2 @setfilename _GDBP__.info
4 @c NOTE: this manual is marked up for preprocessing with a collection
5 @c of m4 macros called "pretex.m4". If you see <_if__> and <_fi__>
6 @c scattered around the source, you have the full source before
7 @c preprocessing; if you don't, you have the source configured for
8 @c _HOST__ architectures (and you can of course get the full source,
9 @c with all configurations, from wherever you got this).
12 THIS IS THE FULL SOURCE. The full source needs to be run through m4
13 before either tex- or info- formatting: for example,
15 m4 pretex.m4 none.m4 m680x0.m4 gdb.texinfo >gdb-680x0.texinfo
17 will produce (assuming your path finds either GNU or SysV m4; Berkeley
18 won't do) a file suitable for formatting. See the text in "pretex.m4"
19 for a fuller explanation (and the macro definitions).
20 To permit maximum flexibility, the full source also does not contain
21 any "info" markup that can be generated automatically; you should first
22 preprocess it as above, then run it through C-u texinfo-master-menu,
23 before actually info-formatting it.
27 @c FOR UPDATES LEADING TO THIS DRAFT, GDB CHANGELOG CONSULTED BETWEEN:
28 @c Tue Feb 26 01:47:07 1991 Cygnus John Gilmore (cygnus at yuba)
29 @c Sat Dec 22 02:51:40 1990 John Gilmore (gnu at cygint)
31 This file documents the GNU debugger _GDBN__.
33 Copyright (C) 1988, 1989, 1990, 1991 Free Software Foundation, Inc.
35 Permission is granted to make and distribute verbatim copies of
36 this manual provided the copyright notice and this permission notice
37 are preserved on all copies.
40 Permission is granted to process this file through Tex and print the
41 results, provided the printed document carries copying permission
42 notice identical to this one except for the removal of this paragraph
43 (this paragraph not being relevant to the printed manual).
46 Permission is granted to copy and distribute modified versions of this
47 manual under the conditions for verbatim copying, provided also that the
48 section entitled ``GNU General Public License'' is included exactly as
49 in the original, and provided that the entire resulting derived work is
50 distributed under the terms of a permission notice identical to this
53 Permission is granted to copy and distribute translations of this manual
54 into another language, under the above conditions for modified versions,
55 except that the section entitled ``GNU General Public License'' may be
56 included in a translation approved by the author instead of in the
60 @setchapternewpage odd
62 @settitle Using _GDBN__ (v4.0)
65 @settitle Using _GDBN__ v4.0 (_HOST__)
72 @subtitle{A Guide to the GNU Source-Level Debugger}
74 @subtitle{On _HOST__ Systems}
77 @c Maybe crank this up to "Fourth Edition" when released at FSF
78 @c @subtitle Third Edition---_GDBN__ version 4.0
79 @subtitle _GDBN__ version 4.0
81 @author{Richard M. Stallman}
82 @author{Roland H. Pesch --- Cygnus Support}
86 \def\$#1${{#1}} % Kluge: collect RCS revision info without $...$
87 \xdef\manvers{\$Revision$} % For use in headers, footers too
89 \hfill Cygnus Support\par
90 \hfill {\it Using _GDBN__}, \manvers\par
91 \hfill \TeX{}info \texinfoversion\par
95 @vskip 0pt plus 1filll
96 Copyright @copyright{} 1988, 1989, 1990, 1991 Free Software Foundation, Inc.
98 Permission is granted to make and distribute verbatim copies of
99 this manual provided the copyright notice and this permission notice
100 are preserved on all copies.
102 Permission is granted to copy and distribute modified versions of this
103 manual under the conditions for verbatim copying, provided also that the
104 section entitled ``GNU General Public License'' is included exactly as
105 in the original, and provided that the entire resulting derived work is
106 distributed under the terms of a permission notice identical to this
109 Permission is granted to copy and distribute translations of this manual
110 into another language, under the above conditions for modified versions,
111 except that the section entitled ``GNU General Public License'' may be
112 included in a translation approved by the author instead of in the
118 @unnumbered Summary of _GDBN__
120 The purpose of a debugger such as _GDBN__ is to allow you to see what is
121 going on ``inside'' another program while it executes---or what another
122 program was doing at the moment it crashed.
124 _GDBN__ can do four main kinds of things (plus other things in support of
125 these) to help you catch bugs in the act:
129 Start your program, specifying anything that might affect its behavior.
132 Make your program stop on specified conditions.
135 Examine what has happened, when your program has stopped.
138 Change things in your program, so you can experiment with correcting the
139 effects of one bug and go on to learn about another.
142 _GDBN__ can be used to debug programs written in C and C++. Pascal support
143 is being implemented, and Fortran support will be added when a GNU
144 Fortran compiler is ready.
146 @node Free Software,,,
147 @unnumberedsec Free Software
148 _GDBN__ is Free Software, protected by the GNU General Public License (GPL).
149 The GPL gives you the freedom to copy or adapt a licensed
150 program---but every person getting a copy also gets with it the
151 freedom to modify that copy (which means that they must get access to
152 the source code), and the freedom to distribute further copies.
153 Typical software companies use copyrights to limit your freedoms; the
154 Free Software Foundation uses the GPL to preserve these freedoms.
156 Fundamentally, the General Public License is a license which says that
157 you have these freedoms and that you can't take these freedoms away
160 For full details, @pxref{License}.
162 @node New Features,,,
163 @unnumbered New Features in _GDBN__ version 4.0
167 TARGETS: Using the new command @samp{target}, you can select at runtime
168 whether you are debugging local files, local processes, standalone
169 systems over the serial port, realtime systems over a TCP/IP
170 connection, etc. _GDBN__ now uses a function vector to mediate access to
171 all the different possible targets, making it much easier to add
172 support for new remote protocols.
175 WATCHPOINTS: _GDBN__ now sports watchpoints as well as breakpoints. You can
176 use a watchpoint to stop execution whenever the value of an expression
177 changes, without having to predict a particular place in the inferior
178 process where this may happen.
181 OBJECT CODE FORMATS: _GDBN__ uses a new scheme called Binary File
182 Descriptors (BFD) to permit it to switch dynamically, without
183 reconfiguration or recompilation, between different object-file
184 formats. Formats currently supported are COFF, a.out, and the new
185 Intel 960 b.out; files may be read as .o's, archive libraries, or core
186 dumps. BFD is available as a subroutine library so that other
187 programs may take advantage of it, and the other GNU binary utilities
188 are being converted to use it.
191 CONFIGURATION: You must still choose a particular machine architecture
192 and operating system for _GDBN__'s host and target systems when _GDBN__ is built.
193 The script @samp{config.gdb} now handles specification of separate host
194 and target configurations.
197 INTERACTION: _GDBN__ now uses the GNU readline interface to read its
198 input; this provides inline editing of commands, using the familiar
199 Emacs or @code{vi} keymaps, and command-history support. The user interface
200 to _GDBN__'s control variables has been simplified and consolidated in two
201 commands, @samp{set} and @samp{show}. Output lines are now broken at
202 readable places, rather than overflowing onto the next line.
205 SOURCE LANGUAGE: _GDBN__ now understands C++ source as well as C. Multiple
206 inheritance is supported when used with G++ 2.0. There is also limited
207 support for C++ exception handling: _GDBN__ can break when an exception is
208 raised, before the stack is peeled back to the exception handler's
209 context. You can suppress output of machine-level addresses,
210 displaying only source language information.
213 PORTS: _GDBN__ has been ported to the following new architectures:
214 AT&T 3b1, Acorn RISC machine, HP300 running HPUX, big- and little-
215 endian MIPS machines, Motorola 88k, Sun 386i, and Sun 3 running SunOS
216 4. In addition, the following are supported as targets only: AMD
217 29k, Intel 960, and Wind River's VxWorks.
220 SHARED LIBRARIES: _GDBN__ 4.0 supports SunOS shared libraries.
223 WORK IN PROGRESS: kernel debugging for BSD and Mach systems; Tahoe and
224 HPPA architecture support.
228 @node Sample Session,,,
229 @chapter A Sample _GDBN__ Session
231 You can use this manual at your leisure to read all about _GDBN__.
232 However, a handful of commands are enough to get started using the
233 debugger. This chapter illustrates these commands.
235 In this sample session, we emphasize user input like this: @var{input},
236 to make it easier to pick out from the surrounding output.
238 @c FIXME: this example may not be appropriate for some configs, where
239 @c FIXME...primary interest is in remote use.
240 We'll be using _GDBN__ to inspect GNU @code{m4} (a generic macro
244 $ @var{cd gm4/common}
247 Reading symbol data from m4...done.
248 (_GDBP__) set width 70
252 _GDBN__ only reads enough symbol data to know where to find the rest
253 when needed; as a result, the first prompt comes up very quickly. We
254 immediately told _GDBN__ to use a narrower display width than usual, so
255 that examples will fit in this manual.
258 (_GDBP__) @var{break m4_changequote}
259 Breakpoint 1 at 0x59d4: file builtin.c, line 812.
263 We've chosen to see how the @code{m4} builtin @samp{changequote} works.
264 Having looked at the source, we knew the relevant subroutine is
265 @samp{m4_changequote}. We've set a breakpoint there with _GDBN__'s
266 @code{break} command.
270 Starting program: /s1/gnu/src/gm4/common/m4
271 @var{`usual' quotes <not these>}
272 usual quotes <not these>
276 Using the @samp{run} command, we've started @code{m4} running under
277 _GDBN__ control; while we don't touch the @samp{m4_changequote}
278 subroutine, the program runs as usual---it filters standard input.
281 @var{changequote(<,>)}
283 Breakpoint 1, m4_changequote (argc=3, argv=0x2b958) at builtin.c:812
284 812 if (bad_argc(TOKEN_DATA_TEXT(argv[0]), argc, 1, 3))
287 To trigger the breakpoint, we called @code{changequote}. _GDBN__
288 suspended execution of @code{m4}, displaying information about the
289 context where it stopped.
294 bad_argc (name=0xf851cfb4<Address 0xf851cfb4 out of bounds>, argc=3,
295 min=1, max=3) at builtin.c:230
296 230 if (min > 0 && argc < min) @{
299 We've used the command @samp{s} (@code{step}) to advance execution to
300 the next source line; since there was a subroutine call, we've stopped
301 in the first line of that subroutine, not in the next line of
302 @code{m4_changequote}.
305 The command @samp{next} would have taken us to the next line of
306 @code{m4_changequote}. To see where we are in the stack, we can use the
307 @samp{backtrace} command (which can also be spelled @samp{bt}).
311 #0 bad_argc (name=0xf851cfb4<Address 0xf851cfb4 out of bounds>,
312 argc=3, min=1, max=3) at builtin.c:230
313 #1 0x59ec in m4_changequote (argc=3, argv=0x2b958) at builtin.c:812
314 #2 0x6e38 in expand_macro (sym=0x2b060) at macro.c:242
315 #3 0x6840 in expand_token (obs=0x0, t=176224, td=0xf7fffb08)
317 #4 0x6794 in expand_input () at macro.c:40
318 #5 0x28dc in main (argc=0, argv=0xf7fffbf8) at m4.c:174
322 We'll tell _GDBN__ to finish execution of this subroutine, to get back
323 to @code{m4_changequote}.
326 (_GDBP__) @var{finish}
327 Run till exit from #0 bad_argc (name=0xf851cfb4<Address 0xf851cfb4 out
329 argc=3, min=1, max=3) at builtin.c:230
330 0x59ec in m4_changequote (argc=3, argv=0x2b958) at builtin.c:812
331 812 if (bad_argc(TOKEN_DATA_TEXT(argv[0]), argc, 1, 3))
332 Value returned is $1 = false
336 We're still in the middle of the first line of @code{m4_changequote};
337 @samp{finish} just allowed the subroutine call to complete.
338 The display beginning ``@code{0x59ec in}@dots'', preceding the
339 display of line @code{812}, is a reminder of that situation from
342 Now that we're past the subroutine call, using the @code{step} command
343 again will get us to the next line of @code{m4_changequote}.
347 815 lquote = (argc >= 2) ? TOKEN_DATA_TEXT(argv[1])[0] : DEF_LQUOTE;
348 (_GDBP__) @var{whatis lquote}
350 (_GDBP__) @var{p lquote}
355 When we stepped to another line, @code{m4} was about to set a variable
356 @samp{lquote}; we inspected its type with @samp{whatis} and its value
357 with @samp{p} (the @samp{print} command). We can see some context by
358 displaying the surrounding source code, with the @samp{l} (@code{list})
363 810 token_data **argv;
365 812 if (bad_argc(TOKEN_DATA_TEXT(argv[0]), argc, 1, 3))
368 815 lquote = (argc >= 2) ? TOKEN_DATA_TEXT(argv[1])[0] : DEF_LQUOTE;
369 816 rquote = (argc >= 3) ? TOKEN_DATA_TEXT(argv[2])[0] : DEF_RQUOTE;
374 816 rquote = (argc >= 3) ? TOKEN_DATA_TEXT(argv[2])[0] : DEF_RQUOTE;
377 (_GDBP__) @var{p lquote}
379 (_GDBP__) @var{p rquote}
384 We proceeded past another line with @samp{s}, and inspected the new
385 values of @code{m4}'s internal variables @code{rquote} and
388 Since we're done with our inspection of this subroutine, we'll tell
389 _GDBN__ to allow @code{m4} to continue running, with the @samp{c}
390 (@code{continue}) command:
396 @var{`usual' quotes <not these>}
397 `usual' quotes not these
399 Program exited normally.
403 _1__@end smallexample
406 Finally, when we ended the @code{m4} run, _GDBN__ told us
407 ``@code{Program exited normally.}'' We ended our _GDBN__ session with
408 the _GDBN__ @samp{quit} command.
412 @chapter Getting In and Out of _GDBN__
414 @node Starting _GDBN__,,,
415 @section Starting _GDBN__
417 _GDBN__ is invoked with the shell command @samp{_GDBP__}. Once started,
418 it reads commands from the terminal until you tell it to exit.
420 You can start by just calling @samp{_GDBP__} with no arguments or
421 options; but the most usual way to start _GDBN__ is with one argument or
422 two, specifying an executable program as the argument:
427 You can also start with both an executable program and a core file specified:
432 You can further control how _GDBN__ starts up by using command-line
435 _GDBN__ itself can remind you of the options available:
440 will display all available options and briefly describe their use
441 (@samp{gdb -h} is a shorter equivalent).
443 All options and command line arguments you give are processed
444 in sequential order. The order makes a difference when the
445 @samp{-x} option is used.
447 @node File Options,,,
448 @subsection Choosing Files
450 As shown above, any arguments other than options specify an executable
451 file and core file; that is, the first argument encountered with no
452 associated option flag is equivalent to a @samp{-se} option, and the
453 second, if any, is equivalent to a @samp{-c} option. Many options have
454 both long and short forms; both are shown here.
457 @item -symbols @var{file}
459 Read symbol table from file @var{file}.
461 @item -exec @var{file}
463 Use file @var{file} as the executable file to execute when
464 appropriate, and for examining pure data in conjunction with a core
468 Read symbol table from file @var{file} and use it as the executable
471 @item -core @var{file}
473 Use file @var{file} as a core dump to examine.
475 @item -command @var{file}
477 Execute _GDBN__ commands from file @var{file}. @xref{Command Files}.
479 @item -directory @var{directory}
480 @itemx -d @var{directory}
481 Add @var{directory} to the path to search for source files.
484 @node Mode Options,,,
485 @subsection Choosing Modes
490 Do not execute commands from any @file{_GDBINIT__} initialization files.
491 Normally, the commands in these files are executed after all the
492 command options and arguments have been processed. @xref{Command
497 ``Quiet''. Do not print the introductory and copyright messages. These
498 messages are also suppressed in batch mode, or if an executable file name is
499 specified on the _GDBN__ command line.
502 Run in batch mode. Exit with code @code{0} after processing all the command
503 files specified with @samp{-x} (and @file{_GDBINIT__}, if not inhibited).
504 Exit with nonzero status if an error occurs in executing the _GDBN__
505 commands in the command files.
507 Batch mode may be useful for running _GDBN__ as a filter, for example to
508 download and run a program on another computer; in order to make this
509 more useful, the message
511 Program exited normally.
514 (which is ordinarily issued whenever a program running under _GDBN__ control
515 terminates) is not issued when running in batch mode.
517 @item -cd @var{directory}
518 Run _GDBN__ using @var{directory} as its working directory,
519 instead of the current directory.
523 This option is used when Emacs runs _GDBN__ as a subprocess. It tells _GDBN__
524 to output the full file name and line number in a standard,
525 recognizable fashion each time a stack frame is displayed (which
526 includes each time the program stops). This recognizable format looks
527 like two @samp{\032} characters, followed by the file name, line number
528 and character position separated by colons, and a newline. The
529 Emacs-to-_GDBN__ interface program uses the two @samp{\032} characters as
530 a signal to display the source code for the frame.
533 Set the line speed (baud rate or bps) of any serial interface used by
534 _GDBN__ for remote debugging.
536 @item -tty @var{device}
537 Run using @code{device} for your program's standard input and output.
541 @node i960-Nindy Remote,,,
542 @subsection _GDBN__ with a Remote i960 (Nindy)
544 ``Nindy'' is the name of a ROM Monitor program for Intel 960 target
545 systems. When _GDBN__ is configured to control a remote Intel 960 using
546 Nindy, you can tell _GDBN__ how to connect to the 960 in several ways:
550 Through command line options specifying device, baud rate, and protocol;
553 By responding to a prompt on startup;
556 By using the @samp{target} command at any point during your _GDBN__ session.
559 @node Nindy Startup,,,
560 @subsubsection Startup with Nindy
562 The command-line options for Nindy are detailed below. If you simply
563 start @code{_GDBP__} without using options to specify a serial port, you are
564 prompted for it, @emph{before} you reach the ordinary _GDBN__ prompt:
566 Attach /dev/ttyNN -- specify NN, or "quit" to quit:
569 You can, if you choose, simply start up with no Nindy connection by
570 responding to the prompt with an empty line. If you do this, and later
571 wish to attach to Nindy, use @samp{target} (@pxref{Target Commands}).
573 @node Nindy Options,,,
574 @subsubsection Options for Nindy
576 These are the startup options for beginning your _GDBN__ session with a
577 Nindy-960 board attached:
581 Specify the serial port name of a serial interface to be used to connect
582 to the target system. This option is only available when _GDBN__ is
583 configured for the Intel 960 target architecture. You may specify
584 @var{port} as any of: a full pathname (e.g. @samp{-r /dev/ttya}), a
585 device name in @samp{/dev} (e.g. @samp{-r ttya}), or simply the unique
586 suffix for a specific @code{tty} (e.g. @samp{-r a}).
589 (An uppercase letter ``O'', not a zero.) Specify that _GDBN__ should use
590 the ``old'' Nindy monitor protocol to connect to the target system.
591 This option is only available when _GDBN__ is configured for the Intel 960
595 @emph{Warning:} if you specify @samp{-O}, but are actually trying to
596 connect to a target system that expects the newer protocol, the connection
597 will fail, appearing to be a speed mismatch. _GDBN__ will repeatedly
598 attempt to reconnect at several different line speeds. You can abort
599 this process with an interrupt.
603 Specify that _GDBN__ should first send a @samp{BREAK} signal to the target
604 system, in an attempt to reset it, before connecting to a Nindy target.
607 @emph{Warning:} Many target systems do not have the hardware that this
608 requires; it only works with a few boards.
613 The standard @samp{-b} option controls the line speed used on the serial
618 @subsubsection Nindy Reset Command
622 For a Nindy target, this command sends a ``break'' to the remote target
623 system; this is only useful if the target has been equipped with a
624 circuit to perform a hard reset (or some other interesting action) when
631 @node EB29K Remote,,,
632 @subsection _GDBN__ with a Remote EB29K
635 @cindex running 29K programs
638 To use _GDBN__ from a Unix system to run programs on AMD's EB29K
639 board in a PC, you must first connect a serial cable between the PC
640 and a serial port on the Unix system. In the following, we assume
641 you've hooked the cable between the PC's @samp{COM1} port and
642 @samp{/dev/ttya} on the Unix system.
644 @node Comms (EB29K),,,
645 @subsubsection Communications Setup
646 The next step is to set up the PC's port, by doing something like the
647 following in DOS on the PC:
649 C:\> MODE com1:9600,n,8,1,none
652 This example---run on an MS DOS 4.0 system---sets the PC port to 9600
653 bps, no parity, eight data bits, one stop bit, and no ``retry'' action;
654 you must match the communications parameters when establishing the Unix
655 end of the connection as well.
656 @c FIXME: Who knows what this "no retry action" crud from the DOS manual may
657 @c mean? It's optional; leave it out? ---pesch@cygnus.com, 25feb91
659 To give control of the PC to the Unix side of the serial line, type
660 the following at the DOS console:
665 (Later, if you wish to return control to the DOS console, you can use
666 the command @samp{CTTY con}---but you must send it over the device that
667 had control, in our example over the @samp{COM1} serial line).
669 From the Unix host, use a communications program such as @code{tip} or
670 @code{cu} to communicate with the PC; for example,
672 cu -s 9600 -l /dev/ttya
675 The @code{cu} options shown specify, respectively, the linespeed and the
676 serial port to use. If you use @code{tip} instead, your command line
677 may look something like the following instead:
682 Your system may define a different name where our example uses
683 @samp{/dev/ttya} as the argument to @code{tip}. The communications
684 parameters, including what port to use, are associated with the
685 @code{tip} argument in the ``remote'' descriptions file---normally the
686 system table @file{/etc/remote}.
