1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (c) 1988 1989 1990 1991 1992 1993 Free Software Foundation, Inc.
5 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
6 @c of @set vars. However, you can override filename with makeinfo -o.
12 @settitle Debugging with @value{GDBN}
15 @settitle Debugging with @value{GDBN} (@value{TARGET})
17 @setchapternewpage odd
28 @c readline appendices use @vindex
32 @c Determine the edition number in *three* places by hand:
33 @c 1. First ifinfo section 2. title page 3. top node
34 @c To find the locations, search for !!set
36 @c GDB CHANGELOG CONSULTED BETWEEN:
37 @c Fri Oct 11 23:27:06 1991 John Gilmore (gnu at cygnus.com)
38 @c Sat Dec 22 02:51:40 1990 John Gilmore (gnu at cygint)
40 @c THIS MANUAL REQUIRES TEXINFO-2 macros and info-makers to format properly.
43 @c This is a dir.info fragment to support semi-automated addition of
44 @c manuals to an info tree. zoo@cygnus.com is developing this facility.
47 * Gdb:: The GNU debugger.
54 This file documents the GNU debugger @value{GDBN}.
56 @c !!set edition, date, version
57 This is Edition 4.09, April 1993,
58 of @cite{Debugging with @value{GDBN}: the GNU Source-Level Debugger}
59 for GDB Version @value{GDBVN}.
61 Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993 Free Software Foundation, Inc.
63 Permission is granted to make and distribute verbatim copies of
64 this manual provided the copyright notice and this permission notice
65 are preserved on all copies.
68 Permission is granted to process this file through TeX and print the
69 results, provided the printed document carries copying permission
70 notice identical to this one except for the removal of this paragraph
71 (this paragraph not being relevant to the printed manual).
74 Permission is granted to copy and distribute modified versions of this
75 manual under the conditions for verbatim copying, provided also that the
76 entire resulting derived work is distributed under the terms of a
77 permission notice identical to this one.
79 Permission is granted to copy and distribute translations of this manual
80 into another language, under the above conditions for modified versions.
84 @title Debugging with @value{GDBN}
85 @subtitle The GNU Source-Level Debugger
87 @subtitle (@value{TARGET})
90 @c !!set edition, date, version
91 @subtitle Edition 4.09, for @value{GDBN} version @value{GDBVN}
93 @author by Richard M. Stallman and Roland H. Pesch
97 \hfill (Send bugs and comments on @value{GDBN} to bug-gdb\@prep.ai.mit.edu.)\par
98 \hfill {\it Debugging with @value{GDBN}}\par
99 \hfill \TeX{}info \texinfoversion\par
100 \hfill pesch\@cygnus.com\par
104 @vskip 0pt plus 1filll
105 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993 Free Software Foundation, Inc.
107 Permission is granted to make and distribute verbatim copies of
108 this manual provided the copyright notice and this permission notice
109 are preserved on all copies.
111 Permission is granted to copy and distribute modified versions of this
112 manual under the conditions for verbatim copying, provided also that the
113 entire resulting derived work is distributed under the terms of a
114 permission notice identical to this one.
116 Permission is granted to copy and distribute translations of this manual
117 into another language, under the above conditions for modified versions.
123 @top Debugging with @value{GDBN}
125 This file describes @value{GDBN}, the GNU symbolic debugger.
127 @c !!set edition, date, version
128 This is Edition 4.09, April 1993, for GDB Version @value{GDBVN}.
131 * Summary:: Summary of @value{GDBN}
133 * New Features:: New features since GDB version 3.5
136 * Sample Session:: A sample @value{GDBN} session
139 * Invocation:: Getting in and out of @value{GDBN}
140 * Commands:: @value{GDBN} commands
141 * Running:: Running programs under @value{GDBN}
142 * Stopping:: Stopping and continuing
143 * Stack:: Examining the stack
144 * Source:: Examining source files
145 * Data:: Examining data
147 * Languages:: Using @value{GDBN} with different languages
150 * C:: C language support
152 @c remnant makeinfo bug, blank line needed after two end-ifs?
154 * Symbols:: Examining the symbol table
155 * Altering:: Altering execution
156 * GDB Files:: @value{GDBN} files
157 * Targets:: Specifying a debugging target
158 * Controlling GDB:: Controlling @value{GDBN}
159 * Sequences:: Canned sequences of commands
161 * Emacs:: Using @value{GDBN} under GNU Emacs
164 * GDB Bugs:: Reporting bugs in @value{GDBN}
165 * Command Line Editing:: Facilities of the readline library
166 * Using History Interactively::
170 @ifclear PRECONFIGURED
171 * Formatting Documentation:: How to format and print GDB documentation
172 * Installing GDB:: Installing GDB
180 @unnumbered Summary of @value{GDBN}
182 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
183 going on ``inside'' another program while it executes---or what another
184 program was doing at the moment it crashed.
186 @value{GDBN} can do four main kinds of things (plus other things in support of
187 these) to help you catch bugs in the act:
191 Start your program, specifying anything that might affect its behavior.
194 Make your program stop on specified conditions.
197 Examine what has happened, when your program has stopped.
200 Change things in your program, so you can experiment with correcting the
201 effects of one bug and go on to learn about another.
206 You can use @value{GDBN} to debug programs written in C or C++.
209 You can use @value{GDBN} to debug programs written in C, C++, and
213 @value{GDBN} can be used to debug programs written in Fortran, although
214 it does not yet support entering expressions, printing values, etc.
215 using Fortran syntax. It may be necessary to refer to some variables
216 with a trailing underscore.
221 * Free Software:: Freely redistributable software
222 * Contributors:: Contributors to GDB
226 @unnumberedsec Free software
228 @value{GDBN} is @dfn{free software}, protected by the GNU General Public License
229 (GPL). The GPL gives you the freedom to copy or adapt a licensed
230 program---but every person getting a copy also gets with it the
231 freedom to modify that copy (which means that they must get access to
232 the source code), and the freedom to distribute further copies.
233 Typical software companies use copyrights to limit your freedoms; the
234 Free Software Foundation uses the GPL to preserve these freedoms.
236 Fundamentally, the General Public License is a license which says that
237 you have these freedoms and that you cannot take these freedoms away
241 @unnumberedsec Contributors to GDB
243 Richard Stallman was the original author of GDB, and of many other GNU
244 programs. Many others have contributed to its development. This
245 section attempts to credit major contributors. One of the virtues of
246 free software is that everyone is free to contribute to it; with
247 regret, we cannot actually acknowledge everyone here. The file
248 @file{ChangeLog} in the GDB distribution approximates a blow-by-blow
251 Changes much prior to version 2.0 are lost in the mists of time.
254 @emph{Plea:} Additions to this section are particularly welcome. If you
255 or your friends (or enemies, to be evenhanded) have been unfairly
256 omitted from this list, we would like to add your names!
259 So that they may not regard their long labor as thankless, we
260 particularly thank those who shepherded GDB through major releases: Fred
261 Fish (release 4.9), Stu Grossman and John Gilmore (releases 4.8, 4.7,
262 4.6, 4.5, 4.4), John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9); Jim
263 Kingdon (releases 3.5, 3.4, 3.3); and Randy Smith (releases 3.2, 3.1,
264 3.0). As major maintainer of GDB for some period, each contributed
265 significantly to the structure, stability, and capabilities of the
268 Richard Stallman, assisted at various times by Pete TerMaat, Chris
269 Hanson, and Richard Mlynarik, handled releases through 2.8.
272 Michael Tiemann is the author of most of the GNU C++ support in GDB,
273 with significant additional contributions from Per Bothner. James
274 Clark wrote the GNU C++ demangler. Early work on C++ was by Peter
275 TerMaat (who also did much general update work leading to release 3.0).
278 GDB 4 uses the BFD subroutine library to examine multiple
279 object-file formats; BFD was a joint project of David V.
280 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
282 David Johnson wrote the original COFF support; Pace Willison did
283 the original support for encapsulated COFF.
285 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
286 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
287 support. Jean-Daniel Fekete contributed Sun 386i support. Chris
288 Hanson improved the HP9000 support. Noboyuki Hikichi and Tomoyuki
289 Hasei contributed Sony/News OS 3 support. David Johnson contributed
290 Encore Umax support. Jyrki Kuoppala contributed Altos 3068 support.
291 Keith Packard contributed NS32K support. Doug Rabson contributed
292 Acorn Risc Machine support. Chris Smith contributed Convex support
293 (and Fortran debugging). Jonathan Stone contributed Pyramid support.
294 Michael Tiemann contributed SPARC support. Tim Tucker contributed
295 support for the Gould NP1 and Gould Powernode. Pace Willison
296 contributed Intel 386 support. Jay Vosburgh contributed Symmetry
299 Rich Schaefer and Peter Schauer helped with support of SunOS shared
302 Jay Fenlason and Roland McGrath ensured that GDB and GAS agree about
303 several machine instruction sets.
305 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped
306 develop remote debugging. Intel Corporation and Wind River Systems
307 contributed remote debugging modules for their products.
309 Brian Fox is the author of the readline libraries providing
310 command-line editing and command history.
312 Andrew Beers of SUNY Buffalo wrote the language-switching code,
314 the Modula-2 support,
316 and contributed the Languages chapter of this manual.
318 Fred Fish wrote most of the support for Unix System Vr4.
320 He also enhanced the command-completion support to cover C++ overloaded
324 Hitachi America, Ltd. sponsored the support for Hitachi microprocessors.
328 @unnumbered New Features since GDB Version 3.5
332 Using the new command @code{target}, you can select at runtime whether
333 you are debugging local files, local processes, standalone systems over
334 a serial port, realtime systems over a TCP/IP connection, etc. The
335 command @code{load} can download programs into a remote system. Serial
336 stubs are available for Motorola 680x0, Intel 80386, and Sparc remote
337 systems; GDB also supports debugging realtime processes running under
338 VxWorks, using SunRPC Remote Procedure Calls over TCP/IP to talk to a
339 debugger stub on the target system. Internally, GDB now uses a function
340 vector to mediate access to different targets; if you need to add your
341 own support for a remote protocol, this makes it much easier.
344 GDB now sports watchpoints as well as breakpoints. You can use a
345 watchpoint to stop execution whenever the value of an expression
346 changes, without having to predict a particular place in your program
347 where this may happen.
350 Commands that issue wide output now insert newlines at places designed
351 to make the output more readable.
353 @item Object Code Formats
354 GDB uses a new library called the Binary File Descriptor (BFD) Library
355 to permit it to switch dynamically, without reconfiguration or
356 recompilation, between different object-file formats. Formats currently
357 supported are COFF, ELF, a.out, Intel 960 b.out, MIPS ECOFF, HPPA SOM
358 (with stabs debugging), and S-records; files may be read as .o files,
359 archive libraries, or core dumps. BFD is available as a subroutine
360 library so that other programs may take advantage of it, and the other
361 GNU binary utilities are being converted to use it.
363 @item Configuration and Ports
364 Compile-time configuration (to select a particular architecture and
365 operating system) is much easier. The script @code{configure} now
366 allows you to configure GDB as either a native debugger or a
367 cross-debugger. @xref{Installing GDB}, for details on how to
371 The user interface to the GDB control variables is simpler,
372 and is consolidated in two commands, @code{set} and @code{show}. Output
373 lines are now broken at readable places, rather than overflowing onto
374 the next line. You can suppress output of machine-level addresses,
375 displaying only source language information.
378 GDB now supports C++ multiple inheritance (if used with a GCC
379 version 2 compiler), and also has limited support for C++ exception
380 handling, with the commands @code{catch} and @code{info catch}: GDB
381 can break when an exception is raised, before the stack is peeled back
382 to the exception handler's context.
386 GDB now has preliminary support for the GNU Modula-2 compiler, currently
387 under development at the State University of New York at Buffalo.
388 Coordinated development of both GDB and the GNU Modula-2 compiler will
389 continue. Other Modula-2 compilers are currently not supported, and
390 attempting to debug programs compiled with them will likely result in an
391 error as the symbol table of the executable is read in.
394 @item Command Rationalization
395 Many GDB commands have been renamed to make them easier to remember
396 and use. In particular, the subcommands of @code{info} and
397 @code{show}/@code{set} are grouped to make the former refer to the state
398 of your program, and the latter refer to the state of GDB itself.
399 @xref{Renamed Commands}, for details on what commands were renamed.
401 @item Shared Libraries
402 GDB 4 can debug programs and core files that use SunOS, SVR4, or IBM RS/6000
406 GDB 4 has a reference card. @xref{Formatting Documentation,,Formatting
407 the Documentation}, for instructions about how to print it.
413 @chapter A Sample @value{GDBN} Session
415 You can use this manual at your leisure to read all about @value{GDBN}.
416 However, a handful of commands are enough to get started using the
417 debugger. This chapter illustrates those commands.
420 In this sample session, we emphasize user input like this: @b{input},
421 to make it easier to pick out from the surrounding output.
424 @c FIXME: this example may not be appropriate for some configs, where
425 @c FIXME...primary interest is in remote use.
427 One of the preliminary versions of GNU @code{m4} (a generic macro
428 processor) exhibits the following bug: sometimes, when we change its
429 quote strings from the default, the commands used to capture one macro
430 definition within another stop working. In the following short @code{m4}
431 session, we define a macro @code{foo} which expands to @code{0000}; we
432 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
433 same thing. However, when we change the open quote string to
434 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
435 procedure fails to define a new synonym @code{baz}:
444 @b{define(bar,defn(`foo'))}
448 @b{changequote(<QUOTE>,<UNQUOTE>)}
450 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
453 m4: End of input: 0: fatal error: EOF in string
457 Let us use @value{GDBN} to try to see what is going on.
460 $ @b{@value{GDBP} m4}
461 @c FIXME: this falsifies the exact text played out, to permit smallbook
462 @c FIXME... format to come out better.
463 GDB is free software and you are welcome to distribute copies
464 of it under certain conditions; type "show copying" to see
466 There is absolutely no warranty for GDB; type "show warranty"
468 GDB @value{GDBVN}, Copyright 1993 Free Software Foundation, Inc...
473 @value{GDBN} reads only enough symbol data to know where to find the rest when
474 needed; as a result, the first prompt comes up very quickly. We now
475 tell @value{GDBN} to use a narrower display width than usual, so that examples
476 will fit in this manual.
479 (@value{GDBP}) @b{set width 70}
483 We need to see how the @code{m4} built-in @code{changequote} works.
484 Having looked at the source, we know the relevant subroutine is
485 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
486 @code{break} command.
489 (@value{GDBP}) @b{break m4_changequote}
490 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
494 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
495 control; as long as control does not reach the @code{m4_changequote}
496 subroutine, the program runs as usual:
499 (@value{GDBP}) @b{run}
500 Starting program: /work/Editorial/gdb/gnu/m4/m4
508 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
509 suspends execution of @code{m4}, displaying information about the
510 context where it stops.
513 @b{changequote(<QUOTE>,<UNQUOTE>)}
515 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
517 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
521 Now we use the command @code{n} (@code{next}) to advance execution to
522 the next line of the current function.
526 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
531 @code{set_quotes} looks like a promising subroutine. We can go into it
532 by using the command @code{s} (@code{step}) instead of @code{next}.
533 @code{step} goes to the next line to be executed in @emph{any}
534 subroutine, so it steps into @code{set_quotes}.
538 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
540 530 if (lquote != def_lquote)
544 The display that shows the subroutine where @code{m4} is now
545 suspended (and its arguments) is called a stack frame display. It
546 shows a summary of the stack. We can use the @code{backtrace}
547 command (which can also be spelled @code{bt}), to see where we are
548 in the stack as a whole: the @code{backtrace} command displays a
549 stack frame for each active subroutine.
552 (@value{GDBP}) @b{bt}
553 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
555 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
557 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
558 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
560 #4 0x79dc in expand_input () at macro.c:40
561 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
565 We will step through a few more lines to see what happens. The first two
566 times, we can use @samp{s}; the next two times we use @code{n} to avoid
567 falling into the @code{xstrdup} subroutine.
571 0x3b5c 532 if (rquote != def_rquote)
573 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
574 def_lquote : xstrdup(lq);
576 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
579 538 len_lquote = strlen(rquote);
583 The last line displayed looks a little odd; we can examine the variables
584 @code{lquote} and @code{rquote} to see if they are in fact the new left
585 and right quotes we specified. We use the command @code{p}
586 (@code{print}) to see their values.
589 (@value{GDBP}) @b{p lquote}
590 $1 = 0x35d40 "<QUOTE>"
591 (@value{GDBP}) @b{p rquote}
592 $2 = 0x35d50 "<UNQUOTE>"
596 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
597 To look at some context, we can display ten lines of source
598 surrounding the current line with the @code{l} (@code{list}) command.
604 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
606 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
609 538 len_lquote = strlen(rquote);
610 539 len_rquote = strlen(lquote);
617 Let us step past the two lines that set @code{len_lquote} and
618 @code{len_rquote}, and then examine the values of those variables.
622 539 len_rquote = strlen(lquote);
625 (@value{GDBP}) @b{p len_lquote}
627 (@value{GDBP}) @b{p len_rquote}
632 That certainly looks wrong, assuming @code{len_lquote} and
633 @code{len_rquote} are meant to be the lengths of @code{lquote} and
634 @code{rquote} respectively. We can set them to better values using
635 the @code{p} command, since it can print the value of
636 any expression---and that expression can include subroutine calls and
640 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
642 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
647 Is that enough to fix the problem of using the new quotes with the
648 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
649 executing with the @code{c} (@code{continue}) command, and then try the
650 example that caused trouble initially:
656 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
663 Success! The new quotes now work just as well as the default ones. The
664 problem seems to have been just the two typos defining the wrong
665 lengths. We allow @code{m4} exit by giving it an EOF as input:
669 Program exited normally.
673 The message @samp{Program exited normally.} is from @value{GDBN}; it
674 indicates @code{m4} has finished executing. We can end our @value{GDBN}
675 session with the @value{GDBN} @code{quit} command.
678 (@value{GDBP}) @b{quit}
683 @chapter Getting In and Out of @value{GDBN}
685 This chapter discusses how to start @value{GDBN}, and how to get out of it.
686 (The essentials: type @samp{@value{GDBP}} to start GDB, and type @kbd{quit}
687 or @kbd{C-d} to exit.)
690 * Invoking GDB:: How to start @value{GDBN}
691 * Quitting GDB:: How to quit @value{GDBN}
692 * Shell Commands:: How to use shell commands inside @value{GDBN}
696 @section Invoking @value{GDBN}
699 For details on starting up @value{GDBP} as a
700 remote debugger attached to a Hitachi microprocessor, see @ref{Hitachi
701 Remote,,@value{GDBN} and Hitachi Microprocessors}.
704 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
705 @value{GDBN} reads commands from the terminal until you tell it to exit.
707 You can also run @code{@value{GDBP}} with a variety of arguments and options,
708 to specify more of your debugging environment at the outset.
711 The command-line options described here are designed
712 to cover a variety of situations; in some environments, some of these
713 options may effectively be unavailable.
716 The most usual way to start @value{GDBN} is with one argument,
717 specifying an executable program:
720 @value{GDBP} @var{program}
725 You can also start with both an executable program and a core file
729 @value{GDBP} @var{program} @var{core}
732 You can, instead, specify a process ID as a second argument, if you want
733 to debug a running process:
736 @value{GDBP} @var{program} 1234
740 would attach @value{GDBN} to process @code{1234} (unless you also have a file
741 named @file{1234}; @value{GDBN} does check for a core file first).
743 Taking advantage of the second command-line argument requires a fairly
744 complete operating system; when you use @value{GDBN} as a remote debugger
745 attached to a bare board, there may not be any notion of ``process'',
746 and there is often no way to get a core dump.
750 You can further control how @value{GDBN} starts up by using command-line
751 options. @value{GDBN} itself can remind you of the options available.
761 to display all available options and briefly describe their use
762 (@samp{@value{GDBP} -h} is a shorter equivalent).
764 All options and command line arguments you give are processed
765 in sequential order. The order makes a difference when the
766 @samp{-x} option is used.
772 * Remote Serial:: @value{GDBN} remote serial protocol
775 * i960-Nindy Remote:: @value{GDBN} with a remote i960 (Nindy)
778 * UDI29K Remote:: @value{GDBN} and the UDI protocol for AMD29K
779 * EB29K Remote:: @value{GDBN} with a remote EB29K
782 * VxWorks Remote:: @value{GDBN} and VxWorks
785 * ST2000 Remote:: @value{GDBN} with a Tandem ST2000
788 * Hitachi Remote:: @value{GDBN} and Hitachi Microprocessors
791 * MIPS Remote:: @value{GDBN} and MIPS boards
794 * Simulator:: Simulated CPU target
797 @c remnant makeinfo bug requires this blank line after *two* end-ifblahs:
799 * File Options:: Choosing files
800 * Mode Options:: Choosing modes
808 @subsection Choosing files
811 When @value{GDBN} starts, it reads any arguments other than options as
812 specifying an executable file and core file (or process ID). This is
813 the same as if the arguments were specified by the @samp{-se} and
814 @samp{-c} options respectively. (@value{GDBN} reads the first argument
815 that does not have an associated option flag as equivalent to the
816 @samp{-se} option followed by that argument; and the second argument
817 that does not have an associated option flag, if any, as equivalent to
818 the @samp{-c} option followed by that argument.)
821 When @value{GDBN} starts, it reads any argument other than options as
822 specifying an executable file. This is the same as if the argument was
823 specified by the @samp{-se} option.
826 Many options have both long and short forms; both are shown in the
827 following list. @value{GDBN} also recognizes the long forms if you truncate
828 them, so long as enough of the option is present to be unambiguous.
829 (If you prefer, you can flag option arguments with @samp{--} rather
830 than @samp{-}, though we illustrate the more usual convention.)
833 @item -symbols=@var{file}
835 Read symbol table from file @var{file}.
837 @item -exec=@var{file}
839 Use file @var{file} as the executable file to execute when
844 appropriate, and for examining pure data in conjunction with a core
849 Read symbol table from file @var{file} and use it as the executable
853 @item -core=@var{file}
855 Use file @var{file} as a core dump to examine.
857 @item -c @var{number}
858 Connect to process ID @var{number}, as with the @code{attach} command
859 (unless there is a file in core-dump format named @var{number}, in which
860 case @samp{-c} specifies that file as a core dump to read).
863 @item -command=@var{file}
865 Execute @value{GDBN} commands from file @var{file}. @xref{Command
866 Files,, Command files}.
868 @item -directory=@var{directory}
869 @itemx -d @var{directory}
870 Add @var{directory} to the path to search for source files.
875 @emph{Warning: this option depends on operating system facilities that are not
876 supported on all systems.}@*
877 If memory-mapped files are available on your system through the @code{mmap}
878 system call, you can use this option
879 to have @value{GDBN} write the symbols from your
880 program into a reusable file in the current directory. If the program you are debugging is
881 called @file{/tmp/fred}, the mapped symbol file will be @file{./fred.syms}.
882 Future @value{GDBN} debugging sessions will notice the presence of this file,
883 and will quickly map in symbol information from it, rather than reading
884 the symbol table from the executable program.
886 @c FIXME! Really host, not target?
887 The @file{.syms} file is specific to the host machine where @value{GDBN}
888 is run. It holds an exact image of the internal @value{GDBN} symbol
889 table. It cannot be shared across multiple host platforms.
894 Read each symbol file's entire symbol table immediately, rather than
895 the default, which is to read it incrementally as it is needed.
896 This makes startup slower, but makes future operations faster.
900 The @code{-mapped} and @code{-readnow} options are typically combined in
901 order to build a @file{.syms} file that contains complete symbol
902 information. (@xref{Files,,Commands to specify files}, for information
903 on @file{.syms} files.) A simple GDB invocation to do nothing but build
904 a @file{.syms} file for future use is:
907 gdb -batch -nx -mapped -readnow programname
912 @subsection Choosing modes
914 You can run @value{GDBN} in various alternative modes---for example, in
915 batch mode or quiet mode.
920 Do not execute commands from any initialization files (normally called
921 @file{@value{GDBINIT}}). Normally, the commands in these files are
922 executed after all the command options and arguments have been
923 processed. @xref{Command Files,,Command files}.
927 ``Quiet''. Do not print the introductory and copyright messages. These
928 messages are also suppressed in batch mode.
931 Run in batch mode. Exit with status @code{0} after processing all the
932 command files specified with @samp{-x} (and all commands from
933 initialization files, if not inhibited with @samp{-n}). Exit with
934 nonzero status if an error occurs in executing the @value{GDBN} commands
935 in the command files.
937 Batch mode may be useful for running @value{GDBN} as a filter, for example to
938 download and run a program on another computer; in order to make this
939 more useful, the message
942 Program exited normally.
946 (which is ordinarily issued whenever a program running under @value{GDBN} control
947 terminates) is not issued when running in batch mode.
949 @item -cd=@var{directory}
950 Run @value{GDBN} using @var{directory} as its working directory,
951 instead of the current directory.
954 @item -context @var{authentication}
955 When the Energize programming system starts up @value{GDBN}, it uses this
956 option to trigger an alternate mode of interaction.
957 @var{authentication} is a pair of numeric codes that identify @value{GDBN}
958 as a client in the Energize environment. Avoid this option when you run
959 @value{GDBN} directly from the command line. See @ref{Energize,,Using
960 @value{GDBN} with Energize} for more discussion of using @value{GDBN} with Energize.
966 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells @value{GDBN}
967 to output the full file name and line number in a standard,
968 recognizable fashion each time a stack frame is displayed (which
969 includes each time your program stops). This recognizable format looks
970 like two @samp{\032} characters, followed by the file name, line number
971 and character position separated by colons, and a newline. The
972 Emacs-to-@value{GDBN} interface program uses the two @samp{\032} characters as
973 a signal to display the source code for the frame.
978 Set the line speed (baud rate or bits per second) of any serial
979 interface used by @value{GDBN} for remote debugging.
981 @item -tty=@var{device}
982 Run using @var{device} for your program's standard input and output.
983 @c FIXME: kingdon thinks there is more to -tty. Investigate.
988 @section Quitting @value{GDBN}
989 @cindex exiting @value{GDBN}
990 @cindex leaving @value{GDBN}
996 To exit @value{GDBN}, use the @code{quit} command (abbreviated @code{q}), or type
997 an end-of-file character (usually @kbd{C-d}).
1001 An interrupt (often @kbd{C-c}) will not exit from @value{GDBN}, but rather
1002 will terminate the action of any @value{GDBN} command that is in progress and
1003 return to @value{GDBN} command level. It is safe to type the interrupt
1004 character at any time because @value{GDBN} does not allow it to take effect
1005 until a time when it is safe.
1008 If you have been using @value{GDBN} to control an attached process or
1009 device, you can release it with the @code{detach} command
1010 (@pxref{Attach, ,Debugging an already-running process}).
1013 @node Shell Commands
1014 @section Shell commands
1016 If you need to execute occasional shell commands during your
1017 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1018 just use the @code{shell} command.
1021 @item shell @var{command string}
1023 @cindex shell escape
1024 Invoke a the standard shell to execute @var{command string}.
1026 If it exists, the environment variable @code{SHELL} determines which
1027 shell to run. Otherwise @value{GDBN} uses @code{/bin/sh}.
1031 The utility @code{make} is often needed in development environments.
1032 You do not have to use the @code{shell} command for this purpose in
1036 @item make @var{make-args}
1038 @cindex calling make
1039 Execute the @code{make} program with the specified
1040 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1044 @chapter @value{GDBN} Commands
1046 You can abbreviate a @value{GDBN} command to the first few letters of the command
1047 name, if that abbreviation is unambiguous; and you can repeat certain
1048 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1049 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1050 show you the alternatives available, if there is more than one possibility).
1053 * Command Syntax:: How to give commands to @value{GDBN}
1054 * Completion:: Command completion
1055 * Help:: How to ask @value{GDBN} for help
1058 @node Command Syntax
1059 @section Command syntax
1061 A @value{GDBN} command is a single line of input. There is no limit on
1062 how long it can be. It starts with a command name, which is followed by
1063 arguments whose meaning depends on the command name. For example, the
1064 command @code{step} accepts an argument which is the number of times to
1065 step, as in @samp{step 5}. You can also use the @code{step} command
1066 with no arguments. Some command names do not allow any arguments.
1068 @cindex abbreviation
1069 @value{GDBN} command names may always be truncated if that abbreviation is
1070 unambiguous. Other possible command abbreviations are listed in the
1071 documentation for individual commands. In some cases, even ambiguous
1072 abbreviations are allowed; for example, @code{s} is specially defined as
1073 equivalent to @code{step} even though there are other commands whose
1074 names start with @code{s}. You can test abbreviations by using them as
1075 arguments to the @code{help} command.
1077 @cindex repeating commands
1079 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1080 repeat the previous command. Certain commands (for example, @code{run})
1081 will not repeat this way; these are commands for which unintentional
1082 repetition might cause trouble and which you are unlikely to want to
1085 The @code{list} and @code{x} commands, when you repeat them with
1086 @key{RET}, construct new arguments rather than repeating
1087 exactly as typed. This permits easy scanning of source or memory.
1089 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1090 output, in a way similar to the common utility @code{more}
1091 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1092 @key{RET} too many in this situation, @value{GDBN} disables command
1093 repetition after any command that generates this sort of display.
1097 Any text from a @kbd{#} to the end of the line is a comment; it does
1098 nothing. This is useful mainly in command files (@pxref{Command
1099 Files,,Command files}).
1102 @section Command completion
1105 @cindex word completion
1106 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1107 only one possibility; it can also show you what the valid possibilities
1108 are for the next word in a command, at any time. This works for @value{GDBN}
1109 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1111 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1112 of a word. If there is only one possibility, @value{GDBN} will fill in the
1113 word, and wait for you to finish the command (or press @key{RET} to
1114 enter it). For example, if you type
1116 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1117 @c complete accuracy in these examples; space introduced for clarity.
1118 @c If texinfo enhancements make it unnecessary, it would be nice to
1119 @c replace " @key" by "@key" in the following...
1121 (@value{GDBP}) info bre @key{TAB}
1125 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1126 the only @code{info} subcommand beginning with @samp{bre}:
1129 (@value{GDBP}) info breakpoints
1133 You can either press @key{RET} at this point, to run the @code{info
1134 breakpoints} command, or backspace and enter something else, if
1135 @samp{breakpoints} does not look like the command you expected. (If you
1136 were sure you wanted @code{info breakpoints} in the first place, you
1137 might as well just type @key{RET} immediately after @samp{info bre},
1138 to exploit command abbreviations rather than command completion).
1140 If there is more than one possibility for the next word when you press
1141 @key{TAB}, @value{GDBN} will sound a bell. You can either supply more
1142 characters and try again, or just press @key{TAB} a second time, and
1143 @value{GDBN} will display all the possible completions for that word. For
1144 example, you might want to set a breakpoint on a subroutine whose name
1145 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1146 just sounds the bell. Typing @key{TAB} again will display all the
1147 function names in your program that begin with those characters, for
1151 (@value{GDBP}) b make_ @key{TAB}
1152 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1153 make_a_section_from_file make_environ
1154 make_abs_section make_function_type
1155 make_blockvector make_pointer_type
1156 make_cleanup make_reference_type
1157 make_command make_symbol_completion_list
1158 (@value{GDBP}) b make_
1162 After displaying the available possibilities, @value{GDBN} copies your
1163 partial input (@samp{b make_} in the example) so you can finish the
1166 If you just want to see the list of alternatives in the first place, you
1167 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1168 means @kbd{@key{META} ?}. You can type this
1170 either by holding down a
1171 key designated as the @key{META} shift on your keyboard (if there is
1172 one) while typing @kbd{?}, or
1174 as @key{ESC} followed by @kbd{?}.
