).
1074 * Conditions:: Making extra conditions on whether to stop.
1075 * Break Commands:: Commands to be executed at a breakpoint.
1076 * Error in Breakpoints:: "Cannot insert breakpoints" error--why, what to do.
1079 @node Set Breaks, Delete Breaks, Breakpoints, Breakpoints
1080 @subsection Setting Breakpoints
1083 Breakpoints are set with the @samp{break} command (abbreviated @samp{b}).
1084 You have several ways to say where the breakpoint should go.
1087 @item break @var{function}
1088 Set a breakpoint at entry to function @var{function}.
1090 @item break @var{+offset}
1091 @itemx break @var{-offset}
1092 Set a breakpoint some number of lines forward or back from the position
1093 at which execution stopped in the currently selected frame.
1095 @item break @var{linenum}
1096 Set a breakpoint at line @var{linenum} in the current source file.
1097 That file is the last file whose source text was printed. This
1098 breakpoint will stop the program just before it executes any of the
1101 @item break @var{filename}:@var{linenum}
1102 Set a breakpoint at line @var{linenum} in source file @var{filename}.
1104 @item break @var{filename}:@var{function}
1105 Set a breakpoint at entry to function @var{function} found in file
1106 @var{filename}. Specifying a file name as well as a function name is
1107 superfluous except when multiple files contain similarly named
1110 @item break *@var{address}
1111 Set a breakpoint at address @var{address}. You can use this to set
1112 breakpoints in parts of the program which do not have debugging
1113 information or source files.
1116 Set a breakpoint at the next instruction to be executed in the selected
1117 stack frame (@pxref{Stack}). In any selected frame but the innermost,
1118 this will cause the program to stop as soon as control returns to that
1119 frame. This is equivalent to a @samp{finish} command in the frame
1120 inside the selected frame. If this is done in the innermost frame, GDB
1121 will stop the next time it reaches the current location; this may be
1122 useful inside of loops.
1124 GDB normally ignores breakpoints when it resumes execution, until at
1125 least one instruction has been executed. If it did not do this, you
1126 would be unable to proceed past a breakpoint without first disabling the
1127 breakpoint. This rule applies whether or not the breakpoint already
1128 existed when the program stopped.
1130 @item break @dots{} if @var{cond}
1131 Set a breakpoint with condition @var{cond}; evaluate the expression
1132 @var{cond} each time the breakpoint is reached, and stop only if the
1133 value is nonzero. @samp{@dots{}} stands for one of the possible
1134 arguments described above (or no argument) specifying where to break.
1135 @xref{Conditions}, for more information on breakpoint conditions.
1137 @item tbreak @var{args}
1139 Set a breakpoint enabled only for one stop. @var{args} are the
1140 same as in the @samp{break} command, and the breakpoint is set in the same
1141 way, but the breakpoint is automatically disabled the first time it
1142 is hit. @xref{Disabling}.
1145 GDB allows you to set any number of breakpoints at the same place in the
1146 program. There is nothing silly or meaningless about this. When the
1147 breakpoints are conditional, this is even useful (@pxref{Conditions}).
1149 @node Delete Breaks, Disabling, Set Breaks, Breakpoints
1150 @subsection Deleting Breakpoints
1152 @cindex clearing breakpoint
1153 @cindex deleting breakpoints
1154 It is often necessary to eliminate a breakpoint once it has done its job
1155 and you no longer want the program to stop there. This is called
1156 @dfn{deleting} the breakpoint. A breakpoint that has been deleted no
1157 longer exists in any sense; it is forgotten.
1159 With the @samp{clear} command you can delete breakpoints according to where
1160 they are in the program. With the @samp{delete} command you can delete
1161 individual breakpoints by specifying their breakpoint numbers.
1163 @b{It is not necessary to delete a breakpoint to proceed past it.} GDB
1164 automatically ignores breakpoints in the first instruction to be executed
1165 when you continue execution without changing the execution address.
1170 Delete any breakpoints at the next instruction to be executed in the
1171 selected stack frame (@pxref{Selection}). When the innermost frame
1172 is selected, this is a good way to delete a breakpoint that the program
1175 @item clear @var{function}
1176 @itemx clear @var{filename}:@var{function}
1177 Delete any breakpoints set at entry to the function @var{function}.
1179 @item clear @var{linenum}
1180 @itemx clear @var{filename}:@var{linenum}
1181 Delete any breakpoints set at or within the code of the specified line.
1183 @item delete @var{bnums}@dots{}
1185 Delete the breakpoints of the numbers specified as arguments.
1188 @node Disabling, Conditions, Delete Breaks, Breakpoints
1189 @subsection Disabling Breakpoints
1191 @cindex disabled breakpoints
1192 @cindex enabled breakpoints
1193 Rather than deleting a breakpoint, you might prefer to @dfn{disable} it.
1194 This makes the breakpoint inoperative as if it had been deleted, but
1195 remembers the information on the breakpoint so that you can @dfn{enable}
1198 You disable and enable breakpoints with the @samp{enable} and
1199 @samp{disable} commands, specifying one or more breakpoint numbers as
1200 arguments. Use @samp{info break} to print a list of breakpoints if you
1201 don't know which breakpoint numbers to use.
1203 A breakpoint can have any of four different states of enablement:
1207 Enabled. The breakpoint will stop the program. A breakpoint made
1208 with the @samp{break} command starts out in this state.
1210 Disabled. The breakpoint has no effect on the program.
1212 Enabled once. The breakpoint will stop the program, but
1213 when it does so it will become disabled. A breakpoint made
1214 with the @samp{tbreak} command starts out in this state.
1216 Enabled for deletion. The breakpoint will stop the program, but
1217 immediately after it does so it will be deleted permanently.
1220 You change the state of enablement of a breakpoint with the following
1224 @item disable breakpoints @var{bnums}@dots{}
1225 @itemx disable @var{bnums}@dots{}
1226 @kindex disable breakpoints
1228 Disable the specified breakpoints. A disabled breakpoint has no
1229 effect but is not forgotten. All options such as ignore-counts,
1230 conditions and commands are remembered in case the breakpoint is
1231 enabled again later.
1233 @item enable breakpoints @var{bnums}@dots{}
1234 @itemx enable @var{bnums}@dots{}
1235 @kindex enable breakpoints
1237 Enable the specified breakpoints. They become effective once again in
1238 stopping the program, until you specify otherwise.
1240 @item enable breakpoints once @var{bnums}@dots{}
1241 @itemx enable once @var{bnums}@dots{}
1242 Enable the specified breakpoints temporarily. Each will be disabled
1243 again the next time it stops the program (unless you have used one of
1244 these commands to specify a different state before that time comes).
1246 @item enable breakpoints delete @var{bnums}@dots{}
1247 @itemx enable delete @var{bnums}@dots{}
1248 Enable the specified breakpoints to work once and then die. Each of
1249 the breakpoints will be deleted the next time it stops the program
1250 (unless you have used one of these commands to specify a different
1251 state before that time comes).
1254 Aside from the automatic disablement or deletion of a breakpoint when it
1255 stops the program, which happens only in certain states, the state of
1256 enablement of a breakpoint changes only when one of the commands above
1259 @node Conditions, Break Commands, Disabling, Breakpoints
1260 @subsection Break Conditions
1261 @cindex conditional breakpoints
1262 @cindex breakpoint conditions
1264 The simplest sort of breakpoint breaks every time the program reaches a
1265 specified place. You can also specify a @dfn{condition} for a
1266 breakpoint. A condition is just a boolean expression in your
1267 programming language. (@xref{Expressions}). A breakpoint with a
1268 condition evaluates the expression each time the program reaches it, and
1269 the program stops only if the condition is true.
1271 Break conditions may have side effects, and may even call functions in your
1272 program. These may sound like strange things to do, but their effects are
1273 completely predictable unless there is another enabled breakpoint at the
1274 same address. (In that case, GDB might see the other breakpoint first and
1275 stop the program without checking the condition of this one.) Note that
1276 breakpoint commands are usually more convenient and flexible for the
1277 purpose of performing side effects when a breakpoint is reached
1278 (@pxref{Break Commands}).
1280 Break conditions can be specified when a breakpoint is set, by using
1281 @samp{if} in the arguments to the @samp{break} command. @xref{Set Breaks}.
1282 They can also be changed at any time with the @samp{condition} command:
1285 @item condition @var{bnum} @var{expression}
1287 Specify @var{expression} as the break condition for breakpoint number
1288 @var{bnum}. From now on, this breakpoint will stop the program only if
1289 the value of @var{expression} is true (nonzero, in C). @var{expression}
1290 is not evaluated at the time the @samp{condition} command is given.
1293 @item condition @var{bnum}
1294 Remove the condition from breakpoint number @var{bnum}. It becomes
1295 an ordinary unconditional breakpoint.
1298 @cindex ignore count (of breakpoint)
1299 A special case of a breakpoint condition is to stop only when the
1300 breakpoint has been reached a certain number of times. This is so
1301 useful that there is a special way to do it, using the @dfn{ignore
1302 count} of the breakpoint. Every breakpoint has an ignore count, which
1303 is an integer. Most of the time, the ignore count is zero, and
1304 therefore has no effect. But if the program reaches a breakpoint whose
1305 ignore count is positive, then instead of stopping, it just decrements
1306 the ignore count by one and continues. As a result, if the ignore count
1307 value is @var{n}, the breakpoint will not stop the next @var{n} times it
1311 @item ignore @var{bnum} @var{count}
1313 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
1314 The next @var{count} times the breakpoint is reached, it will not stop.
1316 To make the breakpoint stop the next time it is reached, specify
1319 @item cont @var{count}
1320 Continue execution of the program, setting the ignore count of the
1321 breakpoint that the program stopped at to @var{count} minus one.
