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 @kindex show version
1089 @c FIXME: chgd to "show version" from "info". Verify John doing same to GDBv4.
1090 Show what version of _GDBN__ is running. You should include this
1091 information in _GDBN__ bug-reports. If multiple versions of _GDBN__ are
1092 in use at your site, you may occasionally want to make sure what version
1093 of _GDBN__ you're running; as _GDBN__ evolves, new commands are
1094 introduced, and old ones may wither away. The version number is also
1095 announced when you start _GDBN__ with no arguments.
1097 This @samp{show} subcommand is one of those with no corresponding
1098 @samp{set} subcommand.
1102 @chapter Running Programs Under _GDBN__
1104 @node Compilation,,,
1105 @section Compiling for Debugging
1107 In order to debug a program most effectively, you need to generate
1108 debugging information when you compile it. This debugging information
1109 is stored in the object file; it describes the data type of each
1110 variable or function and the correspondence between source line numbers
1111 and addresses in the executable code.
1113 To request debugging information, specify the @samp{-g} option when you run
1116 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1117 options together. Using those compilers, you cannot generate optimized
1118 executables containing debugging information.
1120 The GNU C compiler supports @samp{-g} with or without @samp{-O}, making it
1121 possible to debug optimized code. We recommend that you @emph{always} use
1122 @samp{-g} whenever you compile a program. You may think the program is
1123 correct, but there's no sense in pushing your luck.
1125 Some things do not work as well with @samp{-g -O} as with just
1126 @samp{-g}, particularly on machines with instruction scheduling. If in
1127 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1128 please report it as a bug (including a test case!).
1130 Older versions of the GNU C compiler permitted a variant option
1131 @samp{-gg} for debugging information. _GDBN__ no longer supports this
1132 format; if your GNU C compiler has this option, do not use it.
1135 @comment As far as I know, there are no cases in which _GDBN__ will
1136 @comment produce strange output in this case. (but no promises).
1137 If your program includes archives made with the @code{ar} program, and
1138 if the object files used as input to @code{ar} were compiled without the
1139 @samp{-g} option and have names longer than 15 characters, _GDBN__ will get
1140 confused reading the program's symbol table. No error message will be
1141 given, but _GDBN__ may behave strangely. The reason for this problem is a
1142 deficiency in the Unix archive file format, which cannot represent file
1143 names longer than 15 characters.
1145 To avoid this problem, compile the archive members with the @samp{-g}
1146 option or use shorter file names. Alternatively, use a version of GNU
1147 @code{ar} dated more recently than August 1989.
1152 @section Starting your Program
1156 To start your program under _GDBN__, use the @samp{run} command.
1158 Except on VxWorks, you
1163 must first specify the program name with an argument to _GDBN__
1164 (@pxref{Invocation}), or using the @samp{file} or @samp{exec-file}
1165 command (@pxref{Files}).@refill
1167 On targets that support processes, @samp{run} creates an inferior
1168 process and makes that process run your program. On other targets,
1169 @samp{run} jumps to the start of the program.
1171 The execution of a program is affected by certain information it
1172 receives from its superior. _GDBN__ provides ways to specify this
1173 information, which you must do @i{before} starting the program. (You
1174 can change it after starting the program, but such changes will only affect
1175 the program the next time you start it.) This information may be
1176 divided into four categories:
1179 @item The @i{arguments.}
1180 You specify the arguments to give your program as the arguments of the
1181 @samp{run} command. If a shell is available on your target, the shell
1182 is used to pass the arguments, so that you may use normal conventions
1183 (for example regular expression expansion or variable substitution) in
1184 describing the arguments. In Unix systems, you can control which shell
1185 is used with the @code{SHELL} environment variable.
1187 @item The @i{environment.}
1188 Your program normally inherits its environment from _GDBN__, but you can
1189 use the _GDBN__ commands @samp{set environment} and @samp{unset
1190 environment} to change parts of the environment that will be given to
1193 @item The @i{working directory.}
1194 Your program inherits its working directory from _GDBN__. You can set
1195 _GDBN__'s working directory with the @samp{cd} command in _GDBN__.
1197 @item The @i{standard input and output.}
1198 Your program normally uses the same device for standard input and
1199 standard output as _GDBN__ is using. You can redirect input and output
1200 in the @code{run} command line, or you can use the @samp{tty} command to
1201 set a different device for your program.
1204 When you issue the @samp{run} command, your program begins to execute
1205 immediately. @xref{Stopping}, for discussion of how to arrange for your
1208 Note that once your program has been started by the @samp{run} command,
1209 you may evaluate expressions that involve calls to functions in the
1210 inferior, using the @samp{print} or @samp{call} commands. @xref{Data}.
1212 If the modification time of your symbol file has changed since the last
1213 time _GDBN__ read its symbols, _GDBN__ will discard its symbol table and re-read
1214 it. In this process, it tries to retain your current breakpoints.
1217 @section Your Program's Arguments
1219 @cindex arguments (to your program)
1220 The arguments to your program can be specified by the arguments of the
1221 @samp{run} command. They are passed to a shell, which expands wildcard
1222 characters and performs redirection of I/O, and thence to the program.
1223 _GDBN__ uses the shell indicated by your environment variable
1224 @code{SHELL} if it exists; otherwise, _GDBN__ uses @samp{/bin/sh}.
1226 @samp{run} with no arguments uses the same arguments used by the previous
1227 @samp{run}, or those set by the @samp{set args} command.
1232 Specify the arguments to be used the next time your program is run. If
1233 @samp{set args} has no arguments, @samp{run} will execute your program
1234 with no arguments. Once you have run your program with arguments, this
1235 is the only way to run it again without arguments.
1239 Show the arguments to give your program when it is started.
1242 @node Environment,,,
1243 @section Your Program's Environment
1245 @cindex environment (of your program)
1246 The @dfn{environment} consists of a set of environment variables and
1247 their values. Environment variables conventionally record such things as
1248 your user name, your home directory, your terminal type, and your search
1249 path for programs to run. Usually you set up environment variables with
1250 the shell and they are inherited by all the other programs you run. When
1251 debugging, it can be useful to try running the program with a modified
1252 environment without having to start _GDBN__ over again.
1255 @item path @var{directory}
1257 Add @var{directory} to the front of the @code{PATH} environment variable
1258 (the search path for executables), for both _GDBN__ and your program.
1259 You may specify several directory names, separated by @samp{:} or
1260 whitespace. If a directory is already in the path, it is moved to the
1261 front, so it will be searched sooner. You can use the string
1262 @samp{$cwd} to refer to the current working directory.
1264 This command will not repeat if you press @key{RET} a second time after
1267 @item show environment @var{varname}
1268 @kindex show environment
1269 Print the value of environment variable @var{varname} to be given to
1270 your program when it starts.
1272 @item show environment
1273 Print the names and values of all environment variables to be given to
1276 @item set environment @var{varname} @var{value}
1277 @itemx set environment @var{varname} = @var{value}
1278 @kindex set environment
1279 Sets environment variable @var{varname} to @var{value}. The value
1280 changes for your program only, not for _GDBN__ itself. @var{value} may
1281 be any string; the values of environment variables are just strings, and
1282 any interpretation is supplied by your program itself. The @var{value}
1283 parameter is optional; if it is eliminated, the variable is set to a
1285 @c FIXME: I think "any string" here doesn't include leading, trailing
1286 @c FIXME... blanks. Queried J Gilmore. ---pesch@cygnus.com, 4apr91
1288 For example, this command:
1295 tells a Unix program, when subsequently run, that its user is named
1298 @item unset environment @var{varname}
1299 @kindex unset environment
1300 Remove variable @var{varname} from the environment to be passed to your
1301 program. This is different from @samp{set env @var{varname}=};
1302 @samp{unset environment} removes the variable from the environment,
1303 rather than assigning it an empty value.
1306 @node Working Directory,,,
1307 @section Your Program's Working Directory
1309 @cindex working directory (of your program)
1310 Each time you start your program with @samp{run}, it inherits its
1311 working directory from the current working directory of _GDBN__. _GDBN__'s
1312 working directory is initially whatever it inherited from its parent
1313 process (typically the shell), but you can specify a new working
1314 directory in _GDBN__ with the @samp{cd} command.
1316 The _GDBN__ working directory also serves as a default for the commands
1317 that specify files for _GDBN__ to operate on. @xref{Files}.
1320 @item cd @var{directory}
1322 Set _GDBN__'s working directory to @var{directory}.
1326 Print _GDBN__'s working directory.
1329 @node Input/Output,,,
1330 @section Your Program's Input and Output
1335 @cindex controlling terminal
1336 By default, the program you run under _GDBN__ does input and output to the same
1337 terminal that _GDBN__ uses.
1339 You can redirect the program's input and/or output using shell
1340 redirection with the @samp{run} command. For example,
1347 starts the program, diverting its output to the file @file{outfile}.
1350 Another way to specify where the program should do input and output is
1351 with the @samp{tty} command. This command accepts a file name as
1352 argument, and causes this file to be the default for future @samp{run}
1353 commands. It also resets the controlling terminal for the child
1354 process, for future @samp{run} commands. For example,
1361 directs that processes started with subsequent @samp{run} commands
1362 default to do input and output on the terminal @file{/dev/ttyb} and have
1363 that as their controlling terminal.
1365 An explicit redirection in @samp{run} overrides the @samp{tty} command's
1366 effect on input/output redirection, but not its effect on the
1367 controlling terminal.
1369 When you use the @samp{tty} command or redirect input in the @samp{run}
1370 command, only the input @emph{for your program} is affected. The input
1371 for _GDBN__ still comes from your terminal.
1374 @section Debugging an Already-Running Process
1379 @item attach @var{process-id}
1380 If your currently selected target supports processes, this command
1381 attaches to a running process---one that was started outside _GDBN__.
1382 (@samp{info files} will show your active targets). The command takes as
1383 argument a process ID. The usual way to find out the process-id of
1384 a Unix process is with the @code{ps} utility, or with the @code{jobs -l}
1387 @samp{attach} will not repeat if you press @key{RET} a second time after
1388 executing the command.
1391 To use @samp{attach}, you must have permission to send the process a
1392 signal, and it must have the same effective user ID as the _GDBN__
1395 When using @samp{attach}, you should first use the @samp{file} command
1396 to specify the program running in the process and load its symbol table.
1398 The first thing _GDBN__ does after arranging to debug the specified
1399 process is to stop it. You can examine and modify an attached process
1400 with all the _GDBN__ commands that ordinarily available when you start
1401 processes with @samp{run}. You can insert breakpoints; you can step and
1402 continue; you can modify storage. If you would rather the process
1403 continue running, you may use the @samp{continue} command after
1404 attaching _GDBN__ to the process.
1407 When you have finished debugging the attached process, you can use the
1408 @samp{detach} command to release it from _GDBN__'s control. Detaching
1409 the process continues its execution. After the @samp{detach} command,
1410 that process and _GDBN__ become completely independent once more, and you
1411 are ready to @samp{attach} another process or start one with @samp{run}.
1412 @samp{detach} will not repeat if you press @key{RET} again after using
1415 If you exit _GDBN__ or use the @samp{run} command while you have an attached
1416 process, you kill that process. By default, you will be asked for
1417 confirmation if you try to do either of these things; you can control
1418 whether or not this happens by using the @samp{set caution} command
1419 (@pxref{Messages/Warnings}).
1422 @node Kill Process,,,
1423 @section Killing the Child Process
1428 Kill the child process in which your program is running under _GDBN__.
1431 This command is useful if you wish to debug a core dump instead of a
1432 running process. _GDBN__ ignores any core dump file while your program
1436 On some operating systems, you can't execute your program in another
1437 process while breakpoints are active inside _GDBN__. You can use the
1438 @samp{kill} command in this situation to permit running the program
1439 outside the debugger.
1441 The @samp{kill} command is also useful if you wish to recompile and
1442 relink the program, since on many systems it is impossible to modify an
1443 executable file which is running in a process. In this case, when you
1444 next type @samp{run}, _GDBN__ will notice that the file has changed, and
1445 will re-read the symbol table (while trying to preserve your current
1446 breakpoint settings).
1449 @chapter Stopping and Continuing
1451 When you run a program normally, it runs until it terminates. The
1452 principal purpose of using a debugger is so that you can stop it before
1453 that point; or so that if the program runs into trouble you can
1454 investigate and find out why.
1456 @node Breakpoints,,,
1457 @section Breakpoints
1460 A @dfn{breakpoint} makes your program stop whenever a certain point in the
1461 program is reached. You set breakpoints explicitly with _GDBN__ commands,
1462 specifying the place where the program should stop by line number, function
1463 name or exact address in the program. You can add various other conditions
1464 to control whether the program will stop.
1466 Each breakpoint is assigned a number when it is created; these numbers are
1467 successive integers starting with 1. In many of the commands for controlling
1468 various features of breakpoints you use the breakpoint number to say which
1469 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
1470 @dfn{disabled}; if disabled, it has no effect on the program until you
1474 @subsection Setting Breakpoints
1478 Breakpoints are set with the @samp{break} command (abbreviated @samp{b}).
1480 You have several ways to say where the breakpoint should go.
1483 @item break @var{function}
1484 Set a breakpoint at entry to function @var{function}.
1486 @item break @var{+offset}
1487 @itemx break @var{-offset}
1488 Set a breakpoint some number of lines forward or back from the position
1489 at which execution stopped in the currently selected frame.
1491 @item break @var{linenum}
1492 Set a breakpoint at line @var{linenum} in the current source file.
1493 That file is the last file whose source text was printed. This
1494 breakpoint will stop the program just before it executes any of the
1497 @item break @var{filename}:@var{linenum}
1498 Set a breakpoint at line @var{linenum} in source file @var{filename}.
1500 @item break @var{filename}:@var{function}
1501 Set a breakpoint at entry to function @var{function} found in file
1502 @var{filename}. Specifying a file name as well as a function name is
1503 superfluous except when multiple files contain similarly named
1506 @item break *@var{address}
1507 Set a breakpoint at address @var{address}. You can use this to set
1508 breakpoints in parts of the program which do not have debugging
1509 information or source files.
1512 Set a breakpoint at the next instruction to be executed in the selected
1513 stack frame (@pxref{Stack}). In any selected frame but the innermost,
1514 this will cause the program to stop as soon as control returns to that
1515 frame. This is equivalent to a @samp{finish} command in the frame
1516 inside the selected frame. If this is done in the innermost frame, _GDBN__
1517 will stop the next time it reaches the current location; this may be
1518 useful inside of loops.
1520 _GDBN__ normally ignores breakpoints when it resumes execution, until at
1521 least one instruction has been executed. If it did not do this, you
1522 would be unable to proceed past a breakpoint without first disabling the
1523 breakpoint. This rule applies whether or not the breakpoint already
1524 existed when the program stopped.
1526 @item break @dots{} if @var{cond}
1527 Set a breakpoint with condition @var{cond}; evaluate the expression
1528 @var{cond} each time the breakpoint is reached, and stop only if the
1529 value is nonzero. @samp{@dots{}} stands for one of the possible
1530 arguments described above (or no argument) specifying where to break.
1531 @xref{Conditions}, for more information on breakpoint conditions.
1533 @item tbreak @var{args}
1535 Set a breakpoint enabled only for one stop. @var{args} are the
1536 same as in the @samp{break} command, and the breakpoint is set in the same
1537 way, but the breakpoint is automatically disabled the first time it
1538 is hit. @xref{Disabling}.
1540 @item rbreak @var{regex}
1542 Set a breakpoint on all functions matching @var{regex}. This is
1543 useful for setting breakpoints on overloaded functions that are not
1544 members of any special classes. This command sets an unconditional
1545 breakpoint on all matches, printing a list of all breakpoints it set.
1546 Once these breakpoints are set, they are treated just like the
1547 breakpoints set with the @samp{break} command. They can be deleted,
1548 disabled, made conditional, etc., in the standard ways.
1550 @kindex info breakpoints
1552 @item info breakpoints
1553 The command @samp{info breakpoints} prints a list of all breakpoints set
1554 and not deleted, showing their numbers, where in the program they are,
1555 and any special features in use for them. Disabled breakpoints are
1556 included in the list, but marked as disabled. @samp{info break} with a
1557 breakpoint number as argument lists only that breakpoint. The
1558 convenience variable @code{$_} and the default examining-address for the
1559 @samp{x} command are set to the address of the last breakpoint listed
1563 _GDBN__ allows you to set any number of breakpoints at the same place in the
1564 program. There is nothing silly or meaningless about this. When the
1565 breakpoints are conditional, this is even useful (@pxref{Conditions}).
1567 @node Set Watchpoints,,,
1568 @subsection Setting Watchpoints
1570 A @dfn{watchpoint} is a special breakpoint that stops your program when
1571 the value of an expression changes. You can use a watchpoint to stop
1572 execution whenever the value of an expression changes, without having to
1573 predict a particular place in the inferior process where this may
1574 happen. Aside from the different syntax in setting a watchpoint, it is
1575 managed exactly like any other breakpoint and is enabled, disabled, and
1576 deleted using exactly the same commands.
1578 Watchpoints currently execute two orders of magnitude more slowly than
1579 other breakpoints, but this can well be worth it to catch errors where
1580 you have no clue what part of your program is the culprit. Some
1581 processors provide special hardware to implement this feature; future
1582 releases of _GDBN__ will use such hardware if it is available.
1586 @item watch @var{expr}
1587 Set a watchpoint for an expression.
1591 This command prints a list of watchpoints.
1594 @node Exception Handling,,,
1595 @subsection Breakpoints and Exceptions
1596 @cindex exception handlers
1598 Some languages, such as GNU C++, implement exception handling. _GDBN__
1599 can be used to examine what caused the program to raise an exception
1600 and to list the exceptions the program is prepared to handle at a
1601 given point in time.
