1 \input texinfo @c -*-texinfo-*-
2 @c Copyright 1988 1989 1990 1991 1992 1993 1994 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
31 @c !!set GDB manual's edition---not the same as GDB version!
34 @c !!set GDB manual's revision date
35 @set DATE January 1994
37 @c GDB CHANGELOG CONSULTED BETWEEN:
38 @c Fri Oct 11 23:27:06 1991 John Gilmore (gnu at cygnus.com)
39 @c Sat Dec 22 02:51:40 1990 John Gilmore (gnu at cygint)
41 @c THIS MANUAL REQUIRES TEXINFO-2 macros and info-makers to format properly.
44 @c This is a dir.info fragment to support semi-automated addition of
45 @c manuals to an info tree. zoo@cygnus.com is developing this facility.
48 * Gdb: (gdb). The GNU debugger.
55 This file documents the GNU debugger @value{GDBN}.
58 This is Edition @value{EDITION}, @value{DATE},
59 of @cite{Debugging with @value{GDBN}: the GNU Source-Level Debugger}
60 for GDB Version @value{GDBVN}.
62 Copyright (C) 1988, '89, '90, '91, '92, '93 Free Software Foundation, Inc.
64 Permission is granted to make and distribute verbatim copies of
65 this manual provided the copyright notice and this permission notice
66 are preserved on all copies.
69 Permission is granted to process this file through TeX and print the
70 results, provided the printed document carries copying permission
71 notice identical to this one except for the removal of this paragraph
72 (this paragraph not being relevant to the printed manual).
75 Permission is granted to copy and distribute modified versions of this
76 manual under the conditions for verbatim copying, provided also that the
77 entire resulting derived work is distributed under the terms of a
78 permission notice identical to this one.
80 Permission is granted to copy and distribute translations of this manual
81 into another language, under the above conditions for modified versions.
85 @title Debugging with @value{GDBN}
86 @subtitle The GNU Source-Level Debugger
88 @subtitle (@value{TARGET})
91 @subtitle Edition @value{EDITION}, for @value{GDBN} version @value{GDBVN}
92 @subtitle @value{DATE}
93 @author 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, '89, '90, '91, '92, '93 Free Software
108 Published by the Free Software Foundation @*
109 675 Massachusetts Avenue, @*
110 Cambridge, MA 02139 USA @*
111 Printed copies are available for $20 each. @*
112 ISBN 1-882114-11-6 @*
114 Permission is granted to make and distribute verbatim copies of
115 this manual provided the copyright notice and this permission notice
116 are preserved on all copies.
118 Permission is granted to copy and distribute modified versions of this
119 manual under the conditions for verbatim copying, provided also that the
120 entire resulting derived work is distributed under the terms of a
121 permission notice identical to this one.
123 Permission is granted to copy and distribute translations of this manual
124 into another language, under the above conditions for modified versions.
130 @top Debugging with @value{GDBN}
132 This file describes @value{GDBN}, the GNU symbolic debugger.
134 This is Edition @value{EDITION}, @value{DATE}, for GDB Version @value{GDBVN}.
137 * Summary:: Summary of @value{GDBN}
139 * New Features:: New features since GDB version 3.5
142 * Sample Session:: A sample @value{GDBN} session
145 * Invocation:: Getting in and out of @value{GDBN}
146 * Commands:: @value{GDBN} commands
147 * Running:: Running programs under @value{GDBN}
148 * Stopping:: Stopping and continuing
149 * Stack:: Examining the stack
150 * Source:: Examining source files
151 * Data:: Examining data
153 * Languages:: Using @value{GDBN} with different languages
156 * C:: C language support
158 @c remnant makeinfo bug, blank line needed after two end-ifs?
160 * Symbols:: Examining the symbol table
161 * Altering:: Altering execution
162 * GDB Files:: @value{GDBN} files
163 * Targets:: Specifying a debugging target
164 * Controlling GDB:: Controlling @value{GDBN}
165 * Sequences:: Canned sequences of commands
167 * Emacs:: Using @value{GDBN} under GNU Emacs
170 * GDB Bugs:: Reporting bugs in @value{GDBN}
171 * Command Line Editing:: Facilities of the readline library
172 * Using History Interactively::
176 @ifclear PRECONFIGURED
177 * Formatting Documentation:: How to format and print GDB documentation
178 * Installing GDB:: Installing GDB
186 @unnumbered Summary of @value{GDBN}
188 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
189 going on ``inside'' another program while it executes---or what another
190 program was doing at the moment it crashed.
192 @value{GDBN} can do four main kinds of things (plus other things in support of
193 these) to help you catch bugs in the act:
197 Start your program, specifying anything that might affect its behavior.
200 Make your program stop on specified conditions.
203 Examine what has happened, when your program has stopped.
206 Change things in your program, so you can experiment with correcting the
207 effects of one bug and go on to learn about another.
211 You can use @value{GDBN} to debug programs written in C or C++.
212 @c "MOD2" used as a "miscellaneous languages" flag here.
213 @c This is acceptable while there is no real doc for Chill and Pascal.
215 For more information, see @ref{Support,,Supported languages}.
218 For more information, see @ref{C,,C and C++}.
220 Support for Modula-2 and Chill is partial. For information on Modula-2,
221 see @ref{Modula-2,,Modula-2}. There is no further documentation on Chill yet.
223 Debugging Pascal programs which use sets, subranges, file variables, or nested
224 functions does not currently work. @value{GDBN} does not support
225 entering expressions, printing values, or similar features using Pascal syntax.
229 @value{GDBN} can be used to debug programs written in Fortran, although
230 it does not yet support entering expressions, printing values, or
231 similar features using Fortran syntax. It may be necessary to refer to
232 some variables with a trailing underscore.
237 * Free Software:: Freely redistributable software
238 * Contributors:: Contributors to GDB
242 @unnumberedsec Free software
244 @value{GDBN} is @dfn{free software}, protected by the GNU General Public License
245 (GPL). The GPL gives you the freedom to copy or adapt a licensed
246 program---but every person getting a copy also gets with it the
247 freedom to modify that copy (which means that they must get access to
248 the source code), and the freedom to distribute further copies.
249 Typical software companies use copyrights to limit your freedoms; the
250 Free Software Foundation uses the GPL to preserve these freedoms.
252 Fundamentally, the General Public License is a license which says that
253 you have these freedoms and that you cannot take these freedoms away
257 @unnumberedsec Contributors to GDB
259 Richard Stallman was the original author of GDB, and of many other GNU
260 programs. Many others have contributed to its development. This
261 section attempts to credit major contributors. One of the virtues of
262 free software is that everyone is free to contribute to it; with
263 regret, we cannot actually acknowledge everyone here. The file
264 @file{ChangeLog} in the GDB distribution approximates a blow-by-blow
267 Changes much prior to version 2.0 are lost in the mists of time.
270 @emph{Plea:} Additions to this section are particularly welcome. If you
271 or your friends (or enemies, to be evenhanded) have been unfairly
272 omitted from this list, we would like to add your names!
275 So that they may not regard their long labor as thankless, we
276 particularly thank those who shepherded GDB through major releases:
277 Fred Fish (releases 4.12, 4.11, 4.10, and 4.9),
278 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4),
279 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
280 Jim Kingdon (releases 3.5, 3.4, and 3.3);
281 and Randy Smith (releases 3.2, 3.1, and 3.0).
282 As major maintainer of GDB for some period, each
283 contributed significantly to the structure, stability, and capabilities
284 of the entire debugger.
286 Richard Stallman, assisted at various times by Peter TerMaat, Chris
287 Hanson, and Richard Mlynarik, handled releases through 2.8.
290 Michael Tiemann is the author of most of the GNU C++ support in GDB,
291 with significant additional contributions from Per Bothner. James
292 Clark wrote the GNU C++ demangler. Early work on C++ was by Peter
293 TerMaat (who also did much general update work leading to release 3.0).
296 GDB 4 uses the BFD subroutine library to examine multiple
297 object-file formats; BFD was a joint project of David V.
298 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
300 David Johnson wrote the original COFF support; Pace Willison did
301 the original support for encapsulated COFF.
303 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
304 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
306 Jean-Daniel Fekete contributed Sun 386i support.
307 Chris Hanson improved the HP9000 support.
308 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
309 David Johnson contributed Encore Umax support.
310 Jyrki Kuoppala contributed Altos 3068 support.
311 Jeff Law contributed HP PA and SOM support.
312 Keith Packard contributed NS32K support.
313 Doug Rabson contributed Acorn Risc Machine support.
314 Bob Rusk contributed Harris Nighthawk CX-UX support.
315 Chris Smith contributed Convex support (and Fortran debugging).
316 Jonathan Stone contributed Pyramid support.
317 Michael Tiemann contributed SPARC support.
318 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
319 Pace Willison contributed Intel 386 support.
320 Jay Vosburgh contributed Symmetry support.
322 Rich Schaefer and Peter Schauer helped with support of SunOS shared
325 Jay Fenlason and Roland McGrath ensured that GDB and GAS agree about
326 several machine instruction sets.
328 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped
329 develop remote debugging. Intel Corporation and Wind River Systems
330 contributed remote debugging modules for their products.
332 Brian Fox is the author of the readline libraries providing
333 command-line editing and command history.
335 Andrew Beers of SUNY Buffalo wrote the language-switching code,
337 the Modula-2 support,
339 and contributed the Languages chapter of this manual.
341 Fred Fish wrote most of the support for Unix System Vr4.
343 He also enhanced the command-completion support to cover C++ overloaded
347 Hitachi America, Ltd. sponsored the support for Hitachi microprocessors.
349 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
352 Stu Grossman wrote gdbserver.
354 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
355 nearly innumerable bug fixes and cleanups throughout GDB.
359 @unnumbered New Features since GDB Version 3.5
363 Using the new command @code{target}, you can select at runtime whether
364 you are debugging local files, local processes, standalone systems over
365 a serial port, or realtime systems over a TCP/IP connection. The
366 command @code{load} can download programs into a remote system. Serial
367 stubs are available for Motorola 680x0, Intel 80386, and Sparc remote
368 systems; GDB also supports debugging realtime processes running under
369 VxWorks, using SunRPC Remote Procedure Calls over TCP/IP to talk to a
370 debugger stub on the target system. Internally, GDB now uses a function
371 vector to mediate access to different targets; if you need to add your
372 own support for a remote protocol, this makes it much easier.
375 GDB now sports watchpoints as well as breakpoints. You can use a
376 watchpoint to stop execution whenever the value of an expression
377 changes, without having to predict a particular place in your program
378 where this may happen.
381 Commands that issue wide output now insert newlines at places designed
382 to make the output more readable.
384 @item Object Code Formats
385 GDB uses a new library called the Binary File Descriptor (BFD) Library
386 to permit it to switch dynamically, without reconfiguration or
387 recompilation, between different object-file formats. Formats currently
388 supported are COFF, ELF, a.out, Intel 960 b.out, MIPS ECOFF, HPPA SOM
389 (with stabs debugging), and S-records; files may be read as .o files,
390 archive libraries, or core dumps. BFD is available as a subroutine
391 library so that other programs may take advantage of it, and the other
392 GNU binary utilities are being converted to use it.
394 @item Configuration and Ports
395 Compile-time configuration (to select a particular architecture and
396 operating system) is much easier. The script @code{configure} now
397 allows you to configure GDB as either a native debugger or a
398 cross-debugger. @xref{Installing GDB}, for details on how to
402 The user interface to the GDB control variables is simpler,
403 and is consolidated in two commands, @code{set} and @code{show}. Output
404 lines are now broken at readable places, rather than overflowing onto
405 the next line. You can suppress output of machine-level addresses,
406 displaying only source language information.
409 GDB now supports C++ multiple inheritance (if used with a GCC
410 version 2 compiler), and also has limited support for C++ exception
411 handling, with the commands @code{catch} and @code{info catch}: GDB
412 can break when an exception is raised, before the stack is peeled back
413 to the exception handler's context.
417 GDB now has preliminary support for the GNU Modula-2 compiler, currently
418 under development at the State University of New York at Buffalo.
419 Coordinated development of both GDB and the GNU Modula-2 compiler will
420 continue. Other Modula-2 compilers are currently not supported, and
421 attempting to debug programs compiled with them will likely result in an
422 error as the symbol table of the executable is read in.
425 @item Command Rationalization
426 Many GDB commands have been renamed to make them easier to remember
427 and use. In particular, the subcommands of @code{info} and
428 @code{show}/@code{set} are grouped to make the former refer to the state
429 of your program, and the latter refer to the state of GDB itself.
430 @xref{Renamed Commands}, for details on what commands were renamed.
432 @item Shared Libraries
433 GDB 4 can debug programs and core files that use SunOS, SVR4, or IBM RS/6000
437 On some systems, GDB 4 has facilities to debug multi-thread programs.
440 GDB 4 has a reference card. @xref{Formatting Documentation,,Formatting
441 the Documentation}, for instructions about how to print it.
447 @chapter A Sample @value{GDBN} Session
449 You can use this manual at your leisure to read all about @value{GDBN}.
450 However, a handful of commands are enough to get started using the
451 debugger. This chapter illustrates those commands.
454 In this sample session, we emphasize user input like this: @b{input},
455 to make it easier to pick out from the surrounding output.
458 @c FIXME: this example may not be appropriate for some configs, where
459 @c FIXME...primary interest is in remote use.
461 One of the preliminary versions of GNU @code{m4} (a generic macro
462 processor) exhibits the following bug: sometimes, when we change its
463 quote strings from the default, the commands used to capture one macro
464 definition within another stop working. In the following short @code{m4}
465 session, we define a macro @code{foo} which expands to @code{0000}; we
466 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
467 same thing. However, when we change the open quote string to
468 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
469 procedure fails to define a new synonym @code{baz}:
478 @b{define(bar,defn(`foo'))}
482 @b{changequote(<QUOTE>,<UNQUOTE>)}
484 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
487 m4: End of input: 0: fatal error: EOF in string
491 Let us use @value{GDBN} to try to see what is going on.
494 $ @b{@value{GDBP} m4}
495 @c FIXME: this falsifies the exact text played out, to permit smallbook
496 @c FIXME... format to come out better.
497 GDB is free software and you are welcome to distribute copies
498 of it under certain conditions; type "show copying" to see
500 There is absolutely no warranty for GDB; type "show warranty"
502 GDB @value{GDBVN}, Copyright 1993 Free Software Foundation, Inc...
507 @value{GDBN} reads only enough symbol data to know where to find the
508 rest when needed; as a result, the first prompt comes up very quickly.
509 We now tell @value{GDBN} to use a narrower display width than usual, so
510 that examples fit in this manual.
513 (@value{GDBP}) @b{set width 70}
517 We need to see how the @code{m4} built-in @code{changequote} works.
518 Having looked at the source, we know the relevant subroutine is
519 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
520 @code{break} command.
523 (@value{GDBP}) @b{break m4_changequote}
524 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
528 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
529 control; as long as control does not reach the @code{m4_changequote}
530 subroutine, the program runs as usual:
533 (@value{GDBP}) @b{run}
534 Starting program: /work/Editorial/gdb/gnu/m4/m4
542 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
543 suspends execution of @code{m4}, displaying information about the
544 context where it stops.
547 @b{changequote(<QUOTE>,<UNQUOTE>)}
549 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
551 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
555 Now we use the command @code{n} (@code{next}) to advance execution to
556 the next line of the current function.
560 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
565 @code{set_quotes} looks like a promising subroutine. We can go into it
566 by using the command @code{s} (@code{step}) instead of @code{next}.
567 @code{step} goes to the next line to be executed in @emph{any}
568 subroutine, so it steps into @code{set_quotes}.
572 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
574 530 if (lquote != def_lquote)
578 The display that shows the subroutine where @code{m4} is now
579 suspended (and its arguments) is called a stack frame display. It
580 shows a summary of the stack. We can use the @code{backtrace}
581 command (which can also be spelled @code{bt}), to see where we are
582 in the stack as a whole: the @code{backtrace} command displays a
583 stack frame for each active subroutine.
586 (@value{GDBP}) @b{bt}
587 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
589 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
591 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
592 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
594 #4 0x79dc in expand_input () at macro.c:40
595 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
599 We step through a few more lines to see what happens. The first two
600 times, we can use @samp{s}; the next two times we use @code{n} to avoid
601 falling into the @code{xstrdup} subroutine.
605 0x3b5c 532 if (rquote != def_rquote)
607 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
608 def_lquote : xstrdup(lq);
610 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
613 538 len_lquote = strlen(rquote);
617 The last line displayed looks a little odd; we can examine the variables
618 @code{lquote} and @code{rquote} to see if they are in fact the new left
619 and right quotes we specified. We use the command @code{p}
620 (@code{print}) to see their values.
623 (@value{GDBP}) @b{p lquote}
624 $1 = 0x35d40 "<QUOTE>"
625 (@value{GDBP}) @b{p rquote}
626 $2 = 0x35d50 "<UNQUOTE>"
630 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
631 To look at some context, we can display ten lines of source
632 surrounding the current line with the @code{l} (@code{list}) command.
638 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
640 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
643 538 len_lquote = strlen(rquote);
644 539 len_rquote = strlen(lquote);
651 Let us step past the two lines that set @code{len_lquote} and
652 @code{len_rquote}, and then examine the values of those variables.
656 539 len_rquote = strlen(lquote);
659 (@value{GDBP}) @b{p len_lquote}
661 (@value{GDBP}) @b{p len_rquote}
666 That certainly looks wrong, assuming @code{len_lquote} and
667 @code{len_rquote} are meant to be the lengths of @code{lquote} and
668 @code{rquote} respectively. We can set them to better values using
669 the @code{p} command, since it can print the value of
670 any expression---and that expression can include subroutine calls and
674 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
676 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
681 Is that enough to fix the problem of using the new quotes with the
682 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
683 executing with the @code{c} (@code{continue}) command, and then try the
684 example that caused trouble initially:
690 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
697 Success! The new quotes now work just as well as the default ones. The
698 problem seems to have been just the two typos defining the wrong
699 lengths. We allow @code{m4} exit by giving it an EOF as input:
703 Program exited normally.
707 The message @samp{Program exited normally.} is from @value{GDBN}; it
708 indicates @code{m4} has finished executing. We can end our @value{GDBN}
709 session with the @value{GDBN} @code{quit} command.
712 (@value{GDBP}) @b{quit}
717 @chapter Getting In and Out of @value{GDBN}
719 This chapter discusses how to start @value{GDBN}, and how to get out of it.
720 (The essentials: type @samp{@value{GDBP}} to start GDB, and type @kbd{quit}
721 or @kbd{C-d} to exit.)
724 * Invoking GDB:: How to start @value{GDBN}
725 * Quitting GDB:: How to quit @value{GDBN}
726 * Shell Commands:: How to use shell commands inside @value{GDBN}
730 @section Invoking @value{GDBN}
733 For details on starting up @value{GDBP} as a
734 remote debugger attached to a Hitachi microprocessor, see @ref{Hitachi
735 Remote,,@value{GDBN} and Hitachi Microprocessors}.
738 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
739 @value{GDBN} reads commands from the terminal until you tell it to exit.
741 You can also run @code{@value{GDBP}} with a variety of arguments and options,
742 to specify more of your debugging environment at the outset.
745 The command-line options described here are designed
746 to cover a variety of situations; in some environments, some of these
747 options may effectively be unavailable.
750 The most usual way to start @value{GDBN} is with one argument,
751 specifying an executable program:
754 @value{GDBP} @var{program}
759 You can also start with both an executable program and a core file
763 @value{GDBP} @var{program} @var{core}
766 You can, instead, specify a process ID as a second argument, if you want
767 to debug a running process:
770 @value{GDBP} @var{program} 1234
774 would attach @value{GDBN} to process @code{1234} (unless you also have a file
775 named @file{1234}; @value{GDBN} does check for a core file first).
777 Taking advantage of the second command-line argument requires a fairly
778 complete operating system; when you use @value{GDBN} as a remote debugger
779 attached to a bare board, there may not be any notion of ``process'',
780 and there is often no way to get a core dump.
784 You can further control how @value{GDBN} starts up by using command-line
785 options. @value{GDBN} itself can remind you of the options available.
795 to display all available options and briefly describe their use
796 (@samp{@value{GDBP} -h} is a shorter equivalent).
798 All options and command line arguments you give are processed
799 in sequential order. The order makes a difference when the
800 @samp{-x} option is used.
806 * Remote Serial:: @value{GDBN} remote serial protocol
809 * i960-Nindy Remote:: @value{GDBN} with a remote i960 (Nindy)
812 * UDI29K Remote:: The UDI protocol for AMD29K
813 * EB29K Remote:: The EBMON protocol for AMD29K
816 * VxWorks Remote:: @value{GDBN} and VxWorks
819 * ST2000 Remote:: @value{GDBN} with a Tandem ST2000
822 * Hitachi Remote:: @value{GDBN} and Hitachi Microprocessors
825 * MIPS Remote:: @value{GDBN} and MIPS boards
828 * Simulator:: Simulated CPU target
831 @c remnant makeinfo bug requires this blank line after *two* end-ifblahs:
833 * File Options:: Choosing files
834 * Mode Options:: Choosing modes
842 @subsection Choosing files
845 When @value{GDBN} starts, it reads any arguments other than options as
846 specifying an executable file and core file (or process ID). This is
847 the same as if the arguments were specified by the @samp{-se} and
848 @samp{-c} options respectively. (@value{GDBN} reads the first argument
849 that does not have an associated option flag as equivalent to the
850 @samp{-se} option followed by that argument; and the second argument
851 that does not have an associated option flag, if any, as equivalent to
852 the @samp{-c} option followed by that argument.)
855 When @value{GDBN} starts, it reads any argument other than options as
856 specifying an executable file. This is the same as if the argument was
857 specified by the @samp{-se} option.
860 Many options have both long and short forms; both are shown in the
861 following list. @value{GDBN} also recognizes the long forms if you truncate
862 them, so long as enough of the option is present to be unambiguous.
863 (If you prefer, you can flag option arguments with @samp{--} rather
864 than @samp{-}, though we illustrate the more usual convention.)
867 @item -symbols @var{file}
869 Read symbol table from file @var{file}.
871 @item -exec @var{file}
873 Use file @var{file} as the executable file to execute when
878 appropriate, and for examining pure data in conjunction with a core
883 Read symbol table from file @var{file} and use it as the executable
887 @item -core @var{file}
889 Use file @var{file} as a core dump to examine.
891 @item -c @var{number}
892 Connect to process ID @var{number}, as with the @code{attach} command
893 (unless there is a file in core-dump format named @var{number}, in which
894 case @samp{-c} specifies that file as a core dump to read).
897 @item -command @var{file}
899 Execute @value{GDBN} commands from file @var{file}. @xref{Command
900 Files,, Command files}.
902 @item -directory @var{directory}
903 @itemx -d @var{directory}
904 Add @var{directory} to the path to search for source files.
909 @emph{Warning: this option depends on operating system facilities that are not
910 supported on all systems.}@*
911 If memory-mapped files are available on your system through the @code{mmap}
912 system call, you can use this option
913 to have @value{GDBN} write the symbols from your
914 program into a reusable file in the current directory. If the program you are debugging is
915 called @file{/tmp/fred}, the mapped symbol file is @file{./fred.syms}.
916 Future @value{GDBN} debugging sessions notice the presence of this file,
917 and can quickly map in symbol information from it, rather than reading
918 the symbol table from the executable program.
920 @c FIXME! Really host, not target?
921 The @file{.syms} file is specific to the host machine where @value{GDBN}
922 is run. It holds an exact image of the internal @value{GDBN} symbol
923 table. It cannot be shared across multiple host platforms.
928 Read each symbol file's entire symbol table immediately, rather than
929 the default, which is to read it incrementally as it is needed.
930 This makes startup slower, but makes future operations faster.
934 The @code{-mapped} and @code{-readnow} options are typically combined in
935 order to build a @file{.syms} file that contains complete symbol
936 information. (@xref{Files,,Commands to specify files}, for information
937 on @file{.syms} files.) A simple GDB invocation to do nothing but build
938 a @file{.syms} file for future use is:
941 gdb -batch -nx -mapped -readnow programname
946 @subsection Choosing modes
948 You can run @value{GDBN} in various alternative modes---for example, in
949 batch mode or quiet mode.
954 Do not execute commands from any initialization files (normally called
955 @file{@value{GDBINIT}}). Normally, the commands in these files are
956 executed after all the command options and arguments have been
957 processed. @xref{Command Files,,Command files}.
961 ``Quiet''. Do not print the introductory and copyright messages. These
962 messages are also suppressed in batch mode.
965 Run in batch mode. Exit with status @code{0} after processing all the
966 command files specified with @samp{-x} (and all commands from
967 initialization files, if not inhibited with @samp{-n}). Exit with
968 nonzero status if an error occurs in executing the @value{GDBN} commands
969 in the command files.
971 Batch mode may be useful for running @value{GDBN} as a filter, for example to
972 download and run a program on another computer; in order to make this
973 more useful, the message
976 Program exited normally.
980 (which is ordinarily issued whenever a program running under @value{GDBN} control
981 terminates) is not issued when running in batch mode.
983 @item -cd @var{directory}
984 Run @value{GDBN} using @var{directory} as its working directory,
985 instead of the current directory.
988 @item -context @var{authentication}
989 When the Energize programming system starts up @value{GDBN}, it uses this
990 option to trigger an alternate mode of interaction.
991 @var{authentication} is a pair of numeric codes that identify @value{GDBN}
992 as a client in the Energize environment. Avoid this option when you run
993 @value{GDBN} directly from the command line. See @ref{Energize,,Using
994 @value{GDBN} with Energize} for more discussion of using @value{GDBN} with Energize.
1000 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells @value{GDBN}
1001 to output the full file name and line number in a standard,
1002 recognizable fashion each time a stack frame is displayed (which
1003 includes each time your program stops). This recognizable format looks
1004 like two @samp{\032} characters, followed by the file name, line number
1005 and character position separated by colons, and a newline. The
1006 Emacs-to-@value{GDBN} interface program uses the two @samp{\032} characters as
1007 a signal to display the source code for the frame.
1012 Set the line speed (baud rate or bits per second) of any serial
1013 interface used by @value{GDBN} for remote debugging.
1015 @item -tty @var{device}
1016 Run using @var{device} for your program's standard input and output.
1017 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1022 @section Quitting @value{GDBN}
1023 @cindex exiting @value{GDBN}
1024 @cindex leaving @value{GDBN}
1030 To exit @value{GDBN}, use the @code{quit} command (abbreviated @code{q}), or type
1031 an end-of-file character (usually @kbd{C-d}).
1035 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
1036 terminates the action of any @value{GDBN} command that is in progress and
1037 returns to @value{GDBN} command level. It is safe to type the interrupt
1038 character at any time because @value{GDBN} does not allow it to take effect
1039 until a time when it is safe.
1042 If you have been using @value{GDBN} to control an attached process or
1043 device, you can release it with the @code{detach} command
1044 (@pxref{Attach, ,Debugging an already-running process}).
1047 @node Shell Commands
1048 @section Shell commands
1050 If you need to execute occasional shell commands during your
1051 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1052 just use the @code{shell} command.
1055 @item shell @var{command string}
1057 @cindex shell escape
1058 Invoke a the standard shell to execute @var{command string}.
1060 If it exists, the environment variable @code{SHELL} determines which
1061 shell to run. Otherwise @value{GDBN} uses @code{/bin/sh}.
1065 The utility @code{make} is often needed in development environments.
1066 You do not have to use the @code{shell} command for this purpose in
1070 @item make @var{make-args}
1072 @cindex calling make
1073 Execute the @code{make} program with the specified
1074 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1078 @chapter @value{GDBN} Commands
1080 You can abbreviate a @value{GDBN} command to the first few letters of the command
1081 name, if that abbreviation is unambiguous; and you can repeat certain
1082 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1083 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1084 show you the alternatives available, if there is more than one possibility).
1087 * Command Syntax:: How to give commands to @value{GDBN}
1088 * Completion:: Command completion
1089 * Help:: How to ask @value{GDBN} for help
1092 @node Command Syntax
1093 @section Command syntax
1095 A @value{GDBN} command is a single line of input. There is no limit on
1096 how long it can be. It starts with a command name, which is followed by
1097 arguments whose meaning depends on the command name. For example, the
1098 command @code{step} accepts an argument which is the number of times to
1099 step, as in @samp{step 5}. You can also use the @code{step} command
1100 with no arguments. Some command names do not allow any arguments.
1102 @cindex abbreviation
1103 @value{GDBN} command names may always be truncated if that abbreviation is
1104 unambiguous. Other possible command abbreviations are listed in the
1105 documentation for individual commands. In some cases, even ambiguous
1106 abbreviations are allowed; for example, @code{s} is specially defined as
1107 equivalent to @code{step} even though there are other commands whose
1108 names start with @code{s}. You can test abbreviations by using them as
1109 arguments to the @code{help} command.
1111 @cindex repeating commands
1113 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1114 repeat the previous command. Certain commands (for example, @code{run})
1115 will not repeat this way; these are commands whose unintentional
1116 repetition might cause trouble and which you are unlikely to want to
1119 The @code{list} and @code{x} commands, when you repeat them with
1120 @key{RET}, construct new arguments rather than repeating
1121 exactly as typed. This permits easy scanning of source or memory.
1123 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1124 output, in a way similar to the common utility @code{more}
1125 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1126 @key{RET} too many in this situation, @value{GDBN} disables command
1127 repetition after any command that generates this sort of display.
1131 Any text from a @kbd{#} to the end of the line is a comment; it does
1132 nothing. This is useful mainly in command files (@pxref{Command
1133 Files,,Command files}).
1136 @section Command completion
1139 @cindex word completion
1140 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1141 only one possibility; it can also show you what the valid possibilities
1142 are for the next word in a command, at any time. This works for @value{GDBN}
1143 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1145 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1146 of a word. If there is only one possibility, @value{GDBN} fills in the
1147 word, and waits for you to finish the command (or press @key{RET} to
1148 enter it). For example, if you type
1150 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1151 @c complete accuracy in these examples; space introduced for clarity.
1152 @c If texinfo enhancements make it unnecessary, it would be nice to
1153 @c replace " @key" by "@key" in the following...
1155 (@value{GDBP}) info bre @key{TAB}
1159 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1160 the only @code{info} subcommand beginning with @samp{bre}:
1163 (@value{GDBP}) info breakpoints
1167 You can either press @key{RET} at this point, to run the @code{info
1168 breakpoints} command, or backspace and enter something else, if
1169 @samp{breakpoints} does not look like the command you expected. (If you
1170 were sure you wanted @code{info breakpoints} in the first place, you
1171 might as well just type @key{RET} immediately after @samp{info bre},
1172 to exploit command abbreviations rather than command completion).
1174 If there is more than one possibility for the next word when you press
1175 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1176 characters and try again, or just press @key{TAB} a second time;
1177 @value{GDBN} displays all the possible completions for that word. For
1178 example, you might want to set a breakpoint on a subroutine whose name
1179 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1180 just sounds the bell. Typing @key{TAB} again displays all the
1181 function names in your program that begin with those characters, for
1185 (@value{GDBP}) b make_ @key{TAB}
1186 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1187 make_a_section_from_file make_environ
1188 make_abs_section make_function_type
1189 make_blockvector make_pointer_type
1190 make_cleanup make_reference_type
1191 make_command make_symbol_completion_list
1192 (@value{GDBP}) b make_
1196 After displaying the available possibilities, @value{GDBN} copies your
1197 partial input (@samp{b make_} in the example) so you can finish the
1200 If you just want to see the list of alternatives in the first place, you
1201 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1202 means @kbd{@key{META} ?}. You can type this
1204 either by holding down a
1205 key designated as the @key{META} shift on your keyboard (if there is
1206 one) while typing @kbd{?}, or
1208 as @key{ESC} followed by @kbd{?}.
1210 @cindex quotes in commands
1211 @cindex completion of quoted strings
1212 Sometimes the string you need, while logically a ``word'', may contain
1213 parentheses or other characters that @value{GDBN} normally excludes from its
1214 notion of a word. To permit word completion to work in this situation,
1215 you may enclose words in @code{'} (single quote marks) in @value{GDBN} commands.
1218 The most likely situation where you might need this is in typing the
1219 name of a C++ function. This is because C++ allows function overloading
1220 (multiple definitions of the same function, distinguished by argument
1221 type). For example, when you want to set a breakpoint you may need to
1222 distinguish whether you mean the version of @code{name} that takes an
1223 @code{int} parameter, @code{name(int)}, or the version that takes a
1224 @code{float} parameter, @code{name(float)}. To use the word-completion
1225 facilities in this situation, type a single quote @code{'} at the
1226 beginning of the function name. This alerts @value{GDBN} that it may need to
1227 consider more information than usual when you press @key{TAB} or
1228 @kbd{M-?} to request word completion:
1231 (@value{GDBP}) b 'bubble( @key{M-?}
1232 bubble(double,double) bubble(int,int)
1233 (@value{GDBP}) b 'bubble(
1236 In some cases, @value{GDBN} can tell that completing a name requires using
1237 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1238 completing as much as it can) if you do not type the quote in the first
1242 (@value{GDBP}) b bub @key{TAB}
1243 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1244 (@value{GDBP}) b 'bubble(
1248 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1249 you have not yet started typing the argument list when you ask for
1250 completion on an overloaded symbol.