1322 Thus, the program will not stop at this breakpoint until the
1323 @var{count}'th time it is reached.
1325 This command is allowed only when the program stopped due to a
1326 breakpoint. At other times, the argument to @samp{cont} is ignored.
1329 If a breakpoint has a positive ignore count and a condition, the condition
1330 is not checked. Once the ignore count reaches zero, the condition will
1331 start to be checked.
1333 Note that you could achieve the effect of the ignore count with a
1334 condition such as @w{@samp{$foo-- <= 0}} using a debugger convenience
1335 variable that is decremented each time. @xref{Convenience Vars}.
1337 @node Break Commands, Error in Breakpoints, Conditions, Breakpoints
1338 @subsection Commands Executed on Breaking
1340 @cindex breakpoint commands
1341 You can give any breakpoint a series of commands to execute when the
1342 program stops due to that breakpoint. For example, you might want to
1343 print the values of certain expressions, or enable other breakpoints.
1346 @item commands @var{bnum}
1347 Specify commands for breakpoint number @var{bnum}. The commands
1348 themselves appear on the following lines. Type a line containing just
1349 @samp{end} to terminate the commands.
1351 To remove all commands from a breakpoint, use the command
1352 @samp{commands} and follow it immediately by @samp{end}; that is, give
1355 With no arguments, @samp{commands} refers to the last breakpoint set.
1358 It is possible for breakpoint commands to start the program up again.
1359 Simply use the @samp{cont} command, or @samp{step}, or any other command
1360 to resume execution. However, any remaining breakpoint commands are
1361 ignored. When the program stops again, GDB will act according to the
1365 If the first command specified is @samp{silent}, the usual message about
1366 stopping at a breakpoint is not printed. This may be desirable for
1367 breakpoints that are to print a specific message and then continue.
1368 If the remaining commands too print nothing, you will see no sign that
1369 the breakpoint was reached at all. @samp{silent} is not really a command;
1370 it is meaningful only at the beginning of the commands for a breakpoint.
1372 The commands @samp{echo} and @samp{output} that allow you to print precisely
1373 controlled output are often useful in silent breakpoints. @xref{Output}.
1375 For example, here is how you could use breakpoint commands to print the
1376 value of @code{x} at entry to @code{foo} whenever it is positive.
1389 One application for breakpoint commands is to correct one bug so you can
1390 test another. Put a breakpoint just after the erroneous line of code, give
1391 it a condition to detect the case in which something erroneous has been
1392 done, and give it commands to assign correct values to any variables that
1393 need them. End with the @samp{cont} command so that the program does not
1394 stop, and start with the @samp{silent} command so that no output is
1395 produced. Here is an example:
1406 One deficiency in the operation of automatically continuing breakpoints
1407 under Unix appears when your program uses raw mode for the terminal.
1408 GDB switches back to its own terminal modes (not raw) before executing
1409 commands, and then must switch back to raw mode when your program is
1410 continued. This causes any pending terminal input to be lost.
1412 In the GNU system, this will be fixed by changing the behavior of
1415 Under Unix, when you have this problem, you might be able to get around
1416 it by putting your actions into the breakpoint condition instead of
1417 commands. For example
1420 condition 5 (x = y + 4), 0
1424 specifies a condition expression (@xref{Expressions}) that will change
1425 @code{x} as needed, then always have the value 0 so the program will not
1426 stop. Loss of input is avoided here because break conditions are
1427 evaluated without changing the terminal modes. When you want to have
1428 nontrivial conditions for performing the side effects, the operators
1429 @samp{&&}, @samp{||} and @samp{?@dots{}:} may be useful.
1431 @node Error in Breakpoints,, Break Commands, Breakpoints
1432 @subsection ``Cannot Insert Breakpoints'' Error
1434 Under some operating systems, breakpoints cannot be used in a program if
1435 any other process is running that program. Attempting to run or
1436 continue the program with a breakpoint in this case will cause GDB to
1439 When this happens, you have three ways to proceed:
1443 Remove or disable the breakpoints, then continue.
1446 Suspend GDB, and copy the file containing the program to a new name.
1447 Resume GDB and use the @samp{exec-file} command to specify that GDB
1448 should run the program under that name. Then start the program again.
1451 Relink the program so that the text segment is nonsharable, using the
1452 linker option @samp{-N}. The operating system limitation may not apply
1453 to nonsharable executables.
1456 @node Continuing, Stepping, Breakpoints, Stopping
1459 After your program stops, most likely you will want it to run some more if
1460 the bug you are looking for has not happened yet.
1465 Continue running the program at the place where it stopped.
1468 If the program stopped at a breakpoint, the place to continue running
1469 is the address of the breakpoint. You might expect that continuing would
1470 just stop at the same breakpoint immediately. In fact, @samp{cont}
1471 takes special care to prevent that from happening. You do not need
1472 to delete the breakpoint to proceed through it after stopping at it.
1474 You can, however, specify an ignore-count for the breakpoint that the
1475 program stopped at, by means of an argument to the @samp{cont} command.
1478 If the program stopped because of a signal other than @code{SIGINT} or
1479 @code{SIGTRAP}, continuing will cause the program to see that signal.
1480 You may not want this to happen. For example, if the program stopped
1481 due to some sort of memory reference error, you might store correct
1482 values into the erroneous variables and continue, hoping to see more
1483 execution; but the program would probably terminate immediately as
1484 a result of the fatal signal once it sees the signal. To prevent this,
1485 you can continue with @samp{signal 0}. @xref{Signaling}. You can
1486 also act in advance to prevent the program from seeing certain kinds
1487 of signals, using the @samp{handle} command (@pxref{Signals}).
1489 @node Stepping,, Continuing, Stopping
1493 @dfn{Stepping} means setting your program in motion for a limited time, so
1494 that control will return automatically to the debugger after one line of
1495 code or one machine instruction. Breakpoints are active during stepping
1496 and the program will stop for them even if it has not gone as far as the
1497 stepping command specifies.
1502 Continue running the program until control reaches a different line,
1503 then stop it and return control to the debugger. This command is
1504 abbreviated @samp{s}.
1506 This command may be given when control is within a function for which
1507 there is no debugging information. In that case, execution will proceed
1508 until control reaches a different function, or is about to return from
1509 this function. An argument repeats this action.
1511 @item step @var{count}
1512 Continue running as in @samp{step}, but do so @var{count} times. If a
1513 breakpoint is reached or a signal not related to stepping occurs before
1514 @var{count} steps, stepping stops right away.
1518 Similar to @samp{step}, but any function calls appearing within the line of
1519 code are executed without stopping. Execution stops when control reaches a
1520 different line of code at the stack level which was executing when the
1521 @samp{next} command was given. This command is abbreviated @samp{n}.
1523 An argument is a repeat count, as in @samp{step}.
1525 @samp{next} within a function without debugging information acts as does
1526 @samp{step}, but any function calls appearing within the code of the
1527 function are executed without stopping.
1531 Continue running until just after the selected stack frame returns (or
1532 until there is some other reason to stop, such as a fatal signal or a
1533 breakpoint). Print value returned by the selected stack frame (if any).
1535 Contrast this with the @samp{return} command (@pxref{Returning}).
1539 This command is used to avoid single stepping through a loop more than
1540 once. It is like the @samp{next} command, except that when @samp{until}
1541 encounters a jump, it automatically continues execution until the
1542 program counter is greater than the address of the jump.
1544 This means that when you reach the end of a loop after single stepping
1545 though it, @samp{until} will cause the program to continue execution
1546 until the loop is exited. In contrast, a @samp{next} command at the end
1547 of a loop will simply step back to the beginning of the loop, which
1548 would force you to step through the next iteration.
1550 @samp{until} always stops the program if it attempts to exit the current
1553 @samp{until} may produce somewhat counterintuitive results if the order
1554 of the source lines does not match the actual order of execution. For
1555 example, in a typical C @code{for}-loop, the third expression in the
1556 @code{for}-statement (the loop-step expression) is executed after the
1557 statements in the body of the loop, but is written before them.
1558 Therefore, the @samp{until} command would appear to step back to the
1559 beginning of the loop when it advances to this expression. However, it
1560 has not really done so, not in terms of the actual machine code.
1562 Note that @samp{until} with no argument works by means of single
1563 instruction stepping, and hence is slower than @samp{until} with an
1566 @item until @var{location}
1567 Continue running the program until either the specified location is
1568 reached, or the current (innermost) stack frame returns. This form of
1569 the command uses breakpoints, and hence is quicker than @samp{until}
1570 without an argument.
1576 Execute one machine instruction, then stop and return to the debugger.
1578 It is often useful to do @samp{display/i $pc} when stepping by machine
1579 instructions. This will cause the next instruction to be executed to
1580 be displayed automatically at each stop. @xref{Auto Display}.
1582 An argument is a repeat count, as in @samp{step}.
1588 Execute one machine instruction, but if it is a subroutine call,
1589 proceed until the subroutine returns.
1591 An argument is a repeat count, as in @samp{next}.
1594 A typical technique for using stepping is to put a breakpoint
1595 (@pxref{Breakpoints}) at the beginning of the function or the section of
1596 the program in which a problem is believed to lie, and then step through
1597 the suspect area, examining the variables that are interesting, until the
1600 The @samp{cont} command can be used after stepping to resume execution
1601 until the next breakpoint or signal.