687 @c FIXME: What if anything needs doing to match the "n,8,1,none" part of
688 @c the DOS side's comms setup? cu can support -o (odd
689 @c parity), -e (even parity)---apparently no settings for no parity or
690 @c for character size. Taken from stty maybe...? John points out tip
691 @c can set these as internal variables, eg ~s parity=none; man stty
692 @c suggests that it *might* work to stty these options with stdin or
693 @c stdout redirected... ---pesch@cygnus.com, 25feb91
696 Using the @samp{tip} or @samp{cu} connection, change the DOS working
697 directory to the directory containing a copy of your 29K program, then
698 start the PC program @samp{EBMON} (an EB29K control program supplied
699 with your board by AMD). You should see an initial display from
700 @code{EBMON} similar to the one that follows, ending with the
701 @code{EBMON} prompt @samp{#}---
705 G:\> CD \usr\joe\work29k
707 G:\USR\JOE\WORK29K> EBMON
708 Am29000 PC Coprocessor Board Monitor, version 3.0-18
709 Copyright 1990 Advanced Micro Devices, Inc.
710 Written by Gibbons and Associates, Inc.
712 Enter '?' or 'H' for help
714 PC Coprocessor Type = EB29K
716 Memory Base = 0xd0000
718 Data Memory Size = 2048KB
719 Available I-RAM Range = 0x8000 to 0x1fffff
720 Available D-RAM Range = 0x80002000 to 0x801fffff
723 Register Stack Size = 0x800
724 Memory Stack Size = 0x1800
727 Am29027 Available = No
728 Byte Write Available = Yes
733 Then exit the @code{cu} or @code{tip} program (done in the example by
734 typing @code{~.} at the @code{EBMON} prompt). @code{EBMON} will keep
735 running, ready for _GDBN__ to take over.
737 For this example, we've assumed what is probably the most convenient
738 way to make sure the same 29K program is on both the PC and the Unix
739 system: a PC/NFS connection that establishes ``drive @code{G:}'' on the
740 PC as a file system on the Unix host. If you don't have PC/NFS or
741 something similar connecting the two systems, you must arrange some
742 other way---perhaps floppy-disk transfer---of getting the 29K program
743 from the Unix system to the PC; _GDBN__ will @emph{not} download it over the
746 @node _GDBP__-EB29K,,,
747 @subsubsection EB29K cross-debugging
748 Finally, @code{cd} to the directory containing an image of your 29K
749 program on the Unix system, and start _GDBN__---specifying as argument the
750 name of your 29K program:
755 Now you can use the @code{target} command:
757 target amd-eb /dev/ttya 9600 MYFOO
759 @c FIXME: test above 'target amd-eb' as spelled, with caps! caps are meant to
760 @c emphasize that this is the name as seen by DOS (since I think DOS is
761 @c single-minded about case of letters). ---pesch@cygnus.com, 25feb91
764 In this example, we've assumed your program is in a file called
765 @samp{myfoo}. Note that the filename given as the last argument to
766 @samp{target amd-eb} should be the name of the program as it appears to DOS.
767 In our example this is simply @samp{MYFOO}, but in general it can include
768 a DOS path, and depending on your transfer mechanism may not resemble
769 the name on the Unix side.
771 At this point, you can set any breakpoints you wish; when you're ready
772 to see your program run on the 29K board, use the _GDBN__ command
777 To stop debugging the remote program, use the _GDBN__ @samp{detach}
780 To return control of the PC to its console, use @code{tip} or @code{cu}
781 once again, after your _GDBN__ session has concluded, to attach to
782 @code{EBMON}. You can then type the command @samp{q} to shut down
783 @code{EBMON}, returning control to the DOS command-line interpreter.
784 Type @samp{CTTY con} to return command input to the main DOS console,
785 and type @samp{~.} to leave @code{tip} or @code{cu}.
788 @subsubsection Remote Log
790 @cindex log file for EB29K
791 The @samp{target amd-eb} command creates a file @file{eb.log} in the
792 current working directory, to help debug problems with the connection.
793 @file{eb.log} records all the output from @code{EBMON}, including echoes
794 of the commands sent to it. Running @samp{tail -f} on this file in
795 another window often helps to understand trouble with @code{EBMON}, or
796 unexpected events on the PC side of the connection.
800 @node VxWorks Remote,,,
801 @subsection _GDBN__ and VxWorks
802 _GDBN__ enables developers to spawn and debug tasks running on networked
803 VxWorks targets from a Unix host. Already-running tasks spawned from
804 the VxWorks shell can also be debugged. _GDBN__ uses code that runs on
805 both the UNIX host and on the VxWorks target. The program
806 @code{_GDBP__} is installed and executed on the UNIX host.
808 The remote debugging interface (RDB) routines are installed and executed
809 on the VxWorks target. These routines are included in the VxWorks library
810 @code{rdb.a} and are incorporated into the system image when source-level
811 debugging is enabled in the VxWorks configuration.
813 Defining @code{INCLUDE_RDB} in the VxWorks configuration file
814 @code{configAll.h} includes the RDB interface routines and spawns the
815 source debugging task @code{tRdbTask} when VxWorks is booted. For more
816 information on configuring and remaking VxWorks, see the @cite{VxWorks
819 Once you have included the RDB interface in your VxWorks system image
820 and set your Unix execution search path to find _GDBN__, you are ready
821 to run _GDBN__. From your UNIX host, type:
827 _GDBN__ will come up showing the prompt:
833 @node VxWorks connection,,,
834 @subsubsection Connecting to VxWorks
836 The _GDBN__ command @samp{target} lets you connect to a VxWorks target on the
837 network. To connect to a target whose host name is ``@code{tt}'', type:
840 (_GDBP__) target vxworks tt
843 _GDBN__ will display a message similar to the following:
846 Attaching remote machine across net... Success!
849 _GDBN__ will then attempt to read the symbol tables of any object
850 modules loaded into the VxWorks target since it was last booted.
851 _GDBN__ locates these files by searching the directories listed in the
852 command search path (@pxref{Command Path}); if it fails to find an
853 object file, it will display a message such as:
856 prog.o: No such file or directory.
859 This will cause the @samp{target} command to abort. When this happens,
860 you should add the appropriate directory to the search path, with the
861 _GDBN__ command @samp{path}, and execute the @samp{target} command
864 @node VxWorks download,,,
865 @subsubsection VxWorks Download
867 If you have connected to the VxWorks target and you want to debug an
868 object that has not yet been loaded, you can use the _GDBN__ @samp{load}
869 command to download a file from UNIX to VxWorks incrementally. The
870 object file given as an argument to the @samp{load} command is actually
871 opened twice: first by the VxWorks target in order to download the code,
872 then by _GDBN__ in order to read the symbol table. This can lead to
873 problems if the current working directories on the two systems differ.
874 It is simplest to set the working directory on both systems to the
875 directory in which the object file resides, and then to reference the
876 file by its name, without any path. Thus, to load a program
877 @samp{prog.o}, residing in @code{wherever/vw/demo/rdb}, on VxWorks type:
880 -> cd "wherever/vw/demo/rdb"
886 (_GDBP__) cd wherever/vw/demo/rdb
887 (_GDBP__) load prog.o
890 _GDBN__ will display a response similar to the following:
893 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
896 You can also use the @samp{load} command to reload an object module
897 after editing and recompiling the corresponding source file. Note that
898 this will cause _GDBN__ to delete all currently-defined breakpoints,
899 auto-displays, and convenience variables, and to clear the value
900 history. (This is necessary in order to preserve the integrity of
901 debugger data structures that reference the target system's symbol
904 @node VxWorks attach,,,
905 @subsubsection Running Tasks
907 You can also attach to an existing task using the @samp{attach} command as
911 (_GDBP__) attach @var{taskId}
914 where @var{taskId} is the VxWorks hexadecimal task ID. The task can be running
915 or suspended when you attach to it. If running, it will be suspended at
916 the time of attachment.
920 @node Leaving _GDBN__,,,
921 @section Leaving _GDBN__
922 @cindex exiting _GDBN__
924 To exit _GDBN__, use the @samp{quit} command (abbreviated @samp{q}), or type
925 an end-of-file character (usually @kbd{C-d}).
928 An interrupt (often @kbd{C-c}) will not exit from _GDBN__, but rather
929 will terminate the action of any _GDBN__ command that is in progress and
930 return to _GDBN__ command level. It is safe to type the interrupt
931 character at any time because _GDBN__ does not allow it to take effect
932 until a time when it is safe.
934 If you've been using _GDBN__ to control an attached process or device,
935 you can release it with the @samp{detach} command; @pxref{Attach}.
937 @node Shell Commands,,,
938 @section Shell Commands
939 If you just need to execute occasional shell commands during your
940 debugging session, there's no need to leave or suspend _GDBN__; you can
941 just use the @samp{shell} command.
944 @item shell @var{command string}
947 Directs _GDBN__ to invoke an inferior shell to execute @var{command string}.
948 The environment variable @code{SHELL} is used if it exists, otherwise _GDBN__
952 The utility @samp{make} is often needed in development environments.
953 You don't have to use the @samp{shell} command for this purpose in _GDBN__:
956 @item make @var{make-args}
959 Causes _GDBN__ to execute an inferior @code{make} program with the specified
960 arguments. This is equivalent to @samp{shell make @var{make-args}}.
964 @chapter _GDBN__ Commands
966 @node Command Syntax,,,
967 @section Command Syntax
968 A _GDBN__ command is a single line of input. There is no limit on how long
969 it can be. It starts with a command name, which is followed by arguments
970 whose meaning depends on the command name. For example, the command
971 @samp{step} accepts an argument which is the number of times to step,
972 as in @samp{step 5}. You can also use the @samp{step} command with
973 no arguments. Some command names do not allow any arguments.
976 _GDBN__ command names may always be truncated if that abbreviation is
977 unambiguous. Other possible command abbreviations are listed in the
978 documentation for individual commands. Sometimes even ambiguous
979 abbreviations are allowed; for example, @samp{s} is specially defined as
980 equivalent to @samp{step} even though there are other commands whose
981 names start with @samp{s}.
983 @cindex repeating commands
984 A blank line as input to _GDBN__ means to repeat the previous command.
985 Certain commands (for example, @samp{run}) will not repeat this way;
986 these are commands for which unintentional repetition might cause
987 trouble and which you are unlikely to want to repeat.
989 The @samp{list} and @samp{x} commands construct new arguments when
990 repeated, rather than repeating exactly as typed, to permit easy
991 scanning of source or memory.
995 A line of input starting with @samp{#} is a comment; it does nothing.
996 This is useful mainly in command files (@xref{Command Files}).
999 @section Getting Help
1000 @cindex online documentation
1002 You can always ask _GDBN__ itself for information on its commands, using the
1003 command @samp{help}.
1007 Used with no arguments, @samp{help} displays a short list of named
1008 categories of commands:
1011 List of classes of commands:
1013 running -- Running the program
1014 stack -- Examining the stack
1015 data -- Examining data
1016 breakpoints -- Making program stop at certain points
1017 files -- Specifying and examining files
1018 status -- Status inquiries
1019 support -- Support facilities
1020 user-defined -- User-defined commands
1021 aliases -- Aliases of other commands
1022 obscure -- Obscure features
1024 Type "help" followed by a class name for a list of commands in that class.
1025 Type "help" followed by command name for full documentation.
1026 Command name abbreviations are allowed if unambiguous.
1030 @item help @var{category}
1031 Using one of the general help categories as an argument, you can get a
1032 list of the individual commands in a category. For example, here is the
1033 help display for category @samp{status}:
1035 (_GDBP__) help status
1040 show -- Generic command for showing things set with "set"
1041 info -- Generic command for printing status
1043 Type "help" followed by command name for full documentation.
1044 Command name abbreviations are allowed if unambiguous.
1048 @item help @var{command}
1049 With a command name as @samp{help} argument, _GDBN__ will display a
1050 short paragraph on how to use that command.
1053 In addition to @samp{help}, you can use the _GDBN__ commands @samp{info}
1054 and @samp{show} to inquire about the state of your program, or the state
1055 of _GDBN__ itself. Both commands support many ``sub-commands'', or
1056 topics of inquiry; this manual introduces each of them in the
1057 appropriate context. The listings under ``@code{info}'' and under
1058 ``@code{show}'' in the Index point to all the sub-commands
1064 This command is for describing the state of your program; for example,
1065 it can list the arguments given to your program (@samp{info args}), the
1066 registers currently in use (@samp{info registers}), or the breakpoints
1067 you've set (@samp{info breakpoints}). You can get a complete list of
1068 the @code{info} sub-commands with @samp{help info}.
1072 In contrast, @samp{show} is for describing the state of _GDBN__ itself.
1073 You can change most of the things you can @code{show}, by using the
1074 related command @samp{set}; for example, you can control what number
1075 system is used for displays with @samp{set radix}, or simply inquire
1076 which is currently in use with @samp{show radix}.
1079 To display all the settable parameters and their current
1080 values, you can use @samp{show} with no arguments; you may also use
1081 @samp{info set}. Both commands produce the same display.
1082 @c FIXME: "info set" violates the rule that "info" is for state of
1083 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1084 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1087 Here are three miscellaneous @samp{show} subcommands, all of which are
1088 exceptional in lacking corresponding @samp{set} commands:
1091 @kindex show version
1093 Show what version of _GDBN__ is running. You should include this
1094 information in _GDBN__ bug-reports. If multiple versions of _GDBN__ are
1095 in use at your site, you may occasionally want to make sure what version
1096 of _GDBN__ you're running; as _GDBN__ evolves, new commands are
1097 introduced, and old ones may wither away. The version number is also
1098 announced when you start _GDBN__ with no arguments.
1100 @kindex show copying
1102 Display information about permission for copying _GDBN__.
1104 @kindex show warranty
1106 Display the GNU ``NO WARRANTY'' statement.
1110 @chapter Running Programs Under _GDBN__
1112 @node Compilation,,,
1113 @section Compiling for Debugging
1115 In order to debug a program most effectively, you need to generate
1116 debugging information when you compile it. This debugging information
1117 is stored in the object file; it describes the data type of each
1118 variable or function and the correspondence between source line numbers
1119 and addresses in the executable code.
1121 To request debugging information, specify the @samp{-g} option when you run
1124 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1125 options together. Using those compilers, you cannot generate optimized
1126 executables containing debugging information.
1128 The GNU C compiler supports @samp{-g} with or without @samp{-O}, making it
1129 possible to debug optimized code. We recommend that you @emph{always} use
1130 @samp{-g} whenever you compile a program. You may think the program is
1131 correct, but there's no sense in pushing your luck.
1133 Some things do not work as well with @samp{-g -O} as with just
1134 @samp{-g}, particularly on machines with instruction scheduling. If in
1135 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1136 please report it as a bug (including a test case!).
1138 Older versions of the GNU C compiler permitted a variant option
1139 @samp{-gg} for debugging information. _GDBN__ no longer supports this
1140 format; if your GNU C compiler has this option, do not use it.
1143 @comment As far as I know, there are no cases in which _GDBN__ will
1144 @comment produce strange output in this case. (but no promises).
1145 If your program includes archives made with the @code{ar} program, and
1146 if the object files used as input to @code{ar} were compiled without the
1147 @samp{-g} option and have names longer than 15 characters, _GDBN__ will get
1148 confused reading the program's symbol table. No error message will be
1149 given, but _GDBN__ may behave strangely. The reason for this problem is a
1150 deficiency in the Unix archive file format, which cannot represent file
1151 names longer than 15 characters.
1153 To avoid this problem, compile the archive members with the @samp{-g}
1154 option or use shorter file names. Alternatively, use a version of GNU
1155 @code{ar} dated more recently than August 1989.
1160 @section Starting your Program
1164 To start your program under _GDBN__, use the @samp{run} command.
1166 Except on VxWorks, you
1171 must first specify the program name with an argument to _GDBN__
1172 (@pxref{Invocation}), or using the @samp{file} or @samp{exec-file}
1173 command (@pxref{Files}).@refill
1175 On targets that support processes, @samp{run} creates an inferior
1176 process and makes that process run your program. On other targets,
1177 @samp{run} jumps to the start of the program.
1179 The execution of a program is affected by certain information it
1180 receives from its superior. _GDBN__ provides ways to specify this
1181 information, which you must do @i{before} starting the program. (You
1182 can change it after starting the program, but such changes will only affect
1183 the program the next time you start it.) This information may be
1184 divided into four categories:
1187 @item The @i{arguments.}
1188 You specify the arguments to give your program as the arguments of the
1189 @samp{run} command. If a shell is available on your target, the shell
1190 is used to pass the arguments, so that you may use normal conventions
1191 (for example regular expression expansion or variable substitution) in
1192 describing the arguments. In Unix systems, you can control which shell
1193 is used with the @code{SHELL} environment variable.
1195 @item The @i{environment.}
1196 Your program normally inherits its environment from _GDBN__, but you can
1197 use the _GDBN__ commands @samp{set environment} and @samp{unset
1198 environment} to change parts of the environment that will be given to
1201 @item The @i{working directory.}
1202 Your program inherits its working directory from _GDBN__. You can set
1203 _GDBN__'s working directory with the @samp{cd} command in _GDBN__.
1205 @item The @i{standard input and output.}
1206 Your program normally uses the same device for standard input and
1207 standard output as _GDBN__ is using. You can redirect input and output
1208 in the @code{run} command line, or you can use the @samp{tty} command to
1209 set a different device for your program.
1212 When you issue the @samp{run} command, your program begins to execute
1213 immediately. @xref{Stopping}, for discussion of how to arrange for your
1216 Note that once your program has been started by the @samp{run} command,
1217 you may evaluate expressions that involve calls to functions in the
1218 inferior, using the @samp{print} or @samp{call} commands. @xref{Data}.
1220 If the modification time of your symbol file has changed since the last
1221 time _GDBN__ read its symbols, _GDBN__ will discard its symbol table and re-read
1222 it. In this process, it tries to retain your current breakpoints.
1225 @section Your Program's Arguments
1227 @cindex arguments (to your program)
1228 The arguments to your program can be specified by the arguments of the
1229 @samp{run} command. They are passed to a shell, which expands wildcard
1230 characters and performs redirection of I/O, and thence to the program.
1231 _GDBN__ uses the shell indicated by your environment variable
1232 @code{SHELL} if it exists; otherwise, _GDBN__ uses @samp{/bin/sh}.
1234 @samp{run} with no arguments uses the same arguments used by the previous
1235 @samp{run}, or those set by the @samp{set args} command.
1240 Specify the arguments to be used the next time your program is run. If
1241 @samp{set args} has no arguments, @samp{run} will execute your program
1242 with no arguments. Once you have run your program with arguments, this
1243 is the only way to run it again without arguments.
1247 Show the arguments to give your program when it is started.
1250 @node Environment,,,
1251 @section Your Program's Environment
1253 @cindex environment (of your program)
1254 The @dfn{environment} consists of a set of environment variables and
1255 their values. Environment variables conventionally record such things as
1256 your user name, your home directory, your terminal type, and your search
1257 path for programs to run. Usually you set up environment variables with
1258 the shell and they are inherited by all the other programs you run. When
1259 debugging, it can be useful to try running the program with a modified
1260 environment without having to start _GDBN__ over again.
1263 @item path @var{directory}
1265 Add @var{directory} to the front of the @code{PATH} environment variable
1266 (the search path for executables), for both _GDBN__ and your program.
1267 You may specify several directory names, separated by @samp{:} or
1268 whitespace. If a directory is already in the path, it is moved to the
1269 front, so it will be searched sooner. You can use the string
1270 @samp{$cwd} to refer to the current working directory.
1272 This command will not repeat if you press @key{RET} a second time after
1275 @item show environment @var{varname}
1276 @kindex show environment
1277 Print the value of environment variable @var{varname} to be given to
1278 your program when it starts.
1280 @item show environment
1281 Print the names and values of all environment variables to be given to
1284 @item set environment @var{varname} @var{value}
1285 @itemx set environment @var{varname} = @var{value}
1286 @kindex set environment
1287 Sets environment variable @var{varname} to @var{value}. The value
1288 changes for your program only, not for _GDBN__ itself. @var{value} may
1289 be any string; the values of environment variables are just strings, and
1290 any interpretation is supplied by your program itself. The @var{value}
1291 parameter is optional; if it is eliminated, the variable is set to a
1293 @c FIXME: I think "any string" here doesn't include leading, trailing
1294 @c FIXME... blanks. Queried J Gilmore. ---pesch@cygnus.com, 4apr91
1296 For example, this command:
1303 tells a Unix program, when subsequently run, that its user is named
1306 @item unset environment @var{varname}
1307 @kindex unset environment
1308 Remove variable @var{varname} from the environment to be passed to your
1309 program. This is different from @samp{set env @var{varname}=};
1310 @samp{unset environment} removes the variable from the environment,
1311 rather than assigning it an empty value.
1314 @node Working Directory,,,
1315 @section Your Program's Working Directory
1317 @cindex working directory (of your program)
1318 Each time you start your program with @samp{run}, it inherits its
1319 working directory from the current working directory of _GDBN__. _GDBN__'s
1320 working directory is initially whatever it inherited from its parent
1321 process (typically the shell), but you can specify a new working
1322 directory in _GDBN__ with the @samp{cd} command.
1324 The _GDBN__ working directory also serves as a default for the commands
1325 that specify files for _GDBN__ to operate on. @xref{Files}.
1328 @item cd @var{directory}
1330 Set _GDBN__'s working directory to @var{directory}.
1334 Print _GDBN__'s working directory.
1337 @node Input/Output,,,
1338 @section Your Program's Input and Output
1343 @cindex controlling terminal
1344 By default, the program you run under _GDBN__ does input and output to the same
1345 terminal that _GDBN__ uses.
1347 You can redirect the program's input and/or output using shell
1348 redirection with the @samp{run} command. For example,
1355 starts the program, diverting its output to the file @file{outfile}.
1358 Another way to specify where the program should do input and output is
1359 with the @samp{tty} command. This command accepts a file name as
1360 argument, and causes this file to be the default for future @samp{run}
1361 commands. It also resets the controlling terminal for the child
1362 process, for future @samp{run} commands. For example,
1369 directs that processes started with subsequent @samp{run} commands
1370 default to do input and output on the terminal @file{/dev/ttyb} and have
1371 that as their controlling terminal.