1604 @item catch @var{exceptions}
1607 Breakpoints can be set at active exception handlers by using the
1608 @samp{catch} command. @var{exceptions} is a list of names of exceptions
1612 You can use @samp{info catch} to list active exception handlers;
1615 There are currently some limitations to exception handling in _GDBN__.
1616 These will be corrected in a future release.
1620 If you call a function interactively, _GDBN__ will normally return
1621 control to you when the function has finished executing. If the call
1622 raises an exception, however, the call may bypass the mechanism that
1623 returns control to the user and cause the program to simply continue
1624 running until it hits a breakpoint, catches a signal that _GDBN__ is
1625 listening for, or exits.
1627 You cannot raise an exception interactively.
1629 You cannot interactively install an exception handler.
1632 @cindex raise exceptions
1633 Sometimes @samp{catch} is not the best way to debug exception handling:
1634 if you need to know exactly where an exception is raised, it's better to
1635 stop @emph{before} the exception handler is called, since that way you
1636 can see the stack before any unwinding takes place.
1638 To stop just before an exception handler is called, you need some
1639 knowledge of the implementation. In the case of GNU C++ exception are
1640 raised by calling a library function named @code{__raise_exception}
1641 which has the following ANSI C interface:
1644 /* ADDR is where the exception identifier is stored.
1645 ID is the exception identifier. */
1646 void __raise_exception (void **addr, void *id);
1650 To make the debugger catch all exceptions before any stack
1651 unwinding takes place, set a breakpoint on @code{__raise_exception}
1652 (@pxref{Breakpoints}). If you set a breakpoint in an exception handler
1653 instead, it may not be easy to find out where the exception was raised.
1655 With a conditional breakpoint (@xref{Conditions}) that depends on the
1656 value of @code{id}, you can cause the debugger to stop only when a
1657 specific exception is raised. Multiple conditional breakpoints can be
1658 used to stop the program when any of a number of exceptions are raised.
1660 @node Delete Breaks,,,
1661 @subsection Deleting Breakpoints
1663 @cindex clearing breakpoints, watchpoints
1664 @cindex deleting breakpoints, watchpoints
1665 It is often necessary to eliminate a breakpoint once it has done its job
1666 and you no longer want the program to stop there. This is called
1667 @dfn{deleting} the breakpoint. A breakpoint that has been deleted no
1668 longer exists in any sense; it is forgotten.
1670 With the @samp{clear} command you can delete breakpoints according to where
1671 they are in the program. With the @samp{delete} command you can delete
1672 individual breakpoints by specifying their breakpoint numbers.
1674 It is not necessary to delete a breakpoint to proceed past it. _GDBN__
1675 automatically ignores breakpoints on the first instruction to be executed
1676 when you continue execution without changing the execution address.
1681 Delete any breakpoints at the next instruction to be executed in the
1682 selected stack frame (@pxref{Selection}). When the innermost frame
1683 is selected, this is a good way to delete a breakpoint that the program
1686 @item clear @var{function}
1687 @itemx clear @var{filename}:@var{function}
1688 Delete any breakpoints set at entry to the function @var{function}.
1690 @item clear @var{linenum}
1691 @itemx clear @var{filename}:@var{linenum}
1692 Delete any breakpoints set at or within the code of the specified line.
1694 @item delete breakpoints @var{bnums}@dots{}
1695 @itemx delete @var{bnums}@dots{}
1697 @kindex delete breakpoints
1699 Delete the breakpoints of the numbers specified as arguments. If no
1700 argument is specified, delete all breakpoints.
1704 @subsection Disabling Breakpoints
1706 @cindex disabled breakpoints
1707 @cindex enabled breakpoints
1708 Rather than deleting a breakpoint, you might prefer to @dfn{disable} it.
1709 This makes the breakpoint inoperative as if it had been deleted, but
1710 remembers the information on the breakpoint so that you can @dfn{enable}
1713 You disable and enable breakpoints with the @samp{enable} and
1714 @samp{disable} commands, optionally specifying one or more breakpoint
1715 numbers as arguments. Use @samp{info break} to print a list of
1716 breakpoints if you don't know which breakpoint numbers to use.
1718 A breakpoint can have any of four different states of enablement:
1722 Enabled. The breakpoint will stop the program. A breakpoint made
1723 with the @samp{break} command starts out in this state.
1725 Disabled. The breakpoint has no effect on the program.
1727 Enabled once. The breakpoint will stop the program, but
1728 when it does so it will become disabled. A breakpoint made
1729 with the @samp{tbreak} command starts out in this state.
1731 Enabled for deletion. The breakpoint will stop the program, but
1732 immediately after it does so it will be deleted permanently.
1735 You can use the following commands to enable or disable a breakpoint:
1738 @item disable breakpoints @var{bnums}@dots{}
1739 @itemx disable @var{bnums}@dots{}
1741 @kindex disable breakpoints
1743 Disable the specified breakpoints---or all breakpoints, if none are
1744 listed. A disabled breakpoint has no effect but is not forgotten. All
1745 options such as ignore-counts, conditions and commands are remembered in
1746 case the breakpoint is enabled again later.
1748 @item enable breakpoints @var{bnums}@dots{}
1749 @itemx enable @var{bnums}@dots{}
1751 @kindex enable breakpoints
1753 Enable the specified breakpoints (or all defined breakpoints). They
1754 become effective once again in stopping the program, until you specify
1757 @item enable breakpoints once @var{bnums}@dots{}
1758 @itemx enable once @var{bnums}@dots{}
1759 Enable the specified breakpoints temporarily. Each will be disabled
1760 again the next time it stops the program (unless you have used one of
1761 these commands to specify a different state before that time comes).
1763 @item enable breakpoints delete @var{bnums}@dots{}
1764 @itemx enable delete @var{bnums}@dots{}
1765 Enable the specified breakpoints to work once and then die. Each of
1766 the breakpoints will be deleted the next time it stops the program
1767 (unless you have used one of these commands to specify a different
1768 state before that time comes).
1771 Save for a breakpoint set with @samp{tbreak} (@pxref{Set Breaks}),
1772 breakpoints that you set initially enabled; subsequently, they become
1773 disabled or enabled only when you use one of the commands above. (The
1774 command @samp{until} can set and delete a breakpoint of its own, but it
1775 will not change the state of your other breakpoints).
1778 @subsection Break Conditions
1779 @cindex conditional breakpoints
1780 @cindex breakpoint conditions
1782 The simplest sort of breakpoint breaks every time the program reaches a
1783 specified place. You can also specify a @dfn{condition} for a
1784 breakpoint. A condition is just a Boolean expression in your
1785 programming language. (@xref{Expressions}). A breakpoint with a
1786 condition evaluates the expression each time the program reaches it, and
1787 the program stops only if the condition is true.
1789 Break conditions may have side effects, and may even call functions in your
1790 program. These may sound like strange things to do, but their effects are
1791 completely predictable unless there is another enabled breakpoint at the
1792 same address. (In that case, _GDBN__ might see the other breakpoint first and
1793 stop the program without checking the condition of this one.) Note that
1794 breakpoint commands are usually more convenient and flexible for the
1795 purpose of performing side effects when a breakpoint is reached
1796 (@pxref{Break Commands}).
1798 Break conditions can be specified when a breakpoint is set, by using
1799 @samp{if} in the arguments to the @samp{break} command. @xref{Set Breaks}.
1800 They can also be changed at any time with the @samp{condition} command:
1803 @item condition @var{bnum} @var{expression}
1805 Specify @var{expression} as the break condition for breakpoint number
1806 @var{bnum}. From now on, this breakpoint will stop the program only if
1807 the value of @var{expression} is true (nonzero, in C). @var{expression}
1808 is not evaluated at the time the @samp{condition} command is given.
1809 When you call @samp{condition}, the expression you specify is checked
1810 immediately for syntactic correctness, and to determine whether symbols
1811 in it have referents in the context of your breakpoint.
1814 @item condition @var{bnum}
1815 Remove the condition from breakpoint number @var{bnum}. It becomes
1816 an ordinary unconditional breakpoint.
1819 @cindex ignore count (of breakpoint)
1820 A special case of a breakpoint condition is to stop only when the
1821 breakpoint has been reached a certain number of times. This is so
1822 useful that there is a special way to do it, using the @dfn{ignore
1823 count} of the breakpoint. Every breakpoint has an ignore count, which
1824 is an integer. Most of the time, the ignore count is zero, and
1825 therefore has no effect. But if the program reaches a breakpoint whose
1826 ignore count is positive, then instead of stopping, it just decrements
1827 the ignore count by one and continues. As a result, if the ignore count
1828 value is @var{n}, the breakpoint will not stop the next @var{n} times it
1832 @item ignore @var{bnum} @var{count}
1834 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
1835 The next @var{count} times the breakpoint is reached, your program's
1836 execution will not stop; other than to decrement the ignore count, _GDBN__
1839 To make the breakpoint stop the next time it is reached, specify
1842 @item continue @var{count}
1843 @itemx c @var{count}
1844 @itemx fg @var{count}
1845 @kindex continue @var{count}
1846 Continue execution of the program, setting the ignore count of the
1847 breakpoint that the program stopped at to @var{count} minus one.
1848 Thus, the program will not stop at this breakpoint until the
1849 @var{count}'th time it is reached.
1851 An argument to this command is meaningful only when the program stopped
1852 due to a breakpoint. At other times, the argument to @samp{continue} is
1855 The synonym @samp{fg} is provided purely for convenience, and has
1856 exactly the same behavior as other forms of the command.
1859 If a breakpoint has a positive ignore count and a condition, the condition
1860 is not checked. Once the ignore count reaches zero, the condition will
1863 You could achieve the effect of the ignore count with a
1864 condition such as _0__@w{@samp{$foo-- <= 0}}_1__ using a debugger convenience
1865 variable that is decremented each time. @xref{Convenience Vars}.
1867 @node Break Commands,,,
1868 @subsection Breakpoint Command Lists
1870 @cindex breakpoint commands
1871 You can give any breakpoint a series of commands to execute when the
1872 program stops due to that breakpoint. For example, you might want to
1873 print the values of certain expressions, or enable other breakpoints.
1876 @item commands @var{bnum}
1877 @itemx @dots @var{command-list} @dots
1881 Specify a list of commands for breakpoint number @var{bnum}. The commands
1882 themselves appear on the following lines. Type a line containing just
1883 @samp{end} to terminate the commands.
1885 To remove all commands from a breakpoint, use the command
1886 @samp{commands} and follow it immediately by @samp{end}; that is, give
1889 With no arguments, @samp{commands} refers to the last breakpoint set
1890 (not to the breakpoint most recently encountered).
1893 Pressing @key{RET} as a means of repeating the last _GDBN__ command is
1894 disabled from the time you enter @samp{commands} to just after the
1895 corresponding @samp{end}.
1897 You can use breakpoint commands to start the program up again. Simply
1898 use the @samp{continue} command, or @samp{step}, or any other command to
1899 resume execution. However, if you do this, any further commands in the
1900 same breakpoint's command list are ignored. When the program stops
1901 again, _GDBN__ will act according to the cause of that stop.
1904 If the first command specified is @samp{silent}, the usual message about
1905 stopping at a breakpoint is not printed. This may be desirable for
1906 breakpoints that are to print a specific message and then continue.
1907 If the remaining commands too print nothing, you will see no sign that
1908 the breakpoint was reached at all. @samp{silent} is not really a command;
1909 it is meaningful only at the beginning of the commands for a breakpoint.
1911 The commands @samp{echo} and @samp{output} that allow you to print precisely
1912 controlled output are often useful in silent breakpoints. @xref{Output}.
1914 For example, here is how you could use breakpoint commands to print the
1915 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
1928 One application for breakpoint commands is to correct one bug so you can
1929 test another. Put a breakpoint just after the erroneous line of code, give
1930 it a condition to detect the case in which something erroneous has been
1931 done, and give it commands to assign correct values to any variables that
1932 need them. End with the @samp{continue} command so that the program does not
1933 stop, and start with the @samp{silent} command so that no output is
1934 produced. Here is an example:
1945 One deficiency in the operation of automatically continuing breakpoints
1946 under Unix appears when your program uses raw mode for the terminal.
1947 _GDBN__ switches back to its own terminal modes (not raw) before executing
1948 commands, and then must switch back to raw mode when your program is
1949 continued. This causes any pending terminal input to be lost.
1950 In the GNU system, this will be fixed by changing the behavior of
1953 Under Unix, when you have this problem, you might be able to get around
1954 it by putting your actions into the breakpoint condition instead of
1955 commands. For example
1958 condition 5 (x = y + 4), 0
1962 specifies a condition expression (@xref{Expressions}) that will change
1963 @code{x} as needed, then always have the value 0 so the program will not
1964 stop. Loss of input is avoided here because break conditions are
1965 evaluated without changing the terminal modes. When you want to have
1966 nontrivial conditions for performing the side effects, the operators
1967 @samp{&&}, @samp{||} and @samp{?@dots{}:} may be useful.
1969 @node Error in Breakpoints,,,
1970 @subsection ``Cannot Insert Breakpoints''
1972 @c FIXME: "cannot insert breakpoints" error, v unclear.
1973 @c Q in pending mail to Gilmore. ---pesch@cygnus.com, 26mar91
1974 Under some operating systems, breakpoints cannot be used in a program if
1975 any other process is running that program. In this situation,
1976 attempting to run or continue a program with a breakpoint will cause _GDBN__
1977 to stop the other process.
1979 When this happens, you have three ways to proceed:
1983 Remove or disable the breakpoints, then continue.
1986 Suspend _GDBN__, and copy the file containing the program to a new name.
1987 Resume _GDBN__ and use the @samp{exec-file} command to specify that _GDBN__
1988 should run the program under that name. Then start the program again.
1991 Relink the program so that the text segment is nonsharable, using the
1992 linker option @samp{-N}. The operating system limitation may not apply
1993 to nonsharable executables.
2000 @dfn{Stepping} means setting your program in motion for a limited time, so
2001 that control will return automatically to _GDBN__ after one line of
2002 code or one machine instruction. Breakpoints are active during stepping
2003 and the program will stop for them even if it has not gone as far as the
2004 stepping command specifies.
2009 Continue running the program until control reaches a different source
2010 line, then stop it and return control to the debugger. This command is
2011 abbreviated @samp{s}.
2013 This command may be given when control is within a function for which
2014 there is no debugging information. In that case, execution will proceed
2015 until control reaches a different function, or is about to return from
2018 @item step @var{count}
2019 Continue running as in @samp{step}, but do so @var{count} times. If a
2020 breakpoint is reached or a signal not related to stepping occurs before
2021 @var{count} steps, stepping stops right away.
2025 Continue to the next source line in the current stack frame. Similar to
2026 @samp{step}, but any function calls appearing within the line of code
2027 are executed without stopping. Execution stops when control reaches a
2028 different line of code at the stack level which was executing when the
2029 @samp{next} command was given. This command is abbreviated @samp{n}.
2031 An argument is a repeat count, as in @samp{step}.
2033 @samp{next} within a function that lacks debugging information acts as does
2034 @samp{step}, but any function calls appearing within the code of the
2035 function are executed without stopping.
2036 @c FIXME: great; so what does *step* do within a fn that lacks debug info?
2040 Continue running until just after the selected stack frame returns (or
2041 until there is some other reason to stop, such as a fatal signal or a
2042 breakpoint). Print the value returned by the selected stack frame (if
2045 Contrast this with the @samp{return} command (@pxref{Returning}).
2049 Continue running until a source line past the current line, in the
2050 current stack frame, is reached. This command is used to avoid single
2051 stepping through a loop more than once. It is like the @samp{next}
2052 command, except that when @samp{until} encounters a jump, it
2053 automatically continues execution until the program counter is greater
2054 than the address of the jump.
2056 This means that when you reach the end of a loop after single stepping
2057 though it, @samp{until} will cause the program to continue execution
2058 until the loop is exited. In contrast, a @samp{next} command at the end
2059 of a loop will simply step back to the beginning of the loop, which
2060 would force you to step through the next iteration.
2062 @samp{until} always stops the program if it attempts to exit the current
2065 @samp{until} may produce somewhat counterintuitive results if the order
2066 of the source lines does not match the actual order of execution. For
2067 example, in a typical C @code{for}-loop, the third expression in the
2068 @code{for}-statement (the loop-step expression) is executed after the
2069 statements in the body of the loop, but is written before them.
2070 Therefore, the @samp{until} command would appear to step back to the
2071 beginning of the loop when it advances to this expression. However, it
2072 has not really done so, not in terms of the actual machine code.
2074 @samp{until} with no argument works by means of single
2075 instruction stepping, and hence is slower than @samp{until} with an
2078 @item until @var{location}
2079 Continue running the program until either the specified location is
2080 reached, or the current (innermost) stack frame returns. @var{location}
2081 is any of the forms of argument acceptable to @samp{break} (@pxref{Set
2082 Breaks}). This form of the command uses breakpoints, and hence is
2083 quicker than @samp{until} without an argument.
2089 Execute one machine instruction, then stop and return to the debugger.
2091 It is often useful to do @samp{display/i $pc} when stepping by machine
2092 instructions. This will cause the next instruction to be executed to
2093 be displayed automatically at each stop. @xref{Auto Display}.
2095 An argument is a repeat count, as in @samp{step}.
2101 Execute one machine instruction, but if it is a function call,
2102 proceed until the function returns.
2104 An argument is a repeat count, as in @samp{next}.
2107 A typical technique for using stepping is to put a breakpoint
2108 (@pxref{Breakpoints}) at the beginning of the function or the section of
2109 the program in which a problem is believed to lie, run the program until
2110 it stops at that breakpoint, and then step through the suspect area,
2111 examining the variables that are interesting, until you see the problem
2114 The @samp{continue} command can be used after stepping to resume execution
2115 until the next breakpoint or signal.