1603 @node Stack, Source, Stopping, Top
1604 @chapter Examining the Stack
1606 When your program has stopped, the first thing you need to know is where it
1607 stopped and how it got there.
1610 Each time your program performs a function call, the information about
1611 where in the program the call was made from is saved in a block of data
1612 called a @dfn{stack frame}. The frame also contains the arguments of the
1613 call and the local variables of the function that was called. All the
1614 stack frames are allocated in a region of memory called the @dfn{call
1617 When your program stops, the GDB commands for examining the stack allow you
1618 to see all of this information.
1620 One of the stack frames is @dfn{selected} by GDB and many GDB commands
1621 refer implicitly to the selected frame. In particular, whenever you ask
1622 GDB for the value of a variable in the program, the value is found in the
1623 selected frame. There are special GDB commands to select whichever frame
1624 you are interested in.
1626 When the program stops, GDB automatically selects the currently executing
1627 frame and describes it briefly as the @samp{frame} command does
1628 (@pxref{Frame Info, Info}).
1631 * Frames:: Explanation of stack frames and terminology.
1632 * Backtrace:: Summarizing many frames at once.
1633 * Selection:: How to select a stack frame.
1634 * Info: Frame Info, Commands to print information on stack frames.
1637 @node Frames, Backtrace, Stack, Stack
1638 @section Stack Frames
1642 The call stack is divided up into contiguous pieces called @dfn{stack
1643 frames}, or @dfn{frames} for short; each frame is the data associated
1644 with one call to one function. The frame contains the arguments given
1645 to the function, the function's local variables, and the address at
1646 which the function is executing.
1648 @cindex initial frame
1649 @cindex outermost frame
1650 @cindex innermost frame
1651 When your program is started, the stack has only one frame, that of the
1652 function @code{main}. This is called the @dfn{initial} frame or the
1653 @dfn{outermost} frame. Each time a function is called, a new frame is
1654 made. Each time a function returns, the frame for that function invocation
1655 is eliminated. If a function is recursive, there can be many frames for
1656 the same function. The frame for the function in which execution is
1657 actually occurring is called the @dfn{innermost} frame. This is the most
1658 recently created of all the stack frames that still exist.
1660 @cindex frame pointer
1661 Inside your program, stack frames are identified by their addresses. A
1662 stack frame consists of many bytes, each of which has its own address; each
1663 kind of computer has a convention for choosing one of those bytes whose
1664 address serves as the address of the frame. Usually this address is kept
1665 in a register called the @dfn{frame pointer register} while execution is
1666 going on in that frame.
1668 @cindex frame number
1669 GDB assigns numbers to all existing stack frames, starting with zero for
1670 the innermost frame, one for the frame that called it, and so on upward.
1671 These numbers do not really exist in your program; they are to give you a
1672 way of talking about stack frames in GDB commands.
1674 @cindex selected frame
1675 Many GDB commands refer implicitly to one stack frame. GDB records a stack
1676 frame that is called the @dfn{selected} stack frame; you can select any
1677 frame using one set of GDB commands, and then other commands will operate
1678 on that frame. When your program stops, GDB automatically selects the
1681 @cindex frameless execution
1682 Some functions can be compiled to run without a frame reserved for them
1683 on the stack. This is occasionally done with heavily used library
1684 functions to save the frame setup time. GDB has limited facilities for
1685 dealing with these function invocations; if the innermost function
1686 invocation has no stack frame, GDB will give it a virtual stack frame of
1687 0 and correctly allow tracing of the function call chain. Results are
1688 undefined if a function invocation besides the innermost one is
1691 @node Backtrace, Selection, Frames, Stack
1694 A backtrace is a summary of how the program got where it is. It shows one
1695 line per frame, for many frames, starting with the currently executing
1696 frame (frame zero), followed by its caller (frame one), and on up the
1704 Print a backtrace of the entire stack: one line per frame for all
1705 frames in the stack.
1707 You can stop the backtrace at any time by typing the system interrupt
1708 character, normally @kbd{Control-C}.
1710 @item backtrace @var{n}
1712 Similar, but print only the innermost @var{n} frames.
1714 @item backtrace @var{-n}
1716 Similar, but print only the outermost @var{n} frames.
1721 The names @samp{where} and @samp{info stack} are additional aliases
1722 for @samp{backtrace}.
1724 Every line in the backtrace shows the frame number, the function name
1725 and the program counter value.
1727 If the function is in a source file whose symbol table data has been
1728 fully read, the backtrace shows the source file name and line number, as
1729 well as the arguments to the function. (The program counter value is
1730 omitted if it is at the beginning of the code for that line number.)
1732 If the source file's symbol data has not been fully read, just scanned,
1733 this extra information is replaced with an ellipsis. You can force the
1734 symbol data for that frame's source file to be read by selecting the
1735 frame. (@xref{Selection}).
1737 Here is an example of a backtrace. It was made with the command
1738 @samp{bt 3}, so it shows the innermost three frames.
1741 #0 rtx_equal_p (x=(rtx) 0x8e58c, y=(rtx) 0x1086c4) (/gp/rms/cc/rtlanal.c line 337)
1742 #1 0x246b0 in expand_call (...) (...)
1743 #2 0x21cfc in expand_expr (...) (...)
1744 (More stack frames follow...)
1748 The functions @code{expand_call} and @code{expand_expr} are in a file
1749 whose symbol details have not been fully read. Full detail is available
1750 for the function @code{rtx_equal_p}, which is in the file
1751 @file{rtlanal.c}. Its arguments, named @code{x} and @code{y}, are shown
1752 with their typed values.
1754 @node Selection, Frame Info, Backtrace, Stack
1755 @section Selecting a Frame
1757 Most commands for examining the stack and other data in the program work on
1758 whichever stack frame is selected at the moment. Here are the commands for
1759 selecting a stack frame; all of them finish by printing a brief description
1760 of the stack frame just selected.
1765 Select frame number @var{n}. Recall that frame zero is the innermost
1766 (currently executing) frame, frame one is the frame that called the
1767 innermost one, and so on. The highest-numbered frame is @code{main}'s
1770 @item frame @var{addr}
1771 Select the frame at address @var{addr}. This is useful mainly if the
1772 chaining of stack frames has been damaged by a bug, making it
1773 impossible for GDB to assign numbers properly to all frames. In
1774 addition, this can be useful when the program has multiple stacks and
1775 switches between them.
1779 Select the frame @var{n} frames up from the frame previously selected.
1780 For positive numbers @var{n}, this advances toward the outermost
1781 frame, to higher frame numbers, to frames that have existed longer.
1782 @var{n} defaults to one.
1786 Select the frame @var{n} frames down from the frame previously
1787 selected. For positive numbers @var{n}, this advances toward the
1788 innermost frame, to lower frame numbers, to frames that were created
1789 more recently. @var{n} defaults to one.
1792 All of these commands end by printing some information on the frame that
1793 has been selected: the frame number, the function name, the arguments, the
1794 source file and line number of execution in that frame, and the text of
1795 that source line. For example:
1798 #3 main (argc=3, argv=??, env=??) at main.c, line 67
1799 67 read_input_file (argv[i]);
1802 After such a printout, the @samp{list} command with no arguments will print
1803 ten lines centered on the point of execution in the frame. @xref{List}.
1805 @node Frame Info,, Selection, Stack
1806 @section Information on a Frame
1808 There are several other commands to print information about the selected
1813 This command prints a brief description of the selected stack frame.
1814 It can be abbreviated @samp{f}. With an argument, this command is
1815 used to select a stack frame; with no argument, it does not change
1816 which frame is selected, but still prints the same information.
1820 This command prints a verbose description of the selected stack frame,
1821 including the address of the frame, the addresses of the next frame in
1822 (called by this frame) and the next frame out (caller of this frame),
1823 the address of the frame's arguments, the program counter saved in it
1824 (the address of execution in the caller frame), and which registers
1825 were saved in the frame. The verbose description is useful when
1826 something has gone wrong that has made the stack format fail to fit
1827 the usual conventions.
1829 @item info frame @var{addr}
1830 Print a verbose description of the frame at address @var{addr},
1831 without selecting that frame. The selected frame remains unchanged by
1836 Print the arguments of the selected frame, each on a separate line.
1840 Print the local variables of the selected frame, each on a separate
1841 line. These are all variables declared static or automatic within all
1842 program blocks that execution in this frame is currently inside of.
1845 @node Source, Data, Stack, Top
1846 @chapter Examining Source Files
1848 GDB knows which source files your program was compiled from, and
1849 can print parts of their text. When your program stops, GDB
1850 spontaneously prints the line it stopped in. Likewise, when you
1851 select a stack frame (@pxref{Selection}), GDB prints the line
1852 which execution in that frame has stopped in. You can also
1853 print parts of source files by explicit command.
1856 * List:: Using the @samp{list} command to print source files.
1857 * Search:: Commands for searching source files.
1858 * Source Path:: Specifying the directories to search for source files.
1861 @node List, Search, Source, Source
1862 @section Printing Source Lines
1865 To print lines from a source file, use the @samp{list} command
1866 (abbreviated @samp{l}). There are several ways to specify what part
1867 of the file you want to print.
1869 Here are the forms of the @samp{list} command most commonly used:
1872 @item list @var{linenum}
1873 Print ten lines centered around line number @var{linenum} in the
1874 current source file.
1876 @item list @var{function}
1877 Print ten lines centered around the beginning of function
1881 Print ten more lines. If the last lines printed were printed with a
1882 @samp{list} command, this prints ten lines following the last lines
1883 printed; however, if the last line printed was a solitary line printed
1884 as part of displaying a stack frame (@pxref{Stack}), this prints ten
1885 lines centered around that line.