1255 @section Getting help
1256 @cindex online documentation
1259 You can always ask @value{GDBN} itself for information on its commands, using the
1260 command @code{help}.
1266 You can use @code{help} (abbreviated @code{h}) with no arguments to
1267 display a short list of named classes of commands:
1271 List of classes of commands:
1273 running -- Running the program
1274 stack -- Examining the stack
1275 data -- Examining data
1276 breakpoints -- Making program stop at certain points
1277 files -- Specifying and examining files
1278 status -- Status inquiries
1279 support -- Support facilities
1280 user-defined -- User-defined commands
1281 aliases -- Aliases of other commands
1282 obscure -- Obscure features
1284 Type "help" followed by a class name for a list of
1285 commands in that class.
1286 Type "help" followed by command name for full
1288 Command name abbreviations are allowed if unambiguous.
1292 @item help @var{class}
1293 Using one of the general help classes as an argument, you can get a
1294 list of the individual commands in that class. For example, here is the
1295 help display for the class @code{status}:
1298 (@value{GDBP}) help status
1303 @c Line break in "show" line falsifies real output, but needed
1304 @c to fit in smallbook page size.
1305 show -- Generic command for showing things set
1307 info -- Generic command for printing status
1309 Type "help" followed by command name for full
1311 Command name abbreviations are allowed if unambiguous.
1315 @item help @var{command}
1316 With a command name as @code{help} argument, @value{GDBN} displays a
1317 short paragraph on how to use that command.
1320 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1321 and @code{show} to inquire about the state of your program, or the state
1322 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1323 manual introduces each of them in the appropriate context. The listings
1324 under @code{info} and under @code{show} in the Index point to
1325 all the sub-commands. @xref{Index}.
1332 This command (abbreviated @code{i}) is for describing the state of your
1333 program. For example, you can list the arguments given to your program
1334 with @code{info args}, list the registers currently in use with @code{info
1335 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1336 You can get a complete list of the @code{info} sub-commands with
1337 @w{@code{help info}}.
1341 In contrast, @code{show} is for describing the state of @value{GDBN} itself.
1342 You can change most of the things you can @code{show}, by using the
1343 related command @code{set}; for example, you can control what number
1344 system is used for displays with @code{set radix}, or simply inquire
1345 which is currently in use with @code{show radix}.
1348 To display all the settable parameters and their current
1349 values, you can use @code{show} with no arguments; you may also use
1350 @code{info set}. Both commands produce the same display.
1351 @c FIXME: "info set" violates the rule that "info" is for state of
1352 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1353 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1357 Here are three miscellaneous @code{show} subcommands, all of which are
1358 exceptional in lacking corresponding @code{set} commands:
1361 @kindex show version
1362 @cindex version number
1364 Show what version of @value{GDBN} is running. You should include this
1365 information in @value{GDBN} bug-reports. If multiple versions of @value{GDBN} are in
1366 use at your site, you may occasionally want to determine which version
1367 of @value{GDBN} you are running; as @value{GDBN} evolves, new commands are introduced,
1368 and old ones may wither away. The version number is also announced
1369 when you start @value{GDBN}.
1371 @kindex show copying
1373 Display information about permission for copying @value{GDBN}.
1375 @kindex show warranty
1377 Display the GNU ``NO WARRANTY'' statement.
1381 @chapter Running Programs Under @value{GDBN}
1383 When you run a program under @value{GDBN}, you must first generate
1384 debugging information when you compile it.
1386 You may start it with its arguments, if any, in an environment of your
1387 choice. You may redirect your program's input and output, debug an
1388 already running process, or kill a child process.
1392 * Compilation:: Compiling for debugging
1393 * Starting:: Starting your program
1395 * Arguments:: Your program's arguments
1396 * Environment:: Your program's environment
1397 * Working Directory:: Your program's working directory
1398 * Input/Output:: Your program's input and output
1399 * Attach:: Debugging an already-running process
1400 * Kill Process:: Killing the child process
1401 * Process Information:: Additional process information
1402 * Threads:: Debugging programs with multiple threads
1403 * Processes:: Debugging programs with multiple processes
1408 @section Compiling for debugging
1410 In order to debug a program effectively, you need to generate
1411 debugging information when you compile it. This debugging information
1412 is stored in the object file; it describes the data type of each
1413 variable or function and the correspondence between source line numbers
1414 and addresses in the executable code.
1416 To request debugging information, specify the @samp{-g} option when you run
1419 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1420 options together. Using those compilers, you cannot generate optimized
1421 executables containing debugging information.
1423 @value{NGCC}, the GNU C compiler, supports @samp{-g} with or without
1424 @samp{-O}, making it possible to debug optimized code. We recommend
1425 that you @emph{always} use @samp{-g} whenever you compile a program.
1426 You may think your program is correct, but there is no sense in pushing
1429 @cindex optimized code, debugging
1430 @cindex debugging optimized code
1431 When you debug a program compiled with @samp{-g -O}, remember that the
1432 optimizer is rearranging your code; the debugger shows you what is
1433 really there. Do not be too surprised when the execution path does not
1434 exactly match your source file! An extreme example: if you define a
1435 variable, but never use it, @value{GDBN} never sees that
1436 variable---because the compiler optimizes it out of existence.
1438 Some things do not work as well with @samp{-g -O} as with just
1439 @samp{-g}, particularly on machines with instruction scheduling. If in
1440 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1441 please report it as a bug (including a test case!).
1443 Older versions of the GNU C compiler permitted a variant option
1444 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1445 format; if your GNU C compiler has this option, do not use it.
1449 @section Starting your program
1457 Use the @code{run} command to start your program under @value{GDBN}. You must
1458 first specify the program name
1462 with an argument to @value{GDBN} (@pxref{Invocation, ,Getting In and
1463 Out of @value{GDBN}}), or by using the @code{file} or @code{exec-file}
1464 command (@pxref{Files, ,Commands to specify files}).
1469 If you are running your program in an execution environment that
1470 supports processes, @code{run} creates an inferior process and makes
1471 that process run your program. (In environments without processes,
1472 @code{run} jumps to the start of your program.)
1474 The execution of a program is affected by certain information it
1475 receives from its superior. @value{GDBN} provides ways to specify this
1476 information, which you must do @emph{before} starting your program. (You
1477 can change it after starting your program, but such changes only affect
1478 your program the next time you start it.) This information may be
1479 divided into four categories:
1482 @item The @emph{arguments.}
1483 Specify the arguments to give your program as the arguments of the
1484 @code{run} command. If a shell is available on your target, the shell
1485 is used to pass the arguments, so that you may use normal conventions
1486 (such as wildcard expansion or variable substitution) in describing
1487 the arguments. In Unix systems, you can control which shell is used
1488 with the @code{SHELL} environment variable. @xref{Arguments, ,Your
1489 program's arguments}.
1491 @item The @emph{environment.}
1492 Your program normally inherits its environment from @value{GDBN}, but you can
1493 use the @value{GDBN} commands @code{set environment} and @code{unset
1494 environment} to change parts of the environment that affect
1495 your program. @xref{Environment, ,Your program's environment}.
1497 @item The @emph{working directory.}
1498 Your program inherits its working directory from @value{GDBN}. You can set
1499 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1500 @xref{Working Directory, ,Your program's working directory}.
1502 @item The @emph{standard input and output.}
1503 Your program normally uses the same device for standard input and
1504 standard output as @value{GDBN} is using. You can redirect input and output
1505 in the @code{run} command line, or you can use the @code{tty} command to
1506 set a different device for your program.
1507 @xref{Input/Output, ,Your program's input and output}.
1510 @emph{Warning:} While input and output redirection work, you cannot use
1511 pipes to pass the output of the program you are debugging to another
1512 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1517 When you issue the @code{run} command, your program begins to execute
1518 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1519 of how to arrange for your program to stop. Once your program has
1520 stopped, you may call functions in your program, using the @code{print}
1521 or @code{call} commands. @xref{Data, ,Examining Data}.
1523 If the modification time of your symbol file has changed since the last
1524 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1525 table, and reads it again. When it does this, @value{GDBN} tries to retain
1526 your current breakpoints.
1530 @section Your program's arguments
1532 @cindex arguments (to your program)
1533 The arguments to your program can be specified by the arguments of the
1534 @code{run} command. They are passed to a shell, which expands wildcard
1535 characters and performs redirection of I/O, and thence to your program.
1536 Your @code{SHELL} environment variable (if it exists) specifies what
1537 shell @value{GDBN} uses. If you do not define @code{SHELL},
1538 @value{GDBN} uses @code{/bin/sh}.
1540 @code{run} with no arguments uses the same arguments used by the previous
1541 @code{run}, or those set by the @code{set args} command.
1546 Specify the arguments to be used the next time your program is run. If
1547 @code{set args} has no arguments, @code{run} executes your program
1548 with no arguments. Once you have run your program with arguments,
1549 using @code{set args} before the next @code{run} is the only way to run
1550 it again without arguments.
1554 Show the arguments to give your program when it is started.
1558 @section Your program's environment
1560 @cindex environment (of your program)
1561 The @dfn{environment} consists of a set of environment variables and
1562 their values. Environment variables conventionally record such things as
1563 your user name, your home directory, your terminal type, and your search
1564 path for programs to run. Usually you set up environment variables with
1565 the shell and they are inherited by all the other programs you run. When
1566 debugging, it can be useful to try running your program with a modified
1567 environment without having to start @value{GDBN} over again.
1570 @item path @var{directory}
1572 Add @var{directory} to the front of the @code{PATH} environment variable
1573 (the search path for executables), for both @value{GDBN} and your program.
1574 You may specify several directory names, separated by @samp{:} or
1575 whitespace. If @var{directory} is already in the path, it is moved to
1576 the front, so it is searched sooner.
1578 You can use the string @samp{$cwd} to refer to whatever is the current
1579 working directory at the time @value{GDBN} searches the path. If you
1580 use @samp{.} instead, it refers to the directory where you executed the
1581 @code{path} command. @value{GDBN} replaces @samp{.} in the
1582 @var{directory} argument (with the current path) before adding
1583 @var{directory} to the search path.
1584 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1585 @c document that, since repeating it would be a no-op.
1589 Display the list of search paths for executables (the @code{PATH}
1590 environment variable).
1592 @item show environment @r{[}@var{varname}@r{]}
1593 @kindex show environment
1594 Print the value of environment variable @var{varname} to be given to
1595 your program when it starts. If you do not supply @var{varname},
1596 print the names and values of all environment variables to be given to
1597 your program. You can abbreviate @code{environment} as @code{env}.
1599 @item set environment @var{varname} @r{[}=@r{]} @var{value}
1600 @kindex set environment
1601 Set environment variable @var{varname} to @var{value}. The value
1602 changes for your program only, not for @value{GDBN} itself. @var{value} may
1603 be any string; the values of environment variables are just strings, and
1604 any interpretation is supplied by your program itself. The @var{value}
1605 parameter is optional; if it is eliminated, the variable is set to a
1607 @c "any string" here does not include leading, trailing
1608 @c blanks. Gnu asks: does anyone care?
1610 For example, this command:
1617 tells a Unix program, when subsequently run, that its user is named
1618 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1619 are not actually required.)
1621 @item unset environment @var{varname}
1622 @kindex unset environment
1623 Remove variable @var{varname} from the environment to be passed to your
1624 program. This is different from @samp{set env @var{varname} =};
1625 @code{unset environment} removes the variable from the environment,
1626 rather than assigning it an empty value.
1629 @emph{Warning:} @value{GDBN} runs your program using the shell indicated
1630 by your @code{SHELL} environment variable if it exists (or
1631 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1632 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1633 @file{.bashrc} for BASH---any variables you set in that file affect
1634 your program. You may wish to move setting of environment variables to
1635 files that are only run when you sign on, such as @file{.login} or
1638 @node Working Directory
1639 @section Your program's working directory
1641 @cindex working directory (of your program)
1642 Each time you start your program with @code{run}, it inherits its
1643 working directory from the current working directory of @value{GDBN}.
1644 The @value{GDBN} working directory is initially whatever it inherited
1645 from its parent process (typically the shell), but you can specify a new
1646 working directory in @value{GDBN} with the @code{cd} command.
1648 The @value{GDBN} working directory also serves as a default for the commands
1649 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1653 @item cd @var{directory}
1655 Set the @value{GDBN} working directory to @var{directory}.
1659 Print the @value{GDBN} working directory.
1663 @section Your program's input and output
1668 By default, the program you run under @value{GDBN} does input and output to
1669 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal to
1670 its own terminal modes to interact with you, but it records the terminal
1671 modes your program was using and switches back to them when you continue
1672 running your program.
1676 @kindex info terminal
1677 Displays information recorded by @value{GDBN} about the terminal modes your
1681 You can redirect your program's input and/or output using shell
1682 redirection with the @code{run} command. For example,
1689 starts your program, diverting its output to the file @file{outfile}.
1692 @cindex controlling terminal
1693 Another way to specify where your program should do input and output is
1694 with the @code{tty} command. This command accepts a file name as
1695 argument, and causes this file to be the default for future @code{run}
1696 commands. It also resets the controlling terminal for the child
1697 process, for future @code{run} commands. For example,
1704 directs that processes started with subsequent @code{run} commands
1705 default to do input and output on the terminal @file{/dev/ttyb} and have
1706 that as their controlling terminal.
1708 An explicit redirection in @code{run} overrides the @code{tty} command's
1709 effect on the input/output device, but not its effect on the controlling
1712 When you use the @code{tty} command or redirect input in the @code{run}
1713 command, only the input @emph{for your program} is affected. The input
1714 for @value{GDBN} still comes from your terminal.
1717 @section Debugging an already-running process
1722 @item attach @var{process-id}
1723 This command attaches to a running process---one that was started
1724 outside @value{GDBN}. (@code{info files} shows your active
1725 targets.) The command takes as argument a process ID. The usual way to
1726 find out the process-id of a Unix process is with the @code{ps} utility,
1727 or with the @samp{jobs -l} shell command.
1729 @code{attach} does not repeat if you press @key{RET} a second time after
1730 executing the command.
1733 To use @code{attach}, your program must be running in an environment
1734 which supports processes; for example, @code{attach} does not work for
1735 programs on bare-board targets that lack an operating system. You must
1736 also have permission to send the process a signal.
1738 When using @code{attach}, you should first use the @code{file} command
1739 to specify the program running in the process and load its symbol table.
1740 @xref{Files, ,Commands to Specify Files}.
1742 The first thing @value{GDBN} does after arranging to debug the specified
1743 process is to stop it. You can examine and modify an attached process
1744 with all the @value{GDBN} commands that are ordinarily available when you start
1745 processes with @code{run}. You can insert breakpoints; you can step and
1746 continue; you can modify storage. If you would rather the process
1747 continue running, you may use the @code{continue} command after
1748 attaching @value{GDBN} to the process.
1753 When you have finished debugging the attached process, you can use the
1754 @code{detach} command to release it from @value{GDBN} control. Detaching
1755 the process continues its execution. After the @code{detach} command,
1756 that process and @value{GDBN} become completely independent once more, and you
1757 are ready to @code{attach} another process or start one with @code{run}.
1758 @code{detach} does not repeat if you press @key{RET} again after
1759 executing the command.
1762 If you exit @value{GDBN} or use the @code{run} command while you have an
1763 attached process, you kill that process. By default, @value{GDBN} asks
1764 for confirmation if you try to do either of these things; you can
1765 control whether or not you need to confirm by using the @code{set
1766 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
1771 @section Killing the child process
1776 Kill the child process in which your program is running under @value{GDBN}.
1779 This command is useful if you wish to debug a core dump instead of a
1780 running process. @value{GDBN} ignores any core dump file while your program
1784 On some operating systems, a program cannot be executed outside @value{GDBN}
1785 while you have breakpoints set on it inside @value{GDBN}. You can use the
1786 @code{kill} command in this situation to permit running your program
1787 outside the debugger.
1789 The @code{kill} command is also useful if you wish to recompile and
1790 relink your program, since on many systems it is impossible to modify an
1791 executable file while it is running in a process. In this case, when you
1792 next type @code{run}, @value{GDBN} notices that the file has changed, and
1793 reads the symbol table again (while trying to preserve your current
1794 breakpoint settings).
1796 @node Process Information
1797 @section Additional process information
1800 @cindex process image
1801 Some operating systems provide a facility called @samp{/proc} that can
1802 be used to examine the image of a running process using file-system
1803 subroutines. If @value{GDBN} is configured for an operating system with this
1804 facility, the command @code{info proc} is available to report on several
1805 kinds of information about the process running your program.
1810 Summarize available information about the process.
1812 @item info proc mappings
1813 @kindex info proc mappings
1814 Report on the address ranges accessible in the program, with information
1815 on whether your program may read, write, or execute each range.
1817 @item info proc times
1818 @kindex info proc times
1819 Starting time, user CPU time, and system CPU time for your program and
1823 @kindex info proc id
1824 Report on the process IDs related to your program: its own process ID,
1825 the ID of its parent, the process group ID, and the session ID.
1827 @item info proc status
1828 @kindex info proc status
1829 General information on the state of the process. If the process is
1830 stopped, this report includes the reason for stopping, and any signal
1834 Show all the above information about the process.
1838 @section Debugging programs with multiple threads
1840 @cindex threads of execution
1841 @cindex multiple threads
1842 @cindex switching threads
1843 In some operating systems, a single program may have more than one
1844 @dfn{thread} of execution. The precise semantics of threads differ from
1845 one operating system to another, but in general the threads of a single
1846 program are akin to multiple processes---except that they share one
1847 address space (that is, they can all examine and modify the same
1848 variables). On the other hand, each thread has its own registers and
1849 execution stack, and perhaps private memory.
1851 @value{GDBN} provides these facilities for debugging multi-thread
1855 @item automatic notification of new threads
1856 @item @samp{thread @var{threadno}}, a command to switch among threads
1857 @item @samp{info threads}, a command to inquire about existing threads
1858 @item thread-specific breakpoints
1862 @emph{Warning:} These facilities are not yet available on every
1863 @value{GDBN} configuration where the operating system supports threads.
1864 If your @value{GDBN} does not support threads, these commands have no
1865 effect. For example, a system without thread support shows no output
1866 from @samp{info threads}, and always rejects the @code{thread} command,
1870 (@value{GDBP}) info threads
1871 (@value{GDBP}) thread 1
1872 Thread ID 1 not known. Use the "info threads" command to
1873 see the IDs of currently known threads.
1875 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
1876 @c doesn't support threads"?
1879 @cindex focus of debugging
1880 @cindex current thread
1881 The @value{GDBN} thread debugging facility allows you to observe all
1882 threads while your program runs---but whenever @value{GDBN} takes
1883 control, one thread in particular is always the focus of debugging.
1884 This thread is called the @dfn{current thread}. Debugging commands show
1885 program information from the perspective of the current thread.
1887 @kindex New @var{systag}
1888 @cindex thread identifier (system)
1889 @c FIXME-implementors!! It would be more helpful if the [New...] message
1890 @c included GDB's numeric thread handle, so you could just go to that
1891 @c thread without first checking `info threads'.
1892 Whenever @value{GDBN} detects a new thread in your program, it displays
1893 the target system's identification for the thread with a message in the
1894 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
1895 whose form varies depending on the particular system. For example, on
1896 LynxOS, you might see
1899 [New process 35 thread 27]
1903 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
1904 the @var{systag} is simply something like @samp{process 368}, with no
1907 @c FIXME!! (1) Does the [New...] message appear even for the very first
1908 @c thread of a program, or does it only appear for the
1909 @c second---i.e., when it becomes obvious we have a multithread
1911 @c (2) *Is* there necessarily a first thread always? Or do some
1912 @c multithread systems permit starting a program with multiple
1913 @c threads ab initio?
1915 @cindex thread number
1916 @cindex thread identifier (GDB)
1917 For debugging purposes, @value{GDBN} associates its own thread
1918 number---always a single integer---with each thread in your program.
1922 @kindex info threads
1923 Display a summary of all threads currently in your
1924 program. @value{GDBN} displays for each thread (in this order):
1927 @item the thread number assigned by @value{GDBN}
1929 @item the target system's thread identifier (@var{systag})
1931 @item the current stack frame summary for that thread
1935 An asterisk @samp{*} to the left of the @value{GDBN} thread number
1936 indicates the current thread.
1940 @c end table here to get a little more width for example
1943 (@value{GDBP}) info threads
1944 3 process 35 thread 27 0x34e5 in sigpause ()
1945 2 process 35 thread 23 0x34e5 in sigpause ()
1946 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
1951 @item thread @var{threadno}
1952 @kindex thread @var{threadno}
1953 Make thread number @var{threadno} the current thread. The command
1954 argument @var{threadno} is the internal @value{GDBN} thread number, as
1955 shown in the first field of the @samp{info threads} display.
1956 @value{GDBN} responds by displaying the system identifier of the thread
1957 you selected, and its current stack frame summary:
1960 @c FIXME!! This example made up; find a GDB w/threads and get real one
1961 (@value{GDBP}) thread 2
1962 [Switching to process 35 thread 23]
1963 0x34e5 in sigpause ()
1967 As with the @samp{[New @dots{}]} message, the form of the text after
1968 @samp{Switching to} depends on your system's conventions for identifying
1972 @cindex automatic thread selection
1973 @cindex switching threads automatically
1974 @cindex threads, automatic switching
1975 Whenever @value{GDBN} stops your program, due to a breakpoint or a
1976 signal, it automatically selects the thread where that breakpoint or
1977 signal happened. @value{GDBN} alerts you to the context switch with a
1978 message of the form @samp{[Switching to @var{systag}]} to identify the
1981 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
1982 more information about how @value{GDBN} behaves when you stop and start
1983 programs with multiple threads.
1985 @xref{Set Watchpoints,,Setting watchpoints}, for information about
1986 watchpoints in programs with multiple threads.
1990 @section Debugging programs with multiple processes
1992 @cindex fork, debugging programs which call
1993 @cindex multiple processes
1994 @cindex processes, multiple
1995 @value{GDBN} has no special support for debugging programs which create
1996 additional processes using the @code{fork} function. When a program
1997 forks, @value{GDBN} will continue to debug the parent process and the
1998 child process will run unimpeded. If you have set a breakpoint in any
1999 code which the child then executes, the child will get a @code{SIGTRAP}
2000 signal which (unless it catches the signal) will cause it to terminate.
2002 However, if you want to debug the child process there is a workaround
2003 which isn't too painful. Put a call to @code{sleep} in the code which
2004 the child process executes after the fork. It may be useful to sleep
2005 only if a certain environment variable is set, or a certain file exists,
2006 so that the delay need not occur when you don't want to run @value{GDBN}
2007 on the child. While the child is sleeping, use the @code{ps} program to
2008 get its process ID. Then tell @value{GDBN} (a new invocation of
2009 @value{GDBN} if you are also debugging the parent process) to attach to
2010 the child process (see @ref{Attach}). From that point on you can debug
2011 the child process just like any other process which you attached to.
2014 @chapter Stopping and Continuing
2016 The principal purposes of using a debugger are so that you can stop your
2017 program before it terminates; or so that, if your program runs into
2018 trouble, you can investigate and find out why.
2020 Inside @value{GDBN}, your program may stop for any of several reasons, such
2025 a breakpoint, or reaching a new line after a @value{GDBN}
2026 command such as @code{step}. You may then examine and change
2027 variables, set new breakpoints or remove old ones, and then continue
2028 execution. Usually, the messages shown by @value{GDBN} provide ample
2029 explanation of the status of your program---but you can also explicitly
2030 request this information at any time.
2034 @kindex info program
2035 Display information about the status of your program: whether it is
2045 * Breakpoints:: Breakpoints, watchpoints, and exceptions
2048 * Breakpoints:: Breakpoints and watchpoints
2050 @c Remnant makeinfo bug requires blank line after *successful* end-if in menu:
2052 * Continuing and Stepping:: Resuming execution
2057 * Thread Stops:: Stopping and starting multi-thread programs
2061 @c makeinfo node-defaulting requires adjacency of @node and sectioning cmds
2062 @c ...hence distribute @node Breakpoints over two possible @if expansions.
2066 @section Breakpoints, watchpoints, and exceptions
2070 @section Breakpoints and watchpoints
2074 A @dfn{breakpoint} makes your program stop whenever a certain point in
2075 the program is reached. For each breakpoint, you can add
2076 conditions to control in finer detail whether your program stops.
2077 You can set breakpoints with the @code{break} command and its variants
2078 (@pxref{Set Breaks, ,Setting breakpoints}), to specify the place where
2079 your program should stop by line number, function name or exact address
2082 In languages with exception handling (such as GNU C++), you can also set
2083 breakpoints where an exception is raised (@pxref{Exception Handling,,
2084 Breakpoints and exceptions}).
2088 @cindex memory tracing
2089 @cindex breakpoint on memory address
2090 @cindex breakpoint on variable modification
2091 A @dfn{watchpoint} is a special breakpoint that stops your program
2092 when the value of an expression changes. You must use a different
2093 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2094 watchpoints}), but aside from that, you can manage a watchpoint like
2095 any other breakpoint: you enable, disable, and delete both breakpoints
2096 and watchpoints using the same commands.
2098 You can arrange to have values from your program displayed automatically
2099 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2102 @cindex breakpoint numbers
2103 @cindex numbers for breakpoints
2104 @value{GDBN} assigns a number to each breakpoint or watchpoint when you
2105 create it; these numbers are successive integers starting with one. In
2106 many of the commands for controlling various features of breakpoints you
2107 use the breakpoint number to say which breakpoint you want to change.
2108 Each breakpoint may be @dfn{enabled} or @dfn{disabled}; if disabled, it has
2109 no effect on your program until you enable it again.
2112 * Set Breaks:: Setting breakpoints
2113 * Set Watchpoints:: Setting watchpoints
2115 * Exception Handling:: Breakpoints and exceptions
2118 * Delete Breaks:: Deleting breakpoints
2119 * Disabling:: Disabling breakpoints
2120 * Conditions:: Break conditions
2121 * Break Commands:: Breakpoint command lists
2123 * Breakpoint Menus:: Breakpoint menus
2126 * Error in Breakpoints:: ``Cannot insert breakpoints''
2131 @subsection Setting breakpoints
2133 @c FIXME LMB what does GDB do if no code on line of breakpt?
2134 @c consider in particular declaration with/without initialization.
2136 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2141 @cindex latest breakpoint
2142 Breakpoints are set with the @code{break} command (abbreviated
2143 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2144 number of the beakpoint you've set most recently; see @ref{Convenience
2145 Vars,, Convenience variables}, for a discussion of what you can do with
2146 convenience variables.
2148 You have several ways to say where the breakpoint should go.
2151 @item break @var{function}
2152 Set a breakpoint at entry to function @var{function}.
2154 When using source languages that permit overloading of symbols, such as
2155 C++, @var{function} may refer to more than one possible place to break.
2156 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2159 @item break +@var{offset}
2160 @itemx break -@var{offset}
2161 Set a breakpoint some number of lines forward or back from the position
2162 at which execution stopped in the currently selected frame.
2164 @item break @var{linenum}
2165 Set a breakpoint at line @var{linenum} in the current source file.
2166 That file is the last file whose source text was printed. This
2167 breakpoint stops your program just before it executes any of the
2170 @item break @var{filename}:@var{linenum}
2171 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2173 @item break @var{filename}:@var{function}
2174 Set a breakpoint at entry to function @var{function} found in file
2175 @var{filename}. Specifying a file name as well as a function name is
2176 superfluous except when multiple files contain similarly named
2179 @item break *@var{address}
2180 Set a breakpoint at address @var{address}. You can use this to set
2181 breakpoints in parts of your program which do not have debugging
2182 information or source files.
2185 When called without any arguments, @code{break} sets a breakpoint at
2186 the next instruction to be executed in the selected stack frame
2187 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2188 innermost, this makes your program stop as soon as control
2189 returns to that frame. This is similar to the effect of a
2190 @code{finish} command in the frame inside the selected frame---except
2191 that @code{finish} does not leave an active breakpoint. If you use
2192 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2193 the next time it reaches the current location; this may be useful
2196 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2197 least one instruction has been executed. If it did not do this, you
2198 would be unable to proceed past a breakpoint without first disabling the
2199 breakpoint. This rule applies whether or not the breakpoint already
2200 existed when your program stopped.
2202 @item break @dots{} if @var{cond}
2203 Set a breakpoint with condition @var{cond}; evaluate the expression
2204 @var{cond} each time the breakpoint is reached, and stop only if the
2205 value is nonzero---that is, if @var{cond} evaluates as true.
2206 @samp{@dots{}} stands for one of the possible arguments described
2207 above (or no argument) specifying where to break. @xref{Conditions,
2208 ,Break conditions}, for more information on breakpoint conditions.
2210 @item tbreak @var{args}
2212 Set a breakpoint enabled only for one stop. @var{args} are the
2213 same as for the @code{break} command, and the breakpoint is set in the same
2214 way, but the breakpoint is automatically deleted after the first time your
2215 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2217 @item rbreak @var{regex}
2219 @cindex regular expression
2220 @c FIXME what kind of regexp?
2221 Set breakpoints on all functions matching the regular expression
2222 @var{regex}. This command
2223 sets an unconditional breakpoint on all matches, printing a list of all
2224 breakpoints it set. Once these breakpoints are set, they are treated
2225 just like the breakpoints set with the @code{break} command. You can
2226 delete them, disable them, or make them conditional the same way as any
2230 When debugging C++ programs, @code{rbreak} is useful for setting
2231 breakpoints on overloaded functions that are not members of any special
2235 @kindex info breakpoints
2236 @cindex @code{$_} and @code{info breakpoints}
2237 @item info breakpoints @r{[}@var{n}@r{]}
2238 @itemx info break @r{[}@var{n}@r{]}
2239 @itemx info watchpoints @r{[}@var{n}@r{]}
2240 Print a table of all breakpoints and watchpoints set and not
2241 deleted, with the following columns for each breakpoint:
2244 @item Breakpoint Numbers
2246 Breakpoint or watchpoint.
2248 Whether the breakpoint is marked to be disabled or deleted when hit.
2249 @item Enabled or Disabled
2250 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2251 that are not enabled.
2253 Where the breakpoint is in your program, as a memory address
2255 Where the breakpoint is in the source for your program, as a file and
2260 If a breakpoint is conditional, @code{info break} shows the condition on
2261 the line following the affected breakpoint; breakpoint commands, if any,
2262 are listed after that.
2265 @code{info break} with a breakpoint
2266 number @var{n} as argument lists only that breakpoint. The
2267 convenience variable @code{$_} and the default examining-address for
2268 the @code{x} command are set to the address of the last breakpoint
2269 listed (@pxref{Memory, ,Examining memory}).
2272 @value{GDBN} allows you to set any number of breakpoints at the same place in
2273 your program. There is nothing silly or meaningless about this. When
2274 the breakpoints are conditional, this is even useful
2275 (@pxref{Conditions, ,Break conditions}).
2277 @cindex negative breakpoint numbers
2278 @cindex internal @value{GDBN} breakpoints
2279 @value{GDBN} itself sometimes sets breakpoints in your program for special
2280 purposes, such as proper handling of @code{longjmp} (in C programs).
2281 These internal breakpoints are assigned negative numbers, starting with
2282 @code{-1}; @samp{info breakpoints} does not display them.
2284 You can see these breakpoints with the @value{GDBN} maintenance command
2285 @samp{maint info breakpoints}.
2288 @kindex maint info breakpoints
2289 @item maint info breakpoints
2290 Using the same format as @samp{info breakpoints}, display both the
2291 breakpoints you've set explicitly, and those @value{GDBN} is using for
2292 internal purposes. Internal breakpoints are shown with negative
2293 breakpoint numbers. The type column identifies what kind of breakpoint
2298 Normal, explicitly set breakpoint.
2301 Normal, explicitly set watchpoint.
2304 Internal breakpoint, used to handle correctly stepping through
2305 @code{longjmp} calls.
2307 @item longjmp resume
2308 Internal breakpoint at the target of a @code{longjmp}.
2311 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2314 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2320 @node Set Watchpoints
2321 @subsection Setting watchpoints
2322 @cindex setting watchpoints
2324 You can use a watchpoint to stop execution whenever the value of an
2325 expression changes, without having to predict a particular place
2326 where this may happen.
2328 Watchpoints currently execute two orders of magnitude more slowly than
2329 other breakpoints, but this can be well worth it to catch errors where
2330 you have no clue what part of your program is the culprit.
2333 @c this "future releases" promise has been in too long, is getting
2334 @c embarrassing. But...