2120 After your program stops, most likely you will want it to run some more if
2121 the bug you are looking for has not happened yet.
2126 Continue running the program at the place where it stopped.
2129 If the program stopped at a breakpoint, the place to continue running
2130 is the address of the breakpoint. You might expect that continuing would
2131 just stop at the same breakpoint immediately. In fact, @samp{continue}
2132 takes special care to prevent that from happening. You do not need
2133 to delete the breakpoint to proceed through it after stopping at it.
2134 You can, however, specify an ignore-count for the breakpoint that the
2135 program stopped at, by means of an argument to the @samp{continue} command.
2138 If the program stopped because of a signal other than @code{SIGINT} or
2139 @code{SIGTRAP}, continuing will cause the program to see that signal.
2140 You may not want this to happen. For example, if the program stopped
2141 due to some sort of memory reference error, you might store correct
2142 values into the erroneous variables and continue, hoping to see more
2143 execution; but the program would probably terminate immediately as
2144 a result of the fatal signal once it sees the signal. To prevent this,
2145 you can continue with @samp{signal 0}. @xref{Signaling}. You can
2146 also act in advance to control what signals your program will see, using
2147 the @samp{handle} command (@pxref{Signals}).
2153 A signal is an asynchronous event that can happen in a program. The
2154 operating system defines the possible kinds of signals, and gives each
2155 kind a name and a number. For example, in Unix @code{SIGINT} is the
2156 signal a program gets when you type an interrupt (often @kbd{C-c});
2157 @code{SIGSEGV} is the signal a program gets from referencing a place in
2158 memory far away from all the areas in use; @code{SIGALRM} occurs when
2159 the alarm clock timer goes off (which happens only if the program has
2160 requested an alarm).
2162 @cindex fatal signals
2163 Some signals, including @code{SIGALRM}, are a normal part of the
2164 functioning of the program. Others, such as @code{SIGSEGV}, indicate
2165 errors; these signals are @dfn{fatal} (kill the program immediately) if the
2166 program has not specified in advance some other way to handle the signal.
2167 @code{SIGINT} does not indicate an error in the program, but it is normally
2168 fatal so it can carry out the purpose of the interrupt: to kill the program.
2170 _GDBN__ has the ability to detect any occurrence of a signal in the program
2171 running under _GDBN__'s control. You can tell _GDBN__ in advance what to do for
2172 each kind of signal.
2174 @cindex handling signals
2175 Normally, _GDBN__ is set up to ignore non-erroneous signals like @code{SIGALRM}
2176 (so as not to interfere with their role in the functioning of the program)
2177 but to stop the program immediately whenever an error signal happens.
2178 You can change these settings with the @samp{handle} command.
2183 Print a table of all the kinds of signals and how _GDBN__ has been told to
2184 handle each one. You can use this to see the signal numbers of all
2185 the defined types of signals.
2187 @item handle @var{signal} @var{keywords}@dots{}
2189 Change the way _GDBN__ handles signal @var{signal}. @var{signal} can be the
2190 number of a signal or its name (with or without the @samp{SIG} at the
2191 beginning). The @var{keywords} say what change to make.
2195 The keywords allowed by the @samp{handle} command can be abbreviated.
2196 Their full names are:
2200 _GDBN__ should not stop the program when this signal happens. It may
2201 still print a message telling you that the signal has come in.
2204 _GDBN__ should stop the program when this signal happens. This implies
2205 the @samp{print} keyword as well.
2208 _GDBN__ should print a message when this signal happens.
2211 _GDBN__ should not mention the occurrence of the signal at all. This
2212 implies the @samp{nostop} keyword as well.
2215 _GDBN__ should allow the program to see this signal; the program will be
2216 able to handle the signal, or may be terminated if the signal is fatal
2220 _GDBN__ should not allow the program to see this signal.
2224 When a signal has been set to stop the program, the program cannot see the
2225 signal until you continue. It will see the signal then, if @samp{pass} is
2226 in effect for the signal in question @i{at that time}. In other words,
2227 after _GDBN__ reports a signal, you can use the @samp{handle} command with
2228 @samp{pass} or @samp{nopass} to control whether that signal will be seen by
2229 the program when you later continue it.
2231 You can also use the @samp{signal} command to prevent the program from
2232 seeing a signal, or cause it to see a signal it normally would not see,
2233 or to give it any signal at any time. @xref{Signaling}.
2237 @chapter Examining the Stack
2239 When your program has stopped, the first thing you need to know is where it
2240 stopped and how it got there.
2243 Each time your program performs a function call, the information about
2244 where in the program the call was made from is saved in a block of data
2245 called a @dfn{stack frame}. The frame also contains the arguments of the
2246 call and the local variables of the function that was called. All the
2247 stack frames are allocated in a region of memory called the @dfn{call
2250 When your program stops, the _GDBN__ commands for examining the stack allow you
2251 to see all of this information.
2253 @cindex selected frame
2254 One of the stack frames is @dfn{selected} by _GDBN__ and many _GDBN__ commands
2255 refer implicitly to the selected frame. In particular, whenever you ask
2256 _GDBN__ for the value of a variable in the program, the value is found in the
2257 selected frame. There are special _GDBN__ commands to select whichever frame
2258 you are interested in.
2260 When the program stops, _GDBN__ automatically selects the currently executing
2261 frame and describes it briefly as the @samp{frame} command does
2262 (@pxref{Frame Info, Info}).
2265 @section Stack Frames
2269 The call stack is divided up into contiguous pieces called @dfn{stack
2270 frames}, or @dfn{frames} for short; each frame is the data associated
2271 with one call to one function. The frame contains the arguments given
2272 to the function, the function's local variables, and the address at
2273 which the function is executing.
2275 @cindex initial frame
2276 @cindex outermost frame
2277 @cindex innermost frame
2278 When your program is started, the stack has only one frame, that of the
2279 function @code{main}. This is called the @dfn{initial} frame or the
2280 @dfn{outermost} frame. Each time a function is called, a new frame is
2281 made. Each time a function returns, the frame for that function invocation
2282 is eliminated. If a function is recursive, there can be many frames for
2283 the same function. The frame for the function in which execution is
2284 actually occurring is called the @dfn{innermost} frame. This is the most
2285 recently created of all the stack frames that still exist.
2287 @cindex frame pointer
2288 Inside your program, stack frames are identified by their addresses. A
2289 stack frame consists of many bytes, each of which has its own address; each
2290 kind of computer has a convention for choosing one of those bytes whose
2291 address serves as the address of the frame. Usually this address is kept
2292 in a register called the @dfn{frame pointer register} while execution is
2293 going on in that frame.
2295 @cindex frame number
2296 _GDBN__ assigns numbers to all existing stack frames, starting with
2297 @code{0} for the innermost frame, @code{1} for the frame that called it,
2298 and so on upward. These numbers do not really exist in your program;
2299 they are assigned by _GDBN__ to give you a way of designating stack
2300 frames in _GDBN__ commands.
2302 @cindex frameless execution
2303 Some compilers allow functions to be compiled so that they operate
2304 without stack frames. (For example, the @code{_GCC__} option
2305 @samp{-fomit-frame-pointer} will generate functions without a frame.)
2306 This is occasionally done with heavily used library functions to save
2307 the frame setup time. _GDBN__ has limited facilities for dealing with
2308 these function invocations; if the innermost function invocation has no
2309 stack frame, _GDBN__ will give it a virtual stack frame of 0 and
2310 correctly allow tracing of the function call chain. Results are
2311 undefined if a function invocation besides the innermost one is
2317 A backtrace is a summary of how the program got where it is. It shows one
2318 line per frame, for many frames, starting with the currently executing
2319 frame (frame zero), followed by its caller (frame one), and on up the
2327 Print a backtrace of the entire stack: one line per frame for all
2328 frames in the stack.
2330 You can stop the backtrace at any time by typing the system interrupt
2331 character, normally @kbd{Control-C}.
2333 @item backtrace @var{n}
2335 Similar, but print only the innermost @var{n} frames.
2337 @item backtrace -@var{n}
2339 Similar, but print only the outermost @var{n} frames.
2344 The names @samp{where} and @samp{info stack} are additional aliases
2345 for @samp{backtrace}.
2347 Each line in the backtrace shows the frame number and the function name.
2348 The program counter value is also shown---unless you use @samp{set
2349 addressprint off}. The backtrace also shows the source file name and
2350 line number, as well as the arguments to the function. The program
2351 counter value is omitted if it is at the beginning of the code for that
2354 Here is an example of a backtrace. It was made with the command
2355 @samp{bt 3}, so it shows the innermost three frames.
2358 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8) at builtin.c:993
2359 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
2360 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
2362 (More stack frames follow...)
2366 The display for frame @code{#0} doesn't begin with a program counter
2367 value, indicating that the program has stopped at the beginning of the
2368 code for line @code{993} of @code{builtin.c}.
2371 @section Selecting a Frame
2373 Most commands for examining the stack and other data in the program work on
2374 whichever stack frame is selected at the moment. Here are the commands for
2375 selecting a stack frame; all of them finish by printing a brief description
2376 of the stack frame just selected.
2383 Select frame number @var{n}. Recall that frame zero is the innermost
2384 (currently executing) frame, frame one is the frame that called the
2385 innermost one, and so on. The highest-numbered frame is @code{main}'s
2388 @item frame @var{addr}
2390 Select the frame at address @var{addr}. This is useful mainly if the
2391 chaining of stack frames has been damaged by a bug, making it
2392 impossible for _GDBN__ to assign numbers properly to all frames. In
2393 addition, this can be useful when the program has multiple stacks and
2394 switches between them.
2398 Move @var{n} frames up the stack. For positive numbers @var{n}, this
2399 advances toward the outermost frame, to higher frame numbers, to frames
2400 that have existed longer. @var{n} defaults to one.
2404 Move @var{n} frames down the stack. For positive numbers @var{n}, this
2405 advances toward the innermost frame, to lower frame numbers, to frames
2406 that were created more recently. @var{n} defaults to one.
2409 All of these commands end by printing some information on the frame that
2410 has been selected: the frame number, the function name, the arguments, the
2411 source file and line number of execution in that frame, and the text of
2412 that source line. For example:
2416 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc) at env.c:10
2417 10 read_input_file (argv[i]);
2420 After such a printout, the @samp{list} command with no arguments will print
2421 ten lines centered on the point of execution in the frame. @xref{List}.
2424 @item up-silently @var{n}
2425 @itemx down-silently @var{n}
2426 @kindex down-silently
2428 These two commands are variants of @samp{up} and @samp{down},
2429 respectively; they differ in that they do their work silently, without
2430 causing display of the new frame. They are intended primarily for use
2431 in _GDBN__ command scripts, where the output might be unnecessary and
2437 @section Information on a Frame
2439 There are several other commands to print information about the selected
2445 When used without any argument, this command does not change which frame
2446 is selected, but prints a brief description of the currently
2447 selected stack frame. It can be abbreviated @samp{f}. With an
2448 argument, this command is used to select a stack frame (@pxref{Selection}).
2452 This command prints a verbose description of the selected stack frame,
2453 including the address of the frame, the addresses of the next frame down
2454 (called by this frame) and the next frame up (caller of this frame),
2455 the address of the frame's arguments, the program counter saved in it
2456 (the address of execution in the caller frame), and which registers
2457 were saved in the frame. The verbose description is useful when
2458 something has gone wrong that has made the stack format fail to fit
2459 the usual conventions.
2461 @item info frame @var{addr}
2462 Print a verbose description of the frame at address @var{addr},
2463 without selecting that frame. The selected frame remains unchanged by
2468 Print the arguments of the selected frame, each on a separate line.
2472 Print the local variables of the selected frame, each on a separate
2473 line. These are all variables declared static or automatic within all
2474 program blocks that execution in this frame is currently inside of.
2478 @cindex catch exceptions
2479 @cindex exception handlers
2480 Print a list of all the exception handlers that are active in the
2481 current stack frame at the current point of execution. To see other
2482 exception handlers, visit the associated frame (using the @samp{up},
2483 @samp{down}, or @samp{frame} commands); then type @samp{info catch}.
2484 @xref{Exception Handling}.
2488 @chapter Examining Source Files
2490 _GDBN__ can print parts of your program's source, since the debugging
2491 information recorded in your program tells _GDBN__ what source files
2492 were used to built it. When your program stops, _GDBN__
2493 spontaneously prints the line where it stopped. Likewise, when you
2494 select a stack frame (@pxref{Selection}), _GDBN__ prints the line
2495 where execution in that frame has stopped. You can also
2496 print parts of source files by explicit command.
2498 If you use _GDBN__ through its GNU Emacs interface, you may prefer to
2499 use Emacs facilities to view source; @pxref{Emacs}.
2502 @section Printing Source Lines
2506 To print lines from a source file, use the @samp{list} command
2507 (abbreviated @samp{l}). There are several ways to specify what part
2508 of the file you want to print.
2510 Here are the forms of the @samp{list} command most commonly used:
2513 @item list @var{linenum}
2514 Print ten lines centered around line number @var{linenum} in the
2515 current source file.
2517 @item list @var{function}
2518 Print ten lines centered around the beginning of function
2522 Print ten more lines. If the last lines printed were printed with a
2523 @samp{list} command, this prints ten lines following the last lines
2524 printed; however, if the last line printed was a solitary line printed
2525 as part of displaying a stack frame (@pxref{Stack}), this prints ten
2526 lines centered around that line.
2529 Print ten lines just before the lines last printed.
2532 Repeating a @samp{list} command with @key{RET} discards the argument,
2533 so it is equivalent to typing just @samp{list}. This is more useful
2534 than listing the same lines again. An exception is made for an
2535 argument of @samp{-}; that argument is preserved in repetition so that
2536 each repetition moves up in the source file.
2539 In general, the @samp{list} command expects you to supply zero, one or two
2540 @dfn{linespecs}. Linespecs specify source lines; there are several ways
2541 of writing them but the effect is always to specify some source line.
2542 Here is a complete description of the possible arguments for @samp{list}:
2545 @item list @var{linespec}
2546 Print ten lines centered around the line specified by @var{linespec}.
2548 @item list @var{first},@var{last}
2549 Print lines from @var{first} to @var{last}. Both arguments are
2552 @item list ,@var{last}
2553 Print ten lines ending with @var{last}.
2555 @item list @var{first},
2556 Print ten lines starting with @var{first}.
2559 Print ten lines just after the lines last printed.
2562 Print ten lines just before the lines last printed.
2565 As described in the preceding table.
2568 Here are the ways of specifying a single source line---all the
2573 Specifies line @var{number} of the current source file.
2574 When a @samp{list} command has two linespecs, this refers to
2575 the same source file as the first linespec.
2578 Specifies the line @var{offset} lines after the last line printed.
2579 When used as the second linespec in a @samp{list} command that has
2580 two, this specifies the line @var{offset} lines down from the
2584 Specifies the line @var{offset} lines before the last line printed.
2586 @item @var{filename}:@var{number}
2587 Specifies line @var{number} in the source file @var{filename}.
2589 @item @var{function}
2590 @c FIXME: "of the open-brace" is C-centric. When we add other langs...
2591 Specifies the line of the open-brace that begins the body of the
2592 function @var{function}.
2594 @item @var{filename}:@var{function}
2595 Specifies the line of the open-brace that begins the body of the
2596 function @var{function} in the file @var{filename}. You only need the
2597 file name with a function name to avoid ambiguity when there are
2598 identically named functions in different source files.
2600 @item *@var{address}
2601 Specifies the line containing the program address @var{address}.
2602 @var{address} may be any expression.
2606 @section Searching Source Files
2609 @kindex forward-search
2610 @kindex reverse-search
2612 There are two commands for searching through the current source file for a
2615 The command @samp{forward-search @var{regexp}} checks each line, starting
2616 with the one following the last line listed, for a match for @var{regexp}.
2617 It lists the line that is found. You can abbreviate the command name
2618 as @samp{fo}. The synonym @samp{search @var{regexp}} is also supported.
2620 The command @samp{reverse-search @var{regexp}} checks each line, starting
2621 with the one before the last line listed and going backward, for a match
2622 for @var{regexp}. It lists the line that is found. You can abbreviate
2623 this command as @samp{rev}.
2625 @node Source Path,,,
2626 @section Specifying Source Directories
2629 @cindex directories for source files
2630 Executable programs sometimes do not record the directories of the source
2631 files from which they were compiled, just the names. Even when they do,
2632 the directories could be moved between the compilation and your debugging
2633 session. _GDBN__ has a list of directories to search for source files;
2634 this is called the @dfn{source path}. Each time _GDBN__ wants a source file,
2635 it tries all the directories in the list, in the order they are present
2636 in the list, until it finds a file with the desired name. Note that
2637 the executable search path is @emph{not} used for this purpose. Neither is
2638 the current working directory, unless it happens to be in the source
2641 If _GDBN__ can't find a source file in the source path, and the object
2642 program records a directory, _GDBN__ tries that directory too. If the
2643 source path is empty, and there is no record of the compilation
2644 directory, _GDBN__ will, as a last resort, look in the current
2647 Whenever you reset or rearrange the source path, _GDBN__ will clear out
2648 any information it has cached about where source files are found, where
2649 each line is in the file, etc.
2652 When you start _GDBN__, its source path is empty.
2653 To add other directories, use the @samp{directory} command.
2656 @item directory @var{dirname} @dots
2657 Add directory @var{dirname} to the front of the source path. Several
2658 directory names may be given to this command, separated by @samp{:} or
2659 whitespace. You may specify a directory that is already in the source
2660 path; this moves it forward, so it will be searched sooner. You can use
2661 the string @samp{$cwd} to refer to the current working directory, and
2662 @samp{$cdir} to refer to the compilation directory (if one is recorded).