1176 @cindex quotes in commands
1177 @cindex completion of quoted strings
1178 Sometimes the string you need, while logically a ``word'', may contain
1179 parentheses or other characters that @value{GDBN} normally excludes from its
1180 notion of a word. To permit word completion to work in this situation,
1181 you may enclose words in @code{'} (single quote marks) in @value{GDBN} commands.
1184 The most likely situation where you might need this is in typing the
1185 name of a C++ function. This is because C++ allows function overloading
1186 (multiple definitions of the same function, distinguished by argument
1187 type). For example, when you want to set a breakpoint you may need to
1188 distinguish whether you mean the version of @code{name} that takes an
1189 @code{int} parameter, @code{name(int)}, or the version that takes a
1190 @code{float} parameter, @code{name(float)}. To use the word-completion
1191 facilities in this situation, type a single quote @code{'} at the
1192 beginning of the function name. This alerts @value{GDBN} that it may need to
1193 consider more information than usual when you press @key{TAB} or
1194 @kbd{M-?} to request word completion:
1197 (@value{GDBP}) b 'bubble( @key{M-?}
1198 bubble(double,double) bubble(int,int)
1199 (@value{GDBP}) b 'bubble(
1202 In some cases, @value{GDBN} can tell that completing a name will require
1203 quotes. When this happens, @value{GDBN} will insert the quote for you (while
1204 completing as much as it can) if you do not type the quote in the first
1208 (@value{GDBP}) b bub @key{TAB}
1209 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1210 (@value{GDBP}) b 'bubble(
1214 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1215 you have not yet started typing the argument list when you ask for
1216 completion on an overloaded symbol.
1221 @section Getting help
1222 @cindex online documentation
1225 You can always ask @value{GDBN} itself for information on its commands, using the
1226 command @code{help}.
1232 You can use @code{help} (abbreviated @code{h}) with no arguments to
1233 display a short list of named classes of commands:
1237 List of classes of commands:
1239 running -- Running the program
1240 stack -- Examining the stack
1241 data -- Examining data
1242 breakpoints -- Making program stop at certain points
1243 files -- Specifying and examining files
1244 status -- Status inquiries
1245 support -- Support facilities
1246 user-defined -- User-defined commands
1247 aliases -- Aliases of other commands
1248 obscure -- Obscure features
1250 Type "help" followed by a class name for a list of
1251 commands in that class.
1252 Type "help" followed by command name for full
1254 Command name abbreviations are allowed if unambiguous.
1258 @item help @var{class}
1259 Using one of the general help classes as an argument, you can get a
1260 list of the individual commands in that class. For example, here is the
1261 help display for the class @code{status}:
1264 (@value{GDBP}) help status
1269 @c Line break in "show" line falsifies real output, but needed
1270 @c to fit in smallbook page size.
1271 show -- Generic command for showing things set
1273 info -- Generic command for printing status
1275 Type "help" followed by command name for full
1277 Command name abbreviations are allowed if unambiguous.
1281 @item help @var{command}
1282 With a command name as @code{help} argument, @value{GDBN} will display a
1283 short paragraph on how to use that command.
1286 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1287 and @code{show} to inquire about the state of your program, or the state
1288 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1289 manual introduces each of them in the appropriate context. The listings
1290 under @code{info} and under @code{show} in the Index point to
1291 all the sub-commands. @xref{Index}.
1298 This command (abbreviated @code{i}) is for describing the state of your
1299 program. For example, you can list the arguments given to your program
1300 with @code{info args}, list the registers currently in use with @code{info
1301 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1302 You can get a complete list of the @code{info} sub-commands with
1303 @w{@code{help info}}.
1307 In contrast, @code{show} is for describing the state of @value{GDBN} itself.
1308 You can change most of the things you can @code{show}, by using the
1309 related command @code{set}; for example, you can control what number
1310 system is used for displays with @code{set radix}, or simply inquire
1311 which is currently in use with @code{show radix}.
1314 To display all the settable parameters and their current
1315 values, you can use @code{show} with no arguments; you may also use
1316 @code{info set}. Both commands produce the same display.
1317 @c FIXME: "info set" violates the rule that "info" is for state of
1318 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1319 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1323 Here are three miscellaneous @code{show} subcommands, all of which are
1324 exceptional in lacking corresponding @code{set} commands:
1327 @kindex show version
1328 @cindex version number
1330 Show what version of @value{GDBN} is running. You should include this
1331 information in @value{GDBN} bug-reports. If multiple versions of @value{GDBN} are in
1332 use at your site, you may occasionally want to determine which version
1333 of @value{GDBN} you are running; as @value{GDBN} evolves, new commands are introduced,
1334 and old ones may wither away. The version number is also announced
1335 when you start @value{GDBN}.
1337 @kindex show copying
1339 Display information about permission for copying @value{GDBN}.
1341 @kindex show warranty
1343 Display the GNU ``NO WARRANTY'' statement.
1347 @chapter Running Programs Under @value{GDBN}
1349 When you run a program under @value{GDBN}, you must first generate
1350 debugging information when you compile it.
1352 You may start it with its arguments, if any, in an environment of your
1353 choice. You may redirect your program's input and output, debug an
1354 already running process, or kill a child process.
1358 * Compilation:: Compiling for debugging
1359 * Starting:: Starting your program
1361 * Arguments:: Your program's arguments
1362 * Environment:: Your program's environment
1363 * Working Directory:: Your program's working directory
1364 * Input/Output:: Your program's input and output
1365 * Attach:: Debugging an already-running process
1366 * Kill Process:: Killing the child process
1367 * Process Information:: Additional process information
1372 @section Compiling for debugging
1374 In order to debug a program effectively, you need to generate
1375 debugging information when you compile it. This debugging information
1376 is stored in the object file; it describes the data type of each
1377 variable or function and the correspondence between source line numbers
1378 and addresses in the executable code.
1380 To request debugging information, specify the @samp{-g} option when you run
1383 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1384 options together. Using those compilers, you cannot generate optimized
1385 executables containing debugging information.
1387 @value{NGCC}, the GNU C compiler, supports @samp{-g} with or without
1388 @samp{-O}, making it possible to debug optimized code. We recommend
1389 that you @emph{always} use @samp{-g} whenever you compile a program.
1390 You may think your program is correct, but there is no sense in pushing
1393 @cindex optimized code, debugging
1394 @cindex debugging optimized code
1395 When you debug a program compiled with @samp{-g -O}, remember that the
1396 optimizer is rearranging your code; the debugger will show you what is
1397 really there. Do not be too surprised when the execution path does not
1398 exactly match your source file! An extreme example: if you define a
1399 variable, but never use it, @value{GDBN} will never see that
1400 variable---because the compiler optimizes it out of existence.
1402 Some things do not work as well with @samp{-g -O} as with just
1403 @samp{-g}, particularly on machines with instruction scheduling. If in
1404 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1405 please report it as a bug (including a test case!).
1407 Older versions of the GNU C compiler permitted a variant option
1408 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1409 format; if your GNU C compiler has this option, do not use it.
1412 @comment As far as I know, there are no cases in which @value{GDBN} will
1413 @comment produce strange output in this case. (but no promises).
1414 If your program includes archives made with the @code{ar} program, and
1415 if the object files used as input to @code{ar} were compiled without the
1416 @samp{-g} option and have names longer than 15 characters, @value{GDBN} will get
1417 confused reading your program's symbol table. No error message will be
1418 given, but @value{GDBN} may behave strangely. The reason for this problem is a
1419 deficiency in the Unix archive file format, which cannot represent file
1420 names longer than 15 characters.
1422 To avoid this problem, compile the archive members with the @samp{-g}
1423 option or use shorter file names. Alternatively, use a version of GNU
1424 @code{ar} dated more recently than August 1989.
1429 @section Starting your program
1437 Use the @code{run} command to start your program under @value{GDBN}. You must
1438 first specify the program name
1442 with an argument to @value{GDBN} (@pxref{Invocation, ,Getting In and
1443 Out of @value{GDBN}}), or by using the @code{file} or @code{exec-file}
1444 command (@pxref{Files, ,Commands to specify files}).
1449 If you are running your program in an execution environment that
1450 supports processes, @code{run} creates an inferior process and makes
1451 that process run your program. (In environments without processes,
1452 @code{run} jumps to the start of your program.)
1454 The execution of a program is affected by certain information it
1455 receives from its superior. @value{GDBN} provides ways to specify this
1456 information, which you must do @emph{before} starting your program. (You
1457 can change it after starting your program, but such changes will only affect
1458 your program the next time you start it.) This information may be
1459 divided into four categories:
1462 @item The @emph{arguments.}
1463 Specify the arguments to give your program as the arguments of the
1464 @code{run} command. If a shell is available on your target, the shell
1465 is used to pass the arguments, so that you may use normal conventions
1466 (such as wildcard expansion or variable substitution) in describing
1467 the arguments. In Unix systems, you can control which shell is used
1468 with the @code{SHELL} environment variable. @xref{Arguments, ,Your
1469 program's arguments}.
1471 @item The @emph{environment.}
1472 Your program normally inherits its environment from @value{GDBN}, but you can
1473 use the @value{GDBN} commands @code{set environment} and @code{unset
1474 environment} to change parts of the environment that will be given to
1475 your program. @xref{Environment, ,Your program's environment}.
1477 @item The @emph{working directory.}
1478 Your program inherits its working directory from @value{GDBN}. You can set
1479 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1480 @xref{Working Directory, ,Your program's working directory}.
1482 @item The @emph{standard input and output.}
1483 Your program normally uses the same device for standard input and
1484 standard output as @value{GDBN} is using. You can redirect input and output
1485 in the @code{run} command line, or you can use the @code{tty} command to
1486 set a different device for your program.
1487 @xref{Input/Output, ,Your program's input and output}.
1490 @emph{Warning:} While input and output redirection work, you cannot use
1491 pipes to pass the output of the program you are debugging to another
1492 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1497 When you issue the @code{run} command, your program begins to execute
1498 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1499 of how to arrange for your program to stop. Once your program has
1500 stopped, you may calls functions in your program, using the @code{print}
1501 or @code{call} commands. @xref{Data, ,Examining Data}.
1503 If the modification time of your symbol file has changed since the
1504 last time @value{GDBN} read its symbols, @value{GDBN} will discard its symbol table and
1505 re-read it. When it does this, @value{GDBN} tries to retain your current
1510 @section Your program's arguments
1512 @cindex arguments (to your program)
1513 The arguments to your program can be specified by the arguments of the
1514 @code{run} command. They are passed to a shell, which expands wildcard
1515 characters and performs redirection of I/O, and thence to your program.
1516 Your @code{SHELL} environment variable (if it exists) specifies what
1517 shell @value{GDBN} if you do not define @code{SHELL}, @value{GDBN} uses
1520 @code{run} with no arguments uses the same arguments used by the previous
1521 @code{run}, or those set by the @code{set args} command.
1526 Specify the arguments to be used the next time your program is run. If
1527 @code{set args} has no arguments, @code{run} will execute your program
1528 with no arguments. Once you have run your program with arguments,
1529 using @code{set args} before the next @code{run} is the only way to run
1530 it again without arguments.
1534 Show the arguments to give your program when it is started.
1538 @section Your program's environment
1540 @cindex environment (of your program)
1541 The @dfn{environment} consists of a set of environment variables and
1542 their values. Environment variables conventionally record such things as
1543 your user name, your home directory, your terminal type, and your search
1544 path for programs to run. Usually you set up environment variables with
1545 the shell and they are inherited by all the other programs you run. When
1546 debugging, it can be useful to try running your program with a modified
1547 environment without having to start @value{GDBN} over again.
1550 @item path @var{directory}
1552 Add @var{directory} to the front of the @code{PATH} environment variable
1553 (the search path for executables), for both @value{GDBN} and your program.
1554 You may specify several directory names, separated by @samp{:} or
1555 whitespace. If @var{directory} is already in the path, it is moved to
1556 the front, so it will be searched sooner.
1558 You can use the string @samp{$cwd} to refer to whatever is the current
1559 working directory at the time @value{GDBN} searches the path. If you
1560 use @samp{.} instead, it refers to the directory where you executed the
1561 @code{path} command. @value{GDBN} replaces @samp{.} in the
1562 @var{directory} argument (with the current path) before adding
1563 @var{directory} to the search path.
1564 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1565 @c document that, since repeating it would be a no-op.
1569 Display the list of search paths for executables (the @code{PATH}
1570 environment variable).
1572 @item show environment @r{[}@var{varname}@r{]}
1573 @kindex show environment
1574 Print the value of environment variable @var{varname} to be given to
1575 your program when it starts. If you do not supply @var{varname},
1576 print the names and values of all environment variables to be given to
1577 your program. You can abbreviate @code{environment} as @code{env}.
1579 @item set environment @var{varname} @r{[}=@r{]} @var{value}
1580 @kindex set environment
1581 Set environment variable @var{varname} to @var{value}. The value
1582 changes for your program only, not for @value{GDBN} itself. @var{value} may
1583 be any string; the values of environment variables are just strings, and
1584 any interpretation is supplied by your program itself. The @var{value}
1585 parameter is optional; if it is eliminated, the variable is set to a
1587 @c "any string" here does not include leading, trailing
1588 @c blanks. Gnu asks: does anyone care?
1590 For example, this command:
1597 tells a Unix program, when subsequently run, that its user is named
1598 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1599 are not actually required.)
1601 @item unset environment @var{varname}
1602 @kindex unset environment
1603 Remove variable @var{varname} from the environment to be passed to your
1604 program. This is different from @samp{set env @var{varname} =};
1605 @code{unset environment} removes the variable from the environment,
1606 rather than assigning it an empty value.
1609 @emph{Warning:} @value{GDBN} runs your program using the shell indicated
1610 by your @code{SHELL} environment variable if it exists (or
1611 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1612 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1613 @file{.bashrc} for BASH---any variables you set in that file will affect
1614 your program. You may wish to move setting of environment variables to
1615 files that are only run when you sign on, such as @file{.login} or
1618 @node Working Directory
1619 @section Your program's working directory
1621 @cindex working directory (of your program)
1622 Each time you start your program with @code{run}, it inherits its
1623 working directory from the current working directory of @value{GDBN}.
1624 The @value{GDBN} working directory is initially whatever it inherited
1625 from its parent process (typically the shell), but you can specify a new
1626 working directory in @value{GDBN} with the @code{cd} command.
1628 The @value{GDBN} working directory also serves as a default for the commands
1629 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1633 @item cd @var{directory}
1635 Set the @value{GDBN} working directory to @var{directory}.
1639 Print the @value{GDBN} working directory.
1643 @section Your program's input and output
1648 By default, the program you run under @value{GDBN} does input and output to
1649 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal to
1650 its own terminal modes to interact with you, but it records the terminal
1651 modes your program was using and switches back to them when you continue
1652 running your program.
1656 @kindex info terminal
1657 Displays information recorded by @value{GDBN} about the terminal modes your
1661 You can redirect your program's input and/or output using shell
1662 redirection with the @code{run} command. For example,
1669 starts your program, diverting its output to the file @file{outfile}.
1672 @cindex controlling terminal
1673 Another way to specify where your program should do input and output is
1674 with the @code{tty} command. This command accepts a file name as
1675 argument, and causes this file to be the default for future @code{run}
1676 commands. It also resets the controlling terminal for the child
1677 process, for future @code{run} commands. For example,
1684 directs that processes started with subsequent @code{run} commands
1685 default to do input and output on the terminal @file{/dev/ttyb} and have
1686 that as their controlling terminal.
1688 An explicit redirection in @code{run} overrides the @code{tty} command's
1689 effect on the input/output device, but not its effect on the controlling
1692 When you use the @code{tty} command or redirect input in the @code{run}
1693 command, only the input @emph{for your program} is affected. The input
1694 for @value{GDBN} still comes from your terminal.
1697 @section Debugging an already-running process
1702 @item attach @var{process-id}
1703 This command attaches to a running process---one that was started
1704 outside @value{GDBN}. (@code{info files} will show your active
1705 targets.) The command takes as argument a process ID. The usual way to
1706 find out the process-id of a Unix process is with the @code{ps} utility,
1707 or with the @samp{jobs -l} shell command.
1709 @code{attach} will not repeat if you press @key{RET} a second time after
1710 executing the command.
1713 To use @code{attach}, your program must be running in an environment
1714 which supports processes; for example, @code{attach} does not work for
1715 programs on bare-board targets that lack an operating system. You must
1716 also have permission to send the process a signal.
1718 When using @code{attach}, you should first use the @code{file} command
1719 to specify the program running in the process and load its symbol table.
1720 @xref{Files, ,Commands to Specify Files}.
1722 The first thing @value{GDBN} does after arranging to debug the specified
1723 process is to stop it. You can examine and modify an attached process
1724 with all the @value{GDBN} commands that are ordinarily available when you start
1725 processes with @code{run}. You can insert breakpoints; you can step and
1726 continue; you can modify storage. If you would rather the process
1727 continue running, you may use the @code{continue} command after
1728 attaching @value{GDBN} to the process.
1733 When you have finished debugging the attached process, you can use the
1734 @code{detach} command to release it from @value{GDBN} control. Detaching
1735 the process continues its execution. After the @code{detach} command,
1736 that process and @value{GDBN} become completely independent once more, and you
1737 are ready to @code{attach} another process or start one with @code{run}.
1738 @code{detach} will not repeat if you press @key{RET} again after
1739 executing the command.
1742 If you exit @value{GDBN} or use the @code{run} command while you have an attached
1743 process, you kill that process. By default, you will be asked for
1744 confirmation if you try to do either of these things; you can control
1745 whether or not you need to confirm by using the @code{set confirm} command
1746 (@pxref{Messages/Warnings, ,Optional warnings and messages}).
1750 @section Killing the child process
1755 Kill the child process in which your program is running under @value{GDBN}.
1758 This command is useful if you wish to debug a core dump instead of a
1759 running process. @value{GDBN} ignores any core dump file while your program
1763 On some operating systems, a program cannot be executed outside @value{GDBN}
1764 while you have breakpoints set on it inside @value{GDBN}. You can use the
1765 @code{kill} command in this situation to permit running your program
1766 outside the debugger.
1768 The @code{kill} command is also useful if you wish to recompile and
1769 relink your program, since on many systems it is impossible to modify an
1770 executable file while it is running in a process. In this case, when you
1771 next type @code{run}, @value{GDBN} will notice that the file has changed, and
1772 will re-read the symbol table (while trying to preserve your current
1773 breakpoint settings).
1775 @node Process Information
1776 @section Additional process information
1779 @cindex process image
1780 Some operating systems provide a facility called @samp{/proc} that can
1781 be used to examine the image of a running process using file-system
1782 subroutines. If @value{GDBN} is configured for an operating system with this
1783 facility, the command @code{info proc} is available to report on several
1784 kinds of information about the process running your program.
1789 Summarize available information about the process.
1791 @item info proc mappings
1792 @kindex info proc mappings
1793 Report on the address ranges accessible in the program, with information
1794 on whether your program may read, write, or execute each range.
1796 @item info proc times
1797 @kindex info proc times
1798 Starting time, user CPU time, and system CPU time for your program and
1802 @kindex info proc id
1803 Report on the process IDs related to your program: its own process ID,
1804 the ID of its parent, the process group ID, and the session ID.
1806 @item info proc status
1807 @kindex info proc status
1808 General information on the state of the process. If the process is
1809 stopped, this report includes the reason for stopping, and any signal
1813 Show all the above information about the process.
1818 @chapter Stopping and Continuing
1820 The principal purposes of using a debugger are so that you can stop your
1821 program before it terminates; or so that, if your program runs into
1822 trouble, you can investigate and find out why.
1824 Inside @value{GDBN}, your program may stop for any of several reasons, such
1829 a breakpoint, or reaching a new line after a @value{GDBN}
1830 command such as @code{step}. You may then examine and change
1831 variables, set new breakpoints or remove old ones, and then continue
1832 execution. Usually, the messages shown by @value{GDBN} provide ample
1833 explanation of the status of your program---but you can also explicitly
1834 request this information at any time.
1838 @kindex info program
1839 Display information about the status of your program: whether it is
1849 * Breakpoints:: Breakpoints, watchpoints, and exceptions
1852 * Breakpoints:: Breakpoints and watchpoints
1854 @c Remnant makeinfo bug requires blank line after *successful* end-if in menu:
1856 * Continuing and Stepping:: Resuming execution
1862 @c makeinfo node-defaulting requires adjacency of @node and sectioning cmds
1863 @c ...hence distribute @node Breakpoints over two possible @if expansions.
1867 @section Breakpoints, watchpoints, and exceptions
1871 @section Breakpoints and watchpoints
1875 A @dfn{breakpoint} makes your program stop whenever a certain point in
1876 the program is reached. For each breakpoint, you can add various
1877 conditions to control in finer detail whether your program will stop.
1878 You can set breakpoints with the @code{break} command and its variants
1879 (@pxref{Set Breaks, ,Setting breakpoints}), to specify the place where
1880 your program should stop by line number, function name or exact address
1883 In languages with exception handling (such as GNU C++), you can also set
1884 breakpoints where an exception is raised (@pxref{Exception Handling,
1885 ,Breakpoints and exceptions}).
1889 @cindex memory tracing
1890 @cindex breakpoint on memory address
1891 @cindex breakpoint on variable modification
1892 A @dfn{watchpoint} is a special breakpoint that stops your program
1893 when the value of an expression changes. You must use a different
1894 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
1895 watchpoints}), but aside from that, you can manage a watchpoint like
1896 any other breakpoint: you enable, disable, and delete both breakpoints
1897 and watchpoints using the same commands.
1899 You can arrange to have values from your program displayed automatically
1900 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,
1901 ,Automatic display}.
1903 @cindex breakpoint numbers
1904 @cindex numbers for breakpoints
1905 @value{GDBN} assigns a number to each breakpoint or watchpoint when you
1906 create it; these numbers are successive integers starting with one. In
1907 many of the commands for controlling various features of breakpoints you
1908 use the breakpoint number to say which breakpoint you want to change.
1909 Each breakpoint may be @dfn{enabled} or @dfn{disabled}; if disabled, it has
1910 no effect on your program until you enable it again.
1913 * Set Breaks:: Setting breakpoints
1914 * Set Watchpoints:: Setting watchpoints
1916 * Exception Handling:: Breakpoints and exceptions
1919 * Delete Breaks:: Deleting breakpoints
1920 * Disabling:: Disabling breakpoints
1921 * Conditions:: Break conditions
1922 * Break Commands:: Breakpoint command lists
1924 * Breakpoint Menus:: Breakpoint menus
1927 * Error in Breakpoints:: ``Cannot insert breakpoints''
1932 @subsection Setting breakpoints
1934 @c FIXME LMB what does GDB do if no code on line of breakpt?
1935 @c consider in particular declaration with/without initialization.
1937 @c FIXME 2 is there stuff on this already? break at fun start, already init?
1942 @cindex latest breakpoint
1943 Breakpoints are set with the @code{break} command (abbreviated
1944 @code{b}). The debugger convenience variable @samp{$bpnum} records the
1945 number of the beakpoint you've set most recently; see @ref{Convenience
1946 Vars,, Convenience variables}, for a discussion of what you can do with
1947 convenience variables.
1949 You have several ways to say where the breakpoint should go.
1952 @item break @var{function}
1953 Set a breakpoint at entry to function @var{function}.
1955 When using source languages that permit overloading of symbols, such as
1956 C++, @var{function} may refer to more than one possible place to break.
1957 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
1960 @item break +@var{offset}
1961 @itemx break -@var{offset}
1962 Set a breakpoint some number of lines forward or back from the position
1963 at which execution stopped in the currently selected frame.
1965 @item break @var{linenum}
1966 Set a breakpoint at line @var{linenum} in the current source file.
1967 That file is the last file whose source text was printed. This
1968 breakpoint will stop your program just before it executes any of the
1971 @item break @var{filename}:@var{linenum}
1972 Set a breakpoint at line @var{linenum} in source file @var{filename}.
1974 @item break @var{filename}:@var{function}
1975 Set a breakpoint at entry to function @var{function} found in file
1976 @var{filename}. Specifying a file name as well as a function name is
1977 superfluous except when multiple files contain similarly named
1980 @item break *@var{address}
1981 Set a breakpoint at address @var{address}. You can use this to set
1982 breakpoints in parts of your program which do not have debugging
1983 information or source files.
1986 When called without any arguments, @code{break} sets a breakpoint at
1987 the next instruction to be executed in the selected stack frame
1988 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
1989 innermost, this will cause your program to stop as soon as control
1990 returns to that frame. This is similar to the effect of a
1991 @code{finish} command in the frame inside the selected frame---except
1992 that @code{finish} does not leave an active breakpoint. If you use
1993 @code{break} without an argument in the innermost frame, @value{GDBN} will stop
1994 the next time it reaches the current location; this may be useful
1997 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
1998 least one instruction has been executed. If it did not do this, you
1999 would be unable to proceed past a breakpoint without first disabling the
2000 breakpoint. This rule applies whether or not the breakpoint already
2001 existed when your program stopped.
2003 @item break @dots{} if @var{cond}
2004 Set a breakpoint with condition @var{cond}; evaluate the expression
2005 @var{cond} each time the breakpoint is reached, and stop only if the
2006 value is nonzero---that is, if @var{cond} evaluates as true.
2007 @samp{@dots{}} stands for one of the possible arguments described
2008 above (or no argument) specifying where to break. @xref{Conditions,
2009 ,Break conditions}, for more information on breakpoint conditions.
2011 @item tbreak @var{args}
2013 Set a breakpoint enabled only for one stop. @var{args} are the
2014 same as for the @code{break} command, and the breakpoint is set in the same
2015 way, but the breakpoint is automatically disabled after the first time your
2016 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2018 @item rbreak @var{regex}
2020 @cindex regular expression
2021 @c FIXME what kind of regexp?
2022 Set breakpoints on all functions matching the regular expression
2023 @var{regex}. This command
2024 sets an unconditional breakpoint on all matches, printing a list of all
2025 breakpoints it set. Once these breakpoints are set, they are treated
2026 just like the breakpoints set with the @code{break} command. They can
2027 be deleted, disabled, made conditional, etc., in the standard ways.
2030 When debugging C++ programs, @code{rbreak} is useful for setting
2031 breakpoints on overloaded functions that are not members of any special
2035 @kindex info breakpoints
2036 @cindex @code{$_} and @code{info breakpoints}
2037 @item info breakpoints @r{[}@var{n}@r{]}
2038 @itemx info break @r{[}@var{n}@r{]}
2039 @itemx info watchpoints @r{[}@var{n}@r{]}
2040 Print a table of all breakpoints and watchpoints set and not
2041 deleted, with the following columns for each breakpoint:
2044 @item Breakpoint Numbers
2046 Breakpoint or watchpoint.
2048 Whether the breakpoint is marked to be disabled or deleted when hit.
2049 @item Enabled or Disabled
2050 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2051 that are not enabled.
2053 Where the breakpoint is in your program, as a memory address
2055 Where the breakpoint is in the source for your program, as a file and
2060 If a breakpoint is conditional, @code{info break} shows the condition on
2061 the line following the affected breakpoint; breakpoint commands, if any,
2062 are listed after that.
2065 @code{info break} with a breakpoint
2066 number @var{n} as argument lists only that breakpoint. The
2067 convenience variable @code{$_} and the default examining-address for
2068 the @code{x} command are set to the address of the last breakpoint
2069 listed (@pxref{Memory, ,Examining memory}).
2072 @value{GDBN} allows you to set any number of breakpoints at the same place in
2073 your program. There is nothing silly or meaningless about this. When
2074 the breakpoints are conditional, this is even useful
2075 (@pxref{Conditions, ,Break conditions}).
2077 @cindex negative breakpoint numbers
2078 @cindex internal @value{GDBN} breakpoints
2079 @value{GDBN} itself sometimes sets breakpoints in your program for special
2080 purposes, such as proper handling of @code{longjmp} (in C programs).
2081 These internal breakpoints are assigned negative numbers, starting with
2082 @code{-1}; @samp{info breakpoints} does not display them.
2084 You can see these breakpoints with the @value{GDBN} maintenance command
2085 @samp{maint info breakpoints}.
2088 @kindex maint info breakpoints
2089 @item maint info breakpoints
2090 Using the same format as @samp{info breakpoints}, display both the
2091 breakpoints you've set explicitly, and those @value{GDBN} is using for
2092 internal purposes. Internal breakpoints are shown with negative
2093 breakpoint numbers. The type column identifies what kind of breakpoint
2098 Normal, explicitly set breakpoint.
2101 Normal, explicitly set watchpoint.
2104 Internal breakpoint, used to handle correctly stepping through
2105 @code{longjmp} calls.
2107 @item longjmp resume
2108 Internal breakpoint at the target of a @code{longjmp}.
2111 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2114 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2120 @node Set Watchpoints
2121 @subsection Setting watchpoints
2122 @cindex setting watchpoints
2124 You can use a watchpoint to stop execution whenever the value of an
2125 expression changes, without having to predict a particular place
2126 where this may happen.
2128 Watchpoints currently execute two orders of magnitude more slowly than
2129 other breakpoints, but this can well be worth it to catch errors where
2130 you have no clue what part of your program is the culprit. Some
2131 processors provide special hardware to support watchpoint evaluation; future
2132 releases of @value{GDBN} will use such hardware if it is available.
2136 @item watch @var{expr}
2137 Set a watchpoint for an expression.
2139 @kindex info watchpoints
2140 @item info watchpoints
2141 This command prints a list of watchpoints and breakpoints; it is the
2142 same as @code{info break}.
2146 @node Exception Handling
2147 @subsection Breakpoints and exceptions
2148 @cindex exception handlers
2150 Some languages, such as GNU C++, implement exception handling. You can
2151 use @value{GDBN} to examine what caused your program to raise an exception,
2152 and to list the exceptions your program is prepared to handle at a
2153 given point in time.
2156 @item catch @var{exceptions}
2158 You can set breakpoints at active exception handlers by using the
2159 @code{catch} command. @var{exceptions} is a list of names of exceptions
2163 You can use @code{info catch} to list active exception handlers.
2164 @xref{Frame Info, ,Information about a frame}.
2166 There are currently some limitations to exception handling in @value{GDBN}.
2167 These will be corrected in a future release.
2171 If you call a function interactively, @value{GDBN} normally returns
2172 control to you when the function has finished executing. If the call
2173 raises an exception, however, the call may bypass the mechanism that
2174 returns control to you and cause your program to simply continue
2175 running until it hits a breakpoint, catches a signal that @value{GDBN} is
2176 listening for, or exits.
2178 You cannot raise an exception interactively.
2180 You cannot interactively install an exception handler.
2183 @cindex raise exceptions
2184 Sometimes @code{catch} is not the best way to debug exception handling:
2185 if you need to know exactly where an exception is raised, it is better to
2186 stop @emph{before} the exception handler is called, since that way you
2187 can see the stack before any unwinding takes place. If you set a
2188 breakpoint in an exception handler instead, it may not be easy to find
2189 out where the exception was raised.
2191 To stop just before an exception handler is called, you need some
2192 knowledge of the implementation. In the case of GNU C++, exceptions are
2193 raised by calling a library function named @code{__raise_exception}
2194 which has the following ANSI C interface:
2197 /* @var{addr} is where the exception identifier is stored.