1373 An explicit redirection in @samp{run} overrides the @samp{tty} command's
1374 effect on input/output redirection, but not its effect on the
1375 controlling terminal.
1377 When you use the @samp{tty} command or redirect input in the @samp{run}
1378 command, only the input @emph{for your program} is affected. The input
1379 for _GDBN__ still comes from your terminal.
1382 @section Debugging an Already-Running Process
1387 @item attach @var{process-id}
1388 If your currently selected target supports processes, this command
1389 attaches to a running process---one that was started outside _GDBN__.
1390 (@samp{info files} will show your active targets). The command takes as
1391 argument a process ID. The usual way to find out the process-id of
1392 a Unix process is with the @code{ps} utility, or with the @code{jobs -l}
1395 @samp{attach} will not repeat if you press @key{RET} a second time after
1396 executing the command.
1399 To use @samp{attach}, you must have permission to send the process a
1400 signal, and it must have the same effective user ID as the _GDBN__
1403 When using @samp{attach}, you should first use the @samp{file} command
1404 to specify the program running in the process and load its symbol table.
1406 The first thing _GDBN__ does after arranging to debug the specified
1407 process is to stop it. You can examine and modify an attached process
1408 with all the _GDBN__ commands that ordinarily available when you start
1409 processes with @samp{run}. You can insert breakpoints; you can step and
1410 continue; you can modify storage. If you would rather the process
1411 continue running, you may use the @samp{continue} command after
1412 attaching _GDBN__ to the process.
1415 When you have finished debugging the attached process, you can use the
1416 @samp{detach} command to release it from _GDBN__'s control. Detaching
1417 the process continues its execution. After the @samp{detach} command,
1418 that process and _GDBN__ become completely independent once more, and you
1419 are ready to @samp{attach} another process or start one with @samp{run}.
1420 @samp{detach} will not repeat if you press @key{RET} again after using
1423 If you exit _GDBN__ or use the @samp{run} command while you have an attached
1424 process, you kill that process. By default, you will be asked for
1425 confirmation if you try to do either of these things; you can control
1426 whether or not this happens by using the @samp{set caution} command
1427 (@pxref{Messages/Warnings}).
1430 @node Kill Process,,,
1431 @section Killing the Child Process
1436 Kill the child process in which your program is running under _GDBN__.
1439 This command is useful if you wish to debug a core dump instead of a
1440 running process. _GDBN__ ignores any core dump file while your program
1444 On some operating systems, you can't execute your program in another
1445 process while breakpoints are active inside _GDBN__. You can use the
1446 @samp{kill} command in this situation to permit running the program
1447 outside the debugger.
1449 The @samp{kill} command is also useful if you wish to recompile and
1450 relink the program, since on many systems it is impossible to modify an
1451 executable file which is running in a process. In this case, when you
1452 next type @samp{run}, _GDBN__ will notice that the file has changed, and
1453 will re-read the symbol table (while trying to preserve your current
1454 breakpoint settings).
1457 @chapter Stopping and Continuing
1459 When you run a program normally, it runs until it terminates. The
1460 principal purpose of using a debugger is so that you can stop it before
1461 that point; or so that if the program runs into trouble you can
1462 investigate and find out why.
1464 @node Breakpoints,,,
1465 @section Breakpoints
1468 A @dfn{breakpoint} makes your program stop whenever a certain point in the
1469 program is reached. You set breakpoints explicitly with _GDBN__ commands,
1470 specifying the place where the program should stop by line number, function
1471 name or exact address in the program. You can add various other conditions
1472 to control whether the program will stop.
1474 Each breakpoint is assigned a number when it is created; these numbers are
1475 successive integers starting with 1. In many of the commands for controlling
1476 various features of breakpoints you use the breakpoint number to say which
1477 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
1478 @dfn{disabled}; if disabled, it has no effect on the program until you
1482 @subsection Setting Breakpoints
1486 Breakpoints are set with the @samp{break} command (abbreviated @samp{b}).
1488 You have several ways to say where the breakpoint should go.
1491 @item break @var{function}
1492 Set a breakpoint at entry to function @var{function}.
1494 @item break @var{+offset}
1495 @itemx break @var{-offset}
1496 Set a breakpoint some number of lines forward or back from the position
1497 at which execution stopped in the currently selected frame.
1499 @item break @var{linenum}
1500 Set a breakpoint at line @var{linenum} in the current source file.
1501 That file is the last file whose source text was printed. This
1502 breakpoint will stop the program just before it executes any of the
1505 @item break @var{filename}:@var{linenum}
1506 Set a breakpoint at line @var{linenum} in source file @var{filename}.
1508 @item break @var{filename}:@var{function}
1509 Set a breakpoint at entry to function @var{function} found in file
1510 @var{filename}. Specifying a file name as well as a function name is
1511 superfluous except when multiple files contain similarly named
1514 @item break *@var{address}
1515 Set a breakpoint at address @var{address}. You can use this to set
1516 breakpoints in parts of the program which do not have debugging
1517 information or source files.
1520 Set a breakpoint at the next instruction to be executed in the selected
1521 stack frame (@pxref{Stack}). In any selected frame but the innermost,
1522 this will cause the program to stop as soon as control returns to that
1523 frame. This is equivalent to a @samp{finish} command in the frame
1524 inside the selected frame. If this is done in the innermost frame, _GDBN__
1525 will stop the next time it reaches the current location; this may be
1526 useful inside of loops.
1528 _GDBN__ normally ignores breakpoints when it resumes execution, until at
1529 least one instruction has been executed. If it did not do this, you
1530 would be unable to proceed past a breakpoint without first disabling the
1531 breakpoint. This rule applies whether or not the breakpoint already
1532 existed when the program stopped.
1534 @item break @dots{} if @var{cond}
1535 Set a breakpoint with condition @var{cond}; evaluate the expression
1536 @var{cond} each time the breakpoint is reached, and stop only if the
1537 value is nonzero. @samp{@dots{}} stands for one of the possible
1538 arguments described above (or no argument) specifying where to break.
1539 @xref{Conditions}, for more information on breakpoint conditions.
1541 @item tbreak @var{args}
1543 Set a breakpoint enabled only for one stop. @var{args} are the
1544 same as in the @samp{break} command, and the breakpoint is set in the same
1545 way, but the breakpoint is automatically disabled the first time it
1546 is hit. @xref{Disabling}.
1548 @item rbreak @var{regex}
1550 Set a breakpoint on all functions matching @var{regex}. This is
1551 useful for setting breakpoints on overloaded functions that are not
1552 members of any special classes. This command sets an unconditional
1553 breakpoint on all matches, printing a list of all breakpoints it set.
1554 Once these breakpoints are set, they are treated just like the
1555 breakpoints set with the @samp{break} command. They can be deleted,
1556 disabled, made conditional, etc., in the standard ways.
1558 @kindex info breakpoints
1560 @item info breakpoints
1561 The command @samp{info breakpoints} prints a list of all breakpoints set
1562 and not deleted, showing their numbers, where in the program they are,
1563 and any special features in use for them. Disabled breakpoints are
1564 included in the list, but marked as disabled. @samp{info break} with a
1565 breakpoint number as argument lists only that breakpoint. The
1566 convenience variable @code{$_} and the default examining-address for the
1567 @samp{x} command are set to the address of the last breakpoint listed
1571 _GDBN__ allows you to set any number of breakpoints at the same place in the
1572 program. There is nothing silly or meaningless about this. When the
1573 breakpoints are conditional, this is even useful (@pxref{Conditions}).
1575 @node Set Watchpoints,,,
1576 @subsection Setting Watchpoints
1578 A @dfn{watchpoint} is a special breakpoint that stops your program when
1579 the value of an expression changes. You can use a watchpoint to stop
1580 execution whenever the value of an expression changes, without having to
1581 predict a particular place in the inferior process where this may
1582 happen. Aside from the different syntax in setting a watchpoint, it is
1583 managed exactly like any other breakpoint and is enabled, disabled, and
1584 deleted using exactly the same commands.
1586 Watchpoints currently execute two orders of magnitude more slowly than
1587 other breakpoints, but this can well be worth it to catch errors where
1588 you have no clue what part of your program is the culprit. Some
1589 processors provide special hardware to implement this feature; future
1590 releases of _GDBN__ will use such hardware if it is available.
1594 @item watch @var{expr}
1595 Set a watchpoint for an expression.
1599 This command prints a list of watchpoints.
1602 @node Exception Handling,,,
1603 @subsection Breakpoints and Exceptions
1604 @cindex exception handlers
1606 Some languages, such as GNU C++, implement exception handling. _GDBN__
1607 can be used to examine what caused the program to raise an exception
1608 and to list the exceptions the program is prepared to handle at a
1609 given point in time.
1612 @item catch @var{exceptions}
1615 Breakpoints can be set at active exception handlers by using the
1616 @samp{catch} command. @var{exceptions} is a list of names of exceptions
1620 You can use @samp{info catch} to list active exception handlers;
1623 There are currently some limitations to exception handling in _GDBN__.
1624 These will be corrected in a future release.
1628 If you call a function interactively, _GDBN__ will normally return
1629 control to you when the function has finished executing. If the call
1630 raises an exception, however, the call may bypass the mechanism that
1631 returns control to the user and cause the program to simply continue
1632 running until it hits a breakpoint, catches a signal that _GDBN__ is
1633 listening for, or exits.
1635 You cannot raise an exception interactively.
1637 You cannot interactively install an exception handler.
1640 @cindex raise exceptions
1641 Sometimes @samp{catch} is not the best way to debug exception handling:
1642 if you need to know exactly where an exception is raised, it's better to
1643 stop @emph{before} the exception handler is called, since that way you
1644 can see the stack before any unwinding takes place.
1646 To stop just before an exception handler is called, you need some
1647 knowledge of the implementation. In the case of GNU C++ exception are
1648 raised by calling a library function named @code{__raise_exception}
1649 which has the following ANSI C interface:
1652 /* ADDR is where the exception identifier is stored.
1653 ID is the exception identifier. */
1654 void __raise_exception (void **addr, void *id);
1658 To make the debugger catch all exceptions before any stack
1659 unwinding takes place, set a breakpoint on @code{__raise_exception}
1660 (@pxref{Breakpoints}). If you set a breakpoint in an exception handler
1661 instead, it may not be easy to find out where the exception was raised.
1663 With a conditional breakpoint (@xref{Conditions}) that depends on the
1664 value of @code{id}, you can cause the debugger to stop only when a
1665 specific exception is raised. Multiple conditional breakpoints can be
1666 used to stop the program when any of a number of exceptions are raised.
1668 @node Delete Breaks,,,
1669 @subsection Deleting Breakpoints
1671 @cindex clearing breakpoints, watchpoints
1672 @cindex deleting breakpoints, watchpoints
1673 It is often necessary to eliminate a breakpoint once it has done its job
1674 and you no longer want the program to stop there. This is called
1675 @dfn{deleting} the breakpoint. A breakpoint that has been deleted no
1676 longer exists in any sense; it is forgotten.
1678 With the @samp{clear} command you can delete breakpoints according to where
1679 they are in the program. With the @samp{delete} command you can delete
1680 individual breakpoints by specifying their breakpoint numbers.
1682 It is not necessary to delete a breakpoint to proceed past it. _GDBN__
1683 automatically ignores breakpoints on the first instruction to be executed
1684 when you continue execution without changing the execution address.
1689 Delete any breakpoints at the next instruction to be executed in the
1690 selected stack frame (@pxref{Selection}). When the innermost frame
1691 is selected, this is a good way to delete a breakpoint that the program
1694 @item clear @var{function}
1695 @itemx clear @var{filename}:@var{function}
1696 Delete any breakpoints set at entry to the function @var{function}.
1698 @item clear @var{linenum}
1699 @itemx clear @var{filename}:@var{linenum}
1700 Delete any breakpoints set at or within the code of the specified line.
1702 @item delete breakpoints @var{bnums}@dots{}
1703 @itemx delete @var{bnums}@dots{}
1705 @kindex delete breakpoints
1707 Delete the breakpoints of the numbers specified as arguments. If no
1708 argument is specified, delete all breakpoints.
1712 @subsection Disabling Breakpoints
1714 @cindex disabled breakpoints
1715 @cindex enabled breakpoints
1716 Rather than deleting a breakpoint, you might prefer to @dfn{disable} it.
1717 This makes the breakpoint inoperative as if it had been deleted, but
1718 remembers the information on the breakpoint so that you can @dfn{enable}
1721 You disable and enable breakpoints with the @samp{enable} and
1722 @samp{disable} commands, optionally specifying one or more breakpoint
1723 numbers as arguments. Use @samp{info break} to print a list of
1724 breakpoints if you don't know which breakpoint numbers to use.
1726 A breakpoint can have any of four different states of enablement:
1730 Enabled. The breakpoint will stop the program. A breakpoint made
1731 with the @samp{break} command starts out in this state.
1733 Disabled. The breakpoint has no effect on the program.
1735 Enabled once. The breakpoint will stop the program, but
1736 when it does so it will become disabled. A breakpoint made
1737 with the @samp{tbreak} command starts out in this state.
1739 Enabled for deletion. The breakpoint will stop the program, but
1740 immediately after it does so it will be deleted permanently.
1743 You can use the following commands to enable or disable a breakpoint:
1746 @item disable breakpoints @var{bnums}@dots{}
1747 @itemx disable @var{bnums}@dots{}
1749 @kindex disable breakpoints
1751 Disable the specified breakpoints---or all breakpoints, if none are
1752 listed. A disabled breakpoint has no effect but is not forgotten. All
1753 options such as ignore-counts, conditions and commands are remembered in
1754 case the breakpoint is enabled again later.
1756 @item enable breakpoints @var{bnums}@dots{}
1757 @itemx enable @var{bnums}@dots{}
1759 @kindex enable breakpoints
1761 Enable the specified breakpoints (or all defined breakpoints). They
1762 become effective once again in stopping the program, until you specify
1765 @item enable breakpoints once @var{bnums}@dots{}
1766 @itemx enable once @var{bnums}@dots{}
1767 Enable the specified breakpoints temporarily. Each will be disabled
1768 again the next time it stops the program (unless you have used one of
1769 these commands to specify a different state before that time comes).
1771 @item enable breakpoints delete @var{bnums}@dots{}
1772 @itemx enable delete @var{bnums}@dots{}
1773 Enable the specified breakpoints to work once and then die. Each of
1774 the breakpoints will be deleted the next time it stops the program
1775 (unless you have used one of these commands to specify a different
1776 state before that time comes).
1779 Save for a breakpoint set with @samp{tbreak} (@pxref{Set Breaks}),
1780 breakpoints that you set initially enabled; subsequently, they become
1781 disabled or enabled only when you use one of the commands above. (The
1782 command @samp{until} can set and delete a breakpoint of its own, but it
1783 will not change the state of your other breakpoints).
1786 @subsection Break Conditions
1787 @cindex conditional breakpoints
1788 @cindex breakpoint conditions
1790 The simplest sort of breakpoint breaks every time the program reaches a
1791 specified place. You can also specify a @dfn{condition} for a
1792 breakpoint. A condition is just a Boolean expression in your
1793 programming language. (@xref{Expressions}). A breakpoint with a
1794 condition evaluates the expression each time the program reaches it, and
1795 the program stops only if the condition is true.
1797 Break conditions may have side effects, and may even call functions in your
1798 program. These may sound like strange things to do, but their effects are
1799 completely predictable unless there is another enabled breakpoint at the
1800 same address. (In that case, _GDBN__ might see the other breakpoint first and
1801 stop the program without checking the condition of this one.) Note that
1802 breakpoint commands are usually more convenient and flexible for the
1803 purpose of performing side effects when a breakpoint is reached
1804 (@pxref{Break Commands}).
1806 Break conditions can be specified when a breakpoint is set, by using
1807 @samp{if} in the arguments to the @samp{break} command. @xref{Set Breaks}.
1808 They can also be changed at any time with the @samp{condition} command:
1811 @item condition @var{bnum} @var{expression}
1813 Specify @var{expression} as the break condition for breakpoint number
1814 @var{bnum}. From now on, this breakpoint will stop the program only if
1815 the value of @var{expression} is true (nonzero, in C). @var{expression}
1816 is not evaluated at the time the @samp{condition} command is given.
1817 When you call @samp{condition}, the expression you specify is checked
1818 immediately for syntactic correctness, and to determine whether symbols
1819 in it have referents in the context of your breakpoint.
1822 @item condition @var{bnum}
1823 Remove the condition from breakpoint number @var{bnum}. It becomes
1824 an ordinary unconditional breakpoint.
1827 @cindex ignore count (of breakpoint)
1828 A special case of a breakpoint condition is to stop only when the
1829 breakpoint has been reached a certain number of times. This is so
1830 useful that there is a special way to do it, using the @dfn{ignore
1831 count} of the breakpoint. Every breakpoint has an ignore count, which
1832 is an integer. Most of the time, the ignore count is zero, and
1833 therefore has no effect. But if the program reaches a breakpoint whose
1834 ignore count is positive, then instead of stopping, it just decrements
1835 the ignore count by one and continues. As a result, if the ignore count
1836 value is @var{n}, the breakpoint will not stop the next @var{n} times it
1840 @item ignore @var{bnum} @var{count}
1842 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
1843 The next @var{count} times the breakpoint is reached, your program's
1844 execution will not stop; other than to decrement the ignore count, _GDBN__
1847 To make the breakpoint stop the next time it is reached, specify
1850 @item continue @var{count}
1851 @itemx c @var{count}
1852 @itemx fg @var{count}
1853 @kindex continue @var{count}
1854 Continue execution of the program, setting the ignore count of the
1855 breakpoint that the program stopped at to @var{count} minus one.
1856 Thus, the program will not stop at this breakpoint until the
1857 @var{count}'th time it is reached.
1859 An argument to this command is meaningful only when the program stopped
1860 due to a breakpoint. At other times, the argument to @samp{continue} is
1863 The synonym @samp{fg} is provided purely for convenience, and has
1864 exactly the same behavior as other forms of the command.
1867 If a breakpoint has a positive ignore count and a condition, the condition
1868 is not checked. Once the ignore count reaches zero, the condition will
1871 You could achieve the effect of the ignore count with a
1872 condition such as _0__@w{@samp{$foo-- <= 0}}_1__ using a debugger convenience
1873 variable that is decremented each time. @xref{Convenience Vars}.
1875 @node Break Commands,,,
1876 @subsection Breakpoint Command Lists
1878 @cindex breakpoint commands
1879 You can give any breakpoint a series of commands to execute when the
1880 program stops due to that breakpoint. For example, you might want to
1881 print the values of certain expressions, or enable other breakpoints.
1884 @item commands @var{bnum}
1885 @itemx @dots @var{command-list} @dots
1889 Specify a list of commands for breakpoint number @var{bnum}. The commands
1890 themselves appear on the following lines. Type a line containing just
1891 @samp{end} to terminate the commands.
1893 To remove all commands from a breakpoint, use the command
1894 @samp{commands} and follow it immediately by @samp{end}; that is, give
1897 With no arguments, @samp{commands} refers to the last breakpoint set
1898 (not to the breakpoint most recently encountered).
1901 Pressing @key{RET} as a means of repeating the last _GDBN__ command is
1902 disabled from the time you enter @samp{commands} to just after the
1903 corresponding @samp{end}.
1905 You can use breakpoint commands to start the program up again. Simply
1906 use the @samp{continue} command, or @samp{step}, or any other command to
1907 resume execution. However, if you do this, any further commands in the
1908 same breakpoint's command list are ignored. When the program stops
1909 again, _GDBN__ will act according to the cause of that stop.
1912 If the first command specified is @samp{silent}, the usual message about
1913 stopping at a breakpoint is not printed. This may be desirable for
1914 breakpoints that are to print a specific message and then continue.
1915 If the remaining commands too print nothing, you will see no sign that
1916 the breakpoint was reached at all. @samp{silent} is not really a command;
1917 it is meaningful only at the beginning of the commands for a breakpoint.
1919 The commands @samp{echo} and @samp{output} that allow you to print precisely
1920 controlled output are often useful in silent breakpoints. @xref{Output}.
1922 For example, here is how you could use breakpoint commands to print the
1923 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
1936 One application for breakpoint commands is to correct one bug so you can
1937 test another. Put a breakpoint just after the erroneous line of code, give
1938 it a condition to detect the case in which something erroneous has been
1939 done, and give it commands to assign correct values to any variables that
1940 need them. End with the @samp{continue} command so that the program does not
1941 stop, and start with the @samp{silent} command so that no output is
1942 produced. Here is an example:
1953 One deficiency in the operation of automatically continuing breakpoints
1954 under Unix appears when your program uses raw mode for the terminal.
1955 _GDBN__ switches back to its own terminal modes (not raw) before executing
1956 commands, and then must switch back to raw mode when your program is
1957 continued. This causes any pending terminal input to be lost.
1958 In the GNU system, this will be fixed by changing the behavior of
1961 Under Unix, when you have this problem, you might be able to get around
1962 it by putting your actions into the breakpoint condition instead of
1963 commands. For example
1966 condition 5 (x = y + 4), 0
1970 specifies a condition expression (@xref{Expressions}) that will change
1971 @code{x} as needed, then always have the value 0 so the program will not
1972 stop. Loss of input is avoided here because break conditions are
1973 evaluated without changing the terminal modes. When you want to have
1974 nontrivial conditions for performing the side effects, the operators
1975 @samp{&&}, @samp{||} and @samp{?@dots{}:} may be useful.
1977 @node Error in Breakpoints,,,
1978 @subsection ``Cannot Insert Breakpoints''
1980 @c FIXME: "cannot insert breakpoints" error, v unclear.
1981 @c Q in pending mail to Gilmore. ---pesch@cygnus.com, 26mar91
1982 Under some operating systems, breakpoints cannot be used in a program if
1983 any other process is running that program. In this situation,
1984 attempting to run or continue a program with a breakpoint will cause _GDBN__
1985 to stop the other process.