2665 Reset the source path to empty again. This requires confirmation.
2667 The @samp{directory} command will not repeat if you press @key{RET} a
2668 second time after executing it once.
2670 @item info directories
2671 @kindex info directories
2672 Print the source path: show which directories it contains.
2675 If your source path is cluttered with directories that are no longer of
2676 interest, _GDBN__ may sometimes cause confusion by finding the wrong
2677 versions of source. You can correct the situation as follows:
2681 Use @samp{directory} with no argument to reset the source path to empty.
2684 Use @samp{directory} with suitable arguments to add any other
2685 directories you want in the source path. You can add all the directories
2689 @node Machine Code,,,
2690 @section Source and Machine Code
2691 You can use the command @samp{info line} to map source lines to program
2692 addresses, and the command @samp{disassemble} or its synonym
2693 @samp{disasm} to display a range of addresses as machine instructions.
2696 @item info line @var{linespec}
2698 Print the starting and ending addresses of the compiled code for
2699 source line @var{linespec}.
2702 After @samp{info line}, the default address for the @samp{x}
2703 command is changed to the starting address of the line, so that
2704 @samp{x/i} is sufficient to begin examining the machine code
2705 (@pxref{Memory}). Also, this address is saved as the value of the
2706 convenience variable @code{$_} (@pxref{Convenience Vars}).
2712 This specialized command is provided to dump a range of memory as
2713 machine instructions. The default memory range is the function
2714 surrounding the program counter of the selected frame. A single
2715 argument to this command is a program counter value; the function
2716 surrounding this value will be dumped. Two arguments (separated by one
2717 or more spaces) specify a range of addresses (first inclusive, second
2718 exclusive) to be dumped. The two spellings, @samp{disasm} and
2719 @samp{disassemble}, are equivalent.
2723 @chapter Examining Data
2725 @cindex printing data
2726 @cindex examining data
2729 @c "inspect" isn't quite a synonym if you're using Epoch, which we don't
2730 @c document because it's nonstandard... Under Epoch it displays in a
2731 @c different window or something like that.
2732 The usual way to examine data in your program is with the @samp{print}
2733 command (abbreviated @samp{p}), or its synonym @samp{inspect}. It
2734 evaluates and prints the value of any valid expression of the language
2735 the program is written in (for now, C or C++). You type
2742 where @var{exp} is any valid expression (in the source language), and
2743 the value of @var{exp} is printed in a format appropriate to its data
2746 A more low-level way of examining data is with the @samp{x} command.
2747 It examines data in memory at a specified address and prints it in a
2748 specified format. @xref{Memory}.
2750 @node Expressions,,,
2751 @section Expressions
2754 @code{print} and many other _GDBN__ commands accept an expression and
2755 compute its value. Any kind of constant, variable or operator defined
2756 by the programming language you are using is legal in an expression in
2757 _GDBN__. This includes conditional expressions, function calls, casts
2758 and string constants. It unfortunately does not include symbols defined
2759 by preprocessor @code{#define} commands, or C++ expressions involving
2760 @samp{::}, the name resolution operator.
2761 @c FIXME: actually C++ a::b works except in obscure circumstances where it
2762 @c FIXME...can conflict with GDB's own name scope resolution.
2764 Casts are supported in all languages, not just in C, because it is so
2765 useful to cast a number into a pointer so as to examine a structure
2766 at that address in memory.
2768 _GDBN__ supports three kinds of operator in addition to those of programming
2773 @samp{@@} is a binary operator for treating parts of memory as arrays.
2774 @xref{Arrays}, for more information.
2777 @samp{::} allows you to specify a variable in terms of the file or
2778 function where it is defined. @xref{Variables}.
2780 @item @{@var{type}@} @var{addr}
2781 Refers to an object of type @var{type} stored at address @var{addr} in
2782 memory. @var{addr} may be any expression whose value is an integer or
2783 pointer (but parentheses are required around binary operators, just as in
2784 a cast). This construct is allowed regardless of what kind of data is
2785 officially supposed to reside at @var{addr}.@refill
2789 @section Program Variables
2791 The most common kind of expression to use is the name of a variable
2794 Variables in expressions are understood in the selected stack frame
2795 (@pxref{Selection}); they must either be global (or static) or be visible
2796 according to the scope rules of the programming language from the point of
2797 execution in that frame. This means that in the function
2812 the variable @code{a} is usable whenever the program is executing
2813 within the function @code{foo}, but the variable @code{b} is visible
2814 only while the program is executing inside the block in which @code{b}
2817 There is an exception: you can refer to a variable or function whose
2818 scope is a single source file even if the current execution point is not
2819 in this file. But it is possible to have more than one such variable or
2820 function with the same name (in different source files). If that happens,
2821 referring to that name has unpredictable effects. If you wish, you can
2822 specify a variable in a particular file, using the colon-colon construct:
2828 @var{block}::@var{variable}
2832 Here @var{block} is the name of the source file whose variable you want.
2834 @cindex name resolution (C++)
2835 Unfortunately, this use of @samp{::} conflicts with the very similar use
2836 of the same notation in C++; accordingly, _GDBN__ does not support use of
2837 the C++ name resolution operator in _GDBN__ expressions.
2840 @section Artificial Arrays
2842 @cindex artificial array
2844 It is often useful to print out several successive objects of the
2845 same type in memory; a section of an array, or an array of
2846 dynamically determined size for which only a pointer exists in the
2849 This can be done by constructing an @dfn{artificial array} with the
2850 binary operator @samp{@@}. The left operand of @samp{@@} should be
2851 the first element of the desired array, as an individual object.
2852 The right operand should be the length of the array. The result is
2853 an array value whose elements are all of the type of the left argument.
2854 The first element is actually the left argument; the second element
2855 comes from bytes of memory immediately following those that hold the
2856 first element, and so on. Here is an example. If a program says
2859 int *array = (int *) malloc (len * sizeof (int));
2863 you can print the contents of @code{array} with
2869 The left operand of @samp{@@} must reside in memory. Array values made
2870 with @samp{@@} in this way behave just like other arrays in terms of
2871 subscripting, and are coerced to pointers when used in expressions.
2872 Artificial arrays most often appear in expressions via the value history
2873 (@pxref{Value History}), after printing one out.)
2875 @node Output formats,,,
2876 @section Output formats
2878 @cindex formatted output
2879 @cindex output formats
2880 By default, _GDBN__ prints a value according to its data type. Sometimes
2881 this is not what you want. For example, you might want to print a number
2882 in hex, or a pointer in decimal. Or you might want to view data in memory
2883 at a certain address as a character string or as an instruction. To do
2884 these things, specify an @dfn{output format} when you print a value.
2886 The simplest use of output formats is to say how to print a value
2887 already computed. This is done by starting the arguments of the
2888 @samp{print} command with a slash and a format letter. The format
2889 letters supported are:
2893 Regard the bits of the value as an integer, and print the integer in
2897 Print as integer in signed decimal.
2900 Print as integer in unsigned decimal.
2903 Print as integer in octal.
2906 Print as an address, both absolute in hex and as an offset from the
2907 nearest preceding symbol. This format can be used to discover where (in
2908 what function) an unknown address is located:
2910 (_GDBP__) p/a 0x54320
2911 _0__$3 = 0x54320 <_initialize_vx+396>_1__
2916 Regard as an integer and print it as a character constant.
2919 Regard the bits of the value as a floating point number and print
2920 using typical floating point syntax.
2923 For example, to print the program counter in hex (@pxref{Registers}), type
2930 Note that no space is required before the slash; this is because command
2931 names in _GDBN__ cannot contain a slash.
2933 To reprint the last value in the value history with a different format,
2934 you can use the @samp{print} command with just a format and no
2935 expression. For example, @samp{p/x} reprints the last value in hex.
2938 @section Examining Memory
2940 @cindex examining memory
2943 @item x/@var{Nuf} @var{expr}
2944 The command @samp{x} (for `examine') can be used to examine memory
2945 without being constrained by your program's data types. You can specify
2946 the unit size @var{u} of memory to inspect, and a repeat count @var{N} of how
2947 many of those units to display. @samp{x} understands the formats
2948 @var{f} used by @samp{print}; two additional formats, @samp{s} (string)
2949 and @samp{i} (machine instruction) can be used without specifying a unit
2953 For example, @samp{x/3hu 0x54320} is a request to display three halfwords
2954 (@code{h}) of memory, formatted as unsigned decimal integers (@code{u}),
2955 starting at address @code{0x54320}. @samp{x/4wx $sp} prints the four
2956 words (@code{w}) of memory above the stack pointer (here, @samp{$sp};
2957 @pxref{Registers}) in hexadecimal (@code{x}).
2959 Since the letters indicating unit sizes are all distinct from the
2960 letters specifying output formats, you don't have to remember whether
2961 unit size or format comes first; either order will work. The output
2962 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
2964 After the format specification, you indicate the address where _GDBN__
2965 is to begin reading memory for display, using an expression. The
2966 expression need not have a pointer value (though it may); it is always
2967 interpreted as an integer address of a byte of memory.
2968 @xref{Expressions} for more information on expressions.
2970 These are the memory units @var{u} you can specify with @samp{x}:
2974 Examine individual bytes.
2977 Examine halfwords (two bytes each).
2980 Examine words (four bytes each).
2983 Many assemblers and cpu designers still use `word' for a 16-bit quantity,
2984 as a holdover from specific predecessor machines of the 1970's that really
2985 did use two-byte words. But more generally the term `word' has always
2986 referred to the size of quantity that a machine normally operates on and
2987 stores in its registers. This is 32 bits for all the machines that _GDBN__
2991 Examine giant words (8 bytes).
2994 You can combine these unit specifications with any of the formats
2995 described for @samp{print}. @xref{Output formats}.
2997 @samp{x} has two additional output specifications which derive the unit
2998 size from the data inspected:
3002 Print a null-terminated string of characters. Any explicitly specified
3003 unit size is ignored; instead, the unit is however many bytes it takes
3004 to reach a null character (including the null character).
3007 Print a machine instruction in assembler syntax (or nearly). Any
3008 specified unit size is ignored; the number of bytes in an instruction
3009 varies depending on the type of machine, the opcode and the addressing
3010 modes used. The command @samp{disassemble} gives an alternative way of
3011 inspecting machine instructions. @xref{Machine Code}.
3014 If you omit either the format @var{f} or the unit size @var{u}, @samp{x}
3015 will use the same one that was used last. If you don't use any letters
3016 after the slash, you can omit the slash as well.
3018 You can also omit the address to examine. Then the address used is just
3019 after the last unit examined. This is why string and instruction
3020 formats actually compute a unit-size based on the data: so that the next
3021 string or instruction examined will start in the right place.
3023 When the @samp{print} command shows a value that resides in memory,
3024 @samp{print} also sets the default address for the @samp{x} command.
3025 @samp{info line} also sets the default for @samp{x}, to the address of
3026 the start of the machine code for the specified line (@pxref{Machine
3027 Code}), and @samp{info breakpoints} sets it to the address of the last
3028 breakpoint listed (@pxref{Set Breaks}).
3030 When you use @key{RET} to repeat an @samp{x} command, the address
3031 specified previously (if any) is ignored, so that the repeated command
3032 examines the successive locations in memory rather than the same ones.
3034 You can examine several consecutive units of memory with one command by
3035 writing a repeat-count after the slash (before the format letters, if
3036 any). Omitting the repeat count @var{N} displays one unit of the
3037 appropriate size. The repeat count must be a decimal integer. It has
3038 the same effect as repeating the @samp{x} command @var{N} times except
3039 that the output may be more compact, with several units per line. For
3047 prints ten instructions starting with the one to be executed next in the
3048 selected frame. After doing this, you could print a further seven
3056 ---where the format and address are allowed to default.
3060 The addresses and contents printed by the @samp{x} command are not put
3061 in the value history because there is often too much of them and they
3062 would get in the way. Instead, _GDBN__ makes these values available for
3063 subsequent use in expressions as values of the convenience variables
3064 @code{$_} and @code{$__}. After an @samp{x} command, the last address
3065 examined is available for use in expressions in the convenience variable
3066 @code{$_}. The contents of that address, as examined, are available in
3067 the convenience variable @code{$__}.
3069 If the @samp{x} command has a repeat count, the address and contents saved
3070 are from the last memory unit printed; this is not the same as the last
3071 address printed if several units were printed on the last line of output.
3073 @node Auto Display,,,
3074 @section Automatic Display
3075 @cindex automatic display
3076 @cindex display of expressions
3078 If you find that you want to print the value of an expression frequently
3079 (to see how it changes), you might want to add it to the @dfn{automatic
3080 display list} so that _GDBN__ will print its value each time the program stops.
3081 Each expression added to the list is given a number to identify it;
3082 to remove an expression from the list, you specify that number.
3083 The automatic display looks like this:
3087 3: bar[5] = (struct hack *) 0x3804
3091 showing item numbers, expressions and their current values. As with
3092 displays you request manually using @samp{x} or @samp{print}, you can
3093 specify the output format you prefer; in fact, @dfn{display} decides
3094 whether to use @code{print} or @code{x} depending on how elaborate your
3095 format specification is---it uses @code{x} if you specify a unit size,
3096 or one of the two formats (@samp{i} and @samp{s}) that are only
3097 supported by @code{x}; otherwise it uses @code{print}.
3100 @item display @var{exp}
3102 Add the expression @var{exp} to the list of expressions to display
3103 each time the program stops. @xref{Expressions}.
3105 @samp{display} will not repeat if you press @key{RET} again after using it.
3107 @item display/@var{fmt} @var{exp}
3108 For @var{fmt} specifying only a display format and not a size or
3109 count, add the expression @var{exp} to the auto-display list but
3110 arranges to display it each time in the specified format @var{fmt}.
3111 @xref{Output formats}.
3113 @item display/@var{fmt} @var{addr}
3114 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
3115 number of units, add the expression @var{addr} as a memory address to
3116 be examined each time the program stops. Examining means in effect
3117 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory}.
3120 For example, @samp{display/i $pc} can be helpful, to see the machine
3121 instruction about to be executed each time execution stops (@samp{$pc}
3122 is a common name for the program counter; @pxref{Registers}).
3125 @item undisplay @var{dnums}@dots{}
3126 @itemx delete display @var{dnums}@dots{}
3127 @kindex delete display
3129 Remove item numbers @var{dnums} from the list of expressions to display.
3131 @samp{undisplay} will not repeat if you press @key{RET} after using it.
3133 @item disable display @var{dnums}@dots{}
3134 @kindex disable display
3135 Disable the display of item numbers @var{dnums}. A disabled display
3136 item is not printed automatically, but is not forgotten. It may be
3137 enabled again later.
3139 @item enable display @var{dnums}@dots{}
3140 @kindex enable display
3141 Enable display of item numbers @var{dnums}. It becomes effective once
3142 again in auto display of its expression, until you specify otherwise.
3145 Display the current values of the expressions on the list, just as is
3146 done when the program stops.
3149 @kindex info display
3150 Print the list of expressions previously set up to display
3151 automatically, each one with its item number, but without showing the
3152 values. This includes disabled expressions, which are marked as such.
3153 It also includes expressions which would not be displayed right now
3154 because they refer to automatic variables not currently available.
3157 If a display expression refers to local variables, then it does not make
3158 sense outside the lexical context for which it was set up. Such an
3159 expression is disabled when execution enters a context where one of its
3160 variables is not defined. For example, if you give the command
3161 @samp{display name} while inside a function with an argument
3162 @code{name}, then this argument will be displayed while the program
3163 continues to stop inside that function. When it stops elsewhere---where
3164 there is no variable @samp{name}---display is disabled. The next time
3165 your program stops where @samp{name} is meaningful, you can enable the
3166 display expression once again.
3168 @node Print Settings,,,
3169 @section Print Settings
3171 @cindex format options
3172 @cindex print settings
3173 _GDBN__ provides the following ways to control how arrays, structures,
3174 and symbols are printed.
3177 @item set array-max @var{number-of-elements}
3178 @kindex set array-max
3179 If _GDBN__ is printing a large array, it will stop printing after it has
3180 printed the number of elements set by the @samp{set array-max} command.
3181 This limit also applies to the display of strings.
3183 @item show array-max
3184 @kindex show array-max
3185 Display the number of elements of a large array that _GDBN__ will print
3186 before losing patience.
3188 @item set arrayprint
3189 @itemx set arrayprint on
3190 @kindex set arrayprint
3191 _GDBN__ will pretty print arrays. This format is more convenient to read,
3192 but uses more space. The default is off.
3194 @item set arrayprint off.
3195 Return to compressed format for arrays.
3197 @item show arrayprint
3198 @kindex show arrayprint
3199 Show whether compressed or pretty format is selected for displaying
3203 @itemx set demangle on
3204 @kindex set demangle
3205 Print C++ names in their source form rather than in the mangled form
3206 in which they are passed to the assembler and linker for type-safe linkage.
3210 @kindex show demangle
3211 Show whether C++ names will be printed in mangled or demangled form.
3213 @item set asm-demangle
3214 @itemx set asm-demangle on
3215 @kindex set asm-demangle
3216 Print C++ names in their source form rather than their mangled form, even
3217 in assembler code printouts such as instruction disassemblies.