1888 Print ten lines just before the lines last printed.
1891 Repeating a @samp{list} command with @key{RET} discards the argument,
1892 so it is equivalent to typing just @samp{list}. This is more useful
1893 than listing the same lines again. An exception is made for an
1894 argument of @samp{-}; that argument is preserved in repetition so that
1895 each repetition moves up in the file.
1898 In general, the @samp{list} command expects you to supply zero, one or two
1899 @dfn{linespecs}. Linespecs specify source lines; there are several ways
1900 of writing them but the effect is always to specify some source line.
1901 Here is a complete description of the possible arguments for @samp{list}:
1904 @item list @var{linespec}
1905 Print ten lines centered around the line specified by @var{linespec}.
1907 @item list @var{first},@var{last}
1908 Print lines from @var{first} to @var{last}. Both arguments are
1911 @item list ,@var{last}
1912 Print ten lines ending with @var{last}.
1914 @item list @var{first},
1915 Print ten lines starting with @var{first}.
1918 Print ten lines just after the lines last printed.
1921 Print ten lines just before the lines last printed.
1924 As described in the preceding table.
1927 Here are the ways of specifying a single source line---all the
1932 Specifies line @var{linenum} of the current source file.
1933 When a @samp{list} command has two linespecs, this refers to
1934 the same source file as the first linespec.
1937 Specifies the line @var{offset} lines after the last line printed.
1938 When used as the second linespec in a @samp{list} command that has
1939 two, this specifies the line @var{offset} lines down from the
1943 Specifies the line @var{offset} lines before the last line printed.
1945 @item @var{filename}:@var{linenum}
1946 Specifies line @var{linenum} in the source file @var{filename}.
1948 @item @var{function}
1949 Specifies the line of the open-brace that begins the body of the
1950 function @var{function}.
1952 @item @var{filename}:@var{function}
1953 Specifies the line of the open-brace that begins the body of the
1954 function @var{function} in the file @var{filename}. The file name is
1955 needed with a function name only for disambiguation of identically
1956 named functions in different source files.
1958 @item *@var{address}
1959 Specifies the line containing the program address @var{address}.
1960 @var{address} may be any expression.
1963 One other command is used to map source lines to program addresses.
1966 @item info line @var{linenum}
1968 Print the starting and ending addresses of the compiled code for
1969 source line @var{linenum}.
1972 The default examine address for the @samp{x} command is changed to the
1973 starting address of the line, so that @samp{x/i} is sufficient to
1974 begin examining the machine code (@pxref{Memory}). Also, this address
1975 is saved as the value of the convenience variable @code{$_}
1976 (@pxref{Convenience Vars}).
1979 @node Search, Source Path, List, Source
1980 @section Searching Source Files
1982 @kindex forward-search
1983 @kindex reverse-search
1985 There are two commands for searching through the current source file for a
1988 The command @samp{forward-search @var{regexp}} checks each line, starting
1989 with the one following the last line listed, for a match for @var{regexp}.
1990 It lists the line that is found. You can abbreviate the command name
1993 The command @samp{reverse-search @var{regexp}} checks each line, starting
1994 with the one before the last line listed and going backward, for a match
1995 for @var{regexp}. It lists the line that is found. You can abbreviate
1996 this command with as little as @samp{rev}.
1998 @node Source Path,, Search, Source
1999 @section Specifying Source Directories
2002 @cindex directories for source files
2003 Executable programs do not record the directories of the source files
2004 from which they were compiled, just the names. GDB remembers a list of
2005 directories to search for source files; this is called the @dfn{source
2006 path}. Each time GDB wants a source file, it tries all the directories
2007 in the list, in the order they are present in the list, until it finds a
2008 file with the desired name. @b{Note that the executable search path is
2009 @i{not} used for this purpose. Neither is the current working
2010 directory, unless it happens to be in the source path.}
2013 When you start GDB, its source path contains just the current working
2014 directory. To add other directories, use the @samp{directory} command.
2017 @item directory @var{dirnames...}
2018 Add directory @var{dirname} to the end of the source path. Several
2019 directory names may be given to this command, separated by whitespace or
2023 Reset the source path to just the current working directory of GDB.
2024 This requires confirmation.
2026 Since this command deletes directories from the search path, it may
2027 change the directory in which a previously read source file will be
2028 discovered. To make this work correctly, this command also clears out
2029 the tables GDB maintains about the source files it has already found.
2031 @item info directories
2032 @kindex info directories
2033 Print the source path: show which directories it contains.
2036 Because the @samp{directory} command adds to the end of the source path,
2037 it does not affect any file that GDB has already found. If the source
2038 path contains directories that you do not want, and these directories
2039 contain misleading files with names matching your source files, the
2040 way to correct the situation is as follows:
2044 Choose the directory you want at the beginning of the source path.
2045 Use the @samp{cd} command to make that the current working directory.
2048 Use @samp{directory} with no argument to reset the source path to just
2052 Use @samp{directory} with suitable arguments to add any other
2053 directories you want in the source path.
2056 @node Data, Symbols, Source, Top
2057 @chapter Examining Data
2059 @cindex printing data
2060 @cindex examining data
2062 The usual way to examine data in your program is with the @samp{print}
2063 command (abbreviated @samp{p}). It evaluates and prints the value of any
2064 valid expression of the language the program is written in (for now, C).
2072 where @var{exp} is any valid expression, and the value of @var{exp}
2073 is printed in a format appropriate to its data type.
2075 A more low-level way of examining data is with the @samp{x} command.
2076 It examines data in memory at a specified address and prints it in a
2080 * Expressions:: Expressions that can be computed and printed.
2081 * Variables:: Using your program's variables in expressions.
2082 * Assignment:: Setting your program's variables.
2083 * Arrays:: Examining part of memory as an array.
2084 * Format Options:: Controlling how structures and arrays are printed.
2085 * Output formats:: Specifying formats for printing values.
2086 * Memory:: Examining memory explicitly.
2087 * Auto Display:: Printing certain expressions whenever program stops.
2088 * Value History:: Referring to values previously printed.
2089 * Convenience Vars:: Giving names to values for future reference.
2090 * Registers:: Referring to and storing in machine registers.
2093 @node Expressions, Variables, Data, Data
2094 @section Expressions
2097 Many different GDB commands accept an expression and compute its value.
2098 Any kind of constant, variable or operator defined by the programming
2099 language you are using is legal in an expression in GDB. This includes
2100 conditional expressions, function calls, casts and string constants.
2101 It unfortunately does not include symbols defined by preprocessor
2102 @code{#define} commands.
2104 Casts are supported in all languages, not just in C, because it is so
2105 useful to cast a number into a pointer so as to examine a structure
2106 at that address in memory.
2108 GDB supports three kinds of operator in addition to those of programming
2113 @samp{@@} is a binary operator for treating parts of memory as arrays.
2114 @xref{Arrays}, for more information.
2117 @samp{::} allows you to specify a variable in terms of the file or
2118 function it is defined in. @xref{Variables}.
2120 @item @{@var{type}@} @var{addr}
2121 Refers to an object of type @var{type} stored at address @var{addr} in
2122 memory. @var{addr} may be any expression whose value is an integer or
2123 pointer (but parentheses are required around nonunary operators, just as in
2124 a cast). This construct is allowed regardless of what kind of data is
2125 officially supposed to reside at @var{addr}.@refill
2128 @node Variables, Arrays, Expressions, Data
2129 @section Program Variables
2131 The most common kind of expression to use is the name of a variable
2134 Variables in expressions are understood in the selected stack frame
2135 (@pxref{Selection}); they must either be global (or static) or be visible
2136 according to the scope rules of the programming language from the point of
2137 execution in that frame. This means that in the function
2152 the variable @code{a} is usable whenever the program is executing
2153 within the function @code{foo}, but the variable @code{b} is visible
2154 only while the program is executing inside the block in which @code{b}
2157 As a special exception, you can refer to a variable or function whose
2158 scope is a single source file even if the current execution point is not
2159 in this file. But it is possible to have more than one such variable
2160 or function with the same name (if they are in different source files).
2161 In such a case, it is not defined which one you will get. If you wish,
2162 you can specify any one of them using the colon-colon construct:
2165 @var{block}::@var{variable}
2169 Here @var{block} is the name of the source file whose variable you want.
2171 @node Arrays, Format options, Variables, Data
2172 @section Artificial Arrays
2174 @cindex artificial array
2175 It is often useful to print out several successive objects of the
2176 same type in memory; a section of an array, or an array of
2177 dynamically determined size for which only a pointer exists in the
2180 This can be done by constructing an @dfn{artificial array} with the
2181 binary operator @samp{@@}. The left operand of @samp{@@} should be
2182 the first element of the desired array, as an individual object.
2183 The right operand should be the length of the array. The result is
2184 an array value whose elements are all of the type of the left argument.
2185 The first element is actually the left argument; the second element
2186 comes from bytes of memory immediately following those that hold the
2187 first element, and so on. Here is an example. If a program says
2190 int *array = (int *) malloc (len * sizeof (int));
2194 you can print the contents of @code{array} with
2200 The left operand of @samp{@@} must reside in memory. Array values made
2201 with @samp{@@} in this way behave just like other arrays in terms of
2202 subscripting, and are coerced to pointers when used in expressions.
2203 (It would probably appear in an expression via the value history,
2204 after you had printed it out.)