1987 When this happens, you have three ways to proceed:
1991 Remove or disable the breakpoints, then continue.
1994 Suspend _GDBN__, and copy the file containing the program to a new name.
1995 Resume _GDBN__ and use the @samp{exec-file} command to specify that _GDBN__
1996 should run the program under that name. Then start the program again.
1999 Relink the program so that the text segment is nonsharable, using the
2000 linker option @samp{-N}. The operating system limitation may not apply
2001 to nonsharable executables.
2008 @dfn{Stepping} means setting your program in motion for a limited time, so
2009 that control will return automatically to _GDBN__ after one line of
2010 code or one machine instruction. Breakpoints are active during stepping
2011 and the program will stop for them even if it has not gone as far as the
2012 stepping command specifies.
2017 Continue running the program until control reaches a different source
2018 line, then stop it and return control to the debugger. This command is
2019 abbreviated @samp{s}.
2021 This command may be given when control is within a function for which
2022 there is no debugging information. In that case, execution will proceed
2023 until control reaches a different function, or is about to return from
2026 @item step @var{count}
2027 Continue running as in @samp{step}, but do so @var{count} times. If a
2028 breakpoint is reached or a signal not related to stepping occurs before
2029 @var{count} steps, stepping stops right away.
2033 Continue to the next source line in the current stack frame. Similar to
2034 @samp{step}, but any function calls appearing within the line of code
2035 are executed without stopping. Execution stops when control reaches a
2036 different line of code at the stack level which was executing when the
2037 @samp{next} command was given. This command is abbreviated @samp{n}.
2039 An argument is a repeat count, as in @samp{step}.
2041 @samp{next} within a function that lacks debugging information acts as does
2042 @samp{step}, but any function calls appearing within the code of the
2043 function are executed without stopping.
2044 @c FIXME: great; so what does *step* do within a fn that lacks debug info?
2048 Continue running until just after the selected stack frame returns (or
2049 until there is some other reason to stop, such as a fatal signal or a
2050 breakpoint). Print the value returned by the selected stack frame (if
2053 Contrast this with the @samp{return} command (@pxref{Returning}).
2057 Continue running until a source line past the current line, in the
2058 current stack frame, is reached. This command is used to avoid single
2059 stepping through a loop more than once. It is like the @samp{next}
2060 command, except that when @samp{until} encounters a jump, it
2061 automatically continues execution until the program counter is greater
2062 than the address of the jump.
2064 This means that when you reach the end of a loop after single stepping
2065 though it, @samp{until} will cause the program to continue execution
2066 until the loop is exited. In contrast, a @samp{next} command at the end
2067 of a loop will simply step back to the beginning of the loop, which
2068 would force you to step through the next iteration.
2070 @samp{until} always stops the program if it attempts to exit the current
2073 @samp{until} may produce somewhat counterintuitive results if the order
2074 of the source lines does not match the actual order of execution. For
2075 example, in a typical C @code{for}-loop, the third expression in the
2076 @code{for}-statement (the loop-step expression) is executed after the
2077 statements in the body of the loop, but is written before them.
2078 Therefore, the @samp{until} command would appear to step back to the
2079 beginning of the loop when it advances to this expression. However, it
2080 has not really done so, not in terms of the actual machine code.
2082 @samp{until} with no argument works by means of single
2083 instruction stepping, and hence is slower than @samp{until} with an
2086 @item until @var{location}
2087 Continue running the program until either the specified location is
2088 reached, or the current (innermost) stack frame returns. @var{location}
2089 is any of the forms of argument acceptable to @samp{break} (@pxref{Set
2090 Breaks}). This form of the command uses breakpoints, and hence is
2091 quicker than @samp{until} without an argument.
2097 Execute one machine instruction, then stop and return to the debugger.
2099 It is often useful to do @samp{display/i $pc} when stepping by machine
2100 instructions. This will cause the next instruction to be executed to
2101 be displayed automatically at each stop. @xref{Auto Display}.
2103 An argument is a repeat count, as in @samp{step}.
2109 Execute one machine instruction, but if it is a function call,
2110 proceed until the function returns.
2112 An argument is a repeat count, as in @samp{next}.
2115 A typical technique for using stepping is to put a breakpoint
2116 (@pxref{Breakpoints}) at the beginning of the function or the section of
2117 the program in which a problem is believed to lie, run the program until
2118 it stops at that breakpoint, and then step through the suspect area,
2119 examining the variables that are interesting, until you see the problem
2122 The @samp{continue} command can be used after stepping to resume execution
2123 until the next breakpoint or signal.
2128 After your program stops, most likely you will want it to run some more if
2129 the bug you are looking for has not happened yet.
2134 Continue running the program at the place where it stopped.
2137 If the program stopped at a breakpoint, the place to continue running
2138 is the address of the breakpoint. You might expect that continuing would
2139 just stop at the same breakpoint immediately. In fact, @samp{continue}
2140 takes special care to prevent that from happening. You do not need
2141 to delete the breakpoint to proceed through it after stopping at it.
2142 You can, however, specify an ignore-count for the breakpoint that the
2143 program stopped at, by means of an argument to the @samp{continue} command.
2146 If the program stopped because of a signal other than @code{SIGINT} or
2147 @code{SIGTRAP}, continuing will cause the program to see that signal.
2148 You may not want this to happen. For example, if the program stopped
2149 due to some sort of memory reference error, you might store correct
2150 values into the erroneous variables and continue, hoping to see more
2151 execution; but the program would probably terminate immediately as
2152 a result of the fatal signal once it sees the signal. To prevent this,
2153 you can continue with @samp{signal 0}. @xref{Signaling}. You can
2154 also act in advance to control what signals your program will see, using
2155 the @samp{handle} command (@pxref{Signals}).
2161 A signal is an asynchronous event that can happen in a program. The
2162 operating system defines the possible kinds of signals, and gives each
2163 kind a name and a number. For example, in Unix @code{SIGINT} is the
2164 signal a program gets when you type an interrupt (often @kbd{C-c});
2165 @code{SIGSEGV} is the signal a program gets from referencing a place in
2166 memory far away from all the areas in use; @code{SIGALRM} occurs when
2167 the alarm clock timer goes off (which happens only if the program has
2168 requested an alarm).
2170 @cindex fatal signals
2171 Some signals, including @code{SIGALRM}, are a normal part of the
2172 functioning of the program. Others, such as @code{SIGSEGV}, indicate
2173 errors; these signals are @dfn{fatal} (kill the program immediately) if the
2174 program has not specified in advance some other way to handle the signal.
2175 @code{SIGINT} does not indicate an error in the program, but it is normally
2176 fatal so it can carry out the purpose of the interrupt: to kill the program.
2178 _GDBN__ has the ability to detect any occurrence of a signal in the program
2179 running under _GDBN__'s control. You can tell _GDBN__ in advance what to do for
2180 each kind of signal.
2182 @cindex handling signals
2183 Normally, _GDBN__ is set up to ignore non-erroneous signals like @code{SIGALRM}
2184 (so as not to interfere with their role in the functioning of the program)
2185 but to stop the program immediately whenever an error signal happens.
2186 You can change these settings with the @samp{handle} command.
2191 Print a table of all the kinds of signals and how _GDBN__ has been told to
2192 handle each one. You can use this to see the signal numbers of all
2193 the defined types of signals.
2195 @item handle @var{signal} @var{keywords}@dots{}
2197 Change the way _GDBN__ handles signal @var{signal}. @var{signal} can be the
2198 number of a signal or its name (with or without the @samp{SIG} at the
2199 beginning). The @var{keywords} say what change to make.
2203 The keywords allowed by the @samp{handle} command can be abbreviated.
2204 Their full names are:
2208 _GDBN__ should not stop the program when this signal happens. It may
2209 still print a message telling you that the signal has come in.
2212 _GDBN__ should stop the program when this signal happens. This implies
2213 the @samp{print} keyword as well.
2216 _GDBN__ should print a message when this signal happens.
2219 _GDBN__ should not mention the occurrence of the signal at all. This
2220 implies the @samp{nostop} keyword as well.
2223 _GDBN__ should allow the program to see this signal; the program will be
2224 able to handle the signal, or may be terminated if the signal is fatal
2228 _GDBN__ should not allow the program to see this signal.
2232 When a signal has been set to stop the program, the program cannot see the
2233 signal until you continue. It will see the signal then, if @samp{pass} is
2234 in effect for the signal in question @i{at that time}. In other words,
2235 after _GDBN__ reports a signal, you can use the @samp{handle} command with
2236 @samp{pass} or @samp{nopass} to control whether that signal will be seen by
2237 the program when you later continue it.
2239 You can also use the @samp{signal} command to prevent the program from
2240 seeing a signal, or cause it to see a signal it normally would not see,
2241 or to give it any signal at any time. @xref{Signaling}.
2245 @chapter Examining the Stack
2247 When your program has stopped, the first thing you need to know is where it
2248 stopped and how it got there.
2251 Each time your program performs a function call, the information about
2252 where in the program the call was made from is saved in a block of data
2253 called a @dfn{stack frame}. The frame also contains the arguments of the
2254 call and the local variables of the function that was called. All the
2255 stack frames are allocated in a region of memory called the @dfn{call
2258 When your program stops, the _GDBN__ commands for examining the stack allow you
2259 to see all of this information.
2261 @cindex selected frame
2262 One of the stack frames is @dfn{selected} by _GDBN__ and many _GDBN__ commands
2263 refer implicitly to the selected frame. In particular, whenever you ask
2264 _GDBN__ for the value of a variable in the program, the value is found in the
2265 selected frame. There are special _GDBN__ commands to select whichever frame
2266 you are interested in.
2268 When the program stops, _GDBN__ automatically selects the currently executing
2269 frame and describes it briefly as the @samp{frame} command does
2270 (@pxref{Frame Info, Info}).
2273 @section Stack Frames
2277 The call stack is divided up into contiguous pieces called @dfn{stack
2278 frames}, or @dfn{frames} for short; each frame is the data associated
2279 with one call to one function. The frame contains the arguments given
2280 to the function, the function's local variables, and the address at
2281 which the function is executing.
2283 @cindex initial frame
2284 @cindex outermost frame
2285 @cindex innermost frame
2286 When your program is started, the stack has only one frame, that of the
2287 function @code{main}. This is called the @dfn{initial} frame or the
2288 @dfn{outermost} frame. Each time a function is called, a new frame is
2289 made. Each time a function returns, the frame for that function invocation
2290 is eliminated. If a function is recursive, there can be many frames for
2291 the same function. The frame for the function in which execution is
2292 actually occurring is called the @dfn{innermost} frame. This is the most
2293 recently created of all the stack frames that still exist.
2295 @cindex frame pointer
2296 Inside your program, stack frames are identified by their addresses. A
2297 stack frame consists of many bytes, each of which has its own address; each
2298 kind of computer has a convention for choosing one of those bytes whose
2299 address serves as the address of the frame. Usually this address is kept
2300 in a register called the @dfn{frame pointer register} while execution is
2301 going on in that frame.
2303 @cindex frame number
2304 _GDBN__ assigns numbers to all existing stack frames, starting with
2305 @code{0} for the innermost frame, @code{1} for the frame that called it,
2306 and so on upward. These numbers do not really exist in your program;
2307 they are assigned by _GDBN__ to give you a way of designating stack
2308 frames in _GDBN__ commands.
2310 @cindex frameless execution
2311 Some compilers allow functions to be compiled so that they operate
2312 without stack frames. (For example, the @code{_GCC__} option
2313 @samp{-fomit-frame-pointer} will generate functions without a frame.)
2314 This is occasionally done with heavily used library functions to save
2315 the frame setup time. _GDBN__ has limited facilities for dealing with
2316 these function invocations; if the innermost function invocation has no
2317 stack frame, _GDBN__ will give it a virtual stack frame of 0 and
2318 correctly allow tracing of the function call chain. Results are
2319 undefined if a function invocation besides the innermost one is
2325 A backtrace is a summary of how the program got where it is. It shows one
2326 line per frame, for many frames, starting with the currently executing
2327 frame (frame zero), followed by its caller (frame one), and on up the
2335 Print a backtrace of the entire stack: one line per frame for all
2336 frames in the stack.
2338 You can stop the backtrace at any time by typing the system interrupt
2339 character, normally @kbd{Control-C}.
2341 @item backtrace @var{n}
2343 Similar, but print only the innermost @var{n} frames.
2345 @item backtrace -@var{n}
2347 Similar, but print only the outermost @var{n} frames.
2352 The names @samp{where} and @samp{info stack} are additional aliases
2353 for @samp{backtrace}.
2355 Each line in the backtrace shows the frame number and the function name.
2356 The program counter value is also shown---unless you use @samp{set
2357 print address off}. The backtrace also shows the source file name and
2358 line number, as well as the arguments to the function. The program
2359 counter value is omitted if it is at the beginning of the code for that
2362 Here is an example of a backtrace. It was made with the command
2363 @samp{bt 3}, so it shows the innermost three frames.
2366 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8) at builtin.c:993
2367 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
2368 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
2370 (More stack frames follow...)
2374 The display for frame @code{#0} doesn't begin with a program counter
2375 value, indicating that the program has stopped at the beginning of the
2376 code for line @code{993} of @code{builtin.c}.
2379 @section Selecting a Frame
2381 Most commands for examining the stack and other data in the program work on
2382 whichever stack frame is selected at the moment. Here are the commands for
2383 selecting a stack frame; all of them finish by printing a brief description
2384 of the stack frame just selected.
2391 Select frame number @var{n}. Recall that frame zero is the innermost
2392 (currently executing) frame, frame one is the frame that called the
2393 innermost one, and so on. The highest-numbered frame is @code{main}'s
2396 @item frame @var{addr}
2398 Select the frame at address @var{addr}. This is useful mainly if the
2399 chaining of stack frames has been damaged by a bug, making it
2400 impossible for _GDBN__ to assign numbers properly to all frames. In
2401 addition, this can be useful when the program has multiple stacks and
2402 switches between them.
2406 Move @var{n} frames up the stack. For positive numbers @var{n}, this
2407 advances toward the outermost frame, to higher frame numbers, to frames
2408 that have existed longer. @var{n} defaults to one.
2412 Move @var{n} frames down the stack. For positive numbers @var{n}, this
2413 advances toward the innermost frame, to lower frame numbers, to frames
2414 that were created more recently. @var{n} defaults to one.
2417 All of these commands end by printing some information on the frame that
2418 has been selected: the frame number, the function name, the arguments, the
2419 source file and line number of execution in that frame, and the text of
2420 that source line. For example:
2424 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc) at env.c:10
2425 10 read_input_file (argv[i]);
2428 After such a printout, the @samp{list} command with no arguments will print
2429 ten lines centered on the point of execution in the frame. @xref{List}.
2432 @item up-silently @var{n}
2433 @itemx down-silently @var{n}
2434 @kindex down-silently
2436 These two commands are variants of @samp{up} and @samp{down},
2437 respectively; they differ in that they do their work silently, without
2438 causing display of the new frame. They are intended primarily for use
2439 in _GDBN__ command scripts, where the output might be unnecessary and
2445 @section Information on a Frame
2447 There are several other commands to print information about the selected
2453 When used without any argument, this command does not change which frame
2454 is selected, but prints a brief description of the currently
2455 selected stack frame. It can be abbreviated @samp{f}. With an
2456 argument, this command is used to select a stack frame (@pxref{Selection}).
2460 This command prints a verbose description of the selected stack frame,
2461 including the address of the frame, the addresses of the next frame down
2462 (called by this frame) and the next frame up (caller of this frame),
2463 the address of the frame's arguments, the program counter saved in it
2464 (the address of execution in the caller frame), and which registers
2465 were saved in the frame. The verbose description is useful when
2466 something has gone wrong that has made the stack format fail to fit
2467 the usual conventions.
2469 @item info frame @var{addr}
2470 Print a verbose description of the frame at address @var{addr},
2471 without selecting that frame. The selected frame remains unchanged by
2476 Print the arguments of the selected frame, each on a separate line.
2480 Print the local variables of the selected frame, each on a separate
2481 line. These are all variables declared static or automatic within all
2482 program blocks that execution in this frame is currently inside of.
2486 @cindex catch exceptions
2487 @cindex exception handlers
2488 Print a list of all the exception handlers that are active in the
2489 current stack frame at the current point of execution. To see other
2490 exception handlers, visit the associated frame (using the @samp{up},
2491 @samp{down}, or @samp{frame} commands); then type @samp{info catch}.
2492 @xref{Exception Handling}.
2496 @chapter Examining Source Files
2498 _GDBN__ can print parts of your program's source, since the debugging
2499 information recorded in your program tells _GDBN__ what source files
2500 were used to built it. When your program stops, _GDBN__
2501 spontaneously prints the line where it stopped. Likewise, when you
2502 select a stack frame (@pxref{Selection}), _GDBN__ prints the line
2503 where execution in that frame has stopped. You can also
2504 print parts of source files by explicit command.
2506 If you use _GDBN__ through its GNU Emacs interface, you may prefer to
2507 use Emacs facilities to view source; @pxref{Emacs}.
2510 @section Printing Source Lines
2514 To print lines from a source file, use the @samp{list} command
2515 (abbreviated @samp{l}). There are several ways to specify what part
2516 of the file you want to print.
2518 Here are the forms of the @samp{list} command most commonly used:
2521 @item list @var{linenum}
2522 Print ten lines centered around line number @var{linenum} in the
2523 current source file.
2525 @item list @var{function}
2526 Print ten lines centered around the beginning of function
2530 Print ten more lines. If the last lines printed were printed with a
2531 @samp{list} command, this prints ten lines following the last lines
2532 printed; however, if the last line printed was a solitary line printed
2533 as part of displaying a stack frame (@pxref{Stack}), this prints ten
2534 lines centered around that line.
2537 Print ten lines just before the lines last printed.
2540 Repeating a @samp{list} command with @key{RET} discards the argument,
2541 so it is equivalent to typing just @samp{list}. This is more useful
2542 than listing the same lines again. An exception is made for an
2543 argument of @samp{-}; that argument is preserved in repetition so that
2544 each repetition moves up in the source file.
2547 In general, the @samp{list} command expects you to supply zero, one or two
2548 @dfn{linespecs}. Linespecs specify source lines; there are several ways
2549 of writing them but the effect is always to specify some source line.
2550 Here is a complete description of the possible arguments for @samp{list}:
2553 @item list @var{linespec}
2554 Print ten lines centered around the line specified by @var{linespec}.
2556 @item list @var{first},@var{last}
2557 Print lines from @var{first} to @var{last}. Both arguments are
2560 @item list ,@var{last}
2561 Print ten lines ending with @var{last}.
2563 @item list @var{first},
2564 Print ten lines starting with @var{first}.
2567 Print ten lines just after the lines last printed.
2570 Print ten lines just before the lines last printed.
2573 As described in the preceding table.
2576 Here are the ways of specifying a single source line---all the
2581 Specifies line @var{number} of the current source file.
2582 When a @samp{list} command has two linespecs, this refers to
2583 the same source file as the first linespec.
2586 Specifies the line @var{offset} lines after the last line printed.
2587 When used as the second linespec in a @samp{list} command that has
2588 two, this specifies the line @var{offset} lines down from the
2592 Specifies the line @var{offset} lines before the last line printed.
2594 @item @var{filename}:@var{number}
2595 Specifies line @var{number} in the source file @var{filename}.
2597 @item @var{function}
2598 @c FIXME: "of the open-brace" is C-centric. When we add other langs...
2599 Specifies the line of the open-brace that begins the body of the
2600 function @var{function}.
2602 @item @var{filename}:@var{function}
2603 Specifies the line of the open-brace that begins the body of the
2604 function @var{function} in the file @var{filename}. You only need the
2605 file name with a function name to avoid ambiguity when there are
2606 identically named functions in different source files.
2608 @item *@var{address}
2609 Specifies the line containing the program address @var{address}.
2610 @var{address} may be any expression.
2614 @section Searching Source Files
2617 @kindex forward-search
2618 @kindex reverse-search
2620 There are two commands for searching through the current source file for a
2623 The command @samp{forward-search @var{regexp}} checks each line, starting
2624 with the one following the last line listed, for a match for @var{regexp}.
2625 It lists the line that is found. You can abbreviate the command name
2626 as @samp{fo}. The synonym @samp{search @var{regexp}} is also supported.
2628 The command @samp{reverse-search @var{regexp}} checks each line, starting
2629 with the one before the last line listed and going backward, for a match
2630 for @var{regexp}. It lists the line that is found. You can abbreviate
2631 this command as @samp{rev}.
2633 @node Source Path,,,
2634 @section Specifying Source Directories
2637 @cindex directories for source files
2638 Executable programs sometimes do not record the directories of the source
2639 files from which they were compiled, just the names. Even when they do,
2640 the directories could be moved between the compilation and your debugging
2641 session. _GDBN__ has a list of directories to search for source files;
2642 this is called the @dfn{source path}. Each time _GDBN__ wants a source file,
2643 it tries all the directories in the list, in the order they are present
2644 in the list, until it finds a file with the desired name. Note that
2645 the executable search path is @emph{not} used for this purpose. Neither is
2646 the current working directory, unless it happens to be in the source
2649 If _GDBN__ can't find a source file in the source path, and the object
2650 program records a directory, _GDBN__ tries that directory too. If the
2651 source path is empty, and there is no record of the compilation
2652 directory, _GDBN__ will, as a last resort, look in the current
2655 Whenever you reset or rearrange the source path, _GDBN__ will clear out
2656 any information it has cached about where source files are found, where
2657 each line is in the file, etc.
2660 When you start _GDBN__, its source path is empty.
2661 To add other directories, use the @samp{directory} command.
2664 @item directory @var{dirname} @dots
2665 Add directory @var{dirname} to the front of the source path. Several
2666 directory names may be given to this command, separated by @samp{:} or
2667 whitespace. You may specify a directory that is already in the source
2668 path; this moves it forward, so it will be searched sooner. You can use
2669 the string @samp{$cwd} to refer to the current working directory, and
2670 @samp{$cdir} to refer to the compilation directory (if one is recorded).