3220 @item show asm-demangle
3221 @kindex show asm-demangle
3222 Show whether C++ names in assembly listings will be printed in mangled
3226 @itemx set vtblprint on
3227 @kindex set vtblprint
3228 Pretty print C++ virtual function tables. The default is off.
3230 @item set vtblprint off
3231 Do not pretty print C++ virtual function tables.
3233 @item show vtblprint
3234 @kindex show vtblprint
3235 Show whether C++ virtual function tables are pretty printed, or not.
3237 @item set addressprint
3238 @item set addressprint on
3239 @kindex set addressprint
3240 _GDBN__ will print memory addresses in stack traces, structure values, pointer
3241 values, breakpoints, etc. The default is on.
3243 @item set addressprint off
3244 Do not print addresses.
3246 @item show addressprint
3247 @kindex show addressprint
3248 Show whether or not addresses are to be printed.
3250 @item set prettyprint on
3251 @kindex set prettyprint
3252 Cause _GDBN__ to print structures in an indented format with one member per
3266 @item set prettyprint off
3267 Cause _GDBN__ to print structures in a compact format, like this:
3270 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, meat \
3275 This is the default format.
3277 @item show prettyprint
3278 @kindex show prettyprint
3279 Show which format _GDBN__ will use to print structures.
3281 @item set unionprint on
3282 @kindex set unionprint
3283 Tell _GDBN__ to print unions which are contained in structures. This is the
3286 @item set unionprint off
3287 Tell _GDBN__ not to print unions which are contained in structures.
3289 @item show unionprint
3290 @kindex show unionprint
3291 Ask _GDBN__ whether or not it will print unions which are contained in
3294 For example, given the declarations
3297 typedef enum @{Tree, Bug@} Species;
3298 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
3299 typedef enum @{Caterpillar, Cocoon, Butterfly@} Bug_forms;
3309 struct thing foo = @{Tree, @{Acorn@}@};
3313 with @samp{set unionprint on} in effect @samp{p foo} would print
3316 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
3320 and with @samp{set unionprint off} in effect it would print
3323 $1 = @{it = Tree, form = @{...@}@}
3327 @node Value History,,,
3328 @section Value History
3330 @cindex value history
3331 Values printed by the @samp{print} command are saved in _GDBN__'s @dfn{value
3332 history} so that you can refer to them in other expressions. Values are
3333 kept until the symbol table is re-read or discarded (for example with
3334 the @samp{file} or @samp{symbol-file} commands). When the symbol table
3335 changes, the value history is discarded, since the values may contain
3336 pointers back to the types defined in the symbol table.
3340 @cindex history number
3341 The values printed are given @dfn{history numbers} for you to refer to them
3342 by. These are successive integers starting with 1. @samp{print} shows you
3343 the history number assigned to a value by printing @samp{$@var{num} = }
3344 before the value; here @var{num} is the history number.
3346 To refer to any previous value, use @samp{$} followed by the value's
3347 history number. The way @samp{print} labels its output is designed to
3348 remind you of this. Just @code{$} refers to the most recent value in
3349 the history, and @code{$$} refers to the value before that.
3350 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
3351 is the value just prior to @code{$$}, @code{$$1} is equivalent to
3352 @code{$$}, and @code{$$0} is equivalent to @code{$}.
3354 For example, suppose you have just printed a pointer to a structure and
3355 want to see the contents of the structure. It suffices to type
3361 If you have a chain of structures where the component @samp{next} points
3362 to the next one, you can print the contents of the next one with this:
3369 You can print successive links in the chain by repeating this
3370 command---which you can do by just typing @key{RET}.
3372 Note that the history records values, not expressions. If the value of
3373 @code{x} is 4 and you type these commands:
3381 then the value recorded in the value history by the @samp{print} command
3382 remains 4 even though the value of @code{x} has changed.
3388 @kindex info history
3389 These two commands are synonymous. Either form will print the last ten
3390 values in the value history, with their item numbers. This is like
3391 @samp{p@ $$9} repeated ten times, except that @samp{info values} does
3392 not change the history.
3394 @item info values @var{n}
3395 Print ten history values centered on history item number @var{n}.
3398 Print ten history values just after the values last printed.
3401 @node Convenience Vars,,,
3402 @section Convenience Variables
3404 @cindex convenience variables
3405 _GDBN__ provides @dfn{convenience variables} that you can use within
3406 _GDBN__ to hold on to a value and refer to it later. These variables
3407 exist entirely within _GDBN__; they are not part of your program, and
3408 setting a convenience variable has no direct effect on further execution
3409 of your program. That's why you can use them freely.
3411 Convenience variables have names starting with @samp{$}. Any name starting
3412 with @samp{$} can be used for a convenience variable, unless it is one of
3413 the predefined machine-specific register names (@pxref{Registers}).
3415 You can save a value in a convenience variable with an assignment
3416 expression, just as you would set a variable in your program. Example:
3419 set $foo = *object_ptr
3423 would save in @code{$foo} the value contained in the object pointed to by
3426 Using a convenience variable for the first time creates it; but its value
3427 is @code{void} until you assign a new value. You can alter the value with
3428 another assignment at any time.
3430 Convenience variables have no fixed types. You can assign a convenience
3431 variable any type of value, including structures and arrays, even if
3432 that variable already has a value of a different type. The convenience
3433 variable, when used as an expression, has the type of its current value.
3436 @item info convenience
3437 @kindex info convenience
3438 Print a list of convenience variables used so far, and their values.
3439 Abbreviated @samp{i con}.
3442 One of the ways to use a convenience variable is as a counter to be
3443 incremented or a pointer to be advanced. For example, to print
3444 a field from successive elements of an array of structures:
3448 print bar[$i++]->contents
3449 @i{@dots{} repeat that command by typing @key{RET}.}
3452 Some convenience variables are created automatically by _GDBN__ and given
3453 values likely to be useful.
3457 The variable @code{$_} is automatically set by the @samp{x} command to
3458 the last address examined (@pxref{Memory}). Other commands which
3459 provide a default address for @samp{x} to examine also set @code{$_}
3460 to that address; these commands include @samp{info line} and @samp{info
3464 The variable @code{$__} is automatically set by the @samp{x} command
3465 to the value found in the last address examined.
3472 Machine register contents can be referred to in expressions as variables
3473 with names starting with @samp{$}. The names of registers are different
3474 for each machine; use @samp{info registers} to see the names used on
3475 your machine. The names @code{$pc} and @code{$sp} are used on most
3476 machines for the program counter register and the stack pointer. Often
3477 @code{$fp} is used for a register that contains a pointer to the current
3478 stack frame, and @code{$ps} is sometimes used for a register that
3479 contains the processor status. These standard register names may be
3480 available on your machine even though the @code{info registers} command
3481 shows other names. For example, on the SPARC, @code{info registers}
3482 displays the processor status register as @code{$psr} but you can also
3483 refer to it as @code{$ps}.
3485 _GDBN__ always considers the contents of an ordinary register as an
3486 integer when the register is examined in this way. Some machines have
3487 special registers which can hold nothing but floating point; these
3488 registers are considered to have floating point values. There is no way
3489 to refer to the contents of an ordinary register as floating point value
3490 (although you can @emph{print} it as a floating point value with
3491 @samp{print/f $@var{regname}}).
3493 Some registers have distinct ``raw'' and ``virtual'' data formats. This
3494 means that the data format in which the register contents are saved by
3495 the operating system is not the same one that your program normally
3496 sees. For example, the registers of the 68881 floating point
3497 coprocessor are always saved in ``extended'' (raw) format, but all C
3498 programs expect to work with ``double'' (virtual) format. In such
3499 cases, _GDBN__ normally works with the virtual format only (the format that
3500 makes sense for your program), but the @samp{info registers} command
3501 prints the data in both formats.
3503 Normally, register values are relative to the selected stack frame
3504 (@pxref{Selection}). This means that you get the value that the register
3505 would contain if all stack frames farther in were exited and their saved
3506 registers restored. In order to see the contents of hardware registers,
3507 you must select the innermost frame (with @samp{frame 0}).
3509 However, _GDBN__ must deduce where registers are saved, from the machine
3510 code generated by your compiler. If some registers are not saved, or if
3511 _GDBN__ is unable to locate the saved registers, the selected stack
3512 frame will make no difference.
3515 @item info registers
3516 @kindex info registers
3517 Print the names and values of all registers (in the selected stack frame).
3519 @item info registers @var{regname}
3520 Print the relativized value of register @var{regname}. @var{regname}
3521 may be any register name valid on the machine you are using, with
3522 or without the initial @samp{$}.
3525 For example, you could print the program counter in hex with
3532 or print the instruction to be executed next with
3539 or add four to the stack pointer with
3546 The last is a way of removing one word from the stack, on machines where
3547 stacks grow downward in memory (most machines, nowadays). This assumes
3548 that the innermost stack frame is selected. Setting @code{$sp} is
3549 not allowed when other stack frames are selected. (To pop entire frames
3550 off the stack, regardless of machine architecture, use @samp{return};
3554 @chapter Examining the Symbol Table
3556 The commands described in this section allow you to inquire about the
3557 symbols (names of variables, functions and types) defined in your
3558 program. This information is inherent in the text of your program and
3559 does not change as the program executes. _GDBN__ finds it in your
3560 program's symbol table, as indicated when you started _GDBN__
3561 (@pxref{File Options}), or by one of the file-management commands
3565 @item info address @var{symbol}
3566 @kindex info address
3567 Describe where the data for @var{symbol} is stored. For a register
3568 variable, this says which register it is kept in. For a non-register
3569 local variable, this prints the stack-frame offset at which the variable
3572 Note the contrast with @samp{print &@var{symbol}}, which does not work
3573 at all for a register variables, and for a stack local variable prints
3574 the exact address of the current instantiation of the variable.
3576 @item whatis @var{exp}
3578 Print the data type of expression @var{exp}. @var{exp} is not
3579 actually evaluated, and any side-effecting operations (such as
3580 assignments or function calls) inside it do not take place.
3584 Print the data type of @code{$}, the last value in the value history.
3586 @item ptype @var{typename}
3588 Print a description of data type @var{typename}. @var{typename} may be
3589 the name of a type, or for C code it may have the form
3590 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
3591 @samp{enum @var{enum-tag}}.@refill
3593 @item ptype @var{exp}
3594 Print a description of the type of expression @var{exp}. @samp{ptype}
3595 differs from @samp{whatis} by printing a detailed description, instead of just
3596 the name of the type. For example, if your program declares a variable
3599 struct complex {double real; double imag;} v;
3602 compare the output of the two commands:
3605 type = struct complex
3607 type = struct complex {
3614 @kindex info sources
3615 Print the names of all source files in the program for which there
3616 is debugging information.
3618 @item info functions
3619 @kindex info functions
3620 Print the names and data types of all defined functions.
3622 @item info functions @var{regexp}
3623 Print the names and data types of all defined functions
3624 whose names contain a match for regular expression @var{regexp}.
3625 Thus, @samp{info fun step} finds all functions whose names
3626 include @samp{step}; @samp{info fun ^step} finds those whose names
3627 start with @samp{step}.
3629 @item info variables
3630 @kindex info variables
3631 Print the names and data types of all variables that are declared
3632 outside of functions (i.e., excluding local variables).
3634 @item info variables @var{regexp}
3635 Print the names and data types of all variables (except for local
3636 variables) whose names contain a match for regular expression
3641 This was never implemented.
3643 @itemx info methods @var{regexp}
3644 @kindex info methods
3645 The @samp{info-methods} command permits the user to examine all defined
3646 methods within C++ program, or (with the @var{regexp} argument) a
3647 specific set of methods found in the various C++ classes. Many
3648 C++ classes provide a large number of methods. Thus, the output
3649 from the @samp{ptype} command can be overwhelming and hard to use. The
3650 @samp{info-methods} command filters the methods, printing only those
3651 which match the regular-expression @var{regexp}.
3654 @item printsyms @var{filename}
3656 Write a complete dump of the debugger's symbol data into the
3657 file @var{filename}.
3661 @chapter Altering Execution
3663 Once you think you have found an error in the program, you might want to
3664 find out for certain whether correcting the apparent error would lead to
3665 correct results in the rest of the run. You can find the answer by
3666 experiment, using the _GDBN__ features for altering execution of the
3669 For example, you can store new values into variables or memory
3670 locations, give the program a signal, restart it at a different address,
3671 or even return prematurely from a function to its caller.
3674 @section Assignment to Variables
3677 @cindex setting variables
3678 To alter the value of a variable, evaluate an assignment expression.
3679 @xref{Expressions}. For example,
3686 would store the value 4 into the variable @code{x}, and then print the
3687 value of the assignment expression (which is 4). All the assignment
3688 operators of C are supported, including the increment operators
3689 @samp{++} and @samp{--}, and combining assignments such as @samp{+=} and
3693 @kindex set variable
3694 @cindex variables, setting
3695 If you are not interested in seeing the value of the assignment, use the
3696 @samp{set} command instead of the @samp{print} command. @samp{set} is
3697 really the same as @samp{print} except that the expression's value is not
3698 printed and is not put in the value history (@pxref{Value History}). The
3699 expression is evaluated only for its effects.
3701 If the beginning of the argument string of the @samp{set} command
3702 appears identical to a @samp{set} subcommand, use the @samp{set
3703 variable} command instead of just @samp{set}. This command is identical
3704 to @samp{set} except for its lack of subcommands.
3706 _GDBN__ allows more implicit conversions in assignments than C does; you can
3707 freely store an integer value into a pointer variable or vice versa, and
3708 any structure can be converted to any other structure that is the same
3710 @comment FIXME: how do structs align/pad in these conversions?
3711 @comment /pesch@cygnus.com 18dec1990
3713 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
3714 construct to generate a value of specified type at a specified address
3715 (@pxref{Expressions}). For example, @code{@{int@}0x83040} refers
3716 to memory location @code{0x83040} as an integer (which implies a certain size
3717 and representation in memory), and
3720 set @{int@}0x83040 = 4
3724 stores the value 4 into that memory location.
3727 @section Continuing at a Different Address
3729 Ordinarily, when you continue the program, you do so at the place where
3730 it stopped, with the @samp{continue} command. You can instead continue at
3731 an address of your own choosing, with the following commands:
3734 @item jump @var{linenum}
3736 Resume execution at line number @var{linenum}. Execution will stop
3737 immediately if there is a breakpoint there.
3739 The @samp{jump} command does not change the current stack frame, or
3740 the stack pointer, or the contents of any memory location or any
3741 register other than the program counter. If line @var{linenum} is in
3742 a different function from the one currently executing, the results may
3743 be bizarre if the two functions expect different patterns of arguments or
3744 of local variables. For this reason, the @samp{jump} command requests
3745 confirmation if the specified line is not in the function currently
3746 executing. However, even bizarre results are predictable if you are
3747 well acquainted with the machine-language code of the program.
3749 @item jump *@var{address}
3750 Resume execution at the instruction at address @var{address}.
3753 You can get much the same effect as the @code{jump} command by storing a
3754 new value into the register @code{$pc}. The difference is that this
3755 does not start the program running; it only changes the address where it
3756 @emph{will} run when it is continued. For example,
3763 causes the next @samp{continue} command or stepping command to execute at
3764 address 0x485, rather than at the address where the program stopped.
3767 The most common occasion to use the @samp{jump} command is to back up,
3768 perhaps with more breakpoints set, over a portion of a program that has
3773 @section Giving the Program a Signal
3776 @item signal @var{signalnum}
3778 Resume execution where the program stopped, but give it immediately the
3779 signal number @var{signalnum}.
3781 Alternatively, if @var{signalnum} is zero, continue execution without
3782 giving a signal. This is useful when the program stopped on account of
3783 a signal and would ordinary see the signal when resumed with the
3784 @samp{continue} command; @samp{signal 0} causes it to resume without a
3787 This command does not repeat when you press @key{RET} a second time
3788 after using it once.
3793 @section Returning from a Function
3797 @cindex returning from a function
3799 You can cancel execution of a function call with the @samp{return}
3803 This command abandons execution of a function. When you use
3804 @code{return}, _GDBN__ discards the selected stack frame (and all frames
3805 within it). You can think of this as making the discarded frame return
3806 prematurely. If you wish to specify a value to be returned, give that
3807 value as the argument to @code{return}.
3809 This pops the selected stack frame (@pxref{Selection}), and any other
3810 frames inside of it, leaving its caller as the innermost remaining
3811 frame. That frame becomes selected. The specified value is stored in
3812 the registers used for returning values of functions.
3814 The @samp{return} command does not resume execution; it leaves the
3815 program stopped in the state that would exist if the function had just
3816 returned. In contrast, the @samp{finish} command (@pxref{Stepping})
3817 resumes execution until the selected stack frame returns naturally.
3820 @section Calling your Program's Functions
3822 @cindex calling functions
3825 @item call @var{expr}
3826 Evaluate the expression @var{expr} without displaying @code{void}
3830 You can use this variant of the @samp{print} command if you want to
3831 execute a function from your program, but without cluttering the output
3832 with @code{void} returned values. The result is printed and saved in
3833 the value history, if it is not void.
3836 @chapter _GDBN__'s Files
3839 @section Commands to Specify Files
3840 @cindex core dump file
3841 @cindex symbol table
3842 _GDBN__ needs to know the file name of the program to be debugged, both in
3843 order to read its symbol table and in order to start the program. To
3844 debug a core dump of a previous run, _GDBN__ must be told the file name of
3847 The usual way to specify the executable and core dump file names is with
3848 the command arguments given when you start _GDBN__, as discussed in
3851 Occasionally it is necessary to change to a different file during a
3852 _GDBN__ session. Or you may run _GDBN__ and forget to specify the files you
3853 want to use. In these situations the _GDBN__ commands to specify new files
3857 @item file @var{filename}
3858 @cindex executable file
3860 Use @var{filename} as the program to be debugged. It is read for its
3861 symbols and for the contents of pure memory. It is also the program
3862 executed when you use the @samp{run} command. If you do not specify a
3863 directory and the file is not found in _GDBN__'s working directory,
3865 _GDBN__ will use the environment variable @code{PATH} as a list of
3866 directories to search, just as the shell does when looking for a program
3867 to run. You can change the value of this variable, for both _GDBN__ and
3868 your program, using the @code{path} command.