2335 @c FIXME: in future updates, check whether hardware watchpoints in on any
2336 @c platforms yet. As of 26jan94, they're very close on HPPA running
2337 @c Berkeley and on Irix 4.
2338 Some processors provide special hardware to support watchpoint
2339 evaluation; future releases of @value{GDBN} will use such hardware if it
2345 @item watch @var{expr}
2346 Set a watchpoint for an expression.
2348 @kindex info watchpoints
2349 @item info watchpoints
2350 This command prints a list of watchpoints and breakpoints; it is the
2351 same as @code{info break}.
2356 @cindex watchpoints and threads
2357 @cindex threads and watchpoints
2358 @emph{Warning:} in multi-thread programs, watchpoints have only limited
2359 usefulness. With the current watchpoint implementation, @value{GDBN}
2360 can only watch the value of an expression @emph{in a single thread}. If
2361 you are confident that the expression can only change due to the current
2362 thread's activity (and if you are also confident that no other thread
2363 can become current), then you can use watchpoints as usual. However,
2364 @value{GDBN} may not notice when a non-current thread's activity changes
2370 @node Exception Handling
2371 @subsection Breakpoints and exceptions
2372 @cindex exception handlers
2374 Some languages, such as GNU C++, implement exception handling. You can
2375 use @value{GDBN} to examine what caused your program to raise an exception,
2376 and to list the exceptions your program is prepared to handle at a
2377 given point in time.
2380 @item catch @var{exceptions}
2382 You can set breakpoints at active exception handlers by using the
2383 @code{catch} command. @var{exceptions} is a list of names of exceptions
2387 You can use @code{info catch} to list active exception handlers.
2388 @xref{Frame Info, ,Information about a frame}.
2390 There are currently some limitations to exception handling in @value{GDBN}:
2394 If you call a function interactively, @value{GDBN} normally returns
2395 control to you when the function has finished executing. If the call
2396 raises an exception, however, the call may bypass the mechanism that
2397 returns control to you and cause your program to simply continue
2398 running until it hits a breakpoint, catches a signal that @value{GDBN} is
2399 listening for, or exits.
2402 You cannot raise an exception interactively.
2405 You cannot install an exception handler interactively.
2408 @cindex raise exceptions
2409 Sometimes @code{catch} is not the best way to debug exception handling:
2410 if you need to know exactly where an exception is raised, it is better to
2411 stop @emph{before} the exception handler is called, since that way you
2412 can see the stack before any unwinding takes place. If you set a
2413 breakpoint in an exception handler instead, it may not be easy to find
2414 out where the exception was raised.
2416 To stop just before an exception handler is called, you need some
2417 knowledge of the implementation. In the case of GNU C++, exceptions are
2418 raised by calling a library function named @code{__raise_exception}
2419 which has the following ANSI C interface:
2422 /* @var{addr} is where the exception identifier is stored.
2423 ID is the exception identifier. */
2424 void __raise_exception (void **@var{addr}, void *@var{id});
2428 To make the debugger catch all exceptions before any stack
2429 unwinding takes place, set a breakpoint on @code{__raise_exception}
2430 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2432 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2433 that depends on the value of @var{id}, you can stop your program when
2434 a specific exception is raised. You can use multiple conditional
2435 breakpoints to stop your program when any of a number of exceptions are
2440 @subsection Deleting breakpoints
2442 @cindex clearing breakpoints, watchpoints
2443 @cindex deleting breakpoints, watchpoints
2444 It is often necessary to eliminate a breakpoint or watchpoint once it
2445 has done its job and you no longer want your program to stop there. This
2446 is called @dfn{deleting} the breakpoint. A breakpoint that has been
2447 deleted no longer exists; it is forgotten.
2449 With the @code{clear} command you can delete breakpoints according to
2450 where they are in your program. With the @code{delete} command you can
2451 delete individual breakpoints or watchpoints by specifying their
2454 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2455 automatically ignores breakpoints on the first instruction to be executed
2456 when you continue execution without changing the execution address.
2461 Delete any breakpoints at the next instruction to be executed in the
2462 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2463 the innermost frame is selected, this is a good way to delete a
2464 breakpoint where your program just stopped.
2466 @item clear @var{function}
2467 @itemx clear @var{filename}:@var{function}
2468 Delete any breakpoints set at entry to the function @var{function}.
2470 @item clear @var{linenum}
2471 @itemx clear @var{filename}:@var{linenum}
2472 Delete any breakpoints set at or within the code of the specified line.
2474 @item delete @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2475 @cindex delete breakpoints
2478 Delete the breakpoints or watchpoints of the numbers specified as
2479 arguments. If no argument is specified, delete all breakpoints (@value{GDBN}
2480 asks confirmation, unless you have @code{set confirm off}). You
2481 can abbreviate this command as @code{d}.
2485 @subsection Disabling breakpoints
2487 @cindex disabled breakpoints
2488 @cindex enabled breakpoints
2489 Rather than deleting a breakpoint or watchpoint, you might prefer to
2490 @dfn{disable} it. This makes the breakpoint inoperative as if it had
2491 been deleted, but remembers the information on the breakpoint so that
2492 you can @dfn{enable} it again later.
2494 You disable and enable breakpoints and watchpoints with the
2495 @code{enable} and @code{disable} commands, optionally specifying one or
2496 more breakpoint numbers as arguments. Use @code{info break} or
2497 @code{info watch} to print a list of breakpoints or watchpoints if you
2498 do not know which numbers to use.
2500 A breakpoint or watchpoint can have any of four different states of
2505 Enabled. The breakpoint stops your program. A breakpoint set
2506 with the @code{break} command starts out in this state.
2508 Disabled. The breakpoint has no effect on your program.
2510 Enabled once. The breakpoint stops your program, but then becomes
2511 disabled. A breakpoint set with the @code{tbreak} command starts out in
2514 Enabled for deletion. The breakpoint stops your program, but
2515 immediately after it does so it is deleted permanently.
2518 You can use the following commands to enable or disable breakpoints and
2522 @item disable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2523 @kindex disable breakpoints
2526 Disable the specified breakpoints---or all breakpoints, if none are
2527 listed. A disabled breakpoint has no effect but is not forgotten. All
2528 options such as ignore-counts, conditions and commands are remembered in
2529 case the breakpoint is enabled again later. You may abbreviate
2530 @code{disable} as @code{dis}.
2532 @item enable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2533 @kindex enable breakpoints
2535 Enable the specified breakpoints (or all defined breakpoints). They
2536 become effective once again in stopping your program.
2538 @item enable @r{[}breakpoints@r{]} once @var{bnums}@dots{}
2539 Enable the specified breakpoints temporarily. @value{GDBN} disables any
2540 of these breakpoints immediately after stopping your program.
2542 @item enable @r{[}breakpoints@r{]} delete @var{bnums}@dots{}
2543 Enable the specified breakpoints to work once, then die. @value{GDBN}
2544 deletes any of these breakpoints as soon as your program stops there.
2547 Save for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2548 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2549 subsequently, they become disabled or enabled only when you use one of
2550 the commands above. (The command @code{until} can set and delete a
2551 breakpoint of its own, but it does not change the state of your other
2552 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2556 @subsection Break conditions
2557 @cindex conditional breakpoints
2558 @cindex breakpoint conditions
2560 @c FIXME what is scope of break condition expr? Context where wanted?
2561 @c in particular for a watchpoint?
2562 The simplest sort of breakpoint breaks every time your program reaches a
2563 specified place. You can also specify a @dfn{condition} for a
2564 breakpoint. A condition is just a Boolean expression in your
2565 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2566 a condition evaluates the expression each time your program reaches it,
2567 and your program stops only if the condition is @emph{true}.
2569 This is the converse of using assertions for program validation; in that
2570 situation, you want to stop when the assertion is violated---that is,
2571 when the condition is false. In C, if you want to test an assertion expressed
2572 by the condition @var{assert}, you should set the condition
2573 @samp{! @var{assert}} on the appropriate breakpoint.
2575 Conditions are also accepted for watchpoints; you may not need them,
2576 since a watchpoint is inspecting the value of an expression anyhow---but
2577 it might be simpler, say, to just set a watchpoint on a variable name,
2578 and specify a condition that tests whether the new value is an interesting
2581 Break conditions can have side effects, and may even call functions in
2582 your program. This can be useful, for example, to activate functions
2583 that log program progress, or to use your own print functions to
2584 format special data structures. The effects are completely predictable
2585 unless there is another enabled breakpoint at the same address. (In
2586 that case, @value{GDBN} might see the other breakpoint first and stop your
2587 program without checking the condition of this one.) Note that
2588 breakpoint commands are usually more convenient and flexible for the
2589 purpose of performing side effects when a breakpoint is reached
2590 (@pxref{Break Commands, ,Breakpoint command lists}).
2592 Break conditions can be specified when a breakpoint is set, by using
2593 @samp{if} in the arguments to the @code{break} command. @xref{Set
2594 Breaks, ,Setting breakpoints}. They can also be changed at any time
2595 with the @code{condition} command. The @code{watch} command does not
2596 recognize the @code{if} keyword; @code{condition} is the only way to
2597 impose a further condition on a watchpoint.
2600 @item condition @var{bnum} @var{expression}
2602 Specify @var{expression} as the break condition for breakpoint or
2603 watchpoint number @var{bnum}. After you set a condition, breakpoint
2604 @var{bnum} stops your program only if the value of @var{expression} is
2605 true (nonzero, in C). When you use @code{condition}, @value{GDBN}
2606 checks @var{expression} immediately for syntactic correctness, and to
2607 determine whether symbols in it have referents in the context of your
2609 @c FIXME so what does GDB do if there is no referent? Moreover, what
2610 @c about watchpoints?
2612 not actually evaluate @var{expression} at the time the @code{condition}
2613 command is given, however. @xref{Expressions, ,Expressions}.
2615 @item condition @var{bnum}
2616 Remove the condition from breakpoint number @var{bnum}. It becomes
2617 an ordinary unconditional breakpoint.
2620 @cindex ignore count (of breakpoint)
2621 A special case of a breakpoint condition is to stop only when the
2622 breakpoint has been reached a certain number of times. This is so
2623 useful that there is a special way to do it, using the @dfn{ignore
2624 count} of the breakpoint. Every breakpoint has an ignore count, which
2625 is an integer. Most of the time, the ignore count is zero, and
2626 therefore has no effect. But if your program reaches a breakpoint whose
2627 ignore count is positive, then instead of stopping, it just decrements
2628 the ignore count by one and continues. As a result, if the ignore count
2629 value is @var{n}, the breakpoint does not stop the next @var{n} times
2630 your program reaches it.
2633 @item ignore @var{bnum} @var{count}
2635 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
2636 The next @var{count} times the breakpoint is reached, your program's
2637 execution does not stop; other than to decrement the ignore count, @value{GDBN}
2640 To make the breakpoint stop the next time it is reached, specify
2643 When you use @code{continue} to resume execution of your program from a
2644 breakpoint, you can specify an ignore count directly as an argument to
2645 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
2646 Stepping,,Continuing and stepping}.
2648 If a breakpoint has a positive ignore count and a condition, the
2649 condition is not checked. Once the ignore count reaches zero,
2650 @value{GDBN} resumes checking the condition.
2652 You could achieve the effect of the ignore count with a condition such
2653 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
2654 is decremented each time. @xref{Convenience Vars, ,Convenience
2658 @node Break Commands
2659 @subsection Breakpoint command lists
2661 @cindex breakpoint commands
2662 You can give any breakpoint (or watchpoint) a series of commands to
2663 execute when your program stops due to that breakpoint. For example, you
2664 might want to print the values of certain expressions, or enable other
2668 @item commands @r{[}@var{bnum}@r{]}
2669 @itemx @dots{} @var{command-list} @dots{}
2673 Specify a list of commands for breakpoint number @var{bnum}. The commands
2674 themselves appear on the following lines. Type a line containing just
2675 @code{end} to terminate the commands.
2677 To remove all commands from a breakpoint, type @code{commands} and
2678 follow it immediately with @code{end}; that is, give no commands.
2680 With no @var{bnum} argument, @code{commands} refers to the last
2681 breakpoint or watchpoint set (not to the breakpoint most recently
2685 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
2686 disabled within a @var{command-list}.
2688 You can use breakpoint commands to start your program up again. Simply
2689 use the @code{continue} command, or @code{step}, or any other command
2690 that resumes execution.
2692 Any other commands in the command list, after a command that resumes
2693 execution, are ignored. This is because any time you resume execution
2694 (even with a simple @code{next} or @code{step}), you may encounter
2695 another breakpoint---which could have its own command list, leading to
2696 ambiguities about which list to execute.
2699 If the first command you specify in a command list is @code{silent}, the
2700 usual message about stopping at a breakpoint is not printed. This may
2701 be desirable for breakpoints that are to print a specific message and
2702 then continue. If none of the remaining commands print anything, you
2703 see no sign that the breakpoint was reached. @code{silent} is
2704 meaningful only at the beginning of a breakpoint command list.
2706 The commands @code{echo}, @code{output}, and @code{printf} allow you to
2707 print precisely controlled output, and are often useful in silent
2708 breakpoints. @xref{Output, ,Commands for controlled output}.
2710 For example, here is how you could use breakpoint commands to print the
2711 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
2717 printf "x is %d\n",x
2722 One application for breakpoint commands is to compensate for one bug so
2723 you can test for another. Put a breakpoint just after the erroneous line
2724 of code, give it a condition to detect the case in which something
2725 erroneous has been done, and give it commands to assign correct values
2726 to any variables that need them. End with the @code{continue} command
2727 so that your program does not stop, and start with the @code{silent}
2728 command so that no output is produced. Here is an example:
2740 @node Breakpoint Menus
2741 @subsection Breakpoint menus
2743 @cindex symbol overloading
2745 Some programming languages (notably C++) permit a single function name
2746 to be defined several times, for application in different contexts.
2747 This is called @dfn{overloading}. When a function name is overloaded,
2748 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
2749 a breakpoint. If you realize this is a problem, you can use
2750 something like @samp{break @var{function}(@var{types})} to specify which
2751 particular version of the function you want. Otherwise, @value{GDBN} offers
2752 you a menu of numbered choices for different possible breakpoints, and
2753 waits for your selection with the prompt @samp{>}. The first two
2754 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
2755 sets a breakpoint at each definition of @var{function}, and typing
2756 @kbd{0} aborts the @code{break} command without setting any new
2759 For example, the following session excerpt shows an attempt to set a
2760 breakpoint at the overloaded symbol @code{String::after}.
2761 We choose three particular definitions of that function name:
2763 @c FIXME! This is likely to change to show arg type lists, at least
2765 (@value{GDBP}) b String::after
2768 [2] file:String.cc; line number:867
2769 [3] file:String.cc; line number:860
2770 [4] file:String.cc; line number:875
2771 [5] file:String.cc; line number:853
2772 [6] file:String.cc; line number:846
2773 [7] file:String.cc; line number:735
2775 Breakpoint 1 at 0xb26c: file String.cc, line 867.
2776 Breakpoint 2 at 0xb344: file String.cc, line 875.
2777 Breakpoint 3 at 0xafcc: file String.cc, line 846.
2778 Multiple breakpoints were set.
2779 Use the "delete" command to delete unwanted
2786 @node Error in Breakpoints
2787 @subsection ``Cannot insert breakpoints''
2789 @c FIXME: "cannot insert breakpoints" error, v unclear.
2790 @c Q in pending mail to Gilmore. ---pesch@cygnus.com, 26mar91
2791 @c some light may be shed by looking at instances of
2792 @c ONE_PROCESS_WRITETEXT. But error message seems possible otherwise
2793 @c too. pesch, 20sep91
2794 Under some operating systems, breakpoints cannot be used in a program if
2795 any other process is running that program. In this situation,
2796 attempting to run or continue a program with a breakpoint causes @value{GDBN}
2797 to stop the other process.
2799 When this happens, you have three ways to proceed:
2803 Remove or disable the breakpoints, then continue.
2806 Suspend @value{GDBN}, and copy the file containing your program to a new name.
2807 Resume @value{GDBN} and use the @code{exec-file} command to specify that @value{GDBN}
2808 should run your program under that name. Then start your program again.
2810 @c FIXME: RMS commented here "Show example". Maybe when someone
2811 @c explains the first FIXME: in this section...
2814 Relink your program so that the text segment is nonsharable, using the
2815 linker option @samp{-N}. The operating system limitation may not apply
2816 to nonsharable executables.
2820 @node Continuing and Stepping
2821 @section Continuing and stepping
2825 @cindex resuming execution
2826 @dfn{Continuing} means resuming program execution until your program
2827 completes normally. In contrast, @dfn{stepping} means executing just
2828 one more ``step'' of your program, where ``step'' may mean either one
2829 line of source code, or one machine instruction (depending on what
2830 particular command you use). Either when continuing
2831 or when stepping, your program may stop even sooner, due to
2836 a breakpoint or a signal. (If due to a signal, you may want to use
2837 @code{handle}, or use @samp{signal 0} to resume execution.
2838 @xref{Signals, ,Signals}.)
2842 @item continue @r{[}@var{ignore-count}@r{]}
2843 @itemx c @r{[}@var{ignore-count}@r{]}
2844 @itemx fg @r{[}@var{ignore-count}@r{]}
2848 Resume program execution, at the address where your program last stopped;
2849 any breakpoints set at that address are bypassed. The optional argument
2850 @var{ignore-count} allows you to specify a further number of times to
2851 ignore a breakpoint at this location; its effect is like that of
2852 @code{ignore} (@pxref{Conditions, ,Break conditions}).
2854 The argument @var{ignore-count} is meaningful only when your program
2855 stopped due to a breakpoint. At other times, the argument to
2856 @code{continue} is ignored.
2858 The synonyms @code{c} and @code{fg} are provided purely for convenience,
2859 and have exactly the same behavior as @code{continue}.
2862 To resume execution at a different place, you can use @code{return}
2863 (@pxref{Returning, ,Returning from a function}) to go back to the
2864 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
2865 different address}) to go to an arbitrary location in your program.
2867 A typical technique for using stepping is to set a breakpoint
2869 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions})
2872 (@pxref{Breakpoints, ,Breakpoints and watchpoints})
2875 beginning of the function or the section of your program where a
2876 problem is believed to lie, run your program until it stops at that
2877 breakpoint, and then step through the suspect area, examining the
2878 variables that are interesting, until you see the problem happen.
2884 Continue running your program until control reaches a different source
2885 line, then stop it and return control to @value{GDBN}. This command is
2886 abbreviated @code{s}.
2889 @c "without debugging information" is imprecise; actually "without line
2890 @c numbers in the debugging information". (gcc -g1 has debugging info but
2891 @c not line numbers). But it seems complex to try to make that
2892 @c distinction here.
2893 @emph{Warning:} If you use the @code{step} command while control is
2894 within a function that was compiled without debugging information,
2895 execution proceeds until control reaches a function that does have
2896 debugging information. Likewise, it will not step into a function which
2897 is compiled without debugging information. To step through functions
2898 without debugging information, use the @code{stepi} command, described
2902 @item step @var{count}
2903 Continue running as in @code{step}, but do so @var{count} times. If a
2904 breakpoint is reached,
2906 or a signal not related to stepping occurs before @var{count} steps,
2908 stepping stops right away.
2910 @item next @r{[}@var{count}@r{]}
2913 Continue to the next source line in the current (innermost) stack frame.
2914 Similar to @code{step}, but any function calls appearing within the line
2915 of code are executed without stopping. Execution stops when control
2916 reaches a different line of code at the stack level which was executing
2917 when the @code{next} command was given. This command is abbreviated
2920 An argument @var{count} is a repeat count, as for @code{step}.
2922 @code{next} within a function that lacks debugging information acts like
2923 @code{step}, but any function calls appearing within the code of the
2924 function are executed without stopping.
2928 Continue running until just after function in the selected stack frame
2929 returns. Print the returned value (if any).
2931 Contrast this with the @code{return} command (@pxref{Returning,
2932 ,Returning from a function}).
2938 Continue running until a source line past the current line, in the
2939 current stack frame, is reached. This command is used to avoid single
2940 stepping through a loop more than once. It is like the @code{next}
2941 command, except that when @code{until} encounters a jump, it
2942 automatically continues execution until the program counter is greater
2943 than the address of the jump.
2945 This means that when you reach the end of a loop after single stepping
2946 though it, @code{until} makes your program continue execution until it
2947 exits the loop. In contrast, a @code{next} command at the end of a loop
2948 simply steps back to the beginning of the loop, which forces you to step
2949 through the next iteration.
2951 @code{until} always stops your program if it attempts to exit the current
2954 @code{until} may produce somewhat counterintuitive results if the order
2955 of machine code does not match the order of the source lines. For
2956 example, in the following excerpt from a debugging session, the @code{f}
2957 (@code{frame}) command shows that execution is stopped at line
2958 @code{206}; yet when we use @code{until}, we get to line @code{195}:
2962 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
2964 (@value{GDBP}) until
2965 195 for ( ; argc > 0; NEXTARG) @{
2968 This happened because, for execution efficiency, the compiler had
2969 generated code for the loop closure test at the end, rather than the
2970 start, of the loop---even though the test in a C @code{for}-loop is
2971 written before the body of the loop. The @code{until} command appeared
2972 to step back to the beginning of the loop when it advanced to this
2973 expression; however, it has not really gone to an earlier
2974 statement---not in terms of the actual machine code.
2976 @code{until} with no argument works by means of single
2977 instruction stepping, and hence is slower than @code{until} with an
2980 @item until @var{location}
2981 @itemx u @var{location}
2982 Continue running your program until either the specified location is
2983 reached, or the current stack frame returns. @var{location} is any of
2984 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
2985 ,Setting breakpoints}). This form of the command uses breakpoints,
2986 and hence is quicker than @code{until} without an argument.
2992 Execute one machine instruction, then stop and return to the debugger.
2994 It is often useful to do @samp{display/i $pc} when stepping by machine
2995 instructions. This makes @value{GDBN} automatically display the next
2996 instruction to be executed, each time your program stops. @xref{Auto
2997 Display,, Automatic display}.
2999 An argument is a repeat count, as in @code{step}.
3006 Execute one machine instruction, but if it is a function call,
3007 proceed until the function returns.
3009 An argument is a repeat count, as in @code{next}.
3017 A signal is an asynchronous event that can happen in a program. The
3018 operating system defines the possible kinds of signals, and gives each
3019 kind a name and a number. For example, in Unix @code{SIGINT} is the
3020 signal a program gets when you type an interrupt (often @kbd{C-c});
3021 @code{SIGSEGV} is the signal a program gets from referencing a place in
3022 memory far away from all the areas in use; @code{SIGALRM} occurs when
3023 the alarm clock timer goes off (which happens only if your program has
3024 requested an alarm).
3026 @cindex fatal signals
3027 Some signals, including @code{SIGALRM}, are a normal part of the
3028 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3029 errors; these signals are @dfn{fatal} (kill your program immediately) if the
3030 program has not specified in advance some other way to handle the signal.
3031 @code{SIGINT} does not indicate an error in your program, but it is normally
3032 fatal so it can carry out the purpose of the interrupt: to kill the program.
3034 @value{GDBN} has the ability to detect any occurrence of a signal in your
3035 program. You can tell @value{GDBN} in advance what to do for each kind of
3038 @cindex handling signals
3039 Normally, @value{GDBN} is set up to ignore non-erroneous signals like @code{SIGALRM}
3040 (so as not to interfere with their role in the functioning of your program)
3041 but to stop your program immediately whenever an error signal happens.
3042 You can change these settings with the @code{handle} command.
3046 @kindex info signals
3047 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3048 handle each one. You can use this to see the signal numbers of all
3049 the defined types of signals.
3051 @item handle @var{signal} @var{keywords}@dots{}
3053 Change the way @value{GDBN} handles signal @var{signal}. @var{signal} can be the
3054 number of a signal or its name (with or without the @samp{SIG} at the
3055 beginning). The @var{keywords} say what change to make.
3059 The keywords allowed by the @code{handle} command can be abbreviated.
3060 Their full names are:
3064 @value{GDBN} should not stop your program when this signal happens. It may
3065 still print a message telling you that the signal has come in.
3068 @value{GDBN} should stop your program when this signal happens. This implies
3069 the @code{print} keyword as well.
3072 @value{GDBN} should print a message when this signal happens.
3075 @value{GDBN} should not mention the occurrence of the signal at all. This
3076 implies the @code{nostop} keyword as well.
3079 @value{GDBN} should allow your program to see this signal; your program
3080 can handle the signal, or else it may terminate if the signal is fatal
3084 @value{GDBN} should not allow your program to see this signal.
3088 When a signal stops your program, the signal is not visible until you
3089 continue. Your program sees the signal then, if @code{pass} is in
3090 effect for the signal in question @emph{at that time}. In other words,
3091 after @value{GDBN} reports a signal, you can use the @code{handle}
3092 command with @code{pass} or @code{nopass} to control whether your
3093 program sees that signal when you continue.
3095 You can also use the @code{signal} command to prevent your program from
3096 seeing a signal, or cause it to see a signal it normally would not see,
3097 or to give it any signal at any time. For example, if your program stopped
3098 due to some sort of memory reference error, you might store correct
3099 values into the erroneous variables and continue, hoping to see more
3100 execution; but your program would probably terminate immediately as
3101 a result of the fatal signal once it saw the signal. To prevent this,
3102 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3108 @section Stopping and starting multi-thread programs
3110 When your program has multiple threads (@pxref{Threads,, Debugging
3111 programs with multiple threads}), you can choose whether to set
3112 breakpoints on all threads, or on a particular thread.
3115 @cindex breakpoints and threads
3116 @cindex thread breakpoints
3117 @kindex break @dots{} thread @var{threadno}
3118 @item break @var{linespec} thread @var{threadno}
3119 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3120 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3121 to specify that you only want @value{GDBN} to stop the program when a
3122 particular thread reaches this breakpoint. @var{threadno} is one of the
3123 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3124 column of the @samp{info threads} display.
3126 If you do not specify @samp{thread @var{threadno}} when you set a
3127 breakpoint, the breakpoint applies to @emph{all} threads of your
3130 You can use the @code{thread} qualifier on conditional breakpoints as
3131 well; in this case, place @samp{thread @var{threadno}} before the
3132 breakpoint condition, like this:
3135 (gdb) break frik.c:13 thread 28 if bartab > lim
3139 @cindex stopped threads
3140 @cindex threads, stopped
3141 Whenever your program stops under @value{GDBN} for any reason,
3142 @emph{all} threads of execution stop, not just the current thread. This
3143 allows you to examine the overall state of the program, including
3144 switching between threads, without worrying that things may change
3147 @cindex continuing threads
3148 @cindex threads, continuing
3149 Conversely, whenever you restart the program, @emph{all} threads start
3150 executing. @emph{This is true even when single-stepping} with commands
3151 like @code{step} or @code{next}.
3153 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3154 Since thread scheduling is up to your debugging target's operating
3155 system (not controlled by @value{GDBN}), other threads may
3156 execute more than one statement while the current thread completes a
3157 single step. Moreover, in general other threads stop in the middle of a
3158 statement, rather than at a clean statement boundary, when the program
3161 You might even find your program stopped in another thread after
3162 continuing or even single-stepping. This happens whenever some other
3163 thread runs into a breakpoint, a signal, or an exception before the
3164 first thread completes whatever you requested.
3168 @chapter Examining the Stack
3170 When your program has stopped, the first thing you need to know is where it
3171 stopped and how it got there.
3174 Each time your program performs a function call, the information about
3175 where in your program the call was made from is saved in a block of data
3176 called a @dfn{stack frame}. The frame also contains the arguments of the
3177 call and the local variables of the function that was called. All the
3178 stack frames are allocated in a region of memory called the @dfn{call
3181 When your program stops, the @value{GDBN} commands for examining the
3182 stack allow you to see all of this information.
3184 @cindex selected frame
3185 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3186 @value{GDBN} commands refer implicitly to the selected frame. In
3187 particular, whenever you ask @value{GDBN} for the value of a variable in
3188 your program, the value is found in the selected frame. There are
3189 special @value{GDBN} commands to select whichever frame you are
3192 When your program stops, @value{GDBN} automatically selects the
3193 currently executing frame and describes it briefly as the @code{frame}
3194 command does (@pxref{Frame Info, ,Information about a frame}).
3197 * Frames:: Stack frames
3198 * Backtrace:: Backtraces
3199 * Selection:: Selecting a frame
3200 * Frame Info:: Information on a frame
3202 * MIPS Stack:: MIPS machines and the function stack
3207 @section Stack frames
3211 The call stack is divided up into contiguous pieces called @dfn{stack
3212 frames}, or @dfn{frames} for short; each frame is the data associated
3213 with one call to one function. The frame contains the arguments given
3214 to the function, the function's local variables, and the address at
3215 which the function is executing.
3217 @cindex initial frame
3218 @cindex outermost frame
3219 @cindex innermost frame
3220 When your program is started, the stack has only one frame, that of the
3221 function @code{main}. This is called the @dfn{initial} frame or the
3222 @dfn{outermost} frame. Each time a function is called, a new frame is
3223 made. Each time a function returns, the frame for that function invocation
3224 is eliminated. If a function is recursive, there can be many frames for
3225 the same function. The frame for the function in which execution is
3226 actually occurring is called the @dfn{innermost} frame. This is the most
3227 recently created of all the stack frames that still exist.
3229 @cindex frame pointer
3230 Inside your program, stack frames are identified by their addresses. A
3231 stack frame consists of many bytes, each of which has its own address; each
3232 kind of computer has a convention for choosing one of those bytes whose
3233 address serves as the address of the frame. Usually this address is kept
3234 in a register called the @dfn{frame pointer register} while execution is
3235 going on in that frame.
3237 @cindex frame number
3238 @value{GDBN} assigns numbers to all existing stack frames, starting with
3239 zero for the innermost frame, one for the frame that called it,
3240 and so on upward. These numbers do not really exist in your program;
3241 they are assigned by @value{GDBN} to give you a way of designating stack
3242 frames in @value{GDBN} commands.
3244 @c below produces an acceptable overful hbox. --mew 13aug1993
3245 @cindex frameless execution
3246 Some compilers provide a way to compile functions so that they operate
3247 without stack frames. (For example, the @code{@value{GCC}} option
3248 @samp{-fomit-frame-pointer} generates functions without a frame.)
3249 This is occasionally done with heavily used library functions to save
3250 the frame setup time. @value{GDBN} has limited facilities for dealing
3251 with these function invocations. If the innermost function invocation
3252 has no stack frame, @value{GDBN} nevertheless regards it as though
3253 it had a separate frame, which is numbered zero as usual, allowing
3254 correct tracing of the function call chain. However, @value{GDBN} has
3255 no provision for frameless functions elsewhere in the stack.
3260 A backtrace is a summary of how your program got where it is. It shows one
3261 line per frame, for many frames, starting with the currently executing
3262 frame (frame zero), followed by its caller (frame one), and on up the
3270 Print a backtrace of the entire stack: one line per frame for all
3271 frames in the stack.
3273 You can stop the backtrace at any time by typing the system interrupt
3274 character, normally @kbd{C-c}.
3276 @item backtrace @var{n}
3278 Similar, but print only the innermost @var{n} frames.
3280 @item backtrace -@var{n}
3282 Similar, but print only the outermost @var{n} frames.
3288 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3289 are additional aliases for @code{backtrace}.
3291 Each line in the backtrace shows the frame number and the function name.
3292 The program counter value is also shown---unless you use @code{set
3293 print address off}. The backtrace also shows the source file name and
3294 line number, as well as the arguments to the function. The program
3295 counter value is omitted if it is at the beginning of the code for that
3298 Here is an example of a backtrace. It was made with the command
3299 @samp{bt 3}, so it shows the innermost three frames.
3303 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3305 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3306 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3308 (More stack frames follow...)
3313 The display for frame zero does not begin with a program counter
3314 value, indicating that your program has stopped at the beginning of the
3315 code for line @code{993} of @code{builtin.c}.
3318 @section Selecting a frame
3320 Most commands for examining the stack and other data in your program work on
3321 whichever stack frame is selected at the moment. Here are the commands for
3322 selecting a stack frame; all of them finish by printing a brief description
3323 of the stack frame just selected.
3330 Select frame number @var{n}. Recall that frame zero is the innermost
3331 (currently executing) frame, frame one is the frame that called the
3332 innermost one, and so on. The highest-numbered frame is the one for
3335 @item frame @var{addr}
3337 Select the frame at address @var{addr}. This is useful mainly if the
3338 chaining of stack frames has been damaged by a bug, making it
3339 impossible for @value{GDBN} to assign numbers properly to all frames. In
3340 addition, this can be useful when your program has multiple stacks and
3341 switches between them.
3343 @ifclear H8EXCLUSIVE
3344 On the SPARC architecture, @code{frame} needs two addresses to
3345 select an arbitrary frame: a frame pointer and a stack pointer.
3347 On the MIPS and Alpha architecture, it needs two addresses: a stack
3348 pointer and a program counter.
