1 \input texinfo @c -*-texinfo-*-
3 @c Free Software Foundation, Inc.
6 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
7 @c of @set vars. However, you can override filename with makeinfo -o.
12 @settitle Debugging with @value{GDBN}
13 @setchapternewpage odd
24 @c readline appendices use @vindex, @findex and @ftable,
25 @c annotate.texi uses @findex.
32 @c !!set GDB manual's edition---not the same as GDB version!
35 @c !!set GDB manual's revision date
38 @c THIS MANUAL REQUIRES TEXINFO 3.12 OR LATER.
40 @c This is a dir.info fragment to support semi-automated addition of
41 @c manuals to an info tree.
42 @dircategory Programming & development tools.
44 * Gdb: (gdb). The @sc{gnu} debugger.
48 This file documents the @sc{gnu} debugger @value{GDBN}.
51 This is the @value{EDITION} Edition, @value{DATE},
52 of @cite{Debugging with @value{GDBN}: the @sc{gnu} Source-Level Debugger}
53 for @value{GDBN} Version @value{GDBVN}.
55 Copyright (C) 1988-2000 Free Software Foundation, Inc.
57 Permission is granted to make and distribute verbatim copies of
58 this manual provided the copyright notice and this permission notice
59 are preserved on all copies.
62 Permission is granted to process this file through TeX and print the
63 results, provided the printed document carries copying permission
64 notice identical to this one except for the removal of this paragraph
65 (this paragraph not being relevant to the printed manual).
68 Permission is granted to copy and distribute modified versions of this
69 manual under the conditions for verbatim copying, provided also that the
70 entire resulting derived work is distributed under the terms of a
71 permission notice identical to this one.
73 Permission is granted to copy and distribute translations of this manual
74 into another language, under the above conditions for modified versions.
78 @title Debugging with @value{GDBN}
79 @subtitle The @sc{gnu} Source-Level Debugger
81 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
82 @subtitle @value{DATE}
83 @author Richard M. Stallman and Roland H. Pesch
87 \hfill (Send bugs and comments on @value{GDBN} to bug-gdb\@gnu.org.)\par
88 \hfill {\it Debugging with @value{GDBN}}\par
89 \hfill \TeX{}info \texinfoversion\par
93 @vskip 0pt plus 1filll
94 Copyright @copyright{} 1988-2000 Free Software Foundation, Inc.
96 Published by the Free Software Foundation @*
97 59 Temple Place - Suite 330, @*
98 Boston, MA 02111-1307 USA @*
101 Permission is granted to make and distribute verbatim copies of
102 this manual provided the copyright notice and this permission notice
103 are preserved on all copies.
105 Permission is granted to copy and distribute modified versions of this
106 manual under the conditions for verbatim copying, provided also that the
107 entire resulting derived work is distributed under the terms of a
108 permission notice identical to this one.
110 Permission is granted to copy and distribute translations of this manual
111 into another language, under the above conditions for modified versions.
116 @node Top, Summary, (dir), (dir)
118 @top Debugging with @value{GDBN}
120 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
122 This is the @value{EDITION} Edition, @value{DATE}, for @value{GDBN} Version
125 Copyright (C) 1988-2000 Free Software Foundation, Inc.
128 * Summary:: Summary of @value{GDBN}
129 * Sample Session:: A sample @value{GDBN} session
131 * Invocation:: Getting in and out of @value{GDBN}
132 * Commands:: @value{GDBN} commands
133 * Running:: Running programs under @value{GDBN}
134 * Stopping:: Stopping and continuing
135 * Stack:: Examining the stack
136 * Source:: Examining source files
137 * Data:: Examining data
139 * Languages:: Using @value{GDBN} with different languages
141 * Symbols:: Examining the symbol table
142 * Altering:: Altering execution
143 * GDB Files:: @value{GDBN} files
144 * Targets:: Specifying a debugging target
145 * Configurations:: Configuration-specific information
146 * Controlling GDB:: Controlling @value{GDBN}
147 * Sequences:: Canned sequences of commands
148 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
149 * Annotations:: @value{GDBN}'s annotation interface.
150 * GDB/MI:: @value{GDBN}'s Machine Interface.
152 * GDB Bugs:: Reporting bugs in @value{GDBN}
153 * Formatting Documentation:: How to format and print @value{GDBN} documentation
155 * Command Line Editing:: Command Line Editing
156 * Using History Interactively:: Using History Interactively
157 * Installing GDB:: Installing GDB
163 @c the replication sucks, but this avoids a texinfo 3.12 lameness
168 @top Debugging with @value{GDBN}
170 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
172 This is the @value{EDITION} Edition, @value{DATE}, for @value{GDBN} Version
175 Copyright (C) 1988-2000 Free Software Foundation, Inc.
178 * Summary:: Summary of @value{GDBN}
179 * Sample Session:: A sample @value{GDBN} session
181 * Invocation:: Getting in and out of @value{GDBN}
182 * Commands:: @value{GDBN} commands
183 * Running:: Running programs under @value{GDBN}
184 * Stopping:: Stopping and continuing
185 * Stack:: Examining the stack
186 * Source:: Examining source files
187 * Data:: Examining data
189 * Languages:: Using @value{GDBN} with different languages
191 * Symbols:: Examining the symbol table
192 * Altering:: Altering execution
193 * GDB Files:: @value{GDBN} files
194 * Targets:: Specifying a debugging target
195 * Configurations:: Configuration-specific information
196 * Controlling GDB:: Controlling @value{GDBN}
197 * Sequences:: Canned sequences of commands
198 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
199 * Annotations:: @value{GDBN}'s annotation interface.
201 * GDB Bugs:: Reporting bugs in @value{GDBN}
202 * Formatting Documentation:: How to format and print @value{GDBN} documentation
204 * Command Line Editing:: Command Line Editing
205 * Using History Interactively:: Using History Interactively
206 * Installing GDB:: Installing GDB
213 @unnumbered Summary of @value{GDBN}
215 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
216 going on ``inside'' another program while it executes---or what another
217 program was doing at the moment it crashed.
219 @value{GDBN} can do four main kinds of things (plus other things in support of
220 these) to help you catch bugs in the act:
224 Start your program, specifying anything that might affect its behavior.
227 Make your program stop on specified conditions.
230 Examine what has happened, when your program has stopped.
233 Change things in your program, so you can experiment with correcting the
234 effects of one bug and go on to learn about another.
237 You can use @value{GDBN} to debug programs written in C and C++.
238 For more information, see @ref{Support,,Supported languages}.
239 For more information, see @ref{C,,C and C++}.
243 Support for Modula-2 and Chill is partial. For information on Modula-2,
244 see @ref{Modula-2,,Modula-2}. For information on Chill, see @ref{Chill}.
247 Debugging Pascal programs which use sets, subranges, file variables, or
248 nested functions does not currently work. @value{GDBN} does not support
249 entering expressions, printing values, or similar features using Pascal
253 @value{GDBN} can be used to debug programs written in Fortran, although
254 it may be necessary to refer to some variables with a trailing
258 * Free Software:: Freely redistributable software
259 * Contributors:: Contributors to GDB
263 @unnumberedsec Free software
265 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
266 General Public License
267 (GPL). The GPL gives you the freedom to copy or adapt a licensed
268 program---but every person getting a copy also gets with it the
269 freedom to modify that copy (which means that they must get access to
270 the source code), and the freedom to distribute further copies.
271 Typical software companies use copyrights to limit your freedoms; the
272 Free Software Foundation uses the GPL to preserve these freedoms.
274 Fundamentally, the General Public License is a license which says that
275 you have these freedoms and that you cannot take these freedoms away
279 @unnumberedsec Contributors to @value{GDBN}
281 Richard Stallman was the original author of @value{GDBN}, and of many
282 other @sc{gnu} programs. Many others have contributed to its
283 development. This section attempts to credit major contributors. One
284 of the virtues of free software is that everyone is free to contribute
285 to it; with regret, we cannot actually acknowledge everyone here. The
286 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
287 blow-by-blow account.
289 Changes much prior to version 2.0 are lost in the mists of time.
292 @emph{Plea:} Additions to this section are particularly welcome. If you
293 or your friends (or enemies, to be evenhanded) have been unfairly
294 omitted from this list, we would like to add your names!
297 So that they may not regard their many labors as thankless, we
298 particularly thank those who shepherded @value{GDBN} through major
300 Jim Blandy (release 4.18);
301 Jason Molenda (release 4.17);
302 Stan Shebs (release 4.14);
303 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
304 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
305 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
306 Jim Kingdon (releases 3.5, 3.4, and 3.3);
307 and Randy Smith (releases 3.2, 3.1, and 3.0).
309 Richard Stallman, assisted at various times by Peter TerMaat, Chris
310 Hanson, and Richard Mlynarik, handled releases through 2.8.
312 Michael Tiemann is the author of most of the @sc{gnu} C++ support in
313 @value{GDBN}, with significant additional contributions from Per
314 Bothner. James Clark wrote the @sc{gnu} C++ demangler. Early work on
315 C++ was by Peter TerMaat (who also did much general update work leading
318 @value{GDBN} 4 uses the BFD subroutine library to examine multiple
319 object-file formats; BFD was a joint project of David V.
320 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
322 David Johnson wrote the original COFF support; Pace Willison did
323 the original support for encapsulated COFF.
325 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
327 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
328 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
330 Jean-Daniel Fekete contributed Sun 386i support.
331 Chris Hanson improved the HP9000 support.
332 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
333 David Johnson contributed Encore Umax support.
334 Jyrki Kuoppala contributed Altos 3068 support.
335 Jeff Law contributed HP PA and SOM support.
336 Keith Packard contributed NS32K support.
337 Doug Rabson contributed Acorn Risc Machine support.
338 Bob Rusk contributed Harris Nighthawk CX-UX support.
339 Chris Smith contributed Convex support (and Fortran debugging).
340 Jonathan Stone contributed Pyramid support.
341 Michael Tiemann contributed SPARC support.
342 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
343 Pace Willison contributed Intel 386 support.
344 Jay Vosburgh contributed Symmetry support.
346 Andreas Schwab contributed M68K Linux support.
348 Rich Schaefer and Peter Schauer helped with support of SunOS shared
351 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
352 about several machine instruction sets.
354 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
355 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
356 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
357 and RDI targets, respectively.
359 Brian Fox is the author of the readline libraries providing
360 command-line editing and command history.
362 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
363 Modula-2 support, and contributed the Languages chapter of this manual.
365 Fred Fish wrote most of the support for Unix System Vr4.
366 He also enhanced the command-completion support to cover C++ overloaded
369 Hitachi America, Ltd. sponsored the support for H8/300, H8/500, and
372 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
374 Mitsubishi sponsored the support for D10V, D30V, and M32R/D processors.
376 Toshiba sponsored the support for the TX39 Mips processor.
378 Matsushita sponsored the support for the MN10200 and MN10300 processors.
380 Fujitsu sponsored the support for SPARClite and FR30 processors.
382 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
385 Michael Snyder added support for tracepoints.
387 Stu Grossman wrote gdbserver.
389 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
390 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
392 The following people at the Hewlett-Packard Company contributed
393 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
394 (narrow mode), HP's implementation of kernel threads, HP's aC++
395 compiler, and the terminal user interface: Ben Krepp, Richard Title,
396 John Bishop, Susan Macchia, Kathy Mann, Satish Pai, India Paul, Steve
397 Rehrauer, and Elena Zannoni. Kim Haase provided HP-specific
398 information in this manual.
400 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
401 development since 1991. Cygnus engineers who have worked on @value{GDBN}
402 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
403 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
404 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
405 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
406 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
407 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
408 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
409 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
410 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
411 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
412 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
413 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
414 Zuhn have made contributions both large and small.
418 @chapter A Sample @value{GDBN} Session
420 You can use this manual at your leisure to read all about @value{GDBN}.
421 However, a handful of commands are enough to get started using the
422 debugger. This chapter illustrates those commands.
425 In this sample session, we emphasize user input like this: @b{input},
426 to make it easier to pick out from the surrounding output.
429 @c FIXME: this example may not be appropriate for some configs, where
430 @c FIXME...primary interest is in remote use.
432 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
433 processor) exhibits the following bug: sometimes, when we change its
434 quote strings from the default, the commands used to capture one macro
435 definition within another stop working. In the following short @code{m4}
436 session, we define a macro @code{foo} which expands to @code{0000}; we
437 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
438 same thing. However, when we change the open quote string to
439 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
440 procedure fails to define a new synonym @code{baz}:
449 @b{define(bar,defn(`foo'))}
453 @b{changequote(<QUOTE>,<UNQUOTE>)}
455 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
458 m4: End of input: 0: fatal error: EOF in string
462 Let us use @value{GDBN} to try to see what is going on.
465 $ @b{@value{GDBP} m4}
466 @c FIXME: this falsifies the exact text played out, to permit smallbook
467 @c FIXME... format to come out better.
468 @value{GDBN} is free software and you are welcome to distribute copies
469 of it under certain conditions; type "show copying" to see
471 There is absolutely no warranty for @value{GDBN}; type "show warranty"
474 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
479 @value{GDBN} reads only enough symbol data to know where to find the
480 rest when needed; as a result, the first prompt comes up very quickly.
481 We now tell @value{GDBN} to use a narrower display width than usual, so
482 that examples fit in this manual.
485 (@value{GDBP}) @b{set width 70}
489 We need to see how the @code{m4} built-in @code{changequote} works.
490 Having looked at the source, we know the relevant subroutine is
491 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
492 @code{break} command.
495 (@value{GDBP}) @b{break m4_changequote}
496 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
500 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
501 control; as long as control does not reach the @code{m4_changequote}
502 subroutine, the program runs as usual:
505 (@value{GDBP}) @b{run}
506 Starting program: /work/Editorial/gdb/gnu/m4/m4
514 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
515 suspends execution of @code{m4}, displaying information about the
516 context where it stops.
519 @b{changequote(<QUOTE>,<UNQUOTE>)}
521 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
523 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
527 Now we use the command @code{n} (@code{next}) to advance execution to
528 the next line of the current function.
532 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
537 @code{set_quotes} looks like a promising subroutine. We can go into it
538 by using the command @code{s} (@code{step}) instead of @code{next}.
539 @code{step} goes to the next line to be executed in @emph{any}
540 subroutine, so it steps into @code{set_quotes}.
544 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
546 530 if (lquote != def_lquote)
550 The display that shows the subroutine where @code{m4} is now
551 suspended (and its arguments) is called a stack frame display. It
552 shows a summary of the stack. We can use the @code{backtrace}
553 command (which can also be spelled @code{bt}), to see where we are
554 in the stack as a whole: the @code{backtrace} command displays a
555 stack frame for each active subroutine.
558 (@value{GDBP}) @b{bt}
559 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
561 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
563 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
564 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
566 #4 0x79dc in expand_input () at macro.c:40
567 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
571 We step through a few more lines to see what happens. The first two
572 times, we can use @samp{s}; the next two times we use @code{n} to avoid
573 falling into the @code{xstrdup} subroutine.
577 0x3b5c 532 if (rquote != def_rquote)
579 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
580 def_lquote : xstrdup(lq);
582 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
585 538 len_lquote = strlen(rquote);
589 The last line displayed looks a little odd; we can examine the variables
590 @code{lquote} and @code{rquote} to see if they are in fact the new left
591 and right quotes we specified. We use the command @code{p}
592 (@code{print}) to see their values.
595 (@value{GDBP}) @b{p lquote}
596 $1 = 0x35d40 "<QUOTE>"
597 (@value{GDBP}) @b{p rquote}
598 $2 = 0x35d50 "<UNQUOTE>"
602 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
603 To look at some context, we can display ten lines of source
604 surrounding the current line with the @code{l} (@code{list}) command.
610 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
612 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
615 538 len_lquote = strlen(rquote);
616 539 len_rquote = strlen(lquote);
623 Let us step past the two lines that set @code{len_lquote} and
624 @code{len_rquote}, and then examine the values of those variables.
628 539 len_rquote = strlen(lquote);
631 (@value{GDBP}) @b{p len_lquote}
633 (@value{GDBP}) @b{p len_rquote}
638 That certainly looks wrong, assuming @code{len_lquote} and
639 @code{len_rquote} are meant to be the lengths of @code{lquote} and
640 @code{rquote} respectively. We can set them to better values using
641 the @code{p} command, since it can print the value of
642 any expression---and that expression can include subroutine calls and
646 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
648 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
653 Is that enough to fix the problem of using the new quotes with the
654 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
655 executing with the @code{c} (@code{continue}) command, and then try the
656 example that caused trouble initially:
662 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
669 Success! The new quotes now work just as well as the default ones. The
670 problem seems to have been just the two typos defining the wrong
671 lengths. We allow @code{m4} exit by giving it an EOF as input:
675 Program exited normally.
679 The message @samp{Program exited normally.} is from @value{GDBN}; it
680 indicates @code{m4} has finished executing. We can end our @value{GDBN}
681 session with the @value{GDBN} @code{quit} command.
684 (@value{GDBP}) @b{quit}
688 @chapter Getting In and Out of @value{GDBN}
690 This chapter discusses how to start @value{GDBN}, and how to get out of it.
694 type @samp{@value{GDBP}} to start @value{GDBN}.
696 type @kbd{quit} or @kbd{C-d} to exit.
700 * Invoking GDB:: How to start @value{GDBN}
701 * Quitting GDB:: How to quit @value{GDBN}
702 * Shell Commands:: How to use shell commands inside @value{GDBN}
706 @section Invoking @value{GDBN}
708 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
709 @value{GDBN} reads commands from the terminal until you tell it to exit.
711 You can also run @code{@value{GDBP}} with a variety of arguments and options,
712 to specify more of your debugging environment at the outset.
714 The command-line options described here are designed
715 to cover a variety of situations; in some environments, some of these
716 options may effectively be unavailable.
718 The most usual way to start @value{GDBN} is with one argument,
719 specifying an executable program:
722 @value{GDBP} @var{program}
726 You can also start with both an executable program and a core file
730 @value{GDBP} @var{program} @var{core}
733 You can, instead, specify a process ID as a second argument, if you want
734 to debug a running process:
737 @value{GDBP} @var{program} 1234
741 would attach @value{GDBN} to process @code{1234} (unless you also have a file
742 named @file{1234}; @value{GDBN} does check for a core file first).
744 Taking advantage of the second command-line argument requires a fairly
745 complete operating system; when you use @value{GDBN} as a remote
746 debugger attached to a bare board, there may not be any notion of
747 ``process'', and there is often no way to get a core dump. @value{GDBN}
748 will warn you if it is unable to attach or to read core dumps.
750 You can run @code{@value{GDBP}} without printing the front material, which describes
751 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
758 You can further control how @value{GDBN} starts up by using command-line
759 options. @value{GDBN} itself can remind you of the options available.
769 to display all available options and briefly describe their use
770 (@samp{@value{GDBP} -h} is a shorter equivalent).
772 All options and command line arguments you give are processed
773 in sequential order. The order makes a difference when the
774 @samp{-x} option is used.
778 * File Options:: Choosing files
779 * Mode Options:: Choosing modes
783 @subsection Choosing files
785 When @value{GDBN} starts, it reads any arguments other than options as
786 specifying an executable file and core file (or process ID). This is
787 the same as if the arguments were specified by the @samp{-se} and
788 @samp{-c} options respectively. (@value{GDBN} reads the first argument
789 that does not have an associated option flag as equivalent to the
790 @samp{-se} option followed by that argument; and the second argument
791 that does not have an associated option flag, if any, as equivalent to
792 the @samp{-c} option followed by that argument.)
794 If @value{GDBN} has not been configured to included core file support,
795 such as for most embedded targets, then it will complain about a second
796 argument and ignore it.
798 Many options have both long and short forms; both are shown in the
799 following list. @value{GDBN} also recognizes the long forms if you truncate
800 them, so long as enough of the option is present to be unambiguous.
801 (If you prefer, you can flag option arguments with @samp{--} rather
802 than @samp{-}, though we illustrate the more usual convention.)
804 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
805 @c way, both those who look for -foo and --foo in the index, will find
809 @item -symbols @var{file}
811 @cindex @code{--symbols}
813 Read symbol table from file @var{file}.
815 @item -exec @var{file}
817 @cindex @code{--exec}
819 Use file @var{file} as the executable file to execute when appropriate,
820 and for examining pure data in conjunction with a core dump.
824 Read symbol table from file @var{file} and use it as the executable
827 @item -core @var{file}
829 @cindex @code{--core}
831 Use file @var{file} as a core dump to examine.
833 @item -c @var{number}
834 Connect to process ID @var{number}, as with the @code{attach} command
835 (unless there is a file in core-dump format named @var{number}, in which
836 case @samp{-c} specifies that file as a core dump to read).
838 @item -command @var{file}
840 @cindex @code{--command}
842 Execute @value{GDBN} commands from file @var{file}. @xref{Command
843 Files,, Command files}.
845 @item -directory @var{directory}
846 @itemx -d @var{directory}
847 @cindex @code{--directory}
849 Add @var{directory} to the path to search for source files.
853 @cindex @code{--mapped}
855 @emph{Warning: this option depends on operating system facilities that are not
856 supported on all systems.}@*
857 If memory-mapped files are available on your system through the @code{mmap}
858 system call, you can use this option
859 to have @value{GDBN} write the symbols from your
860 program into a reusable file in the current directory. If the program you are debugging is
861 called @file{/tmp/fred}, the mapped symbol file is @file{/tmp/fred.syms}.
862 Future @value{GDBN} debugging sessions notice the presence of this file,
863 and can quickly map in symbol information from it, rather than reading
864 the symbol table from the executable program.
866 The @file{.syms} file is specific to the host machine where @value{GDBN}
867 is run. It holds an exact image of the internal @value{GDBN} symbol
868 table. It cannot be shared across multiple host platforms.
872 @cindex @code{--readnow}
874 Read each symbol file's entire symbol table immediately, rather than
875 the default, which is to read it incrementally as it is needed.
876 This makes startup slower, but makes future operations faster.
880 You typically combine the @code{-mapped} and @code{-readnow} options in
881 order to build a @file{.syms} file that contains complete symbol
882 information. (@xref{Files,,Commands to specify files}, for information
883 on @file{.syms} files.) A simple @value{GDBN} invocation to do nothing
884 but build a @file{.syms} file for future use is:
887 gdb -batch -nx -mapped -readnow programname
891 @subsection Choosing modes
893 You can run @value{GDBN} in various alternative modes---for example, in
894 batch mode or quiet mode.
901 Do not execute commands found in any initialization files (normally
902 called @file{.gdbinit}, or @file{gdb.ini} on PCs). Normally,
903 @value{GDBN} executes the commands in these files after all the command
904 options and arguments have been processed. @xref{Command Files,,Command
910 @cindex @code{--quiet}
911 @cindex @code{--silent}
913 ``Quiet''. Do not print the introductory and copyright messages. These
914 messages are also suppressed in batch mode.
917 @cindex @code{--batch}
918 Run in batch mode. Exit with status @code{0} after processing all the
919 command files specified with @samp{-x} (and all commands from
920 initialization files, if not inhibited with @samp{-n}). Exit with
921 nonzero status if an error occurs in executing the @value{GDBN} commands
922 in the command files.
924 Batch mode may be useful for running @value{GDBN} as a filter, for
925 example to download and run a program on another computer; in order to
926 make this more useful, the message
929 Program exited normally.
933 (which is ordinarily issued whenever a program running under
934 @value{GDBN} control terminates) is not issued when running in batch
939 @cindex @code{--nowindows}
941 ``No windows''. If @value{GDBN} comes with a graphical user interface
942 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
943 interface. If no GUI is available, this option has no effect.
947 @cindex @code{--windows}
949 If @value{GDBN} includes a GUI, then this option requires it to be
952 @item -cd @var{directory}
954 Run @value{GDBN} using @var{directory} as its working directory,
955 instead of the current directory.
959 @cindex @code{--fullname}
961 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
962 subprocess. It tells @value{GDBN} to output the full file name and line
963 number in a standard, recognizable fashion each time a stack frame is
964 displayed (which includes each time your program stops). This
965 recognizable format looks like two @samp{\032} characters, followed by
966 the file name, line number and character position separated by colons,
967 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
968 @samp{\032} characters as a signal to display the source code for the
972 @cindex @code{--epoch}
973 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
974 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
975 routines so as to allow Epoch to display values of expressions in a
978 @item -annotate @var{level}
979 @cindex @code{--annotate}
980 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
981 effect is identical to using @samp{set annotate @var{level}}
982 (@pxref{Annotations}).
983 Annotation level controls how much information does @value{GDBN} print
984 together with its prompt, values of expressions, source lines, and other
985 types of output. Level 0 is the normal, level 1 is for use when
986 @value{GDBN} is run as a subprocess of @sc{gnu} Emacs, level 2 is the
987 maximum annotation suitable for programs that control @value{GDBN}.
990 @cindex @code{--async}
991 Use the asynchronous event loop for the command-line interface.
992 @value{GDBN} processes all events, such as user keyboard input, via a
993 special event loop. This allows @value{GDBN} to accept and process user
994 commands in parallel with the debugged process being
995 run@footnote{@value{GDBN} built with @sc{djgpp} tools for
996 MS-DOS/MS-Windows supports this mode of operation, but the event loop is
997 suspended when the debuggee runs.}, so you don't need to wait for
998 control to return to @value{GDBN} before you type the next command.
999 (@emph{Note:} as of version 5.0, the target side of the asynchronous
1000 operation is not yet in place, so @samp{-async} does not work fully
1002 @c FIXME: when the target side of the event loop is done, the above NOTE
1003 @c should be removed.
1005 When the standard input is connected to a terminal device, @value{GDBN}
1006 uses the asynchronous event loop by default, unless disabled by the
1007 @samp{-noasync} option.
1010 @cindex @code{--noasync}
1011 Disable the asynchronous event loop for the command-line interface.
1013 @item -baud @var{bps}
1015 @cindex @code{--baud}
1017 Set the line speed (baud rate or bits per second) of any serial
1018 interface used by @value{GDBN} for remote debugging.
1020 @item -tty @var{device}
1021 @itemx -t @var{device}
1022 @cindex @code{--tty}
1024 Run using @var{device} for your program's standard input and output.
1025 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1027 @c resolve the situation of these eventually
1029 @c @cindex @code{--tui}
1030 @c Use a Terminal User Interface. For information, use your Web browser to
1031 @c read the file @file{TUI.html}, which is usually installed in the
1032 @c directory @code{/opt/langtools/wdb/doc} on HP-UX systems. Do not use
1033 @c this option if you run @value{GDBN} from Emacs (see @pxref{Emacs, ,Using
1034 @c @value{GDBN} under @sc{gnu} Emacs}).
1037 @c @cindex @code{--xdb}
1038 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1039 @c For information, see the file @file{xdb_trans.html}, which is usually
1040 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1043 @item -interpreter @var{interp}
1044 @cindex @code{--interpreter}
1045 Use the interpreter @var{interp} for interface with the controlling
1046 program or device. This option is meant to be set by programs which
1047 communicate with @value{GDBN} using it as a back end. For example,
1048 @samp{--interpreter=mi} causes @value{GDBN} to use the @dfn{gdbmi
1049 interface} (@pxref{GDB/MI, , The @sc{gdb/mi} Interface}).
1052 @cindex @code{--write}
1053 Open the executable and core files for both reading and writing. This
1054 is equivalent to the @samp{set write on} command inside @value{GDBN}
1058 @cindex @code{--statistics}
1059 This option causes @value{GDBN} to print statistics about time and
1060 memory usage after it completes each command and returns to the prompt.
1063 @cindex @code{--version}
1064 This option causes @value{GDBN} to print its version number and
1065 no-warranty blurb, and exit.
1070 @section Quitting @value{GDBN}
1071 @cindex exiting @value{GDBN}
1072 @cindex leaving @value{GDBN}
1075 @kindex quit @r{[}@var{expression}@r{]}
1076 @kindex q @r{(@code{quit})}
1077 @item quit @r{[}@var{expression}@r{]}
1079 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1080 @code{q}), or type an end-of-file character (usually @kbd{C-d}). If you
1081 do not supply @var{expression}, @value{GDBN} will terminate normally;
1082 otherwise it will terminate using the result of @var{expression} as the
1087 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
1088 terminates the action of any @value{GDBN} command that is in progress and
1089 returns to @value{GDBN} command level. It is safe to type the interrupt
1090 character at any time because @value{GDBN} does not allow it to take effect
1091 until a time when it is safe.
1093 If you have been using @value{GDBN} to control an attached process or
1094 device, you can release it with the @code{detach} command
1095 (@pxref{Attach, ,Debugging an already-running process}).
1097 @node Shell Commands
1098 @section Shell commands
1100 If you need to execute occasional shell commands during your
1101 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1102 just use the @code{shell} command.
1106 @cindex shell escape
1107 @item shell @var{command string}
1108 Invoke a standard shell to execute @var{command string}.
1109 If it exists, the environment variable @code{SHELL} determines which
1110 shell to run. Otherwise @value{GDBN} uses the default shell
1111 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1114 The utility @code{make} is often needed in development environments.
1115 You do not have to use the @code{shell} command for this purpose in
1120 @cindex calling make
1121 @item make @var{make-args}
1122 Execute the @code{make} program with the specified
1123 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1127 @chapter @value{GDBN} Commands
1129 You can abbreviate a @value{GDBN} command to the first few letters of the command
1130 name, if that abbreviation is unambiguous; and you can repeat certain
1131 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1132 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1133 show you the alternatives available, if there is more than one possibility).