2673 Reset the source path to empty again. This requires confirmation.
2675 The @samp{directory} command will not repeat if you press @key{RET} a
2676 second time after executing it once.
2678 @item show directories
2679 @kindex show directories
2680 Print the source path: show which directories it contains.
2683 If your source path is cluttered with directories that are no longer of
2684 interest, _GDBN__ may sometimes cause confusion by finding the wrong
2685 versions of source. You can correct the situation as follows:
2689 Use @samp{directory} with no argument to reset the source path to empty.
2692 Use @samp{directory} with suitable arguments to add any other
2693 directories you want in the source path. You can add all the directories
2697 @node Machine Code,,,
2698 @section Source and Machine Code
2699 You can use the command @samp{info line} to map source lines to program
2700 addresses, and the command @samp{disassemble} or its synonym
2701 @samp{disasm} to display a range of addresses as machine instructions.
2704 @item info line @var{linespec}
2706 Print the starting and ending addresses of the compiled code for
2707 source line @var{linespec}.
2710 After @samp{info line}, the default address for the @samp{x}
2711 command is changed to the starting address of the line, so that
2712 @samp{x/i} is sufficient to begin examining the machine code
2713 (@pxref{Memory}). Also, this address is saved as the value of the
2714 convenience variable @code{$_} (@pxref{Convenience Vars}).
2720 This specialized command is provided to dump a range of memory as
2721 machine instructions. The default memory range is the function
2722 surrounding the program counter of the selected frame. A single
2723 argument to this command is a program counter value; the function
2724 surrounding this value will be dumped. Two arguments (separated by one
2725 or more spaces) specify a range of addresses (first inclusive, second
2726 exclusive) to be dumped. The two spellings, @samp{disasm} and
2727 @samp{disassemble}, are equivalent.
2731 @chapter Examining Data
2733 @cindex printing data
2734 @cindex examining data
2737 @c "inspect" isn't quite a synonym if you're using Epoch, which we don't
2738 @c document because it's nonstandard... Under Epoch it displays in a
2739 @c different window or something like that.
2740 The usual way to examine data in your program is with the @samp{print}
2741 command (abbreviated @samp{p}), or its synonym @samp{inspect}. It
2742 evaluates and prints the value of any valid expression of the language
2743 the program is written in (for now, C or C++). You type
2750 where @var{exp} is any valid expression (in the source language), and
2751 the value of @var{exp} is printed in a format appropriate to its data
2754 A more low-level way of examining data is with the @samp{x} command.
2755 It examines data in memory at a specified address and prints it in a
2756 specified format. @xref{Memory}.
2758 @node Expressions,,,
2759 @section Expressions
2762 @code{print} and many other _GDBN__ commands accept an expression and
2763 compute its value. Any kind of constant, variable or operator defined
2764 by the programming language you are using is legal in an expression in
2765 _GDBN__. This includes conditional expressions, function calls, casts
2766 and string constants. It unfortunately does not include symbols defined
2767 by preprocessor @code{#define} commands, or C++ expressions involving
2768 @samp{::}, the name resolution operator.
2769 @c FIXME: actually C++ a::b works except in obscure circumstances where it
2770 @c FIXME...can conflict with GDB's own name scope resolution.
2772 Casts are supported in all languages, not just in C, because it is so
2773 useful to cast a number into a pointer so as to examine a structure
2774 at that address in memory.
2776 _GDBN__ supports three kinds of operator in addition to those of programming
2781 @samp{@@} is a binary operator for treating parts of memory as arrays.
2782 @xref{Arrays}, for more information.
2785 @samp{::} allows you to specify a variable in terms of the file or
2786 function where it is defined. @xref{Variables}.
2788 @item @{@var{type}@} @var{addr}
2789 Refers to an object of type @var{type} stored at address @var{addr} in
2790 memory. @var{addr} may be any expression whose value is an integer or
2791 pointer (but parentheses are required around binary operators, just as in
2792 a cast). This construct is allowed regardless of what kind of data is
2793 officially supposed to reside at @var{addr}.@refill
2797 @section Program Variables
2799 The most common kind of expression to use is the name of a variable
2802 Variables in expressions are understood in the selected stack frame
2803 (@pxref{Selection}); they must either be global (or static) or be visible
2804 according to the scope rules of the programming language from the point of
2805 execution in that frame. This means that in the function
2820 the variable @code{a} is usable whenever the program is executing
2821 within the function @code{foo}, but the variable @code{b} is visible
2822 only while the program is executing inside the block in which @code{b}
2825 There is an exception: you can refer to a variable or function whose
2826 scope is a single source file even if the current execution point is not
2827 in this file. But it is possible to have more than one such variable or
2828 function with the same name (in different source files). If that happens,
2829 referring to that name has unpredictable effects. If you wish, you can
2830 specify a variable in a particular file, using the colon-colon construct:
2836 @var{block}::@var{variable}
2840 Here @var{block} is the name of the source file whose variable you want.
2842 @cindex name resolution (C++)
2843 Unfortunately, this use of @samp{::} conflicts with the very similar use
2844 of the same notation in C++; accordingly, _GDBN__ does not support use of
2845 the C++ name resolution operator in _GDBN__ expressions.
2848 @section Artificial Arrays
2850 @cindex artificial array
2852 It is often useful to print out several successive objects of the
2853 same type in memory; a section of an array, or an array of
2854 dynamically determined size for which only a pointer exists in the
2857 This can be done by constructing an @dfn{artificial array} with the
2858 binary operator @samp{@@}. The left operand of @samp{@@} should be
2859 the first element of the desired array, as an individual object.
2860 The right operand should be the length of the array. The result is
2861 an array value whose elements are all of the type of the left argument.
2862 The first element is actually the left argument; the second element
2863 comes from bytes of memory immediately following those that hold the
2864 first element, and so on. Here is an example. If a program says
2867 int *array = (int *) malloc (len * sizeof (int));
2871 you can print the contents of @code{array} with
2877 The left operand of @samp{@@} must reside in memory. Array values made
2878 with @samp{@@} in this way behave just like other arrays in terms of
2879 subscripting, and are coerced to pointers when used in expressions.
2880 Artificial arrays most often appear in expressions via the value history
2881 (@pxref{Value History}), after printing one out.)
2883 @node Output formats,,,
2884 @section Output formats
2886 @cindex formatted output
2887 @cindex output formats
2888 By default, _GDBN__ prints a value according to its data type. Sometimes
2889 this is not what you want. For example, you might want to print a number
2890 in hex, or a pointer in decimal. Or you might want to view data in memory
2891 at a certain address as a character string or as an instruction. To do
2892 these things, specify an @dfn{output format} when you print a value.
2894 The simplest use of output formats is to say how to print a value
2895 already computed. This is done by starting the arguments of the
2896 @samp{print} command with a slash and a format letter. The format
2897 letters supported are:
2901 Regard the bits of the value as an integer, and print the integer in
2905 Print as integer in signed decimal.
2908 Print as integer in unsigned decimal.
2911 Print as integer in octal.
2914 Print as an address, both absolute in hex and as an offset from the
2915 nearest preceding symbol. This format can be used to discover where (in
2916 what function) an unknown address is located:
2918 (_GDBP__) p/a 0x54320
2919 _0__$3 = 0x54320 <_initialize_vx+396>_1__
2924 Regard as an integer and print it as a character constant.
2927 Regard the bits of the value as a floating point number and print
2928 using typical floating point syntax.
2931 For example, to print the program counter in hex (@pxref{Registers}), type
2938 Note that no space is required before the slash; this is because command
2939 names in _GDBN__ cannot contain a slash.
2941 To reprint the last value in the value history with a different format,
2942 you can use the @samp{print} command with just a format and no
2943 expression. For example, @samp{p/x} reprints the last value in hex.
2946 @section Examining Memory
2948 @cindex examining memory
2951 @item x/@var{Nuf} @var{expr}
2952 The command @samp{x} (for `examine') can be used to examine memory
2953 without being constrained by your program's data types. You can specify
2954 the unit size @var{u} of memory to inspect, and a repeat count @var{N} of how
2955 many of those units to display. @samp{x} understands the formats
2956 @var{f} used by @samp{print}; two additional formats, @samp{s} (string)
2957 and @samp{i} (machine instruction) can be used without specifying a unit
2961 For example, @samp{x/3hu 0x54320} is a request to display three halfwords
2962 (@code{h}) of memory, formatted as unsigned decimal integers (@code{u}),
2963 starting at address @code{0x54320}. @samp{x/4wx $sp} prints the four
2964 words (@code{w}) of memory above the stack pointer (here, @samp{$sp};
2965 @pxref{Registers}) in hexadecimal (@code{x}).
2967 Since the letters indicating unit sizes are all distinct from the
2968 letters specifying output formats, you don't have to remember whether
2969 unit size or format comes first; either order will work. The output
2970 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
2972 After the format specification, you indicate the address where _GDBN__
2973 is to begin reading memory for display, using an expression. The
2974 expression need not have a pointer value (though it may); it is always
2975 interpreted as an integer address of a byte of memory.
2976 @xref{Expressions} for more information on expressions.
2978 These are the memory units @var{u} you can specify with @samp{x}:
2982 Examine individual bytes.
2985 Examine halfwords (two bytes each).
2988 Examine words (four bytes each).
2991 Many assemblers and cpu designers still use `word' for a 16-bit quantity,
2992 as a holdover from specific predecessor machines of the 1970's that really
2993 did use two-byte words. But more generally the term `word' has always
2994 referred to the size of quantity that a machine normally operates on and
2995 stores in its registers. This is 32 bits for all the machines that _GDBN__
2999 Examine giant words (8 bytes).
3002 You can combine these unit specifications with any of the formats
3003 described for @samp{print}. @xref{Output formats}.
3005 @samp{x} has two additional output specifications which derive the unit
3006 size from the data inspected:
3010 Print a null-terminated string of characters. Any explicitly specified
3011 unit size is ignored; instead, the unit is however many bytes it takes
3012 to reach a null character (including the null character).
3015 Print a machine instruction in assembler syntax (or nearly). Any
3016 specified unit size is ignored; the number of bytes in an instruction
3017 varies depending on the type of machine, the opcode and the addressing
3018 modes used. The command @samp{disassemble} gives an alternative way of
3019 inspecting machine instructions. @xref{Machine Code}.
3022 If you omit either the format @var{f} or the unit size @var{u}, @samp{x}
3023 will use the same one that was used last. If you don't use any letters
3024 after the slash, you can omit the slash as well.
3026 You can also omit the address to examine. Then the address used is just
3027 after the last unit examined. This is why string and instruction
3028 formats actually compute a unit-size based on the data: so that the next
3029 string or instruction examined will start in the right place.
3031 When the @samp{print} command shows a value that resides in memory,
3032 @samp{print} also sets the default address for the @samp{x} command.
3033 @samp{info line} also sets the default for @samp{x}, to the address of
3034 the start of the machine code for the specified line (@pxref{Machine
3035 Code}), and @samp{info breakpoints} sets it to the address of the last
3036 breakpoint listed (@pxref{Set Breaks}).
3038 When you use @key{RET} to repeat an @samp{x} command, the address
3039 specified previously (if any) is ignored, so that the repeated command
3040 examines the successive locations in memory rather than the same ones.
3042 You can examine several consecutive units of memory with one command by
3043 writing a repeat-count after the slash (before the format letters, if
3044 any). Omitting the repeat count @var{N} displays one unit of the
3045 appropriate size. The repeat count must be a decimal integer. It has
3046 the same effect as repeating the @samp{x} command @var{N} times except
3047 that the output may be more compact, with several units per line. For
3055 prints ten instructions starting with the one to be executed next in the
3056 selected frame. After doing this, you could print a further seven
3064 ---where the format and address are allowed to default.
3068 The addresses and contents printed by the @samp{x} command are not put
3069 in the value history because there is often too much of them and they
3070 would get in the way. Instead, _GDBN__ makes these values available for
3071 subsequent use in expressions as values of the convenience variables
3072 @code{$_} and @code{$__}. After an @samp{x} command, the last address
3073 examined is available for use in expressions in the convenience variable
3074 @code{$_}. The contents of that address, as examined, are available in
3075 the convenience variable @code{$__}.
3077 If the @samp{x} command has a repeat count, the address and contents saved
3078 are from the last memory unit printed; this is not the same as the last
3079 address printed if several units were printed on the last line of output.
3081 @node Auto Display,,,
3082 @section Automatic Display
3083 @cindex automatic display
3084 @cindex display of expressions
3086 If you find that you want to print the value of an expression frequently
3087 (to see how it changes), you might want to add it to the @dfn{automatic
3088 display list} so that _GDBN__ will print its value each time the program stops.
3089 Each expression added to the list is given a number to identify it;
3090 to remove an expression from the list, you specify that number.
3091 The automatic display looks like this:
3095 3: bar[5] = (struct hack *) 0x3804
3099 showing item numbers, expressions and their current values. As with
3100 displays you request manually using @samp{x} or @samp{print}, you can
3101 specify the output format you prefer; in fact, @dfn{display} decides
3102 whether to use @code{print} or @code{x} depending on how elaborate your
3103 format specification is---it uses @code{x} if you specify a unit size,
3104 or one of the two formats (@samp{i} and @samp{s}) that are only
3105 supported by @code{x}; otherwise it uses @code{print}.
3108 @item display @var{exp}
3110 Add the expression @var{exp} to the list of expressions to display
3111 each time the program stops. @xref{Expressions}.
3113 @samp{display} will not repeat if you press @key{RET} again after using it.
3115 @item display/@var{fmt} @var{exp}
3116 For @var{fmt} specifying only a display format and not a size or
3117 count, add the expression @var{exp} to the auto-display list but
3118 arranges to display it each time in the specified format @var{fmt}.
3119 @xref{Output formats}.
3121 @item display/@var{fmt} @var{addr}
3122 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
3123 number of units, add the expression @var{addr} as a memory address to
3124 be examined each time the program stops. Examining means in effect
3125 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory}.
3128 For example, @samp{display/i $pc} can be helpful, to see the machine
3129 instruction about to be executed each time execution stops (@samp{$pc}
3130 is a common name for the program counter; @pxref{Registers}).
3133 @item undisplay @var{dnums}@dots{}
3134 @itemx delete display @var{dnums}@dots{}
3135 @kindex delete display
3137 Remove item numbers @var{dnums} from the list of expressions to display.
3139 @samp{undisplay} will not repeat if you press @key{RET} after using it.
3141 @item disable display @var{dnums}@dots{}
3142 @kindex disable display
3143 Disable the display of item numbers @var{dnums}. A disabled display
3144 item is not printed automatically, but is not forgotten. It may be
3145 enabled again later.
3147 @item enable display @var{dnums}@dots{}
3148 @kindex enable display
3149 Enable display of item numbers @var{dnums}. It becomes effective once
3150 again in auto display of its expression, until you specify otherwise.
3153 Display the current values of the expressions on the list, just as is
3154 done when the program stops.
3157 @kindex info display
3158 Print the list of expressions previously set up to display
3159 automatically, each one with its item number, but without showing the
3160 values. This includes disabled expressions, which are marked as such.
3161 It also includes expressions which would not be displayed right now
3162 because they refer to automatic variables not currently available.
3165 If a display expression refers to local variables, then it does not make
3166 sense outside the lexical context for which it was set up. Such an
3167 expression is disabled when execution enters a context where one of its
3168 variables is not defined. For example, if you give the command
3169 @samp{display name} while inside a function with an argument
3170 @code{name}, then this argument will be displayed while the program
3171 continues to stop inside that function. When it stops elsewhere---where
3172 there is no variable @samp{name}---display is disabled. The next time
3173 your program stops where @samp{name} is meaningful, you can enable the
3174 display expression once again.
3176 @node Print Settings,,,
3177 @section Print Settings
3179 @cindex format options
3180 @cindex print settings
3181 _GDBN__ provides the following ways to control how arrays, structures,
3182 and symbols are printed.
3185 @item set array-max @var{number-of-elements}
3186 @kindex set array-max
3187 If _GDBN__ is printing a large array, it will stop printing after it has
3188 printed the number of elements set by the @samp{set array-max} command.
3189 This limit also applies to the display of strings.
3191 @item show array-max
3192 @kindex show array-max
3193 Display the number of elements of a large array that _GDBN__ will print
3194 before losing patience.
3196 @item set print array
3197 @itemx set print array on
3198 @kindex set print array
3199 _GDBN__ will pretty print arrays. This format is more convenient to read,
3200 but uses more space. The default is off.
3202 @item set print array off.
3203 Return to compressed format for arrays.
3205 @item show print array
3206 @kindex show print array
3207 Show whether compressed or pretty format is selected for displaying
3210 @item set print demangle
3211 @itemx set print demangle on
3212 @kindex set print demangle
3213 Print C++ names in their source form rather than in the mangled form
3214 in which they are passed to the assembler and linker for type-safe linkage.
3217 @item show print demangle
3218 @kindex show print demangle
3219 Show whether C++ names will be printed in mangled or demangled form.
3221 @item set print asm-demangle
3222 @itemx set print asm-demangle on
3223 @kindex set print asm-demangle
3224 Print C++ names in their source form rather than their mangled form, even
3225 in assembler code printouts such as instruction disassemblies.
3228 @item show print asm-demangle
3229 @kindex show print asm-demangle
3230 Show whether C++ names in assembly listings will be printed in mangled
3233 @item set print vtbl
3234 @itemx set print vtbl on
3235 @kindex set print vtbl
3236 Pretty print C++ virtual function tables. The default is off.
3238 @item set print vtbl off
3239 Do not pretty print C++ virtual function tables.
3241 @item show print vtbl
3242 @kindex show print vtbl
3243 Show whether C++ virtual function tables are pretty printed, or not.
3245 @item set print address
3246 @item set print address on
3247 @kindex set print address
3248 _GDBN__ will print memory addresses in stack traces, structure values, pointer
3249 values, breakpoints, etc. The default is on.
3251 @item set print address off
3252 Do not print addresses.
3254 @item show print address
3255 @kindex show print address
3256 Show whether or not addresses are to be printed.
3258 @item set print pretty on
3259 @kindex set print pretty
3260 Cause _GDBN__ to print structures in an indented format with one member per
3274 @item set print pretty off
3275 Cause _GDBN__ to print structures in a compact format, like this:
3278 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, meat \
3283 This is the default format.
3285 @item show print pretty
3286 @kindex show print pretty
3287 Show which format _GDBN__ will use to print structures.
3289 @item set print union on
3290 @kindex set print union
3291 Tell _GDBN__ to print unions which are contained in structures. This is the
3294 @item set print union off
3295 Tell _GDBN__ not to print unions which are contained in structures.
3297 @item show print union
3298 @kindex show print union
3299 Ask _GDBN__ whether or not it will print unions which are contained in
3302 For example, given the declarations
3305 typedef enum @{Tree, Bug@} Species;
3306 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
3307 typedef enum @{Caterpillar, Cocoon, Butterfly@} Bug_forms;
3317 struct thing foo = @{Tree, @{Acorn@}@};
3321 with @samp{set print union on} in effect @samp{p foo} would print
3324 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
3328 and with @samp{set print union off} in effect it would print
3331 $1 = @{it = Tree, form = @{...@}@}
3335 @node Value History,,,
3336 @section Value History
3338 @cindex value history
3339 Values printed by the @samp{print} command are saved in _GDBN__'s @dfn{value
3340 history} so that you can refer to them in other expressions. Values are
3341 kept until the symbol table is re-read or discarded (for example with
3342 the @samp{file} or @samp{symbol-file} commands). When the symbol table
3343 changes, the value history is discarded, since the values may contain
3344 pointers back to the types defined in the symbol table.
3348 @cindex history number
3349 The values printed are given @dfn{history numbers} for you to refer to them
3350 by. These are successive integers starting with 1. @samp{print} shows you
3351 the history number assigned to a value by printing @samp{$@var{num} = }
3352 before the value; here @var{num} is the history number.
3354 To refer to any previous value, use @samp{$} followed by the value's
3355 history number. The way @samp{print} labels its output is designed to
3356 remind you of this. Just @code{$} refers to the most recent value in
3357 the history, and @code{$$} refers to the value before that.
3358 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
3359 is the value just prior to @code{$$}, @code{$$1} is equivalent to
3360 @code{$$}, and @code{$$0} is equivalent to @code{$}.
3362 For example, suppose you have just printed a pointer to a structure and
3363 want to see the contents of the structure. It suffices to type
3369 If you have a chain of structures where the component @samp{next} points
3370 to the next one, you can print the contents of the next one with this:
3377 You can print successive links in the chain by repeating this
3378 command---which you can do by just typing @key{RET}.
3380 Note that the history records values, not expressions. If the value of
3381 @code{x} is 4 and you type these commands:
3389 then the value recorded in the value history by the @samp{print} command
3390 remains 4 even though the value of @code{x} has changed.
3395 Print the last ten values in the value history, with their item numbers.
3396 This is like @samp{p@ $$9} repeated ten times, except that @samp{show
3397 values} does not change the history.
3399 @item show values @var{n}
3400 Print ten history values centered on history item number @var{n}.
3403 Print ten history values just after the values last printed.
3406 @node Convenience Vars,,,
3407 @section Convenience Variables
3409 @cindex convenience variables
3410 _GDBN__ provides @dfn{convenience variables} that you can use within
3411 _GDBN__ to hold on to a value and refer to it later. These variables
3412 exist entirely within _GDBN__; they are not part of your program, and
3413 setting a convenience variable has no direct effect on further execution
3414 of your program. That's why you can use them freely.
3416 Convenience variables have names starting with @samp{$}. Any name starting
3417 with @samp{$} can be used for a convenience variable, unless it is one of
3418 the predefined machine-specific register names (@pxref{Registers}).