3350 On the 29k architecture, it needs three addresses: a register stack
3351 pointer, a program counter, and a memory stack pointer.
3352 @c note to future updaters: this is conditioned on a flag
3353 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3354 @c as of 27 Jan 1994.
3359 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3360 advances toward the outermost frame, to higher frame numbers, to frames
3361 that have existed longer. @var{n} defaults to one.
3366 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3367 advances toward the innermost frame, to lower frame numbers, to frames
3368 that were created more recently. @var{n} defaults to one. You may
3369 abbreviate @code{down} as @code{do}.
3372 All of these commands end by printing two lines of output describing the
3373 frame. The first line shows the frame number, the function name, the
3374 arguments, and the source file and line number of execution in that
3375 frame. The second line shows the text of that source line.
3383 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3385 10 read_input_file (argv[i]);
3389 After such a printout, the @code{list} command with no arguments
3390 prints ten lines centered on the point of execution in the frame.
3391 @xref{List, ,Printing source lines}.
3394 @item up-silently @var{n}
3395 @itemx down-silently @var{n}
3396 @kindex down-silently
3398 These two commands are variants of @code{up} and @code{down},
3399 respectively; they differ in that they do their work silently, without
3400 causing display of the new frame. They are intended primarily for use
3401 in @value{GDBN} command scripts, where the output might be unnecessary and
3406 @section Information about a frame
3408 There are several other commands to print information about the selected
3414 When used without any argument, this command does not change which
3415 frame is selected, but prints a brief description of the currently
3416 selected stack frame. It can be abbreviated @code{f}. With an
3417 argument, this command is used to select a stack frame.
3418 @xref{Selection, ,Selecting a frame}.
3424 This command prints a verbose description of the selected stack frame,
3425 including the address of the frame, the addresses of the next frame down
3426 (called by this frame) and the next frame up (caller of this frame), the
3427 language that the source code corresponding to this frame was written in,
3428 the address of the frame's arguments, the program counter saved in it
3429 (the address of execution in the caller frame), and which registers
3430 were saved in the frame. The verbose description is useful when
3431 something has gone wrong that has made the stack format fail to fit
3432 the usual conventions.
3434 @item info frame @var{addr}
3435 @itemx info f @var{addr}
3436 Print a verbose description of the frame at address @var{addr}, without
3437 selecting that frame. The selected frame remains unchanged by this
3438 command. This requires the same kind of address (more than one for some
3439 architectures) that you specify in the @code{frame} command.
3440 @xref{Selection, ,Selecting a frame}.
3444 Print the arguments of the selected frame, each on a separate line.
3448 Print the local variables of the selected frame, each on a separate
3449 line. These are all variables (declared either static or automatic)
3450 accessible at the point of execution of the selected frame.
3455 @cindex catch exceptions
3456 @cindex exception handlers
3457 Print a list of all the exception handlers that are active in the
3458 current stack frame at the current point of execution. To see other
3459 exception handlers, visit the associated frame (using the @code{up},
3460 @code{down}, or @code{frame} commands); then type @code{info catch}.
3461 @xref{Exception Handling, ,Breakpoints and exceptions}.
3467 @section MIPS machines and the function stack
3469 @cindex stack on MIPS
3471 MIPS based computers use an unusual stack frame, which sometimes
3472 requires @value{GDBN} to search backward in the object code to find the
3473 beginning of a function.
3475 @cindex response time, MIPS debugging
3476 To improve response time (especially for embedded applications, where
3477 @value{GDBN} may be restricted to a slow serial line for this search)
3478 you may want to limit the size of this search, using one of these
3480 @c FIXME! So what happens when GDB does *not* find the beginning of a
3483 @cindex @code{heuristic-fence-post} (MIPS)
3485 @item set heuristic-fence-post @var{limit}
3486 Restrict @value{GDBN} to examining at most @var{limit} bytes in its search
3487 for the beginning of a function. A value of @code{0} (the default)
3488 means there is no limit.
3490 @item show heuristic-fence-post
3491 Display the current limit.
3495 These commands are available @emph{only} when @value{GDBN} is configured
3496 for debugging programs on MIPS processors.
3500 @chapter Examining Source Files
3502 @value{GDBN} can print parts of your program's source, since the debugging
3503 information recorded in the program tells @value{GDBN} what source files were
3504 used to build it. When your program stops, @value{GDBN} spontaneously prints
3505 the line where it stopped. Likewise, when you select a stack frame
3506 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
3507 execution in that frame has stopped. You can print other portions of
3508 source files by explicit command.
3511 If you use @value{GDBN} through its GNU Emacs interface, you may prefer to use
3512 Emacs facilities to view source; @pxref{Emacs, ,Using @value{GDBN} under GNU
3517 * List:: Printing source lines
3519 * Search:: Searching source files
3522 * Source Path:: Specifying source directories
3523 * Machine Code:: Source and machine code
3527 @section Printing source lines
3531 To print lines from a source file, use the @code{list} command
3532 (abbreviated @code{l}). There are several ways to specify what part
3533 of the file you want to print.
3535 Here are the forms of the @code{list} command most commonly used:
3538 @item list @var{linenum}
3539 Print lines centered around line number @var{linenum} in the
3540 current source file.
3542 @item list @var{function}
3543 Print lines centered around the beginning of function
3547 Print more lines. If the last lines printed were printed with a
3548 @code{list} command, this prints lines following the last lines
3549 printed; however, if the last line printed was a solitary line printed
3550 as part of displaying a stack frame (@pxref{Stack, ,Examining the
3551 Stack}), this prints lines centered around that line.
3554 Print lines just before the lines last printed.
3557 By default, @value{GDBN} prints ten source lines with any of these forms of
3558 the @code{list} command. You can change this using @code{set listsize}:
3561 @item set listsize @var{count}
3562 @kindex set listsize
3563 Make the @code{list} command display @var{count} source lines (unless
3564 the @code{list} argument explicitly specifies some other number).
3567 @kindex show listsize
3568 Display the number of lines that @code{list} prints.
3571 Repeating a @code{list} command with @key{RET} discards the argument,
3572 so it is equivalent to typing just @code{list}. This is more useful
3573 than listing the same lines again. An exception is made for an
3574 argument of @samp{-}; that argument is preserved in repetition so that
3575 each repetition moves up in the source file.
3578 In general, the @code{list} command expects you to supply zero, one or two
3579 @dfn{linespecs}. Linespecs specify source lines; there are several ways
3580 of writing them but the effect is always to specify some source line.
3581 Here is a complete description of the possible arguments for @code{list}:
3584 @item list @var{linespec}
3585 Print lines centered around the line specified by @var{linespec}.
3587 @item list @var{first},@var{last}
3588 Print lines from @var{first} to @var{last}. Both arguments are
3591 @item list ,@var{last}
3592 Print lines ending with @var{last}.
3594 @item list @var{first},
3595 Print lines starting with @var{first}.
3598 Print lines just after the lines last printed.
3601 Print lines just before the lines last printed.
3604 As described in the preceding table.
3607 Here are the ways of specifying a single source line---all the
3612 Specifies line @var{number} of the current source file.
3613 When a @code{list} command has two linespecs, this refers to
3614 the same source file as the first linespec.
3617 Specifies the line @var{offset} lines after the last line printed.
3618 When used as the second linespec in a @code{list} command that has
3619 two, this specifies the line @var{offset} lines down from the
3623 Specifies the line @var{offset} lines before the last line printed.
3625 @item @var{filename}:@var{number}
3626 Specifies line @var{number} in the source file @var{filename}.
3628 @item @var{function}
3629 @c FIXME: "of the open-brace" is C-centric. When we add other langs...
3630 Specifies the line of the open-brace that begins the body of the
3631 function @var{function}.
3633 @item @var{filename}:@var{function}
3634 Specifies the line of the open-brace that begins the body of the
3635 function @var{function} in the file @var{filename}. You only need the
3636 file name with a function name to avoid ambiguity when there are
3637 identically named functions in different source files.
3639 @item *@var{address}
3640 Specifies the line containing the program address @var{address}.
3641 @var{address} may be any expression.
3646 @section Searching source files
3648 @kindex reverse-search
3650 There are two commands for searching through the current source file for a
3654 @item forward-search @var{regexp}
3655 @itemx search @var{regexp}
3657 @kindex forward-search
3658 The command @samp{forward-search @var{regexp}} checks each line,
3659 starting with the one following the last line listed, for a match for
3660 @var{regexp}. It lists the line that is found. You can use
3661 synonym @samp{search @var{regexp}} or abbreviate the command name as
3664 @item reverse-search @var{regexp}
3665 The command @samp{reverse-search @var{regexp}} checks each line, starting
3666 with the one before the last line listed and going backward, for a match
3667 for @var{regexp}. It lists the line that is found. You can abbreviate
3668 this command as @code{rev}.
3673 @section Specifying source directories
3676 @cindex directories for source files
3677 Executable programs sometimes do not record the directories of the source
3678 files from which they were compiled, just the names. Even when they do,
3679 the directories could be moved between the compilation and your debugging
3680 session. @value{GDBN} has a list of directories to search for source files;
3681 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
3682 it tries all the directories in the list, in the order they are present
3683 in the list, until it finds a file with the desired name. Note that
3684 the executable search path is @emph{not} used for this purpose. Neither is
3685 the current working directory, unless it happens to be in the source
3688 If @value{GDBN} cannot find a source file in the source path, and the
3689 object program records a directory, @value{GDBN} tries that directory
3690 too. If the source path is empty, and there is no record of the
3691 compilation directory, @value{GDBN} looks in the current directory as a
3694 Whenever you reset or rearrange the source path, @value{GDBN} clears out
3695 any information it has cached about where source files are found and where
3696 each line is in the file.
3699 When you start @value{GDBN}, its source path is empty.
3700 To add other directories, use the @code{directory} command.
3703 @item directory @var{dirname} @dots{}
3704 Add directory @var{dirname} to the front of the source path. Several
3705 directory names may be given to this command, separated by @samp{:} or
3706 whitespace. You may specify a directory that is already in the source
3707 path; this moves it forward, so @value{GDBN} searches it sooner.
3713 @cindex compilation directory
3714 @cindex current directory
3715 @cindex working directory
3716 @cindex directory, current
3717 @cindex directory, compilation
3718 You can use the string @samp{$cdir} to refer to the compilation
3719 directory (if one is recorded), and @samp{$cwd} to refer to the current
3720 working directory. @samp{$cwd} is not the same as @samp{.}---the former
3721 tracks the current working directory as it changes during your @value{GDBN}
3722 session, while the latter is immediately expanded to the current
3723 directory at the time you add an entry to the source path.
3726 Reset the source path to empty again. This requires confirmation.
3728 @c RET-repeat for @code{directory} is explicitly disabled, but since
3729 @c repeating it would be a no-op we do not say that. (thanks to RMS)
3731 @item show directories
3732 @kindex show directories
3733 Print the source path: show which directories it contains.
3736 If your source path is cluttered with directories that are no longer of
3737 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
3738 versions of source. You can correct the situation as follows:
3742 Use @code{directory} with no argument to reset the source path to empty.
3745 Use @code{directory} with suitable arguments to reinstall the
3746 directories you want in the source path. You can add all the
3747 directories in one command.
3751 @section Source and machine code
3753 You can use the command @code{info line} to map source lines to program
3754 addresses (and vice versa), and the command @code{disassemble} to display
3755 a range of addresses as machine instructions.
3758 @item info line @var{linespec}
3760 Print the starting and ending addresses of the compiled code for
3761 source line @var{linespec}. You can specify source lines in any of
3762 the ways understood by the @code{list} command (@pxref{List, ,Printing
3766 For example, we can use @code{info line} to discover the location of
3767 the object code for the first line of function
3768 @code{m4_changequote}:
3771 (@value{GDBP}) info line m4_changecom
3772 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
3776 We can also inquire (using @code{*@var{addr}} as the form for
3777 @var{linespec}) what source line covers a particular address:
3779 (@value{GDBP}) info line *0x63ff
3780 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
3783 @cindex @code{$_} and @code{info line}
3784 After @code{info line}, the default address for the @code{x} command
3785 is changed to the starting address of the line, so that @samp{x/i} is
3786 sufficient to begin examining the machine code (@pxref{Memory,
3787 ,Examining memory}). Also, this address is saved as the value of the
3788 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
3794 @cindex assembly instructions
3795 @cindex instructions, assembly
3796 @cindex machine instructions
3797 @cindex listing machine instructions
3798 This specialized command dumps a range of memory as machine
3799 instructions. The default memory range is the function surrounding the
3800 program counter of the selected frame. A single argument to this
3801 command is a program counter value; @value{GDBN} dumps the function
3802 surrounding this value. Two arguments specify a range of addresses
3803 (first inclusive, second exclusive) to dump.
3806 @ifclear H8EXCLUSIVE
3807 We can use @code{disassemble} to inspect the object code
3808 range shown in the last @code{info line} example (the example
3809 shows SPARC machine instructions):
3813 (@value{GDBP}) disas 0x63e4 0x6404
3814 Dump of assembler code from 0x63e4 to 0x6404:
3815 0x63e4 <builtin_init+5340>: ble 0x63f8 <builtin_init+5360>
3816 0x63e8 <builtin_init+5344>: sethi %hi(0x4c00), %o0
3817 0x63ec <builtin_init+5348>: ld [%i1+4], %o0
3818 0x63f0 <builtin_init+5352>: b 0x63fc <builtin_init+5364>
3819 0x63f4 <builtin_init+5356>: ld [%o0+4], %o0
3820 0x63f8 <builtin_init+5360>: or %o0, 0x1a4, %o0
3821 0x63fc <builtin_init+5364>: call 0x9288 <path_search>
3822 0x6400 <builtin_init+5368>: nop
3823 End of assembler dump.
3828 For example, here is the beginning of the output for the
3829 disassembly of a function @code{fact}:
3833 (@value{GDBP}) disas fact
3834 Dump of assembler code for function fact:
3836 0x802c <fact>: 6d f2 mov.w r2,@@-r7
3837 0x802e <fact+2>: 6d f3 mov.w r3,@@-r7
3838 0x8030 <fact+4>: 6d f6 mov.w r6,@@-r7
3839 0x8032 <fact+6>: 0d 76 mov.w r7,r6
3840 0x8034 <fact+8>: 6f 70 00 08 mov.w @@(0x8,r7),r0
3841 0x8038 <fact+12> 19 11 sub.w r1,r1
3849 @chapter Examining Data
3851 @cindex printing data
3852 @cindex examining data
3855 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
3856 @c document because it is nonstandard... Under Epoch it displays in a
3857 @c different window or something like that.
3858 The usual way to examine data in your program is with the @code{print}
3859 command (abbreviated @code{p}), or its synonym @code{inspect}.
3861 It evaluates and prints the value of an expression of the language your
3862 program is written in (@pxref{Languages, ,Using @value{GDBN} with Different
3867 @item print @var{exp}
3868 @itemx print /@var{f} @var{exp}
3869 @var{exp} is an expression (in the source language). By default the
3870 value of @var{exp} is printed in a format appropriate to its data type;
3871 you can choose a different format by specifying @samp{/@var{f}}, where
3872 @var{f} is a letter specifying the format; @pxref{Output Formats,,Output
3876 @itemx print /@var{f}
3877 If you omit @var{exp}, @value{GDBN} displays the last value again (from the
3878 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
3879 conveniently inspect the same value in an alternative format.
3882 A more low-level way of examining data is with the @code{x} command.
3883 It examines data in memory at a specified address and prints it in a
3884 specified format. @xref{Memory, ,Examining memory}.
3886 If you are interested in information about types, or about how the fields
3891 are declared, use the @code{ptype @var{exp}}
3892 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol Table}.
3895 * Expressions:: Expressions
3896 * Variables:: Program variables
3897 * Arrays:: Artificial arrays
3898 * Output Formats:: Output formats
3899 * Memory:: Examining memory
3900 * Auto Display:: Automatic display
3901 * Print Settings:: Print settings
3902 * Value History:: Value history
3903 * Convenience Vars:: Convenience variables
3904 * Registers:: Registers
3906 * Floating Point Hardware:: Floating point hardware
3911 @section Expressions
3914 @code{print} and many other @value{GDBN} commands accept an expression and
3915 compute its value. Any kind of constant, variable or operator defined
3916 by the programming language you are using is valid in an expression in
3917 @value{GDBN}. This includes conditional expressions, function calls, casts
3918 and string constants. It unfortunately does not include symbols defined
3919 by preprocessor @code{#define} commands.
3922 Because C is so widespread, most of the expressions shown in examples in
3923 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
3924 Languages}, for information on how to use expressions in other
3927 In this section, we discuss operators that you can use in @value{GDBN}
3928 expressions regardless of your programming language.
3930 Casts are supported in all languages, not just in C, because it is so
3931 useful to cast a number into a pointer so as to examine a structure
3932 at that address in memory.
3933 @c FIXME: casts supported---Mod2 true?
3936 @value{GDBN} supports these operators in addition to those of programming
3941 @samp{@@} is a binary operator for treating parts of memory as arrays.
3942 @xref{Arrays, ,Artificial arrays}, for more information.
3945 @samp{::} allows you to specify a variable in terms of the file or
3946 function where it is defined. @xref{Variables, ,Program variables}.
3948 @item @{@var{type}@} @var{addr}
3949 @cindex @{@var{type}@}
3950 @cindex type casting memory
3951 @cindex memory, viewing as typed object
3952 @cindex casts, to view memory
3953 Refers to an object of type @var{type} stored at address @var{addr} in
3954 memory. @var{addr} may be any expression whose value is an integer or
3955 pointer (but parentheses are required around binary operators, just as in
3956 a cast). This construct is allowed regardless of what kind of data is
3957 normally supposed to reside at @var{addr}.
3961 @section Program variables
3963 The most common kind of expression to use is the name of a variable
3966 Variables in expressions are understood in the selected stack frame
3967 (@pxref{Selection, ,Selecting a frame}); they must either be global
3968 (or static) or be visible according to the scope rules of the
3969 programming language from the point of execution in that frame. This
3970 means that in the function
3985 you can examine and use the variable @code{a} whenever your program is
3986 executing within the function @code{foo}, but you can only use or
3987 examine the variable @code{b} while your program is executing inside
3988 the block where @code{b} is declared.
3990 @cindex variable name conflict
3991 There is an exception: you can refer to a variable or function whose
3992 scope is a single source file even if the current execution point is not
3993 in this file. But it is possible to have more than one such variable or
3994 function with the same name (in different source files). If that
3995 happens, referring to that name has unpredictable effects. If you wish,
3996 you can specify a static variable in a particular function or file,
3997 using the colon-colon notation:
4001 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4005 @var{file}::@var{variable}
4006 @var{function}::@var{variable}
4010 Here @var{file} or @var{function} is the name of the context for the
4011 static @var{variable}. In the case of file names, you can use quotes to
4012 make sure @value{GDBN} parses the file name as a single word---for example,
4013 to print a global value of @code{x} defined in @file{f2.c}:
4016 (@value{GDBP}) p 'f2.c'::x
4020 @cindex C++ scope resolution
4021 This use of @samp{::} is very rarely in conflict with the very similar
4022 use of the same notation in C++. @value{GDBN} also supports use of the C++
4023 scope resolution operator in @value{GDBN} expressions.
4024 @c FIXME: Um, so what happens in one of those rare cases where it's in
4028 @cindex wrong values
4029 @cindex variable values, wrong
4031 @emph{Warning:} Occasionally, a local variable may appear to have the
4032 wrong value at certain points in a function---just after entry to a new
4033 scope, and just before exit.
4035 You may see this problem when you are stepping by machine instructions.
4036 This is because on most machines, it takes more than one instruction to
4037 set up a stack frame (including local variable definitions); if you are
4038 stepping by machine instructions, variables may appear to have the wrong
4039 values until the stack frame is completely built. On exit, it usually
4040 also takes more than one machine instruction to destroy a stack frame;
4041 after you begin stepping through that group of instructions, local
4042 variable definitions may be gone.
4045 @section Artificial arrays
4047 @cindex artificial array
4049 It is often useful to print out several successive objects of the
4050 same type in memory; a section of an array, or an array of
4051 dynamically determined size for which only a pointer exists in the
4054 You can do this by referring to a contiguous span of memory as an
4055 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4056 operand of @samp{@@} should be the first element of the desired array,
4057 as an individual object. The right operand should be the desired length
4058 of the array. The result is an array value whose elements are all of
4059 the type of the left argument. The first element is actually the left
4060 argument; the second element comes from bytes of memory immediately
4061 following those that hold the first element, and so on. Here is an
4062 example. If a program says
4065 int *array = (int *) malloc (len * sizeof (int));
4069 you can print the contents of @code{array} with
4075 The left operand of @samp{@@} must reside in memory. Array values made
4076 with @samp{@@} in this way behave just like other arrays in terms of
4077 subscripting, and are coerced to pointers when used in expressions.
4078 Artificial arrays most often appear in expressions via the value history
4079 (@pxref{Value History, ,Value history}), after printing one out.
4081 Another way to create an artificial array is to use a cast.
4082 This re-interprets a value as if it were an array.
4083 The value need not be in memory:
4085 (@value{GDBP}) p/x (short[2])0x12345678
4086 $1 = @{0x1234, 0x5678@}
4089 As a convenience, if you leave the array length out (as in
4090 @samp{(@var{type})[])@var{value}}) gdb calculates the size to fill
4091 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4093 (@value{GDBP}) p/x (short[])0x12345678
4094 $2 = @{0x1234, 0x5678@}
4097 Sometimes the artificial array mechanism is not quite enough; in
4098 moderately complex data structures, the elements of interest may not
4099 actually be adjacent---for example, if you are interested in the values
4100 of pointers in an array. One useful work-around in this situation is
4101 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4102 variables}) as a counter in an expression that prints the first
4103 interesting value, and then repeat that expression via @key{RET}. For
4104 instance, suppose you have an array @code{dtab} of pointers to
4105 structures, and you are interested in the values of a field @code{fv}
4106 in each structure. Here is an example of what you might type:
4116 @node Output Formats
4117 @section Output formats
4119 @cindex formatted output
4120 @cindex output formats
4121 By default, @value{GDBN} prints a value according to its data type. Sometimes
4122 this is not what you want. For example, you might want to print a number
4123 in hex, or a pointer in decimal. Or you might want to view data in memory
4124 at a certain address as a character string or as an instruction. To do
4125 these things, specify an @dfn{output format} when you print a value.
4127 The simplest use of output formats is to say how to print a value
4128 already computed. This is done by starting the arguments of the
4129 @code{print} command with a slash and a format letter. The format
4130 letters supported are:
4134 Regard the bits of the value as an integer, and print the integer in
4138 Print as integer in signed decimal.
4141 Print as integer in unsigned decimal.
4144 Print as integer in octal.
4147 Print as integer in binary. The letter @samp{t} stands for ``two''.
4148 @footnote{@samp{b} cannot be used because these format letters are also
4149 used with the @code{x} command, where @samp{b} stands for ``byte'';
4150 @pxref{Memory,,Examining memory}.}
4153 @cindex unknown address, locating
4154 Print as an address, both absolute in hexadecimal and as an offset from
4155 the nearest preceding symbol. You can use this format used to discover
4156 where (in what function) an unknown address is located:
4159 (@value{GDBP}) p/a 0x54320
4160 $3 = 0x54320 <_initialize_vx+396>
4164 Regard as an integer and print it as a character constant.
4167 Regard the bits of the value as a floating point number and print
4168 using typical floating point syntax.
4171 For example, to print the program counter in hex (@pxref{Registers}), type
4178 Note that no space is required before the slash; this is because command
4179 names in @value{GDBN} cannot contain a slash.
4181 To reprint the last value in the value history with a different format,
4182 you can use the @code{print} command with just a format and no
4183 expression. For example, @samp{p/x} reprints the last value in hex.
4186 @section Examining memory
4188 You can use the command @code{x} (for ``examine'') to examine memory in
4189 any of several formats, independently of your program's data types.
4191 @cindex examining memory
4194 @item x/@var{nfu} @var{addr}
4197 Use the @code{x} command to examine memory.
4200 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4201 much memory to display and how to format it; @var{addr} is an
4202 expression giving the address where you want to start displaying memory.
4203 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4204 Several commands set convenient defaults for @var{addr}.
4207 @item @var{n}, the repeat count
4208 The repeat count is a decimal integer; the default is 1. It specifies
4209 how much memory (counting by units @var{u}) to display.
4210 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4213 @item @var{f}, the display format
4214 The display format is one of the formats used by @code{print},
4215 or @samp{s} (null-terminated string) or @samp{i} (machine instruction).
4216 The default is @samp{x} (hexadecimal) initially, or the format from the
4217 last time you used either @code{x} or @code{print}.
4219 @item @var{u}, the unit size
4220 The unit size is any of
4226 Halfwords (two bytes).
4228 Words (four bytes). This is the initial default.
4230 Giant words (eight bytes).
4233 Each time you specify a unit size with @code{x}, that size becomes the
4234 default unit the next time you use @code{x}. (For the @samp{s} and
4235 @samp{i} formats, the unit size is ignored and is normally not written.)
4237 @item @var{addr}, starting display address
4238 @var{addr} is the address where you want @value{GDBN} to begin displaying
4239 memory. The expression need not have a pointer value (though it may);
4240 it is always interpreted as an integer address of a byte of memory.
4241 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4242 @var{addr} is usually just after the last address examined---but several
4243 other commands also set the default address: @code{info breakpoints} (to
4244 the address of the last breakpoint listed), @code{info line} (to the
4245 starting address of a line), and @code{print} (if you use it to display
4246 a value from memory).
4249 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4250 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4251 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4252 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4253 @pxref{Registers}) in hexadecimal (@samp{x}).
4255 Since the letters indicating unit sizes are all distinct from the
4256 letters specifying output formats, you do not have to remember whether
4257 unit size or format comes first; either order works. The output
4258 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4259 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4261 Even though the unit size @var{u} is ignored for the formats @samp{s}
4262 and @samp{i}, you might still want to use a count @var{n}; for example,
4263 @samp{3i} specifies that you want to see three machine instructions,
4264 including any operands. The command @code{disassemble} gives an
4265 alternative way of inspecting machine instructions; @pxref{Machine
4266 Code,,Source and machine code}.
4268 All the defaults for the arguments to @code{x} are designed to make it
4269 easy to continue scanning memory with minimal specifications each time
4270 you use @code{x}. For example, after you have inspected three machine
4271 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4272 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4273 the repeat count @var{n} is used again; the other arguments default as
4274 for successive uses of @code{x}.
4276 @cindex @code{$_}, @code{$__}, and value history
4277 The addresses and contents printed by the @code{x} command are not saved
4278 in the value history because there is often too much of them and they
4279 would get in the way. Instead, @value{GDBN} makes these values available for
4280 subsequent use in expressions as values of the convenience variables
4281 @code{$_} and @code{$__}. After an @code{x} command, the last address
4282 examined is available for use in expressions in the convenience variable
4283 @code{$_}. The contents of that address, as examined, are available in
4284 the convenience variable @code{$__}.
4286 If the @code{x} command has a repeat count, the address and contents saved
4287 are from the last memory unit printed; this is not the same as the last
4288 address printed if several units were printed on the last line of output.
4291 @section Automatic display
4292 @cindex automatic display
4293 @cindex display of expressions
4295 If you find that you want to print the value of an expression frequently
4296 (to see how it changes), you might want to add it to the @dfn{automatic
4297 display list} so that @value{GDBN} prints its value each time your program stops.
4298 Each expression added to the list is given a number to identify it;
4299 to remove an expression from the list, you specify that number.
4300 The automatic display looks like this:
4304 3: bar[5] = (struct hack *) 0x3804
4308 This display shows item numbers, expressions and their current values. As with
4309 displays you request manually using @code{x} or @code{print}, you can
4310 specify the output format you prefer; in fact, @code{display} decides
4311 whether to use @code{print} or @code{x} depending on how elaborate your
4312 format specification is---it uses @code{x} if you specify a unit size,
4313 or one of the two formats (@samp{i} and @samp{s}) that are only
4314 supported by @code{x}; otherwise it uses @code{print}.
4317 @item display @var{exp}
4319 Add the expression @var{exp} to the list of expressions to display
4320 each time your program stops. @xref{Expressions, ,Expressions}.
4322 @code{display} does not repeat if you press @key{RET} again after using it.
4324 @item display/@var{fmt} @var{exp}
4325 For @var{fmt} specifying only a display format and not a size or
4326 count, add the expression @var{exp} to the auto-display list but
4327 arrange to display it each time in the specified format @var{fmt}.
4328 @xref{Output Formats,,Output formats}.
4330 @item display/@var{fmt} @var{addr}
4331 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4332 number of units, add the expression @var{addr} as a memory address to
4333 be examined each time your program stops. Examining means in effect
4334 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4337 For example, @samp{display/i $pc} can be helpful, to see the machine
4338 instruction about to be executed each time execution stops (@samp{$pc}
4339 is a common name for the program counter; @pxref{Registers}).
4342 @item undisplay @var{dnums}@dots{}
4343 @itemx delete display @var{dnums}@dots{}
4344 @kindex delete display
4346 Remove item numbers @var{dnums} from the list of expressions to display.
4348 @code{undisplay} does not repeat if you press @key{RET} after using it.
4349 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4351 @item disable display @var{dnums}@dots{}
4352 @kindex disable display
4353 Disable the display of item numbers @var{dnums}. A disabled display
4354 item is not printed automatically, but is not forgotten. It may be
4355 enabled again later.
4357 @item enable display @var{dnums}@dots{}
4358 @kindex enable display
4359 Enable display of item numbers @var{dnums}. It becomes effective once
4360 again in auto display of its expression, until you specify otherwise.
4363 Display the current values of the expressions on the list, just as is
4364 done when your program stops.
4367 @kindex info display
4368 Print the list of expressions previously set up to display
4369 automatically, each one with its item number, but without showing the
4370 values. This includes disabled expressions, which are marked as such.
4371 It also includes expressions which would not be displayed right now
4372 because they refer to automatic variables not currently available.
4375 If a display expression refers to local variables, then it does not make
4376 sense outside the lexical context for which it was set up. Such an
4377 expression is disabled when execution enters a context where one of its
4378 variables is not defined. For example, if you give the command
4379 @code{display last_char} while inside a function with an argument
4380 @code{last_char}, @value{GDBN} displays this argument while your program
4381 continues to stop inside that function. When it stops elsewhere---where
4382 there is no variable @code{last_char}---the display is disabled
4383 automatically. The next time your program stops where @code{last_char}
4384 is meaningful, you can enable the display expression once again.
4386 @node Print Settings
4387 @section Print settings
4389 @cindex format options
4390 @cindex print settings
4391 @value{GDBN} provides the following ways to control how arrays, structures,
4392 and symbols are printed.
4395 These settings are useful for debugging programs in any language:
4398 @item set print address
4399 @itemx set print address on
4400 @kindex set print address
4401 @value{GDBN} prints memory addresses showing the location of stack
4402 traces, structure values, pointer values, breakpoints, and so forth,
4403 even when it also displays the contents of those addresses. The default
4404 is @code{on}. For example, this is what a stack frame display looks like, with
4405 @code{set print address on}:
4410 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
4412 530 if (lquote != def_lquote)
4416 @item set print address off
4417 Do not print addresses when displaying their contents. For example,
4418 this is the same stack frame displayed with @code{set print address off}:
4422 (@value{GDBP}) set print addr off
4424 #0 set_quotes (lq="<<", rq=">>") at input.c:530
4425 530 if (lquote != def_lquote)
4429 You can use @samp{set print address off} to eliminate all machine
4430 dependent displays from the @value{GDBN} interface. For example, with
4431 @code{print address off}, you should get the same text for backtraces on
4432 all machines---whether or not they involve pointer arguments.
4434 @item show print address
4435 @kindex show print address
4436 Show whether or not addresses are to be printed.
4439 When @value{GDBN} prints a symbolic address, it normally prints the
4440 closest earlier symbol plus an offset. If that symbol does not uniquely
4441 identify the address (for example, it is a name whose scope is a single
4442 source file), you may need to disambiguate. One way to do this is with
4443 @code{info line}, for example @samp{info line *0x4537}. Alternately,
4444 you can set @value{GDBN} to print the source file and line number when
4445 it prints a symbolic address:
4448 @item set print symbol-filename on
4449 @kindex set print symbol-filename
4450 Tell @value{GDBN} to print the source file name and line number of a
4451 symbol in the symbolic form of an address.
4453 @item set print symbol-filename off
4454 Do not print source file name and line number of a symbol. This is the
4457 @item show print symbol-filename
4458 @kindex show print symbol-filename
4459 Show whether or not @value{GDBN} will print the source file name and
4460 line number of a symbol in the symbolic form of an address.
4463 Another situation where it is helpful to show symbol filenames and line
4464 numbers is when disassembling code; @value{GDBN} shows you the line
4465 number and source file that corresponds to each instruction.