2206 @node Format options, Output formats, Arrays, Data
2207 @section Format options
2209 @cindex format options
2210 GDB provides a few ways to control how arrays and structures are
2216 Display the current settings for the format options.
2218 @item set array-max @var{number-of-elements}
2219 @kindex set array-max
2220 If GDB is printing a large array, it will stop printing after it has
2221 printed the number of elements set by the @samp{set array-max} command.
2222 This limit also applies to the display of strings.
2224 @item set prettyprint on
2225 @kindex set prettyprint
2226 Cause GDB to print structures in an indented format with one member per
2240 @item set prettyprint off
2241 Cause GDB to print structures in a compact format, like this:
2244 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, meat = 0x54 "Pork"@}
2247 This is the default format.
2249 @item set unionprint on
2250 @kindex set unionprint
2251 Tell GDB to print unions which are contained in structures. This is the
2253 @item set unionprint off
2254 Tell GDB not to print unions which are contained in structures.
2256 For example, given the declarations
2259 typedef enum @{Tree, Bug@} Species;
2260 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
2261 typedef enum @{Caterpiller, Cocoon, Butterfly@} Bug_forms;
2271 struct thing foo = @{Tree, @{Acorn@}@};
2275 with @samp{set unionprint on} in effect @samp{p foo} would print
2278 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
2282 and with @samp{set unionprint off} in effect it would print
2285 $1 = @{it = Tree, form = @{...@}@}
2289 @node Output formats, Memory, Format options, Data
2290 @section Output formats
2292 @cindex formatted output
2293 @cindex output formats
2294 GDB normally prints all values according to their data types. Sometimes
2295 this is not what you want. For example, you might want to print a number
2296 in hex, or a pointer in decimal. Or you might want to view data in memory
2297 at a certain address as a character string or an instruction. These things
2298 can be done with @dfn{output formats}.
2300 The simplest use of output formats is to say how to print a value
2301 already computed. This is done by starting the arguments of the
2302 @samp{print} command with a slash and a format letter. The format
2303 letters supported are:
2307 Regard the bits of the value as an integer, and print the integer in
2311 Print as integer in signed decimal.
2314 Print as integer in unsigned decimal.
2317 Print as integer in octal.
2320 Print as an address, both absolute in hex and then relative
2321 to a symbol defined as an address below it.
2324 Regard as an integer and print it as a character constant.
2327 Regard the bits of the value as a floating point number and print
2328 using typical floating point syntax.
2331 For example, to print the program counter in hex (@pxref{Registers}), type
2338 Note that no space is required before the slash; this is because command
2339 names in GDB cannot contain a slash.
2341 To reprint the last value in the value history with a different format,
2342 you can use the @samp{print} command with just a format and no
2343 expression. For example, @samp{p/x} reprints the last value in hex.
2345 @node Memory, Auto Display, Output formats, Data
2346 @subsection Examining Memory
2348 @cindex examining memory
2350 The command @samp{x} (for `examine') can be used to examine memory
2351 without reference to the program's data types. The format in which you
2352 wish to examine memory is instead explicitly specified. The allowable
2353 formats are a superset of the formats described in the previous section.
2355 @samp{x} is followed by a slash and an output format specification,
2356 followed by an expression for an address. The expression need not have
2357 a pointer value (though it may); it is used as an integer, as the
2358 address of a byte of memory. @xref{Expressions} for more information on
2359 expressions. For example, @samp{x/4xw $sp} prints the four words of
2360 memory above the stack pointer in hexadecimal.
2362 The output format in this case specifies both how big a unit of memory
2363 to examine and how to print the contents of that unit. It is done
2364 with one or two of the following letters:
2366 These letters specify just the size of unit to examine:
2370 Examine individual bytes.
2373 Examine halfwords (two bytes each).
2376 Examine words (four bytes each).
2379 Many assemblers and cpu designers still use `word' for a 16-bit quantity,
2380 as a holdover from specific predecessor machines of the 1970's that really
2381 did use two-byte words. But more generally the term `word' has always
2382 referred to the size of quantity that a machine normally operates on and
2383 stores in its registers. This is 32 bits for all the machines that GDB
2387 Examine giant words (8 bytes).
2390 These letters specify just the way to print the contents:
2394 Print as integers in unsigned hexadecimal.
2397 Print as integers in signed decimal.
2400 Print as integers in unsigned decimal.
2403 Print as integers in unsigned octal.
2406 Print as an address, both absolute in hex and then relative
2407 to a symbol defined as an address below it.
2410 Print as character constants.
2413 Print as floating point. This works only with sizes @samp{w} and
2417 Print a null-terminated string of characters. The specified unit size
2418 is ignored; instead, the unit is however many bytes it takes to reach
2419 a null character (including the null character).
2422 Print a machine instruction in assembler syntax (or nearly). The
2423 specified unit size is ignored; the number of bytes in an instruction
2424 varies depending on the type of machine, the opcode and the addressing
2428 If either the manner of printing or the size of unit fails to be specified,
2429 the default is to use the same one that was used last. If you don't want
2430 to use any letters after the slash, you can omit the slash as well.
2432 You can also omit the address to examine. Then the address used is
2433 just after the last unit examined. This is why string and instruction
2434 formats actually compute a unit-size based on the data: so that the
2435 next string or instruction examined will start in the right place.
2436 The @samp{print} command sometimes sets the default address for
2437 the @samp{x} command; when the value printed resides in memory, the
2438 default is set to examine the same location. @samp{info line} also
2439 sets the default for @samp{x}, to the address of the start of the
2440 machine code for the specified line and @samp{info breakpoints} sets
2441 it to the address of the last breakpoint listed.
2443 When you use @key{RET} to repeat an @samp{x} command, it does not repeat
2444 exactly the same: the address specified previously (if any) is ignored, so
2445 that the repeated command examines the successive locations in memory
2446 rather than the same ones.
2448 You can examine several consecutive units of memory with one command by
2449 writing a repeat-count after the slash (before the format letters, if any).
2450 The repeat count must be a decimal integer. It has the same effect as
2451 repeating the @samp{x} command that many times except that the output may
2452 be more compact with several units per line. For example,
2459 prints ten instructions starting with the one to be executed next in the
2460 selected frame. After doing this, you could print another ten following
2468 in which the format and address are allowed to default.
2472 The addresses and contents printed by the @samp{x} command are not put in
2473 the value history because there is often too much of them and they would
2474 get in the way. Instead, GDB makes these values available for subsequent
2475 use in expressions as values of the convenience variables @code{$_} and
2478 After an @samp{x} command, the last address examined is available for use
2479 in expressions in the convenience variable @code{$_}. The contents of that
2480 address, as examined, are available in the convenience variable @code{$__}.
2482 If the @samp{x} command has a repeat count, the address and contents saved
2483 are from the last memory unit printed; this is not the same as the last
2484 address printed if several units were printed on the last line of output.
2487 The specialized command @samp{disassemble} is also provided to dump a
2488 range of memory as machine instructions. The default memory range is
2489 the function surrounding the program counter of the selected frame. A
2490 single argument to this command is a program counter value; the function
2491 surrounding this value will be dumped. Two arguments specify a range of
2492 addresss (first inclusive, second exclusive) to be dumped.
2494 @node Auto Display, Value History, Memory, Data
2495 @section Automatic Display
2496 @cindex automatic display
2497 @cindex display of expressions
2499 If you find that you want to print the value of an expression frequently
2500 (to see how it changes), you might want to add it to the @dfn{automatic
2501 display list} so that GDB will print its value each time the program stops.
2502 Each expression added to the list is given a number to identify it;
2503 to remove an expression from the list, you specify that number.
2504 The automatic display looks like this:
2508 3: bar[5] = (struct hack *) 0x3804
2512 showing item numbers, expressions and their current values.
2514 If the expression refers to local variables, then it does not make sense
2515 outside the lexical context for which it was set up. Such an expression
2516 is printed only when execution is inside that lexical context. For
2517 example, if you give the command @samp{display name} while inside a
2518 function with an argument @code{name}, then this argument will be
2519 displayed whenever the program stops inside that function, but not when
2520 it stops elsewhere (since this argument doesn't exist elsewhere).
2523 @item display @var{exp}
2525 Add the expression @var{exp} to the list of expressions to display
2526 each time the program stops. @xref{Expressions}.
2528 @item display/@var{fmt} @var{exp}
2529 For @var{fmt} specifying only a display format and not a size or
2530 count, add the expression @var{exp} to the auto-display list but
2531 arranges to display it each time in the specified format @var{fmt}.
2533 @item display/@var{fmt} @var{addr}
2534 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
2535 number of units, add the expression @var{addr} as a memory address to
2536 be examined each time the program stops. Examining means in effect
2537 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory}.
2539 @item undisplay @var{dnums}@dots{}
2540 @itemx delete display @var{dnums}@dots{}
2541 @kindex delete display
2543 Remove item numbers @var{dnums} from the list of expressions to display.
2545 @item disable display @var{dnums}@dots{}
2546 @kindex disable display
2547 Disable the display of item numbers @var{dnums}. A disabled display
2548 item is not printed automatically, but is not forgotten. It may be
2551 @item enable display @var{dnums}@dots{}
2552 @kindex enable display
2553 Enable display of item numbers @var{dnums}. It becomes effective once
2554 again in auto display of its expression, until you specify otherwise.
2557 Display the current values of the expressions on the list, just as is
2558 done when the program stops.
2561 @kindex info display
2562 Print the list of expressions previously set up to display
2563 automatically, each one with its item number, but without showing the
2564 values. This includes disabled expressions, which are marked as such.
2565 It also includes expressions which would not be displayed right now
2566 because they refer to automatic variables not currently available.