1136 * Command Syntax:: How to give commands to @value{GDBN}
1137 * Completion:: Command completion
1138 * Help:: How to ask @value{GDBN} for help
1141 @node Command Syntax
1142 @section Command syntax
1144 A @value{GDBN} command is a single line of input. There is no limit on
1145 how long it can be. It starts with a command name, which is followed by
1146 arguments whose meaning depends on the command name. For example, the
1147 command @code{step} accepts an argument which is the number of times to
1148 step, as in @samp{step 5}. You can also use the @code{step} command
1149 with no arguments. Some commands do not allow any arguments.
1151 @cindex abbreviation
1152 @value{GDBN} command names may always be truncated if that abbreviation is
1153 unambiguous. Other possible command abbreviations are listed in the
1154 documentation for individual commands. In some cases, even ambiguous
1155 abbreviations are allowed; for example, @code{s} is specially defined as
1156 equivalent to @code{step} even though there are other commands whose
1157 names start with @code{s}. You can test abbreviations by using them as
1158 arguments to the @code{help} command.
1160 @cindex repeating commands
1161 @kindex RET @r{(repeat last command)}
1162 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1163 repeat the previous command. Certain commands (for example, @code{run})
1164 will not repeat this way; these are commands whose unintentional
1165 repetition might cause trouble and which you are unlikely to want to
1168 The @code{list} and @code{x} commands, when you repeat them with
1169 @key{RET}, construct new arguments rather than repeating
1170 exactly as typed. This permits easy scanning of source or memory.
1172 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1173 output, in a way similar to the common utility @code{more}
1174 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1175 @key{RET} too many in this situation, @value{GDBN} disables command
1176 repetition after any command that generates this sort of display.
1178 @kindex # @r{(a comment)}
1180 Any text from a @kbd{#} to the end of the line is a comment; it does
1181 nothing. This is useful mainly in command files (@pxref{Command
1182 Files,,Command files}).
1185 @section Command completion
1188 @cindex word completion
1189 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1190 only one possibility; it can also show you what the valid possibilities
1191 are for the next word in a command, at any time. This works for @value{GDBN}
1192 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1194 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1195 of a word. If there is only one possibility, @value{GDBN} fills in the
1196 word, and waits for you to finish the command (or press @key{RET} to
1197 enter it). For example, if you type
1199 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1200 @c complete accuracy in these examples; space introduced for clarity.
1201 @c If texinfo enhancements make it unnecessary, it would be nice to
1202 @c replace " @key" by "@key" in the following...
1204 (@value{GDBP}) info bre @key{TAB}
1208 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1209 the only @code{info} subcommand beginning with @samp{bre}:
1212 (@value{GDBP}) info breakpoints
1216 You can either press @key{RET} at this point, to run the @code{info
1217 breakpoints} command, or backspace and enter something else, if
1218 @samp{breakpoints} does not look like the command you expected. (If you
1219 were sure you wanted @code{info breakpoints} in the first place, you
1220 might as well just type @key{RET} immediately after @samp{info bre},
1221 to exploit command abbreviations rather than command completion).
1223 If there is more than one possibility for the next word when you press
1224 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1225 characters and try again, or just press @key{TAB} a second time;
1226 @value{GDBN} displays all the possible completions for that word. For
1227 example, you might want to set a breakpoint on a subroutine whose name
1228 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1229 just sounds the bell. Typing @key{TAB} again displays all the
1230 function names in your program that begin with those characters, for
1234 (@value{GDBP}) b make_ @key{TAB}
1235 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1236 make_a_section_from_file make_environ
1237 make_abs_section make_function_type
1238 make_blockvector make_pointer_type
1239 make_cleanup make_reference_type
1240 make_command make_symbol_completion_list
1241 (@value{GDBP}) b make_
1245 After displaying the available possibilities, @value{GDBN} copies your
1246 partial input (@samp{b make_} in the example) so you can finish the
1249 If you just want to see the list of alternatives in the first place, you
1250 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1251 means @kbd{@key{META} ?}. You can type this either by holding down a
1252 key designated as the @key{META} shift on your keyboard (if there is
1253 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1255 @cindex quotes in commands
1256 @cindex completion of quoted strings
1257 Sometimes the string you need, while logically a ``word'', may contain
1258 parentheses or other characters that @value{GDBN} normally excludes from
1259 its notion of a word. To permit word completion to work in this
1260 situation, you may enclose words in @code{'} (single quote marks) in
1261 @value{GDBN} commands.
1263 The most likely situation where you might need this is in typing the
1264 name of a C++ function. This is because C++ allows function overloading
1265 (multiple definitions of the same function, distinguished by argument
1266 type). For example, when you want to set a breakpoint you may need to
1267 distinguish whether you mean the version of @code{name} that takes an
1268 @code{int} parameter, @code{name(int)}, or the version that takes a
1269 @code{float} parameter, @code{name(float)}. To use the word-completion
1270 facilities in this situation, type a single quote @code{'} at the
1271 beginning of the function name. This alerts @value{GDBN} that it may need to
1272 consider more information than usual when you press @key{TAB} or
1273 @kbd{M-?} to request word completion:
1276 (@value{GDBP}) b 'bubble( @kbd{M-?}
1277 bubble(double,double) bubble(int,int)
1278 (@value{GDBP}) b 'bubble(
1281 In some cases, @value{GDBN} can tell that completing a name requires using
1282 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1283 completing as much as it can) if you do not type the quote in the first
1287 (@value{GDBP}) b bub @key{TAB}
1288 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1289 (@value{GDBP}) b 'bubble(
1293 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1294 you have not yet started typing the argument list when you ask for
1295 completion on an overloaded symbol.
1297 For more information about overloaded functions, see @ref{C plus plus
1298 expressions, ,C++ expressions}. You can use the command @code{set
1299 overload-resolution off} to disable overload resolution;
1300 see @ref{Debugging C plus plus, ,@value{GDBN} features for C++}.
1304 @section Getting help
1305 @cindex online documentation
1308 You can always ask @value{GDBN} itself for information on its commands,
1309 using the command @code{help}.
1312 @kindex h @r{(@code{help})}
1315 You can use @code{help} (abbreviated @code{h}) with no arguments to
1316 display a short list of named classes of commands:
1320 List of classes of commands:
1322 aliases -- Aliases of other commands
1323 breakpoints -- Making program stop at certain points
1324 data -- Examining data
1325 files -- Specifying and examining files
1326 internals -- Maintenance commands
1327 obscure -- Obscure features
1328 running -- Running the program
1329 stack -- Examining the stack
1330 status -- Status inquiries
1331 support -- Support facilities
1332 tracepoints -- Tracing of program execution without@*
1333 stopping the program
1334 user-defined -- User-defined commands
1336 Type "help" followed by a class name for a list of
1337 commands in that class.
1338 Type "help" followed by command name for full
1340 Command name abbreviations are allowed if unambiguous.
1343 @c the above line break eliminates huge line overfull...
1345 @item help @var{class}
1346 Using one of the general help classes as an argument, you can get a
1347 list of the individual commands in that class. For example, here is the
1348 help display for the class @code{status}:
1351 (@value{GDBP}) help status
1356 @c Line break in "show" line falsifies real output, but needed
1357 @c to fit in smallbook page size.
1358 info -- Generic command for showing things
1359 about the program being debugged
1360 show -- Generic command for showing things
1363 Type "help" followed by command name for full
1365 Command name abbreviations are allowed if unambiguous.
1369 @item help @var{command}
1370 With a command name as @code{help} argument, @value{GDBN} displays a
1371 short paragraph on how to use that command.
1374 @item apropos @var{args}
1375 The @code{apropos @var{args}} command searches through all of the @value{GDBN}
1376 commands, and their documentation, for the regular expression specified in
1377 @var{args}. It prints out all matches found. For example:
1383 @noindent results in:
1387 set symbol-reloading -- Set dynamic symbol table reloading
1388 multiple times in one run
1389 show symbol-reloading -- Show dynamic symbol table reloading
1390 multiple times in one run
1395 @item complete @var{args}
1396 The @code{complete @var{args}} command lists all the possible completions
1397 for the beginning of a command. Use @var{args} to specify the beginning of the
1398 command you want completed. For example:
1404 @noindent results in:
1415 @noindent This is intended for use by @sc{gnu} Emacs.
1418 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1419 and @code{show} to inquire about the state of your program, or the state
1420 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1421 manual introduces each of them in the appropriate context. The listings
1422 under @code{info} and under @code{show} in the Index point to
1423 all the sub-commands. @xref{Index}.
1428 @kindex i @r{(@code{info})}
1430 This command (abbreviated @code{i}) is for describing the state of your
1431 program. For example, you can list the arguments given to your program
1432 with @code{info args}, list the registers currently in use with @code{info
1433 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1434 You can get a complete list of the @code{info} sub-commands with
1435 @w{@code{help info}}.
1439 You can assign the result of an expression to an environment variable with
1440 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1441 @code{set prompt $}.
1445 In contrast to @code{info}, @code{show} is for describing the state of
1446 @value{GDBN} itself.
1447 You can change most of the things you can @code{show}, by using the
1448 related command @code{set}; for example, you can control what number
1449 system is used for displays with @code{set radix}, or simply inquire
1450 which is currently in use with @code{show radix}.
1453 To display all the settable parameters and their current
1454 values, you can use @code{show} with no arguments; you may also use
1455 @code{info set}. Both commands produce the same display.
1456 @c FIXME: "info set" violates the rule that "info" is for state of
1457 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1458 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1462 Here are three miscellaneous @code{show} subcommands, all of which are
1463 exceptional in lacking corresponding @code{set} commands:
1466 @kindex show version
1467 @cindex version number
1469 Show what version of @value{GDBN} is running. You should include this
1470 information in @value{GDBN} bug-reports. If multiple versions of
1471 @value{GDBN} are in use at your site, you may need to determine which
1472 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1473 commands are introduced, and old ones may wither away. Also, many
1474 system vendors ship variant versions of @value{GDBN}, and there are
1475 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1476 The version number is the same as the one announced when you start
1479 @kindex show copying
1481 Display information about permission for copying @value{GDBN}.
1483 @kindex show warranty
1485 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1486 if your version of @value{GDBN} comes with one.
1491 @chapter Running Programs Under @value{GDBN}
1493 When you run a program under @value{GDBN}, you must first generate
1494 debugging information when you compile it.
1496 You may start @value{GDBN} with its arguments, if any, in an environment
1497 of your choice. If you are doing native debugging, you may redirect
1498 your program's input and output, debug an already running process, or
1499 kill a child process.
1502 * Compilation:: Compiling for debugging
1503 * Starting:: Starting your program
1504 * Arguments:: Your program's arguments
1505 * Environment:: Your program's environment
1507 * Working Directory:: Your program's working directory
1508 * Input/Output:: Your program's input and output
1509 * Attach:: Debugging an already-running process
1510 * Kill Process:: Killing the child process
1512 * Threads:: Debugging programs with multiple threads
1513 * Processes:: Debugging programs with multiple processes
1517 @section Compiling for debugging
1519 In order to debug a program effectively, you need to generate
1520 debugging information when you compile it. This debugging information
1521 is stored in the object file; it describes the data type of each
1522 variable or function and the correspondence between source line numbers
1523 and addresses in the executable code.
1525 To request debugging information, specify the @samp{-g} option when you run
1528 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1529 options together. Using those compilers, you cannot generate optimized
1530 executables containing debugging information.
1532 @value{NGCC}, the @sc{gnu} C compiler, supports @samp{-g} with or
1533 without @samp{-O}, making it possible to debug optimized code. We
1534 recommend that you @emph{always} use @samp{-g} whenever you compile a
1535 program. You may think your program is correct, but there is no sense
1536 in pushing your luck.
1538 @cindex optimized code, debugging
1539 @cindex debugging optimized code
1540 When you debug a program compiled with @samp{-g -O}, remember that the
1541 optimizer is rearranging your code; the debugger shows you what is
1542 really there. Do not be too surprised when the execution path does not
1543 exactly match your source file! An extreme example: if you define a
1544 variable, but never use it, @value{GDBN} never sees that
1545 variable---because the compiler optimizes it out of existence.
1547 Some things do not work as well with @samp{-g -O} as with just
1548 @samp{-g}, particularly on machines with instruction scheduling. If in
1549 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1550 please report it to us as a bug (including a test case!).
1552 Older versions of the @sc{gnu} C compiler permitted a variant option
1553 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1554 format; if your @sc{gnu} C compiler has this option, do not use it.
1558 @section Starting your program
1564 @kindex r @r{(@code{run})}
1567 Use the @code{run} command to start your program under @value{GDBN}.
1568 You must first specify the program name (except on VxWorks) with an
1569 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1570 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1571 (@pxref{Files, ,Commands to specify files}).
1575 If you are running your program in an execution environment that
1576 supports processes, @code{run} creates an inferior process and makes
1577 that process run your program. (In environments without processes,
1578 @code{run} jumps to the start of your program.)
1580 The execution of a program is affected by certain information it
1581 receives from its superior. @value{GDBN} provides ways to specify this
1582 information, which you must do @emph{before} starting your program. (You
1583 can change it after starting your program, but such changes only affect
1584 your program the next time you start it.) This information may be
1585 divided into four categories:
1588 @item The @emph{arguments.}
1589 Specify the arguments to give your program as the arguments of the
1590 @code{run} command. If a shell is available on your target, the shell
1591 is used to pass the arguments, so that you may use normal conventions
1592 (such as wildcard expansion or variable substitution) in describing
1594 In Unix systems, you can control which shell is used with the
1595 @code{SHELL} environment variable.
1596 @xref{Arguments, ,Your program's arguments}.
1598 @item The @emph{environment.}
1599 Your program normally inherits its environment from @value{GDBN}, but you can
1600 use the @value{GDBN} commands @code{set environment} and @code{unset
1601 environment} to change parts of the environment that affect
1602 your program. @xref{Environment, ,Your program's environment}.
1604 @item The @emph{working directory.}
1605 Your program inherits its working directory from @value{GDBN}. You can set
1606 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1607 @xref{Working Directory, ,Your program's working directory}.
1609 @item The @emph{standard input and output.}
1610 Your program normally uses the same device for standard input and
1611 standard output as @value{GDBN} is using. You can redirect input and output
1612 in the @code{run} command line, or you can use the @code{tty} command to
1613 set a different device for your program.
1614 @xref{Input/Output, ,Your program's input and output}.
1617 @emph{Warning:} While input and output redirection work, you cannot use
1618 pipes to pass the output of the program you are debugging to another
1619 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1623 When you issue the @code{run} command, your program begins to execute
1624 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1625 of how to arrange for your program to stop. Once your program has
1626 stopped, you may call functions in your program, using the @code{print}
1627 or @code{call} commands. @xref{Data, ,Examining Data}.
1629 If the modification time of your symbol file has changed since the last
1630 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1631 table, and reads it again. When it does this, @value{GDBN} tries to retain
1632 your current breakpoints.
1635 @section Your program's arguments
1637 @cindex arguments (to your program)
1638 The arguments to your program can be specified by the arguments of the
1640 They are passed to a shell, which expands wildcard characters and
1641 performs redirection of I/O, and thence to your program. Your
1642 @code{SHELL} environment variable (if it exists) specifies what shell
1643 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
1644 the default shell (@file{/bin/sh} on Unix).
1646 On non-Unix systems, the program is usually invoked directly by
1647 @value{GDBN}, which emulates I/O redirection via the appropriate system
1648 calls, and the wildcard characters are expanded by the startup code of
1649 the program, not by the shell.
1651 @code{run} with no arguments uses the same arguments used by the previous
1652 @code{run}, or those set by the @code{set args} command.
1657 Specify the arguments to be used the next time your program is run. If
1658 @code{set args} has no arguments, @code{run} executes your program
1659 with no arguments. Once you have run your program with arguments,
1660 using @code{set args} before the next @code{run} is the only way to run
1661 it again without arguments.
1665 Show the arguments to give your program when it is started.
1669 @section Your program's environment
1671 @cindex environment (of your program)
1672 The @dfn{environment} consists of a set of environment variables and
1673 their values. Environment variables conventionally record such things as
1674 your user name, your home directory, your terminal type, and your search
1675 path for programs to run. Usually you set up environment variables with
1676 the shell and they are inherited by all the other programs you run. When
1677 debugging, it can be useful to try running your program with a modified
1678 environment without having to start @value{GDBN} over again.
1682 @item path @var{directory}
1683 Add @var{directory} to the front of the @code{PATH} environment variable
1684 (the search path for executables), for both @value{GDBN} and your program.
1685 You may specify several directory names, separated by whitespace or by a
1686 system-dependent separator character (@samp{:} on Unix, @samp{;} on
1687 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
1688 is moved to the front, so it is searched sooner.
1690 You can use the string @samp{$cwd} to refer to whatever is the current
1691 working directory at the time @value{GDBN} searches the path. If you
1692 use @samp{.} instead, it refers to the directory where you executed the
1693 @code{path} command. @value{GDBN} replaces @samp{.} in the
1694 @var{directory} argument (with the current path) before adding
1695 @var{directory} to the search path.
1696 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1697 @c document that, since repeating it would be a no-op.
1701 Display the list of search paths for executables (the @code{PATH}
1702 environment variable).
1704 @kindex show environment
1705 @item show environment @r{[}@var{varname}@r{]}
1706 Print the value of environment variable @var{varname} to be given to
1707 your program when it starts. If you do not supply @var{varname},
1708 print the names and values of all environment variables to be given to
1709 your program. You can abbreviate @code{environment} as @code{env}.
1711 @kindex set environment
1712 @item set environment @var{varname} @r{[}=@var{value}@r{]}
1713 Set environment variable @var{varname} to @var{value}. The value
1714 changes for your program only, not for @value{GDBN} itself. @var{value} may
1715 be any string; the values of environment variables are just strings, and
1716 any interpretation is supplied by your program itself. The @var{value}
1717 parameter is optional; if it is eliminated, the variable is set to a
1719 @c "any string" here does not include leading, trailing
1720 @c blanks. Gnu asks: does anyone care?
1722 For example, this command:
1729 tells the debugged program, when subsequently run, that its user is named
1730 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1731 are not actually required.)
1733 @kindex unset environment
1734 @item unset environment @var{varname}
1735 Remove variable @var{varname} from the environment to be passed to your
1736 program. This is different from @samp{set env @var{varname} =};
1737 @code{unset environment} removes the variable from the environment,
1738 rather than assigning it an empty value.
1741 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
1743 by your @code{SHELL} environment variable if it exists (or
1744 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1745 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1746 @file{.bashrc} for BASH---any variables you set in that file affect
1747 your program. You may wish to move setting of environment variables to
1748 files that are only run when you sign on, such as @file{.login} or
1751 @node Working Directory
1752 @section Your program's working directory
1754 @cindex working directory (of your program)
1755 Each time you start your program with @code{run}, it inherits its
1756 working directory from the current working directory of @value{GDBN}.
1757 The @value{GDBN} working directory is initially whatever it inherited
1758 from its parent process (typically the shell), but you can specify a new
1759 working directory in @value{GDBN} with the @code{cd} command.
1761 The @value{GDBN} working directory also serves as a default for the commands
1762 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1767 @item cd @var{directory}
1768 Set the @value{GDBN} working directory to @var{directory}.
1772 Print the @value{GDBN} working directory.
1776 @section Your program's input and output
1781 By default, the program you run under @value{GDBN} does input and output to
1782 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
1783 to its own terminal modes to interact with you, but it records the terminal
1784 modes your program was using and switches back to them when you continue
1785 running your program.
1788 @kindex info terminal
1790 Displays information recorded by @value{GDBN} about the terminal modes your
1794 You can redirect your program's input and/or output using shell
1795 redirection with the @code{run} command. For example,
1802 starts your program, diverting its output to the file @file{outfile}.
1805 @cindex controlling terminal
1806 Another way to specify where your program should do input and output is
1807 with the @code{tty} command. This command accepts a file name as
1808 argument, and causes this file to be the default for future @code{run}
1809 commands. It also resets the controlling terminal for the child
1810 process, for future @code{run} commands. For example,
1817 directs that processes started with subsequent @code{run} commands
1818 default to do input and output on the terminal @file{/dev/ttyb} and have
1819 that as their controlling terminal.
1821 An explicit redirection in @code{run} overrides the @code{tty} command's
1822 effect on the input/output device, but not its effect on the controlling
1825 When you use the @code{tty} command or redirect input in the @code{run}
1826 command, only the input @emph{for your program} is affected. The input
1827 for @value{GDBN} still comes from your terminal.
1830 @section Debugging an already-running process
1835 @item attach @var{process-id}
1836 This command attaches to a running process---one that was started
1837 outside @value{GDBN}. (@code{info files} shows your active
1838 targets.) The command takes as argument a process ID. The usual way to
1839 find out the process-id of a Unix process is with the @code{ps} utility,
1840 or with the @samp{jobs -l} shell command.
1842 @code{attach} does not repeat if you press @key{RET} a second time after
1843 executing the command.
1846 To use @code{attach}, your program must be running in an environment
1847 which supports processes; for example, @code{attach} does not work for
1848 programs on bare-board targets that lack an operating system. You must
1849 also have permission to send the process a signal.
1851 When you use @code{attach}, the debugger finds the program running in
1852 the process first by looking in the current working directory, then (if
1853 the program is not found) by using the source file search path
1854 (@pxref{Source Path, ,Specifying source directories}). You can also use
1855 the @code{file} command to load the program. @xref{Files, ,Commands to
1858 The first thing @value{GDBN} does after arranging to debug the specified
1859 process is to stop it. You can examine and modify an attached process
1860 with all the @value{GDBN} commands that are ordinarily available when
1861 you start processes with @code{run}. You can insert breakpoints; you
1862 can step and continue; you can modify storage. If you would rather the
1863 process continue running, you may use the @code{continue} command after
1864 attaching @value{GDBN} to the process.
1869 When you have finished debugging the attached process, you can use the
1870 @code{detach} command to release it from @value{GDBN} control. Detaching
1871 the process continues its execution. After the @code{detach} command,
1872 that process and @value{GDBN} become completely independent once more, and you
1873 are ready to @code{attach} another process or start one with @code{run}.
1874 @code{detach} does not repeat if you press @key{RET} again after
1875 executing the command.
1878 If you exit @value{GDBN} or use the @code{run} command while you have an
1879 attached process, you kill that process. By default, @value{GDBN} asks
1880 for confirmation if you try to do either of these things; you can
1881 control whether or not you need to confirm by using the @code{set
1882 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
1886 @section Killing the child process
1891 Kill the child process in which your program is running under @value{GDBN}.
1894 This command is useful if you wish to debug a core dump instead of a
1895 running process. @value{GDBN} ignores any core dump file while your program
1898 On some operating systems, a program cannot be executed outside @value{GDBN}
1899 while you have breakpoints set on it inside @value{GDBN}. You can use the
1900 @code{kill} command in this situation to permit running your program
1901 outside the debugger.
1903 The @code{kill} command is also useful if you wish to recompile and
1904 relink your program, since on many systems it is impossible to modify an
1905 executable file while it is running in a process. In this case, when you
1906 next type @code{run}, @value{GDBN} notices that the file has changed, and
1907 reads the symbol table again (while trying to preserve your current
1908 breakpoint settings).
1911 @section Debugging programs with multiple threads
1913 @cindex threads of execution
1914 @cindex multiple threads
1915 @cindex switching threads
1916 In some operating systems, such as HP-UX and Solaris, a single program
1917 may have more than one @dfn{thread} of execution. The precise semantics
1918 of threads differ from one operating system to another, but in general
1919 the threads of a single program are akin to multiple processes---except
1920 that they share one address space (that is, they can all examine and
1921 modify the same variables). On the other hand, each thread has its own
1922 registers and execution stack, and perhaps private memory.
1924 @value{GDBN} provides these facilities for debugging multi-thread
1928 @item automatic notification of new threads
1929 @item @samp{thread @var{threadno}}, a command to switch among threads
1930 @item @samp{info threads}, a command to inquire about existing threads
1931 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
1932 a command to apply a command to a list of threads
1933 @item thread-specific breakpoints
1937 @emph{Warning:} These facilities are not yet available on every
1938 @value{GDBN} configuration where the operating system supports threads.
1939 If your @value{GDBN} does not support threads, these commands have no
1940 effect. For example, a system without thread support shows no output
1941 from @samp{info threads}, and always rejects the @code{thread} command,
1945 (@value{GDBP}) info threads
1946 (@value{GDBP}) thread 1
1947 Thread ID 1 not known. Use the "info threads" command to
1948 see the IDs of currently known threads.
1950 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
1951 @c doesn't support threads"?
1954 @cindex focus of debugging
1955 @cindex current thread
1956 The @value{GDBN} thread debugging facility allows you to observe all
1957 threads while your program runs---but whenever @value{GDBN} takes
1958 control, one thread in particular is always the focus of debugging.
1959 This thread is called the @dfn{current thread}. Debugging commands show
1960 program information from the perspective of the current thread.
1962 @cindex @code{New} @var{systag} message
1963 @cindex thread identifier (system)
1964 @c FIXME-implementors!! It would be more helpful if the [New...] message
1965 @c included GDB's numeric thread handle, so you could just go to that
1966 @c thread without first checking `info threads'.
1967 Whenever @value{GDBN} detects a new thread in your program, it displays
1968 the target system's identification for the thread with a message in the
1969 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
1970 whose form varies depending on the particular system. For example, on
1971 LynxOS, you might see
1974 [New process 35 thread 27]
1978 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
1979 the @var{systag} is simply something like @samp{process 368}, with no
1982 @c FIXME!! (1) Does the [New...] message appear even for the very first
1983 @c thread of a program, or does it only appear for the
1984 @c second---i.e., when it becomes obvious we have a multithread
1986 @c (2) *Is* there necessarily a first thread always? Or do some
1987 @c multithread systems permit starting a program with multiple
1988 @c threads ab initio?
1990 @cindex thread number
1991 @cindex thread identifier (GDB)
1992 For debugging purposes, @value{GDBN} associates its own thread
1993 number---always a single integer---with each thread in your program.
1996 @kindex info threads
1998 Display a summary of all threads currently in your
1999 program. @value{GDBN} displays for each thread (in this order):
2002 @item the thread number assigned by @value{GDBN}
2004 @item the target system's thread identifier (@var{systag})
2006 @item the current stack frame summary for that thread
2010 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2011 indicates the current thread.
2015 @c end table here to get a little more width for example
2018 (@value{GDBP}) info threads
2019 3 process 35 thread 27 0x34e5 in sigpause ()
2020 2 process 35 thread 23 0x34e5 in sigpause ()
2021 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2027 @cindex thread number
2028 @cindex thread identifier (GDB)
2029 For debugging purposes, @value{GDBN} associates its own thread
2030 number---a small integer assigned in thread-creation order---with each
2031 thread in your program.
2033 @cindex @code{New} @var{systag} message, on HP-UX
2034 @cindex thread identifier (system), on HP-UX
2035 @c FIXME-implementors!! It would be more helpful if the [New...] message
2036 @c included GDB's numeric thread handle, so you could just go to that
2037 @c thread without first checking `info threads'.
2038 Whenever @value{GDBN} detects a new thread in your program, it displays
2039 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2040 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2041 whose form varies depending on the particular system. For example, on
2045 [New thread 2 (system thread 26594)]
2049 when @value{GDBN} notices a new thread.
2052 @kindex info threads
2054 Display a summary of all threads currently in your
2055 program. @value{GDBN} displays for each thread (in this order):
2058 @item the thread number assigned by @value{GDBN}
2060 @item the target system's thread identifier (@var{systag})
2062 @item the current stack frame summary for that thread
2066 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2067 indicates the current thread.
2071 @c end table here to get a little more width for example
2074 (@value{GDBP}) info threads
2075 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2077 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2078 from /usr/lib/libc.2
2079 1 system thread 27905 0x7b003498 in _brk () \@*
2080 from /usr/lib/libc.2
2084 @kindex thread @var{threadno}
2085 @item thread @var{threadno}
2086 Make thread number @var{threadno} the current thread. The command
2087 argument @var{threadno} is the internal @value{GDBN} thread number, as
2088 shown in the first field of the @samp{info threads} display.
2089 @value{GDBN} responds by displaying the system identifier of the thread
2090 you selected, and its current stack frame summary:
2093 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2094 (@value{GDBP}) thread 2
2095 [Switching to process 35 thread 23]
2096 0x34e5 in sigpause ()
2100 As with the @samp{[New @dots{}]} message, the form of the text after
2101 @samp{Switching to} depends on your system's conventions for identifying
2104 @kindex thread apply
2105 @item thread apply [@var{threadno}] [@var{all}] @var{args}
2106 The @code{thread apply} command allows you to apply a command to one or
2107 more threads. Specify the numbers of the threads that you want affected
2108 with the command argument @var{threadno}. @var{threadno} is the internal
2109 @value{GDBN} thread number, as shown in the first field of the @samp{info
2110 threads} display. To apply a command to all threads, use
2111 @code{thread apply all} @var{args}.
2114 @cindex automatic thread selection
2115 @cindex switching threads automatically
2116 @cindex threads, automatic switching
2117 Whenever @value{GDBN} stops your program, due to a breakpoint or a
2118 signal, it automatically selects the thread where that breakpoint or
2119 signal happened. @value{GDBN} alerts you to the context switch with a
2120 message of the form @samp{[Switching to @var{systag}]} to identify the
2123 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
2124 more information about how @value{GDBN} behaves when you stop and start
2125 programs with multiple threads.
2127 @xref{Set Watchpoints,,Setting watchpoints}, for information about
2128 watchpoints in programs with multiple threads.