4467 Also, you may wish to see the symbolic form only if the address being
4468 printed is reasonably close to the closest earlier symbol:
4471 @item set print max-symbolic-offset @var{max-offset}
4472 @kindex set print max-symbolic-offset
4473 Tell @value{GDBN} to only display the symbolic form of an address if the
4474 offset between the closest earlier symbol and the address is less than
4475 @var{max-offset}. The default is 0, which means to always print the
4476 symbolic form of an address, if any symbol precedes it.
4478 @item show print max-symbolic-offset
4479 @kindex show print max-symbolic-offset
4480 Ask how large the maximum offset is that @value{GDBN} prints in a
4484 @cindex wild pointer, interpreting
4485 @cindex pointer, finding referent
4486 If you have a pointer and you are not sure where it points, try
4487 @samp{set print symbol-filename on}. Then you can determine the name
4488 and source file location of the variable where it points, using
4489 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
4490 For example, here @value{GDBN} shows that a variable @code{ptt} points
4491 at another variable @code{t}, defined in @file{hi2.c}:
4494 (@value{GDBP}) set print symbol-filename on
4495 (@value{GDBP}) p/a ptt
4496 $4 = 0xe008 <t in hi2.c>
4500 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
4501 does not show the symbol name and filename of the referent, even with
4502 the appropriate @code{set print} options turned on.
4505 Other settings control how different kinds of objects are printed:
4508 @item set print array
4509 @itemx set print array on
4510 @kindex set print array
4511 Pretty-print arrays. This format is more convenient to read,
4512 but uses more space. The default is off.
4514 @item set print array off
4515 Return to compressed format for arrays.
4517 @item show print array
4518 @kindex show print array
4519 Show whether compressed or pretty format is selected for displaying
4522 @item set print elements @var{number-of-elements}
4523 @kindex set print elements
4524 If @value{GDBN} is printing a large array, it stops printing after it has
4525 printed the number of elements set by the @code{set print elements} command.
4526 This limit also applies to the display of strings.
4527 Setting the number of elements to zero means that the printing is unlimited.
4529 @item show print elements
4530 @kindex show print elements
4531 Display the number of elements of a large array that @value{GDBN} prints
4532 before losing patience.
4534 @item set print pretty on
4535 @kindex set print pretty
4536 Cause @value{GDBN} to print structures in an indented format with one member per
4552 @item set print pretty off
4553 Cause @value{GDBN} to print structures in a compact format, like this:
4557 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
4558 meat = 0x54 "Pork"@}
4563 This is the default format.
4565 @item show print pretty
4566 @kindex show print pretty
4567 Show which format @value{GDBN} is using to print structures.
4569 @item set print sevenbit-strings on
4570 @kindex set print sevenbit-strings
4571 Print using only seven-bit characters; if this option is set,
4572 @value{GDBN} displays any eight-bit characters (in strings or
4573 character values) using the notation @code{\}@var{nnn}. This setting is
4574 best if you are working in English (@sc{ascii}) and you use the
4575 high-order bit of characters as a marker or ``meta'' bit.
4577 @item set print sevenbit-strings off
4578 Print full eight-bit characters. This allows the use of more
4579 international character sets, and is the default.
4581 @item show print sevenbit-strings
4582 @kindex show print sevenbit-strings
4583 Show whether or not @value{GDBN} is printing only seven-bit characters.
4585 @item set print union on
4586 @kindex set print union
4587 Tell @value{GDBN} to print unions which are contained in structures. This is the
4590 @item set print union off
4591 Tell @value{GDBN} not to print unions which are contained in structures.
4593 @item show print union
4594 @kindex show print union
4595 Ask @value{GDBN} whether or not it will print unions which are contained in
4598 For example, given the declarations
4601 typedef enum @{Tree, Bug@} Species;
4602 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
4603 typedef enum @{Caterpillar, Cocoon, Butterfly@}
4614 struct thing foo = @{Tree, @{Acorn@}@};
4618 with @code{set print union on} in effect @samp{p foo} would print
4621 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
4625 and with @code{set print union off} in effect it would print
4628 $1 = @{it = Tree, form = @{...@}@}
4635 These settings are of interest when debugging C++ programs:
4638 @item set print demangle
4639 @itemx set print demangle on
4640 @kindex set print demangle
4641 Print C++ names in their source form rather than in the encoded
4642 (``mangled'') form passed to the assembler and linker for type-safe
4643 linkage. The default is @samp{on}.
4645 @item show print demangle
4646 @kindex show print demangle
4647 Show whether C++ names are printed in mangled or demangled form.
4649 @item set print asm-demangle
4650 @itemx set print asm-demangle on
4651 @kindex set print asm-demangle
4652 Print C++ names in their source form rather than their mangled form, even
4653 in assembler code printouts such as instruction disassemblies.
4656 @item show print asm-demangle
4657 @kindex show print asm-demangle
4658 Show whether C++ names in assembly listings are printed in mangled
4661 @item set demangle-style @var{style}
4662 @kindex set demangle-style
4663 @cindex C++ symbol decoding style
4664 @cindex symbol decoding style, C++
4665 Choose among several encoding schemes used by different compilers to
4666 represent C++ names. The choices for @var{style} are currently:
4670 Allow @value{GDBN} to choose a decoding style by inspecting your program.
4673 Decode based on the GNU C++ compiler (@code{g++}) encoding algorithm.
4676 Decode based on the Lucid C++ compiler (@code{lcc}) encoding algorithm.
4679 Decode using the algorithm in the @cite{C++ Annotated Reference Manual}.
4680 @strong{Warning:} this setting alone is not sufficient to allow
4681 debugging @code{cfront}-generated executables. @value{GDBN} would
4682 require further enhancement to permit that.
4685 @item show demangle-style
4686 @kindex show demangle-style
4687 Display the encoding style currently in use for decoding C++ symbols.
4689 @item set print object
4690 @itemx set print object on
4691 @kindex set print object
4692 When displaying a pointer to an object, identify the @emph{actual}
4693 (derived) type of the object rather than the @emph{declared} type, using
4694 the virtual function table.
4696 @item set print object off
4697 Display only the declared type of objects, without reference to the
4698 virtual function table. This is the default setting.
4700 @item show print object
4701 @kindex show print object
4702 Show whether actual, or declared, object types are displayed.
4704 @item set print vtbl
4705 @itemx set print vtbl on
4706 @kindex set print vtbl
4707 Pretty print C++ virtual function tables. The default is off.
4709 @item set print vtbl off
4710 Do not pretty print C++ virtual function tables.
4712 @item show print vtbl
4713 @kindex show print vtbl
4714 Show whether C++ virtual function tables are pretty printed, or not.
4719 @section Value history
4721 @cindex value history
4722 Values printed by the @code{print} command are saved in the @value{GDBN} @dfn{value
4723 history} so that you can refer to them in other expressions. Values are
4724 kept until the symbol table is re-read or discarded (for example with
4725 the @code{file} or @code{symbol-file} commands). When the symbol table
4726 changes, the value history is discarded, since the values may contain
4727 pointers back to the types defined in the symbol table.
4731 @cindex history number
4732 The values printed are given @dfn{history numbers} by which you can
4733 refer to them. These are successive integers starting with one.
4734 @code{print} shows you the history number assigned to a value by
4735 printing @samp{$@var{num} = } before the value; here @var{num} is the
4738 To refer to any previous value, use @samp{$} followed by the value's
4739 history number. The way @code{print} labels its output is designed to
4740 remind you of this. Just @code{$} refers to the most recent value in
4741 the history, and @code{$$} refers to the value before that.
4742 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
4743 is the value just prior to @code{$$}, @code{$$1} is equivalent to
4744 @code{$$}, and @code{$$0} is equivalent to @code{$}.
4746 For example, suppose you have just printed a pointer to a structure and
4747 want to see the contents of the structure. It suffices to type
4753 If you have a chain of structures where the component @code{next} points
4754 to the next one, you can print the contents of the next one with this:
4761 You can print successive links in the chain by repeating this
4762 command---which you can do by just typing @key{RET}.
4764 Note that the history records values, not expressions. If the value of
4765 @code{x} is 4 and you type these commands:
4773 then the value recorded in the value history by the @code{print} command
4774 remains 4 even though the value of @code{x} has changed.
4779 Print the last ten values in the value history, with their item numbers.
4780 This is like @samp{p@ $$9} repeated ten times, except that @code{show
4781 values} does not change the history.
4783 @item show values @var{n}
4784 Print ten history values centered on history item number @var{n}.
4787 Print ten history values just after the values last printed. If no more
4788 values are available, produces no display.
4791 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
4792 same effect as @samp{show values +}.
4794 @node Convenience Vars
4795 @section Convenience variables
4797 @cindex convenience variables
4798 @value{GDBN} provides @dfn{convenience variables} that you can use within
4799 @value{GDBN} to hold on to a value and refer to it later. These variables
4800 exist entirely within @value{GDBN}; they are not part of your program, and
4801 setting a convenience variable has no direct effect on further execution
4802 of your program. That is why you can use them freely.
4804 Convenience variables are prefixed with @samp{$}. Any name preceded by
4805 @samp{$} can be used for a convenience variable, unless it is one of
4806 the predefined machine-specific register names (@pxref{Registers}).
4807 (Value history references, in contrast, are @emph{numbers} preceded
4808 by @samp{$}. @xref{Value History, ,Value history}.)
4810 You can save a value in a convenience variable with an assignment
4811 expression, just as you would set a variable in your program.
4815 set $foo = *object_ptr
4819 would save in @code{$foo} the value contained in the object pointed to by
4822 Using a convenience variable for the first time creates it, but its
4823 value is @code{void} until you assign a new value. You can alter the
4824 value with another assignment at any time.
4826 Convenience variables have no fixed types. You can assign a convenience
4827 variable any type of value, including structures and arrays, even if
4828 that variable already has a value of a different type. The convenience
4829 variable, when used as an expression, has the type of its current value.
4832 @item show convenience
4833 @kindex show convenience
4834 Print a list of convenience variables used so far, and their values.
4835 Abbreviated @code{show con}.
4838 One of the ways to use a convenience variable is as a counter to be
4839 incremented or a pointer to be advanced. For example, to print
4840 a field from successive elements of an array of structures:
4844 print bar[$i++]->contents
4845 @i{@dots{} repeat that command by typing @key{RET}.}
4848 Some convenience variables are created automatically by @value{GDBN} and given
4849 values likely to be useful.
4854 The variable @code{$_} is automatically set by the @code{x} command to
4855 the last address examined (@pxref{Memory, ,Examining memory}). Other
4856 commands which provide a default address for @code{x} to examine also
4857 set @code{$_} to that address; these commands include @code{info line}
4858 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
4859 except when set by the @code{x} command, in which case it is a pointer
4860 to the type of @code{$__}.
4864 The variable @code{$__} is automatically set by the @code{x} command
4865 to the value found in the last address examined. Its type is chosen
4866 to match the format in which the data was printed.
4873 You can refer to machine register contents, in expressions, as variables
4874 with names starting with @samp{$}. The names of registers are different
4875 for each machine; use @code{info registers} to see the names used on
4879 @item info registers
4880 @kindex info registers
4881 Print the names and values of all registers except floating-point
4882 registers (in the selected stack frame).
4884 @item info all-registers
4885 @kindex info all-registers
4886 @cindex floating point registers
4887 Print the names and values of all registers, including floating-point
4890 @item info registers @var{regname} @dots{}
4891 Print the relativized value of each specified register @var{regname}.
4892 @var{regname} may be any register name valid on the machine you are using, with
4893 or without the initial @samp{$}.
4896 @value{GDBN} has four ``standard'' register names that are available (in
4897 expressions) on most machines---whenever they do not conflict with an
4898 architecture's canonical mnemonics for registers. The register names
4899 @code{$pc} and @code{$sp} are used for the program counter register and
4900 the stack pointer. @code{$fp} is used for a register that contains a
4901 pointer to the current stack frame, and @code{$ps} is used for a
4902 register that contains the processor status. For example,
4903 you could print the program counter in hex with
4910 or print the instruction to be executed next with
4917 or add four to the stack pointer@footnote{This is a way of removing
4918 one word from the stack, on machines where stacks grow downward in
4919 memory (most machines, nowadays). This assumes that the innermost
4920 stack frame is selected; setting @code{$sp} is not allowed when other
4921 stack frames are selected. To pop entire frames off the stack,
4922 regardless of machine architecture, use @code{return};
4923 @pxref{Returning, ,Returning from a function}.} with
4929 Whenever possible, these four standard register names are available on
4930 your machine even though the machine has different canonical mnemonics,
4931 so long as there is no conflict. The @code{info registers} command
4932 shows the canonical names. For example, on the SPARC, @code{info
4933 registers} displays the processor status register as @code{$psr} but you
4934 can also refer to it as @code{$ps}.
4936 @value{GDBN} always considers the contents of an ordinary register as an
4937 integer when the register is examined in this way. Some machines have
4938 special registers which can hold nothing but floating point; these
4939 registers are considered to have floating point values. There is no way
4940 to refer to the contents of an ordinary register as floating point value
4941 (although you can @emph{print} it as a floating point value with
4942 @samp{print/f $@var{regname}}).
4944 Some registers have distinct ``raw'' and ``virtual'' data formats. This
4945 means that the data format in which the register contents are saved by
4946 the operating system is not the same one that your program normally
4947 sees. For example, the registers of the 68881 floating point
4948 coprocessor are always saved in ``extended'' (raw) format, but all C
4949 programs expect to work with ``double'' (virtual) format. In such
4950 cases, @value{GDBN} normally works with the virtual format only (the format that
4951 makes sense for your program), but the @code{info registers} command
4952 prints the data in both formats.
4954 Normally, register values are relative to the selected stack frame
4955 (@pxref{Selection, ,Selecting a frame}). This means that you get the
4956 value that the register would contain if all stack frames farther in
4957 were exited and their saved registers restored. In order to see the
4958 true contents of hardware registers, you must select the innermost
4959 frame (with @samp{frame 0}).
4961 However, @value{GDBN} must deduce where registers are saved, from the machine
4962 code generated by your compiler. If some registers are not saved, or if
4963 @value{GDBN} is unable to locate the saved registers, the selected stack
4964 frame makes no difference.
4968 @item set rstack_high_address @var{address}
4969 @kindex set rstack_high_address
4970 @cindex AMD 29K register stack
4971 @cindex register stack, AMD29K
4972 On AMD 29000 family processors, registers are saved in a separate
4973 ``register stack''. There is no way for @value{GDBN} to determine the extent
4974 of this stack. Normally, @value{GDBN} just assumes that the stack is ``large
4975 enough''. This may result in @value{GDBN} referencing memory locations that
4976 do not exist. If necessary, you can get around this problem by
4977 specifying the ending address of the register stack with the @code{set
4978 rstack_high_address} command. The argument should be an address, which
4979 you probably want to precede with @samp{0x} to specify in
4982 @item show rstack_high_address
4983 @kindex show rstack_high_address
4984 Display the current limit of the register stack, on AMD 29000 family
4990 @node Floating Point Hardware
4991 @section Floating point hardware
4992 @cindex floating point
4994 @c FIXME! Really host, not target?
4995 Depending on the host machine architecture, @value{GDBN} may be able to give
4996 you more information about the status of the floating point hardware.
5001 Display hardware-dependent information about the floating
5002 point unit. The exact contents and layout vary depending on the
5003 floating point chip; on some platforms, @samp{info float} is not
5006 @c FIXME: this is a cop-out. Try to get examples, explanations. Only
5007 @c FIXME...supported currently on arm's and 386's. Mark properly with
5008 @c FIXME... m4 macros to isolate general statements from hardware-dep,
5009 @c FIXME... at that point.
5014 @chapter Using @value{GDBN} with Different Languages
5018 Although programming languages generally have common aspects, they are
5019 rarely expressed in the same manner. For instance, in ANSI C,
5020 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
5021 Modula-2, it is accomplished by @code{p^}. Values can also be
5022 represented (and displayed) differently. Hex numbers in C are written
5023 like @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
5026 @cindex working language
5027 Language-specific information is built into @value{GDBN} for some languages,
5028 allowing you to express operations like the above in your program's
5029 native language, and allowing @value{GDBN} to output values in a manner
5030 consistent with the syntax of your program's native language. The
5031 language you use to build expressions, called the @dfn{working
5032 language}, can be selected manually, or @value{GDBN} can set it
5036 * Setting:: Switching between source languages
5037 * Show:: Displaying the language
5039 * Checks:: Type and range checks
5042 * Support:: Supported languages
5046 @section Switching between source languages
5048 There are two ways to control the working language---either have @value{GDBN}
5049 set it automatically, or select it manually yourself. You can use the
5050 @code{set language} command for either purpose. On startup, @value{GDBN}
5051 defaults to setting the language automatically. The working language is
5052 used to determine how expressions you type are interpreted, how values
5055 In addition to the working language, every source file which
5056 @value{GDBN} knows about has its own working language. For some object
5057 file formats, the compiler might indicate which language a particular
5058 source file was in, but most of the time @value{GDBN} infers the
5059 language from the name of the file. The language of a source file
5060 controls whether C++ names are demangled---this way @code{backtrace} can
5061 show each frame appropriately for its own language. There is no way to
5062 set the language of a source file from within @value{GDBN}---the most
5063 common case where this is a problem is if you are using a program, such
5064 as @code{cfront} or @code{f2c}, which generates C but for which the real
5065 source code is in fact in another language. In that case, make the
5066 program use @code{#line} directives in its C output; that way
5067 @value{GDBN} will not only know the correct language, it will also be
5068 able to display the source code of the original program, not the
5072 * Filenames:: Filename extensions and languages.
5073 * Manually:: Setting the working language manually
5074 * Automatically:: Having @value{GDBN} infer the source language
5078 @subsection List of filename extensions and languages
5080 If a source file name ends in one of the following extensions, then
5081 @value{GDBN} infers that its language is the one indicated.
5086 Modula-2 source file
5107 Assembler source file. This actually behaves almost like C, but
5108 @value{GDBN} does not skip over function prologues when stepping.
5112 @subsection Setting the working language
5114 If you allow @value{GDBN} to set the language automatically,
5115 expressions are interpreted the same way in your debugging session and
5118 @kindex set language
5119 If you wish, you may set the language manually. To do this, issue the
5120 command @samp{set language @var{lang}}, where @var{lang} is the name of
5126 @code{c} or @code{modula-2}.
5128 For a list of the supported languages, type @samp{set language}.
5129 @c FIXME: rms: eventually this command should be "help set language".
5132 Setting the language manually prevents @value{GDBN} from updating the working
5133 language automatically. This can lead to confusion if you try
5134 to debug a program when the working language is not the same as the
5135 source language, when an expression is acceptable to both
5136 languages---but means different things. For instance, if the current
5137 source file were written in C, and @value{GDBN} was parsing Modula-2, a
5145 might not have the effect you intended. In C, this means to add
5146 @code{b} and @code{c} and place the result in @code{a}. The result
5147 printed would be the value of @code{a}. In Modula-2, this means to compare
5148 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
5152 @subsection Having @value{GDBN} infer the source language
5154 To have @value{GDBN} set the working language automatically, use
5155 @samp{set language local} or @samp{set language auto}. @value{GDBN}
5156 then infers the working language. That is, when your program stops in a
5157 frame (usually by encountering a breakpoint), @value{GDBN} sets the
5158 working language to the language recorded for the function in that
5159 frame. If the language for a frame is unknown (that is, if the function
5160 or block corresponding to the frame was defined in a source file that
5161 does not have a recognized extension), the current working language is
5162 not changed, and @value{GDBN} issues a warning.
5164 This may not seem necessary for most programs, which are written
5165 entirely in one source language. However, program modules and libraries
5166 written in one source language can be used by a main program written in
5167 a different source language. Using @samp{set language auto} in this
5168 case frees you from having to set the working language manually.
5171 @section Displaying the language
5173 The following commands help you find out which language is the
5174 working language, and also what language source files were written in.
5176 @kindex show language
5181 Display the current working language. This is the
5182 language you can use with commands such as @code{print} to
5183 build and compute expressions that may involve variables in your program.
5186 Among the other information listed here (@pxref{Frame Info, ,Information
5187 about a frame}) is the source language for this frame. This
5188 language becomes the working language if you use an
5189 identifier from this frame.
5192 Among the other information listed here (@pxref{Symbols, ,Examining the
5193 Symbol Table}) is the source language of this source file.
5198 @section Type and range checking
5201 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
5202 checking are included, but they do not yet have any effect. This
5203 section documents the intended facilities.
5205 @c FIXME remove warning when type/range code added
5207 Some languages are designed to guard you against making seemingly common
5208 errors through a series of compile- and run-time checks. These include
5209 checking the type of arguments to functions and operators, and making
5210 sure mathematical overflows are caught at run time. Checks such as
5211 these help to ensure a program's correctness once it has been compiled
5212 by eliminating type mismatches, and providing active checks for range
5213 errors when your program is running.
5215 @value{GDBN} can check for conditions like the above if you wish.
5216 Although @value{GDBN} does not check the statements in your program, it
5217 can check expressions entered directly into @value{GDBN} for evaluation via
5218 the @code{print} command, for example. As with the working language,
5219 @value{GDBN} can also decide whether or not to check automatically based on
5220 your program's source language. @xref{Support, ,Supported languages},
5221 for the default settings of supported languages.
5224 * Type Checking:: An overview of type checking
5225 * Range Checking:: An overview of range checking
5228 @cindex type checking
5229 @cindex checks, type
5231 @subsection An overview of type checking
5233 Some languages, such as Modula-2, are strongly typed, meaning that the
5234 arguments to operators and functions have to be of the correct type,
5235 otherwise an error occurs. These checks prevent type mismatch
5236 errors from ever causing any run-time problems. For example,
5244 The second example fails because the @code{CARDINAL} 1 is not
5245 type-compatible with the @code{REAL} 2.3.
5247 For expressions you use in @value{GDBN} commands, you can tell the @value{GDBN}
5248 type checker to skip checking; to treat any mismatches as errors and
5249 abandon the expression; or only issue warnings when type mismatches
5250 occur, but evaluate the expression anyway. When you choose the last of
5251 these, @value{GDBN} evaluates expressions like the second example above, but
5252 also issues a warning.
5254 Even though you may turn type checking off, other type-based reasons may
5255 prevent @value{GDBN} from evaluating an expression. For instance, @value{GDBN} does not
5256 know how to add an @code{int} and a @code{struct foo}. These particular
5257 type errors have nothing to do with the language in use, and usually
5258 arise from expressions, such as the one described above, which make
5259 little sense to evaluate anyway.
5261 Each language defines to what degree it is strict about type. For
5262 instance, both Modula-2 and C require the arguments to arithmetical
5263 operators to be numbers. In C, enumerated types and pointers can be
5264 represented as numbers, so that they are valid arguments to mathematical
5265 operators. @xref{Support, ,Supported languages}, for further
5266 details on specific languages.
5268 @value{GDBN} provides some additional commands for controlling the type checker:
5271 @kindex set check type
5272 @kindex show check type
5274 @item set check type auto
5275 Set type checking on or off based on the current working language.
5276 @xref{Support, ,Supported languages}, for the default settings for
5279 @item set check type on
5280 @itemx set check type off
5281 Set type checking on or off, overriding the default setting for the
5282 current working language. Issue a warning if the setting does not
5283 match the language default. If any type mismatches occur in
5284 evaluating an expression while typechecking is on, @value{GDBN} prints a
5285 message and aborts evaluation of the expression.
5287 @item set check type warn
5288 Cause the type checker to issue warnings, but to always attempt to
5289 evaluate the expression. Evaluating the expression may still
5290 be impossible for other reasons. For example, @value{GDBN} cannot add
5291 numbers and structures.
5294 Show the current setting of the type checker, and whether or not @value{GDBN} is
5295 setting it automatically.
5298 @cindex range checking
5299 @cindex checks, range
5300 @node Range Checking
5301 @subsection An overview of range checking
5303 In some languages (such as Modula-2), it is an error to exceed the
5304 bounds of a type; this is enforced with run-time checks. Such range
5305 checking is meant to ensure program correctness by making sure
5306 computations do not overflow, or indices on an array element access do
5307 not exceed the bounds of the array.
5309 For expressions you use in @value{GDBN} commands, you can tell
5310 @value{GDBN} to treat range errors in one of three ways: ignore them,
5311 always treat them as errors and abandon the expression, or issue
5312 warnings but evaluate the expression anyway.
5314 A range error can result from numerical overflow, from exceeding an
5315 array index bound, or when you type a constant that is not a member
5316 of any type. Some languages, however, do not treat overflows as an
5317 error. In many implementations of C, mathematical overflow causes the
5318 result to ``wrap around'' to lower values---for example, if @var{m} is
5319 the largest integer value, and @var{s} is the smallest, then
5322 @var{m} + 1 @result{} @var{s}
5325 This, too, is specific to individual languages, and in some cases
5326 specific to individual compilers or machines. @xref{Support, ,
5327 Supported languages}, for further details on specific languages.
5329 @value{GDBN} provides some additional commands for controlling the range checker:
5332 @kindex set check range
5333 @kindex show check range
5335 @item set check range auto
5336 Set range checking on or off based on the current working language.
5337 @xref{Support, ,Supported languages}, for the default settings for
5340 @item set check range on
5341 @itemx set check range off
5342 Set range checking on or off, overriding the default setting for the
5343 current working language. A warning is issued if the setting does not
5344 match the language default. If a range error occurs, then a message
5345 is printed and evaluation of the expression is aborted.
5347 @item set check range warn
5348 Output messages when the @value{GDBN} range checker detects a range error,
5349 but attempt to evaluate the expression anyway. Evaluating the
5350 expression may still be impossible for other reasons, such as accessing
5351 memory that the process does not own (a typical example from many Unix
5355 Show the current setting of the range checker, and whether or not it is
5356 being set automatically by @value{GDBN}.
5361 @section Supported languages
5364 @value{GDBN} 4 supports C, C++, and Modula-2.
5367 @value{GDBN} 4 supports C, and C++.
5369 Some @value{GDBN} features may be used in expressions regardless of the
5370 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
5371 and the @samp{@{type@}addr} construct (@pxref{Expressions,
5372 ,Expressions}) can be used with the constructs of any supported
5375 The following sections detail to what degree each source language is
5376 supported by @value{GDBN}. These sections are not meant to be language
5377 tutorials or references, but serve only as a reference guide to what the
5378 @value{GDBN} expression parser accepts, and what input and output
5379 formats should look like for different languages. There are many good
5380 books written on each of these languages; please look to these for a
5381 language reference or tutorial.
5386 * Modula-2:: Modula-2
5390 @subsection C and C++
5392 @cindex expressions in C or C++
5394 Since C and C++ are so closely related, many features of @value{GDBN} apply
5395 to both languages. Whenever this is the case, we discuss both languages
5399 @c Cancel this below, under same condition, at end of this chapter!
5406 The C++ debugging facilities are jointly implemented by the GNU C++
5407 compiler and @value{GDBN}. Therefore, to debug your C++ code
5408 effectively, you must compile your C++ programs with the GNU C++
5409 compiler, @code{g++}.
5411 For best results when debugging C++ programs, use the stabs debugging
5412 format. You can select that format explicitly with the @code{g++}
5413 command-line options @samp{-gstabs} or @samp{-gstabs+}. See
5414 @ref{Debugging Options,,Options for Debugging Your Program or GNU CC,
5415 gcc.info, Using GNU CC}, for more information.
5419 @chapter C Language Support
5421 @cindex expressions in C
5423 Information specific to the C language is built into @value{GDBN} so that you
5424 can use C expressions while degugging. This also permits @value{GDBN} to
5425 output values in a manner consistent with C conventions.
5428 * C Operators:: C operators
5429 * C Constants:: C constants
5430 * Debugging C:: @value{GDBN} and C
5435 * C Operators:: C and C++ operators
5436 * C Constants:: C and C++ constants
5437 * Cplus expressions:: C++ expressions
5438 * C Defaults:: Default settings for C and C++
5440 * C Checks:: C and C++ type and range checks
5443 * Debugging C:: @value{GDBN} and C
5444 * Debugging C plus plus:: Special features for C++
5449 @cindex C and C++ operators
5451 @subsubsection C and C++ operators
5456 @section C operators
5459 Operators must be defined on values of specific types. For instance,
5460 @code{+} is defined on numbers, but not on structures. Operators are
5461 often defined on groups of types.
5464 For the purposes of C and C++, the following definitions hold:
5469 @emph{Integral types} include @code{int} with any of its storage-class
5470 specifiers; @code{char}; and @code{enum}.
5473 @emph{Floating-point types} include @code{float} and @code{double}.
5476 @emph{Pointer types} include all types defined as @code{(@var{type}
5480 @emph{Scalar types} include all of the above.
5484 The following operators are supported. They are listed here
5485 in order of increasing precedence:
5489 The comma or sequencing operator. Expressions in a comma-separated list
5490 are evaluated from left to right, with the result of the entire
5491 expression being the last expression evaluated.
5494 Assignment. The value of an assignment expression is the value
5495 assigned. Defined on scalar types.
5498 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
5499 and translated to @w{@code{@var{a} = @var{a op b}}}.
5500 @w{@code{@var{op}=}} and @code{=} have the same precendence.
5501 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
5502 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
5505 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
5506 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
5510 Logical @sc{or}. Defined on integral types.
5513 Logical @sc{and}. Defined on integral types.
5516 Bitwise @sc{or}. Defined on integral types.
5519 Bitwise exclusive-@sc{or}. Defined on integral types.
5522 Bitwise @sc{and}. Defined on integral types.
5525 Equality and inequality. Defined on scalar types. The value of these
5526 expressions is 0 for false and non-zero for true.
5528 @item <@r{, }>@r{, }<=@r{, }>=
5529 Less than, greater than, less than or equal, greater than or equal.
5530 Defined on scalar types. The value of these expressions is 0 for false
5531 and non-zero for true.
5534 left shift, and right shift. Defined on integral types.
5537 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5540 Addition and subtraction. Defined on integral types, floating-point types and
5543 @item *@r{, }/@r{, }%
5544 Multiplication, division, and modulus. Multiplication and division are
5545 defined on integral and floating-point types. Modulus is defined on
5549 Increment and decrement. When appearing before a variable, the
5550 operation is performed before the variable is used in an expression;
5551 when appearing after it, the variable's value is used before the
5552 operation takes place.
5555 Pointer dereferencing. Defined on pointer types. Same precedence as
5559 Address operator. Defined on variables. Same precedence as @code{++}.
5562 For debugging C++, @value{GDBN} implements a use of @samp{&} beyond what is
5563 allowed in the C++ language itself: you can use @samp{&(&@var{ref})}
5564 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
5565 where a C++ reference variable (declared with @samp{&@var{ref}}) is
5570 Negative. Defined on integral and floating-point types. Same
5571 precedence as @code{++}.
5574 Logical negation. Defined on integral types. Same precedence as
5578 Bitwise complement operator. Defined on integral types. Same precedence as
5583 Structure member, and pointer-to-structure member. For convenience,
5584 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
5585 pointer based on the stored type information.
5586 Defined on @code{struct} and @code{union} data.
5589 Array indexing. @code{@var{a}[@var{i}]} is defined as
5590 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
5593 Function parameter list. Same precedence as @code{->}.
5597 C++ scope resolution operator. Defined on
5598 @code{struct}, @code{union}, and @code{class} types.
5606 represent the @value{GDBN} scope operator (@pxref{Expressions,
5609 Same precedence as @code{::}, above.
5614 @cindex C and C++ constants
5616 @subsubsection C and C++ constants
5618 @value{GDBN} allows you to express the constants of C and C++ in the
5624 @section C constants
5626 @value{GDBN} allows you to express the constants of C in the
5632 Integer constants are a sequence of digits. Octal constants are
5633 specified by a leading @samp{0} (ie. zero), and hexadecimal constants by
5634 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
5635 @samp{l}, specifying that the constant should be treated as a
5639 Floating point constants are a sequence of digits, followed by a decimal
5640 point, followed by a sequence of digits, and optionally followed by an
5641 exponent. An exponent is of the form:
5642 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
5643 sequence of digits. The @samp{+} is optional for positive exponents.
5646 Enumerated constants consist of enumerated identifiers, or their
5647 integral equivalents.
5650 Character constants are a single character surrounded by single quotes
5651 (@code{'}), or a number---the ordinal value of the corresponding character
5652 (usually its @sc{ASCII} value). Within quotes, the single character may
5653 be represented by a letter or by @dfn{escape sequences}, which are of
5654 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
5655 of the character's ordinal value; or of the form @samp{\@var{x}}, where
5656 @samp{@var{x}} is a predefined special character---for example,
5657 @samp{\n} for newline.
5660 String constants are a sequence of character constants surrounded
5661 by double quotes (@code{"}).
5664 Pointer constants are an integral value. You can also write pointers
5665 to constants using the C operator @samp{&}.
5668 Array constants are comma-separated lists surrounded by braces @samp{@{}
5669 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
5670 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
5671 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
5675 @node Cplus expressions
5676 @subsubsection C++ expressions
5678 @cindex expressions in C++
5679 @value{GDBN} expression handling has a number of extensions to
5680 interpret a significant subset of C++ expressions.