3420 You can save a value in a convenience variable with an assignment
3421 expression, just as you would set a variable in your program. Example:
3424 set $foo = *object_ptr
3428 would save in @code{$foo} the value contained in the object pointed to by
3431 Using a convenience variable for the first time creates it; but its value
3432 is @code{void} until you assign a new value. You can alter the value with
3433 another assignment at any time.
3435 Convenience variables have no fixed types. You can assign a convenience
3436 variable any type of value, including structures and arrays, even if
3437 that variable already has a value of a different type. The convenience
3438 variable, when used as an expression, has the type of its current value.
3441 @item show convenience
3442 @kindex show convenience
3443 Print a list of convenience variables used so far, and their values.
3444 Abbreviated @samp{i con}.
3447 One of the ways to use a convenience variable is as a counter to be
3448 incremented or a pointer to be advanced. For example, to print
3449 a field from successive elements of an array of structures:
3453 print bar[$i++]->contents
3454 @i{@dots{} repeat that command by typing @key{RET}.}
3457 Some convenience variables are created automatically by _GDBN__ and given
3458 values likely to be useful.
3462 The variable @code{$_} is automatically set by the @samp{x} command to
3463 the last address examined (@pxref{Memory}). Other commands which
3464 provide a default address for @samp{x} to examine also set @code{$_}
3465 to that address; these commands include @samp{info line} and @samp{info
3469 The variable @code{$__} is automatically set by the @samp{x} command
3470 to the value found in the last address examined.
3477 Machine register contents can be referred to in expressions as variables
3478 with names starting with @samp{$}. The names of registers are different
3479 for each machine; use @samp{info registers} to see the names used on
3480 your machine. The names @code{$pc} and @code{$sp} are used on most
3481 machines for the program counter register and the stack pointer. Often
3482 @code{$fp} is used for a register that contains a pointer to the current
3483 stack frame, and @code{$ps} is sometimes used for a register that
3484 contains the processor status. These standard register names may be
3485 available on your machine even though the @code{info registers} command
3486 shows other names. For example, on the SPARC, @code{info registers}
3487 displays the processor status register as @code{$psr} but you can also
3488 refer to it as @code{$ps}.
3490 _GDBN__ always considers the contents of an ordinary register as an
3491 integer when the register is examined in this way. Some machines have
3492 special registers which can hold nothing but floating point; these
3493 registers are considered to have floating point values. There is no way
3494 to refer to the contents of an ordinary register as floating point value
3495 (although you can @emph{print} it as a floating point value with
3496 @samp{print/f $@var{regname}}).
3498 Some registers have distinct ``raw'' and ``virtual'' data formats. This
3499 means that the data format in which the register contents are saved by
3500 the operating system is not the same one that your program normally
3501 sees. For example, the registers of the 68881 floating point
3502 coprocessor are always saved in ``extended'' (raw) format, but all C
3503 programs expect to work with ``double'' (virtual) format. In such
3504 cases, _GDBN__ normally works with the virtual format only (the format that
3505 makes sense for your program), but the @samp{info registers} command
3506 prints the data in both formats.
3508 Normally, register values are relative to the selected stack frame
3509 (@pxref{Selection}). This means that you get the value that the register
3510 would contain if all stack frames farther in were exited and their saved
3511 registers restored. In order to see the contents of hardware registers,
3512 you must select the innermost frame (with @samp{frame 0}).
3514 However, _GDBN__ must deduce where registers are saved, from the machine
3515 code generated by your compiler. If some registers are not saved, or if
3516 _GDBN__ is unable to locate the saved registers, the selected stack
3517 frame will make no difference.
3520 @item info registers
3521 @kindex info registers
3522 Print the names and values of all registers (in the selected stack frame).
3524 @item info registers @var{regname}
3525 Print the relativized value of register @var{regname}. @var{regname}
3526 may be any register name valid on the machine you are using, with
3527 or without the initial @samp{$}.
3530 For example, you could print the program counter in hex with
3537 or print the instruction to be executed next with
3544 or add four to the stack pointer with
3551 The last is a way of removing one word from the stack, on machines where
3552 stacks grow downward in memory (most machines, nowadays). This assumes
3553 that the innermost stack frame is selected. Setting @code{$sp} is
3554 not allowed when other stack frames are selected. (To pop entire frames
3555 off the stack, regardless of machine architecture, use @samp{return};
3559 @chapter Examining the Symbol Table
3561 The commands described in this section allow you to inquire about the
3562 symbols (names of variables, functions and types) defined in your
3563 program. This information is inherent in the text of your program and
3564 does not change as the program executes. _GDBN__ finds it in your
3565 program's symbol table, as indicated when you started _GDBN__
3566 (@pxref{File Options}), or by one of the file-management commands
3570 @item info address @var{symbol}
3571 @kindex info address
3572 Describe where the data for @var{symbol} is stored. For a register
3573 variable, this says which register it is kept in. For a non-register
3574 local variable, this prints the stack-frame offset at which the variable
3577 Note the contrast with @samp{print &@var{symbol}}, which does not work
3578 at all for a register variables, and for a stack local variable prints
3579 the exact address of the current instantiation of the variable.
3581 @item whatis @var{exp}
3583 Print the data type of expression @var{exp}. @var{exp} is not
3584 actually evaluated, and any side-effecting operations (such as
3585 assignments or function calls) inside it do not take place.
3589 Print the data type of @code{$}, the last value in the value history.
3591 @item ptype @var{typename}
3593 Print a description of data type @var{typename}. @var{typename} may be
3594 the name of a type, or for C code it may have the form
3595 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
3596 @samp{enum @var{enum-tag}}.@refill
3598 @item ptype @var{exp}
3599 Print a description of the type of expression @var{exp}. @samp{ptype}
3600 differs from @samp{whatis} by printing a detailed description, instead of just
3601 the name of the type. For example, if your program declares a variable
3604 struct complex {double real; double imag;} v;
3607 compare the output of the two commands:
3610 type = struct complex
3612 type = struct complex {
3619 @kindex info sources
3620 Print the names of all source files in the program for which there
3621 is debugging information.
3623 @item info functions
3624 @kindex info functions
3625 Print the names and data types of all defined functions.
3627 @item info functions @var{regexp}
3628 Print the names and data types of all defined functions
3629 whose names contain a match for regular expression @var{regexp}.
3630 Thus, @samp{info fun step} finds all functions whose names
3631 include @samp{step}; @samp{info fun ^step} finds those whose names
3632 start with @samp{step}.
3634 @item info variables
3635 @kindex info variables
3636 Print the names and data types of all variables that are declared
3637 outside of functions (i.e., excluding local variables).
3639 @item info variables @var{regexp}
3640 Print the names and data types of all variables (except for local
3641 variables) whose names contain a match for regular expression
3646 This was never implemented.
3648 @itemx info methods @var{regexp}
3649 @kindex info methods
3650 The @samp{info-methods} command permits the user to examine all defined
3651 methods within C++ program, or (with the @var{regexp} argument) a
3652 specific set of methods found in the various C++ classes. Many
3653 C++ classes provide a large number of methods. Thus, the output
3654 from the @samp{ptype} command can be overwhelming and hard to use. The
3655 @samp{info-methods} command filters the methods, printing only those
3656 which match the regular-expression @var{regexp}.
3659 @item printsyms @var{filename}
3661 Write a complete dump of the debugger's symbol data into the
3662 file @var{filename}.
3666 @chapter Altering Execution
3668 Once you think you have found an error in the program, you might want to
3669 find out for certain whether correcting the apparent error would lead to
3670 correct results in the rest of the run. You can find the answer by
3671 experiment, using the _GDBN__ features for altering execution of the
3674 For example, you can store new values into variables or memory
3675 locations, give the program a signal, restart it at a different address,
3676 or even return prematurely from a function to its caller.
3679 @section Assignment to Variables
3682 @cindex setting variables
3683 To alter the value of a variable, evaluate an assignment expression.
3684 @xref{Expressions}. For example,
3691 would store the value 4 into the variable @code{x}, and then print the
3692 value of the assignment expression (which is 4). All the assignment
3693 operators of C are supported, including the increment operators
3694 @samp{++} and @samp{--}, and combining assignments such as @samp{+=} and
3698 @kindex set variable
3699 @cindex variables, setting
3700 If you are not interested in seeing the value of the assignment, use the
3701 @samp{set} command instead of the @samp{print} command. @samp{set} is
3702 really the same as @samp{print} except that the expression's value is not
3703 printed and is not put in the value history (@pxref{Value History}). The
3704 expression is evaluated only for its effects.
3706 If the beginning of the argument string of the @samp{set} command
3707 appears identical to a @samp{set} subcommand, use the @samp{set
3708 variable} command instead of just @samp{set}. This command is identical
3709 to @samp{set} except for its lack of subcommands.
3711 _GDBN__ allows more implicit conversions in assignments than C does; you can
3712 freely store an integer value into a pointer variable or vice versa, and
3713 any structure can be converted to any other structure that is the same
3715 @comment FIXME: how do structs align/pad in these conversions?
3716 @comment /pesch@cygnus.com 18dec1990
3718 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
3719 construct to generate a value of specified type at a specified address
3720 (@pxref{Expressions}). For example, @code{@{int@}0x83040} refers
3721 to memory location @code{0x83040} as an integer (which implies a certain size
3722 and representation in memory), and
3725 set @{int@}0x83040 = 4
3729 stores the value 4 into that memory location.
3732 @section Continuing at a Different Address
3734 Ordinarily, when you continue the program, you do so at the place where
3735 it stopped, with the @samp{continue} command. You can instead continue at
3736 an address of your own choosing, with the following commands:
3739 @item jump @var{linenum}
3741 Resume execution at line number @var{linenum}. Execution will stop
3742 immediately if there is a breakpoint there.
3744 The @samp{jump} command does not change the current stack frame, or
3745 the stack pointer, or the contents of any memory location or any
3746 register other than the program counter. If line @var{linenum} is in
3747 a different function from the one currently executing, the results may
3748 be bizarre if the two functions expect different patterns of arguments or
3749 of local variables. For this reason, the @samp{jump} command requests
3750 confirmation if the specified line is not in the function currently
3751 executing. However, even bizarre results are predictable if you are
3752 well acquainted with the machine-language code of the program.
3754 @item jump *@var{address}
3755 Resume execution at the instruction at address @var{address}.
3758 You can get much the same effect as the @code{jump} command by storing a
3759 new value into the register @code{$pc}. The difference is that this
3760 does not start the program running; it only changes the address where it
3761 @emph{will} run when it is continued. For example,
3768 causes the next @samp{continue} command or stepping command to execute at
3769 address 0x485, rather than at the address where the program stopped.
3772 The most common occasion to use the @samp{jump} command is to back up,
3773 perhaps with more breakpoints set, over a portion of a program that has
3778 @section Giving the Program a Signal
3781 @item signal @var{signalnum}
3783 Resume execution where the program stopped, but give it immediately the
3784 signal number @var{signalnum}.
3786 Alternatively, if @var{signalnum} is zero, continue execution without
3787 giving a signal. This is useful when the program stopped on account of
3788 a signal and would ordinary see the signal when resumed with the
3789 @samp{continue} command; @samp{signal 0} causes it to resume without a
3792 This command does not repeat when you press @key{RET} a second time
3793 after using it once.
3798 @section Returning from a Function
3802 @cindex returning from a function
3804 You can cancel execution of a function call with the @samp{return}
3808 This command abandons execution of a function. When you use
3809 @code{return}, _GDBN__ discards the selected stack frame (and all frames
3810 within it). You can think of this as making the discarded frame return
3811 prematurely. If you wish to specify a value to be returned, give that
3812 value as the argument to @code{return}.
3814 This pops the selected stack frame (@pxref{Selection}), and any other
3815 frames inside of it, leaving its caller as the innermost remaining
3816 frame. That frame becomes selected. The specified value is stored in
3817 the registers used for returning values of functions.
3819 The @samp{return} command does not resume execution; it leaves the
3820 program stopped in the state that would exist if the function had just
3821 returned. In contrast, the @samp{finish} command (@pxref{Stepping})
3822 resumes execution until the selected stack frame returns naturally.
3825 @section Calling your Program's Functions
3827 @cindex calling functions
3830 @item call @var{expr}
3831 Evaluate the expression @var{expr} without displaying @code{void}
3835 You can use this variant of the @samp{print} command if you want to
3836 execute a function from your program, but without cluttering the output
3837 with @code{void} returned values. The result is printed and saved in
3838 the value history, if it is not void.
3841 @chapter _GDBN__'s Files
3844 @section Commands to Specify Files
3845 @cindex core dump file
3846 @cindex symbol table
3847 _GDBN__ needs to know the file name of the program to be debugged, both in
3848 order to read its symbol table and in order to start the program. To
3849 debug a core dump of a previous run, _GDBN__ must be told the file name of
3852 The usual way to specify the executable and core dump file names is with
3853 the command arguments given when you start _GDBN__, as discussed in
3856 Occasionally it is necessary to change to a different file during a
3857 _GDBN__ session. Or you may run _GDBN__ and forget to specify the files you
3858 want to use. In these situations the _GDBN__ commands to specify new files
3862 @item file @var{filename}
3863 @cindex executable file
3865 Use @var{filename} as the program to be debugged. It is read for its
3866 symbols and for the contents of pure memory. It is also the program
3867 executed when you use the @samp{run} command. If you do not specify a
3868 directory and the file is not found in _GDBN__'s working directory,
3870 _GDBN__ will use the environment variable @code{PATH} as a list of
3871 directories to search, just as the shell does when looking for a program
3872 to run. You can change the value of this variable, for both _GDBN__ and
3873 your program, using the @code{path} command.
3875 @samp{file} with no argument makes _GDBN__ discard any information it
3876 has on both executable file and the symbol table.
3878 @item exec-file @var{filename}
3880 Specify that the program to be run (but not the symbol table) is found
3881 in @var{filename}. _GDBN__ will search the environment variable @code{PATH}
3882 if necessary to locate the program.
3884 @item symbol-file @var{filename}
3886 Read symbol table information from file @var{filename}. @code{PATH} is
3887 searched when necessary. Use the @samp{file} command to get both symbol
3888 table and program to run from the same file.
3890 @samp{symbol-file} with no argument clears out _GDBN__'s information on your
3891 program's symbol table.
3893 The @samp{symbol-file} command causes _GDBN__ to forget the contents of its
3894 convenience variables, the value history, and all breakpoints and
3895 auto-display expressions. This is because they may contain pointers to
3896 the internal data recording symbols and data types, which are part of
3897 the old symbol table data being discarded inside _GDBN__.
3899 @samp{symbol-file} will not repeat if you press @key{RET} again after
3902 On some kinds of object files, the @samp{symbol-file} command does not
3903 actually read the symbol table in full right away. Instead, it scans
3904 the symbol table quickly to find which source files and which symbols
3905 are present. The details are read later, one source file at a time,
3906 when they are needed.
3908 The purpose of this two-stage reading strategy is to make _GDBN__ start up
3909 faster. For the most part, it is invisible except for occasional pauses
3910 while the symbol table details for a particular source file are being
3911 read. (The @samp{set verbose} command can turn these pauses into
3912 messages if desired. @xref{Messages/Warnings}).
3914 When the symbol table is stored in COFF format, @samp{symbol-file} does
3915 read the symbol table data in full right away. We haven't implemented
3916 the two-stage strategy for COFF yet.
3918 When _GDBN__ is configured for a particular environment, it will
3919 understand debugging information in whatever format is the standard
3920 generated for that environment; you may use either a GNU compiler, or
3921 other compilers that adhere to the local conventions. Best results are
3922 usually obtained from GNU compilers; for example, using @code{_GCC__}
3923 you can generate debugging information for optimized code.
3925 @item core-file @var{filename}
3926 @itemx core @var{filename}
3929 Specify the whereabouts of a core dump file to be used as the ``contents
3930 of memory''. Traditionally, core files contain only some parts of the
3931 address space of the process that generated them; _GDBN__ can access the
3932 executable file itself for other parts.
3934 @samp{core-file} with no argument specifies that no core file is
3937 Note that the core file is ignored when your program is actually running
3938 under _GDBN__. So, if you have been running the program and you wish to
3939 debug a core file instead, you must kill the subprocess in which the
3940 program is running. To do this, use the @samp{kill} command
3941 (@pxref{Kill Process}).
3943 @item load @var{filename}
3946 Depending on what remote debugging facilities are configured into
3947 _GDBN__, the @samp{load} command may be available. Where it exists, it
3948 is meant to make @var{filename} (an executable) available for debugging
3949 on the remote system---by downloading, or dynamic linking, for example.
3950 @samp{load} also records @var{filename}'s symbol table in _GDBN__, like
3951 the @samp{add-symbol-file} command.
3953 If @samp{load} is not available on your _GDBN__, attempting to execute
3954 it gets the error message ``@code{You can't do that when your target is
3959 On VxWorks, @samp{load} will dynamically link @var{filename} on the
3960 current target system as well as adding its symbols in _GDBN__.
3964 With the Nindy interface to an Intel 960 board, @samp{load} will
3965 download @var{filename} to the 960 as well as adding its symbols in
3969 @samp{load} will not repeat if you press @key{RET} again after using it.
3971 @item add-symbol-file @var{filename} @var{address}
3972 @kindex add-symbol-file
3973 @cindex dynamic linking
3974 The @samp{add-symbol-file} command reads additional symbol table information
3975 from the file @var{filename}. You would use this command when that file
3976 has been dynamically loaded (by some other means) into the program that
3977 is running. @var{address} should be the memory address at which the
3978 file has been loaded; _GDBN__ cannot figure this out for itself.
3980 The symbol table of the file @var{filename} is added to the symbol table
3981 originally read with the @samp{symbol-file} command. You can use the
3982 @samp{add-symbol-file} command any number of times; the new symbol data thus
3983 read keeps adding to the old. To discard all old symbol data instead,
3984 use the @samp{symbol-file} command.
3986 @samp{add-symbol-file} will not repeat if you press @key{RET} after using it.
3992 @samp{info files} and @samp{info target} are synonymous; both print the
3993 current targets (@pxref{Targets}), including the names of the executable
3994 and core dump files currently in use by _GDBN__, and the files from
3995 which symbols were loaded. The command @samp{help targets} lists all
3996 possible targets rather than current ones.
4000 All file-specifying commands allow both absolute and relative file names
4001 as arguments. _GDBN__ always converts the file name to an absolute path
4002 name and remembers it that way.
4004 @kindex sharedlibrary
4006 @cindex shared libraries
4008 _GDBN__ supports the SunOS shared library format. Symbols from a shared
4009 library cannot be referenced before the shared library has been linked
4010 with the program. (That is to say, until after you type @samp{run} and
4011 the function @code{main} has been entered; or when examining core
4012 files.) Once the shared library has been linked in, you can use the
4016 @item sharedlibrary @var{regex}
4017 @itemx share @var{regex}
4018 Load shared object library symbols for files matching a UNIX regular
4022 @itemx sharedlibrary
4023 Load symbols for all shared libraries.
4026 @itemx info sharedlibrary
4027 @kindex info sharedlibrary
4029 Print the names of the shared libraries which are currently loaded.
4032 @samp{sharedlibrary} does not repeat automatically when you press
4033 @key{RET} after using it once.
4035 @node Symbol Errors,,,
4036 @section Errors Reading Symbol Files
4037 While a symbol file is being read, _GDBN__ will occasionally encounter
4038 problems, such as symbol types it does not recognize, or known bugs in
4039 compiler output. By default, it prints one message about each such
4040 type of problem, no matter how many times the problem occurs. You can
4041 ask it to print more messages, to see how many times the problems occur,
4042 or can shut the messages off entirely, with the @samp{set
4043 complaints} command (@xref{Messages/Warnings}).
4045 The messages currently printed, and their meanings, are:
4048 @item inner block not inside outer block in @var{symbol}
4050 The symbol information shows where symbol scopes begin and end
4051 (such as at the start of a function or a block of statements). This
4052 error indicates that an inner scope block is not fully contained
4053 in its outer scope blocks.
4055 _GDBN__ circumvents the problem by treating the inner block as if it had
4056 the same scope as the outer block. In the error message, @var{symbol}
4057 may be shown as ``@code{(don't know)}'' if the outer block is not a
4060 @item block at @var{address} out of order
4062 The symbol information for symbol scope blocks should occur in
4063 order of increasing addresses. This error indicates that it does not
4066 _GDBN__ does not circumvent this problem, and will have trouble locating
4067 symbols in the source file whose symbols being read. (You can often
4068 determine what source file is affected by specifying @samp{set verbose
4069 on}. @xref{Messages/Warnings}.)
4071 @item bad block start address patched
4073 The symbol information for a symbol scope block has a start address
4074 smaller than the address of the preceding source line. This is known
4075 to occur in the SunOS 4.1.1 (and earlier) C compiler.
4077 _GDBN__ circumvents the problem by treating the symbol scope block as
4078 starting on the previous source line.
4080 @c @item{encountered DBX-style class variable debugging information.
4081 @c You seem to have compiled your program with "g++ -g0" instead of "g++ -g".
4082 @c Therefore _GDBN__ will not know about your class variables}
4084 @c This error indicates that the symbol information produced for a C++
4085 @c program includes zero-size fields, which indicated static fields in
4086 @c a previous release of the G++ compiler. This message is probably
4089 @item bad string table offset in symbol @var{n}
4092 Symbol number @var{n} contains a pointer into the string table which is
4093 larger than the size of the string table.
4095 _GDBN__ circumvents the problem by considering the symbol to have the
4096 name @code{foo}, which may cause other problems if many symbols end up
4099 @item unknown symbol type @code{0x@var{NN}}
4101 The symbol information contains new data types that _GDBN__ does not yet
4102 know how to read. @code{0x@var{NN}} is the symbol type of the misunderstood
4103 information, in hexadecimal.