2198 ID is the exception identifier. */
2199 void __raise_exception (void **@var{addr}, void *@var{id});
2203 To make the debugger catch all exceptions before any stack
2204 unwinding takes place, set a breakpoint on @code{__raise_exception}
2205 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2207 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2208 that depends on the value of @var{id}, you can stop your program when
2209 a specific exception is raised. You can use multiple conditional
2210 breakpoints to stop your program when any of a number of exceptions are
2215 @subsection Deleting breakpoints
2217 @cindex clearing breakpoints, watchpoints
2218 @cindex deleting breakpoints, watchpoints
2219 It is often necessary to eliminate a breakpoint or watchpoint once it
2220 has done its job and you no longer want your program to stop there. This
2221 is called @dfn{deleting} the breakpoint. A breakpoint that has been
2222 deleted no longer exists; it is forgotten.
2224 With the @code{clear} command you can delete breakpoints according to
2225 where they are in your program. With the @code{delete} command you can
2226 delete individual breakpoints or watchpoints by specifying their
2229 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2230 automatically ignores breakpoints on the first instruction to be executed
2231 when you continue execution without changing the execution address.
2236 Delete any breakpoints at the next instruction to be executed in the
2237 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2238 the innermost frame is selected, this is a good way to delete a
2239 breakpoint where your program just stopped.
2241 @item clear @var{function}
2242 @itemx clear @var{filename}:@var{function}
2243 Delete any breakpoints set at entry to the function @var{function}.
2245 @item clear @var{linenum}
2246 @itemx clear @var{filename}:@var{linenum}
2247 Delete any breakpoints set at or within the code of the specified line.
2249 @item delete @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2250 @cindex delete breakpoints
2253 Delete the breakpoints or watchpoints of the numbers specified as
2254 arguments. If no argument is specified, delete all breakpoints (@value{GDBN}
2255 asks confirmation, unless you have @code{set confirm off}). You
2256 can abbreviate this command as @code{d}.
2260 @subsection Disabling breakpoints
2262 @cindex disabled breakpoints
2263 @cindex enabled breakpoints
2264 Rather than deleting a breakpoint or watchpoint, you might prefer to
2265 @dfn{disable} it. This makes the breakpoint inoperative as if it had
2266 been deleted, but remembers the information on the breakpoint so that
2267 you can @dfn{enable} it again later.
2269 You disable and enable breakpoints and watchpoints with the
2270 @code{enable} and @code{disable} commands, optionally specifying one or
2271 more breakpoint numbers as arguments. Use @code{info break} or
2272 @code{info watch} to print a list of breakpoints or watchpoints if you
2273 do not know which numbers to use.
2275 A breakpoint or watchpoint can have any of four different states of
2280 Enabled. The breakpoint will stop your program. A breakpoint set
2281 with the @code{break} command starts out in this state.
2283 Disabled. The breakpoint has no effect on your program.
2285 Enabled once. The breakpoint will stop your program, but
2286 when it does so it will become disabled. A breakpoint set
2287 with the @code{tbreak} command starts out in this state.
2289 Enabled for deletion. The breakpoint will stop your program, but
2290 immediately after it does so it will be deleted permanently.
2293 You can use the following commands to enable or disable breakpoints and
2297 @item disable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2298 @kindex disable breakpoints
2301 Disable the specified breakpoints---or all breakpoints, if none are
2302 listed. A disabled breakpoint has no effect but is not forgotten. All
2303 options such as ignore-counts, conditions and commands are remembered in
2304 case the breakpoint is enabled again later. You may abbreviate
2305 @code{disable} as @code{dis}.
2307 @item enable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2308 @kindex enable breakpoints
2310 Enable the specified breakpoints (or all defined breakpoints). They
2311 become effective once again in stopping your program.
2313 @item enable @r{[}breakpoints@r{]} once @var{bnums}@dots{}
2314 Enable the specified breakpoints temporarily. Each will be disabled
2315 again the next time it stops your program.
2317 @item enable @r{[}breakpoints@r{]} delete @var{bnums}@dots{}
2318 Enable the specified breakpoints to work once and then die. Each of
2319 the breakpoints will be deleted the next time it stops your program.
2322 Save for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2323 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2324 subsequently, they become disabled or enabled only when you use one of
2325 the commands above. (The command @code{until} can set and delete a
2326 breakpoint of its own, but it will not change the state of your other
2327 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2331 @subsection Break conditions
2332 @cindex conditional breakpoints
2333 @cindex breakpoint conditions
2335 @c FIXME what is scope of break condition expr? Context where wanted?
2336 @c in particular for a watchpoint?
2337 The simplest sort of breakpoint breaks every time your program reaches a
2338 specified place. You can also specify a @dfn{condition} for a
2339 breakpoint. A condition is just a Boolean expression in your
2340 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2341 a condition evaluates the expression each time your program reaches it,
2342 and your program stops only if the condition is @emph{true}.
2344 This is the converse of using assertions for program validation; in that
2345 situation, you want to stop when the assertion is violated---that is,
2346 when the condition is false. In C, if you want to test an assertion expressed
2347 by the condition @var{assert}, you should set the condition
2348 @samp{! @var{assert}} on the appropriate breakpoint.
2350 Conditions are also accepted for watchpoints; you may not need them,
2351 since a watchpoint is inspecting the value of an expression anyhow---but
2352 it might be simpler, say, to just set a watchpoint on a variable name,
2353 and specify a condition that tests whether the new value is an interesting
2356 Break conditions can have side effects, and may even call functions in
2357 your program. This can be useful, for example, to activate functions
2358 that log program progress, or to use your own print functions to
2359 format special data structures. The effects are completely predictable
2360 unless there is another enabled breakpoint at the same address. (In
2361 that case, @value{GDBN} might see the other breakpoint first and stop your
2362 program without checking the condition of this one.) Note that
2363 breakpoint commands are usually more convenient and flexible for the
2364 purpose of performing side effects when a breakpoint is reached
2365 (@pxref{Break Commands, ,Breakpoint command lists}).
2367 Break conditions can be specified when a breakpoint is set, by using
2368 @samp{if} in the arguments to the @code{break} command. @xref{Set
2369 Breaks, ,Setting breakpoints}. They can also be changed at any time
2370 with the @code{condition} command. The @code{watch} command does not
2371 recognize the @code{if} keyword; @code{condition} is the only way to
2372 impose a further condition on a watchpoint.
2375 @item condition @var{bnum} @var{expression}
2377 Specify @var{expression} as the break condition for breakpoint or
2378 watchpoint number @var{bnum}. From now on, this breakpoint will stop
2379 your program only if the value of @var{expression} is true (nonzero, in
2380 C). When you use @code{condition}, @value{GDBN} checks @var{expression}
2381 immediately for syntactic correctness, and to determine whether symbols
2382 in it have referents in the context of your breakpoint.
2383 @c FIXME so what does GDB do if there is no referent? Moreover, what
2384 @c about watchpoints?
2386 not actually evaluate @var{expression} at the time the @code{condition}
2387 command is given, however. @xref{Expressions, ,Expressions}.
2389 @item condition @var{bnum}
2390 Remove the condition from breakpoint number @var{bnum}. It becomes
2391 an ordinary unconditional breakpoint.
2394 @cindex ignore count (of breakpoint)
2395 A special case of a breakpoint condition is to stop only when the
2396 breakpoint has been reached a certain number of times. This is so
2397 useful that there is a special way to do it, using the @dfn{ignore
2398 count} of the breakpoint. Every breakpoint has an ignore count, which
2399 is an integer. Most of the time, the ignore count is zero, and
2400 therefore has no effect. But if your program reaches a breakpoint whose
2401 ignore count is positive, then instead of stopping, it just decrements
2402 the ignore count by one and continues. As a result, if the ignore count
2403 value is @var{n}, the breakpoint will not stop the next @var{n} times it
2407 @item ignore @var{bnum} @var{count}
2409 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
2410 The next @var{count} times the breakpoint is reached, your program's
2411 execution will not stop; other than to decrement the ignore count, @value{GDBN}
2414 To make the breakpoint stop the next time it is reached, specify
2417 When you use @code{continue} to resume execution of your program from a
2418 breakpoint, you can specify an ignore count directly as an argument to
2419 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
2420 Stepping,,Continuing and stepping}.
2422 If a breakpoint has a positive ignore count and a condition, the condition
2423 is not checked. Once the ignore count reaches zero, the condition will
2426 You could achieve the effect of the ignore count with a condition such
2427 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
2428 is decremented each time. @xref{Convenience Vars, ,Convenience
2432 @node Break Commands
2433 @subsection Breakpoint command lists
2435 @cindex breakpoint commands
2436 You can give any breakpoint (or watchpoint) a series of commands to
2437 execute when your program stops due to that breakpoint. For example, you
2438 might want to print the values of certain expressions, or enable other
2442 @item commands @r{[}@var{bnum}@r{]}
2443 @itemx @dots{} @var{command-list} @dots{}
2447 Specify a list of commands for breakpoint number @var{bnum}. The commands
2448 themselves appear on the following lines. Type a line containing just
2449 @code{end} to terminate the commands.
2451 To remove all commands from a breakpoint, type @code{commands} and
2452 follow it immediately with @code{end}; that is, give no commands.
2454 With no @var{bnum} argument, @code{commands} refers to the last
2455 breakpoint or watchpoint set (not to the breakpoint most recently
2459 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
2460 disabled within a @var{command-list}.
2462 You can use breakpoint commands to start your program up again. Simply
2463 use the @code{continue} command, or @code{step}, or any other command
2464 that resumes execution.
2466 Any other commands in the command list, after a command that resumes
2467 execution, are ignored. This is because any time you resume execution
2468 (even with a simple @code{next} or @code{step}), you may encounter
2469 another breakpoint---which could have its own command list, leading to
2470 ambiguities about which list to execute.
2473 If the first command you specify in a command list is @code{silent}, the
2474 usual message about stopping at a breakpoint is not printed. This may
2475 be desirable for breakpoints that are to print a specific message and
2476 then continue. If none of the remaining commands print anything, you
2477 will see no sign that the breakpoint was reached. @code{silent} is
2478 meaningful only at the beginning of a breakpoint command list.
2480 The commands @code{echo}, @code{output}, and @code{printf} allow you to
2481 print precisely controlled output, and are often useful in silent
2482 breakpoints. @xref{Output, ,Commands for controlled output}.
2484 For example, here is how you could use breakpoint commands to print the
2485 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
2491 printf "x is %d\n",x
2496 One application for breakpoint commands is to compensate for one bug so
2497 you can test for another. Put a breakpoint just after the erroneous line
2498 of code, give it a condition to detect the case in which something
2499 erroneous has been done, and give it commands to assign correct values
2500 to any variables that need them. End with the @code{continue} command
2501 so that your program does not stop, and start with the @code{silent}
2502 command so that no output is produced. Here is an example:
2514 @c Do we need to mention this at all? I am sort of tempted to mention
2515 @c it in case people are used to seeing this section of the manual. But
2516 @c for new users it is an annoyance--it documents something which isn't
2517 @c there. -kingdon, 6 Jul 93
2518 Previous versions of @value{GDBN} (4.9 and earlier) would flush pending
2519 input when executing breakpoint commands, if your program used raw mode
2520 for the terminal. This is no longer true.
2523 @c I don't think this is true any longer, now that only readline
2524 @c switches to or from raw mode. In any event, it is a (relatively
2525 @c easily fixable) GDB bug if it switches to or from raw mode except
2526 @c when it has to in order to read input from the terminal. kingdon -6 Jul 93.
2527 One deficiency in the operation of automatically continuing breakpoints
2528 under Unix appears when your program uses raw mode for the terminal.
2529 @value{GDBN} switches back to its own terminal modes (not raw) before executing
2530 commands, and then must switch back to raw mode when your program is
2531 continued. This causes any pending terminal input to be lost.
2532 @c FIXME: revisit below when GNU sys avail.
2533 @c In the GNU system, this will be fixed by changing the behavior of
2536 Under Unix, you can get around this problem by writing actions into
2537 the breakpoint condition rather than in commands. For example,
2540 condition 5 (x = y + 4), 0
2544 specifies a condition expression (@pxref{Expressions, ,Expressions}) that will
2545 change @code{x} as needed, then always have the value zero so your
2546 program will not stop. No input is lost here, because @value{GDBN} evaluates
2547 break conditions without changing the terminal modes. When you want
2548 to have nontrivial conditions for performing the side effects, the
2549 operators @samp{&&}, @samp{||} and @samp{?@dots{}:} may be useful.
2553 @node Breakpoint Menus
2554 @subsection Breakpoint menus
2556 @cindex symbol overloading
2558 Some programming languages (notably C++) permit a single function name
2559 to be defined several times, for application in different contexts.
2560 This is called @dfn{overloading}. When a function name is overloaded,
2561 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
2562 a breakpoint. If you realize this will be a problem, you can use
2563 something like @samp{break @var{function}(@var{types})} to specify which
2564 particular version of the function you want. Otherwise, @value{GDBN} offers
2565 you a menu of numbered choices for different possible breakpoints, and
2566 waits for your selection with the prompt @samp{>}. The first two
2567 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
2568 sets a breakpoint at each definition of @var{function}, and typing
2569 @kbd{0} aborts the @code{break} command without setting any new
2572 For example, the following session excerpt shows an attempt to set a
2573 breakpoint at the overloaded symbol @code{String::after}.
2574 We choose three particular definitions of that function name:
2576 @c FIXME! This is likely to change to show arg type lists, at least
2578 (@value{GDBP}) b String::after
2581 [2] file:String.cc; line number:867
2582 [3] file:String.cc; line number:860
2583 [4] file:String.cc; line number:875
2584 [5] file:String.cc; line number:853
2585 [6] file:String.cc; line number:846
2586 [7] file:String.cc; line number:735
2588 Breakpoint 1 at 0xb26c: file String.cc, line 867.
2589 Breakpoint 2 at 0xb344: file String.cc, line 875.
2590 Breakpoint 3 at 0xafcc: file String.cc, line 846.
2591 Multiple breakpoints were set.
2592 Use the "delete" command to delete unwanted
2599 @node Error in Breakpoints
2600 @subsection ``Cannot insert breakpoints''
2602 @c FIXME: "cannot insert breakpoints" error, v unclear.
2603 @c Q in pending mail to Gilmore. ---pesch@cygnus.com, 26mar91
2604 @c some light may be shed by looking at instances of
2605 @c ONE_PROCESS_WRITETEXT. But error message seems possible otherwise
2606 @c too. pesch, 20sep91
2607 Under some operating systems, breakpoints cannot be used in a program if
2608 any other process is running that program. In this situation,
2609 attempting to run or continue a program with a breakpoint causes @value{GDBN}
2610 to stop the other process.
2612 When this happens, you have three ways to proceed:
2616 Remove or disable the breakpoints, then continue.
2619 Suspend @value{GDBN}, and copy the file containing your program to a new name.
2620 Resume @value{GDBN} and use the @code{exec-file} command to specify that @value{GDBN}
2621 should run your program under that name. Then start your program again.
2623 @c FIXME: RMS commented here "Show example". Maybe when someone
2624 @c explains the first FIXME: in this section...
2627 Relink your program so that the text segment is nonsharable, using the
2628 linker option @samp{-N}. The operating system limitation may not apply
2629 to nonsharable executables.
2633 @node Continuing and Stepping
2634 @section Continuing and stepping
2638 @cindex resuming execution
2639 @dfn{Continuing} means resuming program execution until your program
2640 completes normally. In contrast, @dfn{stepping} means executing just
2641 one more ``step'' of your program, where ``step'' may mean either one
2642 line of source code, or one machine instruction (depending on what
2643 particular command you use). Either when continuing
2644 or when stepping, your program may stop even sooner, due to
2649 a breakpoint or to a signal. (If due to a signal, you may want to use
2650 @code{handle}, or use @samp{signal 0} to resume execution.
2651 @xref{Signals, ,Signals}.)
2655 @item continue @r{[}@var{count}@r{]}
2656 @itemx c @r{[}@var{count}@r{]}
2657 @itemx fg @r{[}@var{count}@r{]}
2661 Resume program execution, at the address where your program last
2662 stopped; any breakpoints set at that address are bypassed. The optional
2663 argument @var{count} means to set the ignore count of a breakpoint which
2664 you are stopped at to @var{count} @minus{} 1, just like the @code{ignore}
2665 command (@pxref{Conditions, ,Break conditions}). This means that the
2666 program does not stop at that breakpoint until the @var{count}th time
2669 The argument @var{count} is meaningful only when your program
2670 stopped due to a breakpoint. At other times, the argument to
2671 @code{continue} is ignored.
2673 The synonyms @code{c} and @code{fg} are provided purely for convenience,
2674 and have exactly the same behavior as @code{continue}.
2677 To resume execution at a different place, you can use @code{return}
2678 (@pxref{Returning, ,Returning from a function}) to go back to the
2679 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
2680 different address}) to go to an arbitrary location in your program.
2682 A typical technique for using stepping is to set a breakpoint
2684 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions})
2687 (@pxref{Breakpoints, ,Breakpoints and watchpoints})
2690 beginning of the function or the section of your program where a
2691 problem is believed to lie, run your program until it stops at that
2692 breakpoint, and then step through the suspect area, examining the
2693 variables that are interesting, until you see the problem happen.
2699 Continue running your program until control reaches a different source
2700 line, then stop it and return control to @value{GDBN}. This command is
2701 abbreviated @code{s}.
2704 @emph{Warning:} If you use the @code{step} command while control is
2705 within a function that was compiled without debugging information,
2706 execution proceeds until control reaches a function that does have
2707 debugging information.
2710 @item step @var{count}
2711 Continue running as in @code{step}, but do so @var{count} times. If a
2712 breakpoint is reached,
2714 or a signal not related to stepping occurs before @var{count} steps,
2716 stepping stops right away.
2718 @item next @r{[}@var{count}@r{]}
2721 Continue to the next source line in the current (innermost) stack frame.
2722 Similar to @code{step}, but any function calls appearing within the line
2723 of code are executed without stopping. Execution stops when control
2724 reaches a different line of code at the stack level which was executing
2725 when the @code{next} command was given. This command is abbreviated
2728 An argument @var{count} is a repeat count, as for @code{step}.
2730 @code{next} within a function that lacks debugging information acts like
2731 @code{step}, but any function calls appearing within the code of the
2732 function are executed without stopping.
2736 Continue running until just after function in the selected stack frame
2737 returns. Print the returned value (if any).
2739 Contrast this with the @code{return} command (@pxref{Returning,
2740 ,Returning from a function}).
2746 Continue running until a source line past the current line, in the
2747 current stack frame, is reached. This command is used to avoid single
2748 stepping through a loop more than once. It is like the @code{next}
2749 command, except that when @code{until} encounters a jump, it
2750 automatically continues execution until the program counter is greater
2751 than the address of the jump.
2753 This means that when you reach the end of a loop after single stepping
2754 though it, @code{until} will cause your program to continue execution
2755 until the loop is exited. In contrast, a @code{next} command at the end
2756 of a loop will simply step back to the beginning of the loop, which
2757 would force you to step through the next iteration.
2759 @code{until} always stops your program if it attempts to exit the current
2762 @code{until} may produce somewhat counterintuitive results if the order
2763 of machine code does not match the order of the source lines. For
2764 example, in the following excerpt from a debugging session, the @code{f}
2765 (@code{frame}) command shows that execution is stopped at line
2766 @code{206}; yet when we use @code{until}, we get to line @code{195}:
2770 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
2772 (@value{GDBP}) until
2773 195 for ( ; argc > 0; NEXTARG) @{
2776 This happened because, for execution efficiency, the compiler had
2777 generated code for the loop closure test at the end, rather than the
2778 start, of the loop---even though the test in a C @code{for}-loop is
2779 written before the body of the loop. The @code{until} command appeared
2780 to step back to the beginning of the loop when it advanced to this
2781 expression; however, it has not really gone to an earlier
2782 statement---not in terms of the actual machine code.
2784 @code{until} with no argument works by means of single
2785 instruction stepping, and hence is slower than @code{until} with an
2788 @item until @var{location}
2789 @item u @var{location}
2790 Continue running your program until either the specified location is
2791 reached, or the current stack frame returns. @var{location} is any of
2792 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
2793 ,Setting breakpoints}). This form of the command uses breakpoints,
2794 and hence is quicker than @code{until} without an argument.
2800 Execute one machine instruction, then stop and return to the debugger.
2802 It is often useful to do @samp{display/i $pc} when stepping by machine
2803 instructions. This will cause the next instruction to be executed to
2804 be displayed automatically at each stop. @xref{Auto Display,
2805 ,Automatic display}.
2807 An argument is a repeat count, as in @code{step}.
2814 Execute one machine instruction, but if it is a function call,
2815 proceed until the function returns.
2817 An argument is a repeat count, as in @code{next}.
2825 A signal is an asynchronous event that can happen in a program. The
2826 operating system defines the possible kinds of signals, and gives each
2827 kind a name and a number. For example, in Unix @code{SIGINT} is the
2828 signal a program gets when you type an interrupt (often @kbd{C-c});
2829 @code{SIGSEGV} is the signal a program gets from referencing a place in
2830 memory far away from all the areas in use; @code{SIGALRM} occurs when
2831 the alarm clock timer goes off (which happens only if your program has
2832 requested an alarm).
2834 @cindex fatal signals
2835 Some signals, including @code{SIGALRM}, are a normal part of the
2836 functioning of your program. Others, such as @code{SIGSEGV}, indicate
2837 errors; these signals are @dfn{fatal} (kill your program immediately) if the
2838 program has not specified in advance some other way to handle the signal.
2839 @code{SIGINT} does not indicate an error in your program, but it is normally
2840 fatal so it can carry out the purpose of the interrupt: to kill the program.
2842 @value{GDBN} has the ability to detect any occurrence of a signal in your
2843 program. You can tell @value{GDBN} in advance what to do for each kind of
2846 @cindex handling signals
2847 Normally, @value{GDBN} is set up to ignore non-erroneous signals like @code{SIGALRM}
2848 (so as not to interfere with their role in the functioning of your program)
2849 but to stop your program immediately whenever an error signal happens.
2850 You can change these settings with the @code{handle} command.
2854 @kindex info signals
2855 Print a table of all the kinds of signals and how @value{GDBN} has been told to
2856 handle each one. You can use this to see the signal numbers of all
2857 the defined types of signals.
2859 @item handle @var{signal} @var{keywords}@dots{}
2861 Change the way @value{GDBN} handles signal @var{signal}. @var{signal} can be the
2862 number of a signal or its name (with or without the @samp{SIG} at the
2863 beginning). The @var{keywords} say what change to make.
2867 The keywords allowed by the @code{handle} command can be abbreviated.
2868 Their full names are:
2872 @value{GDBN} should not stop your program when this signal happens. It may
2873 still print a message telling you that the signal has come in.
2876 @value{GDBN} should stop your program when this signal happens. This implies
2877 the @code{print} keyword as well.
2880 @value{GDBN} should print a message when this signal happens.
2883 @value{GDBN} should not mention the occurrence of the signal at all. This
2884 implies the @code{nostop} keyword as well.
2887 @value{GDBN} should allow your program to see this signal; your program will be
2888 able to handle the signal, or may be terminated if the signal is fatal
2892 @value{GDBN} should not allow your program to see this signal.
2896 When a signal stops your program, the signal is not visible until you
2897 continue. Your program will see the signal then, if @code{pass} is in
2898 effect for the signal in question @emph{at that time}. In other words,
2899 after @value{GDBN} reports a signal, you can use the @code{handle}
2900 command with @code{pass} or @code{nopass} to control whether that
2901 signal will be seen by your program when you later continue it.
2903 You can also use the @code{signal} command to prevent your program from
2904 seeing a signal, or cause it to see a signal it normally would not see,
2905 or to give it any signal at any time. For example, if your program stopped
2906 due to some sort of memory reference error, you might store correct
2907 values into the erroneous variables and continue, hoping to see more
2908 execution; but your program would probably terminate immediately as
2909 a result of the fatal signal once it saw the signal. To prevent this,
2910 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
2915 @chapter Examining the Stack
2917 When your program has stopped, the first thing you need to know is where it
2918 stopped and how it got there.
2921 Each time your program performs a function call, the information about
2922 where in your program the call was made from is saved in a block of data
2923 called a @dfn{stack frame}. The frame also contains the arguments of the
2924 call and the local variables of the function that was called. All the
2925 stack frames are allocated in a region of memory called the @dfn{call
2928 When your program stops, the @value{GDBN} commands for examining the
2929 stack allow you to see all of this information.
2931 @cindex selected frame
2932 One of the stack frames is @dfn{selected} by @value{GDBN} and many
2933 @value{GDBN} commands refer implicitly to the selected frame. In
2934 particular, whenever you ask @value{GDBN} for the value of a variable in
2935 your program, the value is found in the selected frame. There are
2936 special @value{GDBN} commands to select whichever frame you are
2939 When your program stops, @value{GDBN} automatically selects the
2940 currently executing frame and describes it briefly as the @code{frame}
2941 command does (@pxref{Frame Info, ,Information about a frame}).
2944 * Frames:: Stack frames
2945 * Backtrace:: Backtraces
2946 * Selection:: Selecting a frame
2947 * Frame Info:: Information on a frame
2949 * MIPS Stack:: MIPS machines and the function stack
2954 @section Stack frames
2958 The call stack is divided up into contiguous pieces called @dfn{stack
2959 frames}, or @dfn{frames} for short; each frame is the data associated
2960 with one call to one function. The frame contains the arguments given
2961 to the function, the function's local variables, and the address at
2962 which the function is executing.
2964 @cindex initial frame
2965 @cindex outermost frame
2966 @cindex innermost frame
2967 When your program is started, the stack has only one frame, that of the
2968 function @code{main}. This is called the @dfn{initial} frame or the
2969 @dfn{outermost} frame. Each time a function is called, a new frame is
2970 made. Each time a function returns, the frame for that function invocation
2971 is eliminated. If a function is recursive, there can be many frames for
2972 the same function. The frame for the function in which execution is
2973 actually occurring is called the @dfn{innermost} frame. This is the most
2974 recently created of all the stack frames that still exist.
2976 @cindex frame pointer
2977 Inside your program, stack frames are identified by their addresses. A
2978 stack frame consists of many bytes, each of which has its own address; each
2979 kind of computer has a convention for choosing one of those bytes whose
2980 address serves as the address of the frame. Usually this address is kept
2981 in a register called the @dfn{frame pointer register} while execution is
2982 going on in that frame.
2984 @cindex frame number
2985 @value{GDBN} assigns numbers to all existing stack frames, starting with
2986 zero for the innermost frame, one for the frame that called it,
2987 and so on upward. These numbers do not really exist in your program;
2988 they are assigned by @value{GDBN} to give you a way of designating stack
2989 frames in @value{GDBN} commands.
2991 @cindex frameless execution
2992 Some compilers provide a way to compile functions so that they operate
2993 without stack frames. (For example, the @code{@value{GCC}} option
2994 @samp{-fomit-frame-pointer} will generate functions without a frame.)
2995 This is occasionally done with heavily used library functions to save
2996 the frame setup time. @value{GDBN} has limited facilities for dealing
2997 with these function invocations. If the innermost function invocation
2998 has no stack frame, @value{GDBN} will nevertheless regard it as though
2999 it had a separate frame, which is numbered zero as usual, allowing
3000 correct tracing of the function call chain. However, @value{GDBN} has
3001 no provision for frameless functions elsewhere in the stack.
3006 A backtrace is a summary of how your program got where it is. It shows one
3007 line per frame, for many frames, starting with the currently executing
3008 frame (frame zero), followed by its caller (frame one), and on up the
3016 Print a backtrace of the entire stack: one line per frame for all
3017 frames in the stack.
3019 You can stop the backtrace at any time by typing the system interrupt
3020 character, normally @kbd{C-c}.
3022 @item backtrace @var{n}
3024 Similar, but print only the innermost @var{n} frames.
3026 @item backtrace -@var{n}
3028 Similar, but print only the outermost @var{n} frames.
3034 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3035 are additional aliases for @code{backtrace}.
3037 Each line in the backtrace shows the frame number and the function name.
3038 The program counter value is also shown---unless you use @code{set
3039 print address off}. The backtrace also shows the source file name and
3040 line number, as well as the arguments to the function. The program
3041 counter value is omitted if it is at the beginning of the code for that
3044 Here is an example of a backtrace. It was made with the command
3045 @samp{bt 3}, so it shows the innermost three frames.
3049 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3051 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3052 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3054 (More stack frames follow...)
3059 The display for frame zero does not begin with a program counter
3060 value, indicating that your program has stopped at the beginning of the
3061 code for line @code{993} of @code{builtin.c}.
3064 @section Selecting a frame
3066 Most commands for examining the stack and other data in your program work on
3067 whichever stack frame is selected at the moment. Here are the commands for
3068 selecting a stack frame; all of them finish by printing a brief description
3069 of the stack frame just selected.
3076 Select frame number @var{n}. Recall that frame zero is the innermost
3077 (currently executing) frame, frame one is the frame that called the
3078 innermost one, and so on. The highest-numbered frame is the one for
3081 @item frame @var{addr}
3083 Select the frame at address @var{addr}. This is useful mainly if the
3084 chaining of stack frames has been damaged by a bug, making it
3085 impossible for @value{GDBN} to assign numbers properly to all frames. In
3086 addition, this can be useful when your program has multiple stacks and
3087 switches between them.
3090 On the SPARC architecture, @code{frame} needs two addresses to
3091 select an arbitrary frame: a frame pointer and a stack pointer.
3092 @c note to future updaters: this is conditioned on a flag
3093 @c FRAME_SPECIFICATION_DYADIC in the tm-*.h files, currently only used
3094 @c by SPARC, hence the specific attribution. Generalize or list all
3095 @c possibilities if more supported machines start doing this.
3100 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3101 advances toward the outermost frame, to higher frame numbers, to frames
3102 that have existed longer. @var{n} defaults to one.
3107 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3108 advances toward the innermost frame, to lower frame numbers, to frames
3109 that were created more recently. @var{n} defaults to one. You may
3110 abbreviate @code{down} as @code{do}.
3113 All of these commands end by printing two lines of output describing the
3114 frame. The first line shows the frame number, the function name, the
3115 arguments, and the source file and line number of execution in that
3116 frame. The second line shows the text of that source line.
3122 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3124 10 read_input_file (argv[i]);
3128 After such a printout, the @code{list} command with no arguments will
3129 print ten lines centered on the point of execution in the frame.
3130 @xref{List, ,Printing source lines}.
3133 @item up-silently @var{n}
3134 @itemx down-silently @var{n}
3135 @kindex down-silently
3137 These two commands are variants of @code{up} and @code{down},
3138 respectively; they differ in that they do their work silently, without
3139 causing display of the new frame. They are intended primarily for use
3140 in @value{GDBN} command scripts, where the output might be unnecessary and
3145 @section Information about a frame
3147 There are several other commands to print information about the selected
3153 When used without any argument, this command does not change which
3154 frame is selected, but prints a brief description of the currently
3155 selected stack frame. It can be abbreviated @code{f}. With an
3156 argument, this command is used to select a stack frame.
3157 @xref{Selection, ,Selecting a frame}.
3163 This command prints a verbose description of the selected stack frame,
3164 including the address of the frame, the addresses of the next frame down
3165 (called by this frame) and the next frame up (caller of this frame), the
3166 language that the source code corresponding to this frame was written in,
3167 the address of the frame's arguments, the program counter saved in it
3168 (the address of execution in the caller frame), and which registers
3169 were saved in the frame. The verbose description is useful when
3170 something has gone wrong that has made the stack format fail to fit
3171 the usual conventions.
3173 @item info frame @var{addr}
3174 @itemx info f @var{addr}
3175 Print a verbose description of the frame at address @var{addr},
3176 without selecting that frame. The selected frame remains unchanged by
3181 Print the arguments of the selected frame, each on a separate line.
3185 Print the local variables of the selected frame, each on a separate
3186 line. These are all variables (declared either static or automatic)
3187 accessible at the point of execution of the selected frame.