2569 @node Value History, Convenience Vars, Auto Display, Data
2570 @section Value History
2572 @cindex value history
2573 Every value printed by the @samp{print} command is saved for the entire
2574 session in GDB's @dfn{value history} so that you can refer to it in
2579 @cindex history number
2580 The values printed are given @dfn{history numbers} for you to refer to them
2581 by. These are successive integers starting with 1. @samp{print} shows you
2582 the history number assigned to a value by printing @samp{$@var{num} = }
2583 before the value; here @var{num} is the history number.
2585 To refer to any previous value, use @samp{$} followed by the value's
2586 history number. The output printed by @samp{print} is designed to remind
2587 you of this. Just @code{$} refers to the most recent value in the history,
2588 and @code{$$} refers to the value before that.
2590 For example, suppose you have just printed a pointer to a structure and
2591 want to see the contents of the structure. It suffices to type
2597 If you have a chain of structures where the component @samp{next} points
2598 to the next one, you can print the contents of the next one with this:
2605 It might be useful to repeat this command many times by typing @key{RET}.
2607 Note that the history records values, not expressions. If the value of
2608 @code{x} is 4 and you type this command:
2616 then the value recorded in the value history by the @samp{print} command
2617 remains 4 even though the value of @code{x} has changed.
2622 Print the last ten values in the value history, with their item
2623 numbers. This is like @samp{p $$9} repeated ten times, except that
2624 @samp{info values} does not change the history.
2626 @item info values @var{n}
2627 Print ten history values centered on history item number @var{n}.
2630 Print ten history values just after the values last printed.
2633 @node Convenience Vars, Registers, Value History, Data
2634 @section Convenience Variables
2636 @cindex convenience variables
2637 GDB provides @dfn{convenience variables} that you can use within GDB to
2638 hold on to a value and refer to it later. These variables exist entirely
2639 within GDB; they are not part of your program, and setting a convenience
2640 variable has no effect on further execution of your program. That's why
2641 you can use them freely.
2643 Convenience variables have names starting with @samp{$}. Any name starting
2644 with @samp{$} can be used for a convenience variable, unless it is one of
2645 the predefined set of register names (@pxref{Registers}).
2647 You can save a value in a convenience variable with an assignment
2648 expression, just as you would set a variable in your program. Example:
2651 set $foo = *object_ptr
2655 would save in @code{$foo} the value contained in the object pointed to by
2658 Using a convenience variable for the first time creates it; but its value
2659 is @code{void} until you assign a new value. You can alter the value with
2660 another assignment at any time.
2662 Convenience variables have no fixed types. You can assign a convenience
2663 variable any type of value, even if it already has a value of a different
2664 type. The convenience variable as an expression has whatever type its
2668 @item info convenience
2669 @kindex info convenience
2670 Print a list of convenience variables used so far, and their values.
2671 Abbreviated @samp{i con}.
2674 One of the ways to use a convenience variable is as a counter to be
2675 incremented or a pointer to be advanced. For example:
2679 print bar[$i++]->contents
2680 @i{@dots{}repeat that command by typing @key{RET}.}
2683 Some convenience variables are created automatically by GDB and given
2684 values likely to be useful.
2688 The variable @code{$_} is automatically set by the @samp{x} command to
2689 the last address examined (@pxref{Memory}). Other commands which
2690 provide a default address for @samp{x} to examine also set @code{$_}
2691 to that address; these commands include @samp{info line} and @samp{info
2695 The variable @code{$__} is automatically set by the @samp{x} command
2696 to the value found in the last address examined.
2699 @node Registers,, Convenience Vars, Data
2703 Machine register contents can be referred to in expressions as variables
2704 with names starting with @samp{$}. The names of registers are different
2705 for each machine; use @samp{info registers} to see the names used on your
2706 machine. The names @code{$pc} and @code{$sp} are used on all machines for
2707 the program counter register and the stack pointer. Often @code{$fp} is
2708 used for a register that contains a pointer to the current stack frame,
2709 and @code{$ps} is used for a register that contains the processor
2710 status. These standard register names may be available on your machine
2711 even though the @code{info registers} command displays them with a
2712 different name. For example, on the SPARC, @code{info registers}
2713 displays the processor status register as @code{$psr} but you can also
2714 refer to it as @code{$ps}.
2716 GDB always considers the contents of an ordinary register as an integer
2717 when the register is examined in this way. Some machines have special
2718 registers which can hold nothing but floating point; these registers are
2719 considered floating point. There is no way to refer to the contents of an
2720 ordinary register as floating point value (although you can @emph{print}
2721 it as a floating point value with @samp{print/f $@var{regname}}).
2723 Some registers have distinct ``raw'' and ``virtual'' data formats. This
2724 means that the data format in which the register contents are saved by the
2725 operating system is not the same one that your program normally sees. For
2726 example, the registers of the 68881 floating point coprocessor are always
2727 saved in ``extended'' format, but all C programs expect to work with
2728 ``double'' format. In such cases, GDB normally works with the virtual
2729 format only (the format that makes sense for your program), but the
2730 @samp{info registers} command prints the data in both formats.
2732 Register values are relative to the selected stack frame
2733 (@pxref{Selection}). This means that you get the value that the register
2734 would contain if all stack frames farther in were exited and their saved
2735 registers restored. In order to see the real contents of all registers,
2736 you must select the innermost frame (with @samp{frame 0}).
2738 Some registers are never saved (typically those numbered zero or one)
2739 because they are used for returning function values; for these registers,
2740 relativization makes no difference.
2743 @item info registers
2744 @kindex info registers
2745 Print the names and relativized values of all registers.
2747 @item info registers @var{regname}
2748 Print the relativized value of register @var{regname}. @var{regname}
2749 may be any register name valid on the machine you are using, with
2750 or without the initial @samp{$}.
2753 @subsection Examples
2755 You could print the program counter in hex with
2762 or print the instruction to be executed next with
2769 or add four to the stack pointer with
2776 The last is a way of removing one word from the stack, on machines where
2777 stacks grow downward in memory (most machines, nowadays). This assumes
2778 that the innermost stack frame is selected. Setting @code{$sp} is
2779 not allowed when other stack frames are selected.
2781 @node Symbols, Altering, Data, Top
2782 @chapter Examining the Symbol Table
2784 The commands described in this section allow you to make inquiries for
2785 information about the symbols (names of variables, functions and types)
2786 defined in your program. This information is found by GDB in the symbol
2787 table loaded by the @samp{symbol-file} command; it is inherent in the text
2788 of your program and does not change as the program executes.
2791 @item whatis @var{exp}
2793 Print the data type of expression @var{exp}. @var{exp} is not
2794 actually evaluated, and any side-effecting operations (such as
2795 assignments or function calls) inside it do not take place.
2799 Print the data type of @code{$}, the last value in the value history.
2801 @item info address @var{symbol}
2802 @kindex info address
2803 Describe where the data for @var{symbol} is stored. For a register
2804 variable, this says which register it is kept in. For a non-register
2805 local variable, this prints the stack-frame offset at which the variable
2808 Note the contrast with @samp{print &@var{symbol}}, which does not work
2809 at all for a register variables, and for a stack local variable prints
2810 the exact address of the current instantiation of the variable.
2812 @item ptype @var{typename}
2814 Print a description of data type @var{typename}. @var{typename} may be
2815 the name of a type, or for C code it may have the form
2816 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
2817 @samp{enum @var{enum-tag}}.@refill
2820 @kindex info sources
2821 Print the names of all source files in the program for which there
2822 is debugging information.
2824 @item info functions
2825 @kindex info functions
2826 Print the names and data types of all defined functions.
2828 @item info functions @var{regexp}
2829 Print the names and data types of all defined functions
2830 whose names contain a match for regular expression @var{regexp}.
2831 Thus, @samp{info fun step} finds all functions whose names
2832 include @samp{step}; @samp{info fun ^step} finds those whose names
2833 start with @samp{step}.
2835 @item info variables
2836 @kindex info variables
2837 Print the names and data types of all variables that are declared
2838 outside of functions (i.e., except for local variables).
2840 @item info variables @var{regexp}
2841 Print the names and data types of all variables (except for local
2842 variables) whose names contain a match for regular expression
2847 Print all data types that are defined in the program.
2849 @item info types @var{regexp}
2850 Print all data types that are defined in the program whose names
2851 contain a match for regular expression @var{regexp}.
2854 This was never implemented.
2856 @itemx info methods @var{regexp}
2857 @kindex info methods
2858 The @samp{info-methods} command permits the user to examine all defined
2859 methods within C++ program, or (with the @var{regexp} argument) a
2860 specific set of methods found in the various C++ classes. Many
2861 C++ classes provide a large number of methods. Thus, the output
2862 from the @samp{ptype} command can be overwhelming and hard to use. The
2863 @samp{info-methods} command filters the methods, printing only those
2864 which match the regular-expression @var{regexp}.
2867 @item printsyms @var{filename}
2869 Write a complete dump of the debugger's symbol data into the
2870 file @var{filename}.
2873 @node Altering, Sequences, Symbols, Top
2874 @chapter Altering Execution
2876 Once you think you have find an error in the program, you might want to
2877 find out for certain whether correcting the apparent error would lead to
2878 correct results in the rest of the run. You can find the answer by
2879 experiment, using the GDB features for altering execution of the
2882 For example, you can store new values into variables or memory
2883 locations, give the program a signal, restart it at a different address,
2884 or even return prematurely from a function to its caller.
2887 * Assignment:: Altering variable values or memory contents.
2888 * Jumping:: Altering control flow.
2889 * Signaling:: Making signals happen in the program.
2890 * Returning:: Making a function return prematurely.