4105 _GDBN__ circumvents the error by ignoring this symbol information. This
4106 will usually allow the program to be debugged, though certain symbols
4107 will not be accessible. If you encounter such a problem and feel like
4108 debugging it, you can debug @code{_GDBP__} with itself, breakpoint on
4109 @samp{complain}, then go up to the function @samp{read_dbx_symtab} and
4110 examine @code{*bufp} to see the symbol.
4112 @c @item stub type has NULL name
4114 @c FIXME, Mike Tiemann needs to write about what this means.
4116 @item const/volatile indicator missing, got 'X'
4118 The symbol information for a C++ type is missing some information that
4119 the compiler should have output for it.
4121 @item C++ type mismatch between compiler and debugger
4123 The debugger could not parse a type specification output by the compiler
4124 for some C++ object.
4129 @chapter Specifying a Debugging Target
4130 @cindex debugging target
4132 A @dfn{target} is an interface between the debugger and a particular
4133 kind of file or process.
4135 Often, you will be able to run _GDBN__ in the same host environment as the
4136 program you are debugging; in that case, the debugging target can just be
4137 specified as a side effect of the @samp{file} or @samp{core} commands.
4138 When you need more flexibility---for example, running _GDBN__ on a
4139 physically separate host, controlling standalone systems over a
4140 serial port, or realtime systems over a TCP/IP connection---you can use
4141 the @samp{target} command.
4143 @node Active Targets,,,
4144 @section Active Targets
4145 @cindex stacking targets
4146 @cindex active targets
4147 @cindex multiple targets
4149 Targets are managed in three @dfn{strata} that correspond to different
4150 classes of target: processes, core files, and executable files. This
4151 allows you to (for example) start a process and inspect its activity
4152 without abandoning your work on a core file.
4154 More than one target can potentially respond to a request. In
4155 particular, when you access memory _GDBN__ will examine the three strata of
4156 targets until it finds a target that can handle that particular address.
4158 Strata are always examined in a fixed order: first a process if there is
4159 one, then a core file if there is one, and finally an executable file if
4160 there is one of those.
4162 When you specify a new target in a given stratum, it replaces any target
4163 previously in that stratum.
4165 To get rid of a target without replacing it, use the @samp{detach}
4166 command. The related command @samp{attach} provides you with a way of
4167 choosing a particular running process as a new target. @xref{Attach}.
4169 @node Target Commands,,,
4170 @section Commands for Managing Targets
4173 @item target @var{type} @var{parameters}
4174 Connects the _GDBN__ host environment to a target machine or process. A
4175 target is typically a protocol for talking to debugging facilities. You
4176 use the argument @var{type} to specify the type or protocol of the
4179 Further @var{parameters} are interpreted by the target protocol, but
4180 typically include things like device names or host names to connect
4181 with, process numbers, and baud rates.
4183 The @samp{target} command will not repeat if you press @key{RET} again
4184 after executing the command.
4187 @kindex help targets
4188 Displays the names of all targets available. To display targets
4189 currently selected, use either @samp{info target} or @samp{info files}
4192 @item help target @var{name}
4193 Describe a particular target, including any parameters necessary to
4197 Here are some common targets (available, or not, depending on the _GDBN__
4201 @item target exec @var{prog}
4203 An executable file. @samp{target exec @var{prog}} is the same as
4204 @samp{exec-file @var{prog}}.
4206 @item target core @var{filename}
4208 A core dump file. @samp{target core @var{filename}} is the same as
4209 @samp{core-file @var{filename}}.
4211 @item target remote @var{dev}
4212 @kindex target remote
4213 Remote serial target in _GDBN__-specific protocol. The argument @var{dev}
4214 specifies what serial device to use for the connection (e.g.
4215 @code{/dev/ttya}). @xref{Remote}.
4218 @item target amd-eb @var{dev} @var{speed} @var{PROG}
4219 @kindex target amd-eb
4221 Remote PC-resident AMD EB29K board, attached over serial lines.
4222 @var{dev} is the serial device, as for @samp{target remote};
4223 @samp{speed} allows you to specify the linespeed; and @var{PROG} is the
4224 name of the program to be debugged, as it appears to DOS on the PC.
4225 @xref{EB29K Remote}.
4229 @item target nindy @var{devicename}
4230 @kindex target nindy
4231 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
4232 the name of the serial device to use for the connection, e.g.
4233 @samp{/dev/ttya}. @xref{i960-Nindy Remote}.
4237 @item target vxworks @var{machinename}
4238 @kindex target vxworks
4239 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
4240 is the target system's machine name or IP address.
4241 @xref{VxWorks Remote}.
4246 Different targets are available on different configurations of _GDBN__; your
4247 configuration may have more or fewer targets.
4251 @section Remote Debugging
4252 @cindex remote debugging
4254 If you are trying to debug a program running on a machine that can't run
4255 _GDBN__ in the usual way, it is often useful to use remote debugging. For
4256 example, you might be debugging an operating system kernel, or debugging
4257 a small system which does not have a general purpose operating system
4258 powerful enough to run a full-featured debugger.
4260 Some configurations of _GDBN__ have special serial or TCP/IP interfaces
4261 to make this work with particular debugging targets. In addition,
4262 _GDBN__ comes with a generic serial protocol (specific to _GDBN__, but
4263 not specific to any particular target system) which you can adapt.
4265 To use the _GDBN__ remote serial protocol, the program to be debugged on
4266 the remote machine needs to contain a debugging device driver which
4267 talks to _GDBN__ over the serial line. Several sample remote debugging
4268 drivers are distributed with _GDBN__; see the @file{README} file in the
4269 _GDBN__ distribution for more information.
4271 For details of this communication protocol, see the comments in the
4272 _GDBN__ source file @file{remote.c}.
4274 To start remote debugging, first run _GDBN__ and specify as an executable file
4275 the program that is running in the remote machine. This tells _GDBN__ how
4276 to find the program's symbols and the contents of its pure text. Then
4277 establish communication using the @samp{target remote} command with a device
4278 name as an argument. For example:
4281 target remote /dev/ttyb
4285 if the serial line is connected to the device named @file{/dev/ttyb}. This
4286 will stop the remote machine if it is not already stopped.
4288 Now you can use all the usual commands to examine and change data and to
4289 step and continue the remote program.
4291 To resume the remote program and stop debugging it, use the @samp{detach}
4294 Other remote targets may be available in your
4295 configuration of _GDBN__; use @samp{help targets} to list them.
4297 @node Controlling _GDBN__,,,
4298 @chapter Controlling _GDBN__
4300 You can alter many aspects of _GDBN__'s interaction with you by using
4301 the @samp{set} command. For commands controlling how _GDBN__ displays
4302 data, @pxref{Print Settings}; other settings are described here.
4307 _GDBN__ indicates its readiness to read a command by printing a string
4308 called the @dfn{prompt}. This string is normally @samp{(_GDBP__)}. You
4309 can change the prompt string with the @samp{set prompt} command. For
4310 instance, when debugging _GDBN__ with _GDBN__, it is useful to change
4311 the prompt in one of the _GDBN__<>s so that you tell which one you are
4315 @item set prompt @var{newprompt}
4317 Directs _GDBN__ to use @var{newprompt} as its prompt string henceforth.
4320 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
4324 @section Command Editing
4326 @cindex command line editing
4327 _GDBN__ reads its input commands via the @dfn{readline} interface. This
4328 GNU library provides consistent behavior for programs which provide a
4329 command line interface to the user. Advantages are @samp{emacs}-style
4330 or @samp{vi}-style inline editing of commands, @samp{csh}-like history
4331 substitution, and a storage and recall of command history across
4334 You may control the behavior of command line editing in _GDBN__ with the
4335 command @samp{set}. You may check the status of any of these settings
4336 with the command @samp{show}.
4342 @itemx set editing on
4343 Enable command line editing (enabled by default).
4345 @item set editing off
4346 Disable command line editing.
4348 @kindex show editing
4350 Show whether command line editing is enabled.
4353 @section Command History
4354 @cindex history substitution
4355 @cindex history file
4356 @kindex set history file
4357 @item set history file @var{filename}
4358 Set the name of the _GDBN__ command history file to @var{filename}. This is
4359 the file from which _GDBN__ will read an initial command history
4360 list or to which it will write this list when it exits. This list is
4361 accessed through history expansion or through the history
4362 command editing characters listed below. This file defaults to the
4363 value of the environment variable @code{GDBHISTFILE}, or to
4364 @file{./.gdb_history} if this variable is not set.
4366 @cindex history write
4367 @kindex set history write
4368 @item set history write
4369 @itemx set history write on
4370 Record command history in a file, whose name may be specified with the
4371 @samp{set history file} command. By default, this option is disabled.
4373 @item set history write off
4374 Stop recording command history in a file.
4376 @cindex history size
4377 @kindex set history size
4378 @item set history size @var{size}
4379 Set the number of commands which _GDBN__ will keep in its history list.
4380 This defaults to the value of the environment variable
4381 @code{HISTSIZE}, or to 256 if this variable is not set.
4384 @cindex history expansion
4385 History expansion assigns special meaning to the character @samp{!}.
4387 (@xref{Event Designators}.)
4389 Since @samp{!} is also the logical not operator in C, history expansion
4390 is off by default. If you decide to enable history expansion with the
4391 @samp{set history expansion on} command, you may sometimes need to
4392 follow @samp{!} (when it is used as logical not, in an expression) with
4393 a space or a tab to prevent it from being expanded. The readline
4394 history facilities will not attempt substitution on the strings
4395 @samp{!=} and @samp{!(}, even when history expansion is enabled.
4397 The commands to control history expansion are:
4401 @kindex set history expansion
4402 @item set history expansion on
4403 @itemx set history expansion
4404 Enable history expansion. History expansion is off by default.
4406 @item set history expansion off
4407 Disable history expansion.
4409 The readline code comes with more complete documentation of
4410 editing and history expansion features. Users unfamiliar with @samp{emacs}
4411 or @samp{vi} may wish to read it.
4413 @xref{Command Line Editing}.
4417 @kindex show history
4419 @itemx show history file
4420 @itemx show history write
4421 @itemx show history size
4422 @itemx show history expansion
4423 These commands display the state of the _GDBN__ history parameters.
4424 @samp{show history} by itself displays all four states.
4430 @kindex show commands
4432 Display the last ten commands in the command history.
4434 @item show commands @var{n}
4435 Print ten commands centered on command number @var{n}.
4437 @item show commands +
4438 Print ten commands just after the commands last printed.
4442 @node Screen Size,,,
4443 @section Screen Size
4444 @cindex size of screen
4445 @cindex pauses in output
4446 Certain commands to _GDBN__ may produce large amounts of information
4447 output to the screen. To help you read all of it, _GDBN__ pauses and
4448 asks you for input at the end of each page of output. Type @key{RET}
4449 when you want to continue the output. _GDBN__ also uses the screen
4450 width setting to determine when to wrap lines of output. Depending on
4451 what is being printed, it tries to break the line at a readable place,
4452 rather than simply letting it overflow onto the following line.
4454 Normally _GDBN__ knows the size of the screen from the termcap data base
4455 together with the value of the @code{TERM} environment variable and the
4456 @code{stty rows} and @code{stty cols} settings. If this is not correct,
4457 you can override it with the @samp{set height} and @samp{set
4461 @item set height @var{lpp}
4463 @itemx set width @var{cpl}
4469 These @samp{set} commands specify a screen height of @var{lpp} lines and
4470 a screen width of @var{cpl} characters. The associated @samp{show}
4471 commands display the current settings.
4473 If you specify a height of zero lines, _GDBN__ will not pause during output
4474 no matter how long the output is. This is useful if output is to a file
4475 or to an editor buffer.
4480 @cindex number representation
4481 @cindex entering numbers
4482 You can always enter numbers in octal, decimal, or hexadecimal in _GDBN__ by
4483 the usual conventions: octal numbers begin with @samp{0}, decimal
4484 numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}.
4485 Numbers that begin with none of these are, by default, entered in base
4486 10; likewise, the default display for numbers---when no particular
4487 format is specified---is base 10. You can change the default base for
4488 both input and output with the @samp{set radix} command.
4492 @item set radix @var{base}
4493 Set the default base for numeric input and display. Supported choices
4494 for @var{base} are decimal 8, 10, 16. @var{base} must itself be
4495 specified either unambiguously or using the current default radix; for
4505 will set the base to decimal. On the other hand, @samp{set radix 10}
4506 will leave the radix unchanged no matter what it was.
4510 Display the current default base for numeric input and display.
4514 @node Messages/Warnings,,,
4515 @section Optional Warnings and Messages
4516 By default, _GDBN__ is silent about its inner workings. If you are running
4517 on a slow machine, you may want to use the @samp{set verbose} command.
4518 It will make _GDBN__ tell you when it does a lengthy internal operation, so
4519 you won't think it has crashed.
4521 Currently, the messages controlled by @samp{set verbose} are those which
4522 announce that the symbol table for a source file is being read
4523 (@pxref{Files}, in the description of the command
4524 @samp{symbol-file}).
4525 @c The following is the right way to do it, but emacs 18.55 doesn't support
4526 @c @ref, and neither the emacs lisp manual version of texinfmt or makeinfo
4529 see @samp{symbol-file} in @ref{Files}).
4534 @item set verbose on
4535 Enables _GDBN__'s output of certain informational messages.
4537 @item set verbose off
4538 Disables _GDBN__'s output of certain informational messages.
4540 @kindex show verbose
4542 Displays whether @samp{set verbose} is on or off.
4545 By default, if _GDBN__ encounters bugs in the symbol table of an object file,
4546 it prints a single message about each type of problem it finds, then
4547 shuts up. You can suppress these messages, or allow more than one such
4548 message to be printed if you want to see how frequent the problems are.
4549 @xref{Symbol Errors}.
4552 @kindex set complaints
4553 @item set complaints @var{limit}
4554 Permits _GDBN__ to output @var{limit} complaints about each type of unusual
4555 symbols before becoming silent about the problem. Set @var{limit} to
4556 zero to suppress all complaints; set it to a large number to prevent
4557 complaints from being suppressed.
4559 @kindex show complaints
4560 @item show complaints
4561 Displays how many symbol complaints _GDBN__ is permitted to produce.
4564 By default, _GDBN__ is cautious, and asks what sometimes seem to be a lot of
4565 stupid questions. For example, if you try to run a program which is
4570 The program being debugged has been started already.
4571 Start it from the beginning? (y or n)
4574 If you're willing to unflinchingly face the consequences of your own
4575 commands, you can disable this ``feature'':
4580 @cindex stupid questions
4581 @item set caution off
4582 Disables cautious questions.
4584 @item set caution on
4585 Enables cautious questions (the default).
4588 @kindex show caution
4589 Displays state of cautious questions.
4593 @chapter Canned Sequences of Commands
4595 Aside from breakpoint commands (@pxref{Break Commands}),_GDBN__ provides two
4596 ways to store sequences of commands for execution as a unit:
4597 user-defined commands and command files.
4600 @section User-Defined Commands
4602 @cindex user-defined command
4603 A @dfn{user-defined command} is a sequence of _GDBN__ commands to which you
4604 assign a new name as a command. This is done with the @samp{define}
4608 @item define @var{commandname}
4610 Define a command named @var{commandname}. If there is already a command
4611 by that name, you are asked to confirm that you want to redefine it.
4613 The definition of the command is made up of other _GDBN__ command lines,
4614 which are given following the @samp{define} command. The end of these
4615 commands is marked by a line containing @samp{end}.
4617 @item document @var{commandname}
4619 Give documentation to the user-defined command @var{commandname}. The
4620 command @var{commandname} must already be defined. This command reads
4621 lines of documentation just as @samp{define} reads the lines of the
4622 command definition, ending with @samp{end}. After the @samp{document}
4623 command is finished, @samp{help} on command @var{commandname} will print
4624 the documentation you have specified.
4626 You may use the @samp{document} command again to change the
4627 documentation of a command. Redefining the command with @samp{define}
4628 does not change the documentation.
4631 User-defined commands do not take arguments. When they are executed, the
4632 commands of the definition are not printed. An error in any command
4633 stops execution of the user-defined command.
4635 Commands that would ask for confirmation if used interactively proceed
4636 without asking when used inside a user-defined command. Many _GDBN__ commands
4637 that normally print messages to say what they are doing omit the messages
4638 when used in a user-defined command.
4640 @node Command Files,,,
4641 @section Command Files
4643 @cindex command files
4644 A command file for _GDBN__ is a file of lines that are _GDBN__ commands. Comments
4645 (lines starting with @samp{#}) may also be included. An empty line in a
4646 command file does nothing; it does not mean to repeat the last command, as
4647 it would from the terminal.
4650 @cindex @file{_GDBINIT__}
4651 When you start _GDBN__, it automatically executes commands from its
4652 @dfn{init files}. These are files named @file{_GDBINIT__}. _GDBN__
4653 reads the init file (if any) in your home directory and then the init
4654 file (if any) in the current working directory. (The init files are not
4655 executed if the @samp{-nx} option is given.) You can also request the
4656 execution of a command file with the @samp{source} command:
4659 @item source @var{filename}
4661 Execute the command file @var{filename}.
4664 The lines in a command file are executed sequentially. They are not
4665 printed as they are executed. An error in any command terminates execution
4666 of the command file.
4668 Commands that would ask for confirmation if used interactively proceed
4669 without asking when used in a command file. Many _GDBN__ commands that
4670 normally print messages to say what they are doing omit the messages
4671 when called from command files.
4674 @section Commands for Controlled Output
4676 During the execution of a command file or a user-defined command, normal
4677 _GDBN__ output is suppressed; the only output that appears is what is
4678 explicitly printed by the commands in the definition. This section
4679 describes three commands useful for generating exactly the output you
4683 @item echo @var{text}
4685 @c I don't consider backslash-space a standard C escape sequence
4686 @c because it's not in ANSI.
4687 Print @var{text}. Nonprinting characters can be included in @var{text}
4688 using C escape sequences, such as @samp{\n} to print a newline. @b{No
4689 newline will be printed unless you specify one.} In addition to the
4690 standard C escape sequences a backslash followed by a space stands for a
4691 space. This is useful for outputting a string with spaces at the
4692 beginning or the end, since leading and trailing spaces are otherwise
4693 trimmed from all arguments. Thus, to print @samp{@ and foo =@ }, use the
4694 command @samp{echo \@ and foo = \@ }.
4695 @c FIXME: verify hard copy actually issues enspaces for '@ '! Will this
4698 A backslash at the end of @var{text} can be used, as in C, to continue
4699 the command onto subsequent lines. For example,
4702 echo This is some text\n\
4703 which is continued\n\
4704 onto several lines.\n
4707 produces the same output as
4710 echo This is some text\n
4711 echo which is continued\n
4712 echo onto several lines.\n
4715 @item output @var{expression}
4717 Print the value of @var{expression} and nothing but that value: no
4718 newlines, no @samp{$@var{nn} = }. The value is not entered in the
4719 value history either. @xref{Expressions} for more information on
4722 @item output/@var{fmt} @var{expression}
4723 Print the value of @var{expression} in format @var{fmt}. You can use
4724 the same formats as for @samp{print}; @pxref{Output formats}, for more
4727 @item printf @var{string}, @var{expressions}@dots{}
4729 Print the values of the @var{expressions} under the control of
4730 @var{string}. The @var{expressions} are separated by commas and may
4731 be either numbers or pointers. Their values are printed as specified
4732 by @var{string}, exactly as if the program were to execute
4735 printf (@var{string}, @var{expressions}@dots{});
4738 For example, you can print two values in hex like this:
4741 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
4744 The only backslash-escape sequences that you can use in the format
4745 string are the simple ones that consist of backslash followed by a
4750 @chapter Using _GDBN__ under GNU Emacs
4753 A special interface allows you to use GNU Emacs to view (and
4754 edit) the source files for the program you are debugging with
4757 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
4758 executable file you want to debug as an argument. This command starts
4759 _GDBN__ as a subprocess of Emacs, with input and output through a newly
4760 created Emacs buffer.
4762 Using _GDBN__ under Emacs is just like using _GDBN__ normally except for two
4767 All ``terminal'' input and output goes through the Emacs buffer.
4770 This applies both to _GDBN__ commands and their output, and to the input
4771 and output done by the program you are debugging.
4773 This is useful because it means that you can copy the text of previous
4774 commands and input them again; you can even use parts of the output
4777 All the facilities of Emacs' Shell mode are available for this purpose.
4781 _GDBN__ displays source code through Emacs.
4784 Each time _GDBN__ displays a stack frame, Emacs automatically finds the
4785 source file for that frame and puts an arrow (_0__@samp{=>}_1__) at the
4786 left margin of the current line. Emacs uses a separate buffer for
4787 source display, and splits the window to show both your _GDBN__ session
4790 Explicit _GDBN__ @samp{list} or search commands still produce output as
4791 usual, but you probably will have no reason to use them.
4794 @emph{Warning:} If the directory where your program resides is not your
4795 current directory, it can be easy to confuse Emacs about the location of
4796 the source files, in which case the auxiliary display buffer will not
4797 appear to show your source. _GDBN__ can find programs by searching your
4798 environment's @samp{PATH} variable, so the _GDBN__ input and output
4799 session will proceed normally; but Emacs doesn't get enough information
4800 back from _GDBN__ to locate the source files in this situation. To
4801 avoid this problem, either start _GDBN__ mode from the directory where
4802 your program resides, or specify a full path name when prompted for the
4803 @kbd{M-x gdb} argument.
4805 A similar confusion can result if you use the _GDBN__ @samp{file} command to
4806 switch to debugging a program in some other location, from an existing
4807 _GDBN__ buffer in Emacs.
4810 By default, @kbd{M-x gdb} calls the program called ``@code{gdb}''. If
4811 you need to call _GDBN__ by a different name (for example, if you keep
4812 several configurations around, with different names) you can set the
4813 Emacs variable @code{gdb-command-name}; for example,
4815 (setq gdb-command-name "mygdb")
4818 (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or
4819 in your @samp{.emacs} file) will make Emacs call the program named
4820 ``@code{mygdb}'' instead.