3192 @cindex catch exceptions
3193 @cindex exception handlers
3194 Print a list of all the exception handlers that are active in the
3195 current stack frame at the current point of execution. To see other
3196 exception handlers, visit the associated frame (using the @code{up},
3197 @code{down}, or @code{frame} commands); then type @code{info catch}.
3198 @xref{Exception Handling, ,Breakpoints and exceptions}.
3204 @section MIPS machines and the function stack
3206 @cindex stack on MIPS
3208 MIPS based computers use an unusual stack frame, which sometimes
3209 requires @value{GDBN} to search backward in the object code to find the
3210 beginning of a function.
3212 @cindex response time, MIPS debugging
3213 To improve response time (especially for embedded applications, where
3214 @value{GDBN} may be restricted to a slow serial line for this search)
3215 you may want to limit the size of this search, using one of these
3217 @c FIXME! So what happens when GDB does *not* find the beginning of a
3220 @cindex @code{heuristic-fence-post} (MIPS)
3222 @item set heuristic-fence-post @var{limit}
3223 Restrict @var{GDBN} to examining at most @var{limit} bytes in its search
3224 for the beginning of a function. A value of @code{0} (the default)
3225 means there is no limit.
3227 @item show heuristic-fence-post
3228 Display the current limit.
3232 These commands are available @emph{only} when @value{GDBN} is configured
3233 for debugging programs on MIPS processors.
3237 @chapter Examining Source Files
3239 @value{GDBN} can print parts of your program's source, since the debugging
3240 information recorded in the program tells @value{GDBN} what source files were
3241 used to build it. When your program stops, @value{GDBN} spontaneously prints
3242 the line where it stopped. Likewise, when you select a stack frame
3243 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
3244 execution in that frame has stopped. You can print other portions of
3245 source files by explicit command.
3248 If you use @value{GDBN} through its GNU Emacs interface, you may prefer to use
3249 Emacs facilities to view source; @pxref{Emacs, ,Using @value{GDBN} under GNU
3254 * List:: Printing source lines
3256 * Search:: Searching source files
3259 * Source Path:: Specifying source directories
3260 * Machine Code:: Source and machine code
3264 @section Printing source lines
3268 To print lines from a source file, use the @code{list} command
3269 (abbreviated @code{l}). There are several ways to specify what part
3270 of the file you want to print.
3272 Here are the forms of the @code{list} command most commonly used:
3275 @item list @var{linenum}
3276 Print lines centered around line number @var{linenum} in the
3277 current source file.
3279 @item list @var{function}
3280 Print lines centered around the beginning of function
3284 Print more lines. If the last lines printed were printed with a
3285 @code{list} command, this prints lines following the last lines
3286 printed; however, if the last line printed was a solitary line printed
3287 as part of displaying a stack frame (@pxref{Stack, ,Examining the
3288 Stack}), this prints lines centered around that line.
3291 Print lines just before the lines last printed.
3294 By default, @value{GDBN} prints ten source lines with any of these forms of
3295 the @code{list} command. You can change this using @code{set listsize}:
3298 @item set listsize @var{count}
3299 @kindex set listsize
3300 Make the @code{list} command display @var{count} source lines (unless
3301 the @code{list} argument explicitly specifies some other number).
3304 @kindex show listsize
3305 Display the number of lines that @code{list} will currently display by
3309 Repeating a @code{list} command with @key{RET} discards the argument,
3310 so it is equivalent to typing just @code{list}. This is more useful
3311 than listing the same lines again. An exception is made for an
3312 argument of @samp{-}; that argument is preserved in repetition so that
3313 each repetition moves up in the source file.
3316 In general, the @code{list} command expects you to supply zero, one or two
3317 @dfn{linespecs}. Linespecs specify source lines; there are several ways
3318 of writing them but the effect is always to specify some source line.
3319 Here is a complete description of the possible arguments for @code{list}:
3322 @item list @var{linespec}
3323 Print lines centered around the line specified by @var{linespec}.
3325 @item list @var{first},@var{last}
3326 Print lines from @var{first} to @var{last}. Both arguments are
3329 @item list ,@var{last}
3330 Print lines ending with @var{last}.
3332 @item list @var{first},
3333 Print lines starting with @var{first}.
3336 Print lines just after the lines last printed.
3339 Print lines just before the lines last printed.
3342 As described in the preceding table.
3345 Here are the ways of specifying a single source line---all the
3350 Specifies line @var{number} of the current source file.
3351 When a @code{list} command has two linespecs, this refers to
3352 the same source file as the first linespec.
3355 Specifies the line @var{offset} lines after the last line printed.
3356 When used as the second linespec in a @code{list} command that has
3357 two, this specifies the line @var{offset} lines down from the
3361 Specifies the line @var{offset} lines before the last line printed.
3363 @item @var{filename}:@var{number}
3364 Specifies line @var{number} in the source file @var{filename}.
3366 @item @var{function}
3367 @c FIXME: "of the open-brace" is C-centric. When we add other langs...
3368 Specifies the line of the open-brace that begins the body of the
3369 function @var{function}.
3371 @item @var{filename}:@var{function}
3372 Specifies the line of the open-brace that begins the body of the
3373 function @var{function} in the file @var{filename}. You only need the
3374 file name with a function name to avoid ambiguity when there are
3375 identically named functions in different source files.
3377 @item *@var{address}
3378 Specifies the line containing the program address @var{address}.
3379 @var{address} may be any expression.
3384 @section Searching source files
3386 @kindex reverse-search
3388 There are two commands for searching through the current source file for a
3392 @item forward-search @var{regexp}
3393 @itemx search @var{regexp}
3395 @kindex forward-search
3396 The command @samp{forward-search @var{regexp}} checks each line,
3397 starting with the one following the last line listed, for a match for
3398 @var{regexp}. It lists the line that is found. You can use
3399 synonym @samp{search @var{regexp}} or abbreviate the command name as
3402 @item reverse-search @var{regexp}
3403 The command @samp{reverse-search @var{regexp}} checks each line, starting
3404 with the one before the last line listed and going backward, for a match
3405 for @var{regexp}. It lists the line that is found. You can abbreviate
3406 this command as @code{rev}.
3411 @section Specifying source directories
3414 @cindex directories for source files
3415 Executable programs sometimes do not record the directories of the source
3416 files from which they were compiled, just the names. Even when they do,
3417 the directories could be moved between the compilation and your debugging
3418 session. @value{GDBN} has a list of directories to search for source files;
3419 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
3420 it tries all the directories in the list, in the order they are present
3421 in the list, until it finds a file with the desired name. Note that
3422 the executable search path is @emph{not} used for this purpose. Neither is
3423 the current working directory, unless it happens to be in the source
3426 If @value{GDBN} cannot find a source file in the source path, and the object
3427 program records a directory, @value{GDBN} tries that directory too. If the
3428 source path is empty, and there is no record of the compilation
3429 directory, @value{GDBN} will, as a last resort, look in the current
3432 Whenever you reset or rearrange the source path, @value{GDBN} will clear out
3433 any information it has cached about where source files are found, where
3434 each line is in the file, etc.
3437 When you start @value{GDBN}, its source path is empty.
3438 To add other directories, use the @code{directory} command.
3441 @item directory @var{dirname} @dots{}
3442 Add directory @var{dirname} to the front of the source path. Several
3443 directory names may be given to this command, separated by @samp{:} or
3444 whitespace. You may specify a directory that is already in the source
3445 path; this moves it forward, so it will be searched sooner.
3447 You can use the string @samp{$cdir} to refer to the compilation
3448 directory (if one is recorded), and @samp{$cwd} to refer to the current
3449 working directory. @samp{$cwd} is not the same as @samp{.}---the former
3450 tracks the current working directory as it changes during your @value{GDBN}
3451 session, while the latter is immediately expanded to the current
3452 directory at the time you add an entry to the source path.
3455 Reset the source path to empty again. This requires confirmation.
3457 @c RET-repeat for @code{directory} is explicitly disabled, but since
3458 @c repeating it would be a no-op we do not say that. (thanks to RMS)
3460 @item show directories
3461 @kindex show directories
3462 Print the source path: show which directories it contains.
3465 If your source path is cluttered with directories that are no longer of
3466 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
3467 versions of source. You can correct the situation as follows:
3471 Use @code{directory} with no argument to reset the source path to empty.
3474 Use @code{directory} with suitable arguments to reinstall the
3475 directories you want in the source path. You can add all the
3476 directories in one command.
3480 @section Source and machine code
3482 You can use the command @code{info line} to map source lines to program
3483 addresses (and vice versa), and the command @code{disassemble} to display
3484 a range of addresses as machine instructions.
3487 @item info line @var{linespec}
3489 Print the starting and ending addresses of the compiled code for
3490 source line @var{linespec}. You can specify source lines in any of
3491 the ways understood by the @code{list} command (@pxref{List, ,Printing
3495 For example, we can use @code{info line} to discover the location of
3496 the object code for the first line of function
3497 @code{m4_changequote}:
3500 (@value{GDBP}) info line m4_changecom
3501 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
3505 We can also inquire (using @code{*@var{addr}} as the form for
3506 @var{linespec}) what source line covers a particular address:
3508 (@value{GDBP}) info line *0x63ff
3509 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
3512 @cindex @code{$_} and @code{info line}
3513 After @code{info line}, the default address for the @code{x} command
3514 is changed to the starting address of the line, so that @samp{x/i} is
3515 sufficient to begin examining the machine code (@pxref{Memory,
3516 ,Examining memory}). Also, this address is saved as the value of the
3517 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
3523 @cindex assembly instructions
3524 @cindex instructions, assembly
3525 @cindex machine instructions
3526 @cindex listing machine instructions
3527 This specialized command dumps a range of memory as machine
3528 instructions. The default memory range is the function surrounding the
3529 program counter of the selected frame. A single argument to this
3530 command is a program counter value; the function surrounding this value
3531 will be dumped. Two arguments specify a range of addresses (first
3532 inclusive, second exclusive) to dump.
3535 @ifclear H8EXCLUSIVE
3536 We can use @code{disassemble} to inspect the object code
3537 range shown in the last @code{info line} example (the example
3538 shows SPARC machine instructions):
3542 (@value{GDBP}) disas 0x63e4 0x6404
3543 Dump of assembler code from 0x63e4 to 0x6404:
3544 0x63e4 <builtin_init+5340>: ble 0x63f8 <builtin_init+5360>
3545 0x63e8 <builtin_init+5344>: sethi %hi(0x4c00), %o0
3546 0x63ec <builtin_init+5348>: ld [%i1+4], %o0
3547 0x63f0 <builtin_init+5352>: b 0x63fc <builtin_init+5364>
3548 0x63f4 <builtin_init+5356>: ld [%o0+4], %o0
3549 0x63f8 <builtin_init+5360>: or %o0, 0x1a4, %o0
3550 0x63fc <builtin_init+5364>: call 0x9288 <path_search>
3551 0x6400 <builtin_init+5368>: nop
3552 End of assembler dump.
3557 For example, here is the beginning of the output for the
3558 disassembly of a function @code{fact}:
3562 (@value{GDBP}) disas fact
3563 Dump of assembler code for function fact:
3565 0x802c <fact>: 6d f2 mov.w r2,@@-r7
3566 0x802e <fact+2>: 6d f3 mov.w r3,@@-r7
3567 0x8030 <fact+4>: 6d f6 mov.w r6,@@-r7
3568 0x8032 <fact+6>: 0d 76 mov.w r7,r6
3569 0x8034 <fact+8>: 6f 70 00 08 mov.w @@(0x8,r7),r0
3570 0x8038 <fact+12> 19 11 sub.w r1,r1
3578 @chapter Examining Data
3580 @cindex printing data
3581 @cindex examining data
3584 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
3585 @c document because it is nonstandard... Under Epoch it displays in a
3586 @c different window or something like that.
3587 The usual way to examine data in your program is with the @code{print}
3588 command (abbreviated @code{p}), or its synonym @code{inspect}.
3590 It evaluates and prints the value of an expression of the language your
3591 program is written in (@pxref{Languages, ,Using @value{GDBN} with Different
3596 @item print @var{exp}
3597 @itemx print /@var{f} @var{exp}
3598 @var{exp} is an expression (in the source language). By default the
3599 value of @var{exp} is printed in a format appropriate to its data type;
3600 you can choose a different format by specifying @samp{/@var{f}}, where
3601 @var{f} is a letter specifying the format; @pxref{Output Formats,,Output
3605 @itemx print /@var{f}
3606 If you omit @var{exp}, @value{GDBN} displays the last value again (from the
3607 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
3608 conveniently inspect the same value in an alternative format.
3611 A more low-level way of examining data is with the @code{x} command.
3612 It examines data in memory at a specified address and prints it in a
3613 specified format. @xref{Memory, ,Examining memory}.
3615 If you are interested in information about types, or about how the fields
3620 are declared, use the @code{ptype @var{exp}}
3621 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol Table}.
3624 * Expressions:: Expressions
3625 * Variables:: Program variables
3626 * Arrays:: Artificial arrays
3627 * Output Formats:: Output formats
3628 * Memory:: Examining memory
3629 * Auto Display:: Automatic display
3630 * Print Settings:: Print settings
3631 * Value History:: Value history
3632 * Convenience Vars:: Convenience variables
3633 * Registers:: Registers
3635 * Floating Point Hardware:: Floating point hardware
3640 @section Expressions
3643 @code{print} and many other @value{GDBN} commands accept an expression and
3644 compute its value. Any kind of constant, variable or operator defined
3645 by the programming language you are using is valid in an expression in
3646 @value{GDBN}. This includes conditional expressions, function calls, casts
3647 and string constants. It unfortunately does not include symbols defined
3648 by preprocessor @code{#define} commands.
3651 Because C is so widespread, most of the expressions shown in examples in
3652 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
3653 Languages}, for information on how to use expressions in other
3656 In this section, we discuss operators that you can use in @value{GDBN}
3657 expressions regardless of your programming language.
3659 Casts are supported in all languages, not just in C, because it is so
3660 useful to cast a number into a pointer so as to examine a structure
3661 at that address in memory.
3662 @c FIXME: casts supported---Mod2 true?
3665 @value{GDBN} supports these operators in addition to those of programming
3670 @samp{@@} is a binary operator for treating parts of memory as arrays.
3671 @xref{Arrays, ,Artificial arrays}, for more information.
3674 @samp{::} allows you to specify a variable in terms of the file or
3675 function where it is defined. @xref{Variables, ,Program variables}.
3677 @item @{@var{type}@} @var{addr}
3678 @cindex @{@var{type}@}
3679 @cindex type casting memory
3680 @cindex memory, viewing as typed object
3681 @cindex casts, to view memory
3682 Refers to an object of type @var{type} stored at address @var{addr} in
3683 memory. @var{addr} may be any expression whose value is an integer or
3684 pointer (but parentheses are required around binary operators, just as in
3685 a cast). This construct is allowed regardless of what kind of data is
3686 normally supposed to reside at @var{addr}.
3690 @section Program variables
3692 The most common kind of expression to use is the name of a variable
3695 Variables in expressions are understood in the selected stack frame
3696 (@pxref{Selection, ,Selecting a frame}); they must either be global
3697 (or static) or be visible according to the scope rules of the
3698 programming language from the point of execution in that frame. This
3699 means that in the function
3714 you can examine and use the variable @code{a} whenever your program is
3715 executing within the function @code{foo}, but you can only use or
3716 examine the variable @code{b} while your program is executing inside
3717 the block where @code{b} is declared.
3719 @cindex variable name conflict
3720 There is an exception: you can refer to a variable or function whose
3721 scope is a single source file even if the current execution point is not
3722 in this file. But it is possible to have more than one such variable or
3723 function with the same name (in different source files). If that
3724 happens, referring to that name has unpredictable effects. If you wish,
3725 you can specify a static variable in a particular function or file,
3726 using the colon-colon notation:
3730 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
3734 @var{file}::@var{variable}
3735 @var{function}::@var{variable}
3739 Here @var{file} or @var{function} is the name of the context for the
3740 static @var{variable}. In the case of file names, you can use quotes to
3741 make sure @value{GDBN} parses the file name as a single word---for example,
3742 to print a global value of @code{x} defined in @file{f2.c}:
3745 (@value{GDBP}) p 'f2.c'::x
3749 @cindex C++ scope resolution
3750 This use of @samp{::} is very rarely in conflict with the very similar
3751 use of the same notation in C++. @value{GDBN} also supports use of the C++
3752 scope resolution operator in @value{GDBN} expressions.
3753 @c FIXME: Um, so what happens in one of those rare cases where it's in
3757 @cindex wrong values
3758 @cindex variable values, wrong
3760 @emph{Warning:} Occasionally, a local variable may appear to have the
3761 wrong value at certain points in a function---just after entry to a new
3762 scope, and just before exit.
3764 You may see this problem when you are stepping by machine instructions.
3765 This is because on most machines, it takes more than one instruction to
3766 set up a stack frame (including local variable definitions); if you are
3767 stepping by machine instructions, variables may appear to have the wrong
3768 values until the stack frame is completely built. On exit, it usually
3769 also takes more than one machine instruction to destroy a stack frame;
3770 after you begin stepping through that group of instructions, local
3771 variable definitions may be gone.
3774 @section Artificial arrays
3776 @cindex artificial array
3778 It is often useful to print out several successive objects of the
3779 same type in memory; a section of an array, or an array of
3780 dynamically determined size for which only a pointer exists in the
3783 You can do this by referring to a contiguous span of memory as an
3784 @dfn{artificial array}, using the binary operator @samp{@@}. The left
3785 operand of @samp{@@} should be the first element of the desired array,
3786 as an individual object. The right operand should be the desired length
3787 of the array. The result is an array value whose elements are all of
3788 the type of the left argument. The first element is actually the left
3789 argument; the second element comes from bytes of memory immediately
3790 following those that hold the first element, and so on. Here is an
3791 example. If a program says
3794 int *array = (int *) malloc (len * sizeof (int));
3798 you can print the contents of @code{array} with
3804 The left operand of @samp{@@} must reside in memory. Array values made
3805 with @samp{@@} in this way behave just like other arrays in terms of
3806 subscripting, and are coerced to pointers when used in expressions.
3807 Artificial arrays most often appear in expressions via the value history
3808 (@pxref{Value History, ,Value history}), after printing one out.)
3810 Sometimes the artificial array mechanism is not quite enough; in
3811 moderately complex data structures, the elements of interest may not
3812 actually be adjacent---for example, if you are interested in the values
3813 of pointers in an array. One useful work-around in this situation is
3814 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
3815 variables}) as a counter in an expression that prints the first
3816 interesting value, and then repeat that expression via @key{RET}. For
3817 instance, suppose you have an array @code{dtab} of pointers to
3818 structures, and you are interested in the values of a field @code{fv}
3819 in each structure. Here is an example of what you might type:
3829 @node Output Formats
3830 @section Output formats
3832 @cindex formatted output
3833 @cindex output formats
3834 By default, @value{GDBN} prints a value according to its data type. Sometimes
3835 this is not what you want. For example, you might want to print a number
3836 in hex, or a pointer in decimal. Or you might want to view data in memory
3837 at a certain address as a character string or as an instruction. To do
3838 these things, specify an @dfn{output format} when you print a value.
3840 The simplest use of output formats is to say how to print a value
3841 already computed. This is done by starting the arguments of the
3842 @code{print} command with a slash and a format letter. The format
3843 letters supported are:
3847 Regard the bits of the value as an integer, and print the integer in
3851 Print as integer in signed decimal.
3854 Print as integer in unsigned decimal.
3857 Print as integer in octal.
3860 Print as integer in binary. The letter @samp{t} stands for ``two''.
3861 @footnote{@samp{b} cannot be used because these format letters are also
3862 used with the @code{x} command, where @samp{b} stands for ``byte'';
3863 @pxref{Memory,,Examining memory}.}
3866 Print as an address, both absolute in hex and as an offset from the
3867 nearest preceding symbol. This format can be used to discover where (in
3868 what function) an unknown address is located:
3871 (@value{GDBP}) p/a 0x54320
3872 $3 = 0x54320 <_initialize_vx+396>
3876 Regard as an integer and print it as a character constant.
3879 Regard the bits of the value as a floating point number and print
3880 using typical floating point syntax.
3883 For example, to print the program counter in hex (@pxref{Registers}), type
3890 Note that no space is required before the slash; this is because command
3891 names in @value{GDBN} cannot contain a slash.
3893 To reprint the last value in the value history with a different format,
3894 you can use the @code{print} command with just a format and no
3895 expression. For example, @samp{p/x} reprints the last value in hex.
3898 @section Examining memory
3900 You can use the command @code{x} (for ``examine'') to examine memory in
3901 any of several formats, independently of your program's data types.
3903 @cindex examining memory
3906 @item x/@var{nfu} @var{addr}
3909 Use the @code{x} command to examine memory.
3912 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
3913 much memory to display and how to format it; @var{addr} is an
3914 expression giving the address where you want to start displaying memory.
3915 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
3916 Several commands set convenient defaults for @var{addr}.
3919 @item @var{n}, the repeat count
3920 The repeat count is a decimal integer; the default is 1. It specifies
3921 how much memory (counting by units @var{u}) to display.
3922 @c This really is **decimal**; unaffected by 'set radix' as of GDB
3925 @item @var{f}, the display format
3926 The display format is one of the formats used by @code{print},
3927 or @samp{s} (null-terminated string) or @samp{i} (machine instruction).
3928 The default is @samp{x} (hexadecimal) initially, or the format from the
3929 last time you used either @code{x} or @code{print}.
3931 @item @var{u}, the unit size
3932 The unit size is any of
3938 Halfwords (two bytes).
3940 Words (four bytes). This is the initial default.
3942 Giant words (eight bytes).
3945 Each time you specify a unit size with @code{x}, that size becomes the
3946 default unit the next time you use @code{x}. (For the @samp{s} and
3947 @samp{i} formats, the unit size is ignored and is normally not written.)
3949 @item @var{addr}, starting display address
3950 @var{addr} is the address where you want @value{GDBN} to begin displaying
3951 memory. The expression need not have a pointer value (though it may);
3952 it is always interpreted as an integer address of a byte of memory.
3953 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
3954 @var{addr} is usually just after the last address examined---but several
3955 other commands also set the default address: @code{info breakpoints} (to
3956 the address of the last breakpoint listed), @code{info line} (to the
3957 starting address of a line), and @code{print} (if you use it to display
3958 a value from memory).
3961 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
3962 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
3963 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
3964 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
3965 @pxref{Registers}) in hexadecimal (@samp{x}).
3967 Since the letters indicating unit sizes are all distinct from the
3968 letters specifying output formats, you do not have to remember whether
3969 unit size or format comes first; either order will work. The output
3970 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
3971 (However, the count @var{n} must come first; @samp{wx4} will not work.)
3973 Even though the unit size @var{u} is ignored for the formats @samp{s}
3974 and @samp{i}, you might still want to use a count @var{n}; for example,
3975 @samp{3i} specifies that you want to see three machine instructions,
3976 including any operands. The command @code{disassemble} gives an
3977 alternative way of inspecting machine instructions; @pxref{Machine
3978 Code,,Source and machine code}.
3980 All the defaults for the arguments to @code{x} are designed to make it
3981 easy to continue scanning memory with minimal specifications each time
3982 you use @code{x}. For example, after you have inspected three machine
3983 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
3984 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
3985 the repeat count @var{n} is used again; the other arguments default as
3986 for successive uses of @code{x}.
3988 @cindex @code{$_}, @code{$__}, and value history
3989 The addresses and contents printed by the @code{x} command are not saved
3990 in the value history because there is often too much of them and they
3991 would get in the way. Instead, @value{GDBN} makes these values available for
3992 subsequent use in expressions as values of the convenience variables
3993 @code{$_} and @code{$__}. After an @code{x} command, the last address
3994 examined is available for use in expressions in the convenience variable
3995 @code{$_}. The contents of that address, as examined, are available in
3996 the convenience variable @code{$__}.
3998 If the @code{x} command has a repeat count, the address and contents saved
3999 are from the last memory unit printed; this is not the same as the last
4000 address printed if several units were printed on the last line of output.
4003 @section Automatic display
4004 @cindex automatic display
4005 @cindex display of expressions
4007 If you find that you want to print the value of an expression frequently
4008 (to see how it changes), you might want to add it to the @dfn{automatic
4009 display list} so that @value{GDBN} will print its value each time your program stops.
4010 Each expression added to the list is given a number to identify it;
4011 to remove an expression from the list, you specify that number.
4012 The automatic display looks like this:
4016 3: bar[5] = (struct hack *) 0x3804
4020 This display shows item numbers, expressions and their current values. As with
4021 displays you request manually using @code{x} or @code{print}, you can
4022 specify the output format you prefer; in fact, @code{display} decides
4023 whether to use @code{print} or @code{x} depending on how elaborate your
4024 format specification is---it uses @code{x} if you specify a unit size,
4025 or one of the two formats (@samp{i} and @samp{s}) that are only
4026 supported by @code{x}; otherwise it uses @code{print}.
4029 @item display @var{exp}
4031 Add the expression @var{exp} to the list of expressions to display
4032 each time your program stops. @xref{Expressions, ,Expressions}.
4034 @code{display} will not repeat if you press @key{RET} again after using it.
4036 @item display/@var{fmt} @var{exp}
4037 For @var{fmt} specifying only a display format and not a size or
4038 count, add the expression @var{exp} to the auto-display list but
4039 arranges to display it each time in the specified format @var{fmt}.
4040 @xref{Output Formats,,Output formats}.
4042 @item display/@var{fmt} @var{addr}
4043 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4044 number of units, add the expression @var{addr} as a memory address to
4045 be examined each time your program stops. Examining means in effect
4046 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4049 For example, @samp{display/i $pc} can be helpful, to see the machine
4050 instruction about to be executed each time execution stops (@samp{$pc}
4051 is a common name for the program counter; @pxref{Registers}).
4054 @item undisplay @var{dnums}@dots{}
4055 @itemx delete display @var{dnums}@dots{}
4056 @kindex delete display
4058 Remove item numbers @var{dnums} from the list of expressions to display.
4060 @code{undisplay} will not repeat if you press @key{RET} after using it.
4061 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4063 @item disable display @var{dnums}@dots{}
4064 @kindex disable display
4065 Disable the display of item numbers @var{dnums}. A disabled display
4066 item is not printed automatically, but is not forgotten. It may be
4067 enabled again later.
4069 @item enable display @var{dnums}@dots{}
4070 @kindex enable display
4071 Enable display of item numbers @var{dnums}. It becomes effective once
4072 again in auto display of its expression, until you specify otherwise.
4075 Display the current values of the expressions on the list, just as is
4076 done when your program stops.
4079 @kindex info display
4080 Print the list of expressions previously set up to display
4081 automatically, each one with its item number, but without showing the
4082 values. This includes disabled expressions, which are marked as such.
4083 It also includes expressions which would not be displayed right now
4084 because they refer to automatic variables not currently available.
4087 If a display expression refers to local variables, then it does not make
4088 sense outside the lexical context for which it was set up. Such an
4089 expression is disabled when execution enters a context where one of its
4090 variables is not defined. For example, if you give the command
4091 @code{display last_char} while inside a function with an argument
4092 @code{last_char}, then this argument will be displayed while your program
4093 continues to stop inside that function. When it stops elsewhere---where
4094 there is no variable @code{last_char}---display is disabled. The next time
4095 your program stops where @code{last_char} is meaningful, you can enable the
4096 display expression once again.
4098 @node Print Settings
4099 @section Print settings
4101 @cindex format options
4102 @cindex print settings
4103 @value{GDBN} provides the following ways to control how arrays, structures,
4104 and symbols are printed.
4107 These settings are useful for debugging programs in any language:
4110 @item set print address
4111 @item set print address on
4112 @kindex set print address
4113 @value{GDBN} will print memory addresses showing the location of stack
4114 traces, structure values, pointer values, breakpoints, and so forth,
4115 even when it also displays the contents of those addresses. The default
4116 is on. For example, this is what a stack frame display looks like, with
4117 @code{set print address on}:
4122 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
4124 530 if (lquote != def_lquote)
4128 @item set print address off
4129 Do not print addresses when displaying their contents. For example,
4130 this is the same stack frame displayed with @code{set print address off}:
4134 (@value{GDBP}) set print addr off
4136 #0 set_quotes (lq="<<", rq=">>") at input.c:530
4137 530 if (lquote != def_lquote)
4141 You can use @samp{set print address off} to eliminate all machine
4142 dependent displays from the @value{GDBN} interface. For example, with
4143 @code{print address off}, you should get the same text for backtraces on
4144 all machines---whether or not they involve pointer arguments.
4146 @item show print address
4147 @kindex show print address
4148 Show whether or not addresses are to be printed.
4151 When @value{GDBN} prints a symbolic address, it normally prints the
4152 closest earlier symbol plus an offset. If that symbol does not uniquely
4153 identify the address (for example, it is a name whose scope is a single
4154 source file), you may need to disambiguate. One way to do this is with
4155 @code{info line}, for example @code{info line *0x4537}. Alternately,
4156 you can set @value{GDBN} to print the source file and line number when
4157 it prints a symbolic address:
4160 @item set print symbol-filename on
4161 @kindex set print symbol-filename
4162 Tell @value{GDBN} to print the source file name and line number of a
4163 symbol in the symbolic form of an address.
4165 @item set print symbol-filename off
4166 Do not print source file name and line number of a symbol. This is the
4169 @item show print symbol-filename
4170 @kindex show print symbol-filename
4171 Show whether or not @value{GDBN} will print the source file name and
4172 line number of a symbol in the symbolic form of an address.
4175 Also, you may wish to see the symbolic form only if the address being
4176 printed is reasonably close to the closest earlier symbol:
4179 @item set print max-symbolic-offset @var{max-offset}
4180 @kindex set print max-symbolic-offset
4181 Tell @value{GDBN} to only display the symbolic form of an address if the
4182 offset between the closest earlier symbol and the address is less than
4183 @var{max-offset}. The default is 0, which means to always print the
4184 symbolic form of an address, if any symbol precedes it.
4186 @item show print max-symbolic-offset
4187 @kindex show print max-symbolic-offset
4188 Ask how large the maximum offset is that @value{GDBN} will print in a
4193 @item set print array
4194 @itemx set print array on
4195 @kindex set print array
4196 @value{GDBN} will pretty-print arrays. This format is more convenient to read,
4197 but uses more space. The default is off.
4199 @item set print array off
4200 Return to compressed format for arrays.
4202 @item show print array
4203 @kindex show print array
4204 Show whether compressed or pretty format is selected for displaying
4207 @item set print elements @var{number-of-elements}
4208 @kindex set print elements
4209 If @value{GDBN} is printing a large array, it will stop printing after it has
4210 printed the number of elements set by the @code{set print elements} command.
4211 This limit also applies to the display of strings.
4212 Setting the number of elements to zero means that the printing is unlimited.
4214 @item show print elements
4215 @kindex show print elements
4216 Display the number of elements of a large array that @value{GDBN} will print
4217 before losing patience.
4219 @item set print pretty on
4220 @kindex set print pretty
4221 Cause @value{GDBN} to print structures in an indented format with one member per
4237 @item set print pretty off
4238 Cause @value{GDBN} to print structures in a compact format, like this:
4242 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
4243 meat = 0x54 "Pork"@}
4248 This is the default format.
4250 @item show print pretty
4251 @kindex show print pretty
4252 Show which format @value{GDBN} will use to print structures.
4254 @item set print sevenbit-strings on
4255 @kindex set print sevenbit-strings
4256 Print using only seven-bit characters; if this option is set,
4257 @value{GDBN} will display any eight-bit characters (in strings or character
4258 values) using the notation @code{\}@var{nnn}. For example, @kbd{M-a} is
4259 displayed as @code{\341}.