2131 @section Debugging programs with multiple processes
2133 @cindex fork, debugging programs which call
2134 @cindex multiple processes
2135 @cindex processes, multiple
2136 On most systems, @value{GDBN} has no special support for debugging
2137 programs which create additional processes using the @code{fork}
2138 function. When a program forks, @value{GDBN} will continue to debug the
2139 parent process and the child process will run unimpeded. If you have
2140 set a breakpoint in any code which the child then executes, the child
2141 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2142 will cause it to terminate.
2144 However, if you want to debug the child process there is a workaround
2145 which isn't too painful. Put a call to @code{sleep} in the code which
2146 the child process executes after the fork. It may be useful to sleep
2147 only if a certain environment variable is set, or a certain file exists,
2148 so that the delay need not occur when you don't want to run @value{GDBN}
2149 on the child. While the child is sleeping, use the @code{ps} program to
2150 get its process ID. Then tell @value{GDBN} (a new invocation of
2151 @value{GDBN} if you are also debugging the parent process) to attach to
2152 the child process (@pxref{Attach}). From that point on you can debug
2153 the child process just like any other process which you attached to.
2155 On HP-UX (11.x and later only?), @value{GDBN} provides support for
2156 debugging programs that create additional processes using the
2157 @code{fork} or @code{vfork} function.
2159 By default, when a program forks, @value{GDBN} will continue to debug
2160 the parent process and the child process will run unimpeded.
2162 If you want to follow the child process instead of the parent process,
2163 use the command @w{@code{set follow-fork-mode}}.
2166 @kindex set follow-fork-mode
2167 @item set follow-fork-mode @var{mode}
2168 Set the debugger response to a program call of @code{fork} or
2169 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2170 process. The @var{mode} can be:
2174 The original process is debugged after a fork. The child process runs
2175 unimpeded. This is the default.
2178 The new process is debugged after a fork. The parent process runs
2182 The debugger will ask for one of the above choices.
2185 @item show follow-fork-mode
2186 Display the current debugger response to a @code{fork} or @code{vfork} call.
2189 If you ask to debug a child process and a @code{vfork} is followed by an
2190 @code{exec}, @value{GDBN} executes the new target up to the first
2191 breakpoint in the new target. If you have a breakpoint set on
2192 @code{main} in your original program, the breakpoint will also be set on
2193 the child process's @code{main}.
2195 When a child process is spawned by @code{vfork}, you cannot debug the
2196 child or parent until an @code{exec} call completes.
2198 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2199 call executes, the new target restarts. To restart the parent process,
2200 use the @code{file} command with the parent executable name as its
2203 You can use the @code{catch} command to make @value{GDBN} stop whenever
2204 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2205 Catchpoints, ,Setting catchpoints}.
2208 @chapter Stopping and Continuing
2210 The principal purposes of using a debugger are so that you can stop your
2211 program before it terminates; or so that, if your program runs into
2212 trouble, you can investigate and find out why.
2214 Inside @value{GDBN}, your program may stop for any of several reasons,
2215 such as a signal, a breakpoint, or reaching a new line after a
2216 @value{GDBN} command such as @code{step}. You may then examine and
2217 change variables, set new breakpoints or remove old ones, and then
2218 continue execution. Usually, the messages shown by @value{GDBN} provide
2219 ample explanation of the status of your program---but you can also
2220 explicitly request this information at any time.
2223 @kindex info program
2225 Display information about the status of your program: whether it is
2226 running or not, what process it is, and why it stopped.
2230 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2231 * Continuing and Stepping:: Resuming execution
2233 * Thread Stops:: Stopping and starting multi-thread programs
2237 @section Breakpoints, watchpoints, and catchpoints
2240 A @dfn{breakpoint} makes your program stop whenever a certain point in
2241 the program is reached. For each breakpoint, you can add conditions to
2242 control in finer detail whether your program stops. You can set
2243 breakpoints with the @code{break} command and its variants (@pxref{Set
2244 Breaks, ,Setting breakpoints}), to specify the place where your program
2245 should stop by line number, function name or exact address in the
2248 In HP-UX, SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can set
2249 breakpoints in shared libraries before the executable is run. There is
2250 a minor limitation on HP-UX systems: you must wait until the executable
2251 is run in order to set breakpoints in shared library routines that are
2252 not called directly by the program (for example, routines that are
2253 arguments in a @code{pthread_create} call).
2256 @cindex memory tracing
2257 @cindex breakpoint on memory address
2258 @cindex breakpoint on variable modification
2259 A @dfn{watchpoint} is a special breakpoint that stops your program
2260 when the value of an expression changes. You must use a different
2261 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2262 watchpoints}), but aside from that, you can manage a watchpoint like
2263 any other breakpoint: you enable, disable, and delete both breakpoints
2264 and watchpoints using the same commands.
2266 You can arrange to have values from your program displayed automatically
2267 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2271 @cindex breakpoint on events
2272 A @dfn{catchpoint} is another special breakpoint that stops your program
2273 when a certain kind of event occurs, such as the throwing of a C++
2274 exception or the loading of a library. As with watchpoints, you use a
2275 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2276 catchpoints}), but aside from that, you can manage a catchpoint like any
2277 other breakpoint. (To stop when your program receives a signal, use the
2278 @code{handle} command; see @ref{Signals, ,Signals}.)
2280 @cindex breakpoint numbers
2281 @cindex numbers for breakpoints
2282 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2283 catchpoint when you create it; these numbers are successive integers
2284 starting with one. In many of the commands for controlling various
2285 features of breakpoints you use the breakpoint number to say which
2286 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2287 @dfn{disabled}; if disabled, it has no effect on your program until you
2290 @cindex breakpoint ranges
2291 @cindex ranges of breakpoints
2292 Some @value{GDBN} commands accept a range of breakpoints on which to
2293 operate. A breakpoint range is either a single breakpoint number, like
2294 @samp{5}, or two such numbers, in increasing order, separated by a
2295 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
2296 all breakpoint in that range are operated on.
2299 * Set Breaks:: Setting breakpoints
2300 * Set Watchpoints:: Setting watchpoints
2301 * Set Catchpoints:: Setting catchpoints
2302 * Delete Breaks:: Deleting breakpoints
2303 * Disabling:: Disabling breakpoints
2304 * Conditions:: Break conditions
2305 * Break Commands:: Breakpoint command lists
2306 * Breakpoint Menus:: Breakpoint menus
2307 * Error in Breakpoints:: ``Cannot insert breakpoints''
2311 @subsection Setting breakpoints
2313 @c FIXME LMB what does GDB do if no code on line of breakpt?
2314 @c consider in particular declaration with/without initialization.
2316 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2319 @kindex b @r{(@code{break})}
2320 @vindex $bpnum@r{, convenience variable}
2321 @cindex latest breakpoint
2322 Breakpoints are set with the @code{break} command (abbreviated
2323 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2324 number of the breakpoints you've set most recently; see @ref{Convenience
2325 Vars,, Convenience variables}, for a discussion of what you can do with
2326 convenience variables.
2328 You have several ways to say where the breakpoint should go.
2331 @item break @var{function}
2332 Set a breakpoint at entry to function @var{function}.
2333 When using source languages that permit overloading of symbols, such as
2334 C++, @var{function} may refer to more than one possible place to break.
2335 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2337 @item break +@var{offset}
2338 @itemx break -@var{offset}
2339 Set a breakpoint some number of lines forward or back from the position
2340 at which execution stopped in the currently selected @dfn{stack frame}.
2341 (@xref{Frames, ,Frames}, for a description of stack frames.)
2343 @item break @var{linenum}
2344 Set a breakpoint at line @var{linenum} in the current source file.
2345 The current source file is the last file whose source text was printed.
2346 The breakpoint will stop your program just before it executes any of the
2349 @item break @var{filename}:@var{linenum}
2350 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2352 @item break @var{filename}:@var{function}
2353 Set a breakpoint at entry to function @var{function} found in file
2354 @var{filename}. Specifying a file name as well as a function name is
2355 superfluous except when multiple files contain similarly named
2358 @item break *@var{address}
2359 Set a breakpoint at address @var{address}. You can use this to set
2360 breakpoints in parts of your program which do not have debugging
2361 information or source files.
2364 When called without any arguments, @code{break} sets a breakpoint at
2365 the next instruction to be executed in the selected stack frame
2366 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2367 innermost, this makes your program stop as soon as control
2368 returns to that frame. This is similar to the effect of a
2369 @code{finish} command in the frame inside the selected frame---except
2370 that @code{finish} does not leave an active breakpoint. If you use
2371 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2372 the next time it reaches the current location; this may be useful
2375 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2376 least one instruction has been executed. If it did not do this, you
2377 would be unable to proceed past a breakpoint without first disabling the
2378 breakpoint. This rule applies whether or not the breakpoint already
2379 existed when your program stopped.
2381 @item break @dots{} if @var{cond}
2382 Set a breakpoint with condition @var{cond}; evaluate the expression
2383 @var{cond} each time the breakpoint is reached, and stop only if the
2384 value is nonzero---that is, if @var{cond} evaluates as true.
2385 @samp{@dots{}} stands for one of the possible arguments described
2386 above (or no argument) specifying where to break. @xref{Conditions,
2387 ,Break conditions}, for more information on breakpoint conditions.
2390 @item tbreak @var{args}
2391 Set a breakpoint enabled only for one stop. @var{args} are the
2392 same as for the @code{break} command, and the breakpoint is set in the same
2393 way, but the breakpoint is automatically deleted after the first time your
2394 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2397 @item hbreak @var{args}
2398 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2399 @code{break} command and the breakpoint is set in the same way, but the
2400 breakpoint requires hardware support and some target hardware may not
2401 have this support. The main purpose of this is EPROM/ROM code
2402 debugging, so you can set a breakpoint at an instruction without
2403 changing the instruction. This can be used with the new trap-generation
2404 provided by SPARClite DSU and some x86-based targets. These targets
2405 will generate traps when a program accesses some data or instruction
2406 address that is assigned to the debug registers. However the hardware
2407 breakpoint registers can take a limited number of breakpoints. For
2408 example, on the DSU, only two data breakpoints can be set at a time, and
2409 @value{GDBN} will reject this command if more than two are used. Delete
2410 or disable unused hardware breakpoints before setting new ones
2411 (@pxref{Disabling, ,Disabling}). @xref{Conditions, ,Break conditions}.
2414 @item thbreak @var{args}
2415 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2416 are the same as for the @code{hbreak} command and the breakpoint is set in
2417 the same way. However, like the @code{tbreak} command,
2418 the breakpoint is automatically deleted after the
2419 first time your program stops there. Also, like the @code{hbreak}
2420 command, the breakpoint requires hardware support and some target hardware
2421 may not have this support. @xref{Disabling, ,Disabling breakpoints}.
2422 See also @ref{Conditions, ,Break conditions}.
2425 @cindex regular expression
2426 @item rbreak @var{regex}
2427 Set breakpoints on all functions matching the regular expression
2428 @var{regex}. This command sets an unconditional breakpoint on all
2429 matches, printing a list of all breakpoints it set. Once these
2430 breakpoints are set, they are treated just like the breakpoints set with
2431 the @code{break} command. You can delete them, disable them, or make
2432 them conditional the same way as any other breakpoint.
2434 The syntax of the regular expression is the standard one used with tools
2435 like @file{grep}. Note that this is different from the syntax used by
2436 shells, so for instance @code{foo*} matches all functions that include
2437 an @code{fo} followed by zero or more @code{o}s. There is an implicit
2438 @code{.*} leading and trailing the regular expression you supply, so to
2439 match only functions that begin with @code{foo}, use @code{^foo}.
2441 When debugging C++ programs, @code{rbreak} is useful for setting
2442 breakpoints on overloaded functions that are not members of any special
2445 @kindex info breakpoints
2446 @cindex @code{$_} and @code{info breakpoints}
2447 @item info breakpoints @r{[}@var{n}@r{]}
2448 @itemx info break @r{[}@var{n}@r{]}
2449 @itemx info watchpoints @r{[}@var{n}@r{]}
2450 Print a table of all breakpoints, watchpoints, and catchpoints set and
2451 not deleted, with the following columns for each breakpoint:
2454 @item Breakpoint Numbers
2456 Breakpoint, watchpoint, or catchpoint.
2458 Whether the breakpoint is marked to be disabled or deleted when hit.
2459 @item Enabled or Disabled
2460 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2461 that are not enabled.
2463 Where the breakpoint is in your program, as a memory address.
2465 Where the breakpoint is in the source for your program, as a file and
2470 If a breakpoint is conditional, @code{info break} shows the condition on
2471 the line following the affected breakpoint; breakpoint commands, if any,
2472 are listed after that.
2475 @code{info break} with a breakpoint
2476 number @var{n} as argument lists only that breakpoint. The
2477 convenience variable @code{$_} and the default examining-address for
2478 the @code{x} command are set to the address of the last breakpoint
2479 listed (@pxref{Memory, ,Examining memory}).
2482 @code{info break} displays a count of the number of times the breakpoint
2483 has been hit. This is especially useful in conjunction with the
2484 @code{ignore} command. You can ignore a large number of breakpoint
2485 hits, look at the breakpoint info to see how many times the breakpoint
2486 was hit, and then run again, ignoring one less than that number. This
2487 will get you quickly to the last hit of that breakpoint.
2490 @value{GDBN} allows you to set any number of breakpoints at the same place in
2491 your program. There is nothing silly or meaningless about this. When
2492 the breakpoints are conditional, this is even useful
2493 (@pxref{Conditions, ,Break conditions}).
2495 @cindex negative breakpoint numbers
2496 @cindex internal @value{GDBN} breakpoints
2497 @value{GDBN} itself sometimes sets breakpoints in your program for special
2498 purposes, such as proper handling of @code{longjmp} (in C programs).
2499 These internal breakpoints are assigned negative numbers, starting with
2500 @code{-1}; @samp{info breakpoints} does not display them.
2502 You can see these breakpoints with the @value{GDBN} maintenance command
2503 @samp{maint info breakpoints}.
2506 @kindex maint info breakpoints
2507 @item maint info breakpoints
2508 Using the same format as @samp{info breakpoints}, display both the
2509 breakpoints you've set explicitly, and those @value{GDBN} is using for
2510 internal purposes. Internal breakpoints are shown with negative
2511 breakpoint numbers. The type column identifies what kind of breakpoint
2516 Normal, explicitly set breakpoint.
2519 Normal, explicitly set watchpoint.
2522 Internal breakpoint, used to handle correctly stepping through
2523 @code{longjmp} calls.
2525 @item longjmp resume
2526 Internal breakpoint at the target of a @code{longjmp}.
2529 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2532 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2535 Shared library events.
2542 @node Set Watchpoints
2543 @subsection Setting watchpoints
2545 @cindex setting watchpoints
2546 @cindex software watchpoints
2547 @cindex hardware watchpoints
2548 You can use a watchpoint to stop execution whenever the value of an
2549 expression changes, without having to predict a particular place where
2552 Depending on your system, watchpoints may be implemented in software or
2553 hardware. @value{GDBN} does software watchpointing by single-stepping your
2554 program and testing the variable's value each time, which is hundreds of
2555 times slower than normal execution. (But this may still be worth it, to
2556 catch errors where you have no clue what part of your program is the
2559 On some systems, such as HP-UX, Linux and some other x86-based targets,
2560 @value{GDBN} includes support for
2561 hardware watchpoints, which do not slow down the running of your
2566 @item watch @var{expr}
2567 Set a watchpoint for an expression. @value{GDBN} will break when @var{expr}
2568 is written into by the program and its value changes.
2571 @item rwatch @var{expr}
2572 Set a watchpoint that will break when watch @var{expr} is read by the program.
2575 @item awatch @var{expr}
2576 Set a watchpoint that will break when @var{expr} is either read or written into
2579 @kindex info watchpoints
2580 @item info watchpoints
2581 This command prints a list of watchpoints, breakpoints, and catchpoints;
2582 it is the same as @code{info break}.
2585 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
2586 watchpoints execute very quickly, and the debugger reports a change in
2587 value at the exact instruction where the change occurs. If @value{GDBN}
2588 cannot set a hardware watchpoint, it sets a software watchpoint, which
2589 executes more slowly and reports the change in value at the next
2590 statement, not the instruction, after the change occurs.
2592 When you issue the @code{watch} command, @value{GDBN} reports
2595 Hardware watchpoint @var{num}: @var{expr}
2599 if it was able to set a hardware watchpoint.
2601 Currently, the @code{awatch} and @code{rwatch} commands can only set
2602 hardware watchpoints, because accesses to data that don't change the
2603 value of the watched expression cannot be detected without examining
2604 every instruction as it is being executed, and @value{GDBN} does not do
2605 that currently. If @value{GDBN} finds that it is unable to set a
2606 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
2607 will print a message like this:
2610 Expression cannot be implemented with read/access watchpoint.
2613 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
2614 data type of the watched expression is wider than what a hardware
2615 watchpoint on the target machine can handle. For example, some systems
2616 can only watch regions that are up to 4 bytes wide; on such systems you
2617 cannot set hardware watchpoints for an expression that yields a
2618 double-precision floating-point number (which is typically 8 bytes
2619 wide). As a work-around, it might be possible to break the large region
2620 into a series of smaller ones and watch them with separate watchpoints.
2622 If you set too many hardware watchpoints, @value{GDBN} might be unable
2623 to insert all of them when you resume the execution of your program.
2624 Since the precise number of active watchpoints is unknown until such
2625 time as the program is about to be resumed, @value{GDBN} might not be
2626 able to warn you about this when you set the watchpoints, and the
2627 warning will be printed only when the program is resumed:
2630 Hardware watchpoint @var{num}: Could not insert watchpoint
2634 If this happens, delete or disable some of the watchpoints.
2636 The SPARClite DSU will generate traps when a program accesses some data
2637 or instruction address that is assigned to the debug registers. For the
2638 data addresses, DSU facilitates the @code{watch} command. However the
2639 hardware breakpoint registers can only take two data watchpoints, and
2640 both watchpoints must be the same kind. For example, you can set two
2641 watchpoints with @code{watch} commands, two with @code{rwatch} commands,
2642 @strong{or} two with @code{awatch} commands, but you cannot set one
2643 watchpoint with one command and the other with a different command.
2644 @value{GDBN} will reject the command if you try to mix watchpoints.
2645 Delete or disable unused watchpoint commands before setting new ones.
2647 If you call a function interactively using @code{print} or @code{call},
2648 any watchpoints you have set will be inactive until @value{GDBN} reaches another
2649 kind of breakpoint or the call completes.
2651 @value{GDBN} automatically deletes watchpoints that watch local
2652 (automatic) variables, or expressions that involve such variables, when
2653 they go out of scope, that is, when the execution leaves the block in
2654 which these variables were defined. In particular, when the program
2655 being debugged terminates, @emph{all} local variables go out of scope,
2656 and so only watchpoints that watch global variables remain set. If you
2657 rerun the program, you will need to set all such watchpoints again. One
2658 way of doing that would be to set a code breakpoint at the entry to the
2659 @code{main} function and when it breaks, set all the watchpoints.
2662 @cindex watchpoints and threads
2663 @cindex threads and watchpoints
2664 @emph{Warning:} In multi-thread programs, watchpoints have only limited
2665 usefulness. With the current watchpoint implementation, @value{GDBN}
2666 can only watch the value of an expression @emph{in a single thread}. If
2667 you are confident that the expression can only change due to the current
2668 thread's activity (and if you are also confident that no other thread
2669 can become current), then you can use watchpoints as usual. However,
2670 @value{GDBN} may not notice when a non-current thread's activity changes
2673 @c FIXME: this is almost identical to the previous paragraph.
2674 @emph{HP-UX Warning:} In multi-thread programs, software watchpoints
2675 have only limited usefulness. If @value{GDBN} creates a software
2676 watchpoint, it can only watch the value of an expression @emph{in a
2677 single thread}. If you are confident that the expression can only
2678 change due to the current thread's activity (and if you are also
2679 confident that no other thread can become current), then you can use
2680 software watchpoints as usual. However, @value{GDBN} may not notice
2681 when a non-current thread's activity changes the expression. (Hardware
2682 watchpoints, in contrast, watch an expression in all threads.)
2685 @node Set Catchpoints
2686 @subsection Setting catchpoints
2687 @cindex catchpoints, setting
2688 @cindex exception handlers
2689 @cindex event handling
2691 You can use @dfn{catchpoints} to cause the debugger to stop for certain
2692 kinds of program events, such as C++ exceptions or the loading of a
2693 shared library. Use the @code{catch} command to set a catchpoint.
2697 @item catch @var{event}
2698 Stop when @var{event} occurs. @var{event} can be any of the following:
2702 The throwing of a C++ exception.
2706 The catching of a C++ exception.
2710 A call to @code{exec}. This is currently only available for HP-UX.
2714 A call to @code{fork}. This is currently only available for HP-UX.
2718 A call to @code{vfork}. This is currently only available for HP-UX.
2721 @itemx load @var{libname}
2723 The dynamic loading of any shared library, or the loading of the library
2724 @var{libname}. This is currently only available for HP-UX.
2727 @itemx unload @var{libname}
2728 @kindex catch unload
2729 The unloading of any dynamically loaded shared library, or the unloading
2730 of the library @var{libname}. This is currently only available for HP-UX.
2733 @item tcatch @var{event}
2734 Set a catchpoint that is enabled only for one stop. The catchpoint is
2735 automatically deleted after the first time the event is caught.
2739 Use the @code{info break} command to list the current catchpoints.
2741 There are currently some limitations to C++ exception handling
2742 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
2746 If you call a function interactively, @value{GDBN} normally returns
2747 control to you when the function has finished executing. If the call
2748 raises an exception, however, the call may bypass the mechanism that
2749 returns control to you and cause your program either to abort or to
2750 simply continue running until it hits a breakpoint, catches a signal
2751 that @value{GDBN} is listening for, or exits. This is the case even if
2752 you set a catchpoint for the exception; catchpoints on exceptions are
2753 disabled within interactive calls.
2756 You cannot raise an exception interactively.
2759 You cannot install an exception handler interactively.
2762 @cindex raise exceptions
2763 Sometimes @code{catch} is not the best way to debug exception handling:
2764 if you need to know exactly where an exception is raised, it is better to
2765 stop @emph{before} the exception handler is called, since that way you
2766 can see the stack before any unwinding takes place. If you set a
2767 breakpoint in an exception handler instead, it may not be easy to find
2768 out where the exception was raised.
2770 To stop just before an exception handler is called, you need some
2771 knowledge of the implementation. In the case of @sc{gnu} C++, exceptions are
2772 raised by calling a library function named @code{__raise_exception}
2773 which has the following ANSI C interface:
2776 /* @var{addr} is where the exception identifier is stored.
2777 @var{id} is the exception identifier. */
2778 void __raise_exception (void **addr, void *id);
2782 To make the debugger catch all exceptions before any stack
2783 unwinding takes place, set a breakpoint on @code{__raise_exception}
2784 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2786 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2787 that depends on the value of @var{id}, you can stop your program when
2788 a specific exception is raised. You can use multiple conditional
2789 breakpoints to stop your program when any of a number of exceptions are
2794 @subsection Deleting breakpoints
2796 @cindex clearing breakpoints, watchpoints, catchpoints
2797 @cindex deleting breakpoints, watchpoints, catchpoints
2798 It is often necessary to eliminate a breakpoint, watchpoint, or
2799 catchpoint once it has done its job and you no longer want your program
2800 to stop there. This is called @dfn{deleting} the breakpoint. A
2801 breakpoint that has been deleted no longer exists; it is forgotten.
2803 With the @code{clear} command you can delete breakpoints according to
2804 where they are in your program. With the @code{delete} command you can
2805 delete individual breakpoints, watchpoints, or catchpoints by specifying
2806 their breakpoint numbers.
2808 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2809 automatically ignores breakpoints on the first instruction to be executed
2810 when you continue execution without changing the execution address.
2815 Delete any breakpoints at the next instruction to be executed in the
2816 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2817 the innermost frame is selected, this is a good way to delete a
2818 breakpoint where your program just stopped.
2820 @item clear @var{function}
2821 @itemx clear @var{filename}:@var{function}
2822 Delete any breakpoints set at entry to the function @var{function}.
2824 @item clear @var{linenum}
2825 @itemx clear @var{filename}:@var{linenum}
2826 Delete any breakpoints set at or within the code of the specified line.
2828 @cindex delete breakpoints
2830 @kindex d @r{(@code{delete})}
2831 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2832 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
2833 ranges specified as arguments. If no argument is specified, delete all
2834 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
2835 confirm off}). You can abbreviate this command as @code{d}.
2839 @subsection Disabling breakpoints
2841 @kindex disable breakpoints
2842 @kindex enable breakpoints
2843 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
2844 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
2845 it had been deleted, but remembers the information on the breakpoint so
2846 that you can @dfn{enable} it again later.
2848 You disable and enable breakpoints, watchpoints, and catchpoints with
2849 the @code{enable} and @code{disable} commands, optionally specifying one
2850 or more breakpoint numbers as arguments. Use @code{info break} or
2851 @code{info watch} to print a list of breakpoints, watchpoints, and
2852 catchpoints if you do not know which numbers to use.
2854 A breakpoint, watchpoint, or catchpoint can have any of four different
2855 states of enablement:
2859 Enabled. The breakpoint stops your program. A breakpoint set
2860 with the @code{break} command starts out in this state.
2862 Disabled. The breakpoint has no effect on your program.
2864 Enabled once. The breakpoint stops your program, but then becomes
2867 Enabled for deletion. The breakpoint stops your program, but
2868 immediately after it does so it is deleted permanently. A breakpoint
2869 set with the @code{tbreak} command starts out in this state.
2872 You can use the following commands to enable or disable breakpoints,
2873 watchpoints, and catchpoints:
2876 @kindex disable breakpoints
2878 @kindex dis @r{(@code{disable})}
2879 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2880 Disable the specified breakpoints---or all breakpoints, if none are
2881 listed. A disabled breakpoint has no effect but is not forgotten. All
2882 options such as ignore-counts, conditions and commands are remembered in
2883 case the breakpoint is enabled again later. You may abbreviate
2884 @code{disable} as @code{dis}.
2886 @kindex enable breakpoints
2888 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2889 Enable the specified breakpoints (or all defined breakpoints). They
2890 become effective once again in stopping your program.
2892 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
2893 Enable the specified breakpoints temporarily. @value{GDBN} disables any
2894 of these breakpoints immediately after stopping your program.
2896 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
2897 Enable the specified breakpoints to work once, then die. @value{GDBN}
2898 deletes any of these breakpoints as soon as your program stops there.
2901 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
2902 @c confusing: tbreak is also initially enabled.
2903 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2904 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2905 subsequently, they become disabled or enabled only when you use one of
2906 the commands above. (The command @code{until} can set and delete a
2907 breakpoint of its own, but it does not change the state of your other
2908 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2912 @subsection Break conditions
2913 @cindex conditional breakpoints
2914 @cindex breakpoint conditions
2916 @c FIXME what is scope of break condition expr? Context where wanted?
2917 @c in particular for a watchpoint?
2918 The simplest sort of breakpoint breaks every time your program reaches a
2919 specified place. You can also specify a @dfn{condition} for a
2920 breakpoint. A condition is just a Boolean expression in your
2921 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2922 a condition evaluates the expression each time your program reaches it,
2923 and your program stops only if the condition is @emph{true}.
2925 This is the converse of using assertions for program validation; in that
2926 situation, you want to stop when the assertion is violated---that is,
2927 when the condition is false. In C, if you want to test an assertion expressed
2928 by the condition @var{assert}, you should set the condition
2929 @samp{! @var{assert}} on the appropriate breakpoint.
2931 Conditions are also accepted for watchpoints; you may not need them,
2932 since a watchpoint is inspecting the value of an expression anyhow---but
2933 it might be simpler, say, to just set a watchpoint on a variable name,
2934 and specify a condition that tests whether the new value is an interesting
2937 Break conditions can have side effects, and may even call functions in
2938 your program. This can be useful, for example, to activate functions
2939 that log program progress, or to use your own print functions to
2940 format special data structures. The effects are completely predictable
2941 unless there is another enabled breakpoint at the same address. (In
2942 that case, @value{GDBN} might see the other breakpoint first and stop your
2943 program without checking the condition of this one.) Note that
2944 breakpoint commands are usually more convenient and flexible than break
2946 purpose of performing side effects when a breakpoint is reached
2947 (@pxref{Break Commands, ,Breakpoint command lists}).
2949 Break conditions can be specified when a breakpoint is set, by using
2950 @samp{if} in the arguments to the @code{break} command. @xref{Set
2951 Breaks, ,Setting breakpoints}. They can also be changed at any time
2952 with the @code{condition} command.
2954 You can also use the @code{if} keyword with the @code{watch} command.
2955 The @code{catch} command does not recognize the @code{if} keyword;
2956 @code{condition} is the only way to impose a further condition on a
2961 @item condition @var{bnum} @var{expression}
2962 Specify @var{expression} as the break condition for breakpoint,
2963 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
2964 breakpoint @var{bnum} stops your program only if the value of
2965 @var{expression} is true (nonzero, in C). When you use
2966 @code{condition}, @value{GDBN} checks @var{expression} immediately for
2967 syntactic correctness, and to determine whether symbols in it have
2968 referents in the context of your breakpoint. If @var{expression} uses
2969 symbols not referenced in the context of the breakpoint, @value{GDBN}
2970 prints an error message:
2973 No symbol "foo" in current context.
2978 not actually evaluate @var{expression} at the time the @code{condition}
2979 command (or a command that sets a breakpoint with a condition, like
2980 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
2982 @item condition @var{bnum}
2983 Remove the condition from breakpoint number @var{bnum}. It becomes
2984 an ordinary unconditional breakpoint.