2893 @node Assignment, Jumping, Altering, Altering
2894 @section Assignment to Variables
2897 @cindex setting variables
2898 To alter the value of a variable, evaluate an assignment expression.
2899 @xref{Expressions}. For example,
2906 would store the value 4 into the variable @code{x}, and then print
2907 the value of the assignment expression (which is 4).
2909 All the assignment operators of C are supported, including the
2910 incrementation operators @samp{++} and @samp{--}, and combining
2911 assignments such as @samp{+=} and @samp{<<=}.
2914 @kindex set variable
2915 If you are not interested in seeing the value of the assignment, use the
2916 @samp{set} command instead of the @samp{print} command. @samp{set} is
2917 really the same as @samp{print} except that the expression's value is not
2918 printed and is not put in the value history (@pxref{Value History}). The
2919 expression is evaluated only for side effects.
2921 Note that if the beginning of the argument string of the @samp{set} command
2922 appears identical to a @samp{set} subcommand, it may be necessary to use
2923 the @samp{set variable} command. This command is identical to @samp{set}
2924 except for its lack of subcommands.
2926 GDB allows more implicit conversions in assignments than C does; you can
2927 freely store an integer value into a pointer variable or vice versa, and
2928 any structure can be converted to any other structure that is the same
2931 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
2932 construct to generate a value of specified type at a specified address
2933 (@pxref{Expressions}). For example, @code{@{int@}0x83040} would refer
2934 to memory location 0x83040 as an integer (which implies a certain size
2935 and representation in memory), and
2938 set @{int@}0x83040 = 4
2941 would store the value 4 into that memory location.
2943 @node Jumping, Signaling, Assignment, Altering
2944 @section Continuing at a Different Address
2946 Ordinarily, when you continue the program, you do so at the place where
2947 it stopped, with the @samp{cont} command. You can instead continue at
2948 an address of your own choosing, with the following commands:
2951 @item jump @var{linenum}
2953 Resume execution at line number @var{linenum}. Execution may stop
2954 immediately if there is a breakpoint there.
2956 The @samp{jump} command does not change the current stack frame, or
2957 the stack pointer, or the contents of any memory location or any
2958 register other than the program counter. If line @var{linenum} is in
2959 a different function from the one currently executing, the results may
2960 be bizarre if the two functions expect different patterns of arguments or
2961 of local variables. For this reason, the @samp{jump} command requests
2962 confirmation if the specified line is not in the function currently
2963 executing. However, even bizarre results are predictable based on
2964 careful study of the machine-language code of the program.
2966 @item jump *@var{address}
2967 Resume execution at the instruction at address @var{address}.
2970 You can get much the same effect as the @code{jump} command by storing a
2971 new value into the register @code{$pc}. The difference is that this
2972 does not start the program running; it only changes the address where it
2973 @emph{will} run when it is continued. For example,
2980 causes the next @samp{cont} command or stepping command to execute at
2981 address 0x485, rather than at the address where the program stopped.
2984 The most common occasion to use the @samp{jump} command is when you have
2985 stepped across a function call with @code{next}, and found that the
2986 return value is incorrect. If all the relevant data appeared correct
2987 before the function call, the error is probably in the function that
2990 In general, your next step would now be to rerun the program and execute
2991 up to this function call, and then step into it to see where it goes
2992 astray. But this may be time consuming. If the function did not have
2993 significant side effects, you could get the same information by resuming
2994 execution just before the function call and stepping through it. To do this,
2995 first put a breakpoint on that function; then, use the @samp{jump} command
2996 to continue on the line with the function call.
2998 @node Signaling, Returning, Jumping, Altering
2999 @section Giving the Program a Signal
3002 @item signal @var{signalnum}
3004 Resume execution where the program stopped, but give it immediately the
3005 signal number @var{signalnum}.
3007 Alternatively, if @var{signalnum} is zero, continue execution without
3008 giving a signal. This is useful when the program stopped on account of
3009 a signal and would ordinary see the signal when resumed with the
3010 @samp{cont} command; @samp{signal 0} causes it to resume without a
3014 @node Returning,, Signaling, Altering
3015 @section Returning from a Function
3017 @cindex returning from a function
3019 You can cancel execution of a function call with the @samp{return}
3020 command. This command has the effect of discarding the selected stack
3021 frame (and all frames within it), so that control moves to the caller of
3022 that function. You can think of this as making the discarded frame
3025 First select the stack frame that you wish to return from
3026 (@pxref{Selection}). Then type the @samp{return} command. If you wish
3027 to specify the value to be returned, give that as an argument.
3029 This pops the selected stack frame (and any other frames inside of it),
3030 leaving its caller as the innermost remaining frame. That frame becomes
3031 selected. The specified value is stored in the registers used for
3032 returning values of functions.
3034 The @samp{return} command does not resume execution; it leaves the
3035 program stopped in the state that would exist if the function had just
3036 returned. Contrast this with the @samp{finish} command
3037 (@pxref{Stepping}), which resumes execution until the selected stack
3038 frame returns @emph{naturally}.
3040 @node Sequences, Options, Altering, Top
3041 @chapter Canned Sequences of Commands
3043 GDB provides two ways to store sequences of commands for execution as a
3044 unit: user-defined commands and command files.
3047 * Define:: User-defined commands.
3048 * Command Files:: Command files.
3049 * Output:: Controlled output commands useful in
3050 user-defined commands and command files.
3053 @node Define, Command Files, Sequences, Sequences
3054 @section User-Defined Commands
3056 @cindex user-defined command
3057 A @dfn{user-defined command} is a sequence of GDB commands to which you
3058 assign a new name as a command. This is done with the @samp{define}
3062 @item define @var{commandname}
3064 Define a command named @var{commandname}. If there is already a command
3065 by that name, you are asked to confirm that you want to redefine it.
3067 The definition of the command is made up of other GDB command lines,
3068 which are given following the @samp{define} command. The end of these
3069 commands is marked by a line containing @samp{end}.
3071 @item document @var{commandname}
3073 Give documentation to the user-defined command @var{commandname}. The
3074 command @var{commandname} must already be defined. This command reads
3075 lines of documentation just as @samp{define} reads the lines of the
3076 command definition, ending with @samp{end}. After the @samp{document}
3077 command is finished, @samp{help} on command @var{commandname} will print
3078 the documentation you have specified.
3080 You may use the @samp{document} command again to change the
3081 documentation of a command. Redefining the command with @samp{define}
3082 does not change the documentation.
3085 User-defined commands do not take arguments. When they are executed, the
3086 commands of the definition are not printed. An error in any command
3087 stops execution of the user-defined command.
3089 Commands that would ask for confirmation if used interactively proceed
3090 without asking when used inside a user-defined command. Many GDB commands
3091 that normally print messages to say what they are doing omit the messages
3092 when used in user-defined command.
3094 @node Command Files, Output, Define, Sequences
3095 @section Command Files
3097 @cindex command files
3098 A command file for GDB is a file of lines that are GDB commands. Comments
3099 (lines starting with @samp{#}) may also be included. An empty line in a
3100 command file does nothing; it does not mean to repeat the last command, as
3101 it would from the terminal.
3104 @cindex @file{.gdbinit}
3105 When GDB starts, it automatically executes its @dfn{init files}, command
3106 files named @file{.gdbinit}. GDB reads the init file (if any) in your home
3107 directory and then the init file (if any) in the current working
3108 directory. (The init files are not executed if the @samp{-nx} option
3109 is given.) You can also request the execution of a command file with the
3110 @samp{source} command:
3113 @item source @var{filename}
3115 Execute the command file @var{filename}.
3118 The lines in a command file are executed sequentially. They are not
3119 printed as they are executed. An error in any command terminates execution
3120 of the command file.
3122 Commands that would ask for confirmation if used interactively proceed
3123 without asking when used in a command file. Many GDB commands that
3124 normally print messages to say what they are doing omit the messages
3125 when used in a command file.
3127 @node Output,, Command Files, Sequences
3128 @section Commands for Controlled Output
3130 During the execution of a command file or a user-defined command, the only
3131 output that appears is what is explicitly printed by the commands of the
3132 definition. This section describes three commands useful for generating
3133 exactly the output you want.
3136 @item echo @var{text}
3138 @comment I don't consider backslash-space a standard C escape sequence
3139 @comment because it's not in ANSI.
3140 Print @var{text}. Nonprinting characters can be included in @var{text}
3141 using C escape sequences, such as @samp{\n} to print a newline. @b{No
3142 newline will be printed unless you specify one.} In addition to the
3143 standard C escape sequences a backslash followed by a space stands for a
3144 space. This is useful for outputting a string with spaces at the
3145 beginning or the end, since leading and trailing spaces are trimmed from
3146 all arguments. Thus, to print @w{`` and foo = ''}, use the command
3147 @w{``echo \ and foo = \ ''}.
3148 @comment AAARGGG! How am I supposed to do @samp{ and foo = } and not
3149 @comment have the spaces be invisible in TeX?
3151 A backslash at the end of @var{text} can be used, as in C, to continue
3152 the command onto subsequent lines. For example,
3155 echo This is some text\n\
3156 which is continued\n\
3157 onto several lines.\n
3160 produces the same output as
3163 echo This is some text\n
3164 echo which is continued\n
3165 echo onto several lines.\n
3168 @item output @var{expression}
3170 Print the value of @var{expression} and nothing but that value: no
3171 newlines, no @samp{$@var{nn} = }. The value is not entered in the
3172 value history either. @xref{Expressions} for more information on
3175 @item output/@var{fmt} @var{expression}
3176 Print the value of @var{expression} in format @var{fmt}.
3177 @xref{Output formats}, for more information.