4261 @item set print sevenbit-strings off
4262 Print using either seven-bit or eight-bit characters, as required. This
4265 @item show print sevenbit-strings
4266 @kindex show print sevenbit-strings
4267 Show whether or not @value{GDBN} will print only seven-bit characters.
4269 @item set print union on
4270 @kindex set print union
4271 Tell @value{GDBN} to print unions which are contained in structures. This is the
4274 @item set print union off
4275 Tell @value{GDBN} not to print unions which are contained in structures.
4277 @item show print union
4278 @kindex show print union
4279 Ask @value{GDBN} whether or not it will print unions which are contained in
4282 For example, given the declarations
4285 typedef enum @{Tree, Bug@} Species;
4286 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
4287 typedef enum @{Caterpillar, Cocoon, Butterfly@}
4298 struct thing foo = @{Tree, @{Acorn@}@};
4302 with @code{set print union on} in effect @samp{p foo} would print
4305 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
4309 and with @code{set print union off} in effect it would print
4312 $1 = @{it = Tree, form = @{...@}@}
4319 These settings are of interest when debugging C++ programs:
4322 @item set print demangle
4323 @itemx set print demangle on
4324 @kindex set print demangle
4325 Print C++ names in their source form rather than in the encoded
4326 (``mangled'') form passed to the assembler and linker for type-safe
4327 linkage. The default is @samp{on}.
4329 @item show print demangle
4330 @kindex show print demangle
4331 Show whether C++ names will be printed in mangled or demangled form.
4333 @item set print asm-demangle
4334 @itemx set print asm-demangle on
4335 @kindex set print asm-demangle
4336 Print C++ names in their source form rather than their mangled form, even
4337 in assembler code printouts such as instruction disassemblies.
4340 @item show print asm-demangle
4341 @kindex show print asm-demangle
4342 Show whether C++ names in assembly listings will be printed in mangled
4345 @item set demangle-style @var{style}
4346 @kindex set demangle-style
4347 @cindex C++ symbol decoding style
4348 @cindex symbol decoding style, C++
4349 Choose among several encoding schemes used by different compilers to
4350 represent C++ names. The choices for @var{style} are currently:
4354 Allow @value{GDBN} to choose a decoding style by inspecting your program.
4357 Decode based on the GNU C++ compiler (@code{g++}) encoding algorithm.
4360 Decode based on the Lucid C++ compiler (@code{lcc}) encoding algorithm.
4363 Decode using the algorithm in the @cite{C++ Annotated Reference Manual}.
4364 @strong{Warning:} this setting alone is not sufficient to allow
4365 debugging @code{cfront}-generated executables. @value{GDBN} would
4366 require further enhancement to permit that.
4369 @item show demangle-style
4370 @kindex show demangle-style
4371 Display the encoding style currently in use for decoding C++ symbols.
4373 @item set print object
4374 @itemx set print object on
4375 @kindex set print object
4376 When displaying a pointer to an object, identify the @emph{actual}
4377 (derived) type of the object rather than the @emph{declared} type, using
4378 the virtual function table.
4380 @item set print object off
4381 Display only the declared type of objects, without reference to the
4382 virtual function table. This is the default setting.
4384 @item show print object
4385 @kindex show print object
4386 Show whether actual, or declared, object types will be displayed.
4388 @item set print vtbl
4389 @itemx set print vtbl on
4390 @kindex set print vtbl
4391 Pretty print C++ virtual function tables. The default is off.
4393 @item set print vtbl off
4394 Do not pretty print C++ virtual function tables.
4396 @item show print vtbl
4397 @kindex show print vtbl
4398 Show whether C++ virtual function tables are pretty printed, or not.
4403 @section Value history
4405 @cindex value history
4406 Values printed by the @code{print} command are saved in the @value{GDBN} @dfn{value
4407 history} so that you can refer to them in other expressions. Values are
4408 kept until the symbol table is re-read or discarded (for example with
4409 the @code{file} or @code{symbol-file} commands). When the symbol table
4410 changes, the value history is discarded, since the values may contain
4411 pointers back to the types defined in the symbol table.
4415 @cindex history number
4416 The values printed are given @dfn{history numbers} for you to refer to them
4417 by. These are successive integers starting with one. @code{print} shows you
4418 the history number assigned to a value by printing @samp{$@var{num} = }
4419 before the value; here @var{num} is the history number.
4421 To refer to any previous value, use @samp{$} followed by the value's
4422 history number. The way @code{print} labels its output is designed to
4423 remind you of this. Just @code{$} refers to the most recent value in
4424 the history, and @code{$$} refers to the value before that.
4425 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
4426 is the value just prior to @code{$$}, @code{$$1} is equivalent to
4427 @code{$$}, and @code{$$0} is equivalent to @code{$}.
4429 For example, suppose you have just printed a pointer to a structure and
4430 want to see the contents of the structure. It suffices to type
4436 If you have a chain of structures where the component @code{next} points
4437 to the next one, you can print the contents of the next one with this:
4444 You can print successive links in the chain by repeating this
4445 command---which you can do by just typing @key{RET}.
4447 Note that the history records values, not expressions. If the value of
4448 @code{x} is 4 and you type these commands:
4456 then the value recorded in the value history by the @code{print} command
4457 remains 4 even though the value of @code{x} has changed.
4462 Print the last ten values in the value history, with their item numbers.
4463 This is like @samp{p@ $$9} repeated ten times, except that @code{show
4464 values} does not change the history.
4466 @item show values @var{n}
4467 Print ten history values centered on history item number @var{n}.
4470 Print ten history values just after the values last printed. If no more
4471 values are available, produces no display.
4474 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
4475 same effect as @samp{show values +}.
4477 @node Convenience Vars
4478 @section Convenience variables
4480 @cindex convenience variables
4481 @value{GDBN} provides @dfn{convenience variables} that you can use within
4482 @value{GDBN} to hold on to a value and refer to it later. These variables
4483 exist entirely within @value{GDBN}; they are not part of your program, and
4484 setting a convenience variable has no direct effect on further execution
4485 of your program. That is why you can use them freely.
4487 Convenience variables are prefixed with @samp{$}. Any name preceded by
4488 @samp{$} can be used for a convenience variable, unless it is one of
4489 the predefined machine-specific register names (@pxref{Registers}).
4490 (Value history references, in contrast, are @emph{numbers} preceded
4491 by @samp{$}. @xref{Value History, ,Value history}.)
4493 You can save a value in a convenience variable with an assignment
4494 expression, just as you would set a variable in your program.
4498 set $foo = *object_ptr
4502 would save in @code{$foo} the value contained in the object pointed to by
4505 Using a convenience variable for the first time creates it; but its value
4506 is @code{void} until you assign a new value. You can alter the value with
4507 another assignment at any time.
4509 Convenience variables have no fixed types. You can assign a convenience
4510 variable any type of value, including structures and arrays, even if
4511 that variable already has a value of a different type. The convenience
4512 variable, when used as an expression, has the type of its current value.
4515 @item show convenience
4516 @kindex show convenience
4517 Print a list of convenience variables used so far, and their values.
4518 Abbreviated @code{show con}.
4521 One of the ways to use a convenience variable is as a counter to be
4522 incremented or a pointer to be advanced. For example, to print
4523 a field from successive elements of an array of structures:
4527 print bar[$i++]->contents
4528 @i{@dots{} repeat that command by typing @key{RET}.}
4531 Some convenience variables are created automatically by @value{GDBN} and given
4532 values likely to be useful.
4537 The variable @code{$_} is automatically set by the @code{x} command to
4538 the last address examined (@pxref{Memory, ,Examining memory}). Other
4539 commands which provide a default address for @code{x} to examine also
4540 set @code{$_} to that address; these commands include @code{info line}
4541 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
4542 except when set by the @code{x} command, in which case it is a pointer
4543 to the type of @code{$__}.
4547 The variable @code{$__} is automatically set by the @code{x} command
4548 to the value found in the last address examined. Its type is chosen
4549 to match the format in which the data was printed.
4556 You can refer to machine register contents, in expressions, as variables
4557 with names starting with @samp{$}. The names of registers are different
4558 for each machine; use @code{info registers} to see the names used on
4562 @item info registers
4563 @kindex info registers
4564 Print the names and values of all registers except floating-point
4565 registers (in the selected stack frame).
4567 @item info all-registers
4568 @kindex info all-registers
4569 @cindex floating point registers
4570 Print the names and values of all registers, including floating-point
4573 @item info registers @var{regname} @dots{}
4574 Print the relativized value of each specified register @var{regname}.
4575 @var{regname} may be any register name valid on the machine you are using, with
4576 or without the initial @samp{$}.
4579 @value{GDBN} has four ``standard'' register names that are available (in
4580 expressions) on most machines---whenever they do not conflict with an
4581 architecture's canonical mnemonics for registers. The register names
4582 @code{$pc} and @code{$sp} are used for the program counter register and
4583 the stack pointer. @code{$fp} is used for a register that contains a
4584 pointer to the current stack frame, and @code{$ps} is used for a
4585 register that contains the processor status. For example,
4586 you could print the program counter in hex with
4593 or print the instruction to be executed next with
4600 or add four to the stack pointer@footnote{This is a way of removing
4601 one word from the stack, on machines where stacks grow downward in
4602 memory (most machines, nowadays). This assumes that the innermost
4603 stack frame is selected; setting @code{$sp} is not allowed when other
4604 stack frames are selected. To pop entire frames off the stack,
4605 regardless of machine architecture, use @code{return};
4606 @pxref{Returning, ,Returning from a function}.} with
4612 Whenever possible, these four standard register names are available on
4613 your machine even though the machine has different canonical mnemonics,
4614 so long as there is no conflict. The @code{info registers} command
4615 shows the canonical names. For example, on the SPARC, @code{info
4616 registers} displays the processor status register as @code{$psr} but you
4617 can also refer to it as @code{$ps}.
4619 @value{GDBN} always considers the contents of an ordinary register as an
4620 integer when the register is examined in this way. Some machines have
4621 special registers which can hold nothing but floating point; these
4622 registers are considered to have floating point values. There is no way
4623 to refer to the contents of an ordinary register as floating point value
4624 (although you can @emph{print} it as a floating point value with
4625 @samp{print/f $@var{regname}}).
4627 Some registers have distinct ``raw'' and ``virtual'' data formats. This
4628 means that the data format in which the register contents are saved by
4629 the operating system is not the same one that your program normally
4630 sees. For example, the registers of the 68881 floating point
4631 coprocessor are always saved in ``extended'' (raw) format, but all C
4632 programs expect to work with ``double'' (virtual) format. In such
4633 cases, @value{GDBN} normally works with the virtual format only (the format that
4634 makes sense for your program), but the @code{info registers} command
4635 prints the data in both formats.
4637 Normally, register values are relative to the selected stack frame
4638 (@pxref{Selection, ,Selecting a frame}). This means that you get the
4639 value that the register would contain if all stack frames farther in
4640 were exited and their saved registers restored. In order to see the
4641 true contents of hardware registers, you must select the innermost
4642 frame (with @samp{frame 0}).
4644 However, @value{GDBN} must deduce where registers are saved, from the machine
4645 code generated by your compiler. If some registers are not saved, or if
4646 @value{GDBN} is unable to locate the saved registers, the selected stack
4647 frame will make no difference.
4651 @item set rstack_high_address @var{address}
4652 @kindex set rstack_high_address
4653 @cindex AMD 29K register stack
4654 @cindex register stack, AMD29K
4655 On AMD 29000 family processors, registers are saved in a separate
4656 ``register stack''. There is no way for @value{GDBN} to determine the extent
4657 of this stack. Normally, @value{GDBN} just assumes that the stack is ``large
4658 enough''. This may result in @value{GDBN} referencing memory locations that
4659 do not exist. If necessary, you can get around this problem by
4660 specifying the ending address of the register stack with the @code{set
4661 rstack_high_address} command. The argument should be an address, which
4662 you will probably want to precede with @samp{0x} to specify in
4665 @item show rstack_high_address
4666 @kindex show rstack_high_address
4667 Display the current limit of the register stack, on AMD 29000 family
4673 @node Floating Point Hardware
4674 @section Floating point hardware
4675 @cindex floating point
4677 @c FIXME! Really host, not target?
4678 Depending on the host machine architecture, @value{GDBN} may be able to give
4679 you more information about the status of the floating point hardware.
4684 Display hardware-dependent information about the floating
4685 point unit. The exact contents and layout vary depending on the
4686 floating point chip; on some platforms, @samp{info float} is not
4689 @c FIXME: this is a cop-out. Try to get examples, explanations. Only
4690 @c FIXME...supported currently on arm's and 386's. Mark properly with
4691 @c FIXME... m4 macros to isolate general statements from hardware-dep,
4692 @c FIXME... at that point.
4697 @chapter Using @value{GDBN} with Different Languages
4701 Although programming languages generally have common aspects, they are
4702 rarely expressed in the same manner. For instance, in ANSI C,
4703 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
4704 Modula-2, it is accomplished by @code{p^}. Values can also be
4705 represented (and displayed) differently. Hex numbers in C are written
4706 like @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
4709 @cindex working language
4710 Language-specific information is built into @value{GDBN} for some languages,
4711 allowing you to express operations like the above in your program's
4712 native language, and allowing @value{GDBN} to output values in a manner
4713 consistent with the syntax of your program's native language. The
4714 language you use to build expressions, called the @dfn{working
4715 language}, can be selected manually, or @value{GDBN} can set it
4719 * Setting:: Switching between source languages
4720 * Show:: Displaying the language
4722 * Checks:: Type and range checks
4725 * Support:: Supported languages
4729 @section Switching between source languages
4731 There are two ways to control the working language---either have @value{GDBN}
4732 set it automatically, or select it manually yourself. You can use the
4733 @code{set language} command for either purpose. On startup, @value{GDBN}
4734 defaults to setting the language automatically.
4737 * Manually:: Setting the working language manually
4738 * Automatically:: Having @value{GDBN} infer the source language
4742 @subsection Setting the working language
4744 If you allow @value{GDBN} to set the language automatically,
4745 expressions are interpreted the same way in your debugging session and
4748 @kindex set language
4749 If you wish, you may set the language manually. To do this, issue the
4750 command @samp{set language @var{lang}}, where @var{lang} is the name of
4756 @code{c} or @code{modula-2}.
4758 For a list of the supported languages, type @samp{set language}.
4759 @c FIXME: rms: eventually this command should be "help set language".
4762 Setting the language manually prevents @value{GDBN} from updating the working
4763 language automatically. This can lead to confusion if you try
4764 to debug a program when the working language is not the same as the
4765 source language, when an expression is acceptable to both
4766 languages---but means different things. For instance, if the current
4767 source file were written in C, and @value{GDBN} was parsing Modula-2, a
4775 might not have the effect you intended. In C, this means to add
4776 @code{b} and @code{c} and place the result in @code{a}. The result
4777 printed would be the value of @code{a}. In Modula-2, this means to compare
4778 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
4782 @subsection Having @value{GDBN} infer the source language
4784 To have @value{GDBN} set the working language automatically, use @samp{set
4785 language local} or @samp{set language auto}. @value{GDBN} then infers the
4786 language that a program was written in by looking at the name of its
4787 source files, and examining their extensions:
4792 Modula-2 source file
4803 This information is recorded for each function or procedure in a source
4804 file. When your program stops in a frame (usually by encountering a
4805 breakpoint), @value{GDBN} sets the working language to the language recorded
4806 for the function in that frame. If the language for a frame is unknown
4807 (that is, if the function or block corresponding to the frame was
4808 defined in a source file that does not have a recognized extension), the
4809 current working language is not changed, and @value{GDBN} issues a warning.
4811 This may not seem necessary for most programs, which are written
4812 entirely in one source language. However, program modules and libraries
4813 written in one source language can be used by a main program written in
4814 a different source language. Using @samp{set language auto} in this
4815 case frees you from having to set the working language manually.
4818 @section Displaying the language
4820 The following commands will help you find out which language is the
4821 working language, and also what language source files were written in.
4823 @kindex show language
4828 Display the current working language. This is the
4829 language you can use with commands such as @code{print} to
4830 build and compute expressions that may involve variables in your program.
4833 Among the other information listed here (@pxref{Frame Info, ,Information
4834 about a frame}) is the source language for this frame. This is the
4835 language that will become the working language if you ever use an
4836 identifier that is in this frame.
4839 Among the other information listed here (@pxref{Symbols, ,Examining the
4840 Symbol Table}) is the source language of this source file.
4845 @section Type and range checking
4848 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
4849 checking are included, but they do not yet have any effect. This
4850 section documents the intended facilities.
4852 @c FIXME remove warning when type/range code added
4854 Some languages are designed to guard you against making seemingly common
4855 errors through a series of compile- and run-time checks. These include
4856 checking the type of arguments to functions and operators, and making
4857 sure mathematical overflows are caught at run time. Checks such as
4858 these help to ensure a program's correctness once it has been compiled
4859 by eliminating type mismatches, and providing active checks for range
4860 errors when your program is running.
4862 @value{GDBN} can check for conditions like the above if you wish.
4863 Although @value{GDBN} will not check the statements in your program, it
4864 can check expressions entered directly into @value{GDBN} for evaluation via
4865 the @code{print} command, for example. As with the working language,
4866 @value{GDBN} can also decide whether or not to check automatically based on
4867 your program's source language. @xref{Support, ,Supported languages},
4868 for the default settings of supported languages.
4871 * Type Checking:: An overview of type checking
4872 * Range Checking:: An overview of range checking
4875 @cindex type checking
4876 @cindex checks, type
4878 @subsection An overview of type checking
4880 Some languages, such as Modula-2, are strongly typed, meaning that the
4881 arguments to operators and functions have to be of the correct type,
4882 otherwise an error occurs. These checks prevent type mismatch
4883 errors from ever causing any run-time problems. For example,
4891 The second example fails because the @code{CARDINAL} 1 is not
4892 type-compatible with the @code{REAL} 2.3.
4894 For expressions you use in @value{GDBN} commands, you can tell the @value{GDBN}
4895 type checker to skip checking; to treat any mismatches as errors and
4896 abandon the expression; or only issue warnings when type mismatches
4897 occur, but evaluate the expression anyway. When you choose the last of
4898 these, @value{GDBN} evaluates expressions like the second example above, but
4899 also issues a warning.
4901 Even though you may turn type checking off, other type-based reasons may
4902 prevent @value{GDBN} from evaluating an expression. For instance, @value{GDBN} does not
4903 know how to add an @code{int} and a @code{struct foo}. These particular
4904 type errors have nothing to do with the language in use, and usually
4905 arise from expressions, such as the one described above, which make
4906 little sense to evaluate anyway.
4908 Each language defines to what degree it is strict about type. For
4909 instance, both Modula-2 and C require the arguments to arithmetical
4910 operators to be numbers. In C, enumerated types and pointers can be
4911 represented as numbers, so that they are valid arguments to mathematical
4912 operators. @xref{Support, ,Supported languages}, for further
4913 details on specific languages.
4915 @value{GDBN} provides some additional commands for controlling the type checker:
4918 @kindex set check type
4919 @kindex show check type
4921 @item set check type auto
4922 Set type checking on or off based on the current working language.
4923 @xref{Support, ,Supported languages}, for the default settings for
4926 @item set check type on
4927 @itemx set check type off
4928 Set type checking on or off, overriding the default setting for the
4929 current working language. Issue a warning if the setting does not
4930 match the language default. If any type mismatches occur in
4931 evaluating an expression while typechecking is on, @value{GDBN} prints a
4932 message and aborts evaluation of the expression.
4934 @item set check type warn
4935 Cause the type checker to issue warnings, but to always attempt to
4936 evaluate the expression. Evaluating the expression may still
4937 be impossible for other reasons. For example, @value{GDBN} cannot add
4938 numbers and structures.
4941 Show the current setting of the type checker, and whether or not @value{GDBN} is
4942 setting it automatically.
4945 @cindex range checking
4946 @cindex checks, range
4947 @node Range Checking
4948 @subsection An overview of range checking
4950 In some languages (such as Modula-2), it is an error to exceed the
4951 bounds of a type; this is enforced with run-time checks. Such range
4952 checking is meant to ensure program correctness by making sure
4953 computations do not overflow, or indices on an array element access do
4954 not exceed the bounds of the array.
4956 For expressions you use in @value{GDBN} commands, you can tell
4957 @value{GDBN} to treat range errors in one of three ways: ignore them,
4958 always treat them as errors and abandon the expression, or issue
4959 warnings but evaluate the expression anyway.
4961 A range error can result from numerical overflow, from exceeding an
4962 array index bound, or when you type a constant that is not a member
4963 of any type. Some languages, however, do not treat overflows as an
4964 error. In many implementations of C, mathematical overflow causes the
4965 result to ``wrap around'' to lower values---for example, if @var{m} is
4966 the largest integer value, and @var{s} is the smallest, then
4969 @var{m} + 1 @result{} @var{s}
4972 This, too, is specific to individual languages, and in some cases
4973 specific to individual compilers or machines. @xref{Support, ,
4974 Supported languages}, for further details on specific languages.
4976 @value{GDBN} provides some additional commands for controlling the range checker:
4979 @kindex set check range
4980 @kindex show check range
4982 @item set check range auto
4983 Set range checking on or off based on the current working language.
4984 @xref{Support, ,Supported languages}, for the default settings for
4987 @item set check range on
4988 @itemx set check range off
4989 Set range checking on or off, overriding the default setting for the
4990 current working language. A warning is issued if the setting does not
4991 match the language default. If a range error occurs, then a message
4992 is printed and evaluation of the expression is aborted.
4994 @item set check range warn
4995 Output messages when the @value{GDBN} range checker detects a range error,
4996 but attempt to evaluate the expression anyway. Evaluating the
4997 expression may still be impossible for other reasons, such as accessing
4998 memory that the process does not own (a typical example from many UNIX
5002 Show the current setting of the range checker, and whether or not it is
5003 being set automatically by @value{GDBN}.
5008 @section Supported languages
5011 @value{GDBN} 4 supports C, C++, and Modula-2.
5014 @value{GDBN} 4 supports C, and C++.
5016 Some @value{GDBN} features may be used in expressions regardless of the
5017 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
5018 and the @samp{@{type@}addr} construct (@pxref{Expressions,
5019 ,Expressions}) can be used with the constructs of any supported
5022 The following sections detail to what degree each source language is
5023 supported by @value{GDBN}. These sections are not meant to be language
5024 tutorials or references, but serve only as a reference guide to what the
5025 @value{GDBN} expression parser will accept, and what input and output
5026 formats should look like for different languages. There are many good
5027 books written on each of these languages; please look to these for a
5028 language reference or tutorial.
5033 * Modula-2:: Modula-2
5037 @subsection C and C++
5039 @cindex expressions in C or C++
5041 Since C and C++ are so closely related, many features of @value{GDBN} apply
5042 to both languages. Whenever this is the case, we discuss both languages
5046 @c Cancel this below, under same condition, at end of this chapter!
5053 The C++ debugging facilities are jointly implemented by the GNU C++
5054 compiler and @value{GDBN}. Therefore, to debug your C++ code effectively,
5055 you must compile your C++ programs with the GNU C++ compiler,
5060 @chapter C Language Support
5062 @cindex expressions in C
5064 Information specific to the C language is built into @value{GDBN} so that you
5065 can use C expressions while degugging. This also permits @value{GDBN} to
5066 output values in a manner consistent with C conventions.
5069 * C Operators:: C operators
5070 * C Constants:: C constants
5071 * Debugging C:: @value{GDBN} and C
5076 * C Operators:: C and C++ operators
5077 * C Constants:: C and C++ constants
5078 * Cplus expressions:: C++ expressions
5079 * C Defaults:: Default settings for C and C++
5081 * C Checks:: C and C++ type and range checks
5084 * Debugging C:: @value{GDBN} and C
5085 * Debugging C plus plus:: Special features for C++
5090 @cindex C and C++ operators
5092 @subsubsection C and C++ operators
5097 @section C operators
5100 Operators must be defined on values of specific types. For instance,
5101 @code{+} is defined on numbers, but not on structures. Operators are
5102 often defined on groups of types.
5105 For the purposes of C and C++, the following definitions hold:
5110 @emph{Integral types} include @code{int} with any of its storage-class
5111 specifiers; @code{char}; and @code{enum}.
5114 @emph{Floating-point types} include @code{float} and @code{double}.
5117 @emph{Pointer types} include all types defined as @code{(@var{type}
5121 @emph{Scalar types} include all of the above.
5125 The following operators are supported. They are listed here
5126 in order of increasing precedence:
5130 The comma or sequencing operator. Expressions in a comma-separated list
5131 are evaluated from left to right, with the result of the entire
5132 expression being the last expression evaluated.
5135 Assignment. The value of an assignment expression is the value
5136 assigned. Defined on scalar types.
5139 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
5140 and translated to @w{@code{@var{a} = @var{a op b}}}.
5141 @w{@code{@var{op}=}} and @code{=} have the same precendence.
5142 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
5143 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
5146 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
5147 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
5151 Logical @sc{or}. Defined on integral types.
5154 Logical @sc{and}. Defined on integral types.
5157 Bitwise @sc{or}. Defined on integral types.
5160 Bitwise exclusive-@sc{or}. Defined on integral types.
5163 Bitwise @sc{and}. Defined on integral types.
5166 Equality and inequality. Defined on scalar types. The value of these
5167 expressions is 0 for false and non-zero for true.
5169 @item <@r{, }>@r{, }<=@r{, }>=
5170 Less than, greater than, less than or equal, greater than or equal.
5171 Defined on scalar types. The value of these expressions is 0 for false
5172 and non-zero for true.
5175 left shift, and right shift. Defined on integral types.
5178 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5181 Addition and subtraction. Defined on integral types, floating-point types and
5184 @item *@r{, }/@r{, }%
5185 Multiplication, division, and modulus. Multiplication and division are
5186 defined on integral and floating-point types. Modulus is defined on
5190 Increment and decrement. When appearing before a variable, the
5191 operation is performed before the variable is used in an expression;
5192 when appearing after it, the variable's value is used before the
5193 operation takes place.
5196 Pointer dereferencing. Defined on pointer types. Same precedence as
5200 Address operator. Defined on variables. Same precedence as @code{++}.
5203 For debugging C++, @value{GDBN} implements a use of @samp{&} beyond what is
5204 allowed in the C++ language itself: you can use @samp{&(&@var{ref})}
5205 (or, if you prefer, simply @samp{&&@var{ref}} to examine the address
5206 where a C++ reference variable (declared with @samp{&@var{ref}}) is
5211 Negative. Defined on integral and floating-point types. Same
5212 precedence as @code{++}.
5215 Logical negation. Defined on integral types. Same precedence as
5219 Bitwise complement operator. Defined on integral types. Same precedence as
5224 Structure member, and pointer-to-structure member. For convenience,
5225 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
5226 pointer based on the stored type information.
5227 Defined on @code{struct} and @code{union} data.
5230 Array indexing. @code{@var{a}[@var{i}]} is defined as
5231 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
5234 Function parameter list. Same precedence as @code{->}.
5238 C++ scope resolution operator. Defined on
5239 @code{struct}, @code{union}, and @code{class} types.
5247 represent the @value{GDBN} scope operator (@pxref{Expressions,
5250 Same precedence as @code{::}, above.
5255 @cindex C and C++ constants
5257 @subsubsection C and C++ constants
5259 @value{GDBN} allows you to express the constants of C and C++ in the
5265 @section C constants
5267 @value{GDBN} allows you to express the constants of C in the
5273 Integer constants are a sequence of digits. Octal constants are
5274 specified by a leading @samp{0} (ie. zero), and hexadecimal constants by
5275 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
5276 @samp{l}, specifying that the constant should be treated as a
5280 Floating point constants are a sequence of digits, followed by a decimal
5281 point, followed by a sequence of digits, and optionally followed by an
5282 exponent. An exponent is of the form:
5283 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
5284 sequence of digits. The @samp{+} is optional for positive exponents.
5287 Enumerated constants consist of enumerated identifiers, or their
5288 integral equivalents.
5291 Character constants are a single character surrounded by single quotes
5292 (@code{'}), or a number---the ordinal value of the corresponding character
5293 (usually its @sc{ASCII} value). Within quotes, the single character may
5294 be represented by a letter or by @dfn{escape sequences}, which are of
5295 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
5296 of the character's ordinal value; or of the form @samp{\@var{x}}, where
5297 @samp{@var{x}} is a predefined special character---for example,
5298 @samp{\n} for newline.
5301 String constants are a sequence of character constants surrounded
5302 by double quotes (@code{"}).
5305 Pointer constants are an integral value. You can also write pointers
5306 to constants using the C operator @samp{&}.
5309 Array constants are comma-separated lists surrounded by braces @samp{@{}
5310 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
5311 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
5312 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
5316 @node Cplus expressions
5317 @subsubsection C++ expressions
5319 @cindex expressions in C++
5320 @value{GDBN} expression handling has a number of extensions to
5321 interpret a significant subset of C++ expressions.
5323 @cindex C++ support, not in @sc{coff}
5324 @cindex @sc{coff} versus C++
5325 @cindex C++ and object formats
5326 @cindex object formats and C++
5327 @cindex a.out and C++
5328 @cindex @sc{ecoff} and C++
5329 @cindex @sc{xcoff} and C++
5330 @cindex @sc{elf}/stabs and C++
5331 @cindex @sc{elf}/@sc{dwarf} and C++
5333 @emph{Warning:} Most of these extensions depend on the use of additional
5334 debugging information in the symbol table, and thus require a rich,
5335 extendable object code format. In particular, if your system uses
5336 a.out, MIPS @sc{ecoff}, RS/6000 @sc{xcoff}, or Sun @sc{elf} with stabs
5337 extensions to the symbol table, these facilities are all available.
5338 Where the object code format is standard @sc{coff}, on the other hand,
5339 most of the C++ support in @value{GDBN} will @emph{not} work, nor can it.
5340 For the standard SVr4 debugging format, @sc{dwarf} in @sc{elf}, the
5341 standard is still evolving, so the C++ support in @value{GDBN} is still
5342 fragile; when this debugging format stabilizes, however, C++ support
5343 will also be available on systems that use it.
5348 @cindex member functions
5350 Member function calls are allowed; you can use expressions like
5353 count = aml->GetOriginal(x, y)
5357 @cindex namespace in C++
5359 While a member function is active (in the selected stack frame), your
5360 expressions have the same namespace available as the member function;
5361 that is, @value{GDBN} allows implicit references to the class instance
5362 pointer @code{this} following the same rules as C++.
5364 @cindex call overloaded functions
5365 @cindex type conversions in C++
5367 You can call overloaded functions; @value{GDBN} will resolve the function
5368 call to the right definition, with one restriction---you must use
5369 arguments of the type required by the function that you want to call.
5370 @value{GDBN} will not perform conversions requiring constructors or
5371 user-defined type operators.
5373 @cindex reference declarations
5375 @value{GDBN} understands variables declared as C++ references; you can use them in
5376 expressions just as you do in C++ source---they are automatically
5379 In the parameter list shown when @value{GDBN} displays a frame, the values of
5380 reference variables are not displayed (unlike other variables); this
5381 avoids clutter, since references are often used for large structures.
5382 The @emph{address} of a reference variable is always shown, unless
5383 you have specified @samp{set print address off}.
5386 @value{GDBN} supports the C++ name resolution operator @code{::}---your
5387 expressions can use it just as expressions in your program do. Since
5388 one scope may be defined in another, you can use @code{::} repeatedly if
5389 necessary, for example in an expression like
5390 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
5391 resolving name scope by reference to source files, in both C and C++
5392 debugging (@pxref{Variables, ,Program variables}).