2987 @cindex ignore count (of breakpoint)
2988 A special case of a breakpoint condition is to stop only when the
2989 breakpoint has been reached a certain number of times. This is so
2990 useful that there is a special way to do it, using the @dfn{ignore
2991 count} of the breakpoint. Every breakpoint has an ignore count, which
2992 is an integer. Most of the time, the ignore count is zero, and
2993 therefore has no effect. But if your program reaches a breakpoint whose
2994 ignore count is positive, then instead of stopping, it just decrements
2995 the ignore count by one and continues. As a result, if the ignore count
2996 value is @var{n}, the breakpoint does not stop the next @var{n} times
2997 your program reaches it.
3001 @item ignore @var{bnum} @var{count}
3002 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3003 The next @var{count} times the breakpoint is reached, your program's
3004 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3007 To make the breakpoint stop the next time it is reached, specify
3010 When you use @code{continue} to resume execution of your program from a
3011 breakpoint, you can specify an ignore count directly as an argument to
3012 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3013 Stepping,,Continuing and stepping}.
3015 If a breakpoint has a positive ignore count and a condition, the
3016 condition is not checked. Once the ignore count reaches zero,
3017 @value{GDBN} resumes checking the condition.
3019 You could achieve the effect of the ignore count with a condition such
3020 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3021 is decremented each time. @xref{Convenience Vars, ,Convenience
3025 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3028 @node Break Commands
3029 @subsection Breakpoint command lists
3031 @cindex breakpoint commands
3032 You can give any breakpoint (or watchpoint or catchpoint) a series of
3033 commands to execute when your program stops due to that breakpoint. For
3034 example, you might want to print the values of certain expressions, or
3035 enable other breakpoints.
3040 @item commands @r{[}@var{bnum}@r{]}
3041 @itemx @dots{} @var{command-list} @dots{}
3043 Specify a list of commands for breakpoint number @var{bnum}. The commands
3044 themselves appear on the following lines. Type a line containing just
3045 @code{end} to terminate the commands.
3047 To remove all commands from a breakpoint, type @code{commands} and
3048 follow it immediately with @code{end}; that is, give no commands.
3050 With no @var{bnum} argument, @code{commands} refers to the last
3051 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
3052 recently encountered).
3055 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
3056 disabled within a @var{command-list}.
3058 You can use breakpoint commands to start your program up again. Simply
3059 use the @code{continue} command, or @code{step}, or any other command
3060 that resumes execution.
3062 Any other commands in the command list, after a command that resumes
3063 execution, are ignored. This is because any time you resume execution
3064 (even with a simple @code{next} or @code{step}), you may encounter
3065 another breakpoint---which could have its own command list, leading to
3066 ambiguities about which list to execute.
3069 If the first command you specify in a command list is @code{silent}, the
3070 usual message about stopping at a breakpoint is not printed. This may
3071 be desirable for breakpoints that are to print a specific message and
3072 then continue. If none of the remaining commands print anything, you
3073 see no sign that the breakpoint was reached. @code{silent} is
3074 meaningful only at the beginning of a breakpoint command list.
3076 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3077 print precisely controlled output, and are often useful in silent
3078 breakpoints. @xref{Output, ,Commands for controlled output}.
3080 For example, here is how you could use breakpoint commands to print the
3081 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
3087 printf "x is %d\n",x
3092 One application for breakpoint commands is to compensate for one bug so
3093 you can test for another. Put a breakpoint just after the erroneous line
3094 of code, give it a condition to detect the case in which something
3095 erroneous has been done, and give it commands to assign correct values
3096 to any variables that need them. End with the @code{continue} command
3097 so that your program does not stop, and start with the @code{silent}
3098 command so that no output is produced. Here is an example:
3109 @node Breakpoint Menus
3110 @subsection Breakpoint menus
3112 @cindex symbol overloading
3114 Some programming languages (notably C++) permit a single function name
3115 to be defined several times, for application in different contexts.
3116 This is called @dfn{overloading}. When a function name is overloaded,
3117 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
3118 a breakpoint. If you realize this is a problem, you can use
3119 something like @samp{break @var{function}(@var{types})} to specify which
3120 particular version of the function you want. Otherwise, @value{GDBN} offers
3121 you a menu of numbered choices for different possible breakpoints, and
3122 waits for your selection with the prompt @samp{>}. The first two
3123 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
3124 sets a breakpoint at each definition of @var{function}, and typing
3125 @kbd{0} aborts the @code{break} command without setting any new
3128 For example, the following session excerpt shows an attempt to set a
3129 breakpoint at the overloaded symbol @code{String::after}.
3130 We choose three particular definitions of that function name:
3132 @c FIXME! This is likely to change to show arg type lists, at least
3135 (@value{GDBP}) b String::after
3138 [2] file:String.cc; line number:867
3139 [3] file:String.cc; line number:860
3140 [4] file:String.cc; line number:875
3141 [5] file:String.cc; line number:853
3142 [6] file:String.cc; line number:846
3143 [7] file:String.cc; line number:735
3145 Breakpoint 1 at 0xb26c: file String.cc, line 867.
3146 Breakpoint 2 at 0xb344: file String.cc, line 875.
3147 Breakpoint 3 at 0xafcc: file String.cc, line 846.
3148 Multiple breakpoints were set.
3149 Use the "delete" command to delete unwanted
3155 @c @ifclear BARETARGET
3156 @node Error in Breakpoints
3157 @subsection ``Cannot insert breakpoints''
3159 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
3161 Under some operating systems, breakpoints cannot be used in a program if
3162 any other process is running that program. In this situation,
3163 attempting to run or continue a program with a breakpoint causes
3164 @value{GDBN} to print an error message:
3167 Cannot insert breakpoints.
3168 The same program may be running in another process.
3171 When this happens, you have three ways to proceed:
3175 Remove or disable the breakpoints, then continue.
3178 Suspend @value{GDBN}, and copy the file containing your program to a new
3179 name. Resume @value{GDBN} and use the @code{exec-file} command to specify
3180 that @value{GDBN} should run your program under that name.
3181 Then start your program again.
3184 Relink your program so that the text segment is nonsharable, using the
3185 linker option @samp{-N}. The operating system limitation may not apply
3186 to nonsharable executables.
3190 A similar message can be printed if you request too many active
3191 hardware-assisted breakpoints and watchpoints:
3193 @c FIXME: the precise wording of this message may change; the relevant
3194 @c source change is not committed yet (Sep 3, 1999).
3196 Stopped; cannot insert breakpoints.
3197 You may have requested too many hardware breakpoints and watchpoints.
3201 This message is printed when you attempt to resume the program, since
3202 only then @value{GDBN} knows exactly how many hardware breakpoints and
3203 watchpoints it needs to insert.
3205 When this message is printed, you need to disable or remove some of the
3206 hardware-assisted breakpoints and watchpoints, and then continue.
3209 @node Continuing and Stepping
3210 @section Continuing and stepping
3214 @cindex resuming execution
3215 @dfn{Continuing} means resuming program execution until your program
3216 completes normally. In contrast, @dfn{stepping} means executing just
3217 one more ``step'' of your program, where ``step'' may mean either one
3218 line of source code, or one machine instruction (depending on what
3219 particular command you use). Either when continuing or when stepping,
3220 your program may stop even sooner, due to a breakpoint or a signal. (If
3221 it stops due to a signal, you may want to use @code{handle}, or use
3222 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
3226 @kindex c @r{(@code{continue})}
3227 @kindex fg @r{(resume foreground execution)}
3228 @item continue @r{[}@var{ignore-count}@r{]}
3229 @itemx c @r{[}@var{ignore-count}@r{]}
3230 @itemx fg @r{[}@var{ignore-count}@r{]}
3231 Resume program execution, at the address where your program last stopped;
3232 any breakpoints set at that address are bypassed. The optional argument
3233 @var{ignore-count} allows you to specify a further number of times to
3234 ignore a breakpoint at this location; its effect is like that of
3235 @code{ignore} (@pxref{Conditions, ,Break conditions}).
3237 The argument @var{ignore-count} is meaningful only when your program
3238 stopped due to a breakpoint. At other times, the argument to
3239 @code{continue} is ignored.
3241 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
3242 debugged program is deemed to be the foreground program) are provided
3243 purely for convenience, and have exactly the same behavior as
3247 To resume execution at a different place, you can use @code{return}
3248 (@pxref{Returning, ,Returning from a function}) to go back to the
3249 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
3250 different address}) to go to an arbitrary location in your program.
3252 A typical technique for using stepping is to set a breakpoint
3253 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and catchpoints}) at the
3254 beginning of the function or the section of your program where a problem
3255 is believed to lie, run your program until it stops at that breakpoint,
3256 and then step through the suspect area, examining the variables that are
3257 interesting, until you see the problem happen.
3261 @kindex s @r{(@code{step})}
3263 Continue running your program until control reaches a different source
3264 line, then stop it and return control to @value{GDBN}. This command is
3265 abbreviated @code{s}.
3268 @c "without debugging information" is imprecise; actually "without line
3269 @c numbers in the debugging information". (gcc -g1 has debugging info but
3270 @c not line numbers). But it seems complex to try to make that
3271 @c distinction here.
3272 @emph{Warning:} If you use the @code{step} command while control is
3273 within a function that was compiled without debugging information,
3274 execution proceeds until control reaches a function that does have
3275 debugging information. Likewise, it will not step into a function which
3276 is compiled without debugging information. To step through functions
3277 without debugging information, use the @code{stepi} command, described
3281 The @code{step} command only stops at the first instruction of a
3282 source line. This prevents the multiple stops that could otherwise occur in
3283 switch statements, for loops, etc. @code{step} continues to stop if a
3284 function that has debugging information is called within the line.
3285 In other words, @code{step} @emph{steps inside} any functions called
3288 Also, the @code{step} command only enters a function if there is line
3289 number information for the function. Otherwise it acts like the
3290 @code{next} command. This avoids problems when using @code{cc -gl}
3291 on MIPS machines. Previously, @code{step} entered subroutines if there
3292 was any debugging information about the routine.
3294 @item step @var{count}
3295 Continue running as in @code{step}, but do so @var{count} times. If a
3296 breakpoint is reached, or a signal not related to stepping occurs before
3297 @var{count} steps, stepping stops right away.
3300 @kindex n @r{(@code{next})}
3301 @item next @r{[}@var{count}@r{]}
3302 Continue to the next source line in the current (innermost) stack frame.
3303 This is similar to @code{step}, but function calls that appear within
3304 the line of code are executed without stopping. Execution stops when
3305 control reaches a different line of code at the original stack level
3306 that was executing when you gave the @code{next} command. This command
3307 is abbreviated @code{n}.
3309 An argument @var{count} is a repeat count, as for @code{step}.
3312 @c FIX ME!! Do we delete this, or is there a way it fits in with
3313 @c the following paragraph? --- Vctoria
3315 @c @code{next} within a function that lacks debugging information acts like
3316 @c @code{step}, but any function calls appearing within the code of the
3317 @c function are executed without stopping.
3319 The @code{next} command only stops at the first instruction of a
3320 source line. This prevents multiple stops that could otherwise occur in
3321 switch statements, for loops, etc.
3325 Continue running until just after function in the selected stack frame
3326 returns. Print the returned value (if any).
3328 Contrast this with the @code{return} command (@pxref{Returning,
3329 ,Returning from a function}).
3332 @kindex u @r{(@code{until})}
3335 Continue running until a source line past the current line, in the
3336 current stack frame, is reached. This command is used to avoid single
3337 stepping through a loop more than once. It is like the @code{next}
3338 command, except that when @code{until} encounters a jump, it
3339 automatically continues execution until the program counter is greater
3340 than the address of the jump.
3342 This means that when you reach the end of a loop after single stepping
3343 though it, @code{until} makes your program continue execution until it
3344 exits the loop. In contrast, a @code{next} command at the end of a loop
3345 simply steps back to the beginning of the loop, which forces you to step
3346 through the next iteration.
3348 @code{until} always stops your program if it attempts to exit the current
3351 @code{until} may produce somewhat counterintuitive results if the order
3352 of machine code does not match the order of the source lines. For
3353 example, in the following excerpt from a debugging session, the @code{f}
3354 (@code{frame}) command shows that execution is stopped at line
3355 @code{206}; yet when we use @code{until}, we get to line @code{195}:
3359 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
3361 (@value{GDBP}) until
3362 195 for ( ; argc > 0; NEXTARG) @{
3365 This happened because, for execution efficiency, the compiler had
3366 generated code for the loop closure test at the end, rather than the
3367 start, of the loop---even though the test in a C @code{for}-loop is
3368 written before the body of the loop. The @code{until} command appeared
3369 to step back to the beginning of the loop when it advanced to this
3370 expression; however, it has not really gone to an earlier
3371 statement---not in terms of the actual machine code.
3373 @code{until} with no argument works by means of single
3374 instruction stepping, and hence is slower than @code{until} with an
3377 @item until @var{location}
3378 @itemx u @var{location}
3379 Continue running your program until either the specified location is
3380 reached, or the current stack frame returns. @var{location} is any of
3381 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
3382 ,Setting breakpoints}). This form of the command uses breakpoints,
3383 and hence is quicker than @code{until} without an argument.
3386 @kindex si @r{(@code{stepi})}
3388 @itemx stepi @var{arg}
3390 Execute one machine instruction, then stop and return to the debugger.
3392 It is often useful to do @samp{display/i $pc} when stepping by machine
3393 instructions. This makes @value{GDBN} automatically display the next
3394 instruction to be executed, each time your program stops. @xref{Auto
3395 Display,, Automatic display}.
3397 An argument is a repeat count, as in @code{step}.
3401 @kindex ni @r{(@code{nexti})}
3403 @itemx nexti @var{arg}
3405 Execute one machine instruction, but if it is a function call,
3406 proceed until the function returns.
3408 An argument is a repeat count, as in @code{next}.
3415 A signal is an asynchronous event that can happen in a program. The
3416 operating system defines the possible kinds of signals, and gives each
3417 kind a name and a number. For example, in Unix @code{SIGINT} is the
3418 signal a program gets when you type an interrupt character (often @kbd{C-c});
3419 @code{SIGSEGV} is the signal a program gets from referencing a place in
3420 memory far away from all the areas in use; @code{SIGALRM} occurs when
3421 the alarm clock timer goes off (which happens only if your program has
3422 requested an alarm).
3424 @cindex fatal signals
3425 Some signals, including @code{SIGALRM}, are a normal part of the
3426 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3427 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
3428 program has not specified in advance some other way to handle the signal.
3429 @code{SIGINT} does not indicate an error in your program, but it is normally
3430 fatal so it can carry out the purpose of the interrupt: to kill the program.
3432 @value{GDBN} has the ability to detect any occurrence of a signal in your
3433 program. You can tell @value{GDBN} in advance what to do for each kind of
3436 @cindex handling signals
3437 Normally, @value{GDBN} is set up to ignore non-erroneous signals like @code{SIGALRM}
3438 (so as not to interfere with their role in the functioning of your program)
3439 but to stop your program immediately whenever an error signal happens.
3440 You can change these settings with the @code{handle} command.
3443 @kindex info signals
3446 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3447 handle each one. You can use this to see the signal numbers of all
3448 the defined types of signals.
3450 @code{info handle} is an alias for @code{info signals}.
3453 @item handle @var{signal} @var{keywords}@dots{}
3454 Change the way @value{GDBN} handles signal @var{signal}. @var{signal} can
3455 be the number of a signal or its name (with or without the @samp{SIG} at the
3456 beginning). The @var{keywords} say what change to make.
3460 The keywords allowed by the @code{handle} command can be abbreviated.
3461 Their full names are:
3465 @value{GDBN} should not stop your program when this signal happens. It may
3466 still print a message telling you that the signal has come in.
3469 @value{GDBN} should stop your program when this signal happens. This implies
3470 the @code{print} keyword as well.
3473 @value{GDBN} should print a message when this signal happens.
3476 @value{GDBN} should not mention the occurrence of the signal at all. This
3477 implies the @code{nostop} keyword as well.
3480 @value{GDBN} should allow your program to see this signal; your program
3481 can handle the signal, or else it may terminate if the signal is fatal
3485 @value{GDBN} should not allow your program to see this signal.
3489 When a signal stops your program, the signal is not visible to the
3491 continue. Your program sees the signal then, if @code{pass} is in
3492 effect for the signal in question @emph{at that time}. In other words,
3493 after @value{GDBN} reports a signal, you can use the @code{handle}
3494 command with @code{pass} or @code{nopass} to control whether your
3495 program sees that signal when you continue.
3497 You can also use the @code{signal} command to prevent your program from
3498 seeing a signal, or cause it to see a signal it normally would not see,
3499 or to give it any signal at any time. For example, if your program stopped
3500 due to some sort of memory reference error, you might store correct
3501 values into the erroneous variables and continue, hoping to see more
3502 execution; but your program would probably terminate immediately as
3503 a result of the fatal signal once it saw the signal. To prevent this,
3504 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3508 @section Stopping and starting multi-thread programs
3510 When your program has multiple threads (@pxref{Threads,, Debugging
3511 programs with multiple threads}), you can choose whether to set
3512 breakpoints on all threads, or on a particular thread.
3515 @cindex breakpoints and threads
3516 @cindex thread breakpoints
3517 @kindex break @dots{} thread @var{threadno}
3518 @item break @var{linespec} thread @var{threadno}
3519 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3520 @var{linespec} specifies source lines; there are several ways of
3521 writing them, but the effect is always to specify some source line.
3523 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3524 to specify that you only want @value{GDBN} to stop the program when a
3525 particular thread reaches this breakpoint. @var{threadno} is one of the
3526 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3527 column of the @samp{info threads} display.
3529 If you do not specify @samp{thread @var{threadno}} when you set a
3530 breakpoint, the breakpoint applies to @emph{all} threads of your
3533 You can use the @code{thread} qualifier on conditional breakpoints as
3534 well; in this case, place @samp{thread @var{threadno}} before the
3535 breakpoint condition, like this:
3538 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
3543 @cindex stopped threads
3544 @cindex threads, stopped
3545 Whenever your program stops under @value{GDBN} for any reason,
3546 @emph{all} threads of execution stop, not just the current thread. This
3547 allows you to examine the overall state of the program, including
3548 switching between threads, without worrying that things may change
3551 @cindex continuing threads
3552 @cindex threads, continuing
3553 Conversely, whenever you restart the program, @emph{all} threads start
3554 executing. @emph{This is true even when single-stepping} with commands
3555 like @code{step} or @code{next}.
3557 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3558 Since thread scheduling is up to your debugging target's operating
3559 system (not controlled by @value{GDBN}), other threads may
3560 execute more than one statement while the current thread completes a
3561 single step. Moreover, in general other threads stop in the middle of a
3562 statement, rather than at a clean statement boundary, when the program
3565 You might even find your program stopped in another thread after
3566 continuing or even single-stepping. This happens whenever some other
3567 thread runs into a breakpoint, a signal, or an exception before the
3568 first thread completes whatever you requested.
3570 On some OSes, you can lock the OS scheduler and thus allow only a single
3574 @item set scheduler-locking @var{mode}
3575 Set the scheduler locking mode. If it is @code{off}, then there is no
3576 locking and any thread may run at any time. If @code{on}, then only the
3577 current thread may run when the inferior is resumed. The @code{step}
3578 mode optimizes for single-stepping. It stops other threads from
3579 ``seizing the prompt'' by preempting the current thread while you are
3580 stepping. Other threads will only rarely (or never) get a chance to run
3581 when you step. They are more likely to run when you @samp{next} over a
3582 function call, and they are completely free to run when you use commands
3583 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
3584 thread hits a breakpoint during its timeslice, they will never steal the
3585 @value{GDBN} prompt away from the thread that you are debugging.
3587 @item show scheduler-locking
3588 Display the current scheduler locking mode.
3593 @chapter Examining the Stack
3595 When your program has stopped, the first thing you need to know is where it
3596 stopped and how it got there.
3599 Each time your program performs a function call, information about the call
3601 That information includes the location of the call in your program,
3602 the arguments of the call,
3603 and the local variables of the function being called.
3604 The information is saved in a block of data called a @dfn{stack frame}.
3605 The stack frames are allocated in a region of memory called the @dfn{call
3608 When your program stops, the @value{GDBN} commands for examining the
3609 stack allow you to see all of this information.
3611 @cindex selected frame
3612 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3613 @value{GDBN} commands refer implicitly to the selected frame. In
3614 particular, whenever you ask @value{GDBN} for the value of a variable in
3615 your program, the value is found in the selected frame. There are
3616 special @value{GDBN} commands to select whichever frame you are
3617 interested in. @xref{Selection, ,Selecting a frame}.
3619 When your program stops, @value{GDBN} automatically selects the
3620 currently executing frame and describes it briefly, similar to the
3621 @code{frame} command (@pxref{Frame Info, ,Information about a frame}).
3624 * Frames:: Stack frames
3625 * Backtrace:: Backtraces
3626 * Selection:: Selecting a frame
3627 * Frame Info:: Information on a frame
3632 @section Stack frames
3634 @cindex frame, definition
3636 The call stack is divided up into contiguous pieces called @dfn{stack
3637 frames}, or @dfn{frames} for short; each frame is the data associated
3638 with one call to one function. The frame contains the arguments given
3639 to the function, the function's local variables, and the address at
3640 which the function is executing.
3642 @cindex initial frame
3643 @cindex outermost frame
3644 @cindex innermost frame
3645 When your program is started, the stack has only one frame, that of the
3646 function @code{main}. This is called the @dfn{initial} frame or the
3647 @dfn{outermost} frame. Each time a function is called, a new frame is
3648 made. Each time a function returns, the frame for that function invocation
3649 is eliminated. If a function is recursive, there can be many frames for
3650 the same function. The frame for the function in which execution is
3651 actually occurring is called the @dfn{innermost} frame. This is the most
3652 recently created of all the stack frames that still exist.
3654 @cindex frame pointer
3655 Inside your program, stack frames are identified by their addresses. A
3656 stack frame consists of many bytes, each of which has its own address; each
3657 kind of computer has a convention for choosing one byte whose
3658 address serves as the address of the frame. Usually this address is kept
3659 in a register called the @dfn{frame pointer register} while execution is
3660 going on in that frame.
3662 @cindex frame number
3663 @value{GDBN} assigns numbers to all existing stack frames, starting with
3664 zero for the innermost frame, one for the frame that called it,
3665 and so on upward. These numbers do not really exist in your program;
3666 they are assigned by @value{GDBN} to give you a way of designating stack
3667 frames in @value{GDBN} commands.
3669 @c The -fomit-frame-pointer below perennially causes hbox overflow
3670 @c underflow problems.
3671 @cindex frameless execution
3672 Some compilers provide a way to compile functions so that they operate
3673 without stack frames. (For example, the @value{GCC} option
3675 @samp{-fomit-frame-pointer}
3677 generates functions without a frame.)
3678 This is occasionally done with heavily used library functions to save
3679 the frame setup time. @value{GDBN} has limited facilities for dealing
3680 with these function invocations. If the innermost function invocation
3681 has no stack frame, @value{GDBN} nevertheless regards it as though
3682 it had a separate frame, which is numbered zero as usual, allowing
3683 correct tracing of the function call chain. However, @value{GDBN} has
3684 no provision for frameless functions elsewhere in the stack.
3687 @kindex frame@r{, command}
3688 @cindex current stack frame
3689 @item frame @var{args}
3690 The @code{frame} command allows you to move from one stack frame to another,
3691 and to print the stack frame you select. @var{args} may be either the
3692 address of the frame or the stack frame number. Without an argument,
3693 @code{frame} prints the current stack frame.
3695 @kindex select-frame
3696 @cindex selecting frame silently
3698 The @code{select-frame} command allows you to move from one stack frame
3699 to another without printing the frame. This is the silent version of
3708 @cindex stack traces
3709 A backtrace is a summary of how your program got where it is. It shows one
3710 line per frame, for many frames, starting with the currently executing
3711 frame (frame zero), followed by its caller (frame one), and on up the
3716 @kindex bt @r{(@code{backtrace})}
3719 Print a backtrace of the entire stack: one line per frame for all
3720 frames in the stack.
3722 You can stop the backtrace at any time by typing the system interrupt
3723 character, normally @kbd{C-c}.
3725 @item backtrace @var{n}
3727 Similar, but print only the innermost @var{n} frames.
3729 @item backtrace -@var{n}
3731 Similar, but print only the outermost @var{n} frames.
3736 @kindex info s @r{(@code{info stack})}
3737 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3738 are additional aliases for @code{backtrace}.
3740 Each line in the backtrace shows the frame number and the function name.
3741 The program counter value is also shown---unless you use @code{set
3742 print address off}. The backtrace also shows the source file name and
3743 line number, as well as the arguments to the function. The program
3744 counter value is omitted if it is at the beginning of the code for that
3747 Here is an example of a backtrace. It was made with the command
3748 @samp{bt 3}, so it shows the innermost three frames.
3752 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3754 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3755 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3757 (More stack frames follow...)
3762 The display for frame zero does not begin with a program counter
3763 value, indicating that your program has stopped at the beginning of the
3764 code for line @code{993} of @code{builtin.c}.
3767 @section Selecting a frame
3769 Most commands for examining the stack and other data in your program work on
3770 whichever stack frame is selected at the moment. Here are the commands for
3771 selecting a stack frame; all of them finish by printing a brief description
3772 of the stack frame just selected.
3775 @kindex frame@r{, selecting}
3776 @kindex f @r{(@code{frame})}
3779 Select frame number @var{n}. Recall that frame zero is the innermost
3780 (currently executing) frame, frame one is the frame that called the
3781 innermost one, and so on. The highest-numbered frame is the one for
3784 @item frame @var{addr}
3786 Select the frame at address @var{addr}. This is useful mainly if the
3787 chaining of stack frames has been damaged by a bug, making it
3788 impossible for @value{GDBN} to assign numbers properly to all frames. In
3789 addition, this can be useful when your program has multiple stacks and
3790 switches between them.
3792 On the SPARC architecture, @code{frame} needs two addresses to
3793 select an arbitrary frame: a frame pointer and a stack pointer.
3795 On the MIPS and Alpha architecture, it needs two addresses: a stack
3796 pointer and a program counter.
3798 On the 29k architecture, it needs three addresses: a register stack
3799 pointer, a program counter, and a memory stack pointer.
3800 @c note to future updaters: this is conditioned on a flag
3801 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3802 @c as of 27 Jan 1994.
3806 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3807 advances toward the outermost frame, to higher frame numbers, to frames
3808 that have existed longer. @var{n} defaults to one.
3811 @kindex do @r{(@code{down})}
3813 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3814 advances toward the innermost frame, to lower frame numbers, to frames
3815 that were created more recently. @var{n} defaults to one. You may
3816 abbreviate @code{down} as @code{do}.
3819 All of these commands end by printing two lines of output describing the
3820 frame. The first line shows the frame number, the function name, the
3821 arguments, and the source file and line number of execution in that
3822 frame. The second line shows the text of that source line.
3830 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3832 10 read_input_file (argv[i]);
3836 After such a printout, the @code{list} command with no arguments
3837 prints ten lines centered on the point of execution in the frame.
3838 @xref{List, ,Printing source lines}.
3841 @kindex down-silently
3843 @item up-silently @var{n}
3844 @itemx down-silently @var{n}
3845 These two commands are variants of @code{up} and @code{down},
3846 respectively; they differ in that they do their work silently, without
3847 causing display of the new frame. They are intended primarily for use
3848 in @value{GDBN} command scripts, where the output might be unnecessary and
3853 @section Information about a frame
3855 There are several other commands to print information about the selected
3861 When used without any argument, this command does not change which
3862 frame is selected, but prints a brief description of the currently
3863 selected stack frame. It can be abbreviated @code{f}. With an
3864 argument, this command is used to select a stack frame.
3865 @xref{Selection, ,Selecting a frame}.
3868 @kindex info f @r{(@code{info frame})}
3871 This command prints a verbose description of the selected stack frame,
3876 the address of the frame
3878 the address of the next frame down (called by this frame)
3880 the address of the next frame up (caller of this frame)
3882 the language in which the source code corresponding to this frame is written
3884 the address of the frame's arguments
3886 the address of the frame's local variables
3888 the program counter saved in it (the address of execution in the caller frame)
3890 which registers were saved in the frame
3893 @noindent The verbose description is useful when
3894 something has gone wrong that has made the stack format fail to fit
3895 the usual conventions.
3897 @item info frame @var{addr}
3898 @itemx info f @var{addr}
3899 Print a verbose description of the frame at address @var{addr}, without
3900 selecting that frame. The selected frame remains unchanged by this
3901 command. This requires the same kind of address (more than one for some
3902 architectures) that you specify in the @code{frame} command.
3903 @xref{Selection, ,Selecting a frame}.
3907 Print the arguments of the selected frame, each on a separate line.
3911 Print the local variables of the selected frame, each on a separate
3912 line. These are all variables (declared either static or automatic)
3913 accessible at the point of execution of the selected frame.
3916 @cindex catch exceptions, list active handlers
3917 @cindex exception handlers, how to list
3919 Print a list of all the exception handlers that are active in the
3920 current stack frame at the current point of execution. To see other
3921 exception handlers, visit the associated frame (using the @code{up},
3922 @code{down}, or @code{frame} commands); then type @code{info catch}.
3923 @xref{Set Catchpoints, , Setting catchpoints}.
3929 @chapter Examining Source Files
3931 @value{GDBN} can print parts of your program's source, since the debugging
3932 information recorded in the program tells @value{GDBN} what source files were
3933 used to build it. When your program stops, @value{GDBN} spontaneously prints
3934 the line where it stopped. Likewise, when you select a stack frame
3935 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
3936 execution in that frame has stopped. You can print other portions of
3937 source files by explicit command.