3870 @samp{file} with no argument makes _GDBN__ discard any information it
3871 has on both executable file and the symbol table.
3873 @item exec-file @var{filename}
3875 Specify that the program to be run (but not the symbol table) is found
3876 in @var{filename}. _GDBN__ will search the environment variable @code{PATH}
3877 if necessary to locate the program.
3879 @item symbol-file @var{filename}
3881 Read symbol table information from file @var{filename}. @code{PATH} is
3882 searched when necessary. Use the @samp{file} command to get both symbol
3883 table and program to run from the same file.
3885 @samp{symbol-file} with no argument clears out _GDBN__'s information on your
3886 program's symbol table.
3888 The @samp{symbol-file} command causes _GDBN__ to forget the contents of its
3889 convenience variables, the value history, and all breakpoints and
3890 auto-display expressions. This is because they may contain pointers to
3891 the internal data recording symbols and data types, which are part of
3892 the old symbol table data being discarded inside _GDBN__.
3894 @samp{symbol-file} will not repeat if you press @key{RET} again after
3897 On some kinds of object files, the @samp{symbol-file} command does not
3898 actually read the symbol table in full right away. Instead, it scans
3899 the symbol table quickly to find which source files and which symbols
3900 are present. The details are read later, one source file at a time,
3901 when they are needed.
3903 The purpose of this two-stage reading strategy is to make _GDBN__ start up
3904 faster. For the most part, it is invisible except for occasional pauses
3905 while the symbol table details for a particular source file are being
3906 read. (The @samp{set verbose} command can turn these pauses into
3907 messages if desired. @xref{Messages/Warnings}).
3909 When the symbol table is stored in COFF format, @samp{symbol-file} does
3910 read the symbol table data in full right away. We haven't implemented
3911 the two-stage strategy for COFF yet.
3913 When _GDBN__ is configured for a particular environment, it will
3914 understand debugging information in whatever format is the standard
3915 generated for that environment; you may use either a GNU compiler, or
3916 other compilers that adhere to the local conventions. Best results are
3917 usually obtained from GNU compilers; for example, using @code{_GCC__}
3918 you can generate debugging information for optimized code.
3920 @item core-file @var{filename}
3921 @itemx core @var{filename}
3924 Specify the whereabouts of a core dump file to be used as the ``contents
3925 of memory''. Traditionally, core files contain only some parts of the
3926 address space of the process that generated them; _GDBN__ can access the
3927 executable file itself for other parts.
3929 @samp{core-file} with no argument specifies that no core file is
3932 Note that the core file is ignored when your program is actually running
3933 under _GDBN__. So, if you have been running the program and you wish to
3934 debug a core file instead, you must kill the subprocess in which the
3935 program is running. To do this, use the @samp{kill} command
3936 (@pxref{Kill Process}).
3938 @item load @var{filename}
3941 Depending on what remote debugging facilities are configured into
3942 _GDBN__, the @samp{load} command may be available. Where it exists, it
3943 is meant to make @var{filename} (an executable) available for debugging
3944 on the remote system---by downloading, or dynamic linking, for example.
3945 @samp{load} also records @var{filename}'s symbol table in _GDBN__, like
3946 the @samp{add-syms} command.
3948 If @samp{load} is not available on your _GDBN__, attempting to execute
3949 it gets the error message ``@code{You can't do that when your target is
3954 On VxWorks, @samp{load} will dynamically link @var{filename} on the
3955 current target system as well as adding its symbols in _GDBN__.
3959 With the Nindy interface to an Intel 960 board, @samp{load} will
3960 download @var{filename} to the 960 as well as adding its symbols in
3964 @samp{load} will not repeat if you press @key{RET} again after using it.
3966 @item add-syms @var{filename} @var{address}
3968 @cindex dynamic linking
3969 The @samp{add-syms} command reads additional symbol table information
3970 from the file @var{filename}. You would use this command when that file
3971 has been dynamically loaded (by some other means) into the program that
3972 is running. @var{address} should be the memory address at which the
3973 file has been loaded; _GDBN__ cannot figure this out for itself.
3975 The symbol table of the file @var{filename} is added to the symbol table
3976 originally read with the @samp{symbol-file} command. You can use the
3977 @samp{add-syms} command any number of times; the new symbol data thus
3978 read keeps adding to the old. To discard all old symbol data instead,
3979 use the @samp{symbol-file} command.
3981 @samp{add-syms} will not repeat if you press @key{RET} after using it.
3987 @samp{info files} and @samp{info target} are synonymous; both print the
3988 current targets (@pxref{Targets}), including the names of the
3989 executable and core dump files currently in use by _GDBN__, and the files
3990 from which symbols were loaded.
3992 Beware: the similar command @samp{info targets} lists all possible
3993 targets rather than current ones.
3997 All file-specifying commands allow both absolute and relative file names
3998 as arguments. _GDBN__ always converts the file name to an absolute path
3999 name and remembers it that way.
4001 @kindex sharedlibrary
4003 @cindex shared libraries
4005 _GDBN__ supports the SunOS shared library format. Symbols from a shared
4006 library cannot be referenced before the shared library has been linked
4007 with the program. (That is to say, until after you type @samp{run} and
4008 the function @code{main} has been entered; or when examining core
4009 files.) Once the shared library has been linked in, you can use the
4013 @item sharedlibrary @var{regex}
4014 @itemx share @var{regex}
4015 Load shared object library symbols for files matching a UNIX regular
4019 @itemx sharedlibrary
4020 Load symbols for all shared libraries.
4023 @itemx info sharedlibrary
4024 @kindex info sharedlibrary
4026 Print the names of the shared libraries which are currently loaded.
4029 @samp{sharedlibrary} does not repeat automatically when you press
4030 @key{RET} after using it once.
4032 @node Symbol Errors,,,
4033 @section Errors Reading Symbol Files
4034 While a symbol file is being read, _GDBN__ will occasionally encounter
4035 problems, such as symbol types it does not recognize, or known bugs in
4036 compiler output. By default, it prints one message about each such
4037 type of problem, no matter how many times the problem occurs. You can
4038 ask it to print more messages, to see how many times the problems occur,
4039 or can shut the messages off entirely, with the @samp{set
4040 complaints} command (@xref{Messages/Warnings}).
4042 The messages currently printed, and their meanings, are:
4045 @item inner block not inside outer block in @var{symbol}
4047 The symbol information shows where symbol scopes begin and end
4048 (such as at the start of a function or a block of statements). This
4049 error indicates that an inner scope block is not fully contained
4050 in its outer scope blocks.
4052 _GDBN__ circumvents the problem by treating the inner block as if it had
4053 the same scope as the outer block. In the error message, @var{symbol}
4054 may be shown as ``@code{(don't know)}'' if the outer block is not a
4057 @item block at @var{address} out of order
4059 The symbol information for symbol scope blocks should occur in
4060 order of increasing addresses. This error indicates that it does not
4063 _GDBN__ does not circumvent this problem, and will have trouble locating
4064 symbols in the source file whose symbols being read. (You can often
4065 determine what source file is affected by specifying @samp{set verbose
4066 on}. @xref{Messages/Warnings}.)
4068 @item bad block start address patched
4070 The symbol information for a symbol scope block has a start address
4071 smaller than the address of the preceding source line. This is known
4072 to occur in the SunOS 4.1.1 (and earlier) C compiler.
4074 _GDBN__ circumvents the problem by treating the symbol scope block as
4075 starting on the previous source line.
4077 @c @item{encountered DBX-style class variable debugging information.
4078 @c You seem to have compiled your program with "g++ -g0" instead of "g++ -g".
4079 @c Therefore _GDBN__ will not know about your class variables}
4081 @c This error indicates that the symbol information produced for a C++
4082 @c program includes zero-size fields, which indicated static fields in
4083 @c a previous release of the G++ compiler. This message is probably
4086 @item bad string table offset in symbol @var{n}
4089 Symbol number @var{n} contains a pointer into the string table which is
4090 larger than the size of the string table.
4092 _GDBN__ circumvents the problem by considering the symbol to have the
4093 name @code{foo}, which may cause other problems if many symbols end up
4096 @item unknown symbol type @code{0x@var{NN}}
4098 The symbol information contains new data types that _GDBN__ does not yet
4099 know how to read. @code{0x@var{NN}} is the symbol type of the misunderstood
4100 information, in hexadecimal.
4102 _GDBN__ circumvents the error by ignoring this symbol information. This
4103 will usually allow the program to be debugged, though certain symbols
4104 will not be accessible. If you encounter such a problem and feel like
4105 debugging it, you can debug @code{_GDBP__} with itself, breakpoint on
4106 @samp{complain}, then go up to the function @samp{read_dbx_symtab} and
4107 examine @code{*bufp} to see the symbol.
4109 @c @item stub type has NULL name
4111 @c FIXME, Mike Tiemann needs to write about what this means.
4113 @item const/volatile indicator missing, got 'X'
4115 The symbol information for a C++ type is missing some information that
4116 the compiler should have output for it.
4118 @item C++ type mismatch between compiler and debugger
4120 The debugger could not parse a type specification output by the compiler
4121 for some C++ object.
4126 @chapter Specifying a Debugging Target
4127 @cindex debugging target
4129 A @dfn{target} is an interface between the debugger and a particular
4130 kind of file or process.
4132 Often, you will be able to run _GDBN__ in the same host environment as the
4133 program you are debugging; in that case, the debugging target can just be
4134 specified as a side effect of the @samp{file} or @samp{core} commands.
4135 When you need more flexibility---for example, running _GDBN__ on a
4136 physically separate host, controlling standalone systems over a
4137 serial port, or realtime systems over a TCP/IP connection---you can use
4138 the @samp{target} command.
4140 @node Active Targets,,,
4141 @section Active Targets
4142 @cindex stacking targets
4143 @cindex active targets
4144 @cindex multiple targets
4146 Targets are managed in three @dfn{strata} that correspond to different
4147 classes of target: processes, core files, and executable files. This
4148 allows you to (for example) start a process and inspect its activity
4149 without abandoning your work on a core file.
4151 More than one target can potentially respond to a request. In
4152 particular, when you access memory _GDBN__ will examine the three strata of
4153 targets until it finds a target that can handle that particular address.
4155 Strata are always examined in a fixed order: first a process if there is
4156 one, then a core file if there is one, and finally an executable file if
4157 there is one of those.
4159 When you specify a new target in a given stratum, it replaces any target
4160 previously in that stratum.
4162 To get rid of a target without replacing it, use the @samp{detach}
4163 command. The related command @samp{attach} provides you with a way of
4164 choosing a particular running process as a new target. @xref{Attach}.
4166 @node Target Commands,,,
4167 @section Commands for Managing Targets
4170 @item target @var{type} @var{parameters}
4171 Connects the _GDBN__ host environment to a target machine or process. A
4172 target is typically a protocol for talking to debugging facilities. You
4173 use the argument @var{type} to specify the type or protocol of the
4176 Further @var{parameters} are interpreted by the target protocol, but
4177 typically include things like device names or host names to connect
4178 with, process numbers, and baud rates. Executing
4184 (without any parameters) will issue a message about what
4185 parameters are required for that target type.
4187 The @samp{target} command will not repeat if you press @key{RET} again
4188 after executing the command.
4191 @kindex info targets
4192 Displays the names of all targets available. Beware: the similar
4193 command @samp{info target} displays targets currently in use rather than
4194 all available ones. @samp{info files} gives the same information as
4195 @samp{info target} (@pxref{Files}).
4198 Here are some common targets (available, or not, depending on the _GDBN__
4202 @item target exec @var{prog}
4204 An executable file. @samp{target exec @var{prog}} is the same as
4205 @samp{exec-file @var{prog}}.
4207 @item target core @var{filename}
4209 A core dump file. @samp{target core @var{filename}} is the same as
4210 @samp{core-file @var{filename}}.
4212 @item target remote @var{dev}
4213 @kindex target remote
4214 Remote serial target in _GDBN__-specific protocol. The argument @var{dev}
4215 specifies what serial device to use for the connection (e.g.
4216 @code{/dev/ttya}). @xref{Remote}.
4219 @item target amd-eb @var{dev} @var{speed} @var{PROG}
4220 @kindex target amd-eb
4222 Remote PC-resident AMD EB29K board, attached over serial lines.
4223 @var{dev} is the serial device, as for @samp{target remote};
4224 @samp{speed} allows you to specify the linespeed; and @var{PROG} is the
4225 name of the program to be debugged, as it appears to DOS on the PC.
4226 @xref{EB29K Remote}.
4230 @item target nindy @var{devicename}
4231 @kindex target nindy
4232 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
4233 the name of the serial device to use for the connection, e.g.
4234 @samp{/dev/ttya}. @xref{i960-Nindy Remote}.
4238 @item target vxworks @var{machinename}
4239 @kindex target vxworks
4240 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
4241 is the target system's machine name or IP address.
4242 @xref{VxWorks Remote}.
4247 Different targets are available on different configurations of _GDBN__; your
4248 configuration may have more or fewer targets.
4252 @section Remote Debugging
4253 @cindex remote debugging
4255 If you are trying to debug a program running on a machine that can't run
4256 _GDBN__ in the usual way, it is often useful to use remote debugging. For
4257 example, you might be debugging an operating system kernel, or debugging
4258 a small system which does not have a general purpose operating system
4259 powerful enough to run a full-featured debugger.
4261 Some configurations of _GDBN__ have special serial or TCP/IP interfaces
4262 to make this work with particular debugging targets. In addition,
4263 _GDBN__ comes with a generic serial protocol (specific to _GDBN__, but
4264 not specific to any particular target system) which you can adapt.
4266 To use the _GDBN__ remote serial protocol, the program to be debugged on
4267 the remote machine needs to contain a debugging device driver which
4268 talks to _GDBN__ over the serial line. Several sample remote debugging
4269 drivers are distributed with _GDBN__; see the @file{README} file in the
4270 _GDBN__ distribution for more information.
4272 For details of this communication protocol, see the comments in the
4273 _GDBN__ source file @file{remote.c}.
4275 To start remote debugging, first run _GDBN__ and specify as an executable file
4276 the program that is running in the remote machine. This tells _GDBN__ how
4277 to find the program's symbols and the contents of its pure text. Then
4278 establish communication using the @samp{target remote} command with a device
4279 name as an argument. For example:
4282 target remote /dev/ttyb
4286 if the serial line is connected to the device named @file{/dev/ttyb}. This
4287 will stop the remote machine if it is not already stopped.
4289 Now you can use all the usual commands to examine and change data and to
4290 step and continue the remote program.
4292 To resume the remote program and stop debugging it, use the @samp{detach}
4295 Other remote targets may be available in your
4296 configuration of _GDBN__; use @samp{info targets} to list them.
4298 @node Controlling _GDBN__,,,
4299 @chapter Controlling _GDBN__
4301 You can alter many aspects of _GDBN__'s interaction with you by using
4302 the @samp{set} command. For commands controlling how _GDBN__ displays
4303 data, @pxref{Print Settings}; other settings are described here.
4308 _GDBN__ indicates its readiness to read a command by printing a string
4309 called the @dfn{prompt}. This string is normally @samp{(_GDBP__)}. You
4310 can change the prompt string with the @samp{set prompt} command. For
4311 instance, when debugging _GDBN__ with _GDBN__, it is useful to change
4312 the prompt in one of the _GDBN__<>s so that you tell which one you are
4316 @item set prompt @var{newprompt}
4318 Directs _GDBN__ to use @var{newprompt} as its prompt string henceforth.
4321 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
4325 @section Command Editing
4327 @cindex command line editing
4328 _GDBN__ reads its input commands via the @dfn{readline} interface. This
4329 GNU library provides consistent behavior for programs which provide a
4330 command line interface to the user. Advantages are @samp{emacs}-style
4331 or @samp{vi}-style inline editing of commands, @samp{csh}-like history
4332 substitution, and a storage and recall of command history across
4335 You may control the behavior of command line editing in _GDBN__ with the
4336 command @samp{set}. You may check the status of any of these settings
4337 with the command @samp{show}.
4343 @itemx set editing on
4344 Enable command line editing (enabled by default).
4346 @item set editing off
4347 Disable command line editing.
4349 @kindex show editing
4351 Show whether command line editing is enabled.
4354 @section Command History
4355 @cindex history substitution
4356 @cindex history file
4357 @kindex set history file
4358 @item set history file @var{filename}
4359 Set the name of the _GDBN__ command history file to @var{filename}. This is
4360 the file from which _GDBN__ will read an initial command history
4361 list or to which it will write this list when it exits. This list is
4362 accessed through history expansion or through the history
4363 command editing characters listed below. This file defaults to the
4364 value of the environment variable @code{GDBHISTFILE}, or to
4365 @file{./.gdb_history} if this variable is not set.
4367 @cindex history write
4368 @kindex set history write
4369 @item set history write
4370 @itemx set history write on
4371 Record command history in a file, whose name may be specified with the
4372 @samp{set history file} command. By default, this option is disabled.
4374 @item set history write off
4375 Stop recording command history in a file.
4377 @cindex history size
4378 @kindex set history size
4379 @item set history size @var{size}
4380 Set the number of commands which _GDBN__ will keep in its history list.
4381 This defaults to the value of the environment variable
4382 @code{HISTSIZE}, or to 256 if this variable is not set.
4385 @cindex history expansion
4386 History expansion assigns special meaning to the character @samp{!}.
4388 (@xref{Event Designators}.)