5396 @subsubsection C and C++ defaults
5397 @cindex C and C++ defaults
5399 If you allow @value{GDBN} to set type and range checking automatically, they
5400 both default to @code{off} whenever the working language changes to
5401 C or C++. This happens regardless of whether you, or @value{GDBN},
5402 selected the working language.
5404 If you allow @value{GDBN} to set the language automatically, it sets the
5405 working language to C or C++ on entering code compiled from a source file
5406 whose name ends with @file{.c}, @file{.C}, or @file{.cc}.
5407 @xref{Automatically, ,Having @value{GDBN} infer the source language}, for
5411 @c Type checking is (a) primarily motivated by Modula-2, and (b)
5412 @c unimplemented. If (b) changes, it might make sense to let this node
5413 @c appear even if Mod-2 does not, but meanwhile ignore it. pesch 16jul93.
5415 @subsubsection C and C++ type and range checks
5416 @cindex C and C++ checks
5418 By default, when @value{GDBN} parses C or C++ expressions, type checking
5419 is not used. However, if you turn type checking on, @value{GDBN} will
5420 consider two variables type equivalent if:
5424 The two variables are structured and have the same structure, union, or
5428 Two two variables have the same type name, or types that have been
5429 declared equivalent through @code{typedef}.
5432 @c leaving this out because neither J Gilmore nor R Pesch understand it.
5435 The two @code{struct}, @code{union}, or @code{enum} variables are
5436 declared in the same declaration. (Note: this may not be true for all C
5441 Range checking, if turned on, is done on mathematical operations. Array
5442 indices are not checked, since they are often used to index a pointer
5443 that is not itself an array.
5449 @subsubsection @value{GDBN} and C
5453 @section @value{GDBN} and C
5456 The @code{set print union} and @code{show print union} commands apply to
5457 the @code{union} type. When set to @samp{on}, any @code{union} that is
5458 inside a @code{struct}
5462 will also be printed.
5463 Otherwise, it will appear as @samp{@{...@}}.
5465 The @code{@@} operator aids in the debugging of dynamic arrays, formed
5466 with pointers and a memory allocation function. @xref{Expressions,
5470 @node Debugging C plus plus
5471 @subsubsection @value{GDBN} features for C++
5473 @cindex commands for C++
5474 Some @value{GDBN} commands are particularly useful with C++, and some are
5475 designed specifically for use with C++. Here is a summary:
5478 @cindex break in overloaded functions
5479 @item @r{breakpoint menus}
5480 When you want a breakpoint in a function whose name is overloaded,
5481 @value{GDBN} breakpoint menus help you specify which function definition
5482 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
5484 @cindex overloading in C++
5485 @item rbreak @var{regex}
5486 Setting breakpoints using regular expressions is helpful for setting
5487 breakpoints on overloaded functions that are not members of any special
5489 @xref{Set Breaks, ,Setting breakpoints}.
5491 @cindex C++ exception handling
5492 @item catch @var{exceptions}
5494 Debug C++ exception handling using these commands. @xref{Exception
5495 Handling, ,Breakpoints and exceptions}.
5498 @item ptype @var{typename}
5499 Print inheritance relationships as well as other information for type
5501 @xref{Symbols, ,Examining the Symbol Table}.
5503 @cindex C++ symbol display
5504 @item set print demangle
5505 @itemx show print demangle
5506 @itemx set print asm-demangle
5507 @itemx show print asm-demangle
5508 Control whether C++ symbols display in their source form, both when
5509 displaying code as C++ source and when displaying disassemblies.
5510 @xref{Print Settings, ,Print settings}.
5512 @item set print object
5513 @itemx show print object
5514 Choose whether to print derived (actual) or declared types of objects.
5515 @xref{Print Settings, ,Print settings}.
5517 @item set print vtbl
5518 @itemx show print vtbl
5519 Control the format for printing virtual function tables.
5520 @xref{Print Settings, ,Print settings}.
5522 @item @r{Overloaded symbol names}
5523 You can specify a particular definition of an overloaded symbol, using
5524 the same notation that is used to declare such symbols in C++: type
5525 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
5526 also use the @value{GDBN} command-line word completion facilities to list the
5527 available choices, or to finish the type list for you.
5528 @xref{Completion,, Command completion}, for details on how to do this.
5531 @c cancels "raisesections" under same conditions near bgn of chapter
5537 @subsection Modula-2
5540 The extensions made to @value{GDBN} to support Modula-2 only support
5541 output from the GNU Modula-2 compiler (which is currently being
5542 developed). Other Modula-2 compilers are not currently supported, and
5543 attempting to debug executables produced by them will most likely
5544 result in an error as @value{GDBN} reads in the executable's symbol
5547 @cindex expressions in Modula-2
5549 * M2 Operators:: Built-in operators
5550 * Built-In Func/Proc:: Built-in functions and procedures
5551 * M2 Constants:: Modula-2 constants
5552 * M2 Defaults:: Default settings for Modula-2
5553 * Deviations:: Deviations from standard Modula-2
5554 * M2 Checks:: Modula-2 type and range checks
5555 * M2 Scope:: The scope operators @code{::} and @code{.}
5556 * GDB/M2:: @value{GDBN} and Modula-2
5560 @subsubsection Operators
5561 @cindex Modula-2 operators
5563 Operators must be defined on values of specific types. For instance,
5564 @code{+} is defined on numbers, but not on structures. Operators are
5565 often defined on groups of types. For the purposes of Modula-2, the
5566 following definitions hold:
5571 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
5575 @emph{Character types} consist of @code{CHAR} and its subranges.
5578 @emph{Floating-point types} consist of @code{REAL}.
5581 @emph{Pointer types} consist of anything declared as @code{POINTER TO
5585 @emph{Scalar types} consist of all of the above.
5588 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
5591 @emph{Boolean types} consist of @code{BOOLEAN}.
5595 The following operators are supported, and appear in order of
5596 increasing precedence:
5600 Function argument or array index separator.
5603 Assignment. The value of @var{var} @code{:=} @var{value} is
5607 Less than, greater than on integral, floating-point, or enumerated
5611 Less than, greater than, less than or equal to, greater than or equal to
5612 on integral, floating-point and enumerated types, or set inclusion on
5613 set types. Same precedence as @code{<}.
5615 @item =@r{, }<>@r{, }#
5616 Equality and two ways of expressing inequality, valid on scalar types.
5617 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
5618 available for inequality, since @code{#} conflicts with the script
5622 Set membership. Defined on set types and the types of their members.
5623 Same precedence as @code{<}.
5626 Boolean disjunction. Defined on boolean types.
5629 Boolean conjuction. Defined on boolean types.
5632 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5635 Addition and subtraction on integral and floating-point types, or union
5636 and difference on set types.
5639 Multiplication on integral and floating-point types, or set intersection
5643 Division on floating-point types, or symmetric set difference on set
5644 types. Same precedence as @code{*}.
5647 Integer division and remainder. Defined on integral types. Same
5648 precedence as @code{*}.
5651 Negative. Defined on @code{INTEGER} and @code{REAL} data.
5654 Pointer dereferencing. Defined on pointer types.
5657 Boolean negation. Defined on boolean types. Same precedence as
5661 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
5662 precedence as @code{^}.
5665 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
5668 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
5672 @value{GDBN} and Modula-2 scope operators.
5676 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
5677 will treat the use of the operator @code{IN}, or the use of operators
5678 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
5679 @code{<=}, and @code{>=} on sets as an error.
5682 @cindex Modula-2 built-ins
5683 @node Built-In Func/Proc
5684 @subsubsection Built-in functions and procedures
5686 Modula-2 also makes available several built-in procedures and functions.
5687 In describing these, the following metavariables are used:
5692 represents an @code{ARRAY} variable.
5695 represents a @code{CHAR} constant or variable.
5698 represents a variable or constant of integral type.
5701 represents an identifier that belongs to a set. Generally used in the
5702 same function with the metavariable @var{s}. The type of @var{s} should
5703 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}.
5706 represents a variable or constant of integral or floating-point type.
5709 represents a variable or constant of floating-point type.
5715 represents a variable.
5718 represents a variable or constant of one of many types. See the
5719 explanation of the function for details.
5722 All Modula-2 built-in procedures also return a result, described below.
5726 Returns the absolute value of @var{n}.
5729 If @var{c} is a lower case letter, it returns its upper case
5730 equivalent, otherwise it returns its argument
5733 Returns the character whose ordinal value is @var{i}.
5736 Decrements the value in the variable @var{v}. Returns the new value.
5738 @item DEC(@var{v},@var{i})
5739 Decrements the value in the variable @var{v} by @var{i}. Returns the
5742 @item EXCL(@var{m},@var{s})
5743 Removes the element @var{m} from the set @var{s}. Returns the new
5746 @item FLOAT(@var{i})
5747 Returns the floating point equivalent of the integer @var{i}.
5750 Returns the index of the last member of @var{a}.
5753 Increments the value in the variable @var{v}. Returns the new value.
5755 @item INC(@var{v},@var{i})
5756 Increments the value in the variable @var{v} by @var{i}. Returns the
5759 @item INCL(@var{m},@var{s})
5760 Adds the element @var{m} to the set @var{s} if it is not already
5761 there. Returns the new set.
5764 Returns the maximum value of the type @var{t}.
5767 Returns the minimum value of the type @var{t}.
5770 Returns boolean TRUE if @var{i} is an odd number.
5773 Returns the ordinal value of its argument. For example, the ordinal
5774 value of a character is its ASCII value (on machines supporting the
5775 ASCII character set). @var{x} must be of an ordered type, which include
5776 integral, character and enumerated types.
5779 Returns the size of its argument. @var{x} can be a variable or a type.
5781 @item TRUNC(@var{r})
5782 Returns the integral part of @var{r}.
5784 @item VAL(@var{t},@var{i})
5785 Returns the member of the type @var{t} whose ordinal value is @var{i}.
5789 @emph{Warning:} Sets and their operations are not yet supported, so
5790 @value{GDBN} will treat the use of procedures @code{INCL} and @code{EXCL} as
5794 @cindex Modula-2 constants
5796 @subsubsection Constants
5798 @value{GDBN} allows you to express the constants of Modula-2 in the following
5804 Integer constants are simply a sequence of digits. When used in an
5805 expression, a constant is interpreted to be type-compatible with the
5806 rest of the expression. Hexadecimal integers are specified by a
5807 trailing @samp{H}, and octal integers by a trailing @samp{B}.
5810 Floating point constants appear as a sequence of digits, followed by a
5811 decimal point and another sequence of digits. An optional exponent can
5812 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
5813 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
5814 digits of the floating point constant must be valid decimal (base 10)
5818 Character constants consist of a single character enclosed by a pair of
5819 like quotes, either single (@code{'}) or double (@code{"}). They may
5820 also be expressed by their ordinal value (their ASCII value, usually)
5821 followed by a @samp{C}.
5824 String constants consist of a sequence of characters enclosed by a
5825 pair of like quotes, either single (@code{'}) or double (@code{"}).
5826 Escape sequences in the style of C are also allowed. @xref{C
5827 Constants, ,C and C++ constants}, for a brief explanation of escape
5831 Enumerated constants consist of an enumerated identifier.
5834 Boolean constants consist of the identifiers @code{TRUE} and
5838 Pointer constants consist of integral values only.
5841 Set constants are not yet supported.
5845 @subsubsection Modula-2 defaults
5846 @cindex Modula-2 defaults
5848 If type and range checking are set automatically by @value{GDBN}, they
5849 both default to @code{on} whenever the working language changes to
5850 Modula-2. This happens regardless of whether you, or @value{GDBN},
5851 selected the working language.
5853 If you allow @value{GDBN} to set the language automatically, then entering
5854 code compiled from a file whose name ends with @file{.mod} will set the
5855 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
5856 the language automatically}, for further details.
5859 @subsubsection Deviations from standard Modula-2
5860 @cindex Modula-2, deviations from
5862 A few changes have been made to make Modula-2 programs easier to debug.
5863 This is done primarily via loosening its type strictness:
5867 Unlike in standard Modula-2, pointer constants can be formed by
5868 integers. This allows you to modify pointer variables during
5869 debugging. (In standard Modula-2, the actual address contained in a
5870 pointer variable is hidden from you; it can only be modified
5871 through direct assignment to another pointer variable or expression that
5872 returned a pointer.)
5875 C escape sequences can be used in strings and characters to represent
5876 non-printable characters. @value{GDBN} will print out strings with these
5877 escape sequences embedded. Single non-printable characters are
5878 printed using the @samp{CHR(@var{nnn})} format.
5881 The assignment operator (@code{:=}) returns the value of its right-hand
5885 All built-in procedures both modify @emph{and} return their argument.
5889 @subsubsection Modula-2 type and range checks
5890 @cindex Modula-2 checks
5893 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
5896 @c FIXME remove warning when type/range checks added
5898 @value{GDBN} considers two Modula-2 variables type equivalent if:
5902 They are of types that have been declared equivalent via a @code{TYPE
5903 @var{t1} = @var{t2}} statement
5906 They have been declared on the same line. (Note: This is true of the
5907 GNU Modula-2 compiler, but it may not be true of other compilers.)
5910 As long as type checking is enabled, any attempt to combine variables
5911 whose types are not equivalent is an error.
5913 Range checking is done on all mathematical operations, assignment, array
5914 index bounds, and all built-in functions and procedures.
5917 @subsubsection The scope operators @code{::} and @code{.}
5920 @cindex colon, doubled as scope operator
5923 @c Info cannot handle :: but TeX can.
5929 There are a few subtle differences between the Modula-2 scope operator
5930 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
5935 @var{module} . @var{id}
5936 @var{scope} :: @var{id}
5940 where @var{scope} is the name of a module or a procedure,
5941 @var{module} the name of a module, and @var{id} is any declared
5942 identifier within your program, except another module.
5944 Using the @code{::} operator makes @value{GDBN} search the scope
5945 specified by @var{scope} for the identifier @var{id}. If it is not
5946 found in the specified scope, then @value{GDBN} will search all scopes
5947 enclosing the one specified by @var{scope}.
5949 Using the @code{.} operator makes @value{GDBN} search the current scope for
5950 the identifier specified by @var{id} that was imported from the
5951 definition module specified by @var{module}. With this operator, it is
5952 an error if the identifier @var{id} was not imported from definition
5953 module @var{module}, or if @var{id} is not an identifier in
5957 @subsubsection @value{GDBN} and Modula-2
5959 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
5960 Five subcommands of @code{set print} and @code{show print} apply
5961 specifically to C and C++: @samp{vtbl}, @samp{demangle},
5962 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
5963 apply to C++, and the last to the C @code{union} type, which has no direct
5964 analogue in Modula-2.
5966 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
5967 while using any language, is not useful with Modula-2. Its
5968 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
5969 created in Modula-2 as they can in C or C++. However, because an
5970 address can be specified by an integral constant, the construct
5971 @samp{@{@var{type}@}@var{adrexp}} is still useful. (@pxref{Expressions, ,Expressions})
5973 @cindex @code{#} in Modula-2
5974 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
5975 interpreted as the beginning of a comment. Use @code{<>} instead.
5981 @chapter Examining the Symbol Table
5983 The commands described in this section allow you to inquire about the
5984 symbols (names of variables, functions and types) defined in your
5985 program. This information is inherent in the text of your program and
5986 does not change as your program executes. @value{GDBN} finds it in your
5987 program's symbol table, in the file indicated when you started @value{GDBN}
5988 (@pxref{File Options, ,Choosing files}), or by one of the
5989 file-management commands (@pxref{Files, ,Commands to specify files}).
5991 @c FIXME! This might be intentionally specific to C and C++; if so, move
5992 @c to someplace in C section of lang chapter.
5993 @cindex symbol names
5994 @cindex names of symbols
5995 @cindex quoting names
5996 Occasionally, you may need to refer to symbols that contain unusual
5997 characters, which @value{GDBN} ordinarily treats as word delimiters. The
5998 most frequent case is in referring to static variables in other
5999 source files (@pxref{Variables,,Program variables}). File names
6000 are recorded in object files as debugging symbols, but @value{GDBN} would
6001 ordinarily parse a typical file name, like @file{foo.c}, as the three words
6002 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
6003 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
6010 looks up the value of @code{x} in the scope of the file @file{foo.c}.
6013 @item info address @var{symbol}
6014 @kindex info address
6015 Describe where the data for @var{symbol} is stored. For a register
6016 variable, this says which register it is kept in. For a non-register
6017 local variable, this prints the stack-frame offset at which the variable
6020 Note the contrast with @samp{print &@var{symbol}}, which does not work
6021 at all for a register variables, and for a stack local variable prints
6022 the exact address of the current instantiation of the variable.
6024 @item whatis @var{exp}
6026 Print the data type of expression @var{exp}. @var{exp} is not
6027 actually evaluated, and any side-effecting operations (such as
6028 assignments or function calls) inside it do not take place.
6029 @xref{Expressions, ,Expressions}.
6032 Print the data type of @code{$}, the last value in the value history.
6034 @item ptype @var{typename}
6036 Print a description of data type @var{typename}. @var{typename} may be
6037 the name of a type, or for C code it may have the form
6038 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
6039 @samp{enum @var{enum-tag}}.
6041 @item ptype @var{exp}
6043 Print a description of the type of expression @var{exp}. @code{ptype}
6044 differs from @code{whatis} by printing a detailed description, instead
6045 of just the name of the type.
6047 For example, for this variable declaration:
6050 struct complex @{double real; double imag;@} v;
6054 the two commands give this output:
6058 (@value{GDBP}) whatis v
6059 type = struct complex
6060 (@value{GDBP}) ptype v
6061 type = struct complex @{
6069 As with @code{whatis}, using @code{ptype} without an argument refers to
6070 the type of @code{$}, the last value in the value history.
6072 @item info types @var{regexp}
6075 Print a brief description of all types whose name matches @var{regexp}
6076 (or all types in your program, if you supply no argument). Each
6077 complete typename is matched as though it were a complete line; thus,
6078 @samp{i type value} gives information on all types in your program whose
6079 name includes the string @code{value}, but @samp{i type ^value$} gives
6080 information only on types whose complete name is @code{value}.
6082 This command differs from @code{ptype} in two ways: first, like
6083 @code{whatis}, it does not print a detailed description; second, it
6084 lists all source files where a type is defined.
6088 Show the name of the current source file---that is, the source file for
6089 the function containing the current point of execution---and the language
6093 @kindex info sources
6094 Print the names of all source files in your program for which there is
6095 debugging information, organized into two lists: files whose symbols
6096 have already been read, and files whose symbols will be read when needed.
6098 @item info functions
6099 @kindex info functions
6100 Print the names and data types of all defined functions.
6102 @item info functions @var{regexp}
6103 Print the names and data types of all defined functions
6104 whose names contain a match for regular expression @var{regexp}.
6105 Thus, @samp{info fun step} finds all functions whose names
6106 include @code{step}; @samp{info fun ^step} finds those whose names
6107 start with @code{step}.
6109 @item info variables
6110 @kindex info variables
6111 Print the names and data types of all variables that are declared
6112 outside of functions (i.e., excluding local variables).
6114 @item info variables @var{regexp}
6115 Print the names and data types of all variables (except for local
6116 variables) whose names contain a match for regular expression
6120 This was never implemented.
6122 @itemx info methods @var{regexp}
6123 @kindex info methods
6124 The @code{info methods} command permits the user to examine all defined
6125 methods within C++ program, or (with the @var{regexp} argument) a
6126 specific set of methods found in the various C++ classes. Many
6127 C++ classes provide a large number of methods. Thus, the output
6128 from the @code{ptype} command can be overwhelming and hard to use. The
6129 @code{info-methods} command filters the methods, printing only those
6130 which match the regular-expression @var{regexp}.
6133 @item maint print symbols @var{filename}
6134 @itemx maint print psymbols @var{filename}
6135 @itemx maint print msymbols @var{filename}
6136 @kindex maint print symbols
6138 @kindex maint print psymbols
6139 @cindex partial symbol dump
6140 Write a dump of debugging symbol data into the file @var{filename}.
6141 These commands are used to debug the @value{GDBN} symbol-reading code. Only
6142 symbols with debugging data are included. If you use @samp{maint print
6143 symbols}, @value{GDBN} includes all the symbols for which it has already
6144 collected full details: that is, @var{filename} reflects symbols for
6145 only those files whose symbols @value{GDBN} has read. You can use the
6146 command @code{info sources} to find out which files these are. If you
6147 use @samp{maint print psymbols} instead, the dump shows information about
6148 symbols that @value{GDBN} only knows partially---that is, symbols defined in
6149 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
6150 @samp{maint print msymbols} dumps just the minimal symbol information
6151 required for each object file from which @value{GDBN} has read some symbols.
6152 @xref{Files, ,Commands to specify files}, for a discussion of how
6153 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
6157 @chapter Altering Execution
6159 Once you think you have found an error in your program, you might want to
6160 find out for certain whether correcting the apparent error would lead to
6161 correct results in the rest of the run. You can find the answer by
6162 experiment, using the @value{GDBN} features for altering execution of the
6165 For example, you can store new values into variables or memory
6168 give your program a signal, restart it
6171 restart your program
6173 at a different address, or even return prematurely from a function to
6177 * Assignment:: Assignment to variables
6178 * Jumping:: Continuing at a different address
6180 * Signaling:: Giving your program a signal
6183 * Returning:: Returning from a function
6184 * Calling:: Calling your program's functions
6185 * Patching:: Patching your program
6189 @section Assignment to variables
6192 @cindex setting variables
6193 To alter the value of a variable, evaluate an assignment expression.
6194 @xref{Expressions, ,Expressions}. For example,
6201 stores the value 4 into the variable @code{x}, and then prints the
6202 value of the assignment expression (which is 4).
6204 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
6205 information on operators in supported languages.
6208 @kindex set variable
6209 @cindex variables, setting
6210 If you are not interested in seeing the value of the assignment, use the
6211 @code{set} command instead of the @code{print} command. @code{set} is
6212 really the same as @code{print} except that the expression's value is
6213 not printed and is not put in the value history (@pxref{Value History,
6214 ,Value history}). The expression is evaluated only for its effects.
6216 If the beginning of the argument string of the @code{set} command
6217 appears identical to a @code{set} subcommand, use the @code{set
6218 variable} command instead of just @code{set}. This command is identical
6219 to @code{set} except for its lack of subcommands. For example, if
6220 your program has a variable @code{width}, you get
6221 an error if you try to set a new value with just @samp{set width=13},
6222 because @value{GDBN} has the command @code{set width}:
6225 (@value{GDBP}) whatis width
6227 (@value{GDBP}) p width
6229 (@value{GDBP}) set width=47
6230 Invalid syntax in expression.
6234 The invalid expression, of course, is @samp{=47}. In
6235 order to actually set the program's variable @code{width}, use
6238 (@value{GDBP}) set var width=47
6241 @value{GDBN} allows more implicit conversions in assignments than C; you can
6242 freely store an integer value into a pointer variable or vice versa,
6243 and you can convert any structure to any other structure that is the
6244 same length or shorter.
6245 @comment FIXME: how do structs align/pad in these conversions?
6246 @comment /pesch@cygnus.com 18dec1990
6248 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
6249 construct to generate a value of specified type at a specified address
6250 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
6251 to memory location @code{0x83040} as an integer (which implies a certain size
6252 and representation in memory), and
6255 set @{int@}0x83040 = 4
6259 stores the value 4 into that memory location.
6262 @section Continuing at a different address
6264 Ordinarily, when you continue your program, you do so at the place where
6265 it stopped, with the @code{continue} command. You can instead continue at
6266 an address of your own choosing, with the following commands:
6269 @item jump @var{linespec}
6271 Resume execution at line @var{linespec}. Execution will stop
6272 immediately if there is a breakpoint there. @xref{List, ,Printing
6273 source lines}, for a description of the different forms of
6276 The @code{jump} command does not change the current stack frame, or
6277 the stack pointer, or the contents of any memory location or any
6278 register other than the program counter. If line @var{linespec} is in
6279 a different function from the one currently executing, the results may
6280 be bizarre if the two functions expect different patterns of arguments or
6281 of local variables. For this reason, the @code{jump} command requests
6282 confirmation if the specified line is not in the function currently
6283 executing. However, even bizarre results are predictable if you are
6284 well acquainted with the machine-language code of your program.
6286 @item jump *@var{address}
6287 Resume execution at the instruction at address @var{address}.
6290 You can get much the same effect as the @code{jump} command by storing a
6291 new value into the register @code{$pc}. The difference is that this
6292 does not start your program running; it only changes the address where it
6293 @emph{will} run when it is continued. For example,
6300 causes the next @code{continue} command or stepping command to execute at
6301 address @code{0x485}, rather than at the address where your program stopped.
6302 @xref{Continuing and Stepping, ,Continuing and stepping}.
6304 The most common occasion to use the @code{jump} command is to back up,
6305 perhaps with more breakpoints set, over a portion of a program that has
6306 already executed, in order to examine its execution in more detail.
6311 @section Giving your program a signal
6314 @item signal @var{signalnum}
6316 Resume execution where your program stopped, but give it immediately the
6317 signal number @var{signalnum}.
6319 Alternatively, if @var{signalnum} is zero, continue execution without
6320 giving a signal. This is useful when your program stopped on account of
6321 a signal and would ordinary see the signal when resumed with the
6322 @code{continue} command; @samp{signal 0} causes it to resume without a
6325 @code{signal} does not repeat when you press @key{RET} a second time
6326 after executing the command.
6332 @section Returning from a function
6336 @itemx return @var{expression}
6337 @cindex returning from a function
6339 You can cancel execution of a function call with the @code{return}
6340 command. If you give an
6341 @var{expression} argument, its value is used as the function's return
6345 When you use @code{return}, @value{GDBN} discards the selected stack frame
6346 (and all frames within it). You can think of this as making the
6347 discarded frame return prematurely. If you wish to specify a value to
6348 be returned, give that value as the argument to @code{return}.
6350 This pops the selected stack frame (@pxref{Selection, ,Selecting a
6351 frame}), and any other frames inside of it, leaving its caller as the
6352 innermost remaining frame. That frame becomes selected. The
6353 specified value is stored in the registers used for returning values
6356 The @code{return} command does not resume execution; it leaves the
6357 program stopped in the state that would exist if the function had just
6358 returned. In contrast, the @code{finish} command (@pxref{Continuing
6359 and Stepping, ,Continuing and stepping}) resumes execution until the
6360 selected stack frame returns naturally.
6363 @section Calling program functions
6365 @cindex calling functions
6368 @item call @var{expr}
6369 Evaluate the expression @var{expr} without displaying @code{void}
6373 You can use this variant of the @code{print} command if you want to
6374 execute a function from your program, but without cluttering the output
6375 with @code{void} returned values. The result is printed and saved in
6376 the value history, if it is not void.
6379 @section Patching programs
6380 @cindex patching binaries
6381 @cindex writing into executables
6383 @cindex writing into corefiles
6386 By default, @value{GDBN} opens the file containing your program's executable
6391 read-only. This prevents accidental alterations
6392 to machine code; but it also prevents you from intentionally patching
6393 your program's binary.
6395 If you'd like to be able to patch the binary, you can specify that
6396 explicitly with the @code{set write} command. For example, you might
6397 want to turn on internal debugging flags, or even to make emergency
6402 @itemx set write off
6404 If you specify @samp{set write on}, @value{GDBN} will open executable
6408 files for both reading and writing; if you specify @samp{set write
6409 off} (the default), @value{GDBN} will open them read-only.
6411 If you have already loaded a file, you must load it again (using the
6416 command) after changing @code{set write}, for your new setting to take
6421 Display whether executable files
6425 will be opened for writing as well as reading.
6429 @chapter @value{GDBN} Files
6431 @value{GDBN} needs to know the file name of the program to be debugged, both in
6432 order to read its symbol table and in order to start your program.
6434 To debug a core dump of a previous run, you must also tell @value{GDBN}
6435 the name of the core dump file.
6439 * Files:: Commands to specify files
6440 * Symbol Errors:: Errors reading symbol files
6444 @section Commands to specify files
6445 @cindex symbol table
6448 @cindex core dump file
6449 The usual way to specify executable and core dump file names is with
6450 the command arguments given when you start @value{GDBN} (@pxref{Invocation,
6451 ,Getting In and Out of @value{GDBN}}.
6454 The usual way to specify an executable file name is with
6455 the command argument given when you start @value{GDBN}, (@pxref{Invocation,
6456 ,Getting In and Out of @value{GDBN}}.
6459 Occasionally it is necessary to change to a different file during a
6460 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
6461 a file you want to use. In these situations the @value{GDBN} commands
6462 to specify new files are useful.
6465 @item file @var{filename}
6466 @cindex executable file
6468 Use @var{filename} as the program to be debugged. It is read for its
6469 symbols and for the contents of pure memory. It is also the program
6470 executed when you use the @code{run} command. If you do not specify a
6471 directory and the file is not found in the @value{GDBN} working directory, @value{GDBN}
6472 uses the environment variable @code{PATH} as a list of directories to
6473 search, just as the shell does when looking for a program to run. You
6474 can change the value of this variable, for both @value{GDBN} and your program,
6475 using the @code{path} command.
6477 On systems with memory-mapped files, an auxiliary symbol table file
6478 @file{@var{filename}.syms} may be available for @var{filename}. If it
6479 is, @value{GDBN} will map in the symbol table from
6480 @file{@var{filename}.syms}, starting up more quickly. See the
6481 descriptions of the options @samp{-mapped} and @samp{-readnow} (available
6482 on the command line, and with the commands @code{file}, @code{symbol-file},
6483 or @code{add-symbol-file}), for more information.
6486 @code{file} with no argument makes @value{GDBN} discard any information it
6487 has on both executable file and the symbol table.
6489 @item exec-file @r{[} @var{filename} @r{]}
6491 Specify that the program to be run (but not the symbol table) is found
6492 in @var{filename}. @value{GDBN} will search the environment variable @code{PATH}
6493 if necessary to locate your program. Omitting @var{filename} means to
6494 discard information on the executable file.
6496 @item symbol-file @r{[} @var{filename} @r{]}
6498 Read symbol table information from file @var{filename}. @code{PATH} is
6499 searched when necessary. Use the @code{file} command to get both symbol
6500 table and program to run from the same file.
6502 @code{symbol-file} with no argument clears out @value{GDBN} information on your
6503 program's symbol table.
6505 The @code{symbol-file} command causes @value{GDBN} to forget the contents of its
6506 convenience variables, the value history, and all breakpoints and
6507 auto-display expressions. This is because they may contain pointers to
6508 the internal data recording symbols and data types, which are part of
6509 the old symbol table data being discarded inside @value{GDBN}.
6511 @code{symbol-file} will not repeat if you press @key{RET} again after
6514 When @value{GDBN} is configured for a particular environment, it will
6515 understand debugging information in whatever format is the standard
6516 generated for that environment; you may use either a GNU compiler, or
6517 other compilers that adhere to the local conventions. Best results are
6518 usually obtained from GNU compilers; for example, using @code{@value{GCC}}
6519 you can generate debugging information for optimized code.
6521 On some kinds of object files, the @code{symbol-file} command does not
6522 normally read the symbol table in full right away. Instead, it scans
6523 the symbol table quickly to find which source files and which symbols
6524 are present. The details are read later, one source file at a time,
6527 The purpose of this two-stage reading strategy is to make @value{GDBN} start up
6528 faster. For the most part, it is invisible except for occasional
6529 pauses while the symbol table details for a particular source file are
6530 being read. (The @code{set verbose} command can turn these pauses
6531 into messages if desired. @xref{Messages/Warnings, ,Optional warnings
6534 We have not implemented the two-stage strategy for COFF yet. When the
6535 symbol table is stored in COFF format, @code{symbol-file} reads the
6536 symbol table data in full right away.