3939 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
3940 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
3941 @value{GDBN} under @sc{gnu} Emacs}.
3944 * List:: Printing source lines
3945 * Search:: Searching source files
3946 * Source Path:: Specifying source directories
3947 * Machine Code:: Source and machine code
3951 @section Printing source lines
3954 @kindex l @r{(@code{list})}
3955 To print lines from a source file, use the @code{list} command
3956 (abbreviated @code{l}). By default, ten lines are printed.
3957 There are several ways to specify what part of the file you want to print.
3959 Here are the forms of the @code{list} command most commonly used:
3962 @item list @var{linenum}
3963 Print lines centered around line number @var{linenum} in the
3964 current source file.
3966 @item list @var{function}
3967 Print lines centered around the beginning of function
3971 Print more lines. If the last lines printed were printed with a
3972 @code{list} command, this prints lines following the last lines
3973 printed; however, if the last line printed was a solitary line printed
3974 as part of displaying a stack frame (@pxref{Stack, ,Examining the
3975 Stack}), this prints lines centered around that line.
3978 Print lines just before the lines last printed.
3981 By default, @value{GDBN} prints ten source lines with any of these forms of
3982 the @code{list} command. You can change this using @code{set listsize}:
3985 @kindex set listsize
3986 @item set listsize @var{count}
3987 Make the @code{list} command display @var{count} source lines (unless
3988 the @code{list} argument explicitly specifies some other number).
3990 @kindex show listsize
3992 Display the number of lines that @code{list} prints.
3995 Repeating a @code{list} command with @key{RET} discards the argument,
3996 so it is equivalent to typing just @code{list}. This is more useful
3997 than listing the same lines again. An exception is made for an
3998 argument of @samp{-}; that argument is preserved in repetition so that
3999 each repetition moves up in the source file.
4002 In general, the @code{list} command expects you to supply zero, one or two
4003 @dfn{linespecs}. Linespecs specify source lines; there are several ways
4004 of writing them, but the effect is always to specify some source line.
4005 Here is a complete description of the possible arguments for @code{list}:
4008 @item list @var{linespec}
4009 Print lines centered around the line specified by @var{linespec}.
4011 @item list @var{first},@var{last}
4012 Print lines from @var{first} to @var{last}. Both arguments are
4015 @item list ,@var{last}
4016 Print lines ending with @var{last}.
4018 @item list @var{first},
4019 Print lines starting with @var{first}.
4022 Print lines just after the lines last printed.
4025 Print lines just before the lines last printed.
4028 As described in the preceding table.
4031 Here are the ways of specifying a single source line---all the
4036 Specifies line @var{number} of the current source file.
4037 When a @code{list} command has two linespecs, this refers to
4038 the same source file as the first linespec.
4041 Specifies the line @var{offset} lines after the last line printed.
4042 When used as the second linespec in a @code{list} command that has
4043 two, this specifies the line @var{offset} lines down from the
4047 Specifies the line @var{offset} lines before the last line printed.
4049 @item @var{filename}:@var{number}
4050 Specifies line @var{number} in the source file @var{filename}.
4052 @item @var{function}
4053 Specifies the line that begins the body of the function @var{function}.
4054 For example: in C, this is the line with the open brace.
4056 @item @var{filename}:@var{function}
4057 Specifies the line of the open-brace that begins the body of the
4058 function @var{function} in the file @var{filename}. You only need the
4059 file name with a function name to avoid ambiguity when there are
4060 identically named functions in different source files.
4062 @item *@var{address}
4063 Specifies the line containing the program address @var{address}.
4064 @var{address} may be any expression.
4068 @section Searching source files
4070 @kindex reverse-search
4072 There are two commands for searching through the current source file for a
4077 @kindex forward-search
4078 @item forward-search @var{regexp}
4079 @itemx search @var{regexp}
4080 The command @samp{forward-search @var{regexp}} checks each line,
4081 starting with the one following the last line listed, for a match for
4082 @var{regexp}. It lists the line that is found. You can use the
4083 synonym @samp{search @var{regexp}} or abbreviate the command name as
4086 @item reverse-search @var{regexp}
4087 The command @samp{reverse-search @var{regexp}} checks each line, starting
4088 with the one before the last line listed and going backward, for a match
4089 for @var{regexp}. It lists the line that is found. You can abbreviate
4090 this command as @code{rev}.
4094 @section Specifying source directories
4097 @cindex directories for source files
4098 Executable programs sometimes do not record the directories of the source
4099 files from which they were compiled, just the names. Even when they do,
4100 the directories could be moved between the compilation and your debugging
4101 session. @value{GDBN} has a list of directories to search for source files;
4102 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
4103 it tries all the directories in the list, in the order they are present
4104 in the list, until it finds a file with the desired name. Note that
4105 the executable search path is @emph{not} used for this purpose. Neither is
4106 the current working directory, unless it happens to be in the source
4109 If @value{GDBN} cannot find a source file in the source path, and the
4110 object program records a directory, @value{GDBN} tries that directory
4111 too. If the source path is empty, and there is no record of the
4112 compilation directory, @value{GDBN} looks in the current directory as a
4115 Whenever you reset or rearrange the source path, @value{GDBN} clears out
4116 any information it has cached about where source files are found and where
4117 each line is in the file.
4121 When you start @value{GDBN}, its source path includes only @samp{cdir}
4122 and @samp{cwd}, in that order.
4123 To add other directories, use the @code{directory} command.
4126 @item directory @var{dirname} @dots{}
4127 @item dir @var{dirname} @dots{}
4128 Add directory @var{dirname} to the front of the source path. Several
4129 directory names may be given to this command, separated by @samp{:}
4130 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
4131 part of absolute file names) or
4132 whitespace. You may specify a directory that is already in the source
4133 path; this moves it forward, so @value{GDBN} searches it sooner.
4137 @vindex $cdir@r{, convenience variable}
4138 @vindex $cwdr@r{, convenience variable}
4139 @cindex compilation directory
4140 @cindex current directory
4141 @cindex working directory
4142 @cindex directory, current
4143 @cindex directory, compilation
4144 You can use the string @samp{$cdir} to refer to the compilation
4145 directory (if one is recorded), and @samp{$cwd} to refer to the current
4146 working directory. @samp{$cwd} is not the same as @samp{.}---the former
4147 tracks the current working directory as it changes during your @value{GDBN}
4148 session, while the latter is immediately expanded to the current
4149 directory at the time you add an entry to the source path.
4152 Reset the source path to empty again. This requires confirmation.
4154 @c RET-repeat for @code{directory} is explicitly disabled, but since
4155 @c repeating it would be a no-op we do not say that. (thanks to RMS)
4157 @item show directories
4158 @kindex show directories
4159 Print the source path: show which directories it contains.
4162 If your source path is cluttered with directories that are no longer of
4163 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
4164 versions of source. You can correct the situation as follows:
4168 Use @code{directory} with no argument to reset the source path to empty.
4171 Use @code{directory} with suitable arguments to reinstall the
4172 directories you want in the source path. You can add all the
4173 directories in one command.
4177 @section Source and machine code
4179 You can use the command @code{info line} to map source lines to program
4180 addresses (and vice versa), and the command @code{disassemble} to display
4181 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
4182 mode, the @code{info line} command causes the arrow to point to the
4183 line specified. Also, @code{info line} prints addresses in symbolic form as
4188 @item info line @var{linespec}
4189 Print the starting and ending addresses of the compiled code for
4190 source line @var{linespec}. You can specify source lines in any of
4191 the ways understood by the @code{list} command (@pxref{List, ,Printing
4195 For example, we can use @code{info line} to discover the location of
4196 the object code for the first line of function
4197 @code{m4_changequote}:
4199 @c FIXME: I think this example should also show the addresses in
4200 @c symbolic form, as they usually would be displayed.
4202 (@value{GDBP}) info line m4_changequote
4203 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
4207 We can also inquire (using @code{*@var{addr}} as the form for
4208 @var{linespec}) what source line covers a particular address:
4210 (@value{GDBP}) info line *0x63ff
4211 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
4214 @cindex @code{$_} and @code{info line}
4215 @kindex x@r{(examine), and} info line
4216 After @code{info line}, the default address for the @code{x} command
4217 is changed to the starting address of the line, so that @samp{x/i} is
4218 sufficient to begin examining the machine code (@pxref{Memory,
4219 ,Examining memory}). Also, this address is saved as the value of the
4220 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
4225 @cindex assembly instructions
4226 @cindex instructions, assembly
4227 @cindex machine instructions
4228 @cindex listing machine instructions
4230 This specialized command dumps a range of memory as machine
4231 instructions. The default memory range is the function surrounding the
4232 program counter of the selected frame. A single argument to this
4233 command is a program counter value; @value{GDBN} dumps the function
4234 surrounding this value. Two arguments specify a range of addresses
4235 (first inclusive, second exclusive) to dump.
4238 The following example shows the disassembly of a range of addresses of
4239 HP PA-RISC 2.0 code:
4242 (@value{GDBP}) disas 0x32c4 0x32e4
4243 Dump of assembler code from 0x32c4 to 0x32e4:
4244 0x32c4 <main+204>: addil 0,dp
4245 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
4246 0x32cc <main+212>: ldil 0x3000,r31
4247 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
4248 0x32d4 <main+220>: ldo 0(r31),rp
4249 0x32d8 <main+224>: addil -0x800,dp
4250 0x32dc <main+228>: ldo 0x588(r1),r26
4251 0x32e0 <main+232>: ldil 0x3000,r31
4252 End of assembler dump.
4255 Some architectures have more than one commonly-used set of instruction
4256 mnemonics or other syntax.
4259 @kindex set disassembly-flavor
4260 @cindex assembly instructions
4261 @cindex instructions, assembly
4262 @cindex machine instructions
4263 @cindex listing machine instructions
4264 @cindex Intel disassembly flavor
4265 @cindex AT&T disassembly flavor
4266 @item set disassembly-flavor @var{instruction-set}
4267 Select the instruction set to use when disassembling the
4268 program via the @code{disassemble} or @code{x/i} commands.
4270 Currently this command is only defined for the Intel x86 family. You
4271 can set @var{instruction-set} to either @code{intel} or @code{att}.
4272 The default is @code{att}, the AT&T flavor used by default by Unix
4273 assemblers for x86-based targets.
4278 @chapter Examining Data
4280 @cindex printing data
4281 @cindex examining data
4284 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
4285 @c document because it is nonstandard... Under Epoch it displays in a
4286 @c different window or something like that.
4287 The usual way to examine data in your program is with the @code{print}
4288 command (abbreviated @code{p}), or its synonym @code{inspect}. It
4289 evaluates and prints the value of an expression of the language your
4290 program is written in (@pxref{Languages, ,Using @value{GDBN} with
4291 Different Languages}).
4294 @item print @var{expr}
4295 @itemx print /@var{f} @var{expr}
4296 @var{expr} is an expression (in the source language). By default the
4297 value of @var{expr} is printed in a format appropriate to its data type;
4298 you can choose a different format by specifying @samp{/@var{f}}, where
4299 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
4303 @itemx print /@var{f}
4304 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
4305 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
4306 conveniently inspect the same value in an alternative format.
4309 A more low-level way of examining data is with the @code{x} command.
4310 It examines data in memory at a specified address and prints it in a
4311 specified format. @xref{Memory, ,Examining memory}.
4313 If you are interested in information about types, or about how the
4314 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
4315 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
4319 * Expressions:: Expressions
4320 * Variables:: Program variables
4321 * Arrays:: Artificial arrays
4322 * Output Formats:: Output formats
4323 * Memory:: Examining memory
4324 * Auto Display:: Automatic display
4325 * Print Settings:: Print settings
4326 * Value History:: Value history
4327 * Convenience Vars:: Convenience variables
4328 * Registers:: Registers
4329 * Floating Point Hardware:: Floating point hardware
4333 @section Expressions
4336 @code{print} and many other @value{GDBN} commands accept an expression and
4337 compute its value. Any kind of constant, variable or operator defined
4338 by the programming language you are using is valid in an expression in
4339 @value{GDBN}. This includes conditional expressions, function calls, casts
4340 and string constants. It unfortunately does not include symbols defined
4341 by preprocessor @code{#define} commands.
4343 @value{GDBN} supports array constants in expressions input by
4344 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
4345 you can use the command @code{print @{1, 2, 3@}} to build up an array in
4346 memory that is @code{malloc}ed in the target program.
4348 Because C is so widespread, most of the expressions shown in examples in
4349 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
4350 Languages}, for information on how to use expressions in other
4353 In this section, we discuss operators that you can use in @value{GDBN}
4354 expressions regardless of your programming language.
4356 Casts are supported in all languages, not just in C, because it is so
4357 useful to cast a number into a pointer in order to examine a structure
4358 at that address in memory.
4359 @c FIXME: casts supported---Mod2 true?
4361 @value{GDBN} supports these operators, in addition to those common
4362 to programming languages:
4366 @samp{@@} is a binary operator for treating parts of memory as arrays.
4367 @xref{Arrays, ,Artificial arrays}, for more information.
4370 @samp{::} allows you to specify a variable in terms of the file or
4371 function where it is defined. @xref{Variables, ,Program variables}.
4373 @cindex @{@var{type}@}
4374 @cindex type casting memory
4375 @cindex memory, viewing as typed object
4376 @cindex casts, to view memory
4377 @item @{@var{type}@} @var{addr}
4378 Refers to an object of type @var{type} stored at address @var{addr} in
4379 memory. @var{addr} may be any expression whose value is an integer or
4380 pointer (but parentheses are required around binary operators, just as in
4381 a cast). This construct is allowed regardless of what kind of data is
4382 normally supposed to reside at @var{addr}.
4386 @section Program variables
4388 The most common kind of expression to use is the name of a variable
4391 Variables in expressions are understood in the selected stack frame
4392 (@pxref{Selection, ,Selecting a frame}); they must be either:
4396 global (or file-static)
4403 visible according to the scope rules of the
4404 programming language from the point of execution in that frame
4407 @noindent This means that in the function
4422 you can examine and use the variable @code{a} whenever your program is
4423 executing within the function @code{foo}, but you can only use or
4424 examine the variable @code{b} while your program is executing inside
4425 the block where @code{b} is declared.
4427 @cindex variable name conflict
4428 There is an exception: you can refer to a variable or function whose
4429 scope is a single source file even if the current execution point is not
4430 in this file. But it is possible to have more than one such variable or
4431 function with the same name (in different source files). If that
4432 happens, referring to that name has unpredictable effects. If you wish,
4433 you can specify a static variable in a particular function or file,
4434 using the colon-colon notation:
4436 @cindex colon-colon, context for variables/functions
4438 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4439 @cindex @code{::}, context for variables/functions
4442 @var{file}::@var{variable}
4443 @var{function}::@var{variable}
4447 Here @var{file} or @var{function} is the name of the context for the
4448 static @var{variable}. In the case of file names, you can use quotes to
4449 make sure @value{GDBN} parses the file name as a single word---for example,
4450 to print a global value of @code{x} defined in @file{f2.c}:
4453 (@value{GDBP}) p 'f2.c'::x
4456 @cindex C++ scope resolution
4457 This use of @samp{::} is very rarely in conflict with the very similar
4458 use of the same notation in C++. @value{GDBN} also supports use of the C++
4459 scope resolution operator in @value{GDBN} expressions.
4460 @c FIXME: Um, so what happens in one of those rare cases where it's in
4463 @cindex wrong values
4464 @cindex variable values, wrong
4466 @emph{Warning:} Occasionally, a local variable may appear to have the
4467 wrong value at certain points in a function---just after entry to a new
4468 scope, and just before exit.
4470 You may see this problem when you are stepping by machine instructions.
4471 This is because, on most machines, it takes more than one instruction to
4472 set up a stack frame (including local variable definitions); if you are
4473 stepping by machine instructions, variables may appear to have the wrong
4474 values until the stack frame is completely built. On exit, it usually
4475 also takes more than one machine instruction to destroy a stack frame;
4476 after you begin stepping through that group of instructions, local
4477 variable definitions may be gone.
4479 This may also happen when the compiler does significant optimizations.
4480 To be sure of always seeing accurate values, turn off all optimization
4483 @cindex ``No symbol "foo" in current context''
4484 Another possible effect of compiler optimizations is to optimize
4485 unused variables out of existence, or assign variables to registers (as
4486 opposed to memory addresses). Depending on the support for such cases
4487 offered by the debug info format used by the compiler, @value{GDBN}
4488 might not be able to display values for such local variables. If that
4489 happens, @value{GDBN} will print a message like this:
4492 No symbol "foo" in current context.
4495 To solve such problems, either recompile without optimizations, or use a
4496 different debug info format, if the compiler supports several such
4497 formats. For example, @value{NGCC}, the @sc{gnu} C/C++ compiler usually
4498 supports the @samp{-gstabs} option. @samp{-gstabs} produces debug info
4499 in a format that is superior to formats such as COFF. You may be able
4500 to use DWARF-2 (@samp{-gdwarf-2}), which is also an effective form for
4501 debug info. See @ref{Debugging Options,,Options for Debugging Your
4502 Program or @sc{gnu} CC, gcc.info, Using @sc{gnu} CC}, for more
4507 @section Artificial arrays
4509 @cindex artificial array
4510 @kindex @@@r{, referencing memory as an array}
4511 It is often useful to print out several successive objects of the
4512 same type in memory; a section of an array, or an array of
4513 dynamically determined size for which only a pointer exists in the
4516 You can do this by referring to a contiguous span of memory as an
4517 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4518 operand of @samp{@@} should be the first element of the desired array
4519 and be an individual object. The right operand should be the desired length
4520 of the array. The result is an array value whose elements are all of
4521 the type of the left argument. The first element is actually the left
4522 argument; the second element comes from bytes of memory immediately
4523 following those that hold the first element, and so on. Here is an
4524 example. If a program says
4527 int *array = (int *) malloc (len * sizeof (int));
4531 you can print the contents of @code{array} with
4537 The left operand of @samp{@@} must reside in memory. Array values made
4538 with @samp{@@} in this way behave just like other arrays in terms of
4539 subscripting, and are coerced to pointers when used in expressions.
4540 Artificial arrays most often appear in expressions via the value history
4541 (@pxref{Value History, ,Value history}), after printing one out.
4543 Another way to create an artificial array is to use a cast.
4544 This re-interprets a value as if it were an array.
4545 The value need not be in memory:
4547 (@value{GDBP}) p/x (short[2])0x12345678
4548 $1 = @{0x1234, 0x5678@}
4551 As a convenience, if you leave the array length out (as in
4552 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
4553 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4555 (@value{GDBP}) p/x (short[])0x12345678
4556 $2 = @{0x1234, 0x5678@}
4559 Sometimes the artificial array mechanism is not quite enough; in
4560 moderately complex data structures, the elements of interest may not
4561 actually be adjacent---for example, if you are interested in the values
4562 of pointers in an array. One useful work-around in this situation is
4563 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4564 variables}) as a counter in an expression that prints the first
4565 interesting value, and then repeat that expression via @key{RET}. For
4566 instance, suppose you have an array @code{dtab} of pointers to
4567 structures, and you are interested in the values of a field @code{fv}
4568 in each structure. Here is an example of what you might type:
4578 @node Output Formats
4579 @section Output formats
4581 @cindex formatted output
4582 @cindex output formats
4583 By default, @value{GDBN} prints a value according to its data type. Sometimes
4584 this is not what you want. For example, you might want to print a number
4585 in hex, or a pointer in decimal. Or you might want to view data in memory
4586 at a certain address as a character string or as an instruction. To do
4587 these things, specify an @dfn{output format} when you print a value.
4589 The simplest use of output formats is to say how to print a value
4590 already computed. This is done by starting the arguments of the
4591 @code{print} command with a slash and a format letter. The format
4592 letters supported are:
4596 Regard the bits of the value as an integer, and print the integer in
4600 Print as integer in signed decimal.
4603 Print as integer in unsigned decimal.
4606 Print as integer in octal.
4609 Print as integer in binary. The letter @samp{t} stands for ``two''.
4610 @footnote{@samp{b} cannot be used because these format letters are also
4611 used with the @code{x} command, where @samp{b} stands for ``byte'';
4612 see @ref{Memory,,Examining memory}.}
4615 @cindex unknown address, locating
4616 Print as an address, both absolute in hexadecimal and as an offset from
4617 the nearest preceding symbol. You can use this format used to discover
4618 where (in what function) an unknown address is located:
4621 (@value{GDBP}) p/a 0x54320
4622 $3 = 0x54320 <_initialize_vx+396>
4626 Regard as an integer and print it as a character constant.
4629 Regard the bits of the value as a floating point number and print
4630 using typical floating point syntax.
4633 For example, to print the program counter in hex (@pxref{Registers}), type
4640 Note that no space is required before the slash; this is because command
4641 names in @value{GDBN} cannot contain a slash.
4643 To reprint the last value in the value history with a different format,
4644 you can use the @code{print} command with just a format and no
4645 expression. For example, @samp{p/x} reprints the last value in hex.
4648 @section Examining memory
4650 You can use the command @code{x} (for ``examine'') to examine memory in
4651 any of several formats, independently of your program's data types.
4653 @cindex examining memory
4655 @kindex x @r{(examine memory)}
4656 @item x/@var{nfu} @var{addr}
4659 Use the @code{x} command to examine memory.
4662 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4663 much memory to display and how to format it; @var{addr} is an
4664 expression giving the address where you want to start displaying memory.
4665 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4666 Several commands set convenient defaults for @var{addr}.
4669 @item @var{n}, the repeat count
4670 The repeat count is a decimal integer; the default is 1. It specifies
4671 how much memory (counting by units @var{u}) to display.
4672 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4675 @item @var{f}, the display format
4676 The display format is one of the formats used by @code{print},
4677 @samp{s} (null-terminated string), or @samp{i} (machine instruction).
4678 The default is @samp{x} (hexadecimal) initially.
4679 The default changes each time you use either @code{x} or @code{print}.
4681 @item @var{u}, the unit size
4682 The unit size is any of
4688 Halfwords (two bytes).
4690 Words (four bytes). This is the initial default.
4692 Giant words (eight bytes).
4695 Each time you specify a unit size with @code{x}, that size becomes the
4696 default unit the next time you use @code{x}. (For the @samp{s} and
4697 @samp{i} formats, the unit size is ignored and is normally not written.)
4699 @item @var{addr}, starting display address
4700 @var{addr} is the address where you want @value{GDBN} to begin displaying
4701 memory. The expression need not have a pointer value (though it may);
4702 it is always interpreted as an integer address of a byte of memory.
4703 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4704 @var{addr} is usually just after the last address examined---but several
4705 other commands also set the default address: @code{info breakpoints} (to
4706 the address of the last breakpoint listed), @code{info line} (to the
4707 starting address of a line), and @code{print} (if you use it to display
4708 a value from memory).
4711 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4712 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4713 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4714 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4715 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
4717 Since the letters indicating unit sizes are all distinct from the
4718 letters specifying output formats, you do not have to remember whether
4719 unit size or format comes first; either order works. The output
4720 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4721 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4723 Even though the unit size @var{u} is ignored for the formats @samp{s}
4724 and @samp{i}, you might still want to use a count @var{n}; for example,
4725 @samp{3i} specifies that you want to see three machine instructions,
4726 including any operands. The command @code{disassemble} gives an
4727 alternative way of inspecting machine instructions; see @ref{Machine
4728 Code,,Source and machine code}.
4730 All the defaults for the arguments to @code{x} are designed to make it
4731 easy to continue scanning memory with minimal specifications each time
4732 you use @code{x}. For example, after you have inspected three machine
4733 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4734 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4735 the repeat count @var{n} is used again; the other arguments default as
4736 for successive uses of @code{x}.
4738 @cindex @code{$_}, @code{$__}, and value history
4739 The addresses and contents printed by the @code{x} command are not saved
4740 in the value history because there is often too much of them and they
4741 would get in the way. Instead, @value{GDBN} makes these values available for
4742 subsequent use in expressions as values of the convenience variables
4743 @code{$_} and @code{$__}. After an @code{x} command, the last address
4744 examined is available for use in expressions in the convenience variable
4745 @code{$_}. The contents of that address, as examined, are available in
4746 the convenience variable @code{$__}.
4748 If the @code{x} command has a repeat count, the address and contents saved
4749 are from the last memory unit printed; this is not the same as the last
4750 address printed if several units were printed on the last line of output.
4753 @section Automatic display
4754 @cindex automatic display
4755 @cindex display of expressions
4757 If you find that you want to print the value of an expression frequently
4758 (to see how it changes), you might want to add it to the @dfn{automatic
4759 display list} so that @value{GDBN} prints its value each time your program stops.
4760 Each expression added to the list is given a number to identify it;
4761 to remove an expression from the list, you specify that number.
4762 The automatic display looks like this:
4766 3: bar[5] = (struct hack *) 0x3804
4770 This display shows item numbers, expressions and their current values. As with
4771 displays you request manually using @code{x} or @code{print}, you can
4772 specify the output format you prefer; in fact, @code{display} decides
4773 whether to use @code{print} or @code{x} depending on how elaborate your
4774 format specification is---it uses @code{x} if you specify a unit size,
4775 or one of the two formats (@samp{i} and @samp{s}) that are only
4776 supported by @code{x}; otherwise it uses @code{print}.
4780 @item display @var{expr}
4781 Add the expression @var{expr} to the list of expressions to display
4782 each time your program stops. @xref{Expressions, ,Expressions}.
4784 @code{display} does not repeat if you press @key{RET} again after using it.
4786 @item display/@var{fmt} @var{expr}
4787 For @var{fmt} specifying only a display format and not a size or
4788 count, add the expression @var{expr} to the auto-display list but
4789 arrange to display it each time in the specified format @var{fmt}.
4790 @xref{Output Formats,,Output formats}.
4792 @item display/@var{fmt} @var{addr}
4793 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4794 number of units, add the expression @var{addr} as a memory address to
4795 be examined each time your program stops. Examining means in effect
4796 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4799 For example, @samp{display/i $pc} can be helpful, to see the machine
4800 instruction about to be executed each time execution stops (@samp{$pc}
4801 is a common name for the program counter; @pxref{Registers, ,Registers}).
4804 @kindex delete display
4806 @item undisplay @var{dnums}@dots{}
4807 @itemx delete display @var{dnums}@dots{}
4808 Remove item numbers @var{dnums} from the list of expressions to display.
4810 @code{undisplay} does not repeat if you press @key{RET} after using it.
4811 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4813 @kindex disable display
4814 @item disable display @var{dnums}@dots{}
4815 Disable the display of item numbers @var{dnums}. A disabled display
4816 item is not printed automatically, but is not forgotten. It may be
4817 enabled again later.
4819 @kindex enable display
4820 @item enable display @var{dnums}@dots{}
4821 Enable display of item numbers @var{dnums}. It becomes effective once
4822 again in auto display of its expression, until you specify otherwise.
4825 Display the current values of the expressions on the list, just as is
4826 done when your program stops.
4828 @kindex info display
4830 Print the list of expressions previously set up to display
4831 automatically, each one with its item number, but without showing the
4832 values. This includes disabled expressions, which are marked as such.
4833 It also includes expressions which would not be displayed right now
4834 because they refer to automatic variables not currently available.
4837 If a display expression refers to local variables, then it does not make
4838 sense outside the lexical context for which it was set up. Such an
4839 expression is disabled when execution enters a context where one of its
4840 variables is not defined. For example, if you give the command
4841 @code{display last_char} while inside a function with an argument
4842 @code{last_char}, @value{GDBN} displays this argument while your program
4843 continues to stop inside that function. When it stops elsewhere---where
4844 there is no variable @code{last_char}---the display is disabled
4845 automatically. The next time your program stops where @code{last_char}
4846 is meaningful, you can enable the display expression once again.
4848 @node Print Settings
4849 @section Print settings
4851 @cindex format options
4852 @cindex print settings
4853 @value{GDBN} provides the following ways to control how arrays, structures,
4854 and symbols are printed.
4857 These settings are useful for debugging programs in any language:
4860 @kindex set print address
4861 @item set print address
4862 @itemx set print address on
4863 @value{GDBN} prints memory addresses showing the location of stack
4864 traces, structure values, pointer values, breakpoints, and so forth,
4865 even when it also displays the contents of those addresses. The default
4866 is @code{on}. For example, this is what a stack frame display looks like with
4867 @code{set print address on}:
4872 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
4874 530 if (lquote != def_lquote)
4878 @item set print address off
4879 Do not print addresses when displaying their contents. For example,
4880 this is the same stack frame displayed with @code{set print address off}:
4884 (@value{GDBP}) set print addr off
4886 #0 set_quotes (lq="<<", rq=">>") at input.c:530
4887 530 if (lquote != def_lquote)
4891 You can use @samp{set print address off} to eliminate all machine
4892 dependent displays from the @value{GDBN} interface. For example, with
4893 @code{print address off}, you should get the same text for backtraces on
4894 all machines---whether or not they involve pointer arguments.
4896 @kindex show print address
4897 @item show print address
4898 Show whether or not addresses are to be printed.
4901 When @value{GDBN} prints a symbolic address, it normally prints the
4902 closest earlier symbol plus an offset. If that symbol does not uniquely
4903 identify the address (for example, it is a name whose scope is a single
4904 source file), you may need to clarify. One way to do this is with
4905 @code{info line}, for example @samp{info line *0x4537}. Alternately,
4906 you can set @value{GDBN} to print the source file and line number when
4907 it prints a symbolic address:
4910 @kindex set print symbol-filename
4911 @item set print symbol-filename on
4912 Tell @value{GDBN} to print the source file name and line number of a
4913 symbol in the symbolic form of an address.