5682 @cindex C++ support, not in @sc{coff}
5683 @cindex @sc{coff} versus C++
5684 @cindex C++ and object formats
5685 @cindex object formats and C++
5686 @cindex a.out and C++
5687 @cindex @sc{ecoff} and C++
5688 @cindex @sc{xcoff} and C++
5689 @cindex @sc{elf}/stabs and C++
5690 @cindex @sc{elf}/@sc{dwarf} and C++
5691 @c FIXME!! GDB may eventually be able to debug C++ using DWARF; check
5692 @c periodically whether this has happened...
5694 @emph{Warning:} @value{GDBN} can only debug C++ code if you compile with
5695 the GNU C++ compiler. Moreover, C++ debugging depends on the use of
5696 additional debugging information in the symbol table, and thus requires
5697 special support. @value{GDBN} has this support @emph{only} with the
5698 stabs debug format. In particular, if your compiler generates a.out,
5699 MIPS @sc{ecoff}, RS/6000 @sc{xcoff}, or @sc{elf} with stabs extensions
5700 to the symbol table, these facilities are all available. (With GNU CC,
5701 you can use the @samp{-gstabs} option to request stabs debugging
5702 extensions explicitly.) Where the object code format is standard
5703 @sc{coff} or @sc{dwarf} in @sc{elf}, on the other hand, most of the C++
5704 support in @value{GDBN} does @emph{not} work.
5709 @cindex member functions
5711 Member function calls are allowed; you can use expressions like
5714 count = aml->GetOriginal(x, y)
5718 @cindex namespace in C++
5720 While a member function is active (in the selected stack frame), your
5721 expressions have the same namespace available as the member function;
5722 that is, @value{GDBN} allows implicit references to the class instance
5723 pointer @code{this} following the same rules as C++.
5725 @cindex call overloaded functions
5726 @cindex type conversions in C++
5728 You can call overloaded functions; @value{GDBN} resolves the function
5729 call to the right definition, with one restriction---you must use
5730 arguments of the type required by the function that you want to call.
5731 @value{GDBN} does not perform conversions requiring constructors or
5732 user-defined type operators.
5734 @cindex reference declarations
5736 @value{GDBN} understands variables declared as C++ references; you can use them in
5737 expressions just as you do in C++ source---they are automatically
5740 In the parameter list shown when @value{GDBN} displays a frame, the values of
5741 reference variables are not displayed (unlike other variables); this
5742 avoids clutter, since references are often used for large structures.
5743 The @emph{address} of a reference variable is always shown, unless
5744 you have specified @samp{set print address off}.
5747 @value{GDBN} supports the C++ name resolution operator @code{::}---your
5748 expressions can use it just as expressions in your program do. Since
5749 one scope may be defined in another, you can use @code{::} repeatedly if
5750 necessary, for example in an expression like
5751 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
5752 resolving name scope by reference to source files, in both C and C++
5753 debugging (@pxref{Variables, ,Program variables}).
5757 @subsubsection C and C++ defaults
5758 @cindex C and C++ defaults
5760 If you allow @value{GDBN} to set type and range checking automatically, they
5761 both default to @code{off} whenever the working language changes to
5762 C or C++. This happens regardless of whether you, or @value{GDBN},
5763 selected the working language.
5765 If you allow @value{GDBN} to set the language automatically, it sets the
5766 working language to C or C++ on entering code compiled from a source file
5767 whose name ends with @file{.c}, @file{.C}, or @file{.cc}.
5768 @xref{Automatically, ,Having @value{GDBN} infer the source language}, for
5772 @c Type checking is (a) primarily motivated by Modula-2, and (b)
5773 @c unimplemented. If (b) changes, it might make sense to let this node
5774 @c appear even if Mod-2 does not, but meanwhile ignore it. pesch 16jul93.
5776 @subsubsection C and C++ type and range checks
5777 @cindex C and C++ checks
5779 By default, when @value{GDBN} parses C or C++ expressions, type checking
5780 is not used. However, if you turn type checking on, @value{GDBN}
5781 considers two variables type equivalent if:
5785 The two variables are structured and have the same structure, union, or
5789 Two two variables have the same type name, or types that have been
5790 declared equivalent through @code{typedef}.
5793 @c leaving this out because neither J Gilmore nor R Pesch understand it.
5796 The two @code{struct}, @code{union}, or @code{enum} variables are
5797 declared in the same declaration. (Note: this may not be true for all C
5802 Range checking, if turned on, is done on mathematical operations. Array
5803 indices are not checked, since they are often used to index a pointer
5804 that is not itself an array.
5810 @subsubsection @value{GDBN} and C
5814 @section @value{GDBN} and C
5817 The @code{set print union} and @code{show print union} commands apply to
5818 the @code{union} type. When set to @samp{on}, any @code{union} that is
5819 inside a @code{struct}
5824 Otherwise, it appears as @samp{@{...@}}.
5826 The @code{@@} operator aids in the debugging of dynamic arrays, formed
5827 with pointers and a memory allocation function. @xref{Expressions,
5831 @node Debugging C plus plus
5832 @subsubsection @value{GDBN} features for C++
5834 @cindex commands for C++
5835 Some @value{GDBN} commands are particularly useful with C++, and some are
5836 designed specifically for use with C++. Here is a summary:
5839 @cindex break in overloaded functions
5840 @item @r{breakpoint menus}
5841 When you want a breakpoint in a function whose name is overloaded,
5842 @value{GDBN} breakpoint menus help you specify which function definition
5843 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
5845 @cindex overloading in C++
5846 @item rbreak @var{regex}
5847 Setting breakpoints using regular expressions is helpful for setting
5848 breakpoints on overloaded functions that are not members of any special
5850 @xref{Set Breaks, ,Setting breakpoints}.
5852 @cindex C++ exception handling
5853 @item catch @var{exceptions}
5855 Debug C++ exception handling using these commands. @xref{Exception
5856 Handling, ,Breakpoints and exceptions}.
5859 @item ptype @var{typename}
5860 Print inheritance relationships as well as other information for type
5862 @xref{Symbols, ,Examining the Symbol Table}.
5864 @cindex C++ symbol display
5865 @item set print demangle
5866 @itemx show print demangle
5867 @itemx set print asm-demangle
5868 @itemx show print asm-demangle
5869 Control whether C++ symbols display in their source form, both when
5870 displaying code as C++ source and when displaying disassemblies.
5871 @xref{Print Settings, ,Print settings}.
5873 @item set print object
5874 @itemx show print object
5875 Choose whether to print derived (actual) or declared types of objects.
5876 @xref{Print Settings, ,Print settings}.
5878 @item set print vtbl
5879 @itemx show print vtbl
5880 Control the format for printing virtual function tables.
5881 @xref{Print Settings, ,Print settings}.
5883 @item @r{Overloaded symbol names}
5884 You can specify a particular definition of an overloaded symbol, using
5885 the same notation that is used to declare such symbols in C++: type
5886 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
5887 also use the @value{GDBN} command-line word completion facilities to list the
5888 available choices, or to finish the type list for you.
5889 @xref{Completion,, Command completion}, for details on how to do this.
5892 @c cancels "raisesections" under same conditions near bgn of chapter
5898 @subsection Modula-2
5901 The extensions made to @value{GDBN} to support Modula-2 only support
5902 output from the GNU Modula-2 compiler (which is currently being
5903 developed). Other Modula-2 compilers are not currently supported, and
5904 attempting to debug executables produced by them is most likely
5905 to give an error as @value{GDBN} reads in the executable's symbol
5908 @cindex expressions in Modula-2
5910 * M2 Operators:: Built-in operators
5911 * Built-In Func/Proc:: Built-in functions and procedures
5912 * M2 Constants:: Modula-2 constants
5913 * M2 Defaults:: Default settings for Modula-2
5914 * Deviations:: Deviations from standard Modula-2
5915 * M2 Checks:: Modula-2 type and range checks
5916 * M2 Scope:: The scope operators @code{::} and @code{.}
5917 * GDB/M2:: @value{GDBN} and Modula-2
5921 @subsubsection Operators
5922 @cindex Modula-2 operators
5924 Operators must be defined on values of specific types. For instance,
5925 @code{+} is defined on numbers, but not on structures. Operators are
5926 often defined on groups of types. For the purposes of Modula-2, the
5927 following definitions hold:
5932 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
5936 @emph{Character types} consist of @code{CHAR} and its subranges.
5939 @emph{Floating-point types} consist of @code{REAL}.
5942 @emph{Pointer types} consist of anything declared as @code{POINTER TO
5946 @emph{Scalar types} consist of all of the above.
5949 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
5952 @emph{Boolean types} consist of @code{BOOLEAN}.
5956 The following operators are supported, and appear in order of
5957 increasing precedence:
5961 Function argument or array index separator.
5964 Assignment. The value of @var{var} @code{:=} @var{value} is
5968 Less than, greater than on integral, floating-point, or enumerated
5972 Less than, greater than, less than or equal to, greater than or equal to
5973 on integral, floating-point and enumerated types, or set inclusion on
5974 set types. Same precedence as @code{<}.
5976 @item =@r{, }<>@r{, }#
5977 Equality and two ways of expressing inequality, valid on scalar types.
5978 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
5979 available for inequality, since @code{#} conflicts with the script
5983 Set membership. Defined on set types and the types of their members.
5984 Same precedence as @code{<}.
5987 Boolean disjunction. Defined on boolean types.
5990 Boolean conjuction. Defined on boolean types.
5993 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5996 Addition and subtraction on integral and floating-point types, or union
5997 and difference on set types.
6000 Multiplication on integral and floating-point types, or set intersection
6004 Division on floating-point types, or symmetric set difference on set
6005 types. Same precedence as @code{*}.
6008 Integer division and remainder. Defined on integral types. Same
6009 precedence as @code{*}.
6012 Negative. Defined on @code{INTEGER} and @code{REAL} data.
6015 Pointer dereferencing. Defined on pointer types.
6018 Boolean negation. Defined on boolean types. Same precedence as
6022 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
6023 precedence as @code{^}.
6026 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
6029 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
6033 @value{GDBN} and Modula-2 scope operators.
6037 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
6038 treats the use of the operator @code{IN}, or the use of operators
6039 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
6040 @code{<=}, and @code{>=} on sets as an error.
6043 @cindex Modula-2 built-ins
6044 @node Built-In Func/Proc
6045 @subsubsection Built-in functions and procedures
6047 Modula-2 also makes available several built-in procedures and functions.
6048 In describing these, the following metavariables are used:
6053 represents an @code{ARRAY} variable.
6056 represents a @code{CHAR} constant or variable.
6059 represents a variable or constant of integral type.
6062 represents an identifier that belongs to a set. Generally used in the
6063 same function with the metavariable @var{s}. The type of @var{s} should
6064 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
6067 represents a variable or constant of integral or floating-point type.
6070 represents a variable or constant of floating-point type.
6076 represents a variable.
6079 represents a variable or constant of one of many types. See the
6080 explanation of the function for details.
6083 All Modula-2 built-in procedures also return a result, described below.
6087 Returns the absolute value of @var{n}.
6090 If @var{c} is a lower case letter, it returns its upper case
6091 equivalent, otherwise it returns its argument
6094 Returns the character whose ordinal value is @var{i}.
6097 Decrements the value in the variable @var{v}. Returns the new value.
6099 @item DEC(@var{v},@var{i})
6100 Decrements the value in the variable @var{v} by @var{i}. Returns the
6103 @item EXCL(@var{m},@var{s})
6104 Removes the element @var{m} from the set @var{s}. Returns the new
6107 @item FLOAT(@var{i})
6108 Returns the floating point equivalent of the integer @var{i}.
6111 Returns the index of the last member of @var{a}.
6114 Increments the value in the variable @var{v}. Returns the new value.
6116 @item INC(@var{v},@var{i})
6117 Increments the value in the variable @var{v} by @var{i}. Returns the
6120 @item INCL(@var{m},@var{s})
6121 Adds the element @var{m} to the set @var{s} if it is not already
6122 there. Returns the new set.
6125 Returns the maximum value of the type @var{t}.
6128 Returns the minimum value of the type @var{t}.
6131 Returns boolean TRUE if @var{i} is an odd number.
6134 Returns the ordinal value of its argument. For example, the ordinal
6135 value of a character is its ASCII value (on machines supporting the
6136 ASCII character set). @var{x} must be of an ordered type, which include
6137 integral, character and enumerated types.
6140 Returns the size of its argument. @var{x} can be a variable or a type.
6142 @item TRUNC(@var{r})
6143 Returns the integral part of @var{r}.
6145 @item VAL(@var{t},@var{i})
6146 Returns the member of the type @var{t} whose ordinal value is @var{i}.
6150 @emph{Warning:} Sets and their operations are not yet supported, so
6151 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
6155 @cindex Modula-2 constants
6157 @subsubsection Constants
6159 @value{GDBN} allows you to express the constants of Modula-2 in the following
6165 Integer constants are simply a sequence of digits. When used in an
6166 expression, a constant is interpreted to be type-compatible with the
6167 rest of the expression. Hexadecimal integers are specified by a
6168 trailing @samp{H}, and octal integers by a trailing @samp{B}.
6171 Floating point constants appear as a sequence of digits, followed by a
6172 decimal point and another sequence of digits. An optional exponent can
6173 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
6174 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
6175 digits of the floating point constant must be valid decimal (base 10)
6179 Character constants consist of a single character enclosed by a pair of
6180 like quotes, either single (@code{'}) or double (@code{"}). They may
6181 also be expressed by their ordinal value (their ASCII value, usually)
6182 followed by a @samp{C}.
6185 String constants consist of a sequence of characters enclosed by a
6186 pair of like quotes, either single (@code{'}) or double (@code{"}).
6187 Escape sequences in the style of C are also allowed. @xref{C
6188 Constants, ,C and C++ constants}, for a brief explanation of escape
6192 Enumerated constants consist of an enumerated identifier.
6195 Boolean constants consist of the identifiers @code{TRUE} and
6199 Pointer constants consist of integral values only.
6202 Set constants are not yet supported.
6206 @subsubsection Modula-2 defaults
6207 @cindex Modula-2 defaults
6209 If type and range checking are set automatically by @value{GDBN}, they
6210 both default to @code{on} whenever the working language changes to
6211 Modula-2. This happens regardless of whether you, or @value{GDBN},
6212 selected the working language.
6214 If you allow @value{GDBN} to set the language automatically, then entering
6215 code compiled from a file whose name ends with @file{.mod} sets the
6216 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
6217 the language automatically}, for further details.
6220 @subsubsection Deviations from standard Modula-2
6221 @cindex Modula-2, deviations from
6223 A few changes have been made to make Modula-2 programs easier to debug.
6224 This is done primarily via loosening its type strictness:
6228 Unlike in standard Modula-2, pointer constants can be formed by
6229 integers. This allows you to modify pointer variables during
6230 debugging. (In standard Modula-2, the actual address contained in a
6231 pointer variable is hidden from you; it can only be modified
6232 through direct assignment to another pointer variable or expression that
6233 returned a pointer.)
6236 C escape sequences can be used in strings and characters to represent
6237 non-printable characters. @value{GDBN} prints out strings with these
6238 escape sequences embedded. Single non-printable characters are
6239 printed using the @samp{CHR(@var{nnn})} format.
6242 The assignment operator (@code{:=}) returns the value of its right-hand
6246 All built-in procedures both modify @emph{and} return their argument.
6250 @subsubsection Modula-2 type and range checks
6251 @cindex Modula-2 checks
6254 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
6257 @c FIXME remove warning when type/range checks added
6259 @value{GDBN} considers two Modula-2 variables type equivalent if:
6263 They are of types that have been declared equivalent via a @code{TYPE
6264 @var{t1} = @var{t2}} statement
6267 They have been declared on the same line. (Note: This is true of the
6268 GNU Modula-2 compiler, but it may not be true of other compilers.)
6271 As long as type checking is enabled, any attempt to combine variables
6272 whose types are not equivalent is an error.
6274 Range checking is done on all mathematical operations, assignment, array
6275 index bounds, and all built-in functions and procedures.
6278 @subsubsection The scope operators @code{::} and @code{.}
6281 @cindex colon, doubled as scope operator
6284 @c Info cannot handle :: but TeX can.
6290 There are a few subtle differences between the Modula-2 scope operator
6291 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
6296 @var{module} . @var{id}
6297 @var{scope} :: @var{id}
6301 where @var{scope} is the name of a module or a procedure,
6302 @var{module} the name of a module, and @var{id} is any declared
6303 identifier within your program, except another module.
6305 Using the @code{::} operator makes @value{GDBN} search the scope
6306 specified by @var{scope} for the identifier @var{id}. If it is not
6307 found in the specified scope, then @value{GDBN} searches all scopes
6308 enclosing the one specified by @var{scope}.
6310 Using the @code{.} operator makes @value{GDBN} search the current scope for
6311 the identifier specified by @var{id} that was imported from the
6312 definition module specified by @var{module}. With this operator, it is
6313 an error if the identifier @var{id} was not imported from definition
6314 module @var{module}, or if @var{id} is not an identifier in
6318 @subsubsection @value{GDBN} and Modula-2
6320 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
6321 Five subcommands of @code{set print} and @code{show print} apply
6322 specifically to C and C++: @samp{vtbl}, @samp{demangle},
6323 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
6324 apply to C++, and the last to the C @code{union} type, which has no direct
6325 analogue in Modula-2.
6327 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
6328 while using any language, is not useful with Modula-2. Its
6329 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
6330 created in Modula-2 as they can in C or C++. However, because an
6331 address can be specified by an integral constant, the construct
6332 @samp{@{@var{type}@}@var{adrexp}} is still useful. (@pxref{Expressions, ,Expressions})
6334 @cindex @code{#} in Modula-2
6335 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
6336 interpreted as the beginning of a comment. Use @code{<>} instead.
6342 @chapter Examining the Symbol Table
6344 The commands described in this section allow you to inquire about the
6345 symbols (names of variables, functions and types) defined in your
6346 program. This information is inherent in the text of your program and
6347 does not change as your program executes. @value{GDBN} finds it in your
6348 program's symbol table, in the file indicated when you started @value{GDBN}
6349 (@pxref{File Options, ,Choosing files}), or by one of the
6350 file-management commands (@pxref{Files, ,Commands to specify files}).
6352 @c FIXME! This might be intentionally specific to C and C++; if so, move
6353 @c to someplace in C section of lang chapter.
6354 @cindex symbol names
6355 @cindex names of symbols
6356 @cindex quoting names
6357 Occasionally, you may need to refer to symbols that contain unusual
6358 characters, which @value{GDBN} ordinarily treats as word delimiters. The
6359 most frequent case is in referring to static variables in other
6360 source files (@pxref{Variables,,Program variables}). File names
6361 are recorded in object files as debugging symbols, but @value{GDBN} would
6362 ordinarily parse a typical file name, like @file{foo.c}, as the three words
6363 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
6364 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
6371 looks up the value of @code{x} in the scope of the file @file{foo.c}.
6374 @item info address @var{symbol}
6375 @kindex info address
6376 Describe where the data for @var{symbol} is stored. For a register
6377 variable, this says which register it is kept in. For a non-register
6378 local variable, this prints the stack-frame offset at which the variable
6381 Note the contrast with @samp{print &@var{symbol}}, which does not work
6382 at all for a register variable, and for a stack local variable prints
6383 the exact address of the current instantiation of the variable.
6385 @item whatis @var{exp}
6387 Print the data type of expression @var{exp}. @var{exp} is not
6388 actually evaluated, and any side-effecting operations (such as
6389 assignments or function calls) inside it do not take place.
6390 @xref{Expressions, ,Expressions}.
6393 Print the data type of @code{$}, the last value in the value history.
6395 @item ptype @var{typename}
6397 Print a description of data type @var{typename}. @var{typename} may be
6398 the name of a type, or for C code it may have the form
6400 @samp{class @var{class-name}},
6402 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
6403 @samp{enum @var{enum-tag}}.
6405 @item ptype @var{exp}
6407 Print a description of the type of expression @var{exp}. @code{ptype}
6408 differs from @code{whatis} by printing a detailed description, instead
6409 of just the name of the type.
6411 For example, for this variable declaration:
6414 struct complex @{double real; double imag;@} v;
6418 the two commands give this output:
6422 (@value{GDBP}) whatis v
6423 type = struct complex
6424 (@value{GDBP}) ptype v
6425 type = struct complex @{
6433 As with @code{whatis}, using @code{ptype} without an argument refers to
6434 the type of @code{$}, the last value in the value history.
6436 @item info types @var{regexp}
6439 Print a brief description of all types whose name matches @var{regexp}
6440 (or all types in your program, if you supply no argument). Each
6441 complete typename is matched as though it were a complete line; thus,
6442 @samp{i type value} gives information on all types in your program whose
6443 name includes the string @code{value}, but @samp{i type ^value$} gives
6444 information only on types whose complete name is @code{value}.
6446 This command differs from @code{ptype} in two ways: first, like
6447 @code{whatis}, it does not print a detailed description; second, it
6448 lists all source files where a type is defined.
6452 Show the name of the current source file---that is, the source file for
6453 the function containing the current point of execution---and the language
6457 @kindex info sources
6458 Print the names of all source files in your program for which there is
6459 debugging information, organized into two lists: files whose symbols
6460 have already been read, and files whose symbols will be read when needed.
6462 @item info functions
6463 @kindex info functions
6464 Print the names and data types of all defined functions.
6466 @item info functions @var{regexp}
6467 Print the names and data types of all defined functions
6468 whose names contain a match for regular expression @var{regexp}.
6469 Thus, @samp{info fun step} finds all functions whose names
6470 include @code{step}; @samp{info fun ^step} finds those whose names
6471 start with @code{step}.
6473 @item info variables
6474 @kindex info variables
6475 Print the names and data types of all variables that are declared
6476 outside of functions (i.e., excluding local variables).
6478 @item info variables @var{regexp}
6479 Print the names and data types of all variables (except for local
6480 variables) whose names contain a match for regular expression
6484 This was never implemented.
6486 @itemx info methods @var{regexp}
6487 @kindex info methods
6488 The @code{info methods} command permits the user to examine all defined
6489 methods within C++ program, or (with the @var{regexp} argument) a
6490 specific set of methods found in the various C++ classes. Many
6491 C++ classes provide a large number of methods. Thus, the output
6492 from the @code{ptype} command can be overwhelming and hard to use. The
6493 @code{info-methods} command filters the methods, printing only those
6494 which match the regular-expression @var{regexp}.
6497 @item maint print symbols @var{filename}
6498 @itemx maint print psymbols @var{filename}
6499 @itemx maint print msymbols @var{filename}
6500 @kindex maint print symbols
6502 @kindex maint print psymbols
6503 @cindex partial symbol dump
6504 Write a dump of debugging symbol data into the file @var{filename}.
6505 These commands are used to debug the @value{GDBN} symbol-reading code. Only
6506 symbols with debugging data are included. If you use @samp{maint print
6507 symbols}, @value{GDBN} includes all the symbols for which it has already
6508 collected full details: that is, @var{filename} reflects symbols for
6509 only those files whose symbols @value{GDBN} has read. You can use the
6510 command @code{info sources} to find out which files these are. If you
6511 use @samp{maint print psymbols} instead, the dump shows information about
6512 symbols that @value{GDBN} only knows partially---that is, symbols defined in
6513 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
6514 @samp{maint print msymbols} dumps just the minimal symbol information
6515 required for each object file from which @value{GDBN} has read some symbols.
6516 @xref{Files, ,Commands to specify files}, for a discussion of how
6517 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
6521 @chapter Altering Execution
6523 Once you think you have found an error in your program, you might want to
6524 find out for certain whether correcting the apparent error would lead to
6525 correct results in the rest of the run. You can find the answer by
6526 experiment, using the @value{GDBN} features for altering execution of the
6529 For example, you can store new values into variables or memory
6532 give your program a signal, restart it
6535 restart your program
6537 at a different address, or even return prematurely from a function to
6541 * Assignment:: Assignment to variables
6542 * Jumping:: Continuing at a different address
6544 * Signaling:: Giving your program a signal
6547 * Returning:: Returning from a function
6548 * Calling:: Calling your program's functions
6549 * Patching:: Patching your program
6553 @section Assignment to variables
6556 @cindex setting variables
6557 To alter the value of a variable, evaluate an assignment expression.
6558 @xref{Expressions, ,Expressions}. For example,
6565 stores the value 4 into the variable @code{x}, and then prints the
6566 value of the assignment expression (which is 4).
6568 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
6569 information on operators in supported languages.
6572 @kindex set variable
6573 @cindex variables, setting
6574 If you are not interested in seeing the value of the assignment, use the
6575 @code{set} command instead of the @code{print} command. @code{set} is
6576 really the same as @code{print} except that the expression's value is
6577 not printed and is not put in the value history (@pxref{Value History,
6578 ,Value history}). The expression is evaluated only for its effects.
6580 If the beginning of the argument string of the @code{set} command
6581 appears identical to a @code{set} subcommand, use the @code{set
6582 variable} command instead of just @code{set}. This command is identical
6583 to @code{set} except for its lack of subcommands. For example, if
6584 your program has a variable @code{width}, you get
6585 an error if you try to set a new value with just @samp{set width=13},
6586 because @value{GDBN} has the command @code{set width}:
6589 (@value{GDBP}) whatis width
6591 (@value{GDBP}) p width
6593 (@value{GDBP}) set width=47
6594 Invalid syntax in expression.
6598 The invalid expression, of course, is @samp{=47}. In
6599 order to actually set the program's variable @code{width}, use
6602 (@value{GDBP}) set var width=47
6605 @value{GDBN} allows more implicit conversions in assignments than C; you can
6606 freely store an integer value into a pointer variable or vice versa,
6607 and you can convert any structure to any other structure that is the
6608 same length or shorter.
6609 @comment FIXME: how do structs align/pad in these conversions?
6610 @comment /pesch@cygnus.com 18dec1990
6612 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
6613 construct to generate a value of specified type at a specified address
6614 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
6615 to memory location @code{0x83040} as an integer (which implies a certain size
6616 and representation in memory), and
6619 set @{int@}0x83040 = 4
6623 stores the value 4 into that memory location.
6626 @section Continuing at a different address
6628 Ordinarily, when you continue your program, you do so at the place where
6629 it stopped, with the @code{continue} command. You can instead continue at
6630 an address of your own choosing, with the following commands:
6633 @item jump @var{linespec}
6635 Resume execution at line @var{linespec}. Execution stops again
6636 immediately if there is a breakpoint there. @xref{List, ,Printing
6637 source lines}, for a description of the different forms of
6640 The @code{jump} command does not change the current stack frame, or
6641 the stack pointer, or the contents of any memory location or any
6642 register other than the program counter. If line @var{linespec} is in
6643 a different function from the one currently executing, the results may
6644 be bizarre if the two functions expect different patterns of arguments or
6645 of local variables. For this reason, the @code{jump} command requests
6646 confirmation if the specified line is not in the function currently
6647 executing. However, even bizarre results are predictable if you are
6648 well acquainted with the machine-language code of your program.
6650 @item jump *@var{address}
6651 Resume execution at the instruction at address @var{address}.
6654 You can get much the same effect as the @code{jump} command by storing a
6655 new value into the register @code{$pc}. The difference is that this
6656 does not start your program running; it only changes the address where it
6657 @emph{will} run when you continue. For example,
6664 makes the next @code{continue} command or stepping command execute at
6665 address @code{0x485}, rather than at the address where your program stopped.
6666 @xref{Continuing and Stepping, ,Continuing and stepping}.
6668 The most common occasion to use the @code{jump} command is to back up,
6669 perhaps with more breakpoints set, over a portion of a program that has
6670 already executed, in order to examine its execution in more detail.
6675 @section Giving your program a signal
6678 @item signal @var{signal}
6680 Resume execution where your program stopped, but immediately give it the
6681 signal @var{signal}. @var{signal} can be the name or the number of a
6682 signal. For example, on many systems @code{signal 2} and @code{signal
6683 SIGINT} are both ways of sending an interrupt signal.
6685 Alternatively, if @var{signal} is zero, continue execution without
6686 giving a signal. This is useful when your program stopped on account of
6687 a signal and would ordinary see the signal when resumed with the
6688 @code{continue} command; @samp{signal 0} causes it to resume without a
6691 @code{signal} does not repeat when you press @key{RET} a second time
6692 after executing the command.
6696 Invoking the @code{signal} command is not the same as invoking the
6697 @code{kill} utility from the shell. Sending a signal with @code{kill}
6698 causes @value{GDBN} to decide what to do with the signal depending on
6699 the signal handling tables (@pxref{Signals}). The @code{signal} command
6700 passes the signal directly to your program.
6705 @section Returning from a function
6709 @itemx return @var{expression}
6710 @cindex returning from a function
6712 You can cancel execution of a function call with the @code{return}
6713 command. If you give an
6714 @var{expression} argument, its value is used as the function's return
6718 When you use @code{return}, @value{GDBN} discards the selected stack frame
6719 (and all frames within it). You can think of this as making the
6720 discarded frame return prematurely. If you wish to specify a value to
6721 be returned, give that value as the argument to @code{return}.
6723 This pops the selected stack frame (@pxref{Selection, ,Selecting a
6724 frame}), and any other frames inside of it, leaving its caller as the
6725 innermost remaining frame. That frame becomes selected. The
6726 specified value is stored in the registers used for returning values
6729 The @code{return} command does not resume execution; it leaves the
6730 program stopped in the state that would exist if the function had just
6731 returned. In contrast, the @code{finish} command (@pxref{Continuing
6732 and Stepping, ,Continuing and stepping}) resumes execution until the
6733 selected stack frame returns naturally.
6736 @section Calling program functions
6738 @cindex calling functions
6741 @item call @var{expr}
6742 Evaluate the expression @var{expr} without displaying @code{void}
6746 You can use this variant of the @code{print} command if you want to
6747 execute a function from your program, but without cluttering the output
6748 with @code{void} returned values. The result is printed and saved in
6749 the value history, if it is not void.
6752 @section Patching programs
6753 @cindex patching binaries
6754 @cindex writing into executables
6756 @cindex writing into corefiles
6759 By default, @value{GDBN} opens the file containing your program's executable
6764 read-only. This prevents accidental alterations
6765 to machine code; but it also prevents you from intentionally patching
6766 your program's binary.
6768 If you'd like to be able to patch the binary, you can specify that
6769 explicitly with the @code{set write} command. For example, you might
6770 want to turn on internal debugging flags, or even to make emergency
6775 @itemx set write off
6777 If you specify @samp{set write on}, @value{GDBN} opens executable
6781 files for both reading and writing; if you specify @samp{set write
6782 off} (the default), @value{GDBN} opens them read-only.
6784 If you have already loaded a file, you must load it again (using the
6789 command) after changing @code{set write}, for your new setting to take
6794 Display whether executable files
6798 are opened for writing as well as reading.
6802 @chapter @value{GDBN} Files
6804 @value{GDBN} needs to know the file name of the program to be debugged, both in
6805 order to read its symbol table and in order to start your program.
6807 To debug a core dump of a previous run, you must also tell @value{GDBN}
6808 the name of the core dump file.
6812 * Files:: Commands to specify files
6813 * Symbol Errors:: Errors reading symbol files
6817 @section Commands to specify files
6818 @cindex symbol table
6821 @cindex core dump file
6822 The usual way to specify executable and core dump file names is with
6823 the command arguments given when you start @value{GDBN} (@pxref{Invocation,
6824 ,Getting In and Out of @value{GDBN}}.
6827 The usual way to specify an executable file name is with
6828 the command argument given when you start @value{GDBN}, (@pxref{Invocation,
6829 ,Getting In and Out of @value{GDBN}}.
6832 Occasionally it is necessary to change to a different file during a
6833 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
6834 a file you want to use. In these situations the @value{GDBN} commands
6835 to specify new files are useful.
6838 @item file @var{filename}
6839 @cindex executable file
6841 Use @var{filename} as the program to be debugged. It is read for its
6842 symbols and for the contents of pure memory. It is also the program
6843 executed when you use the @code{run} command. If you do not specify a
6844 directory and the file is not found in the @value{GDBN} working directory, @value{GDBN}
6845 uses the environment variable @code{PATH} as a list of directories to
6846 search, just as the shell does when looking for a program to run. You
6847 can change the value of this variable, for both @value{GDBN} and your program,
6848 using the @code{path} command.
6850 On systems with memory-mapped files, an auxiliary file
6851 @file{@var{filename}.syms} may hold symbol table information for
6852 @var{filename}. If so, @value{GDBN} maps in the symbol table from
6853 @file{@var{filename}.syms}, starting up more quickly. See the
6854 descriptions of the options @samp{-mapped} and @samp{-readnow}
6855 (available on the command line, and with the commands @code{file},
6856 @code{symbol-file}, or @code{add-symbol-file}), for more information.
6859 @code{file} with no argument makes @value{GDBN} discard any information it
6860 has on both executable file and the symbol table.