6538 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
6539 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
6541 @cindex reading symbols immediately
6542 @cindex symbols, reading immediately
6544 @cindex memory-mapped symbol file
6545 @cindex saving symbol table
6546 You can override the @value{GDBN} two-stage strategy for reading symbol
6547 tables by using the @samp{-readnow} option with any of the commands that
6548 load symbol table information, if you want to be sure @value{GDBN} has the
6549 entire symbol table available.
6552 If memory-mapped files are available on your system through the
6553 @code{mmap} system call, you can use another option, @samp{-mapped}, to
6554 cause @value{GDBN} to write the symbols for your program into a reusable
6555 file. Future @value{GDBN} debugging sessions will map in symbol information
6556 from this auxiliary symbol file (if the program has not changed), rather
6557 than spending time reading the symbol table from the executable
6558 program. Using the @samp{-mapped} option has the same effect as
6559 starting @value{GDBN} with the @samp{-mapped} command-line option.
6561 You can use both options together, to make sure the auxiliary symbol
6562 file has all the symbol information for your program.
6564 The auxiliary symbol file for a program called @var{myprog} is called
6565 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
6566 than the corresponding executable), @value{GDBN} will always attempt to use
6567 it when you debug @var{myprog}; no special options or commands are
6570 The @file{.syms} file is specific to the host machine where you run
6571 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
6572 symbol table. It cannot be shared across multiple host platforms.
6574 @c FIXME: for now no mention of directories, since this seems to be in
6575 @c flux. 13mar1992 status is that in theory GDB would look either in
6576 @c current dir or in same dir as myprog; but issues like competing
6577 @c GDB's, or clutter in system dirs, mean that in practice right now
6578 @c only current dir is used. FFish says maybe a special GDB hierarchy
6579 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
6582 @item core-file @r{[} @var{filename} @r{]}
6585 Specify the whereabouts of a core dump file to be used as the ``contents
6586 of memory''. Traditionally, core files contain only some parts of the
6587 address space of the process that generated them; @value{GDBN} can access the
6588 executable file itself for other parts.
6590 @code{core-file} with no argument specifies that no core file is
6593 Note that the core file is ignored when your program is actually running
6594 under @value{GDBN}. So, if you have been running your program and you wish to
6595 debug a core file instead, you must kill the subprocess in which the
6596 program is running. To do this, use the @code{kill} command
6597 (@pxref{Kill Process, ,Killing the child process}).
6600 @item load @var{filename}
6603 Depending on what remote debugging facilities are configured into
6604 @value{GDBN}, the @code{load} command may be available. Where it exists, it
6605 is meant to make @var{filename} (an executable) available for debugging
6606 on the remote system---by downloading, or dynamic linking, for example.
6607 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
6608 the @code{add-symbol-file} command.
6610 If your @value{GDBN} does not have a @code{load} command, attempting to
6611 execute it gets the error message ``@code{You can't do that when your
6612 target is @dots{}}''
6616 On VxWorks, @code{load} will dynamically link @var{filename} on the
6617 current target system as well as adding its symbols in @value{GDBN}.
6621 @cindex download to Nindy-960
6622 With the Nindy interface to an Intel 960 board, @code{load} will
6623 download @var{filename} to the 960 as well as adding its symbols in
6628 @cindex download to H8/300 or H8/500
6629 @cindex H8/300 or H8/500 download
6630 @cindex download to Hitachi SH
6631 @cindex Hitachi SH download
6632 When you select remote debugging to a Hitachi SH, H8/300, or H8/500 board
6633 (@pxref{Hitachi Remote,,@value{GDBN} and Hitachi Microprocessors}),
6634 the @code{load} command downloads your program to the Hitachi board and also
6635 opens it as the current executable target for @value{GDBN} on your host
6636 (like the @code{file} command).
6639 @code{load} will not repeat if you press @key{RET} again after using it.
6642 @item add-symbol-file @var{filename} @var{address}
6643 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
6644 @kindex add-symbol-file
6645 @cindex dynamic linking
6646 The @code{add-symbol-file} command reads additional symbol table information
6647 from the file @var{filename}. You would use this command when @var{filename}
6648 has been dynamically loaded (by some other means) into the program that
6649 is running. @var{address} should be the memory address at which the
6650 file has been loaded; @value{GDBN} cannot figure this out for itself.
6651 You can specify @var{address} as an expression.
6653 The symbol table of the file @var{filename} is added to the symbol table
6654 originally read with the @code{symbol-file} command. You can use the
6655 @code{add-symbol-file} command any number of times; the new symbol data thus
6656 read keeps adding to the old. To discard all old symbol data instead,
6657 use the @code{symbol-file} command.
6659 @code{add-symbol-file} will not repeat if you press @key{RET} after using it.
6661 You can use the @samp{-mapped} and @samp{-readnow} options just as with
6662 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
6663 table information for @var{filename}.
6670 @code{info files} and @code{info target} are synonymous; both print
6671 the current target (@pxref{Targets, ,Specifying a Debugging Target}),
6674 names of the executable and core dump files
6677 name of the executable file
6679 currently in use by @value{GDBN}, and the files from which symbols were
6680 loaded. The command @code{help targets} lists all possible targets
6681 rather than current ones.
6684 All file-specifying commands allow both absolute and relative file names
6685 as arguments. @value{GDBN} always converts the file name to an absolute path
6686 name and remembers it that way.
6689 @cindex shared libraries
6690 @value{GDBN} supports SunOS, SVR4, and IBM RS/6000 shared libraries.
6691 @value{GDBN} automatically loads symbol definitions from shared libraries
6692 when you use the @code{run} command, or when you examine a core file.
6693 (Before you issue the @code{run} command, @value{GDBN} will not understand
6694 references to a function in a shared library, however---unless you are
6695 debugging a core file).
6696 @c FIXME: next @value{GDBN} release should permit some refs to undef
6697 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared lib
6701 @itemx info sharedlibrary
6702 @kindex info sharedlibrary
6704 Print the names of the shared libraries which are currently loaded.
6706 @item sharedlibrary @var{regex}
6707 @itemx share @var{regex}
6708 @kindex sharedlibrary
6710 This is an obsolescent command; you can use it to explicitly
6711 load shared object library symbols for files matching a UNIX regular
6712 expression, but as with files loaded automatically, it will only load
6713 shared libraries required by your program for a core file or after
6714 typing @code{run}. If @var{regex} is omitted all shared libraries
6715 required by your program are loaded.
6720 @section Errors reading symbol files
6722 While reading a symbol file, @value{GDBN} will occasionally encounter problems,
6723 such as symbol types it does not recognize, or known bugs in compiler
6724 output. By default, @value{GDBN} does not notify you of such problems, since
6725 they are relatively common and primarily of interest to people
6726 debugging compilers. If you are interested in seeing information
6727 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
6728 only one message about each such type of problem, no matter how many
6729 times the problem occurs; or you can ask @value{GDBN} to print more messages,
6730 to see how many times the problems occur, with the @code{set
6731 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
6734 The messages currently printed, and their meanings, include:
6737 @item inner block not inside outer block in @var{symbol}
6739 The symbol information shows where symbol scopes begin and end
6740 (such as at the start of a function or a block of statements). This
6741 error indicates that an inner scope block is not fully contained
6742 in its outer scope blocks.
6744 @value{GDBN} circumvents the problem by treating the inner block as if it had
6745 the same scope as the outer block. In the error message, @var{symbol}
6746 may be shown as ``@code{(don't know)}'' if the outer block is not a
6749 @item block at @var{address} out of order
6751 The symbol information for symbol scope blocks should occur in
6752 order of increasing addresses. This error indicates that it does not
6755 @value{GDBN} does not circumvent this problem, and will have trouble
6756 locating symbols in the source file whose symbols it is reading. (You
6757 can often determine what source file is affected by specifying
6758 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
6761 @item bad block start address patched
6763 The symbol information for a symbol scope block has a start address
6764 smaller than the address of the preceding source line. This is known
6765 to occur in the SunOS 4.1.1 (and earlier) C compiler.
6767 @value{GDBN} circumvents the problem by treating the symbol scope block as
6768 starting on the previous source line.
6770 @item bad string table offset in symbol @var{n}
6773 Symbol number @var{n} contains a pointer into the string table which is
6774 larger than the size of the string table.
6776 @value{GDBN} circumvents the problem by considering the symbol to have the
6777 name @code{foo}, which may cause other problems if many symbols end up
6780 @item unknown symbol type @code{0x@var{nn}}
6782 The symbol information contains new data types that @value{GDBN} does not yet
6783 know how to read. @code{0x@var{nn}} is the symbol type of the misunderstood
6784 information, in hexadecimal.
6786 @value{GDBN} circumvents the error by ignoring this symbol information. This
6787 will usually allow your program to be debugged, though certain symbols
6788 will not be accessible. If you encounter such a problem and feel like
6789 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint on
6790 @code{complain}, then go up to the function @code{read_dbx_symtab} and
6791 examine @code{*bufp} to see the symbol.
6793 @item stub type has NULL name
6794 @value{GDBN} could not find the full definition for
6803 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
6805 The symbol information for a C++ member function is missing some
6806 information that recent versions of the compiler should have output
6810 @item info mismatch between compiler and debugger
6812 @value{GDBN} could not parse a type specification output by the compiler.
6816 @chapter Specifying a Debugging Target
6817 @cindex debugging target
6820 A @dfn{target} is the execution environment occupied by your program.
6822 Often, @value{GDBN} runs in the same host environment as your program; in
6823 that case, the debugging target is specified as a side effect when you
6824 use the @code{file} or @code{core} commands. When you need more
6825 flexibility---for example, running @value{GDBN} on a physically separate
6826 host, or controlling a standalone system over a serial port or a
6827 realtime system over a TCP/IP connection---you
6832 can use the @code{target} command to specify one of the target types
6833 configured for @value{GDBN} (@pxref{Target Commands, ,Commands for managing
6837 * Active Targets:: Active targets
6838 * Target Commands:: Commands for managing targets
6839 * Remote:: Remote debugging
6842 @node Active Targets
6843 @section Active targets
6844 @cindex stacking targets
6845 @cindex active targets
6846 @cindex multiple targets
6849 There are three classes of targets: processes, core files, and
6850 executable files. @value{GDBN} can work concurrently on up to three active
6851 targets, one in each class. This allows you to (for example) start a
6852 process and inspect its activity without abandoning your work on a core
6855 For example, if you execute @samp{gdb a.out}, then the executable file
6856 @code{a.out} is the only active target. If you designate a core file as
6857 well---presumably from a prior run that crashed and coredumped---then
6858 @value{GDBN} has two active targets and will use them in tandem, looking
6859 first in the corefile target, then in the executable file, to satisfy
6860 requests for memory addresses. (Typically, these two classes of target
6861 are complementary, since core files contain only a program's
6862 read-write memory---variables and so on---plus machine status, while
6863 executable files contain only the program text and initialized data.)
6866 When you type @code{run}, your executable file becomes an active process
6867 target as well. When a process target is active, all @value{GDBN} commands
6868 requesting memory addresses refer to that target; addresses in an
6872 executable file target are obscured while the process
6876 Use the @code{exec-file} command to select a
6877 new executable target (@pxref{Files, ,Commands to specify
6881 Use the @code{core-file} and @code{exec-file} commands to select a
6882 new core file or executable target (@pxref{Files, ,Commands to specify
6883 files}). To specify as a target a process that is already running, use
6884 the @code{attach} command (@pxref{Attach, ,Debugging an
6885 already-running process}).
6888 @node Target Commands
6889 @section Commands for managing targets
6892 @item target @var{type} @var{parameters}
6893 Connects the @value{GDBN} host environment to a target
6898 machine or process. A target is typically a protocol for talking to
6899 debugging facilities. You use the argument @var{type} to specify the
6900 type or protocol of the target machine.
6902 Further @var{parameters} are interpreted by the target protocol, but
6903 typically include things like device names or host names to connect
6904 with, process numbers, and baud rates.
6907 The @code{target} command will not repeat if you press @key{RET} again
6908 after executing the command.
6912 Displays the names of all targets available. To display targets
6913 currently selected, use either @code{info target} or @code{info files}
6914 (@pxref{Files, ,Commands to specify files}).
6916 @item help target @var{name}
6917 Describe a particular target, including any parameters necessary to
6921 Here are some common targets (available, or not, depending on the GDB
6925 @item target exec @var{program}
6927 An executable file. @samp{target exec @var{program}} is the same as
6928 @samp{exec-file @var{program}}.
6931 @item target core @var{filename}
6933 A core dump file. @samp{target core @var{filename}} is the same as
6934 @samp{core-file @var{filename}}.
6938 @item target remote @var{dev}
6939 @kindex target remote
6940 Remote serial target in GDB-specific protocol. The argument @var{dev}
6941 specifies what serial device to use for the connection (e.g.
6942 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}.
6948 CPU simulator. @xref{Simulator,,Simulated CPU Target}.
6952 @item target udi @var{keyword}
6954 Remote AMD29K target, using the AMD UDI protocol. The @var{keyword}
6955 argument specifies which 29K board or simulator to use. @xref{UDI29K
6956 Remote,,@value{GDBN} and the UDI protocol for AMD29K}.
6958 @item target amd-eb @var{dev} @var{speed} @var{PROG}
6959 @kindex target amd-eb
6961 Remote PC-resident AMD EB29K board, attached over serial lines.
6962 @var{dev} is the serial device, as for @code{target remote};
6963 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
6964 name of the program to be debugged, as it appears to DOS on the PC.
6965 @xref{EB29K Remote, ,@value{GDBN} with a remote EB29K}.
6971 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
6972 @ifclear H8EXCLUSIVE
6973 @c Unix only, not currently of interest for H8-only manual
6974 Use special commands @code{device} and @code{speed} to control the serial
6975 line and the communications speed used.
6977 @xref{Hitachi Remote,,@value{GDBN} and Hitachi Microprocessors}.
6981 @item target nindy @var{devicename}
6982 @kindex target nindy
6983 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
6984 the name of the serial device to use for the connection, e.g.
6985 @file{/dev/ttya}. @xref{i960-Nindy Remote, ,@value{GDBN} with a remote i960 (Nindy)}.
6989 @item target st2000 @var{dev} @var{speed}
6990 @kindex target st2000
6991 A Tandem ST2000 phone switch, running Tandem's STDBUG protocol. @var{dev}
6992 is the name of the device attached to the ST2000 serial line;
6993 @var{speed} is the communication line speed. The arguments are not used
6994 if @value{GDBN} is configured to connect to the ST2000 using TCP or Telnet.
6995 @xref{ST2000 Remote,,@value{GDBN} with a Tandem ST2000}.
6999 @item target vxworks @var{machinename}
7000 @kindex target vxworks
7001 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
7002 is the target system's machine name or IP address.
7003 @xref{VxWorks Remote, ,@value{GDBN} and VxWorks}.
7008 Different targets are available on different configurations of @value{GDBN}; your
7009 configuration may have more or fewer targets.
7013 @section Remote debugging
7014 @cindex remote debugging
7016 If you are trying to debug a program running on a machine that cannot run
7017 GDB in the usual way, it is often useful to use remote debugging. For
7018 example, you might use remote debugging on an operating system kernel, or on
7019 a small system which does not have a general purpose operating system
7020 powerful enough to run a full-featured debugger.
7022 Some configurations of GDB have special serial or TCP/IP interfaces
7023 to make this work with particular debugging targets. In addition,
7024 GDB comes with a generic serial protocol (specific to GDB, but
7025 not specific to any particular target system) which you can use if you
7026 write the remote stubs---the code that will run on the remote system to
7027 communicate with GDB.
7029 Other remote targets may be available in your
7030 configuration of GDB; use @code{help targets} to list them.
7033 @c Text on starting up GDB in various specific cases; it goes up front
7034 @c in manuals configured for any of those particular situations, here
7038 * Remote Serial:: @value{GDBN} remote serial protocol
7041 * i960-Nindy Remote:: @value{GDBN} with a remote i960 (Nindy)
7044 * UDI29K Remote:: @value{GDBN} and the UDI protocol for AMD29K
7045 * EB29K Remote:: @value{GDBN} with a remote EB29K
7048 * VxWorks Remote:: @value{GDBN} and VxWorks
7051 * ST2000 Remote:: @value{GDBN} with a Tandem ST2000
7054 * Hitachi Remote:: @value{GDBN} and Hitachi Microprocessors
7057 * MIPS Remote:: @value{GDBN} and MIPS boards
7060 * Simulator:: Simulated CPU target
7064 @include remote.texi
7067 @node Controlling GDB
7068 @chapter Controlling @value{GDBN}
7070 You can alter the way @value{GDBN} interacts with you by using
7071 the @code{set} command. For commands controlling how @value{GDBN} displays
7072 data, @pxref{Print Settings, ,Print settings}; other settings are described here.
7076 * Editing:: Command editing
7077 * History:: Command history
7078 * Screen Size:: Screen size
7080 * Messages/Warnings:: Optional warnings and messages
7087 @value{GDBN} indicates its readiness to read a command by printing a string
7088 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
7089 can change the prompt string with the @code{set prompt} command. For
7090 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
7091 the prompt in one of the @value{GDBN} sessions so that you can always tell which
7092 one you are talking to.
7095 @item set prompt @var{newprompt}
7097 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
7100 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
7104 @section Command editing
7106 @cindex command line editing
7108 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
7109 GNU library provides consistent behavior for programs which provide a
7110 command line interface to the user. Advantages are @code{emacs}-style
7111 or @code{vi}-style inline editing of commands, @code{csh}-like history
7112 substitution, and a storage and recall of command history across
7115 You may control the behavior of command line editing in @value{GDBN} with the
7122 @itemx set editing on
7123 Enable command line editing (enabled by default).
7125 @item set editing off
7126 Disable command line editing.
7128 @kindex show editing
7130 Show whether command line editing is enabled.
7134 @section Command history
7136 @value{GDBN} can keep track of the commands you type during your
7137 debugging sessions, so that you can be certain of precisely what
7138 happened. Use these commands to manage the @value{GDBN} command
7142 @cindex history substitution
7143 @cindex history file
7144 @kindex set history filename
7145 @item set history filename @var{fname}
7146 Set the name of the @value{GDBN} command history file to @var{fname}. This is
7147 the file from which @value{GDBN} will read an initial command history
7148 list or to which it will write this list when it exits. This list is
7149 accessed through history expansion or through the history
7150 command editing characters listed below. This file defaults to the
7151 value of the environment variable @code{GDBHISTFILE}, or to
7152 @file{./.gdb_history} if this variable is not set.
7154 @cindex history save
7155 @kindex set history save
7156 @item set history save
7157 @itemx set history save on
7158 Record command history in a file, whose name may be specified with the
7159 @code{set history filename} command. By default, this option is disabled.
7161 @item set history save off
7162 Stop recording command history in a file.
7164 @cindex history size
7165 @kindex set history size
7166 @item set history size @var{size}
7167 Set the number of commands which @value{GDBN} will keep in its history list.
7168 This defaults to the value of the environment variable
7169 @code{HISTSIZE}, or to 256 if this variable is not set.
7172 @cindex history expansion
7173 History expansion assigns special meaning to the character @kbd{!}.
7174 @ifset have-readline-appendices
7175 @xref{Event Designators}.
7178 Since @kbd{!} is also the logical not operator in C, history expansion
7179 is off by default. If you decide to enable history expansion with the
7180 @code{set history expansion on} command, you may sometimes need to
7181 follow @kbd{!} (when it is used as logical not, in an expression) with
7182 a space or a tab to prevent it from being expanded. The readline
7183 history facilities will not attempt substitution on the strings
7184 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
7186 The commands to control history expansion are:
7190 @kindex set history expansion
7191 @item set history expansion on
7192 @itemx set history expansion
7193 Enable history expansion. History expansion is off by default.
7195 @item set history expansion off
7196 Disable history expansion.
7198 The readline code comes with more complete documentation of
7199 editing and history expansion features. Users unfamiliar with @code{emacs}
7200 or @code{vi} may wish to read it.
7201 @ifset have-readline-appendices
7202 @xref{Command Line Editing}.
7206 @kindex show history
7208 @itemx show history filename
7209 @itemx show history save
7210 @itemx show history size
7211 @itemx show history expansion
7212 These commands display the state of the @value{GDBN} history parameters.
7213 @code{show history} by itself displays all four states.
7218 @kindex show commands
7220 Display the last ten commands in the command history.
7222 @item show commands @var{n}
7223 Print ten commands centered on command number @var{n}.
7225 @item show commands +
7226 Print ten commands just after the commands last printed.
7230 @section Screen size
7231 @cindex size of screen
7232 @cindex pauses in output
7234 Certain commands to @value{GDBN} may produce large amounts of
7235 information output to the screen. To help you read all of it,
7236 @value{GDBN} pauses and asks you for input at the end of each page of
7237 output. Type @key{RET} when you want to continue the output, or @kbd{q}
7238 to discard the remaining output. Also, the screen width setting
7239 determines when to wrap lines of output. Depending on what is being
7240 printed, @value{GDBN} tries to break the line at a readable place,
7241 rather than simply letting it overflow onto the following line.
7243 Normally @value{GDBN} knows the size of the screen from the termcap data base
7244 together with the value of the @code{TERM} environment variable and the
7245 @code{stty rows} and @code{stty cols} settings. If this is not correct,
7246 you can override it with the @code{set height} and @code{set
7250 @item set height @var{lpp}
7252 @itemx set width @var{cpl}
7258 These @code{set} commands specify a screen height of @var{lpp} lines and
7259 a screen width of @var{cpl} characters. The associated @code{show}
7260 commands display the current settings.
7262 If you specify a height of zero lines, @value{GDBN} will not pause during output
7263 no matter how long the output is. This is useful if output is to a file
7264 or to an editor buffer.
7266 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
7267 from wrapping its output.
7272 @cindex number representation
7273 @cindex entering numbers
7275 You can always enter numbers in octal, decimal, or hexadecimal in @value{GDBN} by
7276 the usual conventions: octal numbers begin with @samp{0}, decimal
7277 numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}.
7278 Numbers that begin with none of these are, by default, entered in base
7279 10; likewise, the default display for numbers---when no particular
7280 format is specified---is base 10. You can change the default base for
7281 both input and output with the @code{set radix} command.
7285 @item set radix @var{base}
7286 Set the default base for numeric input and display. Supported choices
7287 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
7288 specified either unambiguously or using the current default radix; for
7298 will set the base to decimal. On the other hand, @samp{set radix 10}
7299 will leave the radix unchanged no matter what it was.
7303 Display the current default base for numeric input and display.
7306 @node Messages/Warnings
7307 @section Optional warnings and messages
7309 By default, @value{GDBN} is silent about its inner workings. If you are running
7310 on a slow machine, you may want to use the @code{set verbose} command.
7311 It will make @value{GDBN} tell you when it does a lengthy internal operation, so
7312 you will not think it has crashed.
7314 Currently, the messages controlled by @code{set verbose} are those
7315 which announce that the symbol table for a source file is being read;
7316 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
7320 @item set verbose on
7321 Enables @value{GDBN} output of certain informational messages.
7323 @item set verbose off
7324 Disables @value{GDBN} output of certain informational messages.
7326 @kindex show verbose
7328 Displays whether @code{set verbose} is on or off.
7331 By default, if @value{GDBN} encounters bugs in the symbol table of an object
7332 file, it is silent; but if you are debugging a compiler, you may find
7333 this information useful (@pxref{Symbol Errors, ,Errors reading symbol files}).
7336 @kindex set complaints
7337 @item set complaints @var{limit}
7338 Permits @value{GDBN} to output @var{limit} complaints about each type of unusual
7339 symbols before becoming silent about the problem. Set @var{limit} to
7340 zero to suppress all complaints; set it to a large number to prevent
7341 complaints from being suppressed.
7343 @kindex show complaints
7344 @item show complaints
7345 Displays how many symbol complaints @value{GDBN} is permitted to produce.
7348 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
7349 lot of stupid questions to confirm certain commands. For example, if
7350 you try to run a program which is already running:
7354 The program being debugged has been started already.
7355 Start it from the beginning? (y or n)
7358 If you are willing to unflinchingly face the consequences of your own
7359 commands, you can disable this ``feature'':
7364 @cindex confirmation
7365 @cindex stupid questions
7366 @item set confirm off
7367 Disables confirmation requests.
7369 @item set confirm on
7370 Enables confirmation requests (the default).
7373 @kindex show confirm
7374 Displays state of confirmation requests.
7377 @c FIXME this does not really belong here. But where *does* it belong?
7378 @cindex reloading symbols
7379 Some systems allow individual object files that make up your program to
7380 be replaced without stopping and restarting your program.
7382 For example, in VxWorks you can simply recompile a defective object file
7383 and keep on running.
7385 If you are running on one of these systems, you can allow @value{GDBN} to
7386 reload the symbols for automatically relinked modules:
7389 @kindex set symbol-reloading
7390 @item set symbol-reloading on
7391 Replace symbol definitions for the corresponding source file when an
7392 object file with a particular name is seen again.
7394 @item set symbol-reloading off
7395 Do not replace symbol definitions when re-encountering object files of
7396 the same name. This is the default state; if you are not running on a
7397 system that permits automatically relinking modules, you should leave
7398 @code{symbol-reloading} off, since otherwise @value{GDBN} may discard symbols
7399 when linking large programs, that may contain several modules (from
7400 different directories or libraries) with the same name.
7402 @item show symbol-reloading
7403 Show the current @code{on} or @code{off} setting.
7407 @chapter Canned Sequences of Commands
7409 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
7410 command lists}), @value{GDBN} provides two ways to store sequences of commands
7411 for execution as a unit: user-defined commands and command files.
7414 * Define:: User-defined commands
7415 * Hooks:: User-defined command hooks
7416 * Command Files:: Command files
7417 * Output:: Commands for controlled output
7421 @section User-defined commands
7423 @cindex user-defined command
7424 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to which you
7425 assign a new name as a command. This is done with the @code{define}
7429 @item define @var{commandname}
7431 Define a command named @var{commandname}. If there is already a command
7432 by that name, you are asked to confirm that you want to redefine it.
7434 The definition of the command is made up of other @value{GDBN} command lines,
7435 which are given following the @code{define} command. The end of these
7436 commands is marked by a line containing @code{end}.
7438 @item document @var{commandname}
7440 Give documentation to the user-defined command @var{commandname}. The
7441 command @var{commandname} must already be defined. This command reads
7442 lines of documentation just as @code{define} reads the lines of the
7443 command definition, ending with @code{end}. After the @code{document}
7444 command is finished, @code{help} on command @var{commandname} will print
7445 the documentation you have specified.
7447 You may use the @code{document} command again to change the
7448 documentation of a command. Redefining the command with @code{define}
7449 does not change the documentation.
7451 @item help user-defined
7452 @kindex help user-defined
7453 List all user-defined commands, with the first line of the documentation
7457 @itemx show user @var{commandname}
7459 Display the @value{GDBN} commands used to define @var{commandname} (but not its
7460 documentation). If no @var{commandname} is given, display the
7461 definitions for all user-defined commands.
7464 User-defined commands do not take arguments. When they are executed, the
7465 commands of the definition are not printed. An error in any command
7466 stops execution of the user-defined command.
7468 Commands that would ask for confirmation if used interactively proceed
7469 without asking when used inside a user-defined command. Many @value{GDBN} commands
7470 that normally print messages to say what they are doing omit the messages
7471 when used in a user-defined command.
7474 @section User-defined command hooks
7475 @cindex command files
7477 You may define @emph{hooks}, which are a special kind of user-defined
7478 command. Whenever you run the command @samp{foo}, if the user-defined
7479 command @samp{hook-foo} exists, it is executed (with no arguments)
7480 before that command.
7482 In addition, a pseudo-command, @samp{stop} exists. Defining
7483 (@samp{hook-stop}) makes the associated commands execute every time
7484 execution stops in your program: before breakpoint commands are run,
7485 displays are printed, or the stack frame is printed.
7488 For example, to ignore @code{SIGALRM} signals while
7489 single-stepping, but treat them normally during normal execution,
7494 handle SIGALRM nopass
7501 define hook-continue
7507 You can define a hook for any single-word command in @value{GDBN}, but
7508 not for command aliases; you should define a hook for the basic command
7509 name, e.g. @code{backtrace} rather than @code{bt}.
7510 @c FIXME! So how does Joe User discover whether a command is an alias
7512 If an error occurs during the execution of your hook, execution of
7513 @value{GDBN} commands stops and @value{GDBN} issues a prompt
7514 (before the command that you actually typed had a chance to run).
7516 If you try to define a hook which does not match any known command, you
7517 will get a warning from the @code{define} command.
7520 @section Command files
7522 @cindex command files
7523 A command file for @value{GDBN} is a file of lines that are @value{GDBN} commands. Comments
7524 (lines starting with @kbd{#}) may also be included. An empty line in a
7525 command file does nothing; it does not mean to repeat the last command, as
7526 it would from the terminal.
7529 @cindex @file{@value{GDBINIT}}
7530 When you start @value{GDBN}, it automatically executes commands from its
7531 @dfn{init files}. These are files named @file{@value{GDBINIT}}. @value{GDBN} reads
7532 the init file (if any) in your home directory and then the init file
7533 (if any) in the current working directory. (The init files are not
7534 executed if you use the @samp{-nx} option; @pxref{Mode Options,
7538 @cindex init file name
7539 On some configurations of @value{GDBN}, the init file is known by a
7540 different name (these are typically environments where a specialized
7541 form of GDB may need to coexist with other forms, hence a different name
7542 for the specialized version's init file). These are the environments
7543 with special init file names:
7548 VxWorks (Wind River Systems real-time OS): @samp{.vxgdbinit}
7550 @kindex .os68gdbinit
7552 OS68K (Enea Data Systems real-time OS): @samp{.os68gdbinit}
7556 ES-1800 (Ericsson Telecom AB M68000 emulator): @samp{.esgdbinit}
7560 You can also request the execution of a command file with the
7561 @code{source} command:
7564 @item source @var{filename}
7566 Execute the command file @var{filename}.
7569 The lines in a command file are executed sequentially. They are not
7570 printed as they are executed. An error in any command terminates execution
7571 of the command file.
7573 Commands that would ask for confirmation if used interactively proceed
7574 without asking when used in a command file. Many @value{GDBN} commands that
7575 normally print messages to say what they are doing omit the messages
7576 when called from command files.
7579 @section Commands for controlled output
7581 During the execution of a command file or a user-defined command, normal
7582 @value{GDBN} output is suppressed; the only output that appears is what is
7583 explicitly printed by the commands in the definition. This section
7584 describes three commands useful for generating exactly the output you
7588 @item echo @var{text}
7590 @c I do not consider backslash-space a standard C escape sequence
7591 @c because it is not in ANSI.
7592 Print @var{text}. Nonprinting characters can be included in
7593 @var{text} using C escape sequences, such as @samp{\n} to print a
7594 newline. @strong{No newline will be printed unless you specify one.}
7595 In addition to the standard C escape sequences, a backslash followed
7596 by a space stands for a space. This is useful for displaying a
7597 string with spaces at the beginning or the end, since leading and
7598 trailing spaces are otherwise trimmed from all arguments.
7599 To print @samp{@w{ }and foo =@w{ }}, use the command
7600 @samp{echo \@w{ }and foo = \@w{ }}.
7602 A backslash at the end of @var{text} can be used, as in C, to continue
7603 the command onto subsequent lines. For example,
7606 echo This is some text\n\
7607 which is continued\n\
7608 onto several lines.\n
7611 produces the same output as
7614 echo This is some text\n
7615 echo which is continued\n
7616 echo onto several lines.\n
7619 @item output @var{expression}
7621 Print the value of @var{expression} and nothing but that value: no
7622 newlines, no @samp{$@var{nn} = }. The value is not entered in the
7623 value history either. @xref{Expressions, ,Expressions}, for more information on
7626 @item output/@var{fmt} @var{expression}
7627 Print the value of @var{expression} in format @var{fmt}. You can use
7628 the same formats as for @code{print}. @xref{Output Formats,,Output
7629 formats}, for more information.