4915 @item set print symbol-filename off
4916 Do not print source file name and line number of a symbol. This is the
4919 @kindex show print symbol-filename
4920 @item show print symbol-filename
4921 Show whether or not @value{GDBN} will print the source file name and
4922 line number of a symbol in the symbolic form of an address.
4925 Another situation where it is helpful to show symbol filenames and line
4926 numbers is when disassembling code; @value{GDBN} shows you the line
4927 number and source file that corresponds to each instruction.
4929 Also, you may wish to see the symbolic form only if the address being
4930 printed is reasonably close to the closest earlier symbol:
4933 @kindex set print max-symbolic-offset
4934 @item set print max-symbolic-offset @var{max-offset}
4935 Tell @value{GDBN} to only display the symbolic form of an address if the
4936 offset between the closest earlier symbol and the address is less than
4937 @var{max-offset}. The default is 0, which tells @value{GDBN}
4938 to always print the symbolic form of an address if any symbol precedes it.
4940 @kindex show print max-symbolic-offset
4941 @item show print max-symbolic-offset
4942 Ask how large the maximum offset is that @value{GDBN} prints in a
4946 @cindex wild pointer, interpreting
4947 @cindex pointer, finding referent
4948 If you have a pointer and you are not sure where it points, try
4949 @samp{set print symbol-filename on}. Then you can determine the name
4950 and source file location of the variable where it points, using
4951 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
4952 For example, here @value{GDBN} shows that a variable @code{ptt} points
4953 at another variable @code{t}, defined in @file{hi2.c}:
4956 (@value{GDBP}) set print symbol-filename on
4957 (@value{GDBP}) p/a ptt
4958 $4 = 0xe008 <t in hi2.c>
4962 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
4963 does not show the symbol name and filename of the referent, even with
4964 the appropriate @code{set print} options turned on.
4967 Other settings control how different kinds of objects are printed:
4970 @kindex set print array
4971 @item set print array
4972 @itemx set print array on
4973 Pretty print arrays. This format is more convenient to read,
4974 but uses more space. The default is off.
4976 @item set print array off
4977 Return to compressed format for arrays.
4979 @kindex show print array
4980 @item show print array
4981 Show whether compressed or pretty format is selected for displaying
4984 @kindex set print elements
4985 @item set print elements @var{number-of-elements}
4986 Set a limit on how many elements of an array @value{GDBN} will print.
4987 If @value{GDBN} is printing a large array, it stops printing after it has
4988 printed the number of elements set by the @code{set print elements} command.
4989 This limit also applies to the display of strings.
4990 When @value{GDBN} starts, this limit is set to 200.
4991 Setting @var{number-of-elements} to zero means that the printing is unlimited.
4993 @kindex show print elements
4994 @item show print elements
4995 Display the number of elements of a large array that @value{GDBN} will print.
4996 If the number is 0, then the printing is unlimited.
4998 @kindex set print null-stop
4999 @item set print null-stop
5000 Cause @value{GDBN} to stop printing the characters of an array when the first
5001 @sc{null} is encountered. This is useful when large arrays actually
5002 contain only short strings.
5005 @kindex set print pretty
5006 @item set print pretty on
5007 Cause @value{GDBN} to print structures in an indented format with one member
5008 per line, like this:
5023 @item set print pretty off
5024 Cause @value{GDBN} to print structures in a compact format, like this:
5028 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
5029 meat = 0x54 "Pork"@}
5034 This is the default format.
5036 @kindex show print pretty
5037 @item show print pretty
5038 Show which format @value{GDBN} is using to print structures.
5040 @kindex set print sevenbit-strings
5041 @item set print sevenbit-strings on
5042 Print using only seven-bit characters; if this option is set,
5043 @value{GDBN} displays any eight-bit characters (in strings or
5044 character values) using the notation @code{\}@var{nnn}. This setting is
5045 best if you are working in English (@sc{ascii}) and you use the
5046 high-order bit of characters as a marker or ``meta'' bit.
5048 @item set print sevenbit-strings off
5049 Print full eight-bit characters. This allows the use of more
5050 international character sets, and is the default.
5052 @kindex show print sevenbit-strings
5053 @item show print sevenbit-strings
5054 Show whether or not @value{GDBN} is printing only seven-bit characters.
5056 @kindex set print union
5057 @item set print union on
5058 Tell @value{GDBN} to print unions which are contained in structures. This
5059 is the default setting.
5061 @item set print union off
5062 Tell @value{GDBN} not to print unions which are contained in structures.
5064 @kindex show print union
5065 @item show print union
5066 Ask @value{GDBN} whether or not it will print unions which are contained in
5069 For example, given the declarations
5072 typedef enum @{Tree, Bug@} Species;
5073 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
5074 typedef enum @{Caterpillar, Cocoon, Butterfly@}
5085 struct thing foo = @{Tree, @{Acorn@}@};
5089 with @code{set print union on} in effect @samp{p foo} would print
5092 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
5096 and with @code{set print union off} in effect it would print
5099 $1 = @{it = Tree, form = @{...@}@}
5105 These settings are of interest when debugging C++ programs:
5109 @kindex set print demangle
5110 @item set print demangle
5111 @itemx set print demangle on
5112 Print C++ names in their source form rather than in the encoded
5113 (``mangled'') form passed to the assembler and linker for type-safe
5114 linkage. The default is on.
5116 @kindex show print demangle
5117 @item show print demangle
5118 Show whether C++ names are printed in mangled or demangled form.
5120 @kindex set print asm-demangle
5121 @item set print asm-demangle
5122 @itemx set print asm-demangle on
5123 Print C++ names in their source form rather than their mangled form, even
5124 in assembler code printouts such as instruction disassemblies.
5127 @kindex show print asm-demangle
5128 @item show print asm-demangle
5129 Show whether C++ names in assembly listings are printed in mangled
5132 @kindex set demangle-style
5133 @cindex C++ symbol decoding style
5134 @cindex symbol decoding style, C++
5135 @item set demangle-style @var{style}
5136 Choose among several encoding schemes used by different compilers to
5137 represent C++ names. The choices for @var{style} are currently:
5141 Allow @value{GDBN} to choose a decoding style by inspecting your program.
5144 Decode based on the @sc{gnu} C++ compiler (@code{g++}) encoding algorithm.
5145 This is the default.
5148 Decode based on the HP ANSI C++ (@code{aCC}) encoding algorithm.
5151 Decode based on the Lucid C++ compiler (@code{lcc}) encoding algorithm.
5154 Decode using the algorithm in the @cite{C++ Annotated Reference Manual}.
5155 @strong{Warning:} this setting alone is not sufficient to allow
5156 debugging @code{cfront}-generated executables. @value{GDBN} would
5157 require further enhancement to permit that.
5160 If you omit @var{style}, you will see a list of possible formats.
5162 @kindex show demangle-style
5163 @item show demangle-style
5164 Display the encoding style currently in use for decoding C++ symbols.
5166 @kindex set print object
5167 @item set print object
5168 @itemx set print object on
5169 When displaying a pointer to an object, identify the @emph{actual}
5170 (derived) type of the object rather than the @emph{declared} type, using
5171 the virtual function table.
5173 @item set print object off
5174 Display only the declared type of objects, without reference to the
5175 virtual function table. This is the default setting.
5177 @kindex show print object
5178 @item show print object
5179 Show whether actual, or declared, object types are displayed.
5181 @kindex set print static-members
5182 @item set print static-members
5183 @itemx set print static-members on
5184 Print static members when displaying a C++ object. The default is on.
5186 @item set print static-members off
5187 Do not print static members when displaying a C++ object.
5189 @kindex show print static-members
5190 @item show print static-members
5191 Show whether C++ static members are printed, or not.
5193 @c These don't work with HP ANSI C++ yet.
5194 @kindex set print vtbl
5195 @item set print vtbl
5196 @itemx set print vtbl on
5197 Pretty print C++ virtual function tables. The default is off.
5198 (The @code{vtbl} commands do not work on programs compiled with the HP
5199 ANSI C++ compiler (@code{aCC}).)
5201 @item set print vtbl off
5202 Do not pretty print C++ virtual function tables.
5204 @kindex show print vtbl
5205 @item show print vtbl
5206 Show whether C++ virtual function tables are pretty printed, or not.
5210 @section Value history
5212 @cindex value history
5213 Values printed by the @code{print} command are saved in the @value{GDBN}
5214 @dfn{value history}. This allows you to refer to them in other expressions.
5215 Values are kept until the symbol table is re-read or discarded
5216 (for example with the @code{file} or @code{symbol-file} commands).
5217 When the symbol table changes, the value history is discarded,
5218 since the values may contain pointers back to the types defined in the
5223 @cindex history number
5224 The values printed are given @dfn{history numbers} by which you can
5225 refer to them. These are successive integers starting with one.
5226 @code{print} shows you the history number assigned to a value by
5227 printing @samp{$@var{num} = } before the value; here @var{num} is the
5230 To refer to any previous value, use @samp{$} followed by the value's
5231 history number. The way @code{print} labels its output is designed to
5232 remind you of this. Just @code{$} refers to the most recent value in
5233 the history, and @code{$$} refers to the value before that.
5234 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
5235 is the value just prior to @code{$$}, @code{$$1} is equivalent to
5236 @code{$$}, and @code{$$0} is equivalent to @code{$}.
5238 For example, suppose you have just printed a pointer to a structure and
5239 want to see the contents of the structure. It suffices to type
5245 If you have a chain of structures where the component @code{next} points
5246 to the next one, you can print the contents of the next one with this:
5253 You can print successive links in the chain by repeating this
5254 command---which you can do by just typing @key{RET}.
5256 Note that the history records values, not expressions. If the value of
5257 @code{x} is 4 and you type these commands:
5265 then the value recorded in the value history by the @code{print} command
5266 remains 4 even though the value of @code{x} has changed.
5271 Print the last ten values in the value history, with their item numbers.
5272 This is like @samp{p@ $$9} repeated ten times, except that @code{show
5273 values} does not change the history.
5275 @item show values @var{n}
5276 Print ten history values centered on history item number @var{n}.
5279 Print ten history values just after the values last printed. If no more
5280 values are available, @code{show values +} produces no display.
5283 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
5284 same effect as @samp{show values +}.
5286 @node Convenience Vars
5287 @section Convenience variables
5289 @cindex convenience variables
5290 @value{GDBN} provides @dfn{convenience variables} that you can use within
5291 @value{GDBN} to hold on to a value and refer to it later. These variables
5292 exist entirely within @value{GDBN}; they are not part of your program, and
5293 setting a convenience variable has no direct effect on further execution
5294 of your program. That is why you can use them freely.
5296 Convenience variables are prefixed with @samp{$}. Any name preceded by
5297 @samp{$} can be used for a convenience variable, unless it is one of
5298 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
5299 (Value history references, in contrast, are @emph{numbers} preceded
5300 by @samp{$}. @xref{Value History, ,Value history}.)
5302 You can save a value in a convenience variable with an assignment
5303 expression, just as you would set a variable in your program.
5307 set $foo = *object_ptr
5311 would save in @code{$foo} the value contained in the object pointed to by
5314 Using a convenience variable for the first time creates it, but its
5315 value is @code{void} until you assign a new value. You can alter the
5316 value with another assignment at any time.
5318 Convenience variables have no fixed types. You can assign a convenience
5319 variable any type of value, including structures and arrays, even if
5320 that variable already has a value of a different type. The convenience
5321 variable, when used as an expression, has the type of its current value.
5324 @kindex show convenience
5325 @item show convenience
5326 Print a list of convenience variables used so far, and their values.
5327 Abbreviated @code{show conv}.
5330 One of the ways to use a convenience variable is as a counter to be
5331 incremented or a pointer to be advanced. For example, to print
5332 a field from successive elements of an array of structures:
5336 print bar[$i++]->contents
5340 Repeat that command by typing @key{RET}.
5342 Some convenience variables are created automatically by @value{GDBN} and given
5343 values likely to be useful.
5346 @vindex $_@r{, convenience variable}
5348 The variable @code{$_} is automatically set by the @code{x} command to
5349 the last address examined (@pxref{Memory, ,Examining memory}). Other
5350 commands which provide a default address for @code{x} to examine also
5351 set @code{$_} to that address; these commands include @code{info line}
5352 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
5353 except when set by the @code{x} command, in which case it is a pointer
5354 to the type of @code{$__}.
5356 @vindex $__@r{, convenience variable}
5358 The variable @code{$__} is automatically set by the @code{x} command
5359 to the value found in the last address examined. Its type is chosen
5360 to match the format in which the data was printed.
5363 @vindex $_exitcode@r{, convenience variable}
5364 The variable @code{$_exitcode} is automatically set to the exit code when
5365 the program being debugged terminates.
5368 On HP-UX systems, if you refer to a function or variable name that
5369 begins with a dollar sign, @value{GDBN} searches for a user or system
5370 name first, before it searches for a convenience variable.
5376 You can refer to machine register contents, in expressions, as variables
5377 with names starting with @samp{$}. The names of registers are different
5378 for each machine; use @code{info registers} to see the names used on
5382 @kindex info registers
5383 @item info registers
5384 Print the names and values of all registers except floating-point
5385 registers (in the selected stack frame).
5387 @kindex info all-registers
5388 @cindex floating point registers
5389 @item info all-registers
5390 Print the names and values of all registers, including floating-point
5393 @item info registers @var{regname} @dots{}
5394 Print the @dfn{relativized} value of each specified register @var{regname}.
5395 As discussed in detail below, register values are normally relative to
5396 the selected stack frame. @var{regname} may be any register name valid on
5397 the machine you are using, with or without the initial @samp{$}.
5400 @value{GDBN} has four ``standard'' register names that are available (in
5401 expressions) on most machines---whenever they do not conflict with an
5402 architecture's canonical mnemonics for registers. The register names
5403 @code{$pc} and @code{$sp} are used for the program counter register and
5404 the stack pointer. @code{$fp} is used for a register that contains a
5405 pointer to the current stack frame, and @code{$ps} is used for a
5406 register that contains the processor status. For example,
5407 you could print the program counter in hex with
5414 or print the instruction to be executed next with
5421 or add four to the stack pointer@footnote{This is a way of removing
5422 one word from the stack, on machines where stacks grow downward in
5423 memory (most machines, nowadays). This assumes that the innermost
5424 stack frame is selected; setting @code{$sp} is not allowed when other
5425 stack frames are selected. To pop entire frames off the stack,
5426 regardless of machine architecture, use @code{return};
5427 see @ref{Returning, ,Returning from a function}.} with
5433 Whenever possible, these four standard register names are available on
5434 your machine even though the machine has different canonical mnemonics,
5435 so long as there is no conflict. The @code{info registers} command
5436 shows the canonical names. For example, on the SPARC, @code{info
5437 registers} displays the processor status register as @code{$psr} but you
5438 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
5439 is an alias for the @sc{eflags} register.
5441 @value{GDBN} always considers the contents of an ordinary register as an
5442 integer when the register is examined in this way. Some machines have
5443 special registers which can hold nothing but floating point; these
5444 registers are considered to have floating point values. There is no way
5445 to refer to the contents of an ordinary register as floating point value
5446 (although you can @emph{print} it as a floating point value with
5447 @samp{print/f $@var{regname}}).
5449 Some registers have distinct ``raw'' and ``virtual'' data formats. This
5450 means that the data format in which the register contents are saved by
5451 the operating system is not the same one that your program normally
5452 sees. For example, the registers of the 68881 floating point
5453 coprocessor are always saved in ``extended'' (raw) format, but all C
5454 programs expect to work with ``double'' (virtual) format. In such
5455 cases, @value{GDBN} normally works with the virtual format only (the format
5456 that makes sense for your program), but the @code{info registers} command
5457 prints the data in both formats.
5459 Normally, register values are relative to the selected stack frame
5460 (@pxref{Selection, ,Selecting a frame}). This means that you get the
5461 value that the register would contain if all stack frames farther in
5462 were exited and their saved registers restored. In order to see the
5463 true contents of hardware registers, you must select the innermost
5464 frame (with @samp{frame 0}).
5466 However, @value{GDBN} must deduce where registers are saved, from the machine
5467 code generated by your compiler. If some registers are not saved, or if
5468 @value{GDBN} is unable to locate the saved registers, the selected stack
5469 frame makes no difference.
5471 @node Floating Point Hardware
5472 @section Floating point hardware
5473 @cindex floating point
5475 Depending on the configuration, @value{GDBN} may be able to give
5476 you more information about the status of the floating point hardware.
5481 Display hardware-dependent information about the floating
5482 point unit. The exact contents and layout vary depending on the
5483 floating point chip. Currently, @samp{info float} is supported on
5484 the ARM and x86 machines.
5488 @chapter Using @value{GDBN} with Different Languages
5491 Although programming languages generally have common aspects, they are
5492 rarely expressed in the same manner. For instance, in ANSI C,
5493 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
5494 Modula-2, it is accomplished by @code{p^}. Values can also be
5495 represented (and displayed) differently. Hex numbers in C appear as
5496 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
5498 @cindex working language
5499 Language-specific information is built into @value{GDBN} for some languages,
5500 allowing you to express operations like the above in your program's
5501 native language, and allowing @value{GDBN} to output values in a manner
5502 consistent with the syntax of your program's native language. The
5503 language you use to build expressions is called the @dfn{working
5507 * Setting:: Switching between source languages
5508 * Show:: Displaying the language
5509 * Checks:: Type and range checks
5510 * Support:: Supported languages
5514 @section Switching between source languages
5516 There are two ways to control the working language---either have @value{GDBN}
5517 set it automatically, or select it manually yourself. You can use the
5518 @code{set language} command for either purpose. On startup, @value{GDBN}
5519 defaults to setting the language automatically. The working language is
5520 used to determine how expressions you type are interpreted, how values
5523 In addition to the working language, every source file that
5524 @value{GDBN} knows about has its own working language. For some object
5525 file formats, the compiler might indicate which language a particular
5526 source file is in. However, most of the time @value{GDBN} infers the
5527 language from the name of the file. The language of a source file
5528 controls whether C++ names are demangled---this way @code{backtrace} can
5529 show each frame appropriately for its own language. There is no way to
5530 set the language of a source file from within @value{GDBN}, but you can
5531 set the language associated with a filename extension. @xref{Show, ,
5532 Displaying the language}.
5534 This is most commonly a problem when you use a program, such
5535 as @code{cfront} or @code{f2c}, that generates C but is written in
5536 another language. In that case, make the
5537 program use @code{#line} directives in its C output; that way
5538 @value{GDBN} will know the correct language of the source code of the original
5539 program, and will display that source code, not the generated C code.
5542 * Filenames:: Filename extensions and languages.
5543 * Manually:: Setting the working language manually
5544 * Automatically:: Having @value{GDBN} infer the source language
5548 @subsection List of filename extensions and languages
5550 If a source file name ends in one of the following extensions, then
5551 @value{GDBN} infers that its language is the one indicated.
5576 Modula-2 source file
5580 Assembler source file. This actually behaves almost like C, but
5581 @value{GDBN} does not skip over function prologues when stepping.
5584 In addition, you may set the language associated with a filename
5585 extension. @xref{Show, , Displaying the language}.
5588 @subsection Setting the working language
5590 If you allow @value{GDBN} to set the language automatically,
5591 expressions are interpreted the same way in your debugging session and
5594 @kindex set language
5595 If you wish, you may set the language manually. To do this, issue the
5596 command @samp{set language @var{lang}}, where @var{lang} is the name of
5598 @code{c} or @code{modula-2}.
5599 For a list of the supported languages, type @samp{set language}.
5601 Setting the language manually prevents @value{GDBN} from updating the working
5602 language automatically. This can lead to confusion if you try
5603 to debug a program when the working language is not the same as the
5604 source language, when an expression is acceptable to both
5605 languages---but means different things. For instance, if the current
5606 source file were written in C, and @value{GDBN} was parsing Modula-2, a
5614 might not have the effect you intended. In C, this means to add
5615 @code{b} and @code{c} and place the result in @code{a}. The result
5616 printed would be the value of @code{a}. In Modula-2, this means to compare
5617 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
5620 @subsection Having @value{GDBN} infer the source language
5622 To have @value{GDBN} set the working language automatically, use
5623 @samp{set language local} or @samp{set language auto}. @value{GDBN}
5624 then infers the working language. That is, when your program stops in a
5625 frame (usually by encountering a breakpoint), @value{GDBN} sets the
5626 working language to the language recorded for the function in that
5627 frame. If the language for a frame is unknown (that is, if the function
5628 or block corresponding to the frame was defined in a source file that
5629 does not have a recognized extension), the current working language is
5630 not changed, and @value{GDBN} issues a warning.
5632 This may not seem necessary for most programs, which are written
5633 entirely in one source language. However, program modules and libraries
5634 written in one source language can be used by a main program written in
5635 a different source language. Using @samp{set language auto} in this
5636 case frees you from having to set the working language manually.
5639 @section Displaying the language
5641 The following commands help you find out which language is the
5642 working language, and also what language source files were written in.
5644 @kindex show language
5645 @kindex info frame@r{, show the source language}
5646 @kindex info source@r{, show the source language}
5649 Display the current working language. This is the
5650 language you can use with commands such as @code{print} to
5651 build and compute expressions that may involve variables in your program.
5654 Display the source language for this frame. This language becomes the
5655 working language if you use an identifier from this frame.
5656 @xref{Frame Info, ,Information about a frame}, to identify the other
5657 information listed here.
5660 Display the source language of this source file.
5661 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
5662 information listed here.
5665 In unusual circumstances, you may have source files with extensions
5666 not in the standard list. You can then set the extension associated
5667 with a language explicitly:
5669 @kindex set extension-language
5670 @kindex info extensions
5672 @item set extension-language @var{.ext} @var{language}
5673 Set source files with extension @var{.ext} to be assumed to be in
5674 the source language @var{language}.
5676 @item info extensions
5677 List all the filename extensions and the associated languages.
5681 @section Type and range checking
5684 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
5685 checking are included, but they do not yet have any effect. This
5686 section documents the intended facilities.
5688 @c FIXME remove warning when type/range code added
5690 Some languages are designed to guard you against making seemingly common
5691 errors through a series of compile- and run-time checks. These include
5692 checking the type of arguments to functions and operators, and making
5693 sure mathematical overflows are caught at run time. Checks such as
5694 these help to ensure a program's correctness once it has been compiled
5695 by eliminating type mismatches, and providing active checks for range
5696 errors when your program is running.
5698 @value{GDBN} can check for conditions like the above if you wish.
5699 Although @value{GDBN} does not check the statements in your program, it
5700 can check expressions entered directly into @value{GDBN} for evaluation via
5701 the @code{print} command, for example. As with the working language,
5702 @value{GDBN} can also decide whether or not to check automatically based on
5703 your program's source language. @xref{Support, ,Supported languages},
5704 for the default settings of supported languages.
5707 * Type Checking:: An overview of type checking
5708 * Range Checking:: An overview of range checking
5711 @cindex type checking
5712 @cindex checks, type
5714 @subsection An overview of type checking
5716 Some languages, such as Modula-2, are strongly typed, meaning that the
5717 arguments to operators and functions have to be of the correct type,
5718 otherwise an error occurs. These checks prevent type mismatch
5719 errors from ever causing any run-time problems. For example,
5727 The second example fails because the @code{CARDINAL} 1 is not
5728 type-compatible with the @code{REAL} 2.3.
5730 For the expressions you use in @value{GDBN} commands, you can tell the
5731 @value{GDBN} type checker to skip checking;
5732 to treat any mismatches as errors and abandon the expression;
5733 or to only issue warnings when type mismatches occur,
5734 but evaluate the expression anyway. When you choose the last of
5735 these, @value{GDBN} evaluates expressions like the second example above, but
5736 also issues a warning.
5738 Even if you turn type checking off, there may be other reasons
5739 related to type that prevent @value{GDBN} from evaluating an expression.
5740 For instance, @value{GDBN} does not know how to add an @code{int} and
5741 a @code{struct foo}. These particular type errors have nothing to do
5742 with the language in use, and usually arise from expressions, such as
5743 the one described above, which make little sense to evaluate anyway.
5745 Each language defines to what degree it is strict about type. For
5746 instance, both Modula-2 and C require the arguments to arithmetical
5747 operators to be numbers. In C, enumerated types and pointers can be
5748 represented as numbers, so that they are valid arguments to mathematical
5749 operators. @xref{Support, ,Supported languages}, for further
5750 details on specific languages.
5752 @value{GDBN} provides some additional commands for controlling the type checker:
5754 @kindex set check@r{, type}
5755 @kindex set check type
5756 @kindex show check type
5758 @item set check type auto
5759 Set type checking on or off based on the current working language.
5760 @xref{Support, ,Supported languages}, for the default settings for
5763 @item set check type on
5764 @itemx set check type off
5765 Set type checking on or off, overriding the default setting for the
5766 current working language. Issue a warning if the setting does not
5767 match the language default. If any type mismatches occur in
5768 evaluating an expression while type checking is on, @value{GDBN} prints a
5769 message and aborts evaluation of the expression.
5771 @item set check type warn
5772 Cause the type checker to issue warnings, but to always attempt to
5773 evaluate the expression. Evaluating the expression may still
5774 be impossible for other reasons. For example, @value{GDBN} cannot add
5775 numbers and structures.
5778 Show the current setting of the type checker, and whether or not @value{GDBN}
5779 is setting it automatically.
5782 @cindex range checking
5783 @cindex checks, range
5784 @node Range Checking
5785 @subsection An overview of range checking
5787 In some languages (such as Modula-2), it is an error to exceed the
5788 bounds of a type; this is enforced with run-time checks. Such range
5789 checking is meant to ensure program correctness by making sure
5790 computations do not overflow, or indices on an array element access do
5791 not exceed the bounds of the array.
5793 For expressions you use in @value{GDBN} commands, you can tell
5794 @value{GDBN} to treat range errors in one of three ways: ignore them,
5795 always treat them as errors and abandon the expression, or issue
5796 warnings but evaluate the expression anyway.
5798 A range error can result from numerical overflow, from exceeding an
5799 array index bound, or when you type a constant that is not a member
5800 of any type. Some languages, however, do not treat overflows as an
5801 error. In many implementations of C, mathematical overflow causes the
5802 result to ``wrap around'' to lower values---for example, if @var{m} is
5803 the largest integer value, and @var{s} is the smallest, then
5806 @var{m} + 1 @result{} @var{s}
5809 This, too, is specific to individual languages, and in some cases
5810 specific to individual compilers or machines. @xref{Support, ,
5811 Supported languages}, for further details on specific languages.
5813 @value{GDBN} provides some additional commands for controlling the range checker:
5815 @kindex set check@r{, range}
5816 @kindex set check range
5817 @kindex show check range
5819 @item set check range auto
5820 Set range checking on or off based on the current working language.
5821 @xref{Support, ,Supported languages}, for the default settings for
5824 @item set check range on
5825 @itemx set check range off
5826 Set range checking on or off, overriding the default setting for the
5827 current working language. A warning is issued if the setting does not
5828 match the language default. If a range error occurs and range checking is on,
5829 then a message is printed and evaluation of the expression is aborted.
5831 @item set check range warn
5832 Output messages when the @value{GDBN} range checker detects a range error,
5833 but attempt to evaluate the expression anyway. Evaluating the
5834 expression may still be impossible for other reasons, such as accessing
5835 memory that the process does not own (a typical example from many Unix
5839 Show the current setting of the range checker, and whether or not it is
5840 being set automatically by @value{GDBN}.
5844 @section Supported languages
5846 @value{GDBN} supports C, C++, Fortran, Java, Chill, assembly, and Modula-2.
5847 @c This is false ...
5848 Some @value{GDBN} features may be used in expressions regardless of the
5849 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
5850 and the @samp{@{type@}addr} construct (@pxref{Expressions,
5851 ,Expressions}) can be used with the constructs of any supported
5854 The following sections detail to what degree each source language is
5855 supported by @value{GDBN}. These sections are not meant to be language
5856 tutorials or references, but serve only as a reference guide to what the
5857 @value{GDBN} expression parser accepts, and what input and output
5858 formats should look like for different languages. There are many good
5859 books written on each of these languages; please look to these for a
5860 language reference or tutorial.
5864 * Modula-2:: Modula-2
5869 @subsection C and C++
5872 @cindex expressions in C or C++
5874 Since C and C++ are so closely related, many features of @value{GDBN} apply
5875 to both languages. Whenever this is the case, we discuss those languages
5879 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
5880 @cindex @sc{gnu} C++
5881 The C++ debugging facilities are jointly implemented by the C++
5882 compiler and @value{GDBN}. Therefore, to debug your C++ code
5883 effectively, you must compile your C++ programs with a supported
5884 C++ compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C++
5885 compiler (@code{aCC}).
5887 For best results when using @sc{gnu} C++, use the stabs debugging
5888 format. You can select that format explicitly with the @code{g++}
5889 command-line options @samp{-gstabs} or @samp{-gstabs+}. See
5890 @ref{Debugging Options,,Options for Debugging Your Program or @sc{gnu}
5891 CC, gcc.info, Using @sc{gnu} CC}, for more information.
5894 * C Operators:: C and C++ operators
5895 * C Constants:: C and C++ constants
5896 * C plus plus expressions:: C++ expressions
5897 * C Defaults:: Default settings for C and C++
5898 * C Checks:: C and C++ type and range checks
5899 * Debugging C:: @value{GDBN} and C
5900 * Debugging C plus plus:: @value{GDBN} features for C++
5904 @subsubsection C and C++ operators
5906 @cindex C and C++ operators
5908 Operators must be defined on values of specific types. For instance,
5909 @code{+} is defined on numbers, but not on structures. Operators are
5910 often defined on groups of types.