4390 Since @samp{!} is also the logical not operator in C, history expansion
4391 is off by default. If you decide to enable history expansion with the
4392 @samp{set history expansion on} command, you may sometimes need to
4393 follow @samp{!} (when it is used as logical not, in an expression) with
4394 a space or a tab to prevent it from being expanded. The readline
4395 history facilities will not attempt substitution on the strings
4396 @samp{!=} and @samp{!(}, even when history expansion is enabled.
4398 The commands to control history expansion are:
4402 @kindex set history expansion
4403 @item set history expansion on
4404 @itemx set history expansion
4405 Enable history expansion. History expansion is off by default.
4407 @item set history expansion off
4408 Disable history expansion.
4410 The readline code comes with more complete documentation of
4411 editing and history expansion features. Users unfamiliar with @samp{emacs}
4412 or @samp{vi} may wish to read it.
4414 @xref{Command Line Editing}.
4418 @kindex show history
4420 @itemx show history file
4421 @itemx show history write
4422 @itemx show history size
4423 @itemx show history expansion
4424 These commands display the state of the _GDBN__ history parameters.
4425 @samp{show history} by itself displays all four states.
4431 @kindex info editing
4433 Display the last ten commands in the command history.
4435 @item info editing @var{n}
4436 Print ten commands centered on command number @var{n}.
4438 @item info editing +
4439 Print ten commands just after the commands last printed.
4443 @node Screen Size,,,
4444 @section Screen Size
4445 @cindex size of screen
4446 @cindex pauses in output
4447 Certain commands to _GDBN__ may produce large amounts of information
4448 output to the screen. To help you read all of it, _GDBN__ pauses and
4449 asks you for input at the end of each page of output. Type @key{RET}
4450 when you want to continue the output. _GDBN__ also uses the screen
4451 width setting to determine when to wrap lines of output. Depending on
4452 what is being printed, it tries to break the line at a readable place,
4453 rather than simply letting it overflow onto the following line.
4455 Normally _GDBN__ knows the size of the screen from the termcap data base
4456 together with the value of the @code{TERM} environment variable and the
4457 @code{stty rows} and @code{stty cols} settings. If this is not correct,
4458 you can override it with the @samp{set screen-height} and @samp{set
4459 screen-width} commands:
4462 @item set screen-height @var{lpp}
4463 @itemx show screen-height
4464 @itemx set screen-width @var{cpl}
4465 @itemx show screen-width
4466 @kindex set screen-height
4467 @kindex set screen-width
4468 @kindex show screen-width
4469 @kindex show screen-height
4470 These @samp{set} commands specify a screen height of @var{lpp} lines and
4471 a screen width of @var{cpl} characters. The associated @samp{show}
4472 commands display the current settings.
4474 If you specify a height of zero lines, _GDBN__ will not pause during output
4475 no matter how long the output is. This is useful if output is to a file
4476 or to an editor buffer.
4481 @cindex number representation
4482 @cindex entering numbers
4483 You can always enter numbers in octal, decimal, or hexadecimal in _GDBN__ by
4484 the usual conventions: octal numbers begin with @samp{0}, decimal
4485 numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}.
4486 Numbers that begin with none of these are, by default, entered in base
4487 10; likewise, the default display for numbers---when no particular
4488 format is specified---is base 10. You can change the default base for
4489 both input and output with the @samp{set radix} command.
4493 @item set radix @var{base}
4494 Set the default base for numeric input and display. Supported choices
4495 for @var{base} are decimal 8, 10, 16. @var{base} must itself be
4496 specified either unambiguously or using the current default radix; for
4506 will set the base to decimal. On the other hand, @samp{set radix 10}
4507 will leave the radix unchanged no matter what it was.
4511 Display the current default base for numeric input and display.
4515 @node Messages/Warnings,,,
4516 @section Optional Warnings and Messages
4517 By default, _GDBN__ is silent about its inner workings. If you are running
4518 on a slow machine, you may want to use the @samp{set verbose} command.
4519 It will make _GDBN__ tell you when it does a lengthy internal operation, so
4520 you won't think it has crashed.
4522 Currently, the messages controlled by @samp{set verbose} are those which
4523 announce that the symbol table for a source file is being read
4524 (@pxref{Files}, in the description of the command
4525 @samp{symbol-file}).
4526 @c The following is the right way to do it, but emacs 18.55 doesn't support
4527 @c @ref, and neither the emacs lisp manual version of texinfmt or makeinfo
4530 see @samp{symbol-file} in @ref{Files}).
4535 @item set verbose on
4536 Enables _GDBN__'s output of certain informational messages.
4538 @item set verbose off
4539 Disables _GDBN__'s output of certain informational messages.
4541 @kindex show verbose
4543 Displays whether @samp{set verbose} is on or off.
4546 By default, if _GDBN__ encounters bugs in the symbol table of an object file,
4547 it prints a single message about each type of problem it finds, then
4548 shuts up. You can suppress these messages, or allow more than one such
4549 message to be printed if you want to see how frequent the problems are.
4550 @xref{Symbol Errors}.
4553 @kindex set complaints
4554 @item set complaints @var{limit}
4555 Permits _GDBN__ to output @var{limit} complaints about each type of unusual
4556 symbols before becoming silent about the problem. Set @var{limit} to
4557 zero to suppress all complaints; set it to a large number to prevent
4558 complaints from being suppressed.
4560 @kindex show complaints
4561 @item show complaints
4562 Displays how many symbol complaints _GDBN__ is permitted to produce.
4565 By default, _GDBN__ is cautious, and asks what sometimes seem to be a lot of
4566 stupid questions. For example, if you try to run a program which is
4571 The program being debugged has been started already.
4572 Start it from the beginning? (y or n)
4575 If you're willing to unflinchingly face the consequences of your own
4576 commands, you can disable this ``feature'':
4581 @cindex stupid questions
4582 @item set caution off
4583 Disables cautious questions.
4585 @item set caution on
4586 Enables cautious questions (the default).
4589 @kindex show caution
4590 Displays state of cautious questions.
4594 @chapter Canned Sequences of Commands
4596 Aside from breakpoint commands (@pxref{Break Commands}),_GDBN__ provides two
4597 ways to store sequences of commands for execution as a unit:
4598 user-defined commands and command files.
4601 @section User-Defined Commands
4603 @cindex user-defined command
4604 A @dfn{user-defined command} is a sequence of _GDBN__ commands to which you
4605 assign a new name as a command. This is done with the @samp{define}
4609 @item define @var{commandname}
4611 Define a command named @var{commandname}. If there is already a command
4612 by that name, you are asked to confirm that you want to redefine it.
4614 The definition of the command is made up of other _GDBN__ command lines,
4615 which are given following the @samp{define} command. The end of these
4616 commands is marked by a line containing @samp{end}.
4618 @item document @var{commandname}
4620 Give documentation to the user-defined command @var{commandname}. The
4621 command @var{commandname} must already be defined. This command reads
4622 lines of documentation just as @samp{define} reads the lines of the
4623 command definition, ending with @samp{end}. After the @samp{document}
4624 command is finished, @samp{help} on command @var{commandname} will print
4625 the documentation you have specified.
4627 You may use the @samp{document} command again to change the
4628 documentation of a command. Redefining the command with @samp{define}
4629 does not change the documentation.
4632 User-defined commands do not take arguments. When they are executed, the
4633 commands of the definition are not printed. An error in any command
4634 stops execution of the user-defined command.
4636 Commands that would ask for confirmation if used interactively proceed
4637 without asking when used inside a user-defined command. Many _GDBN__ commands
4638 that normally print messages to say what they are doing omit the messages
4639 when used in a user-defined command.
4641 @node Command Files,,,
4642 @section Command Files
4644 @cindex command files
4645 A command file for _GDBN__ is a file of lines that are _GDBN__ commands. Comments
4646 (lines starting with @samp{#}) may also be included. An empty line in a
4647 command file does nothing; it does not mean to repeat the last command, as
4648 it would from the terminal.
4651 @cindex @file{_GDBINIT__}
4652 When you start _GDBN__, it automatically executes commands from its
4653 @dfn{init files}. These are files named @file{_GDBINIT__}. _GDBN__
4654 reads the init file (if any) in your home directory and then the init
4655 file (if any) in the current working directory. (The init files are not
4656 executed if the @samp{-nx} option is given.) You can also request the
4657 execution of a command file with the @samp{source} command:
4660 @item source @var{filename}
4662 Execute the command file @var{filename}.
4665 The lines in a command file are executed sequentially. They are not
4666 printed as they are executed. An error in any command terminates execution
4667 of the command file.
4669 Commands that would ask for confirmation if used interactively proceed
4670 without asking when used in a command file. Many _GDBN__ commands that
4671 normally print messages to say what they are doing omit the messages
4672 when called from command files.
4675 @section Commands for Controlled Output
4677 During the execution of a command file or a user-defined command, normal
4678 _GDBN__ output is suppressed; the only output that appears is what is
4679 explicitly printed by the commands in the definition. This section
4680 describes three commands useful for generating exactly the output you
4684 @item echo @var{text}
4686 @c I don't consider backslash-space a standard C escape sequence
4687 @c because it's not in ANSI.
4688 Print @var{text}. Nonprinting characters can be included in @var{text}
4689 using C escape sequences, such as @samp{\n} to print a newline. @b{No
4690 newline will be printed unless you specify one.} In addition to the
4691 standard C escape sequences a backslash followed by a space stands for a
4692 space. This is useful for outputting a string with spaces at the
4693 beginning or the end, since leading and trailing spaces are otherwise
4694 trimmed from all arguments. Thus, to print @samp{@ and foo =@ }, use the
4695 command @samp{echo \@ and foo = \@ }.
4696 @c FIXME: verify hard copy actually issues enspaces for '@ '! Will this
4699 A backslash at the end of @var{text} can be used, as in C, to continue
4700 the command onto subsequent lines. For example,
4703 echo This is some text\n\
4704 which is continued\n\
4705 onto several lines.\n
4708 produces the same output as
4711 echo This is some text\n
4712 echo which is continued\n
4713 echo onto several lines.\n
4716 @item output @var{expression}
4718 Print the value of @var{expression} and nothing but that value: no
4719 newlines, no @samp{$@var{nn} = }. The value is not entered in the
4720 value history either. @xref{Expressions} for more information on
4723 @item output/@var{fmt} @var{expression}
4724 Print the value of @var{expression} in format @var{fmt}. You can use
4725 the same formats as for @samp{print}; @pxref{Output formats}, for more
4728 @item printf @var{string}, @var{expressions}@dots{}
4730 Print the values of the @var{expressions} under the control of
4731 @var{string}. The @var{expressions} are separated by commas and may
4732 be either numbers or pointers. Their values are printed as specified
4733 by @var{string}, exactly as if the program were to execute
4736 printf (@var{string}, @var{expressions}@dots{});
4739 For example, you can print two values in hex like this:
4742 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
4745 The only backslash-escape sequences that you can use in the format
4746 string are the simple ones that consist of backslash followed by a
4751 @chapter Using _GDBN__ under GNU Emacs
4754 A special interface allows you to use GNU Emacs to view (and
4755 edit) the source files for the program you are debugging with
4758 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
4759 executable file you want to debug as an argument. This command starts
4760 _GDBN__ as a subprocess of Emacs, with input and output through a newly
4761 created Emacs buffer.
4763 Using _GDBN__ under Emacs is just like using _GDBN__ normally except for two
4768 All ``terminal'' input and output goes through the Emacs buffer.
4771 This applies both to _GDBN__ commands and their output, and to the input
4772 and output done by the program you are debugging.
4774 This is useful because it means that you can copy the text of previous
4775 commands and input them again; you can even use parts of the output
4778 All the facilities of Emacs' Shell mode are available for this purpose.
4782 _GDBN__ displays source code through Emacs.
4785 Each time _GDBN__ displays a stack frame, Emacs automatically finds the
4786 source file for that frame and puts an arrow (_0__@samp{=>}_1__) at the
4787 left margin of the current line. Emacs uses a separate buffer for
4788 source display, and splits the window to show both your _GDBN__ session
4791 Explicit _GDBN__ @samp{list} or search commands still produce output as
4792 usual, but you probably will have no reason to use them.
4795 @emph{Warning:} If the directory where your program resides is not your
4796 current directory, it can be easy to confuse Emacs about the location of
4797 the source files, in which case the auxiliary display buffer will not
4798 appear to show your source. _GDBN__ can find programs by searching your
4799 environment's @samp{PATH} variable, so the _GDBN__ input and output
4800 session will proceed normally; but Emacs doesn't get enough information
4801 back from _GDBN__ to locate the source files in this situation. To
4802 avoid this problem, either start _GDBN__ mode from the directory where
4803 your program resides, or specify a full path name when prompted for the
4804 @kbd{M-x gdb} argument.
4806 A similar confusion can result if you use the _GDBN__ @samp{file} command to
4807 switch to debugging a program in some other location, from an existing
4808 _GDBN__ buffer in Emacs.
4811 By default, @kbd{M-x gdb} calls the program called ``@code{gdb}''. If
4812 you need to call _GDBN__ by a different name (for example, if you keep
4813 several configurations around, with different names) you can set the
4814 Emacs variable @code{gdb-command-name}; for example,
4816 (setq gdb-command-name "mygdb")
4819 (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or
4820 in your @samp{.emacs} file) will make Emacs call the program named
4821 ``@code{mygdb}'' instead.
4823 In the _GDBN__ I/O buffer, you can use these special Emacs commands in
4824 addition to the standard Shell mode commands:
4828 Describe the features of Emacs' _GDBN__ Mode.
4831 Execute to another source line, like the _GDBN__ @samp{step} command; also
4832 update the display window to show the current file and location.
4835 Execute to next source line in this function, skipping all function
4836 calls, like the _GDBN__ @samp{next} command. Then update the display window
4837 to show the current file and location.
4840 Execute one instruction, like the _GDBN__ @samp{stepi} command; update
4841 display window accordingly.
4844 Execute to next instruction, using the _GDBN__ @samp{nexti} command; update
4845 display window accordingly.
4848 Execute until exit from the selected stack frame, like the _GDBN__
4849 @samp{finish} command.
4852 @c C-c C-p in emacs 19
4853 Continue execution of the program, like the _GDBN__ @samp{continue} command.
4856 @c C-c C-u in emacs 19
4857 Go up the number of frames indicated by the numeric argument
4858 (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}),
4859 like the _GDBN__ @samp{up} command.@refill
4862 @c C-c C-d in emacs 19
4863 Go down the number of frames indicated by the numeric argument, like the
4864 _GDBN__ @samp{down} command.
4867 Read the number where the cursor is positioned, and insert it at the end
4868 of the _GDBN__ I/O buffer. For example, if you wish to disassemble code
4869 around an address that was displayed earlier, type @kbd{disassemble};
4870 then move the cursor to the address display, and pick up the
4871 argument for @samp{disassemble} by typing @kbd{C-x &}.
4873 You can customize this further on the fly by defining elements of the list
4874 @samp{gdb-print-command}; once it is defined, you can format or
4875 otherwise process numbers picked up by @kbd{C-x &} before they are
4876 inserted. A numeric argument to @kbd{C-x &} will both flag that you
4877 wish special formatting, and act as an index to pick an element of the
4878 list. If the list element is a string, the number to be inserted is
4879 formatted using the Emacs function @samp{format}; otherwise the number
4880 is passed as an argument to the corresponding list element.
4882 @item M-x gdb-display-frame
4883 Explicitly request display of the source code surrounding the current
4884 frame location, in another window. _GDBN__ does this display automatically;
4885 but if, for example, you accidentally kill the buffer where it is
4886 displayed, this command is a way of getting it back.
4889 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
4890 tells _GDBN__ to set a breakpoint on the source line point is on.
4892 The source files displayed in Emacs are in ordinary Emacs buffers
4893 which are visiting the source files in the usual way. You can edit
4894 the files with these buffers if you wish; but keep in mind that _GDBN__
4895 communicates with Emacs in terms of line numbers. If you add or
4896 delete lines from the text, the line numbers that _GDBN__ knows will cease
4897 to correspond properly to the code.
4899 @c The following dropped because Epoch is nonstandard. Reactivate
4900 @c if/when v19 does something similar. ---pesch@cygnus.com 19dec1990
4902 @kindex emacs epoch environment
4906 Version 18 of Emacs has a built-in window system called the @samp{epoch}
4907 environment. Users of this environment can use a new command,
4908 @samp{inspect} which performs identically to @samp{print} except that
4909 each value is printed in its own window.
4912 @node _GDBN__ Bugs,,,
4913 @c node-name, next, previous, up
4914 @chapter Reporting Bugs in _GDBN__
4915 @cindex Bugs in _GDBN__
4916 @cindex Reporting Bugs in _GDBN__
4918 Your bug reports play an essential role in making _GDBN__ reliable.
4920 Reporting a bug may help you by bringing a solution to your problem, or it
4921 may not. But in any case the principal function of a bug report is to help
4922 the entire community by making the next version of _GDBN__ work better. Bug
4923 reports are your contribution to the maintenance of _GDBN__.
4925 In order for a bug report to serve its purpose, you must include the
4926 information that enables us to fix the bug.
4928 @node Bug Criteria,,,
4929 @section Have You Found a Bug?
4930 @cindex Bug Criteria
4932 If you are not sure whether you have found a bug, here are some guidelines:
4936 @cindex Fatal Signal
4938 If the debugger gets a fatal signal, for any input whatever, that is a
4939 _GDBN__ bug. Reliable debuggers never crash.
4942 @cindex error on Valid Input
4943 If _GDBN__ produces an error message for valid input, that is a bug.
4946 @cindex Invalid Input
4947 If _GDBN__ does not produce an error message for invalid input,
4948 that is a bug. However, you should note that your idea of
4949 ``invalid input'' might be our idea of ``an extension'' or ``support
4950 for traditional practice''.