7631 @item printf @var{string}, @var{expressions}@dots{}
7633 Print the values of the @var{expressions} under the control of
7634 @var{string}. The @var{expressions} are separated by commas and may be
7635 either numbers or pointers. Their values are printed as specified by
7636 @var{string}, exactly as if your program were to execute the C
7640 printf (@var{string}, @var{expressions}@dots{});
7643 For example, you can print two values in hex like this:
7646 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
7649 The only backslash-escape sequences that you can use in the format
7650 string are the simple ones that consist of backslash followed by a
7656 @chapter Using @value{GDBN} under GNU Emacs
7659 A special interface allows you to use GNU Emacs to view (and
7660 edit) the source files for the program you are debugging with
7663 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
7664 executable file you want to debug as an argument. This command starts
7665 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
7666 created Emacs buffer.
7668 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
7673 All ``terminal'' input and output goes through the Emacs buffer.
7676 This applies both to @value{GDBN} commands and their output, and to the input
7677 and output done by the program you are debugging.
7679 This is useful because it means that you can copy the text of previous
7680 commands and input them again; you can even use parts of the output
7683 All the facilities of Emacs' Shell mode are available for interacting
7684 with your program. In particular, you can send signals the usual
7685 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
7690 @value{GDBN} displays source code through Emacs.
7693 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
7694 source file for that frame and puts an arrow (@samp{=>}) at the
7695 left margin of the current line. Emacs uses a separate buffer for
7696 source display, and splits the screen to show both your @value{GDBN} session
7699 Explicit @value{GDBN} @code{list} or search commands still produce output as
7700 usual, but you probably will have no reason to use them.
7703 @emph{Warning:} If the directory where your program resides is not your
7704 current directory, it can be easy to confuse Emacs about the location of
7705 the source files, in which case the auxiliary display buffer will not
7706 appear to show your source. @value{GDBN} can find programs by searching your
7707 environment's @code{PATH} variable, so the @value{GDBN} input and output
7708 session will proceed normally; but Emacs does not get enough information
7709 back from @value{GDBN} to locate the source files in this situation. To
7710 avoid this problem, either start @value{GDBN} mode from the directory where
7711 your program resides, or specify a full path name when prompted for the
7712 @kbd{M-x gdb} argument.
7714 A similar confusion can result if you use the @value{GDBN} @code{file} command to
7715 switch to debugging a program in some other location, from an existing
7716 @value{GDBN} buffer in Emacs.
7719 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
7720 you need to call @value{GDBN} by a different name (for example, if you keep
7721 several configurations around, with different names) you can set the
7722 Emacs variable @code{gdb-command-name}; for example,
7725 (setq gdb-command-name "mygdb")
7729 (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or
7730 in your @file{.emacs} file) will make Emacs call the program named
7731 ``@code{mygdb}'' instead.
7733 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
7734 addition to the standard Shell mode commands:
7738 Describe the features of Emacs' @value{GDBN} Mode.
7741 Execute to another source line, like the @value{GDBN} @code{step} command; also
7742 update the display window to show the current file and location.
7745 Execute to next source line in this function, skipping all function
7746 calls, like the @value{GDBN} @code{next} command. Then update the display window
7747 to show the current file and location.
7750 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
7751 display window accordingly.
7754 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
7755 display window accordingly.
7758 Execute until exit from the selected stack frame, like the @value{GDBN}
7759 @code{finish} command.
7762 Continue execution of your program, like the @value{GDBN} @code{continue}
7765 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
7768 Go up the number of frames indicated by the numeric argument
7769 (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}),
7770 like the @value{GDBN} @code{up} command.
7772 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
7775 Go down the number of frames indicated by the numeric argument, like the
7776 @value{GDBN} @code{down} command.
7778 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
7781 Read the number where the cursor is positioned, and insert it at the end
7782 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
7783 around an address that was displayed earlier, type @kbd{disassemble};
7784 then move the cursor to the address display, and pick up the
7785 argument for @code{disassemble} by typing @kbd{C-x &}.
7787 You can customize this further by defining elements of the list
7788 @code{gdb-print-command}; once it is defined, you can format or
7789 otherwise process numbers picked up by @kbd{C-x &} before they are
7790 inserted. A numeric argument to @kbd{C-x &} will both indicate that you
7791 wish special formatting, and act as an index to pick an element of the
7792 list. If the list element is a string, the number to be inserted is
7793 formatted using the Emacs function @code{format}; otherwise the number
7794 is passed as an argument to the corresponding list element.
7797 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
7798 tells @value{GDBN} to set a breakpoint on the source line point is on.
7800 If you accidentally delete the source-display buffer, an easy way to get
7801 it back is to type the command @code{f} in the @value{GDBN} buffer, to
7802 request a frame display; when you run under Emacs, this will recreate
7803 the source buffer if necessary to show you the context of the current
7806 The source files displayed in Emacs are in ordinary Emacs buffers
7807 which are visiting the source files in the usual way. You can edit
7808 the files with these buffers if you wish; but keep in mind that @value{GDBN}
7809 communicates with Emacs in terms of line numbers. If you add or
7810 delete lines from the text, the line numbers that @value{GDBN} knows will cease
7811 to correspond properly with the code.
7813 @c The following dropped because Epoch is nonstandard. Reactivate
7814 @c if/when v19 does something similar. ---pesch@cygnus.com 19dec1990
7816 @kindex emacs epoch environment
7820 Version 18 of Emacs has a built-in window system called the @code{epoch}
7821 environment. Users of this environment can use a new command,
7822 @code{inspect} which performs identically to @code{print} except that
7823 each value is printed in its own window.
7829 @chapter Using @value{GDBN} with Energize
7832 The Energize Programming System is an integrated development environment
7833 that includes a point-and-click interface to many programming tools.
7834 When you use @value{GDBN} in this environment, you can use the standard
7835 Energize graphical interface to drive @value{GDBN}; you can also, if you
7836 choose, type @value{GDBN} commands as usual in a debugging window. Even if
7837 you use the graphical interface, the debugging window (which uses Emacs,
7838 and resembles the standard Emacs interface to @value{GDBN}) displays the
7839 equivalent commands, so that the history of your debugging session is
7842 When Energize starts up a @value{GDBN} session, it uses one of the
7843 command-line options @samp{-energize} or @samp{-cadillac} (``cadillac''
7844 is the name of the communications protocol used by the Energize system).
7845 This option makes @value{GDBN} run as one of the tools in the Energize Tool
7846 Set: it sends all output to the Energize kernel, and accept input from
7849 See the user manual for the Energize Programming System for
7850 information on how to use the Energize graphical interface and the other
7851 development tools that Energize integrates with @value{GDBN}.
7856 @chapter Reporting Bugs in @value{GDBN}
7857 @cindex bugs in @value{GDBN}
7858 @cindex reporting bugs in @value{GDBN}
7860 Your bug reports play an essential role in making @value{GDBN} reliable.
7862 Reporting a bug may help you by bringing a solution to your problem, or it
7863 may not. But in any case the principal function of a bug report is to help
7864 the entire community by making the next version of @value{GDBN} work better. Bug
7865 reports are your contribution to the maintenance of @value{GDBN}.
7867 In order for a bug report to serve its purpose, you must include the
7868 information that enables us to fix the bug.
7871 * Bug Criteria:: Have you found a bug?
7872 * Bug Reporting:: How to report bugs
7876 @section Have you found a bug?
7877 @cindex bug criteria
7879 If you are not sure whether you have found a bug, here are some guidelines:
7883 @cindex fatal signal
7884 @cindex debugger crash
7885 @cindex crash of debugger
7886 If the debugger gets a fatal signal, for any input whatever, that is a
7887 @value{GDBN} bug. Reliable debuggers never crash.
7890 @cindex error on valid input
7891 If @value{GDBN} produces an error message for valid input, that is a bug.
7894 @cindex invalid input
7895 If @value{GDBN} does not produce an error message for invalid input,
7896 that is a bug. However, you should note that your idea of
7897 ``invalid input'' might be our idea of ``an extension'' or ``support
7898 for traditional practice''.
7901 If you are an experienced user of debugging tools, your suggestions
7902 for improvement of @value{GDBN} are welcome in any case.
7906 @section How to report bugs
7908 @cindex @value{GDBN} bugs, reporting
7910 A number of companies and individuals offer support for GNU products.
7911 If you obtained @value{GDBN} from a support organization, we recommend you
7912 contact that organization first.
7914 You can find contact information for many support companies and
7915 individuals in the file @file{etc/SERVICE} in the GNU Emacs
7918 In any event, we also recommend that you send bug reports for @value{GDBN} to one
7922 bug-gdb@@prep.ai.mit.edu
7923 @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb
7926 @strong{Do not send bug reports to @samp{info-gdb}, or to
7927 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do not want to
7928 receive bug reports. Those that do, have arranged to receive @samp{bug-gdb}.
7930 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
7931 serves as a repeater. The mailing list and the newsgroup carry exactly
7932 the same messages. Often people think of posting bug reports to the
7933 newsgroup instead of mailing them. This appears to work, but it has one
7934 problem which can be crucial: a newsgroup posting often lacks a mail
7935 path back to the sender. Thus, if we need to ask for more information,
7936 we may be unable to reach you. For this reason, it is better to send
7937 bug reports to the mailing list.
7939 As a last resort, send bug reports on paper to:
7943 Free Software Foundation
7948 The fundamental principle of reporting bugs usefully is this:
7949 @strong{report all the facts}. If you are not sure whether to state a
7950 fact or leave it out, state it!
7952 Often people omit facts because they think they know what causes the
7953 problem and assume that some details do not matter. Thus, you might
7954 assume that the name of the variable you use in an example does not matter.
7955 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
7956 stray memory reference which happens to fetch from the location where that
7957 name is stored in memory; perhaps, if the name were different, the contents
7958 of that location would fool the debugger into doing the right thing despite
7959 the bug. Play it safe and give a specific, complete example. That is the
7960 easiest thing for you to do, and the most helpful.
7962 Keep in mind that the purpose of a bug report is to enable us to fix
7963 the bug if it is new to us. It is not as important as what happens if
7964 the bug is already known. Therefore, always write your bug reports on
7965 the assumption that the bug has not been reported previously.
7967 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7968 bell?'' Those bug reports are useless, and we urge everyone to
7969 @emph{refuse to respond to them} except to chide the sender to report
7972 To enable us to fix the bug, you should include all these things:
7976 The version of @value{GDBN}. @value{GDBN} announces it if you start with no
7977 arguments; you can also print it at any time using @code{show version}.
7979 Without this, we will not know whether there is any point in looking for
7980 the bug in the current version of @value{GDBN}.
7983 The type of machine you are using, and the operating system name and
7987 What compiler (and its version) was used to compile @value{GDBN}---e.g.
7988 ``@value{GCC}--2.0''.
7991 What compiler (and its version) was used to compile the program you
7992 are debugging---e.g. ``@value{GCC}--2.0''.
7995 The command arguments you gave the compiler to compile your example and
7996 observe the bug. For example, did you use @samp{-O}? To guarantee
7997 you will not omit something important, list them all. A copy of the
7998 Makefile (or the output from make) is sufficient.
8000 If we were to try to guess the arguments, we would probably guess wrong
8001 and then we might not encounter the bug.
8004 A complete input script, and all necessary source files, that will
8008 A description of what behavior you observe that you believe is
8009 incorrect. For example, ``It gets a fatal signal.''
8011 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we will
8012 certainly notice it. But if the bug is incorrect output, we might not
8013 notice unless it is glaringly wrong. We are human, after all. You
8014 might as well not give us a chance to make a mistake.
8016 Even if the problem you experience is a fatal signal, you should still
8017 say so explicitly. Suppose something strange is going on, such as,
8018 your copy of @value{GDBN} is out of synch, or you have encountered a
8019 bug in the C library on your system. (This has happened!) Your copy
8020 might crash and ours would not. If you told us to expect a crash,
8021 then when ours fails to crash, we would know that the bug was not
8022 happening for us. If you had not told us to expect a crash, then we
8023 would not be able to draw any conclusion from our observations.
8026 If you wish to suggest changes to the @value{GDBN} source, send us context
8027 diffs. If you even discuss something in the @value{GDBN} source, refer to
8028 it by context, not by line number.
8030 The line numbers in our development sources will not match those in your
8031 sources. Your line numbers would convey no useful information to us.
8034 Here are some things that are not necessary:
8038 A description of the envelope of the bug.
8040 Often people who encounter a bug spend a lot of time investigating
8041 which changes to the input file will make the bug go away and which
8042 changes will not affect it.
8044 This is often time consuming and not very useful, because the way we
8045 will find the bug is by running a single example under the debugger
8046 with breakpoints, not by pure deduction from a series of examples.
8047 We recommend that you save your time for something else.
8049 Of course, if you can find a simpler example to report @emph{instead}
8050 of the original one, that is a convenience for us. Errors in the
8051 output will be easier to spot, running under the debugger will take
8054 However, simplification is not vital; if you do not want to do this,
8055 report the bug anyway and send us the entire test case you used.
8058 A patch for the bug.
8060 A patch for the bug does help us if it is a good one. But do not omit
8061 the necessary information, such as the test case, on the assumption that
8062 a patch is all we need. We might see problems with your patch and decide
8063 to fix the problem another way, or we might not understand it at all.
8065 Sometimes with a program as complicated as @value{GDBN} it is very hard to
8066 construct an example that will make the program follow a certain path
8067 through the code. If you do not send us the example, we will not be able
8068 to construct one, so we will not be able to verify that the bug is fixed.
8070 And if we cannot understand what bug you are trying to fix, or why your
8071 patch should be an improvement, we will not install it. A test case will
8072 help us to understand.
8075 A guess about what the bug is or what it depends on.
8077 Such guesses are usually wrong. Even we cannot guess right about such
8078 things without first using the debugger to find the facts.
8081 @c The readline documentation is distributed with the readline code
8082 @c and consists of the two following files:
8085 @c Use -I with makeinfo to point to the appropriate directory,
8086 @c environment var TEXINPUTS with TeX.
8087 @include rluser.texinfo
8088 @include inc-hist.texi
8091 @node Renamed Commands
8092 @appendix Renamed Commands
8094 The following commands were renamed in GDB 4, in order to make the
8095 command set as a whole more consistent and easier to use and remember:
8098 @kindex delete environment
8099 @kindex info copying
8100 @kindex info convenience
8101 @kindex info directories
8102 @kindex info editing
8103 @kindex info history
8104 @kindex info targets
8106 @kindex info version
8107 @kindex info warranty
8108 @kindex set addressprint
8109 @kindex set arrayprint
8110 @kindex set prettyprint
8111 @kindex set screen-height
8112 @kindex set screen-width
8113 @kindex set unionprint
8114 @kindex set vtblprint
8115 @kindex set demangle
8116 @kindex set asm-demangle
8117 @kindex set sevenbit-strings
8118 @kindex set array-max
8120 @kindex set history write
8121 @kindex show addressprint
8122 @kindex show arrayprint
8123 @kindex show prettyprint
8124 @kindex show screen-height
8125 @kindex show screen-width
8126 @kindex show unionprint
8127 @kindex show vtblprint
8128 @kindex show demangle
8129 @kindex show asm-demangle
8130 @kindex show sevenbit-strings
8131 @kindex show array-max
8132 @kindex show caution
8133 @kindex show history write
8138 @c END TEXI2ROFF-KILL
8140 OLD COMMAND NEW COMMAND
8142 --------------- -------------------------------
8143 @c END TEXI2ROFF-KILL
8144 add-syms add-symbol-file
8145 delete environment unset environment
8146 info convenience show convenience
8147 info copying show copying
8148 info directories show directories
8149 info editing show commands
8150 info history show values
8151 info targets help target
8152 info values show values
8153 info version show version
8154 info warranty show warranty
8155 set/show addressprint set/show print address
8156 set/show array-max set/show print elements
8157 set/show arrayprint set/show print array
8158 set/show asm-demangle set/show print asm-demangle
8159 set/show caution set/show confirm
8160 set/show demangle set/show print demangle
8161 set/show history write set/show history save
8162 set/show prettyprint set/show print pretty
8163 set/show screen-height set/show height
8164 set/show screen-width set/show width
8165 set/show sevenbit-strings set/show print sevenbit-strings
8166 set/show unionprint set/show print union
8167 set/show vtblprint set/show print vtbl
8169 unset [No longer an alias for delete]
8175 \vskip \parskip\vskip \baselineskip
8176 \halign{\tt #\hfil &\qquad#&\tt #\hfil\cr
8177 {\bf Old Command} &&{\bf New Command}\cr
8178 add-syms &&add-symbol-file\cr
8179 delete environment &&unset environment\cr
8180 info convenience &&show convenience\cr
8181 info copying &&show copying\cr
8182 info directories &&show directories \cr
8183 info editing &&show commands\cr
8184 info history &&show values\cr
8185 info targets &&help target\cr
8186 info values &&show values\cr
8187 info version &&show version\cr
8188 info warranty &&show warranty\cr
8189 set{\rm / }show addressprint &&set{\rm / }show print address\cr
8190 set{\rm / }show array-max &&set{\rm / }show print elements\cr
8191 set{\rm / }show arrayprint &&set{\rm / }show print array\cr
8192 set{\rm / }show asm-demangle &&set{\rm / }show print asm-demangle\cr
8193 set{\rm / }show caution &&set{\rm / }show confirm\cr
8194 set{\rm / }show demangle &&set{\rm / }show print demangle\cr
8195 set{\rm / }show history write &&set{\rm / }show history save\cr
8196 set{\rm / }show prettyprint &&set{\rm / }show print pretty\cr
8197 set{\rm / }show screen-height &&set{\rm / }show height\cr
8198 set{\rm / }show screen-width &&set{\rm / }show width\cr
8199 set{\rm / }show sevenbit-strings &&set{\rm / }show print sevenbit-strings\cr
8200 set{\rm / }show unionprint &&set{\rm / }show print union\cr
8201 set{\rm / }show vtblprint &&set{\rm / }show print vtbl\cr
8203 unset &&\rm(No longer an alias for delete)\cr
8206 @c END TEXI2ROFF-KILL
8209 @ifclear PRECONFIGURED
8210 @node Formatting Documentation
8211 @appendix Formatting Documentation
8213 @cindex GDB reference card
8214 @cindex reference card
8215 The GDB 4 release includes an already-formatted reference card, ready
8216 for printing with PostScript or GhostScript, in the @file{gdb}
8217 subdirectory of the main source directory@footnote{In
8218 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
8219 release.}. If you can use PostScript or GhostScript with your printer,
8220 you can print the reference card immediately with @file{refcard.ps}.
8222 The release also includes the source for the reference card. You
8223 can format it, using @TeX{}, by typing:
8229 The GDB reference card is designed to print in landscape mode on US
8230 ``letter'' size paper; that is, on a sheet 11 inches wide by 8.5 inches
8231 high. You will need to specify this form of printing as an option to
8232 your @sc{dvi} output program.
8234 @cindex documentation
8236 All the documentation for GDB comes as part of the machine-readable
8237 distribution. The documentation is written in Texinfo format, which is
8238 a documentation system that uses a single source file to produce both
8239 on-line information and a printed manual. You can use one of the Info
8240 formatting commands to create the on-line version of the documentation
8241 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
8243 GDB includes an already formatted copy of the on-line Info version of
8244 this manual in the @file{gdb} subdirectory. The main Info file is
8245 @file{gdb-@var{version-number}/gdb/gdb.info}, and it refers to
8246 subordinate files matching @samp{gdb.info*} in the same directory. If
8247 necessary, you can print out these files, or read them with any editor;
8248 but they are easier to read using the @code{info} subsystem in GNU Emacs
8249 or the standalone @code{info} program, available as part of the GNU
8250 Texinfo distribution.
8252 If you want to format these Info files yourself, you need one of the
8253 Info formatting programs, such as @code{texinfo-format-buffer} or
8256 If you have @code{makeinfo} installed, and are in the top level GDB
8257 source directory (@file{gdb-@value{GDBVN}}, in the case of version @value{GDBVN}), you can
8258 make the Info file by typing:
8265 If you want to typeset and print copies of this manual, you need @TeX{},
8266 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
8267 Texinfo definitions file.
8269 @TeX{} is a typesetting program; it does not print files directly, but
8270 produces output files called @sc{dvi} files. To print a typeset
8271 document, you need a program to print @sc{dvi} files. If your system
8272 has @TeX{} installed, chances are it has such a program. The precise
8273 command to use depends on your system; @kbd{lpr -d} is common; another
8274 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
8275 require a file name without any extension or a @samp{.dvi} extension.
8277 @TeX{} also requires a macro definitions file called
8278 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
8279 written in Texinfo format. On its own, @TeX{} cannot read, much less
8280 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
8281 and is located in the @file{gdb-@var{version-number}/texinfo}
8284 If you have @TeX{} and a @sc{dvi} printer program installed, you can
8285 typeset and print this manual. First switch to the the @file{gdb}
8286 subdirectory of the main source directory (for example, to
8287 @file{gdb-@value{GDBVN}/gdb}) and then type:
8293 @node Installing GDB
8294 @appendix Installing GDB
8295 @cindex configuring GDB
8296 @cindex installation
8298 GDB comes with a @code{configure} script that automates the process
8299 of preparing GDB for installation; you can then use @code{make} to
8300 build the @code{gdb} program.
8302 @c irrelevant in info file; it's as current as the code it lives with.
8303 @footnote{If you have a more recent version of GDB than @value{GDBVN},
8304 look at the @file{README} file in the sources; we may have improved the
8305 installation procedures since publishing this manual.}
8308 The GDB distribution includes all the source code you need for GDB in
8309 a single directory, whose name is usually composed by appending the
8310 version number to @samp{gdb}.
8312 For example, the GDB version @value{GDBVN} distribution is in the
8313 @file{gdb-@value{GDBVN}} directory. That directory contains:
8316 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
8317 script for configuring GDB and all its supporting libraries.
8319 @item gdb-@value{GDBVN}/gdb
8320 the source specific to GDB itself
8322 @item gdb-@value{GDBVN}/bfd
8323 source for the Binary File Descriptor library
8325 @item gdb-@value{GDBVN}/include
8328 @item gdb-@value{GDBVN}/libiberty
8329 source for the @samp{-liberty} free software library
8331 @item gdb-@value{GDBVN}/opcodes
8332 source for the library of opcode tables and disassemblers
8334 @item gdb-@value{GDBVN}/readline
8335 source for the GNU command-line interface
8337 @item gdb-@value{GDBVN}/glob
8338 source for the GNU filename pattern-matching subroutine
8340 @item gdb-@value{GDBVN}/mmalloc
8341 source for the GNU memory-mapped malloc package
8344 The simplest way to configure and build GDB is to run @code{configure}
8345 from the @file{gdb-@var{version-number}} source directory, which in
8346 this example is the @file{gdb-@value{GDBVN}} directory.
8348 First switch to the @file{gdb-@var{version-number}} source directory
8349 if you are not already in it; then run @code{configure}. Pass the
8350 identifier for the platform on which GDB will run as an
8356 cd gdb-@value{GDBVN}
8357 ./configure @var{host}
8362 where @var{host} is an identifier such as @samp{sun4} or
8363 @samp{decstation}, that identifies the platform where GDB will run.
8364 (You can often leave off @var{host}; @code{configure} tries to guess the
8365 correct value by examining your system.)
8367 Running @samp{configure @var{host}} and then running @code{make} builds the
8368 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
8369 libraries, then @code{gdb} itself. The configured source files, and the
8370 binaries, are left in the corresponding source directories.
8372 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
8373 system does not recognize this automatically when you run a different
8374 shell, you may need to run @code{sh} on it explicitly:
8377 sh configure @var{host}
8380 If you run @code{configure} from a directory that contains source
8381 directories for multiple libraries or programs, such as the
8382 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
8383 creates configuration files for every directory level underneath (unless
8384 you tell it not to, with the @samp{--norecursion} option).
8386 You can run the @code{configure} script from any of the
8387 subordinate directories in the GDB distribution if you only want to
8388 configure that subdirectory, but be sure to specify a path to it.
8390 For example, with version @value{GDBVN}, type the following to configure only
8391 the @code{bfd} subdirectory:
8395 cd gdb-@value{GDBVN}/bfd
8396 ../configure @var{host}
8400 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
8401 However, you should make sure that the shell on your path (named by
8402 the @samp{SHELL} environment variable) is publicly readable. Remember
8403 that GDB uses the shell to start your program---some systems refuse to
8404 let GDB debug child processes whose programs are not readable.
8407 * Separate Objdir:: Compiling GDB in another directory
8408 * Config Names:: Specifying names for hosts and targets
8409 * configure Options:: Summary of options for configure
8412 @node Separate Objdir
8413 @section Compiling GDB in another directory
8415 If you want to run GDB versions for several host or target machines,
8416 you need a different @code{gdb} compiled for each combination of
8417 host and target. @code{configure} is designed to make this easy by
8418 allowing you to generate each configuration in a separate subdirectory,
8419 rather than in the source directory. If your @code{make} program
8420 handles the @samp{VPATH} feature (GNU @code{make} does), running
8421 @code{make} in each of these directories builds the @code{gdb}
8422 program specified there.
8424 To build @code{gdb} in a separate directory, run @code{configure}
8425 with the @samp{--srcdir} option to specify where to find the source.
8426 (You also need to specify a path to find @code{configure}
8427 itself from your working directory. If the path to @code{configure}
8428 would be the same as the argument to @samp{--srcdir}, you can leave out
8429 the @samp{--srcdir} option; it will be assumed.)
8431 For example, with version @value{GDBVN}, you can build GDB in a separate
8432 directory for a Sun 4 like this:
8436 cd gdb-@value{GDBVN}
8439 ../gdb-@value{GDBVN}/configure sun4
8444 When @code{configure} builds a configuration using a remote source
8445 directory, it creates a tree for the binaries with the same structure
8446 (and using the same names) as the tree under the source directory. In
8447 the example, you'd find the Sun 4 library @file{libiberty.a} in the
8448 directory @file{gdb-sun4/libiberty}, and GDB itself in
8449 @file{gdb-sun4/gdb}.
8451 One popular reason to build several GDB configurations in separate
8452 directories is to configure GDB for cross-compiling (where GDB
8453 runs on one machine---the host---while debugging programs that run on
8454 another machine---the target). You specify a cross-debugging target by
8455 giving the @samp{--target=@var{target}} option to @code{configure}.
8457 When you run @code{make} to build a program or library, you must run
8458 it in a configured directory---whatever directory you were in when you
8459 called @code{configure} (or one of its subdirectories).
8461 The @code{Makefile} that @code{configure} generates in each source
8462 directory also runs recursively. If you type @code{make} in a source
8463 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
8464 directory configured with @samp{--srcdir=@var{path}/gdb-@value{GDBVN}}), you
8465 will build all the required libraries, and then build GDB.
8467 When you have multiple hosts or targets configured in separate
8468 directories, you can run @code{make} on them in parallel (for example,
8469 if they are NFS-mounted on each of the hosts); they will not interfere
8473 @section Specifying names for hosts and targets
8475 The specifications used for hosts and targets in the @code{configure}
8476 script are based on a three-part naming scheme, but some short predefined
8477 aliases are also supported. The full naming scheme encodes three pieces
8478 of information in the following pattern:
8481 @var{architecture}-@var{vendor}-@var{os}
8484 For example, you can use the alias @code{sun4} as a @var{host} argument,
8485 or as the value for @var{target} in a @code{--target=@var{target}}
8486 option. The equivalent full name is @samp{sparc-sun-sunos4}.
8488 The @code{configure} script accompanying GDB does not provide
8489 any query facility to list all supported host and target names or
8490 aliases. @code{configure} calls the Bourne shell script
8491 @code{config.sub} to map abbreviations to full names; you can read the
8492 script, if you wish, or you can use it to test your guesses on
8493 abbreviations---for example:
8496 % sh config.sub sun4
8497 sparc-sun-sunos4.1.1
8498 % sh config.sub sun3
8500 % sh config.sub decstation
8502 % sh config.sub hp300bsd
8504 % sh config.sub i386v
8506 % sh config.sub i786v
8507 Invalid configuration `i786v': machine `i786v' not recognized
8511 @code{config.sub} is also distributed in the GDB source
8512 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
8514 @node configure Options
8515 @section @code{configure} options
8517 Here is a summary of the @code{configure} options and arguments that
8518 are most often useful for building @value{GDBN}. @code{configure} also has
8519 several other options not listed here. @inforef{What Configure
8520 Does,,configure.info}, for a full explanation of @code{configure}.
8521 @c FIXME: Would this be more, or less, useful as an xref (ref to printed
8522 @c manual in the printed manual, ref to info file only from the info file)?
8525 configure @r{[}--help@r{]}
8526 @r{[}--prefix=@var{dir}@r{]}
8527 @r{[}--srcdir=@var{path}@r{]}
8528 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
8529 @r{[}--target=@var{target}@r{]} @var{host}
8533 You may introduce options with a single @samp{-} rather than
8534 @samp{--} if you prefer; but you may abbreviate option names if you use
8539 Display a quick summary of how to invoke @code{configure}.
8541 @item -prefix=@var{dir}
8542 Configure the source to install programs and files under directory
8545 @item --srcdir=@var{path}
8546 @strong{Warning: using this option requires GNU @code{make}, or another
8547 @code{make} that implements the @code{VPATH} feature.}@*
8548 Use this option to make configurations in directories separate from the
8549 GDB source directories. Among other things, you can use this to
8550 build (or maintain) several configurations simultaneously, in separate
8551 directories. @code{configure} writes configuration specific files in
8552 the current directory, but arranges for them to use the source in the
8553 directory @var{path}. @code{configure} will create directories under
8554 the working directory in parallel to the source directories below
8558 Configure only the directory level where @code{configure} is executed; do not
8559 propagate configuration to subdirectories.
8562 @emph{Remove} files otherwise built during configuration.
8564 @c This does not work (yet if ever). FIXME.
8565 @c @item --parse=@var{lang} @dots{}
8566 @c Configure the GDB expression parser to parse the listed languages.
8567 @c @samp{all} configures GDB for all supported languages. To get a
8568 @c list of all supported languages, omit the argument. Without this
8569 @c option, GDB is configured to parse all supported languages.
8571 @item --target=@var{target}
8572 Configure GDB for cross-debugging programs running on the specified
8573 @var{target}. Without this option, GDB is configured to debug
8574 programs that run on the same machine (@var{host}) as GDB itself.
8576 There is no convenient way to generate a list of all available targets.
8578 @item @var{host} @dots{}
8579 Configure GDB to run on the specified @var{host}.
8581 There is no convenient way to generate a list of all available hosts.
8585 @code{configure} accepts other options, for compatibility with
8586 configuring other GNU tools recursively; but these are the only
8587 options that affect GDB or its supporting libraries.
8596 % I think something like @colophon should be in texinfo. In the
8598 \long\def\colophon{\hbox to0pt{}\vfill
8599 \centerline{The body of this manual is set in}
8600 \centerline{\fontname\tenrm,}
8601 \centerline{with headings in {\bf\fontname\tenbf}}
8602 \centerline{and examples in {\tt\fontname\tentt}.}
8603 \centerline{{\it\fontname\tenit\/},}
8604 \centerline{{\bf\fontname\tenbf}, and}
8605 \centerline{{\sl\fontname\tensl\/}}
8606 \centerline{are used for emphasis.}\vfill}
8608 % Blame: pesch@cygnus.com, 1991.