4822 In the _GDBN__ I/O buffer, you can use these special Emacs commands in
4823 addition to the standard Shell mode commands:
4827 Describe the features of Emacs' _GDBN__ Mode.
4830 Execute to another source line, like the _GDBN__ @samp{step} command; also
4831 update the display window to show the current file and location.
4834 Execute to next source line in this function, skipping all function
4835 calls, like the _GDBN__ @samp{next} command. Then update the display window
4836 to show the current file and location.
4839 Execute one instruction, like the _GDBN__ @samp{stepi} command; update
4840 display window accordingly.
4843 Execute to next instruction, using the _GDBN__ @samp{nexti} command; update
4844 display window accordingly.
4847 Execute until exit from the selected stack frame, like the _GDBN__
4848 @samp{finish} command.
4851 @c C-c C-p in emacs 19
4852 Continue execution of the program, like the _GDBN__ @samp{continue} command.
4855 @c C-c C-u in emacs 19
4856 Go up the number of frames indicated by the numeric argument
4857 (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}),
4858 like the _GDBN__ @samp{up} command.@refill
4861 @c C-c C-d in emacs 19
4862 Go down the number of frames indicated by the numeric argument, like the
4863 _GDBN__ @samp{down} command.
4866 Read the number where the cursor is positioned, and insert it at the end
4867 of the _GDBN__ I/O buffer. For example, if you wish to disassemble code
4868 around an address that was displayed earlier, type @kbd{disassemble};
4869 then move the cursor to the address display, and pick up the
4870 argument for @samp{disassemble} by typing @kbd{C-x &}.
4872 You can customize this further on the fly by defining elements of the list
4873 @samp{gdb-print-command}; once it is defined, you can format or
4874 otherwise process numbers picked up by @kbd{C-x &} before they are
4875 inserted. A numeric argument to @kbd{C-x &} will both flag that you
4876 wish special formatting, and act as an index to pick an element of the
4877 list. If the list element is a string, the number to be inserted is
4878 formatted using the Emacs function @samp{format}; otherwise the number
4879 is passed as an argument to the corresponding list element.
4881 @item M-x gdb-display-frame
4882 Explicitly request display of the source code surrounding the current
4883 frame location, in another window. _GDBN__ does this display automatically;
4884 but if, for example, you accidentally kill the buffer where it is
4885 displayed, this command is a way of getting it back.
4888 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
4889 tells _GDBN__ to set a breakpoint on the source line point is on.
4891 The source files displayed in Emacs are in ordinary Emacs buffers
4892 which are visiting the source files in the usual way. You can edit
4893 the files with these buffers if you wish; but keep in mind that _GDBN__
4894 communicates with Emacs in terms of line numbers. If you add or
4895 delete lines from the text, the line numbers that _GDBN__ knows will cease
4896 to correspond properly to the code.
4898 @c The following dropped because Epoch is nonstandard. Reactivate
4899 @c if/when v19 does something similar. ---pesch@cygnus.com 19dec1990
4901 @kindex emacs epoch environment
4905 Version 18 of Emacs has a built-in window system called the @samp{epoch}
4906 environment. Users of this environment can use a new command,
4907 @samp{inspect} which performs identically to @samp{print} except that
4908 each value is printed in its own window.
4911 @node _GDBN__ Bugs,,,
4912 @c node-name, next, previous, up
4913 @chapter Reporting Bugs in _GDBN__
4914 @cindex Bugs in _GDBN__
4915 @cindex Reporting Bugs in _GDBN__
4917 Your bug reports play an essential role in making _GDBN__ reliable.
4919 Reporting a bug may help you by bringing a solution to your problem, or it
4920 may not. But in any case the principal function of a bug report is to help
4921 the entire community by making the next version of _GDBN__ work better. Bug
4922 reports are your contribution to the maintenance of _GDBN__.
4924 In order for a bug report to serve its purpose, you must include the
4925 information that enables us to fix the bug.
4927 @node Bug Criteria,,,
4928 @section Have You Found a Bug?
4929 @cindex Bug Criteria
4931 If you are not sure whether you have found a bug, here are some guidelines:
4935 @cindex Fatal Signal
4937 If the debugger gets a fatal signal, for any input whatever, that is a
4938 _GDBN__ bug. Reliable debuggers never crash.
4941 @cindex error on Valid Input
4942 If _GDBN__ produces an error message for valid input, that is a bug.
4945 @cindex Invalid Input
4946 If _GDBN__ does not produce an error message for invalid input,
4947 that is a bug. However, you should note that your idea of
4948 ``invalid input'' might be our idea of ``an extension'' or ``support
4949 for traditional practice''.
4952 If you are an experienced user of debugging tools, your suggestions
4953 for improvement of _GDBN__ are welcome in any case.
4956 @node Bug Reporting,,,
4957 @section How to Report Bugs
4959 @cindex Compiler Bugs, Reporting
4961 A number of companies and individuals offer support for GNU products.
4962 If you obtained _GDBN__ from a support organization, we recommend you
4963 contact that organization first.
4965 Contact information for many support companies and individuals is
4966 available in the file @samp{etc/SERVICE} in the GNU Emacs distribution.
4968 In any event, we also recommend that you send bug reports for _GDBN__ to one
4972 bug-gdb@@prep.ai.mit.edu
4973 @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb
4976 @strong{Do not send bug reports to @samp{info-gdb}, or to
4977 @samp{help-gdb}, or to any newsgroups.} Most users of _GDBN__ do not want to
4978 receive bug reports. Those that do, have arranged to receive @samp{bug-gdb}.
4980 The mailing list @samp{bug-gdb} has a newsgroup which serves as a
4981 repeater. The mailing list and the newsgroup carry exactly the same
4982 messages. Often people think of posting bug reports to the newsgroup
4983 instead of mailing them. This appears to work, but it has one problem
4984 which can be crucial: a newsgroup posting does not contain a mail path
4985 back to the sender. Thus, if we need to ask for more information, we
4986 may be unable to reach you. For this reason, it is better to send bug
4987 reports to the mailing list.
4989 As a last resort, send bug reports on paper to:
4997 The fundamental principle of reporting bugs usefully is this:
4998 @strong{report all the facts}. If you are not sure whether to state a
4999 fact or leave it out, state it!
5001 Often people omit facts because they think they know what causes the
5002 problem and assume that some details don't matter. Thus, you might
5003 assume that the name of the variable you use in an example does not matter.
5004 Well, probably it doesn't, but one cannot be sure. Perhaps the bug is a
5005 stray memory reference which happens to fetch from the location where that
5006 name is stored in memory; perhaps, if the name were different, the contents
5007 of that location would fool the debugger into doing the right thing despite
5008 the bug. Play it safe and give a specific, complete example. That is the
5009 easiest thing for you to do, and the most helpful.
5011 Keep in mind that the purpose of a bug report is to enable us to fix
5012 the bug if it is new to us. It isn't as important what happens if
5013 the bug is already known. Therefore, always write your bug reports on
5014 the assumption that the bug has not been reported previously.
5016 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5017 bell?'' Those bug reports are useless, and we urge everyone to
5018 @emph{refuse to respond to them} except to chide the sender to report
5021 To enable us to fix the bug, you should include all these things:
5025 The version of _GDBN__. _GDBN__ announces it if you start with no
5026 arguments; you can also print it at any time using @samp{show version}.
5028 Without this, we won't know whether there is any point in looking for
5029 the bug in the current version of _GDBN__.
5032 A complete input script, and all necessary source files, that will
5036 What compiler (and its version) was used to compile _GDBN__---e.g.
5040 The command arguments you gave the compiler to compile your example and
5041 observe the bug. For example, did you use @samp{-O}? To guarantee
5042 you won't omit something important, list them all.
5044 If we were to try to guess the arguments, we would probably guess wrong
5045 and then we would not encounter the bug.
5048 The type of machine you are using, and the operating system name and
5052 A description of what behavior you observe that you believe is
5053 incorrect. For example, ``It gets a fatal signal.''
5055 Of course, if the bug is that _GDBN__ gets a fatal signal, then we
5056 will certainly notice it. But if the bug is incorrect output, we might
5057 not notice unless it is glaringly wrong.
5059 Even if the problem you experience is a fatal signal, you should still
5060 say so explicitly. Suppose something strange is going on, such as,
5061 your copy of _GDBN__ is out of synch, or you have encountered a
5062 bug in the C library on your system. (This has happened!) Your copy
5063 might crash and ours would not. If you told us to expect a crash,
5064 then when ours fails to crash, we would know that the bug was not
5065 happening for us. If you had not told us to expect a crash, then we
5066 would not be able to draw any conclusion from our observations.
5069 If you wish to suggest changes to the _GDBN__ source, send us context
5070 diffs. If you even discuss something in the _GDBN__ source, refer to
5071 it by context, not by line number.
5073 The line numbers in our development sources won't match those in your
5074 sources. Your line numbers would convey no useful information to us.
5078 Here are some things that are not necessary:
5082 A description of the envelope of the bug.
5084 Often people who encounter a bug spend a lot of time investigating
5085 which changes to the input file will make the bug go away and which
5086 changes will not affect it.
5088 This is often time consuming and not very useful, because the way we
5089 will find the bug is by running a single example under the debugger
5090 with breakpoints, not by pure deduction from a series of examples.
5091 We recommend that you save your time for something else.
5093 Of course, if you can find a simpler example to report @emph{instead}
5094 of the original one, that is a convenience for us. Errors in the
5095 output will be easier to spot, running under the debugger will take
5098 However, simplification is not vital; if you don't want to do this,
5099 report the bug anyway and send us the entire test case you used.
5102 A patch for the bug.
5104 A patch for the bug does help us if it is a good one. But don't omit
5105 the necessary information, such as the test case, on the assumption that
5106 a patch is all we need. We might see problems with your patch and decide
5107 to fix the problem another way, or we might not understand it at all.
5109 Sometimes with a program as complicated as _GDBN__ it is very hard to
5110 construct an example that will make the program follow a certain path
5111 through the code. If you don't send us the example, we won't be able
5112 to construct one, so we won't be able to verify that the bug is fixed.
5114 And if we can't understand what bug you are trying to fix, or why your
5115 patch should be an improvement, we won't install it. A test case will
5116 help us to understand.
5119 A guess about what the bug is or what it depends on.
5121 Such guesses are usually wrong. Even we can't guess right about such
5122 things without first using the debugger to find the facts.
5126 @include readline/inc-readline.texinfo
5127 @include readline/inc-history.texinfo
5130 @node Installing _GDBN__,,,
5131 @appendix Installing _GDBN__
5132 @cindex configuring _GDBN__
5133 @cindex installation
5135 The script @samp{config.gdb} automates the process of preparing _GDBN__ for
5136 installation; you can then use @samp{make} to actually build it. For
5145 is all that's required to install _GDBN__ on a Sun 3 running SunOS 4.
5149 @item config.gdb @var{machine}
5150 @itemx config.gdb -srcdir=@var{dir} @var{machine}
5151 This is the most usual way of configuring _GDBN__; to debug programs running
5152 on the same machine as _GDBN__ itself. If you wish to build the _GDBN__ binaries
5153 in a completely different directory from the sources, specify a path to
5154 the source directory using the @samp{-srcdir} option.
5156 @item config.gdb -host
5157 @cindex host environments
5158 Display a list of supported host environments for _GDBN__.
5160 @item config.gdb @var{host} @var{target}
5161 @itemx config.gdb -srcdir=@var{dir} @var{host} @var{target}
5162 @cindex cross-debugging
5163 _GDBN__ can also be used as a cross-debugger, running on a machine of one
5164 type while debugging a program running on a machine of another type.
5165 You configure it this way by specifying first the @var{host}, then the
5166 @var{target} environment on the @code{config.gdb} argument list; the
5167 @var{host} is where _GDBN__ runs, and the @var{target} is where your program
5168 runs. @xref{Remote}. Again, you can use @samp{-srcdir} to specify a
5169 path to the _GDBN__ source.
5171 @item config.gdb -target
5172 @cindex target environments
5173 Display a list of supported target environments for _GDBN__.
5177 @unnumbered GNU GENERAL PUBLIC LICENSE
5178 @center Version 1, February 1989
5181 Copyright @copyright{} 1989 Free Software Foundation, Inc.
5182 675 Mass Ave, Cambridge, MA 02139, USA
5184 Everyone is permitted to copy and distribute verbatim copies
5185 of this license document, but changing it is not allowed.
5188 @unnumberedsec Preamble
5190 The license agreements of most software companies try to keep users
5191 at the mercy of those companies. By contrast, our General Public
5192 License is intended to guarantee your freedom to share and change free
5193 software---to make sure the software is free for all its users. The
5194 General Public License applies to the Free Software Foundation's
5195 software and to any other program whose authors commit to using it.
5196 You can use it for your programs, too.
5198 When we speak of free software, we are referring to freedom, not
5199 price. Specifically, the General Public License is designed to make
5200 sure that you have the freedom to give away or sell copies of free
5201 software, that you receive source code or can get it if you want it,
5202 that you can change the software or use pieces of it in new free
5203 programs; and that you know you can do these things.
5205 To protect your rights, we need to make restrictions that forbid
5206 anyone to deny you these rights or to ask you to surrender the rights.
5207 These restrictions translate to certain responsibilities for you if you
5208 distribute copies of the software, or if you modify it.
5210 For example, if you distribute copies of a such a program, whether
5211 gratis or for a fee, you must give the recipients all the rights that
5212 you have. You must make sure that they, too, receive or can get the
5213 source code. And you must tell them their rights.
5215 We protect your rights with two steps: (1) copyright the software, and
5216 (2) offer you this license which gives you legal permission to copy,
5217 distribute and/or modify the software.
5219 Also, for each author's protection and ours, we want to make certain
5220 that everyone understands that there is no warranty for this free
5221 software. If the software is modified by someone else and passed on, we
5222 want its recipients to know that what they have is not the original, so
5223 that any problems introduced by others will not reflect on the original
5224 authors' reputations.
5226 The precise terms and conditions for copying, distribution and
5227 modification follow.
5230 @unnumberedsec TERMS AND CONDITIONS
5233 @center TERMS AND CONDITIONS
5238 This License Agreement applies to any program or other work which
5239 contains a notice placed by the copyright holder saying it may be
5240 distributed under the terms of this General Public License. The
5241 ``Program'', below, refers to any such program or work, and a ``work based
5242 on the Program'' means either the Program or any work containing the
5243 Program or a portion of it, either verbatim or with modifications. Each
5244 licensee is addressed as ``you''.
5247 You may copy and distribute verbatim copies of the Program's source
5248 code as you receive it, in any medium, provided that you conspicuously and
5249 appropriately publish on each copy an appropriate copyright notice and
5250 disclaimer of warranty; keep intact all the notices that refer to this
5251 General Public License and to the absence of any warranty; and give any
5252 other recipients of the Program a copy of this General Public License
5253 along with the Program. You may charge a fee for the physical act of
5254 transferring a copy.
5257 You may modify your copy or copies of the Program or any portion of
5258 it, and copy and distribute such modifications under the terms of Paragraph
5259 1 above, provided that you also do the following:
5263 cause the modified files to carry prominent notices stating that
5264 you changed the files and the date of any change; and
5267 cause the whole of any work that you distribute or publish, that
5268 in whole or in part contains the Program or any part thereof, either
5269 with or without modifications, to be licensed at no charge to all
5270 third parties under the terms of this General Public License (except
5271 that you may choose to grant warranty protection to some or all
5272 third parties, at your option).
5275 If the modified program normally reads commands interactively when
5276 run, you must cause it, when started running for such interactive use
5277 in the simplest and most usual way, to print or display an
5278 announcement including an appropriate copyright notice and a notice
5279 that there is no warranty (or else, saying that you provide a
5280 warranty) and that users may redistribute the program under these
5281 conditions, and telling the user how to view a copy of this General
5285 You may charge a fee for the physical act of transferring a
5286 copy, and you may at your option offer warranty protection in
5290 Mere aggregation of another independent work with the Program (or its
5291 derivative) on a volume of a storage or distribution medium does not bring
5292 the other work under the scope of these terms.
5295 You may copy and distribute the Program (or a portion or derivative of
5296 it, under Paragraph 2) in object code or executable form under the terms of
5297 Paragraphs 1 and 2 above provided that you also do one of the following:
5301 accompany it with the complete corresponding machine-readable
5302 source code, which must be distributed under the terms of
5303 Paragraphs 1 and 2 above; or,
5306 accompany it with a written offer, valid for at least three
5307 years, to give any third party free (except for a nominal charge
5308 for the cost of distribution) a complete machine-readable copy of the
5309 corresponding source code, to be distributed under the terms of
5310 Paragraphs 1 and 2 above; or,
5313 accompany it with the information you received as to where the
5314 corresponding source code may be obtained. (This alternative is
5315 allowed only for noncommercial distribution and only if you
5316 received the program in object code or executable form alone.)
5319 Source code for a work means the preferred form of the work for making
5320 modifications to it. For an executable file, complete source code means
5321 all the source code for all modules it contains; but, as a special
5322 exception, it need not include source code for modules which are standard
5323 libraries that accompany the operating system on which the executable
5324 file runs, or for standard header files or definitions files that
5325 accompany that operating system.
5328 You may not copy, modify, sublicense, distribute or transfer the
5329 Program except as expressly provided under this General Public License.
5330 Any attempt otherwise to copy, modify, sublicense, distribute or transfer
5331 the Program is void, and will automatically terminate your rights to use
5332 the Program under this License. However, parties who have received
5333 copies, or rights to use copies, from you under this General Public
5334 License will not have their licenses terminated so long as such parties
5335 remain in full compliance.
5338 By copying, distributing or modifying the Program (or any work based
5339 on the Program) you indicate your acceptance of this license to do so,
5340 and all its terms and conditions.
5343 Each time you redistribute the Program (or any work based on the
5344 Program), the recipient automatically receives a license from the original
5345 licensor to copy, distribute or modify the Program subject to these
5346 terms and conditions. You may not impose any further restrictions on the
5347 recipients' exercise of the rights granted herein.
5350 The Free Software Foundation may publish revised and/or new versions
5351 of the General Public License from time to time. Such new versions will
5352 be similar in spirit to the present version, but may differ in detail to
5353 address new problems or concerns.
5355 Each version is given a distinguishing version number. If the Program
5356 specifies a version number of the license which applies to it and ``any
5357 later version'', you have the option of following the terms and conditions
5358 either of that version or of any later version published by the Free
5359 Software Foundation. If the Program does not specify a version number of
5360 the license, you may choose any version ever published by the Free Software
5364 If you wish to incorporate parts of the Program into other free
5365 programs whose distribution conditions are different, write to the author
5366 to ask for permission. For software which is copyrighted by the Free
5367 Software Foundation, write to the Free Software Foundation; we sometimes
5368 make exceptions for this. Our decision will be guided by the two goals
5369 of preserving the free status of all derivatives of our free software and
5370 of promoting the sharing and reuse of software generally.
5373 @heading NO WARRANTY
5380 BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
5381 FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
5382 OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
5383 PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
5384 OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
5385 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
5386 TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
5387 PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
5388 REPAIR OR CORRECTION.
5391 IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL
5392 ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
5393 REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
5394 INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES
5395 ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT
5396 LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES
5397 SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE
5398 WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN
5399 ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
5403 @heading END OF TERMS AND CONDITIONS
5406 @center END OF TERMS AND CONDITIONS
5410 @unnumberedsec Applying These Terms to Your New Programs
5412 If you develop a new program, and you want it to be of the greatest
5413 possible use to humanity, the best way to achieve this is to make it
5414 free software which everyone can redistribute and change under these
5417 To do so, attach the following notices to the program. It is safest to
5418 attach them to the start of each source file to most effectively convey
5419 the exclusion of warranty; and each file should have at least the
5420 ``copyright'' line and a pointer to where the full notice is found.
5423 @var{one line to give the program's name and a brief idea of what it does.}
5424 Copyright (C) 19@var{yy} @var{name of author}
5426 This program is free software; you can redistribute it and/or modify
5427 it under the terms of the GNU General Public License as published by
5428 the Free Software Foundation; either version 1, or (at your option)
5431 This program is distributed in the hope that it will be useful,
5432 but WITHOUT ANY WARRANTY; without even the implied warranty of
5433 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
5434 GNU General Public License for more details.
5436 You should have received a copy of the GNU General Public License
5437 along with this program; if not, write to the Free Software
5438 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
5441 Also add information on how to contact you by electronic and paper mail.
5443 If the program is interactive, make it output a short notice like this
5444 when it starts in an interactive mode:
5447 Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
5448 Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
5449 This is free software, and you are welcome to redistribute it
5450 under certain conditions; type `show c' for details.
5453 The hypothetical commands `show w' and `show c' should show the
5454 appropriate parts of the General Public License. Of course, the
5455 commands you use may be called something other than `show w' and `show
5456 c'; they could even be mouse-clicks or menu items---whatever suits your
5459 You should also get your employer (if you work as a programmer) or your
5460 school, if any, to sign a ``copyright disclaimer'' for the program, if
5461 necessary. Here is a sample; alter the names:
5464 Yoyodyne, Inc., hereby disclaims all copyright interest in the
5465 program `Gnomovision' (a program to direct compilers to make passes
5466 at assemblers) written by James Hacker.
5468 @var{signature of Ty Coon}, 1 April 1989
5469 Ty Coon, President of Vice
5472 That's all there is to it!
5480 % I think something like @colophon should be in texinfo. In the
5482 \long\def\colophon{\hbox to0pt{}\vfill
5483 \centerline{The body of this manual is set in}
5484 \centerline{\fontname\tenrm,}
5485 \centerline{with headings in {\bf\fontname\tenbf}}
5486 \centerline{and examples in {\tt\fontname\tentt}.}
5487 \centerline{{\it\fontname\tenit\/} and}
5488 \centerline{{\sl\fontname\tensl\/}}
5489 \centerline{are used for emphasis.}\vfill}
5491 % Blame: pesch@cygnus.com, 28mar91.