3179 @item printf @var{string}, @var{expressions}@dots{}
3181 Print the values of the @var{expressions} under the control of
3182 @var{string}. The @var{expressions} are separated by commas and may
3183 be either numbers or pointers. Their values are printed as specified
3184 by @var{string}, exactly as if the program were to execute
3187 printf (@var{string}, @var{expressions}@dots{});
3190 For example, you can print two values in hex like this:
3193 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
3196 The only backslash-escape sequences that you can use in the string are
3197 the simple ones that consist of backslash followed by a letter.
3200 @node Options, Emacs, Sequences, Top
3201 @chapter Options and Arguments for GDB
3203 When you invoke GDB, you can specify arguments telling it what files to
3204 operate on and what other things to do.
3207 * Mode Options:: Options controlling modes of operation.
3208 * File Options:: Options to specify files (executable, coredump, commands)
3209 * Other Arguments:: Any other arguments without options
3213 @node Mode Options, File Options, Options, Options
3214 @section Mode Options
3218 Do not execute commands from the init files @file{.gdbinit}.
3219 Normally, the commands in these files are executed after all the
3220 command options and arguments have been processed. @xref{Command
3224 ``Quiet''. Do not print the usual introductory messages.
3227 Run in batch mode. Exit with code 0 after processing all the command
3228 files specified with @samp{-x} (and @file{.gdbinit}, if not inhibited).
3229 Exit with nonzero status if an error occurs in executing the GDB
3230 commands in the command files.
3233 This option is used when Emacs runs GDB as a subprocess. It tells GDB
3234 to output the full file name and line number in a standard,
3235 recognizable fashion each time a stack frame is displayed (which
3236 includes each time the program stops). This recognizable format looks
3237 like two @samp{\032} characters, followed by the file name, line number
3238 and character position separated by colons, and a newline. The
3239 Emacs-to-GDB interface program uses the two @samp{\032} characters as
3240 a signal to display the source code for the frame.
3243 @node File Options, Other Arguments, Mode Options, Options
3244 @section File-specifying Options
3246 All the options and command line arguments given are processed
3247 in sequential order. The order makes a difference when the
3248 @samp{-x} option is used.
3252 Read symbol table from file @var{file}.
3255 Use file @var{file} as the executable file to execute when
3256 appropriate, and for examining pure data in conjunction with a core
3259 @item -se @var{file}
3260 Read symbol table from file @var{file} and use it as the executable
3264 Use file @var{file} as a core dump to examine.
3267 Execute GDB commands from file @var{file}.
3269 @item -d @var{directory}
3270 Add @var{directory} to the path to search for source files.
3273 @node Other Arguments,, File Options, Options
3274 @section Other Arguments
3276 If there are arguments to GDB that are not options or associated with
3277 options, the first one specifies the symbol table and executable file name
3278 (as if it were preceded by @samp{-se}) and the second one specifies a core
3279 dump file name (as if it were preceded by @samp{-c}).
3281 @node Emacs, Remote, Options, Top
3282 @chapter Using GDB under GNU Emacs
3284 A special interface allows you to use GNU Emacs to view (and
3285 edit) the source files for the program you are debugging with
3288 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
3289 executable file you want to debug as an argument. This command starts
3290 GDB as a subprocess of Emacs, with input and output through a newly
3291 created Emacs buffer.
3293 Using GDB under Emacs is just like using GDB normally except for two
3298 All ``terminal'' input and output goes through the Emacs buffer. This
3299 applies both to GDB commands and their output, and to the input and
3300 output done by the program you are debugging.
3302 This is useful because it means that you can copy the text of previous
3303 commands and input them again; you can even use parts of the output
3306 All the facilities of Emacs's Shell mode are available for this purpose.
3309 GDB displays source code through Emacs. Each time GDB displays a
3310 stack frame, Emacs automatically finds the source file for that frame
3311 and puts an arrow (@samp{=>}) at the left margin of the current line.
3313 Explicit GDB @samp{list} or search commands still produce output as
3314 usual, but you probably will have no reason to use them.
3317 In the GDB I/O buffer, you can use these special Emacs commands:
3321 Execute to another source line, like the GDB @samp{step} command.
3324 Execute to next source line in this function, skipping all function
3325 calls, like the GDB @samp{next} command.
3328 Execute one instruction, like the GDB @samp{stepi} command.
3331 Execute until exit from the selected stack frame, like the GDB
3332 @samp{finish} command.
3335 @comment C-c C-p in emacs 19
3336 Continue execution of the program, like the GDB @samp{cont} command.
3339 @comment C-c C-u in emacs 19
3340 Go up the number of frames indicated by the numeric argument
3341 (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}),
3342 like the GDB @samp{up} command.@refill
3345 @comment C-c C-d in emacs 19
3346 Go down the number of frames indicated by the numeric argument, like the
3347 GDB @samp{down} command.
3350 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
3351 tells GDB to set a breakpoint on the source line point is on.
3353 The source files displayed in Emacs are in ordinary Emacs buffers
3354 which are visiting the source files in the usual way. You can edit
3355 the files with these buffers if you wish; but keep in mind that GDB
3356 communicates with Emacs in terms of line numbers. If you add or
3357 delete lines from the text, the line numbers that GDB knows will cease
3358 to correspond properly to the code.
3360 @node Remote, Commands, Emacs, Top
3361 @chapter Remote Kernel Debugging
3363 If you are trying to debug a program running on a machine that can't run
3364 GDB in the usual way, it is often useful to use remote debugging. For
3365 example, you might be debugging an operating system kernel, or debugging
3366 a small system which does not have a general purpose operating system
3367 powerful enough to run a full-featured debugger. Currently GDB supports
3368 remote debugging over a serial connection.
3370 The program to be debugged on the remote machine needs to contain a
3371 debugging device driver which talks to GDB over the serial line using the
3372 protocol described below. The same version of GDB that is used ordinarily
3373 can be used for this. Several sample remote debugging drivers are
3374 distributed with GDB; see the @file{README} file in the GDB distribution for
3378 * Remote Commands:: Commands used to start and finish remote debugging.
3381 For details of the communication protocol, see the comments in the GDB
3382 source file @file{remote.c}.
3384 @node Remote Commands,, Remote, Remote
3385 @section Commands for Remote Debugging
3387 To start remote debugging, first run GDB and specify as an executable file
3388 the program that is running in the remote machine. This tells GDB how
3389 to find the program's symbols and the contents of its pure text. Then
3390 establish communication using the @samp{attach} command with a device
3391 name rather than a pid as an argument. For example:
3398 if the serial line is connected to the device named @file{/dev/ttyd}. This
3399 will stop the remote machine if it is not already stopped.
3401 Now you can use all the usual commands to examine and change data and to
3402 step and continue the remote program.
3404 To resume the remote program and stop debugging it, use the @samp{detach}
3408 This material will be merged in when better Readline documentation
3411 @node GDB Readline, History Top ,Readline Top, Command Editing
3412 @subsection GDB Readline
3414 You may control the behavior of command line editing in GDB with the
3420 @itemx set editing on
3421 Enable command line editing (enabled by default).
3423 @item set editing off
3424 Disable command line editing.
3426 @kindex set history file
3427 @item set history file @var{filename}
3428 Set the name of the GDB command history file to @var{filename}. This is
3429 the file from which GDB will read an initial command history
3430 list or to which it will write this list when it exits. This list is
3431 accessed through history expansion or through the history
3432 command editing characters listed below. This file defaults to the
3433 value of the environmental variable @code{GDBHISTFILE}, or to
3434 @code{./.gdb_history} if this variable is not set.
3436 @kindex set history write
3437 @item set history write
3438 @itemx set history write on
3439 Enable the writing of the command history to the command history file
3440 named above. This is enabled by default.
3442 @item set history write off
3443 Disable the writing of the command history to the command history file.
3445 @kindex set history size
3446 @item set history size @var{size}
3447 Set the number of commands which GDB will keep in its history list.
3448 This defaults to the value of the environmental variable
3449 @code{HISTSIZE}, or to 256 if this variable is not set.
3451 @kindex info editing
3453 Display the current settings relating to command line editing, and also
3454 display the last ten commands in the command history.
3456 @item info editing @var{n}
3457 Print ten commands centered on command number @var{n}.
3459 @item info editing +
3460 Print ten commands just after the commands last printed.
3463 @node GDB History, , History Top, Command editing
3464 @comment node-name, next, previous, up
3465 Note that because of the additional meaning of @code{!} to GDB (as the
3466 logical not operator in C), history expansion is off by default. If you
3467 decide to enable history expansion with the @samp{set history expansion
3468 on} command, you will need to follow @samp{!} with a space or a tab to
3469 prevent it from being expanded.
3471 The commands to control history expansion are:
3475 @kindex set history expansion
3476 @item set history expansion on
3477 @itemx set history expansion
3478 Enable history expansion.
3480 @item set history expansion off
3481 Disable history expansion. History expansion is off by default.
3486 @node Commands, Concepts, Remote, Top
3487 @unnumbered Command Index
3491 @node Concepts, , Commands, Top
3492 @unnumbered Concept Index
3502 Occasionally it is useful to execute a shell command from within GDB.
3503 This can be done with the @samp{shell} command.
3506 @item shell @var{shell command string}
3508 @cindex shell escape
3509 Directs GDB to invoke an inferior shell to execute @var{shell command string}.
3510 The environment variable @code{SHELL} is used if it exists, otherwise GDB
3511 uses @samp{/bin/sh}.