6862 @item exec-file @r{[} @var{filename} @r{]}
6864 Specify that the program to be run (but not the symbol table) is found
6865 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
6866 if necessary to locate your program. Omitting @var{filename} means to
6867 discard information on the executable file.
6869 @item symbol-file @r{[} @var{filename} @r{]}
6871 Read symbol table information from file @var{filename}. @code{PATH} is
6872 searched when necessary. Use the @code{file} command to get both symbol
6873 table and program to run from the same file.
6875 @code{symbol-file} with no argument clears out @value{GDBN} information on your
6876 program's symbol table.
6878 The @code{symbol-file} command causes @value{GDBN} to forget the contents of its
6879 convenience variables, the value history, and all breakpoints and
6880 auto-display expressions. This is because they may contain pointers to
6881 the internal data recording symbols and data types, which are part of
6882 the old symbol table data being discarded inside @value{GDBN}.
6884 @code{symbol-file} does not repeat if you press @key{RET} again after
6887 When @value{GDBN} is configured for a particular environment, it
6888 understands debugging information in whatever format is the standard
6889 generated for that environment; you may use either a GNU compiler, or
6890 other compilers that adhere to the local conventions. Best results are
6891 usually obtained from GNU compilers; for example, using @code{@value{GCC}}
6892 you can generate debugging information for optimized code.
6894 On some kinds of object files, the @code{symbol-file} command does not
6895 normally read the symbol table in full right away. Instead, it scans
6896 the symbol table quickly to find which source files and which symbols
6897 are present. The details are read later, one source file at a time,
6900 The purpose of this two-stage reading strategy is to make @value{GDBN} start up
6901 faster. For the most part, it is invisible except for occasional
6902 pauses while the symbol table details for a particular source file are
6903 being read. (The @code{set verbose} command can turn these pauses
6904 into messages if desired. @xref{Messages/Warnings, ,Optional warnings
6907 We have not implemented the two-stage strategy for COFF yet. When the
6908 symbol table is stored in COFF format, @code{symbol-file} reads the
6909 symbol table data in full right away.
6911 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
6912 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
6914 @cindex reading symbols immediately
6915 @cindex symbols, reading immediately
6917 @cindex memory-mapped symbol file
6918 @cindex saving symbol table
6919 You can override the @value{GDBN} two-stage strategy for reading symbol
6920 tables by using the @samp{-readnow} option with any of the commands that
6921 load symbol table information, if you want to be sure @value{GDBN} has the
6922 entire symbol table available.
6925 If memory-mapped files are available on your system through the
6926 @code{mmap} system call, you can use another option, @samp{-mapped}, to
6927 cause @value{GDBN} to write the symbols for your program into a reusable
6928 file. Future @value{GDBN} debugging sessions map in symbol information
6929 from this auxiliary symbol file (if the program has not changed), rather
6930 than spending time reading the symbol table from the executable
6931 program. Using the @samp{-mapped} option has the same effect as
6932 starting @value{GDBN} with the @samp{-mapped} command-line option.
6934 You can use both options together, to make sure the auxiliary symbol
6935 file has all the symbol information for your program.
6937 The auxiliary symbol file for a program called @var{myprog} is called
6938 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
6939 than the corresponding executable), @value{GDBN} always attempts to use
6940 it when you debug @var{myprog}; no special options or commands are
6943 The @file{.syms} file is specific to the host machine where you run
6944 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
6945 symbol table. It cannot be shared across multiple host platforms.
6947 @c FIXME: for now no mention of directories, since this seems to be in
6948 @c flux. 13mar1992 status is that in theory GDB would look either in
6949 @c current dir or in same dir as myprog; but issues like competing
6950 @c GDB's, or clutter in system dirs, mean that in practice right now
6951 @c only current dir is used. FFish says maybe a special GDB hierarchy
6952 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
6955 @item core-file @r{[} @var{filename} @r{]}
6958 Specify the whereabouts of a core dump file to be used as the ``contents
6959 of memory''. Traditionally, core files contain only some parts of the
6960 address space of the process that generated them; @value{GDBN} can access the
6961 executable file itself for other parts.
6963 @code{core-file} with no argument specifies that no core file is
6966 Note that the core file is ignored when your program is actually running
6967 under @value{GDBN}. So, if you have been running your program and you wish to
6968 debug a core file instead, you must kill the subprocess in which the
6969 program is running. To do this, use the @code{kill} command
6970 (@pxref{Kill Process, ,Killing the child process}).
6973 @item load @var{filename}
6976 Depending on what remote debugging facilities are configured into
6977 @value{GDBN}, the @code{load} command may be available. Where it exists, it
6978 is meant to make @var{filename} (an executable) available for debugging
6979 on the remote system---by downloading, or dynamic linking, for example.
6980 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
6981 the @code{add-symbol-file} command.
6983 If your @value{GDBN} does not have a @code{load} command, attempting to
6984 execute it gets the error message ``@code{You can't do that when your
6985 target is @dots{}}''
6988 The file is loaded at whatever address is specified in the executable.
6989 For some object file formats, you can specify the load address when you
6990 link the program; for other formats, like a.out, the object file format
6991 specifies a fixed address.
6992 @c FIXME! This would be a good place for an xref to the GNU linker doc.
6995 On VxWorks, @code{load} links @var{filename} dynamically on the
6996 current target system as well as adding its symbols in @value{GDBN}.
7000 @cindex download to Nindy-960
7001 With the Nindy interface to an Intel 960 board, @code{load}
7002 downloads @var{filename} to the 960 as well as adding its symbols in
7007 @cindex download to H8/300 or H8/500
7008 @cindex H8/300 or H8/500 download
7009 @cindex download to Hitachi SH
7010 @cindex Hitachi SH download
7011 When you select remote debugging to a Hitachi SH, H8/300, or H8/500 board
7012 (@pxref{Hitachi Remote,,@value{GDBN} and Hitachi Microprocessors}),
7013 the @code{load} command downloads your program to the Hitachi board and also
7014 opens it as the current executable target for @value{GDBN} on your host
7015 (like the @code{file} command).
7018 @code{load} does not repeat if you press @key{RET} again after using it.
7021 @item add-symbol-file @var{filename} @var{address}
7022 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7023 @kindex add-symbol-file
7024 @cindex dynamic linking
7025 The @code{add-symbol-file} command reads additional symbol table information
7026 from the file @var{filename}. You would use this command when @var{filename}
7027 has been dynamically loaded (by some other means) into the program that
7028 is running. @var{address} should be the memory address at which the
7029 file has been loaded; @value{GDBN} cannot figure this out for itself.
7030 You can specify @var{address} as an expression.
7032 The symbol table of the file @var{filename} is added to the symbol table
7033 originally read with the @code{symbol-file} command. You can use the
7034 @code{add-symbol-file} command any number of times; the new symbol data thus
7035 read keeps adding to the old. To discard all old symbol data instead,
7036 use the @code{symbol-file} command.
7038 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
7040 You can use the @samp{-mapped} and @samp{-readnow} options just as with
7041 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
7042 table information for @var{filename}.
7049 @code{info files} and @code{info target} are synonymous; both print
7050 the current target (@pxref{Targets, ,Specifying a Debugging Target}),
7053 names of the executable and core dump files
7056 name of the executable file
7058 currently in use by @value{GDBN}, and the files from which symbols were
7059 loaded. The command @code{help target} lists all possible targets
7060 rather than current ones.
7063 All file-specifying commands allow both absolute and relative file names
7064 as arguments. @value{GDBN} always converts the file name to an absolute file
7065 name and remembers it that way.
7068 @cindex shared libraries
7069 @value{GDBN} supports SunOS, SVr4, Irix 5, and IBM RS/6000 shared libraries.
7070 @value{GDBN} automatically loads symbol definitions from shared libraries
7071 when you use the @code{run} command, or when you examine a core file.
7072 (Before you issue the @code{run} command, @value{GDBN} does not understand
7073 references to a function in a shared library, however---unless you are
7074 debugging a core file).
7075 @c FIXME: some @value{GDBN} release may permit some refs to undef
7076 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
7077 @c FIXME...lib; check this from time to time when updating manual
7081 @itemx info sharedlibrary
7082 @kindex info sharedlibrary
7084 Print the names of the shared libraries which are currently loaded.
7086 @item sharedlibrary @var{regex}
7087 @itemx share @var{regex}
7088 @kindex sharedlibrary
7090 This is an obsolescent command; you can use it to explicitly load shared
7091 object library symbols for files matching a Unix regular expression, but
7092 as with files loaded automatically, it only loads shared libraries
7093 required by your program for a core file or after typing @code{run}. If
7094 @var{regex} is omitted all shared libraries required by your program are
7100 @section Errors reading symbol files
7102 While reading a symbol file, @value{GDBN} occasionally encounters problems,
7103 such as symbol types it does not recognize, or known bugs in compiler
7104 output. By default, @value{GDBN} does not notify you of such problems, since
7105 they are relatively common and primarily of interest to people
7106 debugging compilers. If you are interested in seeing information
7107 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
7108 only one message about each such type of problem, no matter how many
7109 times the problem occurs; or you can ask @value{GDBN} to print more messages,
7110 to see how many times the problems occur, with the @code{set
7111 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
7114 The messages currently printed, and their meanings, include:
7117 @item inner block not inside outer block in @var{symbol}
7119 The symbol information shows where symbol scopes begin and end
7120 (such as at the start of a function or a block of statements). This
7121 error indicates that an inner scope block is not fully contained
7122 in its outer scope blocks.
7124 @value{GDBN} circumvents the problem by treating the inner block as if it had
7125 the same scope as the outer block. In the error message, @var{symbol}
7126 may be shown as ``@code{(don't know)}'' if the outer block is not a
7129 @item block at @var{address} out of order
7131 The symbol information for symbol scope blocks should occur in
7132 order of increasing addresses. This error indicates that it does not
7135 @value{GDBN} does not circumvent this problem, and has trouble
7136 locating symbols in the source file whose symbols it is reading. (You
7137 can often determine what source file is affected by specifying
7138 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
7141 @item bad block start address patched
7143 The symbol information for a symbol scope block has a start address
7144 smaller than the address of the preceding source line. This is known
7145 to occur in the SunOS 4.1.1 (and earlier) C compiler.
7147 @value{GDBN} circumvents the problem by treating the symbol scope block as
7148 starting on the previous source line.
7150 @item bad string table offset in symbol @var{n}
7153 Symbol number @var{n} contains a pointer into the string table which is
7154 larger than the size of the string table.
7156 @value{GDBN} circumvents the problem by considering the symbol to have the
7157 name @code{foo}, which may cause other problems if many symbols end up
7160 @item unknown symbol type @code{0x@var{nn}}
7162 The symbol information contains new data types that @value{GDBN} does not yet
7163 know how to read. @code{0x@var{nn}} is the symbol type of the misunderstood
7164 information, in hexadecimal.
7166 @value{GDBN} circumvents the error by ignoring this symbol information. This
7167 usually allows you to debug your program, though certain symbols
7168 are not accessible. If you encounter such a problem and feel like
7169 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint on
7170 @code{complain}, then go up to the function @code{read_dbx_symtab} and
7171 examine @code{*bufp} to see the symbol.
7173 @item stub type has NULL name
7174 @value{GDBN} could not find the full definition for
7183 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
7185 The symbol information for a C++ member function is missing some
7186 information that recent versions of the compiler should have output
7190 @item info mismatch between compiler and debugger
7192 @value{GDBN} could not parse a type specification output by the compiler.
7196 @chapter Specifying a Debugging Target
7197 @cindex debugging target
7200 A @dfn{target} is the execution environment occupied by your program.
7202 Often, @value{GDBN} runs in the same host environment as your program; in
7203 that case, the debugging target is specified as a side effect when you
7204 use the @code{file} or @code{core} commands. When you need more
7205 flexibility---for example, running @value{GDBN} on a physically separate
7206 host, or controlling a standalone system over a serial port or a
7207 realtime system over a TCP/IP connection---you
7212 can use the @code{target} command to specify one of the target types
7213 configured for @value{GDBN} (@pxref{Target Commands, ,Commands for managing
7217 * Active Targets:: Active targets
7218 * Target Commands:: Commands for managing targets
7219 * Remote:: Remote debugging
7222 @node Active Targets
7223 @section Active targets
7224 @cindex stacking targets
7225 @cindex active targets
7226 @cindex multiple targets
7229 There are three classes of targets: processes, core files, and
7230 executable files. @value{GDBN} can work concurrently on up to three active
7231 targets, one in each class. This allows you to (for example) start a
7232 process and inspect its activity without abandoning your work on a core
7235 For example, if you execute @samp{gdb a.out}, then the executable file
7236 @code{a.out} is the only active target. If you designate a core file as
7237 well---presumably from a prior run that crashed and coredumped---then
7238 @value{GDBN} has two active targets and uses them in tandem, looking
7239 first in the corefile target, then in the executable file, to satisfy
7240 requests for memory addresses. (Typically, these two classes of target
7241 are complementary, since core files contain only a program's
7242 read-write memory---variables and so on---plus machine status, while
7243 executable files contain only the program text and initialized data.)
7246 When you type @code{run}, your executable file becomes an active process
7247 target as well. When a process target is active, all @value{GDBN} commands
7248 requesting memory addresses refer to that target; addresses in an
7252 executable file target are obscured while the process
7256 Use the @code{exec-file} command to select a
7257 new executable target (@pxref{Files, ,Commands to specify
7261 Use the @code{core-file} and @code{exec-file} commands to select a
7262 new core file or executable target (@pxref{Files, ,Commands to specify
7263 files}). To specify as a target a process that is already running, use
7264 the @code{attach} command (@pxref{Attach, ,Debugging an
7265 already-running process}).
7268 @node Target Commands
7269 @section Commands for managing targets
7272 @item target @var{type} @var{parameters}
7273 Connects the @value{GDBN} host environment to a target
7278 machine or process. A target is typically a protocol for talking to
7279 debugging facilities. You use the argument @var{type} to specify the
7280 type or protocol of the target machine.
7282 Further @var{parameters} are interpreted by the target protocol, but
7283 typically include things like device names or host names to connect
7284 with, process numbers, and baud rates.
7287 The @code{target} command does not repeat if you press @key{RET} again
7288 after executing the command.
7292 Displays the names of all targets available. To display targets
7293 currently selected, use either @code{info target} or @code{info files}
7294 (@pxref{Files, ,Commands to specify files}).
7296 @item help target @var{name}
7297 Describe a particular target, including any parameters necessary to
7301 Here are some common targets (available, or not, depending on the GDB
7305 @item target exec @var{program}
7307 An executable file. @samp{target exec @var{program}} is the same as
7308 @samp{exec-file @var{program}}.
7311 @item target core @var{filename}
7313 A core dump file. @samp{target core @var{filename}} is the same as
7314 @samp{core-file @var{filename}}.
7318 @item target remote @var{dev}
7319 @kindex target remote
7320 Remote serial target in GDB-specific protocol. The argument @var{dev}
7321 specifies what serial device to use for the connection (e.g.
7322 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}.
7328 CPU simulator. @xref{Simulator,,Simulated CPU Target}.
7332 @item target udi @var{keyword}
7334 Remote AMD29K target, using the AMD UDI protocol. The @var{keyword}
7335 argument specifies which 29K board or simulator to use. @xref{UDI29K
7336 Remote,,The UDI protocol for AMD29K}.
7338 @item target amd-eb @var{dev} @var{speed} @var{PROG}
7339 @kindex target amd-eb
7341 Remote PC-resident AMD EB29K board, attached over serial lines.
7342 @var{dev} is the serial device, as for @code{target remote};
7343 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
7344 name of the program to be debugged, as it appears to DOS on the PC.
7345 @xref{EB29K Remote, ,The EBMON protocol for AMD29K}.
7351 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
7352 @ifclear H8EXCLUSIVE
7353 @c Unix only, not currently of interest for H8-only manual
7354 Use special commands @code{device} and @code{speed} to control the serial
7355 line and the communications speed used.
7357 @xref{Hitachi Remote,,@value{GDBN} and Hitachi Microprocessors}.
7361 @item target nindy @var{devicename}
7362 @kindex target nindy
7363 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
7364 the name of the serial device to use for the connection, e.g.
7365 @file{/dev/ttya}. @xref{i960-Nindy Remote, ,@value{GDBN} with a remote i960 (Nindy)}.
7369 @item target st2000 @var{dev} @var{speed}
7370 @kindex target st2000
7371 A Tandem ST2000 phone switch, running Tandem's STDBUG protocol. @var{dev}
7372 is the name of the device attached to the ST2000 serial line;
7373 @var{speed} is the communication line speed. The arguments are not used
7374 if @value{GDBN} is configured to connect to the ST2000 using TCP or Telnet.
7375 @xref{ST2000 Remote,,@value{GDBN} with a Tandem ST2000}.
7379 @item target vxworks @var{machinename}
7380 @kindex target vxworks
7381 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
7382 is the target system's machine name or IP address.
7383 @xref{VxWorks Remote, ,@value{GDBN} and VxWorks}.
7388 Different targets are available on different configurations of @value{GDBN}; your
7389 configuration may have more or fewer targets.
7393 @section Remote debugging
7394 @cindex remote debugging
7396 If you are trying to debug a program running on a machine that cannot run
7397 GDB in the usual way, it is often useful to use remote debugging. For
7398 example, you might use remote debugging on an operating system kernel, or on
7399 a small system which does not have a general purpose operating system
7400 powerful enough to run a full-featured debugger.
7402 Some configurations of GDB have special serial or TCP/IP interfaces
7403 to make this work with particular debugging targets. In addition,
7404 GDB comes with a generic serial protocol (specific to GDB, but
7405 not specific to any particular target system) which you can use if you
7406 write the remote stubs---the code that runs on the remote system to
7407 communicate with GDB.
7409 Other remote targets may be available in your
7410 configuration of GDB; use @code{help target} to list them.
7413 @c Text on starting up GDB in various specific cases; it goes up front
7414 @c in manuals configured for any of those particular situations, here
7418 * Remote Serial:: @value{GDBN} remote serial protocol
7421 * i960-Nindy Remote:: @value{GDBN} with a remote i960 (Nindy)
7424 * UDI29K Remote:: The UDI protocol for AMD29K
7425 * EB29K Remote:: The EBMON protocol for AMD29K
7428 * VxWorks Remote:: @value{GDBN} and VxWorks
7431 * ST2000 Remote:: @value{GDBN} with a Tandem ST2000
7434 * Hitachi Remote:: @value{GDBN} and Hitachi Microprocessors
7437 * MIPS Remote:: @value{GDBN} and MIPS boards
7440 * Simulator:: Simulated CPU target
7444 @include remote.texi
7447 @node Controlling GDB
7448 @chapter Controlling @value{GDBN}
7450 You can alter the way @value{GDBN} interacts with you by using
7451 the @code{set} command. For commands controlling how @value{GDBN} displays
7452 data, @pxref{Print Settings, ,Print settings}; other settings are described here.
7456 * Editing:: Command editing
7457 * History:: Command history
7458 * Screen Size:: Screen size
7460 * Messages/Warnings:: Optional warnings and messages
7467 @value{GDBN} indicates its readiness to read a command by printing a string
7468 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
7469 can change the prompt string with the @code{set prompt} command. For
7470 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
7471 the prompt in one of the @value{GDBN} sessions so that you can always tell which
7472 one you are talking to.
7475 @item set prompt @var{newprompt}
7477 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
7480 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
7484 @section Command editing
7486 @cindex command line editing
7488 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
7489 GNU library provides consistent behavior for programs which provide a
7490 command line interface to the user. Advantages are @code{emacs}-style
7491 or @code{vi}-style inline editing of commands, @code{csh}-like history
7492 substitution, and a storage and recall of command history across
7495 You may control the behavior of command line editing in @value{GDBN} with the
7502 @itemx set editing on
7503 Enable command line editing (enabled by default).
7505 @item set editing off
7506 Disable command line editing.
7508 @kindex show editing
7510 Show whether command line editing is enabled.
7514 @section Command history
7516 @value{GDBN} can keep track of the commands you type during your
7517 debugging sessions, so that you can be certain of precisely what
7518 happened. Use these commands to manage the @value{GDBN} command
7522 @cindex history substitution
7523 @cindex history file
7524 @kindex set history filename
7526 @item set history filename @var{fname}
7527 Set the name of the @value{GDBN} command history file to @var{fname}.
7528 This is the file where @value{GDBN} reads an initial command history
7529 list, and where it writes the command history from this session when it
7530 exits. You can access this list through history expansion or through
7531 the history command editing characters listed below. This file defaults
7532 to the value of the environment variable @code{GDBHISTFILE}, or to
7533 @file{./.gdb_history} if this variable is not set.
7535 @cindex history save
7536 @kindex set history save
7537 @item set history save
7538 @itemx set history save on
7539 Record command history in a file, whose name may be specified with the
7540 @code{set history filename} command. By default, this option is disabled.
7542 @item set history save off
7543 Stop recording command history in a file.
7545 @cindex history size
7546 @kindex set history size
7547 @item set history size @var{size}
7548 Set the number of commands which @value{GDBN} keeps in its history list.
7549 This defaults to the value of the environment variable
7550 @code{HISTSIZE}, or to 256 if this variable is not set.
7553 @cindex history expansion
7554 History expansion assigns special meaning to the character @kbd{!}.
7555 @ifset have-readline-appendices
7556 @xref{Event Designators}.
7559 Since @kbd{!} is also the logical not operator in C, history expansion
7560 is off by default. If you decide to enable history expansion with the
7561 @code{set history expansion on} command, you may sometimes need to
7562 follow @kbd{!} (when it is used as logical not, in an expression) with
7563 a space or a tab to prevent it from being expanded. The readline
7564 history facilities do not attempt substitution on the strings
7565 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
7567 The commands to control history expansion are:
7571 @kindex set history expansion
7572 @item set history expansion on
7573 @itemx set history expansion
7574 Enable history expansion. History expansion is off by default.
7576 @item set history expansion off
7577 Disable history expansion.
7579 The readline code comes with more complete documentation of
7580 editing and history expansion features. Users unfamiliar with @code{emacs}
7581 or @code{vi} may wish to read it.
7582 @ifset have-readline-appendices
7583 @xref{Command Line Editing}.
7587 @kindex show history
7589 @itemx show history filename
7590 @itemx show history save
7591 @itemx show history size
7592 @itemx show history expansion
7593 These commands display the state of the @value{GDBN} history parameters.
7594 @code{show history} by itself displays all four states.
7599 @kindex show commands
7601 Display the last ten commands in the command history.
7603 @item show commands @var{n}
7604 Print ten commands centered on command number @var{n}.
7606 @item show commands +
7607 Print ten commands just after the commands last printed.
7611 @section Screen size
7612 @cindex size of screen
7613 @cindex pauses in output
7615 Certain commands to @value{GDBN} may produce large amounts of
7616 information output to the screen. To help you read all of it,
7617 @value{GDBN} pauses and asks you for input at the end of each page of
7618 output. Type @key{RET} when you want to continue the output, or @kbd{q}
7619 to discard the remaining output. Also, the screen width setting
7620 determines when to wrap lines of output. Depending on what is being
7621 printed, @value{GDBN} tries to break the line at a readable place,
7622 rather than simply letting it overflow onto the following line.
7624 Normally @value{GDBN} knows the size of the screen from the termcap data base
7625 together with the value of the @code{TERM} environment variable and the
7626 @code{stty rows} and @code{stty cols} settings. If this is not correct,
7627 you can override it with the @code{set height} and @code{set
7631 @item set height @var{lpp}
7633 @itemx set width @var{cpl}
7639 These @code{set} commands specify a screen height of @var{lpp} lines and
7640 a screen width of @var{cpl} characters. The associated @code{show}
7641 commands display the current settings.
7643 If you specify a height of zero lines, @value{GDBN} does not pause during output
7644 no matter how long the output is. This is useful if output is to a file
7645 or to an editor buffer.
7647 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
7648 from wrapping its output.
7653 @cindex number representation
7654 @cindex entering numbers
7656 You can always enter numbers in octal, decimal, or hexadecimal in @value{GDBN} by
7657 the usual conventions: octal numbers begin with @samp{0}, decimal
7658 numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}.
7659 Numbers that begin with none of these are, by default, entered in base
7660 10; likewise, the default display for numbers---when no particular
7661 format is specified---is base 10. You can change the default base for
7662 both input and output with the @code{set radix} command.
7666 @item set radix @var{base}
7667 Set the default base for numeric input and display. Supported choices
7668 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
7669 specified either unambiguously or using the current default radix; for
7679 sets the base to decimal. On the other hand, @samp{set radix 10}
7680 leaves the radix unchanged no matter what it was.
7684 Display the current default base for numeric input and display.
7687 @node Messages/Warnings
7688 @section Optional warnings and messages
7690 By default, @value{GDBN} is silent about its inner workings. If you are running
7691 on a slow machine, you may want to use the @code{set verbose} command.
7692 It makes @value{GDBN} tell you when it does a lengthy internal operation, so
7693 you will not think it has crashed.
7695 Currently, the messages controlled by @code{set verbose} are those
7696 which announce that the symbol table for a source file is being read;
7697 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
7701 @item set verbose on
7702 Enables @value{GDBN} output of certain informational messages.
7704 @item set verbose off
7705 Disables @value{GDBN} output of certain informational messages.
7707 @kindex show verbose
7709 Displays whether @code{set verbose} is on or off.
7712 By default, if @value{GDBN} encounters bugs in the symbol table of an object
7713 file, it is silent; but if you are debugging a compiler, you may find
7714 this information useful (@pxref{Symbol Errors, ,Errors reading symbol files}).
7717 @kindex set complaints
7718 @item set complaints @var{limit}
7719 Permits @value{GDBN} to output @var{limit} complaints about each type of unusual
7720 symbols before becoming silent about the problem. Set @var{limit} to
7721 zero to suppress all complaints; set it to a large number to prevent
7722 complaints from being suppressed.
7724 @kindex show complaints
7725 @item show complaints
7726 Displays how many symbol complaints @value{GDBN} is permitted to produce.
7729 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
7730 lot of stupid questions to confirm certain commands. For example, if
7731 you try to run a program which is already running:
7735 The program being debugged has been started already.
7736 Start it from the beginning? (y or n)
7739 If you are willing to unflinchingly face the consequences of your own
7740 commands, you can disable this ``feature'':
7745 @cindex confirmation
7746 @cindex stupid questions
7747 @item set confirm off
7748 Disables confirmation requests.
7750 @item set confirm on
7751 Enables confirmation requests (the default).
7754 @kindex show confirm
7755 Displays state of confirmation requests.
7758 @c FIXME this does not really belong here. But where *does* it belong?
7759 @cindex reloading symbols
7760 Some systems allow individual object files that make up your program to
7761 be replaced without stopping and restarting your program.
7763 For example, in VxWorks you can simply recompile a defective object file
7764 and keep on running.
7766 If you are running on one of these systems, you can allow @value{GDBN} to
7767 reload the symbols for automatically relinked modules:
7770 @kindex set symbol-reloading
7771 @item set symbol-reloading on
7772 Replace symbol definitions for the corresponding source file when an
7773 object file with a particular name is seen again.
7775 @item set symbol-reloading off
7776 Do not replace symbol definitions when re-encountering object files of
7777 the same name. This is the default state; if you are not running on a
7778 system that permits automatically relinking modules, you should leave
7779 @code{symbol-reloading} off, since otherwise @value{GDBN} may discard symbols
7780 when linking large programs, that may contain several modules (from
7781 different directories or libraries) with the same name.
7783 @item show symbol-reloading
7784 Show the current @code{on} or @code{off} setting.
7788 @chapter Canned Sequences of Commands
7790 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
7791 command lists}), @value{GDBN} provides two ways to store sequences of commands
7792 for execution as a unit: user-defined commands and command files.
7795 * Define:: User-defined commands
7796 * Hooks:: User-defined command hooks
7797 * Command Files:: Command files
7798 * Output:: Commands for controlled output
7802 @section User-defined commands
7804 @cindex user-defined command
7805 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to which you
7806 assign a new name as a command. This is done with the @code{define}
7810 @item define @var{commandname}
7812 Define a command named @var{commandname}. If there is already a command
7813 by that name, you are asked to confirm that you want to redefine it.
7815 The definition of the command is made up of other @value{GDBN} command lines,
7816 which are given following the @code{define} command. The end of these
7817 commands is marked by a line containing @code{end}.
7819 @item document @var{commandname}
7821 Give documentation to the user-defined command @var{commandname}. The
7822 command @var{commandname} must already be defined. This command reads
7823 lines of documentation just as @code{define} reads the lines of the
7824 command definition, ending with @code{end}. After the @code{document}
7825 command is finished, @code{help} on command @var{commandname} displays
7826 the documentation you have specified.
7828 You may use the @code{document} command again to change the
7829 documentation of a command. Redefining the command with @code{define}
7830 does not change the documentation.
7832 @item help user-defined
7833 @kindex help user-defined
7834 List all user-defined commands, with the first line of the documentation
7838 @itemx show user @var{commandname}
7840 Display the @value{GDBN} commands used to define @var{commandname} (but not its
7841 documentation). If no @var{commandname} is given, display the
7842 definitions for all user-defined commands.
7845 User-defined commands do not take arguments. When they are executed, the
7846 commands of the definition are not printed. An error in any command
7847 stops execution of the user-defined command.
7849 Commands that would ask for confirmation if used interactively proceed
7850 without asking when used inside a user-defined command. Many @value{GDBN} commands
7851 that normally print messages to say what they are doing omit the messages
7852 when used in a user-defined command.
7855 @section User-defined command hooks
7856 @cindex command files
7858 You may define @emph{hooks}, which are a special kind of user-defined
7859 command. Whenever you run the command @samp{foo}, if the user-defined
7860 command @samp{hook-foo} exists, it is executed (with no arguments)
7861 before that command.
7863 In addition, a pseudo-command, @samp{stop} exists. Defining
7864 (@samp{hook-stop}) makes the associated commands execute every time
7865 execution stops in your program: before breakpoint commands are run,
7866 displays are printed, or the stack frame is printed.
7869 For example, to ignore @code{SIGALRM} signals while
7870 single-stepping, but treat them normally during normal execution,
7875 handle SIGALRM nopass
7882 define hook-continue
7888 You can define a hook for any single-word command in @value{GDBN}, but
7889 not for command aliases; you should define a hook for the basic command
7890 name, e.g. @code{backtrace} rather than @code{bt}.
7891 @c FIXME! So how does Joe User discover whether a command is an alias
7893 If an error occurs during the execution of your hook, execution of
7894 @value{GDBN} commands stops and @value{GDBN} issues a prompt
7895 (before the command that you actually typed had a chance to run).
7897 If you try to define a hook which does not match any known command, you
7898 get a warning from the @code{define} command.
7901 @section Command files
7903 @cindex command files
7904 A command file for @value{GDBN} is a file of lines that are @value{GDBN} commands. Comments
7905 (lines starting with @kbd{#}) may also be included. An empty line in a
7906 command file does nothing; it does not mean to repeat the last command, as
7907 it would from the terminal.
7910 @cindex @file{@value{GDBINIT}}
7911 When you start @value{GDBN}, it automatically executes commands from its
7912 @dfn{init files}. These are files named @file{@value{GDBINIT}}.
7913 @value{GDBN} reads the init file (if any) in your home directory, then
7914 processes command line options and operands, and then reads the init
7915 file (if any) in the current working directory. This is so the init
7916 file in your home directory can set options (such as @code{set
7917 complaints}) which affect the processing of the command line options and
7918 operands. The init files are not executed if you use the @samp{-nx}
7919 option; @pxref{Mode Options, ,Choosing modes}.
7922 @cindex init file name
7923 On some configurations of @value{GDBN}, the init file is known by a
7924 different name (these are typically environments where a specialized
7925 form of GDB may need to coexist with other forms, hence a different name
7926 for the specialized version's init file). These are the environments
7927 with special init file names:
7932 VxWorks (Wind River Systems real-time OS): @samp{.vxgdbinit}
7934 @kindex .os68gdbinit
7936 OS68K (Enea Data Systems real-time OS): @samp{.os68gdbinit}
7940 ES-1800 (Ericsson Telecom AB M68000 emulator): @samp{.esgdbinit}
7944 You can also request the execution of a command file with the
7945 @code{source} command:
7948 @item source @var{filename}
7950 Execute the command file @var{filename}.
7953 The lines in a command file are executed sequentially. They are not
7954 printed as they are executed. An error in any command terminates execution
7955 of the command file.
7957 Commands that would ask for confirmation if used interactively proceed
7958 without asking when used in a command file. Many @value{GDBN} commands that
7959 normally print messages to say what they are doing omit the messages
7960 when called from command files.
7963 @section Commands for controlled output
7965 During the execution of a command file or a user-defined command, normal
7966 @value{GDBN} output is suppressed; the only output that appears is what is
7967 explicitly printed by the commands in the definition. This section
7968 describes three commands useful for generating exactly the output you
7972 @item echo @var{text}
7974 @c I do not consider backslash-space a standard C escape sequence
7975 @c because it is not in ANSI.