5912 For the purposes of C and C++, the following definitions hold:
5917 @emph{Integral types} include @code{int} with any of its storage-class
5918 specifiers; @code{char}; @code{enum}; and, for C++, @code{bool}.
5921 @emph{Floating-point types} include @code{float}, @code{double}, and
5922 @code{long double} (if supported by the target platform).
5925 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
5928 @emph{Scalar types} include all of the above.
5933 The following operators are supported. They are listed here
5934 in order of increasing precedence:
5938 The comma or sequencing operator. Expressions in a comma-separated list
5939 are evaluated from left to right, with the result of the entire
5940 expression being the last expression evaluated.
5943 Assignment. The value of an assignment expression is the value
5944 assigned. Defined on scalar types.
5947 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
5948 and translated to @w{@code{@var{a} = @var{a op b}}}.
5949 @w{@code{@var{op}=}} and @code{=} have the same precedence.
5950 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
5951 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
5954 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
5955 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
5959 Logical @sc{or}. Defined on integral types.
5962 Logical @sc{and}. Defined on integral types.
5965 Bitwise @sc{or}. Defined on integral types.
5968 Bitwise exclusive-@sc{or}. Defined on integral types.
5971 Bitwise @sc{and}. Defined on integral types.
5974 Equality and inequality. Defined on scalar types. The value of these
5975 expressions is 0 for false and non-zero for true.
5977 @item <@r{, }>@r{, }<=@r{, }>=
5978 Less than, greater than, less than or equal, greater than or equal.
5979 Defined on scalar types. The value of these expressions is 0 for false
5980 and non-zero for true.
5983 left shift, and right shift. Defined on integral types.
5986 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5989 Addition and subtraction. Defined on integral types, floating-point types and
5992 @item *@r{, }/@r{, }%
5993 Multiplication, division, and modulus. Multiplication and division are
5994 defined on integral and floating-point types. Modulus is defined on
5998 Increment and decrement. When appearing before a variable, the
5999 operation is performed before the variable is used in an expression;
6000 when appearing after it, the variable's value is used before the
6001 operation takes place.
6004 Pointer dereferencing. Defined on pointer types. Same precedence as
6008 Address operator. Defined on variables. Same precedence as @code{++}.
6010 For debugging C++, @value{GDBN} implements a use of @samp{&} beyond what is
6011 allowed in the C++ language itself: you can use @samp{&(&@var{ref})}
6012 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
6013 where a C++ reference variable (declared with @samp{&@var{ref}}) is
6017 Negative. Defined on integral and floating-point types. Same
6018 precedence as @code{++}.
6021 Logical negation. Defined on integral types. Same precedence as
6025 Bitwise complement operator. Defined on integral types. Same precedence as
6030 Structure member, and pointer-to-structure member. For convenience,
6031 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
6032 pointer based on the stored type information.
6033 Defined on @code{struct} and @code{union} data.
6036 Dereferences of pointers to members.
6039 Array indexing. @code{@var{a}[@var{i}]} is defined as
6040 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
6043 Function parameter list. Same precedence as @code{->}.
6046 C++ scope resolution operator. Defined on @code{struct}, @code{union},
6047 and @code{class} types.
6050 Doubled colons also represent the @value{GDBN} scope operator
6051 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
6055 If an operator is redefined in the user code, @value{GDBN} usually
6056 attempts to invoke the redefined version instead of using the operator's
6064 @subsubsection C and C++ constants
6066 @cindex C and C++ constants
6068 @value{GDBN} allows you to express the constants of C and C++ in the
6073 Integer constants are a sequence of digits. Octal constants are
6074 specified by a leading @samp{0} (i.e. zero), and hexadecimal constants by
6075 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
6076 @samp{l}, specifying that the constant should be treated as a
6080 Floating point constants are a sequence of digits, followed by a decimal
6081 point, followed by a sequence of digits, and optionally followed by an
6082 exponent. An exponent is of the form:
6083 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
6084 sequence of digits. The @samp{+} is optional for positive exponents.
6085 A floating-point constant may also end with a letter @samp{f} or
6086 @samp{F}, specifying that the constant should be treated as being of
6087 the @code{float} (as opposed to the default @code{double}) type; or with
6088 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
6092 Enumerated constants consist of enumerated identifiers, or their
6093 integral equivalents.
6096 Character constants are a single character surrounded by single quotes
6097 (@code{'}), or a number---the ordinal value of the corresponding character
6098 (usually its @sc{ascii} value). Within quotes, the single character may
6099 be represented by a letter or by @dfn{escape sequences}, which are of
6100 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
6101 of the character's ordinal value; or of the form @samp{\@var{x}}, where
6102 @samp{@var{x}} is a predefined special character---for example,
6103 @samp{\n} for newline.
6106 String constants are a sequence of character constants surrounded by
6107 double quotes (@code{"}). Any valid character constant (as described
6108 above) may appear. Double quotes within the string must be preceded by
6109 a backslash, so for instance @samp{"a\"b'c"} is a string of five
6113 Pointer constants are an integral value. You can also write pointers
6114 to constants using the C operator @samp{&}.
6117 Array constants are comma-separated lists surrounded by braces @samp{@{}
6118 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
6119 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
6120 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
6124 * C plus plus expressions::
6131 @node C plus plus expressions
6132 @subsubsection C++ expressions
6134 @cindex expressions in C++
6135 @value{GDBN} expression handling can interpret most C++ expressions.
6137 @cindex C++ support, not in @sc{coff}
6138 @cindex @sc{coff} versus C++
6139 @cindex C++ and object formats
6140 @cindex object formats and C++
6141 @cindex a.out and C++
6142 @cindex @sc{ecoff} and C++
6143 @cindex @sc{xcoff} and C++
6144 @cindex @sc{elf}/stabs and C++
6145 @cindex @sc{elf}/@sc{dwarf} and C++
6146 @c FIXME!! GDB may eventually be able to debug C++ using DWARF; check
6147 @c periodically whether this has happened...
6149 @emph{Warning:} @value{GDBN} can only debug C++ code if you use the
6150 proper compiler. Typically, C++ debugging depends on the use of
6151 additional debugging information in the symbol table, and thus requires
6152 special support. In particular, if your compiler generates a.out, MIPS
6153 @sc{ecoff}, RS/6000 @sc{xcoff}, or @sc{elf} with stabs extensions to the
6154 symbol table, these facilities are all available. (With @sc{gnu} CC,
6155 you can use the @samp{-gstabs} option to request stabs debugging
6156 extensions explicitly.) Where the object code format is standard
6157 @sc{coff} or @sc{dwarf} in @sc{elf}, on the other hand, most of the C++
6158 support in @value{GDBN} does @emph{not} work.
6163 @cindex member functions
6165 Member function calls are allowed; you can use expressions like
6168 count = aml->GetOriginal(x, y)
6171 @vindex this@r{, inside C@t{++} member functions}
6172 @cindex namespace in C++
6174 While a member function is active (in the selected stack frame), your
6175 expressions have the same namespace available as the member function;
6176 that is, @value{GDBN} allows implicit references to the class instance
6177 pointer @code{this} following the same rules as C++.
6179 @cindex call overloaded functions
6180 @cindex overloaded functions, calling
6181 @cindex type conversions in C++
6183 You can call overloaded functions; @value{GDBN} resolves the function
6184 call to the right definition, with some restrictions. @value{GDBN} does not
6185 perform overload resolution involving user-defined type conversions,
6186 calls to constructors, or instantiations of templates that do not exist
6187 in the program. It also cannot handle ellipsis argument lists or
6190 It does perform integral conversions and promotions, floating-point
6191 promotions, arithmetic conversions, pointer conversions, conversions of
6192 class objects to base classes, and standard conversions such as those of
6193 functions or arrays to pointers; it requires an exact match on the
6194 number of function arguments.
6196 Overload resolution is always performed, unless you have specified
6197 @code{set overload-resolution off}. @xref{Debugging C plus plus,
6198 ,@value{GDBN} features for C++}.
6200 You must specify @code{set overload-resolution off} in order to use an
6201 explicit function signature to call an overloaded function, as in
6203 p 'foo(char,int)'('x', 13)
6206 The @value{GDBN} command-completion facility can simplify this;
6207 see @ref{Completion, ,Command completion}.
6209 @cindex reference declarations
6211 @value{GDBN} understands variables declared as C++ references; you can use
6212 them in expressions just as you do in C++ source---they are automatically
6215 In the parameter list shown when @value{GDBN} displays a frame, the values of
6216 reference variables are not displayed (unlike other variables); this
6217 avoids clutter, since references are often used for large structures.
6218 The @emph{address} of a reference variable is always shown, unless
6219 you have specified @samp{set print address off}.
6222 @value{GDBN} supports the C++ name resolution operator @code{::}---your
6223 expressions can use it just as expressions in your program do. Since
6224 one scope may be defined in another, you can use @code{::} repeatedly if
6225 necessary, for example in an expression like
6226 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
6227 resolving name scope by reference to source files, in both C and C++
6228 debugging (@pxref{Variables, ,Program variables}).
6231 In addition, when used with HP's C++ compiler, @value{GDBN} supports
6232 calling virtual functions correctly, printing out virtual bases of
6233 objects, calling functions in a base subobject, casting objects, and
6234 invoking user-defined operators.
6237 @subsubsection C and C++ defaults
6239 @cindex C and C++ defaults
6241 If you allow @value{GDBN} to set type and range checking automatically, they
6242 both default to @code{off} whenever the working language changes to
6243 C or C++. This happens regardless of whether you or @value{GDBN}
6244 selects the working language.
6246 If you allow @value{GDBN} to set the language automatically, it
6247 recognizes source files whose names end with @file{.c}, @file{.C}, or
6248 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
6249 these files, it sets the working language to C or C++.
6250 @xref{Automatically, ,Having @value{GDBN} infer the source language},
6251 for further details.
6253 @c Type checking is (a) primarily motivated by Modula-2, and (b)
6254 @c unimplemented. If (b) changes, it might make sense to let this node
6255 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
6258 @subsubsection C and C++ type and range checks
6260 @cindex C and C++ checks
6262 By default, when @value{GDBN} parses C or C++ expressions, type checking
6263 is not used. However, if you turn type checking on, @value{GDBN}
6264 considers two variables type equivalent if:
6268 The two variables are structured and have the same structure, union, or
6272 The two variables have the same type name, or types that have been
6273 declared equivalent through @code{typedef}.
6276 @c leaving this out because neither J Gilmore nor R Pesch understand it.
6279 The two @code{struct}, @code{union}, or @code{enum} variables are
6280 declared in the same declaration. (Note: this may not be true for all C
6285 Range checking, if turned on, is done on mathematical operations. Array
6286 indices are not checked, since they are often used to index a pointer
6287 that is not itself an array.
6290 @subsubsection @value{GDBN} and C
6292 The @code{set print union} and @code{show print union} commands apply to
6293 the @code{union} type. When set to @samp{on}, any @code{union} that is
6294 inside a @code{struct} or @code{class} is also printed. Otherwise, it
6295 appears as @samp{@{...@}}.
6297 The @code{@@} operator aids in the debugging of dynamic arrays, formed
6298 with pointers and a memory allocation function. @xref{Expressions,
6302 * Debugging C plus plus::
6305 @node Debugging C plus plus
6306 @subsubsection @value{GDBN} features for C++
6308 @cindex commands for C++
6310 Some @value{GDBN} commands are particularly useful with C++, and some are
6311 designed specifically for use with C++. Here is a summary:
6314 @cindex break in overloaded functions
6315 @item @r{breakpoint menus}
6316 When you want a breakpoint in a function whose name is overloaded,
6317 @value{GDBN} breakpoint menus help you specify which function definition
6318 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
6320 @cindex overloading in C++
6321 @item rbreak @var{regex}
6322 Setting breakpoints using regular expressions is helpful for setting
6323 breakpoints on overloaded functions that are not members of any special
6325 @xref{Set Breaks, ,Setting breakpoints}.
6327 @cindex C++ exception handling
6330 Debug C++ exception handling using these commands. @xref{Set
6331 Catchpoints, , Setting catchpoints}.
6334 @item ptype @var{typename}
6335 Print inheritance relationships as well as other information for type
6337 @xref{Symbols, ,Examining the Symbol Table}.
6339 @cindex C++ symbol display
6340 @item set print demangle
6341 @itemx show print demangle
6342 @itemx set print asm-demangle
6343 @itemx show print asm-demangle
6344 Control whether C++ symbols display in their source form, both when
6345 displaying code as C++ source and when displaying disassemblies.
6346 @xref{Print Settings, ,Print settings}.
6348 @item set print object
6349 @itemx show print object
6350 Choose whether to print derived (actual) or declared types of objects.
6351 @xref{Print Settings, ,Print settings}.
6353 @item set print vtbl
6354 @itemx show print vtbl
6355 Control the format for printing virtual function tables.
6356 @xref{Print Settings, ,Print settings}.
6357 (The @code{vtbl} commands do not work on programs compiled with the HP
6358 ANSI C++ compiler (@code{aCC}).)
6360 @kindex set overload-resolution
6361 @cindex overloaded functions, overload resolution
6362 @item set overload-resolution on
6363 Enable overload resolution for C++ expression evaluation. The default
6364 is on. For overloaded functions, @value{GDBN} evaluates the arguments
6365 and searches for a function whose signature matches the argument types,
6366 using the standard C++ conversion rules (see @ref{C plus plus expressions, ,C++
6367 expressions}, for details). If it cannot find a match, it emits a
6370 @item set overload-resolution off
6371 Disable overload resolution for C++ expression evaluation. For
6372 overloaded functions that are not class member functions, @value{GDBN}
6373 chooses the first function of the specified name that it finds in the
6374 symbol table, whether or not its arguments are of the correct type. For
6375 overloaded functions that are class member functions, @value{GDBN}
6376 searches for a function whose signature @emph{exactly} matches the
6379 @item @r{Overloaded symbol names}
6380 You can specify a particular definition of an overloaded symbol, using
6381 the same notation that is used to declare such symbols in C++: type
6382 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
6383 also use the @value{GDBN} command-line word completion facilities to list the
6384 available choices, or to finish the type list for you.
6385 @xref{Completion,, Command completion}, for details on how to do this.
6389 @subsection Modula-2
6391 @cindex Modula-2, @value{GDBN} support
6393 The extensions made to @value{GDBN} to support Modula-2 only support
6394 output from the @sc{gnu} Modula-2 compiler (which is currently being
6395 developed). Other Modula-2 compilers are not currently supported, and
6396 attempting to debug executables produced by them is most likely
6397 to give an error as @value{GDBN} reads in the executable's symbol
6400 @cindex expressions in Modula-2
6402 * M2 Operators:: Built-in operators
6403 * Built-In Func/Proc:: Built-in functions and procedures
6404 * M2 Constants:: Modula-2 constants
6405 * M2 Defaults:: Default settings for Modula-2
6406 * Deviations:: Deviations from standard Modula-2
6407 * M2 Checks:: Modula-2 type and range checks
6408 * M2 Scope:: The scope operators @code{::} and @code{.}
6409 * GDB/M2:: @value{GDBN} and Modula-2
6413 @subsubsection Operators
6414 @cindex Modula-2 operators
6416 Operators must be defined on values of specific types. For instance,
6417 @code{+} is defined on numbers, but not on structures. Operators are
6418 often defined on groups of types. For the purposes of Modula-2, the
6419 following definitions hold:
6424 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
6428 @emph{Character types} consist of @code{CHAR} and its subranges.
6431 @emph{Floating-point types} consist of @code{REAL}.
6434 @emph{Pointer types} consist of anything declared as @code{POINTER TO
6438 @emph{Scalar types} consist of all of the above.
6441 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
6444 @emph{Boolean types} consist of @code{BOOLEAN}.
6448 The following operators are supported, and appear in order of
6449 increasing precedence:
6453 Function argument or array index separator.
6456 Assignment. The value of @var{var} @code{:=} @var{value} is
6460 Less than, greater than on integral, floating-point, or enumerated
6464 Less than or equal to, greater than or equal to
6465 on integral, floating-point and enumerated types, or set inclusion on
6466 set types. Same precedence as @code{<}.
6468 @item =@r{, }<>@r{, }#
6469 Equality and two ways of expressing inequality, valid on scalar types.
6470 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
6471 available for inequality, since @code{#} conflicts with the script
6475 Set membership. Defined on set types and the types of their members.
6476 Same precedence as @code{<}.
6479 Boolean disjunction. Defined on boolean types.
6482 Boolean conjunction. Defined on boolean types.
6485 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
6488 Addition and subtraction on integral and floating-point types, or union
6489 and difference on set types.
6492 Multiplication on integral and floating-point types, or set intersection
6496 Division on floating-point types, or symmetric set difference on set
6497 types. Same precedence as @code{*}.
6500 Integer division and remainder. Defined on integral types. Same
6501 precedence as @code{*}.
6504 Negative. Defined on @code{INTEGER} and @code{REAL} data.
6507 Pointer dereferencing. Defined on pointer types.
6510 Boolean negation. Defined on boolean types. Same precedence as
6514 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
6515 precedence as @code{^}.
6518 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
6521 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
6525 @value{GDBN} and Modula-2 scope operators.
6529 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
6530 treats the use of the operator @code{IN}, or the use of operators
6531 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
6532 @code{<=}, and @code{>=} on sets as an error.
6535 @cindex Modula-2 built-ins
6536 @node Built-In Func/Proc
6537 @subsubsection Built-in functions and procedures
6539 Modula-2 also makes available several built-in procedures and functions.
6540 In describing these, the following metavariables are used:
6545 represents an @code{ARRAY} variable.
6548 represents a @code{CHAR} constant or variable.
6551 represents a variable or constant of integral type.
6554 represents an identifier that belongs to a set. Generally used in the
6555 same function with the metavariable @var{s}. The type of @var{s} should
6556 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
6559 represents a variable or constant of integral or floating-point type.
6562 represents a variable or constant of floating-point type.
6568 represents a variable.
6571 represents a variable or constant of one of many types. See the
6572 explanation of the function for details.
6575 All Modula-2 built-in procedures also return a result, described below.
6579 Returns the absolute value of @var{n}.
6582 If @var{c} is a lower case letter, it returns its upper case
6583 equivalent, otherwise it returns its argument.
6586 Returns the character whose ordinal value is @var{i}.
6589 Decrements the value in the variable @var{v} by one. Returns the new value.
6591 @item DEC(@var{v},@var{i})
6592 Decrements the value in the variable @var{v} by @var{i}. Returns the
6595 @item EXCL(@var{m},@var{s})
6596 Removes the element @var{m} from the set @var{s}. Returns the new
6599 @item FLOAT(@var{i})
6600 Returns the floating point equivalent of the integer @var{i}.
6603 Returns the index of the last member of @var{a}.
6606 Increments the value in the variable @var{v} by one. Returns the new value.
6608 @item INC(@var{v},@var{i})
6609 Increments the value in the variable @var{v} by @var{i}. Returns the
6612 @item INCL(@var{m},@var{s})
6613 Adds the element @var{m} to the set @var{s} if it is not already
6614 there. Returns the new set.
6617 Returns the maximum value of the type @var{t}.
6620 Returns the minimum value of the type @var{t}.
6623 Returns boolean TRUE if @var{i} is an odd number.
6626 Returns the ordinal value of its argument. For example, the ordinal
6627 value of a character is its @sc{ascii} value (on machines supporting the
6628 @sc{ascii} character set). @var{x} must be of an ordered type, which include
6629 integral, character and enumerated types.
6632 Returns the size of its argument. @var{x} can be a variable or a type.
6634 @item TRUNC(@var{r})
6635 Returns the integral part of @var{r}.
6637 @item VAL(@var{t},@var{i})
6638 Returns the member of the type @var{t} whose ordinal value is @var{i}.
6642 @emph{Warning:} Sets and their operations are not yet supported, so
6643 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
6647 @cindex Modula-2 constants
6649 @subsubsection Constants
6651 @value{GDBN} allows you to express the constants of Modula-2 in the following
6657 Integer constants are simply a sequence of digits. When used in an
6658 expression, a constant is interpreted to be type-compatible with the
6659 rest of the expression. Hexadecimal integers are specified by a
6660 trailing @samp{H}, and octal integers by a trailing @samp{B}.
6663 Floating point constants appear as a sequence of digits, followed by a
6664 decimal point and another sequence of digits. An optional exponent can
6665 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
6666 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
6667 digits of the floating point constant must be valid decimal (base 10)
6671 Character constants consist of a single character enclosed by a pair of
6672 like quotes, either single (@code{'}) or double (@code{"}). They may
6673 also be expressed by their ordinal value (their @sc{ascii} value, usually)
6674 followed by a @samp{C}.
6677 String constants consist of a sequence of characters enclosed by a
6678 pair of like quotes, either single (@code{'}) or double (@code{"}).
6679 Escape sequences in the style of C are also allowed. @xref{C
6680 Constants, ,C and C++ constants}, for a brief explanation of escape
6684 Enumerated constants consist of an enumerated identifier.
6687 Boolean constants consist of the identifiers @code{TRUE} and
6691 Pointer constants consist of integral values only.
6694 Set constants are not yet supported.
6698 @subsubsection Modula-2 defaults
6699 @cindex Modula-2 defaults
6701 If type and range checking are set automatically by @value{GDBN}, they
6702 both default to @code{on} whenever the working language changes to
6703 Modula-2. This happens regardless of whether you or @value{GDBN}
6704 selected the working language.
6706 If you allow @value{GDBN} to set the language automatically, then entering
6707 code compiled from a file whose name ends with @file{.mod} sets the
6708 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
6709 the language automatically}, for further details.
6712 @subsubsection Deviations from standard Modula-2
6713 @cindex Modula-2, deviations from
6715 A few changes have been made to make Modula-2 programs easier to debug.
6716 This is done primarily via loosening its type strictness:
6720 Unlike in standard Modula-2, pointer constants can be formed by
6721 integers. This allows you to modify pointer variables during
6722 debugging. (In standard Modula-2, the actual address contained in a
6723 pointer variable is hidden from you; it can only be modified
6724 through direct assignment to another pointer variable or expression that
6725 returned a pointer.)
6728 C escape sequences can be used in strings and characters to represent
6729 non-printable characters. @value{GDBN} prints out strings with these
6730 escape sequences embedded. Single non-printable characters are
6731 printed using the @samp{CHR(@var{nnn})} format.
6734 The assignment operator (@code{:=}) returns the value of its right-hand
6738 All built-in procedures both modify @emph{and} return their argument.
6742 @subsubsection Modula-2 type and range checks
6743 @cindex Modula-2 checks
6746 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
6749 @c FIXME remove warning when type/range checks added
6751 @value{GDBN} considers two Modula-2 variables type equivalent if:
6755 They are of types that have been declared equivalent via a @code{TYPE
6756 @var{t1} = @var{t2}} statement
6759 They have been declared on the same line. (Note: This is true of the
6760 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
6763 As long as type checking is enabled, any attempt to combine variables
6764 whose types are not equivalent is an error.
6766 Range checking is done on all mathematical operations, assignment, array
6767 index bounds, and all built-in functions and procedures.
6770 @subsubsection The scope operators @code{::} and @code{.}
6772 @cindex @code{.}, Modula-2 scope operator
6773 @cindex colon, doubled as scope operator
6775 @vindex colon-colon@r{, in Modula-2}
6776 @c Info cannot handle :: but TeX can.
6779 @vindex ::@r{, in Modula-2}
6782 There are a few subtle differences between the Modula-2 scope operator
6783 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
6788 @var{module} . @var{id}
6789 @var{scope} :: @var{id}
6793 where @var{scope} is the name of a module or a procedure,
6794 @var{module} the name of a module, and @var{id} is any declared
6795 identifier within your program, except another module.
6797 Using the @code{::} operator makes @value{GDBN} search the scope
6798 specified by @var{scope} for the identifier @var{id}. If it is not
6799 found in the specified scope, then @value{GDBN} searches all scopes
6800 enclosing the one specified by @var{scope}.
6802 Using the @code{.} operator makes @value{GDBN} search the current scope for
6803 the identifier specified by @var{id} that was imported from the
6804 definition module specified by @var{module}. With this operator, it is
6805 an error if the identifier @var{id} was not imported from definition
6806 module @var{module}, or if @var{id} is not an identifier in
6810 @subsubsection @value{GDBN} and Modula-2
6812 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
6813 Five subcommands of @code{set print} and @code{show print} apply
6814 specifically to C and C++: @samp{vtbl}, @samp{demangle},
6815 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
6816 apply to C++, and the last to the C @code{union} type, which has no direct
6817 analogue in Modula-2.
6819 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
6820 with any language, is not useful with Modula-2. Its
6821 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
6822 created in Modula-2 as they can in C or C++. However, because an
6823 address can be specified by an integral constant, the construct
6824 @samp{@{@var{type}@}@var{adrexp}} is still useful.
6826 @cindex @code{#} in Modula-2
6827 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
6828 interpreted as the beginning of a comment. Use @code{<>} instead.
6833 The extensions made to @value{GDBN} to support Chill only support output
6834 from the @sc{gnu} Chill compiler. Other Chill compilers are not currently
6835 supported, and attempting to debug executables produced by them is most
6836 likely to give an error as @value{GDBN} reads in the executable's symbol
6839 @c This used to say "... following Chill related topics ...", but since
6840 @c menus are not shown in the printed manual, it would look awkward.
6841 This section covers the Chill related topics and the features
6842 of @value{GDBN} which support these topics.
6845 * How modes are displayed:: How modes are displayed
6846 * Locations:: Locations and their accesses
6847 * Values and their Operations:: Values and their Operations
6848 * Chill type and range checks::
6852 @node How modes are displayed
6853 @subsubsection How modes are displayed
6855 The Chill Datatype- (Mode) support of @value{GDBN} is directly related
6856 with the functionality of the @sc{gnu} Chill compiler, and therefore deviates
6857 slightly from the standard specification of the Chill language. The
6860 @c FIXME: this @table's contents effectively disable @code by using @r
6861 @c on every @item. So why does it need @code?
6863 @item @r{@emph{Discrete modes:}}
6866 @emph{Integer Modes} which are predefined by @code{BYTE, UBYTE, INT,
6869 @emph{Boolean Mode} which is predefined by @code{BOOL},
6871 @emph{Character Mode} which is predefined by @code{CHAR},
6873 @emph{Set Mode} which is displayed by the keyword @code{SET}.
6875 (@value{GDBP}) ptype x
6876 type = SET (karli = 10, susi = 20, fritzi = 100)
6878 If the type is an unnumbered set the set element values are omitted.
6880 @emph{Range Mode} which is displayed by
6882 @code{type = <basemode>(<lower bound> : <upper bound>)}
6884 where @code{<lower bound>, <upper bound>} can be of any discrete literal
6885 expression (e.g. set element names).
6888 @item @r{@emph{Powerset Mode:}}
6889 A Powerset Mode is displayed by the keyword @code{POWERSET} followed by
6890 the member mode of the powerset. The member mode can be any discrete mode.
6892 (@value{GDBP}) ptype x
6893 type = POWERSET SET (egon, hugo, otto)
6896 @item @r{@emph{Reference Modes:}}
6899 @emph{Bound Reference Mode} which is displayed by the keyword @code{REF}
6900 followed by the mode name to which the reference is bound.
6902 @emph{Free Reference Mode} which is displayed by the keyword @code{PTR}.
6905 @item @r{@emph{Procedure mode}}
6906 The procedure mode is displayed by @code{type = PROC(<parameter list>)
6907 <return mode> EXCEPTIONS (<exception list>)}. The @code{<parameter
6908 list>} is a list of the parameter modes. @code{<return mode>} indicates
6909 the mode of the result of the procedure if any. The exceptionlist lists
6910 all possible exceptions which can be raised by the procedure.
6913 @item @r{@emph{Instance mode}}
6914 The instance mode is represented by a structure, which has a static
6915 type, and is therefore not really of interest.
6918 @item @r{@emph{Synchronization Modes:}}
6921 @emph{Event Mode} which is displayed by
6923 @code{EVENT (<event length>)}
6925 where @code{(<event length>)} is optional.
6927 @emph{Buffer Mode} which is displayed by
6929 @code{BUFFER (<buffer length>)<buffer element mode>}
6931 where @code{(<buffer length>)} is optional.
6934 @item @r{@emph{Timing Modes:}}
6937 @emph{Duration Mode} which is predefined by @code{DURATION}
6939 @emph{Absolute Time Mode} which is predefined by @code{TIME}
6942 @item @r{@emph{Real Modes:}}
6943 Real Modes are predefined with @code{REAL} and @code{LONG_REAL}.
6945 @item @r{@emph{String Modes:}}
6948 @emph{Character String Mode} which is displayed by
6950 @code{CHARS(<string length>)}
6952 followed by the keyword @code{VARYING} if the String Mode is a varying
6955 @emph{Bit String Mode} which is displayed by
6962 @item @r{@emph{Array Mode:}}
6963 The Array Mode is displayed by the keyword @code{ARRAY(<range>)}
6964 followed by the element mode (which may in turn be an array mode).
6966 (@value{GDBP}) ptype x
6969 SET (karli = 10, susi = 20, fritzi = 100)
6972 @item @r{@emph{Structure Mode}}
6973 The Structure mode is displayed by the keyword @code{STRUCT(<field
6974 list>)}. The @code{<field list>} consists of names and modes of fields
6975 of the structure. Variant structures have the keyword @code{CASE <field>
6976 OF <variant fields> ESAC} in their field list. Since the current version
6977 of the GNU Chill compiler doesn't implement tag processing (no runtime
6978 checks of variant fields, and therefore no debugging info), the output
6979 always displays all variant fields.