4953 If you are an experienced user of debugging tools, your suggestions
4954 for improvement of _GDBN__ are welcome in any case.
4957 @node Bug Reporting,,,
4958 @section How to Report Bugs
4960 @cindex Compiler Bugs, Reporting
4962 A number of companies and individuals offer support for GNU products.
4963 If you obtained _GDBN__ from a support organization, we recommend you
4964 contact that organization first.
4966 Contact information for many support companies and individuals is
4967 available in the file @samp{etc/SERVICE} in the GNU Emacs distribution.
4969 In any event, we also recommend that you send bug reports for _GDBN__ to one
4973 bug-gdb@@prep.ai.mit.edu
4974 @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb
4977 @strong{Do not send bug reports to @samp{info-gdb}, or to
4978 @samp{help-gdb}, or to any newsgroups.} Most users of _GDBN__ do not want to
4979 receive bug reports. Those that do, have arranged to receive @samp{bug-gdb}.
4981 The mailing list @samp{bug-gdb} has a newsgroup which serves as a
4982 repeater. The mailing list and the newsgroup carry exactly the same
4983 messages. Often people think of posting bug reports to the newsgroup
4984 instead of mailing them. This appears to work, but it has one problem
4985 which can be crucial: a newsgroup posting does not contain a mail path
4986 back to the sender. Thus, if we need to ask for more information, we
4987 may be unable to reach you. For this reason, it is better to send bug
4988 reports to the mailing list.
4990 As a last resort, send bug reports on paper to:
4998 The fundamental principle of reporting bugs usefully is this:
4999 @strong{report all the facts}. If you are not sure whether to state a
5000 fact or leave it out, state it!
5002 Often people omit facts because they think they know what causes the
5003 problem and assume that some details don't matter. Thus, you might
5004 assume that the name of the variable you use in an example does not matter.
5005 Well, probably it doesn't, but one cannot be sure. Perhaps the bug is a
5006 stray memory reference which happens to fetch from the location where that
5007 name is stored in memory; perhaps, if the name were different, the contents
5008 of that location would fool the debugger into doing the right thing despite
5009 the bug. Play it safe and give a specific, complete example. That is the
5010 easiest thing for you to do, and the most helpful.
5012 Keep in mind that the purpose of a bug report is to enable us to fix
5013 the bug if it is new to us. It isn't as important what happens if
5014 the bug is already known. Therefore, always write your bug reports on
5015 the assumption that the bug has not been reported previously.
5017 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5018 bell?'' Those bug reports are useless, and we urge everyone to
5019 @emph{refuse to respond to them} except to chide the sender to report
5022 To enable us to fix the bug, you should include all these things:
5026 The version of _GDBN__. _GDBN__ announces it if you start with no
5027 arguments; you can also print it at any time using @samp{show version}.
5029 Without this, we won't know whether there is any point in looking for
5030 the bug in the current version of _GDBN__.
5033 A complete input script, and all necessary source files, that will
5037 What compiler (and its version) was used to compile _GDBN__---e.g.
5041 The command arguments you gave the compiler to compile your example and
5042 observe the bug. For example, did you use @samp{-O}? To guarantee
5043 you won't omit something important, list them all.
5045 If we were to try to guess the arguments, we would probably guess wrong
5046 and then we would not encounter the bug.
5049 The type of machine you are using, and the operating system name and
5053 A description of what behavior you observe that you believe is
5054 incorrect. For example, ``It gets a fatal signal.''
5056 Of course, if the bug is that _GDBN__ gets a fatal signal, then we
5057 will certainly notice it. But if the bug is incorrect output, we might
5058 not notice unless it is glaringly wrong.
5060 Even if the problem you experience is a fatal signal, you should still
5061 say so explicitly. Suppose something strange is going on, such as,
5062 your copy of _GDBN__ is out of synch, or you have encountered a
5063 bug in the C library on your system. (This has happened!) Your copy
5064 might crash and ours would not. If you told us to expect a crash,
5065 then when ours fails to crash, we would know that the bug was not
5066 happening for us. If you had not told us to expect a crash, then we
5067 would not be able to draw any conclusion from our observations.
5070 If you wish to suggest changes to the _GDBN__ source, send us context
5071 diffs. If you even discuss something in the _GDBN__ source, refer to
5072 it by context, not by line number.
5074 The line numbers in our development sources won't match those in your
5075 sources. Your line numbers would convey no useful information to us.
5079 Here are some things that are not necessary:
5083 A description of the envelope of the bug.
5085 Often people who encounter a bug spend a lot of time investigating
5086 which changes to the input file will make the bug go away and which
5087 changes will not affect it.
5089 This is often time consuming and not very useful, because the way we
5090 will find the bug is by running a single example under the debugger
5091 with breakpoints, not by pure deduction from a series of examples.
5092 We recommend that you save your time for something else.
5094 Of course, if you can find a simpler example to report @emph{instead}
5095 of the original one, that is a convenience for us. Errors in the
5096 output will be easier to spot, running under the debugger will take
5099 However, simplification is not vital; if you don't want to do this,
5100 report the bug anyway and send us the entire test case you used.
5103 A patch for the bug.
5105 A patch for the bug does help us if it is a good one. But don't omit
5106 the necessary information, such as the test case, on the assumption that
5107 a patch is all we need. We might see problems with your patch and decide
5108 to fix the problem another way, or we might not understand it at all.
5110 Sometimes with a program as complicated as _GDBN__ it is very hard to
5111 construct an example that will make the program follow a certain path
5112 through the code. If you don't send us the example, we won't be able
5113 to construct one, so we won't be able to verify that the bug is fixed.
5115 And if we can't understand what bug you are trying to fix, or why your
5116 patch should be an improvement, we won't install it. A test case will
5117 help us to understand.
5120 A guess about what the bug is or what it depends on.
5122 Such guesses are usually wrong. Even we can't guess right about such
5123 things without first using the debugger to find the facts.
5127 @include readline/inc-readline.texinfo
5128 @include readline/inc-history.texinfo
5131 @node Installing _GDBN__,,,
5132 @appendix Installing _GDBN__
5133 @cindex configuring _GDBN__
5134 @cindex installation
5136 The script @samp{config.gdb} automates the process of preparing _GDBN__ for
5137 installation; you can then use @samp{make} to actually build it. For
5146 is all that's required to install _GDBN__ on a Sun 3 running SunOS 4.
5150 @item config.gdb @var{machine}
5151 @itemx config.gdb -srcdir=@var{dir} @var{machine}
5152 This is the most usual way of configuring _GDBN__; to debug programs running
5153 on the same machine as _GDBN__ itself. If you wish to build the _GDBN__ binaries
5154 in a completely different directory from the sources, specify a path to
5155 the source directory using the @samp{-srcdir} option.
5157 @item config.gdb -host
5158 @cindex host environments
5159 Display a list of supported host environments for _GDBN__.
5161 @item config.gdb @var{host} @var{target}
5162 @itemx config.gdb -srcdir=@var{dir} @var{host} @var{target}
5163 @cindex cross-debugging
5164 _GDBN__ can also be used as a cross-debugger, running on a machine of one
5165 type while debugging a program running on a machine of another type.
5166 You configure it this way by specifying first the @var{host}, then the
5167 @var{target} environment on the @code{config.gdb} argument list; the
5168 @var{host} is where _GDBN__ runs, and the @var{target} is where your program
5169 runs. @xref{Remote}. Again, you can use @samp{-srcdir} to specify a
5170 path to the _GDBN__ source.
5172 @item config.gdb -target
5173 @cindex target environments
5174 Display a list of supported target environments for _GDBN__.
5178 @unnumbered GNU GENERAL PUBLIC LICENSE
5179 @center Version 1, February 1989
5182 Copyright @copyright{} 1989 Free Software Foundation, Inc.
5183 675 Mass Ave, Cambridge, MA 02139, USA
5185 Everyone is permitted to copy and distribute verbatim copies
5186 of this license document, but changing it is not allowed.
5189 @unnumberedsec Preamble
5191 The license agreements of most software companies try to keep users
5192 at the mercy of those companies. By contrast, our General Public
5193 License is intended to guarantee your freedom to share and change free
5194 software---to make sure the software is free for all its users. The
5195 General Public License applies to the Free Software Foundation's
5196 software and to any other program whose authors commit to using it.
5197 You can use it for your programs, too.
5199 When we speak of free software, we are referring to freedom, not
5200 price. Specifically, the General Public License is designed to make
5201 sure that you have the freedom to give away or sell copies of free
5202 software, that you receive source code or can get it if you want it,
5203 that you can change the software or use pieces of it in new free
5204 programs; and that you know you can do these things.
5206 To protect your rights, we need to make restrictions that forbid
5207 anyone to deny you these rights or to ask you to surrender the rights.
5208 These restrictions translate to certain responsibilities for you if you
5209 distribute copies of the software, or if you modify it.
5211 For example, if you distribute copies of a such a program, whether
5212 gratis or for a fee, you must give the recipients all the rights that
5213 you have. You must make sure that they, too, receive or can get the
5214 source code. And you must tell them their rights.
5216 We protect your rights with two steps: (1) copyright the software, and
5217 (2) offer you this license which gives you legal permission to copy,
5218 distribute and/or modify the software.
5220 Also, for each author's protection and ours, we want to make certain
5221 that everyone understands that there is no warranty for this free
5222 software. If the software is modified by someone else and passed on, we
5223 want its recipients to know that what they have is not the original, so
5224 that any problems introduced by others will not reflect on the original
5225 authors' reputations.
5227 The precise terms and conditions for copying, distribution and
5228 modification follow.
5231 @unnumberedsec TERMS AND CONDITIONS
5234 @center TERMS AND CONDITIONS
5239 This License Agreement applies to any program or other work which
5240 contains a notice placed by the copyright holder saying it may be
5241 distributed under the terms of this General Public License. The
5242 ``Program'', below, refers to any such program or work, and a ``work based
5243 on the Program'' means either the Program or any work containing the
5244 Program or a portion of it, either verbatim or with modifications. Each
5245 licensee is addressed as ``you''.
5248 You may copy and distribute verbatim copies of the Program's source
5249 code as you receive it, in any medium, provided that you conspicuously and
5250 appropriately publish on each copy an appropriate copyright notice and
5251 disclaimer of warranty; keep intact all the notices that refer to this
5252 General Public License and to the absence of any warranty; and give any
5253 other recipients of the Program a copy of this General Public License
5254 along with the Program. You may charge a fee for the physical act of
5255 transferring a copy.
5258 You may modify your copy or copies of the Program or any portion of
5259 it, and copy and distribute such modifications under the terms of Paragraph
5260 1 above, provided that you also do the following:
5264 cause the modified files to carry prominent notices stating that
5265 you changed the files and the date of any change; and
5268 cause the whole of any work that you distribute or publish, that
5269 in whole or in part contains the Program or any part thereof, either
5270 with or without modifications, to be licensed at no charge to all
5271 third parties under the terms of this General Public License (except
5272 that you may choose to grant warranty protection to some or all
5273 third parties, at your option).
5276 If the modified program normally reads commands interactively when
5277 run, you must cause it, when started running for such interactive use
5278 in the simplest and most usual way, to print or display an
5279 announcement including an appropriate copyright notice and a notice
5280 that there is no warranty (or else, saying that you provide a
5281 warranty) and that users may redistribute the program under these
5282 conditions, and telling the user how to view a copy of this General
5286 You may charge a fee for the physical act of transferring a
5287 copy, and you may at your option offer warranty protection in
5291 Mere aggregation of another independent work with the Program (or its
5292 derivative) on a volume of a storage or distribution medium does not bring
5293 the other work under the scope of these terms.
5296 You may copy and distribute the Program (or a portion or derivative of
5297 it, under Paragraph 2) in object code or executable form under the terms of
5298 Paragraphs 1 and 2 above provided that you also do one of the following:
5302 accompany it with the complete corresponding machine-readable
5303 source code, which must be distributed under the terms of
5304 Paragraphs 1 and 2 above; or,
5307 accompany it with a written offer, valid for at least three
5308 years, to give any third party free (except for a nominal charge
5309 for the cost of distribution) a complete machine-readable copy of the
5310 corresponding source code, to be distributed under the terms of
5311 Paragraphs 1 and 2 above; or,
5314 accompany it with the information you received as to where the
5315 corresponding source code may be obtained. (This alternative is
5316 allowed only for noncommercial distribution and only if you
5317 received the program in object code or executable form alone.)
5320 Source code for a work means the preferred form of the work for making
5321 modifications to it. For an executable file, complete source code means
5322 all the source code for all modules it contains; but, as a special
5323 exception, it need not include source code for modules which are standard
5324 libraries that accompany the operating system on which the executable
5325 file runs, or for standard header files or definitions files that
5326 accompany that operating system.
5329 You may not copy, modify, sublicense, distribute or transfer the
5330 Program except as expressly provided under this General Public License.
5331 Any attempt otherwise to copy, modify, sublicense, distribute or transfer
5332 the Program is void, and will automatically terminate your rights to use
5333 the Program under this License. However, parties who have received
5334 copies, or rights to use copies, from you under this General Public
5335 License will not have their licenses terminated so long as such parties
5336 remain in full compliance.
5339 By copying, distributing or modifying the Program (or any work based
5340 on the Program) you indicate your acceptance of this license to do so,
5341 and all its terms and conditions.
5344 Each time you redistribute the Program (or any work based on the
5345 Program), the recipient automatically receives a license from the original
5346 licensor to copy, distribute or modify the Program subject to these
5347 terms and conditions. You may not impose any further restrictions on the
5348 recipients' exercise of the rights granted herein.
5351 The Free Software Foundation may publish revised and/or new versions
5352 of the General Public License from time to time. Such new versions will
5353 be similar in spirit to the present version, but may differ in detail to
5354 address new problems or concerns.
5356 Each version is given a distinguishing version number. If the Program
5357 specifies a version number of the license which applies to it and ``any
5358 later version'', you have the option of following the terms and conditions
5359 either of that version or of any later version published by the Free
5360 Software Foundation. If the Program does not specify a version number of
5361 the license, you may choose any version ever published by the Free Software
5365 If you wish to incorporate parts of the Program into other free
5366 programs whose distribution conditions are different, write to the author
5367 to ask for permission. For software which is copyrighted by the Free
5368 Software Foundation, write to the Free Software Foundation; we sometimes
5369 make exceptions for this. Our decision will be guided by the two goals
5370 of preserving the free status of all derivatives of our free software and
5371 of promoting the sharing and reuse of software generally.
5374 @heading NO WARRANTY
5381 BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
5382 FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
5383 OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
5384 PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
5385 OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
5386 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
5387 TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
5388 PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
5389 REPAIR OR CORRECTION.
5392 IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL
5393 ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
5394 REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
5395 INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES
5396 ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT
5397 LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES
5398 SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE
5399 WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN
5400 ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
5404 @heading END OF TERMS AND CONDITIONS
5407 @center END OF TERMS AND CONDITIONS
5411 @unnumberedsec Applying These Terms to Your New Programs
5413 If you develop a new program, and you want it to be of the greatest
5414 possible use to humanity, the best way to achieve this is to make it
5415 free software which everyone can redistribute and change under these
5418 To do so, attach the following notices to the program. It is safest to
5419 attach them to the start of each source file to most effectively convey
5420 the exclusion of warranty; and each file should have at least the
5421 ``copyright'' line and a pointer to where the full notice is found.
5424 @var{one line to give the program's name and a brief idea of what it does.}
5425 Copyright (C) 19@var{yy} @var{name of author}
5427 This program is free software; you can redistribute it and/or modify
5428 it under the terms of the GNU General Public License as published by
5429 the Free Software Foundation; either version 1, or (at your option)
5432 This program is distributed in the hope that it will be useful,
5433 but WITHOUT ANY WARRANTY; without even the implied warranty of
5434 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
5435 GNU General Public License for more details.
5437 You should have received a copy of the GNU General Public License
5438 along with this program; if not, write to the Free Software
5439 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
5442 Also add information on how to contact you by electronic and paper mail.
5444 If the program is interactive, make it output a short notice like this
5445 when it starts in an interactive mode:
5448 Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
5449 Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
5450 This is free software, and you are welcome to redistribute it
5451 under certain conditions; type `show c' for details.
5454 The hypothetical commands `show w' and `show c' should show the
5455 appropriate parts of the General Public License. Of course, the
5456 commands you use may be called something other than `show w' and `show
5457 c'; they could even be mouse-clicks or menu items---whatever suits your
5460 You should also get your employer (if you work as a programmer) or your
5461 school, if any, to sign a ``copyright disclaimer'' for the program, if
5462 necessary. Here is a sample; alter the names:
5465 Yoyodyne, Inc., hereby disclaims all copyright interest in the
5466 program `Gnomovision' (a program to direct compilers to make passes
5467 at assemblers) written by James Hacker.
5469 @var{signature of Ty Coon}, 1 April 1989
5470 Ty Coon, President of Vice
5473 That's all there is to it!
5481 % I think something like @colophon should be in texinfo. In the
5483 \long\def\colophon{\hbox to0pt{}\vfill
5484 \centerline{The body of this manual is set in}
5485 \centerline{\fontname\tenrm,}
5486 \centerline{with headings in {\bf\fontname\tenbf}}
5487 \centerline{and examples in {\tt\fontname\tentt}.}
5488 \centerline{{\it\fontname\tenit\/} and}
5489 \centerline{{\sl\fontname\tensl\/}}
5490 \centerline{are used for emphasis.}\vfill}
5492 % Blame: pesch@cygnus.com, 28mar91.