7976 Print @var{text}. Nonprinting characters can be included in
7977 @var{text} using C escape sequences, such as @samp{\n} to print a
7978 newline. @strong{No newline is printed unless you specify one.}
7979 In addition to the standard C escape sequences, a backslash followed
7980 by a space stands for a space. This is useful for displaying a
7981 string with spaces at the beginning or the end, since leading and
7982 trailing spaces are otherwise trimmed from all arguments.
7983 To print @samp{@w{ }and foo =@w{ }}, use the command
7984 @samp{echo \@w{ }and foo = \@w{ }}.
7986 A backslash at the end of @var{text} can be used, as in C, to continue
7987 the command onto subsequent lines. For example,
7990 echo This is some text\n\
7991 which is continued\n\
7992 onto several lines.\n
7995 produces the same output as
7998 echo This is some text\n
7999 echo which is continued\n
8000 echo onto several lines.\n
8003 @item output @var{expression}
8005 Print the value of @var{expression} and nothing but that value: no
8006 newlines, no @samp{$@var{nn} = }. The value is not entered in the
8007 value history either. @xref{Expressions, ,Expressions}, for more information on
8010 @item output/@var{fmt} @var{expression}
8011 Print the value of @var{expression} in format @var{fmt}. You can use
8012 the same formats as for @code{print}. @xref{Output Formats,,Output
8013 formats}, for more information.
8015 @item printf @var{string}, @var{expressions}@dots{}
8017 Print the values of the @var{expressions} under the control of
8018 @var{string}. The @var{expressions} are separated by commas and may be
8019 either numbers or pointers. Their values are printed as specified by
8020 @var{string}, exactly as if your program were to execute the C
8024 printf (@var{string}, @var{expressions}@dots{});
8027 For example, you can print two values in hex like this:
8030 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
8033 The only backslash-escape sequences that you can use in the format
8034 string are the simple ones that consist of backslash followed by a
8040 @chapter Using @value{GDBN} under GNU Emacs
8043 A special interface allows you to use GNU Emacs to view (and
8044 edit) the source files for the program you are debugging with
8047 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
8048 executable file you want to debug as an argument. This command starts
8049 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
8050 created Emacs buffer.
8052 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
8057 All ``terminal'' input and output goes through the Emacs buffer.
8060 This applies both to @value{GDBN} commands and their output, and to the input
8061 and output done by the program you are debugging.
8063 This is useful because it means that you can copy the text of previous
8064 commands and input them again; you can even use parts of the output
8067 All the facilities of Emacs' Shell mode are available for interacting
8068 with your program. In particular, you can send signals the usual
8069 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
8074 @value{GDBN} displays source code through Emacs.
8077 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
8078 source file for that frame and puts an arrow (@samp{=>}) at the
8079 left margin of the current line. Emacs uses a separate buffer for
8080 source display, and splits the screen to show both your @value{GDBN} session
8083 Explicit @value{GDBN} @code{list} or search commands still produce output as
8084 usual, but you probably have no reason to use them from Emacs.
8087 @emph{Warning:} If the directory where your program resides is not your
8088 current directory, it can be easy to confuse Emacs about the location of
8089 the source files, in which case the auxiliary display buffer does not
8090 appear to show your source. @value{GDBN} can find programs by searching your
8091 environment's @code{PATH} variable, so the @value{GDBN} input and output
8092 session proceeds normally; but Emacs does not get enough information
8093 back from @value{GDBN} to locate the source files in this situation. To
8094 avoid this problem, either start @value{GDBN} mode from the directory where
8095 your program resides, or specify an absolute file name when prompted for the
8096 @kbd{M-x gdb} argument.
8098 A similar confusion can result if you use the @value{GDBN} @code{file} command to
8099 switch to debugging a program in some other location, from an existing
8100 @value{GDBN} buffer in Emacs.
8103 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
8104 you need to call @value{GDBN} by a different name (for example, if you keep
8105 several configurations around, with different names) you can set the
8106 Emacs variable @code{gdb-command-name}; for example,
8109 (setq gdb-command-name "mygdb")
8113 (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or
8114 in your @file{.emacs} file) makes Emacs call the program named
8115 ``@code{mygdb}'' instead.
8117 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
8118 addition to the standard Shell mode commands:
8122 Describe the features of Emacs' @value{GDBN} Mode.
8125 Execute to another source line, like the @value{GDBN} @code{step} command; also
8126 update the display window to show the current file and location.
8129 Execute to next source line in this function, skipping all function
8130 calls, like the @value{GDBN} @code{next} command. Then update the display window
8131 to show the current file and location.
8134 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
8135 display window accordingly.
8138 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
8139 display window accordingly.
8142 Execute until exit from the selected stack frame, like the @value{GDBN}
8143 @code{finish} command.
8146 Continue execution of your program, like the @value{GDBN} @code{continue}
8149 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
8152 Go up the number of frames indicated by the numeric argument
8153 (@pxref{Arguments, , Numeric Arguments, emacs, The GNU Emacs Manual}),
8154 like the @value{GDBN} @code{up} command.
8156 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
8159 Go down the number of frames indicated by the numeric argument, like the
8160 @value{GDBN} @code{down} command.
8162 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
8165 Read the number where the cursor is positioned, and insert it at the end
8166 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
8167 around an address that was displayed earlier, type @kbd{disassemble};
8168 then move the cursor to the address display, and pick up the
8169 argument for @code{disassemble} by typing @kbd{C-x &}.
8171 You can customize this further by defining elements of the list
8172 @code{gdb-print-command}; once it is defined, you can format or
8173 otherwise process numbers picked up by @kbd{C-x &} before they are
8174 inserted. A numeric argument to @kbd{C-x &} indicates that you
8175 wish special formatting, and also acts as an index to pick an element of the
8176 list. If the list element is a string, the number to be inserted is
8177 formatted using the Emacs function @code{format}; otherwise the number
8178 is passed as an argument to the corresponding list element.
8181 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
8182 tells @value{GDBN} to set a breakpoint on the source line point is on.
8184 If you accidentally delete the source-display buffer, an easy way to get
8185 it back is to type the command @code{f} in the @value{GDBN} buffer, to
8186 request a frame display; when you run under Emacs, this recreates
8187 the source buffer if necessary to show you the context of the current
8190 The source files displayed in Emacs are in ordinary Emacs buffers
8191 which are visiting the source files in the usual way. You can edit
8192 the files with these buffers if you wish; but keep in mind that @value{GDBN}
8193 communicates with Emacs in terms of line numbers. If you add or
8194 delete lines from the text, the line numbers that @value{GDBN} knows cease
8195 to correspond properly with the code.
8197 @c The following dropped because Epoch is nonstandard. Reactivate
8198 @c if/when v19 does something similar. ---pesch@cygnus.com 19dec1990
8200 @kindex emacs epoch environment
8204 Version 18 of Emacs has a built-in window system called the @code{epoch}
8205 environment. Users of this environment can use a new command,
8206 @code{inspect} which performs identically to @code{print} except that
8207 each value is printed in its own window.
8213 @chapter Using @value{GDBN} with Energize
8216 The Energize Programming System is an integrated development environment
8217 that includes a point-and-click interface to many programming tools.
8218 When you use @value{GDBN} in this environment, you can use the standard
8219 Energize graphical interface to drive @value{GDBN}; you can also, if you
8220 choose, type @value{GDBN} commands as usual in a debugging window. Even if
8221 you use the graphical interface, the debugging window (which uses Emacs,
8222 and resembles the standard Emacs interface to @value{GDBN}) displays the
8223 equivalent commands, so that the history of your debugging session is
8226 When Energize starts up a @value{GDBN} session, it uses one of the
8227 command-line options @samp{-energize} or @samp{-cadillac} (``cadillac''
8228 is the name of the communications protocol used by the Energize system).
8229 This option makes @value{GDBN} run as one of the tools in the Energize Tool
8230 Set: it sends all output to the Energize kernel, and accept input from
8233 See the user manual for the Energize Programming System for
8234 information on how to use the Energize graphical interface and the other
8235 development tools that Energize integrates with @value{GDBN}.
8240 @chapter Reporting Bugs in @value{GDBN}
8241 @cindex bugs in @value{GDBN}
8242 @cindex reporting bugs in @value{GDBN}
8244 Your bug reports play an essential role in making @value{GDBN} reliable.
8246 Reporting a bug may help you by bringing a solution to your problem, or it
8247 may not. But in any case the principal function of a bug report is to help
8248 the entire community by making the next version of @value{GDBN} work better. Bug
8249 reports are your contribution to the maintenance of @value{GDBN}.
8251 In order for a bug report to serve its purpose, you must include the
8252 information that enables us to fix the bug.
8255 * Bug Criteria:: Have you found a bug?
8256 * Bug Reporting:: How to report bugs
8260 @section Have you found a bug?
8261 @cindex bug criteria
8263 If you are not sure whether you have found a bug, here are some guidelines:
8267 @cindex fatal signal
8268 @cindex debugger crash
8269 @cindex crash of debugger
8270 If the debugger gets a fatal signal, for any input whatever, that is a
8271 @value{GDBN} bug. Reliable debuggers never crash.
8274 @cindex error on valid input
8275 If @value{GDBN} produces an error message for valid input, that is a bug.
8278 @cindex invalid input
8279 If @value{GDBN} does not produce an error message for invalid input,
8280 that is a bug. However, you should note that your idea of
8281 ``invalid input'' might be our idea of ``an extension'' or ``support
8282 for traditional practice''.
8285 If you are an experienced user of debugging tools, your suggestions
8286 for improvement of @value{GDBN} are welcome in any case.
8290 @section How to report bugs
8292 @cindex @value{GDBN} bugs, reporting
8294 A number of companies and individuals offer support for GNU products.
8295 If you obtained @value{GDBN} from a support organization, we recommend you
8296 contact that organization first.
8298 You can find contact information for many support companies and
8299 individuals in the file @file{etc/SERVICE} in the GNU Emacs
8302 In any event, we also recommend that you send bug reports for @value{GDBN} to one
8306 bug-gdb@@prep.ai.mit.edu
8307 @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb
8310 @strong{Do not send bug reports to @samp{info-gdb}, or to
8311 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do not want to
8312 receive bug reports. Those that do, have arranged to receive @samp{bug-gdb}.
8314 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
8315 serves as a repeater. The mailing list and the newsgroup carry exactly
8316 the same messages. Often people think of posting bug reports to the
8317 newsgroup instead of mailing them. This appears to work, but it has one
8318 problem which can be crucial: a newsgroup posting often lacks a mail
8319 path back to the sender. Thus, if we need to ask for more information,
8320 we may be unable to reach you. For this reason, it is better to send
8321 bug reports to the mailing list.
8323 As a last resort, send bug reports on paper to:
8327 Free Software Foundation
8332 The fundamental principle of reporting bugs usefully is this:
8333 @strong{report all the facts}. If you are not sure whether to state a
8334 fact or leave it out, state it!
8336 Often people omit facts because they think they know what causes the
8337 problem and assume that some details do not matter. Thus, you might
8338 assume that the name of the variable you use in an example does not matter.
8339 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
8340 stray memory reference which happens to fetch from the location where that
8341 name is stored in memory; perhaps, if the name were different, the contents
8342 of that location would fool the debugger into doing the right thing despite
8343 the bug. Play it safe and give a specific, complete example. That is the
8344 easiest thing for you to do, and the most helpful.
8346 Keep in mind that the purpose of a bug report is to enable us to fix
8347 the bug if it is new to us. It is not as important as what happens if
8348 the bug is already known. Therefore, always write your bug reports on
8349 the assumption that the bug has not been reported previously.
8351 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8352 bell?'' Those bug reports are useless, and we urge everyone to
8353 @emph{refuse to respond to them} except to chide the sender to report
8356 To enable us to fix the bug, you should include all these things:
8360 The version of @value{GDBN}. @value{GDBN} announces it if you start with no
8361 arguments; you can also print it at any time using @code{show version}.
8363 Without this, we will not know whether there is any point in looking for
8364 the bug in the current version of @value{GDBN}.
8367 The type of machine you are using, and the operating system name and
8371 What compiler (and its version) was used to compile @value{GDBN}---e.g.
8372 ``@value{GCC}--2.0''.
8375 What compiler (and its version) was used to compile the program you
8376 are debugging---e.g. ``@value{GCC}--2.0''.
8379 The command arguments you gave the compiler to compile your example and
8380 observe the bug. For example, did you use @samp{-O}? To guarantee
8381 you will not omit something important, list them all. A copy of the
8382 Makefile (or the output from make) is sufficient.
8384 If we were to try to guess the arguments, we would probably guess wrong
8385 and then we might not encounter the bug.
8388 A complete input script, and all necessary source files, that will
8392 A description of what behavior you observe that you believe is
8393 incorrect. For example, ``It gets a fatal signal.''
8395 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we will
8396 certainly notice it. But if the bug is incorrect output, we might not
8397 notice unless it is glaringly wrong. We are human, after all. You
8398 might as well not give us a chance to make a mistake.
8400 Even if the problem you experience is a fatal signal, you should still
8401 say so explicitly. Suppose something strange is going on, such as,
8402 your copy of @value{GDBN} is out of synch, or you have encountered a
8403 bug in the C library on your system. (This has happened!) Your copy
8404 might crash and ours would not. If you told us to expect a crash,
8405 then when ours fails to crash, we would know that the bug was not
8406 happening for us. If you had not told us to expect a crash, then we
8407 would not be able to draw any conclusion from our observations.
8410 If you wish to suggest changes to the @value{GDBN} source, send us context
8411 diffs. If you even discuss something in the @value{GDBN} source, refer to
8412 it by context, not by line number.
8414 The line numbers in our development sources will not match those in your
8415 sources. Your line numbers would convey no useful information to us.
8418 Here are some things that are not necessary:
8422 A description of the envelope of the bug.
8424 Often people who encounter a bug spend a lot of time investigating
8425 which changes to the input file will make the bug go away and which
8426 changes will not affect it.
8428 This is often time consuming and not very useful, because the way we
8429 will find the bug is by running a single example under the debugger
8430 with breakpoints, not by pure deduction from a series of examples.
8431 We recommend that you save your time for something else.
8433 Of course, if you can find a simpler example to report @emph{instead}
8434 of the original one, that is a convenience for us. Errors in the
8435 output will be easier to spot, running under the debugger will take
8436 less time, and so on.
8438 However, simplification is not vital; if you do not want to do this,
8439 report the bug anyway and send us the entire test case you used.
8442 A patch for the bug.
8444 A patch for the bug does help us if it is a good one. But do not omit
8445 the necessary information, such as the test case, on the assumption that
8446 a patch is all we need. We might see problems with your patch and decide
8447 to fix the problem another way, or we might not understand it at all.
8449 Sometimes with a program as complicated as @value{GDBN} it is very hard to
8450 construct an example that will make the program follow a certain path
8451 through the code. If you do not send us the example, we will not be able
8452 to construct one, so we will not be able to verify that the bug is fixed.
8454 And if we cannot understand what bug you are trying to fix, or why your
8455 patch should be an improvement, we will not install it. A test case will
8456 help us to understand.
8459 A guess about what the bug is or what it depends on.
8461 Such guesses are usually wrong. Even we cannot guess right about such
8462 things without first using the debugger to find the facts.
8465 @c The readline documentation is distributed with the readline code
8466 @c and consists of the two following files:
8469 @c Use -I with makeinfo to point to the appropriate directory,
8470 @c environment var TEXINPUTS with TeX.
8471 @include rluser.texinfo
8472 @include inc-hist.texi
8475 @node Renamed Commands
8476 @appendix Renamed Commands
8478 The following commands were renamed in GDB 4, in order to make the
8479 command set as a whole more consistent and easier to use and remember:
8482 @kindex delete environment
8483 @kindex info copying
8484 @kindex info convenience
8485 @kindex info directories
8486 @kindex info editing
8487 @kindex info history
8488 @kindex info targets
8490 @kindex info version
8491 @kindex info warranty
8492 @kindex set addressprint
8493 @kindex set arrayprint
8494 @kindex set prettyprint
8495 @kindex set screen-height
8496 @kindex set screen-width
8497 @kindex set unionprint
8498 @kindex set vtblprint
8499 @kindex set demangle
8500 @kindex set asm-demangle
8501 @kindex set sevenbit-strings
8502 @kindex set array-max
8504 @kindex set history write
8505 @kindex show addressprint
8506 @kindex show arrayprint
8507 @kindex show prettyprint
8508 @kindex show screen-height
8509 @kindex show screen-width
8510 @kindex show unionprint
8511 @kindex show vtblprint
8512 @kindex show demangle
8513 @kindex show asm-demangle
8514 @kindex show sevenbit-strings
8515 @kindex show array-max
8516 @kindex show caution
8517 @kindex show history write
8522 @c END TEXI2ROFF-KILL
8524 OLD COMMAND NEW COMMAND
8526 --------------- -------------------------------
8527 @c END TEXI2ROFF-KILL
8528 add-syms add-symbol-file
8529 delete environment unset environment
8530 info convenience show convenience
8531 info copying show copying
8532 info directories show directories
8533 info editing show commands
8534 info history show values
8535 info targets help target
8536 info values show values
8537 info version show version
8538 info warranty show warranty
8539 set/show addressprint set/show print address
8540 set/show array-max set/show print elements
8541 set/show arrayprint set/show print array
8542 set/show asm-demangle set/show print asm-demangle
8543 set/show caution set/show confirm
8544 set/show demangle set/show print demangle
8545 set/show history write set/show history save
8546 set/show prettyprint set/show print pretty
8547 set/show screen-height set/show height
8548 set/show screen-width set/show width
8549 set/show sevenbit-strings set/show print sevenbit-strings
8550 set/show unionprint set/show print union
8551 set/show vtblprint set/show print vtbl
8553 unset [No longer an alias for delete]
8559 \vskip \parskip\vskip \baselineskip
8560 \halign{\tt #\hfil &\qquad#&\tt #\hfil\cr
8561 {\bf Old Command} &&{\bf New Command}\cr
8562 add-syms &&add-symbol-file\cr
8563 delete environment &&unset environment\cr
8564 info convenience &&show convenience\cr
8565 info copying &&show copying\cr
8566 info directories &&show directories \cr
8567 info editing &&show commands\cr
8568 info history &&show values\cr
8569 info targets &&help target\cr
8570 info values &&show values\cr
8571 info version &&show version\cr
8572 info warranty &&show warranty\cr
8573 set{\rm / }show addressprint &&set{\rm / }show print address\cr
8574 set{\rm / }show array-max &&set{\rm / }show print elements\cr
8575 set{\rm / }show arrayprint &&set{\rm / }show print array\cr
8576 set{\rm / }show asm-demangle &&set{\rm / }show print asm-demangle\cr
8577 set{\rm / }show caution &&set{\rm / }show confirm\cr
8578 set{\rm / }show demangle &&set{\rm / }show print demangle\cr
8579 set{\rm / }show history write &&set{\rm / }show history save\cr
8580 set{\rm / }show prettyprint &&set{\rm / }show print pretty\cr
8581 set{\rm / }show screen-height &&set{\rm / }show height\cr
8582 set{\rm / }show screen-width &&set{\rm / }show width\cr
8583 set{\rm / }show sevenbit-strings &&set{\rm / }show print sevenbit-strings\cr
8584 set{\rm / }show unionprint &&set{\rm / }show print union\cr
8585 set{\rm / }show vtblprint &&set{\rm / }show print vtbl\cr
8587 unset &&\rm(No longer an alias for delete)\cr
8590 @c END TEXI2ROFF-KILL
8593 @ifclear PRECONFIGURED
8594 @node Formatting Documentation
8595 @appendix Formatting Documentation
8597 @cindex GDB reference card
8598 @cindex reference card
8599 The GDB 4 release includes an already-formatted reference card, ready
8600 for printing with PostScript or Ghostscript, in the @file{gdb}
8601 subdirectory of the main source directory@footnote{In
8602 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
8603 release.}. If you can use PostScript or Ghostscript with your printer,
8604 you can print the reference card immediately with @file{refcard.ps}.
8606 The release also includes the source for the reference card. You
8607 can format it, using @TeX{}, by typing:
8613 The GDB reference card is designed to print in landscape mode on US
8614 ``letter'' size paper; that is, on a sheet 11 inches wide by 8.5 inches
8615 high. You will need to specify this form of printing as an option to
8616 your @sc{dvi} output program.
8618 @cindex documentation
8620 All the documentation for GDB comes as part of the machine-readable
8621 distribution. The documentation is written in Texinfo format, which is
8622 a documentation system that uses a single source file to produce both
8623 on-line information and a printed manual. You can use one of the Info
8624 formatting commands to create the on-line version of the documentation
8625 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
8627 GDB includes an already formatted copy of the on-line Info version of
8628 this manual in the @file{gdb} subdirectory. The main Info file is
8629 @file{gdb-@var{version-number}/gdb/gdb.info}, and it refers to
8630 subordinate files matching @samp{gdb.info*} in the same directory. If
8631 necessary, you can print out these files, or read them with any editor;
8632 but they are easier to read using the @code{info} subsystem in GNU Emacs
8633 or the standalone @code{info} program, available as part of the GNU
8634 Texinfo distribution.
8636 If you want to format these Info files yourself, you need one of the
8637 Info formatting programs, such as @code{texinfo-format-buffer} or
8640 If you have @code{makeinfo} installed, and are in the top level GDB
8641 source directory (@file{gdb-@value{GDBVN}}, in the case of version @value{GDBVN}), you can
8642 make the Info file by typing:
8649 If you want to typeset and print copies of this manual, you need @TeX{},
8650 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
8651 Texinfo definitions file.
8653 @TeX{} is a typesetting program; it does not print files directly, but
8654 produces output files called @sc{dvi} files. To print a typeset
8655 document, you need a program to print @sc{dvi} files. If your system
8656 has @TeX{} installed, chances are it has such a program. The precise
8657 command to use depends on your system; @kbd{lpr -d} is common; another
8658 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
8659 require a file name without any extension or a @samp{.dvi} extension.
8661 @TeX{} also requires a macro definitions file called
8662 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
8663 written in Texinfo format. On its own, @TeX{} cannot read, much less
8664 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
8665 and is located in the @file{gdb-@var{version-number}/texinfo}
8668 If you have @TeX{} and a @sc{dvi} printer program installed, you can
8669 typeset and print this manual. First switch to the the @file{gdb}
8670 subdirectory of the main source directory (for example, to
8671 @file{gdb-@value{GDBVN}/gdb}) and then type:
8677 @node Installing GDB
8678 @appendix Installing GDB
8679 @cindex configuring GDB
8680 @cindex installation
8682 GDB comes with a @code{configure} script that automates the process
8683 of preparing GDB for installation; you can then use @code{make} to
8684 build the @code{gdb} program.
8686 @c irrelevant in info file; it's as current as the code it lives with.
8687 @footnote{If you have a more recent version of GDB than @value{GDBVN},
8688 look at the @file{README} file in the sources; we may have improved the
8689 installation procedures since publishing this manual.}
8692 The GDB distribution includes all the source code you need for GDB in
8693 a single directory, whose name is usually composed by appending the
8694 version number to @samp{gdb}.
8696 For example, the GDB version @value{GDBVN} distribution is in the
8697 @file{gdb-@value{GDBVN}} directory. That directory contains:
8700 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
8701 script for configuring GDB and all its supporting libraries.
8703 @item gdb-@value{GDBVN}/gdb
8704 the source specific to GDB itself
8706 @item gdb-@value{GDBVN}/bfd
8707 source for the Binary File Descriptor library
8709 @item gdb-@value{GDBVN}/include
8712 @item gdb-@value{GDBVN}/libiberty
8713 source for the @samp{-liberty} free software library
8715 @item gdb-@value{GDBVN}/opcodes
8716 source for the library of opcode tables and disassemblers
8718 @item gdb-@value{GDBVN}/readline
8719 source for the GNU command-line interface
8721 @item gdb-@value{GDBVN}/glob
8722 source for the GNU filename pattern-matching subroutine
8724 @item gdb-@value{GDBVN}/mmalloc
8725 source for the GNU memory-mapped malloc package
8728 The simplest way to configure and build GDB is to run @code{configure}
8729 from the @file{gdb-@var{version-number}} source directory, which in
8730 this example is the @file{gdb-@value{GDBVN}} directory.
8732 First switch to the @file{gdb-@var{version-number}} source directory
8733 if you are not already in it; then run @code{configure}. Pass the
8734 identifier for the platform on which GDB will run as an
8740 cd gdb-@value{GDBVN}
8741 ./configure @var{host}
8746 where @var{host} is an identifier such as @samp{sun4} or
8747 @samp{decstation}, that identifies the platform where GDB will run.
8748 (You can often leave off @var{host}; @code{configure} tries to guess the
8749 correct value by examining your system.)
8751 Running @samp{configure @var{host}} and then running @code{make} builds the
8752 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
8753 libraries, then @code{gdb} itself. The configured source files, and the
8754 binaries, are left in the corresponding source directories.
8756 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
8757 system does not recognize this automatically when you run a different
8758 shell, you may need to run @code{sh} on it explicitly:
8761 sh configure @var{host}
8764 If you run @code{configure} from a directory that contains source
8765 directories for multiple libraries or programs, such as the
8766 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
8767 creates configuration files for every directory level underneath (unless
8768 you tell it not to, with the @samp{--norecursion} option).
8770 You can run the @code{configure} script from any of the
8771 subordinate directories in the GDB distribution if you only want to
8772 configure that subdirectory, but be sure to specify a path to it.
8774 For example, with version @value{GDBVN}, type the following to configure only
8775 the @code{bfd} subdirectory:
8779 cd gdb-@value{GDBVN}/bfd
8780 ../configure @var{host}
8784 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
8785 However, you should make sure that the shell on your path (named by
8786 the @samp{SHELL} environment variable) is publicly readable. Remember
8787 that GDB uses the shell to start your program---some systems refuse to
8788 let GDB debug child processes whose programs are not readable.
8791 * Separate Objdir:: Compiling GDB in another directory
8792 * Config Names:: Specifying names for hosts and targets
8793 * configure Options:: Summary of options for configure
8796 @node Separate Objdir
8797 @section Compiling GDB in another directory
8799 If you want to run GDB versions for several host or target machines,
8800 you need a different @code{gdb} compiled for each combination of
8801 host and target. @code{configure} is designed to make this easy by
8802 allowing you to generate each configuration in a separate subdirectory,
8803 rather than in the source directory. If your @code{make} program
8804 handles the @samp{VPATH} feature (GNU @code{make} does), running
8805 @code{make} in each of these directories builds the @code{gdb}
8806 program specified there.
8808 To build @code{gdb} in a separate directory, run @code{configure}
8809 with the @samp{--srcdir} option to specify where to find the source.
8810 (You also need to specify a path to find @code{configure}
8811 itself from your working directory. If the path to @code{configure}
8812 would be the same as the argument to @samp{--srcdir}, you can leave out
8813 the @samp{--srcdir} option; it is assumed.)
8815 For example, with version @value{GDBVN}, you can build GDB in a separate
8816 directory for a Sun 4 like this:
8820 cd gdb-@value{GDBVN}
8823 ../gdb-@value{GDBVN}/configure sun4
8828 When @code{configure} builds a configuration using a remote source
8829 directory, it creates a tree for the binaries with the same structure
8830 (and using the same names) as the tree under the source directory. In
8831 the example, you'd find the Sun 4 library @file{libiberty.a} in the
8832 directory @file{gdb-sun4/libiberty}, and GDB itself in
8833 @file{gdb-sun4/gdb}.
8835 One popular reason to build several GDB configurations in separate
8836 directories is to configure GDB for cross-compiling (where GDB
8837 runs on one machine---the host---while debugging programs that run on
8838 another machine---the target). You specify a cross-debugging target by
8839 giving the @samp{--target=@var{target}} option to @code{configure}.
8841 When you run @code{make} to build a program or library, you must run
8842 it in a configured directory---whatever directory you were in when you
8843 called @code{configure} (or one of its subdirectories).
8845 The @code{Makefile} that @code{configure} generates in each source
8846 directory also runs recursively. If you type @code{make} in a source
8847 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
8848 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
8849 will build all the required libraries, and then build GDB.
8851 When you have multiple hosts or targets configured in separate
8852 directories, you can run @code{make} on them in parallel (for example,
8853 if they are NFS-mounted on each of the hosts); they will not interfere
8857 @section Specifying names for hosts and targets
8859 The specifications used for hosts and targets in the @code{configure}
8860 script are based on a three-part naming scheme, but some short predefined
8861 aliases are also supported. The full naming scheme encodes three pieces
8862 of information in the following pattern:
8865 @var{architecture}-@var{vendor}-@var{os}
8868 For example, you can use the alias @code{sun4} as a @var{host} argument,
8869 or as the value for @var{target} in a @code{--target=@var{target}}
8870 option. The equivalent full name is @samp{sparc-sun-sunos4}.
8872 The @code{configure} script accompanying GDB does not provide
8873 any query facility to list all supported host and target names or
8874 aliases. @code{configure} calls the Bourne shell script
8875 @code{config.sub} to map abbreviations to full names; you can read the
8876 script, if you wish, or you can use it to test your guesses on
8877 abbreviations---for example:
8880 % sh config.sub sun4
8881 sparc-sun-sunos4.1.1
8882 % sh config.sub sun3
8884 % sh config.sub decstation
8886 % sh config.sub hp300bsd
8888 % sh config.sub i386v
8890 % sh config.sub i786v
8891 Invalid configuration `i786v': machine `i786v' not recognized
8895 @code{config.sub} is also distributed in the GDB source
8896 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
8898 @node configure Options
8899 @section @code{configure} options
8901 Here is a summary of the @code{configure} options and arguments that
8902 are most often useful for building @value{GDBN}. @code{configure} also has
8903 several other options not listed here. @inforef{What Configure
8904 Does,,configure.info}, for a full explanation of @code{configure}.
8905 @c FIXME: Would this be more, or less, useful as an xref (ref to printed
8906 @c manual in the printed manual, ref to info file only from the info file)?
8909 configure @r{[}--help@r{]}
8910 @r{[}--prefix=@var{dir}@r{]}
8911 @r{[}--srcdir=@var{dirname}@r{]}
8912 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
8913 @r{[}--target=@var{target}@r{]} @var{host}
8917 You may introduce options with a single @samp{-} rather than
8918 @samp{--} if you prefer; but you may abbreviate option names if you use
8923 Display a quick summary of how to invoke @code{configure}.
8925 @item -prefix=@var{dir}
8926 Configure the source to install programs and files under directory
8929 @c avoid splitting the warning from the explanation:
8931 @item --srcdir=@var{dirname}
8932 @strong{Warning: using this option requires GNU @code{make}, or another
8933 @code{make} that implements the @code{VPATH} feature.}@*
8934 Use this option to make configurations in directories separate from the
8935 GDB source directories. Among other things, you can use this to
8936 build (or maintain) several configurations simultaneously, in separate
8937 directories. @code{configure} writes configuration specific files in
8938 the current directory, but arranges for them to use the source in the
8939 directory @var{dirname}. @code{configure} creates directories under
8940 the working directory in parallel to the source directories below
8944 Configure only the directory level where @code{configure} is executed; do not
8945 propagate configuration to subdirectories.
8948 @emph{Remove} files otherwise built during configuration.
8950 @c This does not work (yet if ever). FIXME.
8951 @c @item --parse=@var{lang} @dots{}
8952 @c Configure the GDB expression parser to parse the listed languages.
8953 @c @samp{all} configures GDB for all supported languages. To get a
8954 @c list of all supported languages, omit the argument. Without this
8955 @c option, GDB is configured to parse all supported languages.
8957 @item --target=@var{target}
8958 Configure GDB for cross-debugging programs running on the specified
8959 @var{target}. Without this option, GDB is configured to debug
8960 programs that run on the same machine (@var{host}) as GDB itself.
8962 There is no convenient way to generate a list of all available targets.
8964 @item @var{host} @dots{}
8965 Configure GDB to run on the specified @var{host}.
8967 There is no convenient way to generate a list of all available hosts.
8971 @code{configure} accepts other options, for compatibility with
8972 configuring other GNU tools recursively; but these are the only
8973 options that affect GDB or its supporting libraries.
8982 % I think something like @colophon should be in texinfo. In the
8984 \long\def\colophon{\hbox to0pt{}\vfill
8985 \centerline{The body of this manual is set in}
8986 \centerline{\fontname\tenrm,}
8987 \centerline{with headings in {\bf\fontname\tenbf}}
8988 \centerline{and examples in {\tt\fontname\tentt}.}
8989 \centerline{{\it\fontname\tenit\/},}
8990 \centerline{{\bf\fontname\tenbf}, and}
8991 \centerline{{\sl\fontname\tensl\/}}
8992 \centerline{are used for emphasis.}\vfill}
8994 % Blame: pesch@cygnus.com, 1991.