6981 (@value{GDBP}) ptype str
6996 @subsubsection Locations and their accesses
6998 A location in Chill is an object which can contain values.
7000 A value of a location is generally accessed by the (declared) name of
7001 the location. The output conforms to the specification of values in
7002 Chill programs. How values are specified
7003 is the topic of the next section, @ref{Values and their Operations}.
7005 The pseudo-location @code{RESULT} (or @code{result}) can be used to
7006 display or change the result of a currently-active procedure:
7013 This does the same as the Chill action @code{RESULT EXPR} (which
7014 is not available in @value{GDBN}).
7016 Values of reference mode locations are printed by @code{PTR(<hex
7017 value>)} in case of a free reference mode, and by @code{(REF <reference
7018 mode>) (<hex-value>)} in case of a bound reference. @code{<hex value>}
7019 represents the address where the reference points to. To access the
7020 value of the location referenced by the pointer, use the dereference
7023 Values of procedure mode locations are displayed by
7026 (<argument modes> ) <return mode> @} <address> <name of procedure
7029 @code{<argument modes>} is a list of modes according to the parameter
7030 specification of the procedure and @code{<address>} shows the address of
7034 Locations of instance modes are displayed just like a structure with two
7035 fields specifying the @emph{process type} and the @emph{copy number} of
7036 the investigated instance location@footnote{This comes from the current
7037 implementation of instances. They are implemented as a structure (no
7038 na). The output should be something like @code{[<name of the process>;
7039 <instance number>]}.}. The field names are @code{__proc_type} and
7042 Locations of synchronization modes are displayed like a structure with
7043 the field name @code{__event_data} in case of a event mode location, and
7044 like a structure with the field @code{__buffer_data} in case of a buffer
7045 mode location (refer to previous paragraph).
7047 Structure Mode locations are printed by @code{[.<field name>: <value>,
7048 ...]}. The @code{<field name>} corresponds to the structure mode
7049 definition and the layout of @code{<value>} varies depending of the mode
7050 of the field. If the investigated structure mode location is of variant
7051 structure mode, the variant parts of the structure are enclosed in curled
7052 braces (@samp{@{@}}). Fields enclosed by @samp{@{,@}} are residing
7053 on the same memory location and represent the current values of the
7054 memory location in their specific modes. Since no tag processing is done
7055 all variants are displayed. A variant field is printed by
7056 @code{(<variant name>) = .<field name>: <value>}. (who implements the
7059 (@value{GDBP}) print str1 $4 = [.as: 0, .bs: karli, .<TAG>: { (karli) =
7060 [.cs: []], (susi) = [.ds: susi]}]
7064 Substructures of string mode-, array mode- or structure mode-values
7065 (e.g. array slices, fields of structure locations) are accessed using
7066 certain operations which are described in the next section, @ref{Values
7067 and their Operations}.
7069 A location value may be interpreted as having a different mode using the
7070 location conversion. This mode conversion is written as @code{<mode
7071 name>(<location>)}. The user has to consider that the sizes of the modes
7072 have to be equal otherwise an error occurs. Furthermore, no range
7073 checking of the location against the destination mode is performed, and
7074 therefore the result can be quite confusing.
7077 (@value{GDBP}) print int (s(3 up 4)) XXX TO be filled in !! XXX
7080 @node Values and their Operations
7081 @subsubsection Values and their Operations
7083 Values are used to alter locations, to investigate complex structures in
7084 more detail or to filter relevant information out of a large amount of
7085 data. There are several (mode dependent) operations defined which enable
7086 such investigations. These operations are not only applicable to
7087 constant values but also to locations, which can become quite useful
7088 when debugging complex structures. During parsing the command line
7089 (e.g. evaluating an expression) @value{GDBN} treats location names as
7090 the values behind these locations.
7092 This section describes how values have to be specified and which
7093 operations are legal to be used with such values.
7096 @item Literal Values
7097 Literal values are specified in the same manner as in @sc{gnu} Chill programs.
7098 For detailed specification refer to the @sc{gnu} Chill implementation Manual
7100 @c FIXME: if the Chill Manual is a Texinfo documents, the above should
7101 @c be converted to a @ref.
7106 @emph{Integer Literals} are specified in the same manner as in Chill
7107 programs (refer to the Chill Standard z200/88 chpt 5.2.4.2)
7109 @emph{Boolean Literals} are defined by @code{TRUE} and @code{FALSE}.
7111 @emph{Character Literals} are defined by @code{'<character>'}. (e.g.
7114 @emph{Set Literals} are defined by a name which was specified in a set
7115 mode. The value delivered by a Set Literal is the set value. This is
7116 comparable to an enumeration in C/C++ language.
7118 @emph{Emptiness Literal} is predefined by @code{NULL}. The value of the
7119 emptiness literal delivers either the empty reference value, the empty
7120 procedure value or the empty instance value.
7123 @emph{Character String Literals} are defined by a sequence of characters
7124 enclosed in single- or double quotes. If a single- or double quote has
7125 to be part of the string literal it has to be stuffed (specified twice).
7127 @emph{Bitstring Literals} are specified in the same manner as in Chill
7128 programs (refer z200/88 chpt 5.2.4.8).
7130 @emph{Floating point literals} are specified in the same manner as in
7131 (gnu-)Chill programs (refer @sc{gnu} Chill implementation Manual chapter 1.5).
7136 A tuple is specified by @code{<mode name>[<tuple>]}, where @code{<mode
7137 name>} can be omitted if the mode of the tuple is unambiguous. This
7138 unambiguity is derived from the context of a evaluated expression.
7139 @code{<tuple>} can be one of the following:
7142 @item @emph{Powerset Tuple}
7143 @item @emph{Array Tuple}
7144 @item @emph{Structure Tuple}
7145 Powerset tuples, array tuples and structure tuples are specified in the
7146 same manner as in Chill programs refer to z200/88 chpt 5.2.5.
7149 @item String Element Value
7150 A string element value is specified by
7152 @code{<string value>(<index>)}
7154 where @code{<index>} is a integer expression. It delivers a character
7155 value which is equivalent to the character indexed by @code{<index>} in
7158 @item String Slice Value
7159 A string slice value is specified by @code{<string value>(<slice
7160 spec>)}, where @code{<slice spec>} can be either a range of integer
7161 expressions or specified by @code{<start expr> up <size>}.
7162 @code{<size>} denotes the number of elements which the slice contains.
7163 The delivered value is a string value, which is part of the specified
7166 @item Array Element Values
7167 An array element value is specified by @code{<array value>(<expr>)} and
7168 delivers a array element value of the mode of the specified array.
7170 @item Array Slice Values
7171 An array slice is specified by @code{<array value>(<slice spec>)}, where
7172 @code{<slice spec>} can be either a range specified by expressions or by
7173 @code{<start expr> up <size>}. @code{<size>} denotes the number of
7174 arrayelements the slice contains. The delivered value is an array value
7175 which is part of the specified array.
7177 @item Structure Field Values
7178 A structure field value is derived by @code{<structure value>.<field
7179 name>}, where @code{<field name>} indicates the name of a field specified
7180 in the mode definition of the structure. The mode of the delivered value
7181 corresponds to this mode definition in the structure definition.
7183 @item Procedure Call Value
7184 The procedure call value is derived from the return value of the
7185 procedure@footnote{If a procedure call is used for instance in an
7186 expression, then this procedure is called with all its side
7187 effects. This can lead to confusing results if used carelessly.}.
7189 Values of duration mode locations are represented by @code{ULONG} literals.
7191 Values of time mode locations appear as
7193 @code{TIME(<secs>:<nsecs>)}
7198 This is not implemented yet:
7199 @item Built-in Value
7201 The following built in functions are provided:
7213 @item @code{UPPER()}
7214 @item @code{LOWER()}
7215 @item @code{LENGTH()}
7219 @item @code{ARCSIN()}
7220 @item @code{ARCCOS()}
7221 @item @code{ARCTAN()}
7228 For a detailed description refer to the GNU Chill implementation manual
7232 @item Zero-adic Operator Value
7233 The zero-adic operator value is derived from the instance value for the
7234 current active process.
7236 @item Expression Values
7237 The value delivered by an expression is the result of the evaluation of
7238 the specified expression. If there are error conditions (mode
7239 incompatibility, etc.) the evaluation of expressions is aborted with a
7240 corresponding error message. Expressions may be parenthesised which
7241 causes the evaluation of this expression before any other expression
7242 which uses the result of the parenthesised expression. The following
7243 operators are supported by @value{GDBN}:
7246 @item @code{OR, ORIF, XOR}
7247 @itemx @code{AND, ANDIF}
7249 Logical operators defined over operands of boolean mode.
7252 Equality and inequality operators defined over all modes.
7256 Relational operators defined over predefined modes.
7259 @itemx @code{*, /, MOD, REM}
7260 Arithmetic operators defined over predefined modes.
7263 Change sign operator.
7266 String concatenation operator.
7269 String repetition operator.
7272 Referenced location operator which can be used either to take the
7273 address of a location (@code{->loc}), or to dereference a reference
7274 location (@code{loc->}).
7276 @item @code{OR, XOR}
7279 Powerset and bitstring operators.
7283 Powerset inclusion operators.
7286 Membership operator.
7290 @node Chill type and range checks
7291 @subsubsection Chill type and range checks
7293 @value{GDBN} considers two Chill variables mode equivalent if the sizes
7294 of the two modes are equal. This rule applies recursively to more
7295 complex datatypes which means that complex modes are treated
7296 equivalent if all element modes (which also can be complex modes like
7297 structures, arrays, etc.) have the same size.
7299 Range checking is done on all mathematical operations, assignment, array
7300 index bounds and all built in procedures.
7302 Strong type checks are forced using the @value{GDBN} command @code{set
7303 check strong}. This enforces strong type and range checks on all
7304 operations where Chill constructs are used (expressions, built in
7305 functions, etc.) in respect to the semantics as defined in the z.200
7306 language specification.
7308 All checks can be disabled by the @value{GDBN} command @code{set check
7312 @c Deviations from the Chill Standard Z200/88
7313 see last paragraph ?
7316 @node Chill defaults
7317 @subsubsection Chill defaults
7319 If type and range checking are set automatically by @value{GDBN}, they
7320 both default to @code{on} whenever the working language changes to
7321 Chill. This happens regardless of whether you or @value{GDBN}
7322 selected the working language.
7324 If you allow @value{GDBN} to set the language automatically, then entering
7325 code compiled from a file whose name ends with @file{.ch} sets the
7326 working language to Chill. @xref{Automatically, ,Having @value{GDBN} set
7327 the language automatically}, for further details.
7330 @chapter Examining the Symbol Table
7332 The commands described in this chapter allow you to inquire about the
7333 symbols (names of variables, functions and types) defined in your
7334 program. This information is inherent in the text of your program and
7335 does not change as your program executes. @value{GDBN} finds it in your
7336 program's symbol table, in the file indicated when you started @value{GDBN}
7337 (@pxref{File Options, ,Choosing files}), or by one of the
7338 file-management commands (@pxref{Files, ,Commands to specify files}).
7340 @cindex symbol names
7341 @cindex names of symbols
7342 @cindex quoting names
7343 Occasionally, you may need to refer to symbols that contain unusual
7344 characters, which @value{GDBN} ordinarily treats as word delimiters. The
7345 most frequent case is in referring to static variables in other
7346 source files (@pxref{Variables,,Program variables}). File names
7347 are recorded in object files as debugging symbols, but @value{GDBN} would
7348 ordinarily parse a typical file name, like @file{foo.c}, as the three words
7349 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
7350 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
7357 looks up the value of @code{x} in the scope of the file @file{foo.c}.
7360 @kindex info address
7361 @item info address @var{symbol}
7362 Describe where the data for @var{symbol} is stored. For a register
7363 variable, this says which register it is kept in. For a non-register
7364 local variable, this prints the stack-frame offset at which the variable
7367 Note the contrast with @samp{print &@var{symbol}}, which does not work
7368 at all for a register variable, and for a stack local variable prints
7369 the exact address of the current instantiation of the variable.
7372 @item whatis @var{expr}
7373 Print the data type of expression @var{expr}. @var{expr} is not
7374 actually evaluated, and any side-effecting operations (such as
7375 assignments or function calls) inside it do not take place.
7376 @xref{Expressions, ,Expressions}.
7379 Print the data type of @code{$}, the last value in the value history.
7382 @item ptype @var{typename}
7383 Print a description of data type @var{typename}. @var{typename} may be
7384 the name of a type, or for C code it may have the form @samp{class
7385 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
7386 @var{union-tag}} or @samp{enum @var{enum-tag}}.
7388 @item ptype @var{expr}
7390 Print a description of the type of expression @var{expr}. @code{ptype}
7391 differs from @code{whatis} by printing a detailed description, instead
7392 of just the name of the type.
7394 For example, for this variable declaration:
7397 struct complex @{double real; double imag;@} v;
7401 the two commands give this output:
7405 (@value{GDBP}) whatis v
7406 type = struct complex
7407 (@value{GDBP}) ptype v
7408 type = struct complex @{
7416 As with @code{whatis}, using @code{ptype} without an argument refers to
7417 the type of @code{$}, the last value in the value history.
7420 @item info types @var{regexp}
7422 Print a brief description of all types whose names match @var{regexp}
7423 (or all types in your program, if you supply no argument). Each
7424 complete typename is matched as though it were a complete line; thus,
7425 @samp{i type value} gives information on all types in your program whose
7426 names include the string @code{value}, but @samp{i type ^value$} gives
7427 information only on types whose complete name is @code{value}.
7429 This command differs from @code{ptype} in two ways: first, like
7430 @code{whatis}, it does not print a detailed description; second, it
7431 lists all source files where a type is defined.
7435 Show the name of the current source file---that is, the source file for
7436 the function containing the current point of execution---and the language
7439 @kindex info sources
7441 Print the names of all source files in your program for which there is
7442 debugging information, organized into two lists: files whose symbols
7443 have already been read, and files whose symbols will be read when needed.
7445 @kindex info functions
7446 @item info functions
7447 Print the names and data types of all defined functions.
7449 @item info functions @var{regexp}
7450 Print the names and data types of all defined functions
7451 whose names contain a match for regular expression @var{regexp}.
7452 Thus, @samp{info fun step} finds all functions whose names
7453 include @code{step}; @samp{info fun ^step} finds those whose names
7454 start with @code{step}.
7456 @kindex info variables
7457 @item info variables
7458 Print the names and data types of all variables that are declared
7459 outside of functions (i.e., excluding local variables).
7461 @item info variables @var{regexp}
7462 Print the names and data types of all variables (except for local
7463 variables) whose names contain a match for regular expression
7467 This was never implemented.
7468 @kindex info methods
7470 @itemx info methods @var{regexp}
7471 The @code{info methods} command permits the user to examine all defined
7472 methods within C++ program, or (with the @var{regexp} argument) a
7473 specific set of methods found in the various C++ classes. Many
7474 C++ classes provide a large number of methods. Thus, the output
7475 from the @code{ptype} command can be overwhelming and hard to use. The
7476 @code{info-methods} command filters the methods, printing only those
7477 which match the regular-expression @var{regexp}.
7480 @cindex reloading symbols
7481 Some systems allow individual object files that make up your program to
7482 be replaced without stopping and restarting your program. For example,
7483 in VxWorks you can simply recompile a defective object file and keep on
7484 running. If you are running on one of these systems, you can allow
7485 @value{GDBN} to reload the symbols for automatically relinked modules:
7488 @kindex set symbol-reloading
7489 @item set symbol-reloading on
7490 Replace symbol definitions for the corresponding source file when an
7491 object file with a particular name is seen again.
7493 @item set symbol-reloading off
7494 Do not replace symbol definitions when encountering object files of the
7495 same name more than once. This is the default state; if you are not
7496 running on a system that permits automatic relinking of modules, you
7497 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
7498 may discard symbols when linking large programs, that may contain
7499 several modules (from different directories or libraries) with the same
7502 @kindex show symbol-reloading
7503 @item show symbol-reloading
7504 Show the current @code{on} or @code{off} setting.
7507 @kindex set opaque-type-resolution
7508 @item set opaque-type-resolution on
7509 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
7510 declared as a pointer to a @code{struct}, @code{class}, or
7511 @code{union}---for example, @code{struct MyType *}---that is used in one
7512 source file although the full declaration of @code{struct MyType} is in
7513 another source file. The default is on.
7515 A change in the setting of this subcommand will not take effect until
7516 the next time symbols for a file are loaded.
7518 @item set opaque-type-resolution off
7519 Tell @value{GDBN} not to resolve opaque types. In this case, the type
7520 is printed as follows:
7522 @{<no data fields>@}
7525 @kindex show opaque-type-resolution
7526 @item show opaque-type-resolution
7527 Show whether opaque types are resolved or not.
7529 @kindex maint print symbols
7531 @kindex maint print psymbols
7532 @cindex partial symbol dump
7533 @item maint print symbols @var{filename}
7534 @itemx maint print psymbols @var{filename}
7535 @itemx maint print msymbols @var{filename}
7536 Write a dump of debugging symbol data into the file @var{filename}.
7537 These commands are used to debug the @value{GDBN} symbol-reading code. Only
7538 symbols with debugging data are included. If you use @samp{maint print
7539 symbols}, @value{GDBN} includes all the symbols for which it has already
7540 collected full details: that is, @var{filename} reflects symbols for
7541 only those files whose symbols @value{GDBN} has read. You can use the
7542 command @code{info sources} to find out which files these are. If you
7543 use @samp{maint print psymbols} instead, the dump shows information about
7544 symbols that @value{GDBN} only knows partially---that is, symbols defined in
7545 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
7546 @samp{maint print msymbols} dumps just the minimal symbol information
7547 required for each object file from which @value{GDBN} has read some symbols.
7548 @xref{Files, ,Commands to specify files}, for a discussion of how
7549 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
7553 @chapter Altering Execution
7555 Once you think you have found an error in your program, you might want to
7556 find out for certain whether correcting the apparent error would lead to
7557 correct results in the rest of the run. You can find the answer by
7558 experiment, using the @value{GDBN} features for altering execution of the
7561 For example, you can store new values into variables or memory
7562 locations, give your program a signal, restart it at a different
7563 address, or even return prematurely from a function.
7566 * Assignment:: Assignment to variables
7567 * Jumping:: Continuing at a different address
7568 * Signaling:: Giving your program a signal
7569 * Returning:: Returning from a function
7570 * Calling:: Calling your program's functions
7571 * Patching:: Patching your program
7575 @section Assignment to variables
7578 @cindex setting variables
7579 To alter the value of a variable, evaluate an assignment expression.
7580 @xref{Expressions, ,Expressions}. For example,
7587 stores the value 4 into the variable @code{x}, and then prints the
7588 value of the assignment expression (which is 4).
7589 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
7590 information on operators in supported languages.
7592 @kindex set variable
7593 @cindex variables, setting
7594 If you are not interested in seeing the value of the assignment, use the
7595 @code{set} command instead of the @code{print} command. @code{set} is
7596 really the same as @code{print} except that the expression's value is
7597 not printed and is not put in the value history (@pxref{Value History,
7598 ,Value history}). The expression is evaluated only for its effects.
7600 If the beginning of the argument string of the @code{set} command
7601 appears identical to a @code{set} subcommand, use the @code{set
7602 variable} command instead of just @code{set}. This command is identical
7603 to @code{set} except for its lack of subcommands. For example, if your
7604 program has a variable @code{width}, you get an error if you try to set
7605 a new value with just @samp{set width=13}, because @value{GDBN} has the
7606 command @code{set width}:
7609 (@value{GDBP}) whatis width
7611 (@value{GDBP}) p width
7613 (@value{GDBP}) set width=47
7614 Invalid syntax in expression.
7618 The invalid expression, of course, is @samp{=47}. In
7619 order to actually set the program's variable @code{width}, use
7622 (@value{GDBP}) set var width=47
7625 Because the @code{set} command has many subcommands that can conflict
7626 with the names of program variables, it is a good idea to use the
7627 @code{set variable} command instead of just @code{set}. For example, if
7628 your program has a variable @code{g}, you run into problems if you try
7629 to set a new value with just @samp{set g=4}, because @value{GDBN} has
7630 the command @code{set gnutarget}, abbreviated @code{set g}:
7634 (@value{GDBP}) whatis g
7638 (@value{GDBP}) set g=4
7642 The program being debugged has been started already.
7643 Start it from the beginning? (y or n) y
7644 Starting program: /home/smith/cc_progs/a.out
7645 "/home/smith/cc_progs/a.out": can't open to read symbols:
7647 (@value{GDBP}) show g
7648 The current BFD target is "=4".
7653 The program variable @code{g} did not change, and you silently set the
7654 @code{gnutarget} to an invalid value. In order to set the variable
7658 (@value{GDBP}) set var g=4
7661 @value{GDBN} allows more implicit conversions in assignments than C; you can
7662 freely store an integer value into a pointer variable or vice versa,
7663 and you can convert any structure to any other structure that is the
7664 same length or shorter.
7665 @comment FIXME: how do structs align/pad in these conversions?
7666 @comment /doc@cygnus.com 18dec1990
7668 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
7669 construct to generate a value of specified type at a specified address
7670 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
7671 to memory location @code{0x83040} as an integer (which implies a certain size
7672 and representation in memory), and
7675 set @{int@}0x83040 = 4
7679 stores the value 4 into that memory location.
7682 @section Continuing at a different address
7684 Ordinarily, when you continue your program, you do so at the place where
7685 it stopped, with the @code{continue} command. You can instead continue at
7686 an address of your own choosing, with the following commands:
7690 @item jump @var{linespec}
7691 Resume execution at line @var{linespec}. Execution stops again
7692 immediately if there is a breakpoint there. @xref{List, ,Printing
7693 source lines}, for a description of the different forms of
7694 @var{linespec}. It is common practice to use the @code{tbreak} command
7695 in conjunction with @code{jump}. @xref{Set Breaks, ,Setting
7698 The @code{jump} command does not change the current stack frame, or
7699 the stack pointer, or the contents of any memory location or any
7700 register other than the program counter. If line @var{linespec} is in
7701 a different function from the one currently executing, the results may
7702 be bizarre if the two functions expect different patterns of arguments or
7703 of local variables. For this reason, the @code{jump} command requests
7704 confirmation if the specified line is not in the function currently
7705 executing. However, even bizarre results are predictable if you are
7706 well acquainted with the machine-language code of your program.
7708 @item jump *@var{address}
7709 Resume execution at the instruction at address @var{address}.
7712 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
7713 On many systems, you can get much the same effect as the @code{jump}
7714 command by storing a new value into the register @code{$pc}. The
7715 difference is that this does not start your program running; it only
7716 changes the address of where it @emph{will} run when you continue. For
7724 makes the next @code{continue} command or stepping command execute at
7725 address @code{0x485}, rather than at the address where your program stopped.
7726 @xref{Continuing and Stepping, ,Continuing and stepping}.
7728 The most common occasion to use the @code{jump} command is to back
7729 up---perhaps with more breakpoints set---over a portion of a program
7730 that has already executed, in order to examine its execution in more
7735 @section Giving your program a signal
7739 @item signal @var{signal}
7740 Resume execution where your program stopped, but immediately give it the
7741 signal @var{signal}. @var{signal} can be the name or the number of a
7742 signal. For example, on many systems @code{signal 2} and @code{signal
7743 SIGINT} are both ways of sending an interrupt signal.
7745 Alternatively, if @var{signal} is zero, continue execution without
7746 giving a signal. This is useful when your program stopped on account of
7747 a signal and would ordinary see the signal when resumed with the
7748 @code{continue} command; @samp{signal 0} causes it to resume without a
7751 @code{signal} does not repeat when you press @key{RET} a second time
7752 after executing the command.
7756 Invoking the @code{signal} command is not the same as invoking the
7757 @code{kill} utility from the shell. Sending a signal with @code{kill}
7758 causes @value{GDBN} to decide what to do with the signal depending on
7759 the signal handling tables (@pxref{Signals}). The @code{signal} command
7760 passes the signal directly to your program.
7764 @section Returning from a function
7767 @cindex returning from a function
7770 @itemx return @var{expression}
7771 You can cancel execution of a function call with the @code{return}
7772 command. If you give an
7773 @var{expression} argument, its value is used as the function's return
7777 When you use @code{return}, @value{GDBN} discards the selected stack frame
7778 (and all frames within it). You can think of this as making the
7779 discarded frame return prematurely. If you wish to specify a value to
7780 be returned, give that value as the argument to @code{return}.
7782 This pops the selected stack frame (@pxref{Selection, ,Selecting a
7783 frame}), and any other frames inside of it, leaving its caller as the
7784 innermost remaining frame. That frame becomes selected. The
7785 specified value is stored in the registers used for returning values
7788 The @code{return} command does not resume execution; it leaves the
7789 program stopped in the state that would exist if the function had just
7790 returned. In contrast, the @code{finish} command (@pxref{Continuing
7791 and Stepping, ,Continuing and stepping}) resumes execution until the
7792 selected stack frame returns naturally.
7795 @section Calling program functions
7797 @cindex calling functions
7800 @item call @var{expr}
7801 Evaluate the expression @var{expr} without displaying @code{void}
7805 You can use this variant of the @code{print} command if you want to
7806 execute a function from your program, but without cluttering the output
7807 with @code{void} returned values. If the result is not void, it
7808 is printed and saved in the value history.
7810 For the A29K, a user-controlled variable @code{call_scratch_address},
7811 specifies the location of a scratch area to be used when @value{GDBN}
7812 calls a function in the target. This is necessary because the usual
7813 method of putting the scratch area on the stack does not work in systems
7814 that have separate instruction and data spaces.
7817 @section Patching programs
7819 @cindex patching binaries
7820 @cindex writing into executables
7821 @cindex writing into corefiles
7823 By default, @value{GDBN} opens the file containing your program's
7824 executable code (or the corefile) read-only. This prevents accidental
7825 alterations to machine code; but it also prevents you from intentionally
7826 patching your program's binary.
7828 If you'd like to be able to patch the binary, you can specify that
7829 explicitly with the @code{set write} command. For example, you might
7830 want to turn on internal debugging flags, or even to make emergency
7836 @itemx set write off
7837 If you specify @samp{set write on}, @value{GDBN} opens executable and
7838 core files for both reading and writing; if you specify @samp{set write
7839 off} (the default), @value{GDBN} opens them read-only.
7841 If you have already loaded a file, you must load it again (using the
7842 @code{exec-file} or @code{core-file} command) after changing @code{set
7843 write}, for your new setting to take effect.
7847 Display whether executable files and core files are opened for writing
7852 @chapter @value{GDBN} Files
7854 @value{GDBN} needs to know the file name of the program to be debugged,
7855 both in order to read its symbol table and in order to start your
7856 program. To debug a core dump of a previous run, you must also tell
7857 @value{GDBN} the name of the core dump file.
7860 * Files:: Commands to specify files
7861 * Symbol Errors:: Errors reading symbol files
7865 @section Commands to specify files
7867 @cindex symbol table
7868 @cindex core dump file
7870 You may want to specify executable and core dump file names. The usual
7871 way to do this is at start-up time, using the arguments to
7872 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
7873 Out of @value{GDBN}}).
7875 Occasionally it is necessary to change to a different file during a
7876 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
7877 a file you want to use. In these situations the @value{GDBN} commands
7878 to specify new files are useful.
7881 @cindex executable file
7883 @item file @var{filename}
7884 Use @var{filename} as the program to be debugged. It is read for its
7885 symbols and for the contents of pure memory. It is also the program
7886 executed when you use the @code{run} command. If you do not specify a
7887 directory and the file is not found in the @value{GDBN} working directory,
7888 @value{GDBN} uses the environment variable @code{PATH} as a list of
7889 directories to search, just as the shell does when looking for a program
7890 to run. You can change the value of this variable, for both @value{GDBN}
7891 and your program, using the @code{path} command.
7893 On systems with memory-mapped files, an auxiliary file named
7894 @file{@var{filename}.syms} may hold symbol table information for
7895 @var{filename}. If so, @value{GDBN} maps in the symbol table from
7896 @file{@var{filename}.syms}, starting up more quickly. See the
7897 descriptions of the file options @samp{-mapped} and @samp{-readnow}
7898 (available on the command line, and with the commands @code{file},
7899 @code{symbol-file}, or @code{add-symbol-file}, described below),
7900 for more information.
7903 @code{file} with no argument makes @value{GDBN} discard any information it
7904 has on both executable file and the symbol table.
7907 @item exec-file @r{[} @var{filename} @r{]}
7908 Specify that the program to be run (but not the symbol table) is found
7909 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
7910 if necessary to locate your program. Omitting @var{filename} means to
7911 discard information on the executable file.
7914 @item symbol-file @r{[} @var{filename} @r{]}
7915 Read symbol table information from file @var{filename}. @code{PATH} is
7916 searched when necessary. Use the @code{file} command to get both symbol
7917 table and program to run from the same file.
7919 @code{symbol-file} with no argument clears out @value{GDBN} information on your
7920 program's symbol table.
7922 The @code{symbol-file} command causes @value{GDBN} to forget the contents
7923 of its convenience variables, the value history, and all breakpoints and
7924 auto-display expressions. This is because they may contain pointers to
7925 the internal data recording symbols and data types, which are part of
7926 the old symbol table data being discarded inside @value{GDBN}.
7928 @code{symbol-file} does not repeat if you press @key{RET} again after
7931 When @value{GDBN} is configured for a particular environment, it
7932 understands debugging information in whatever format is the standard
7933 generated for that environment; you may use either a @sc{gnu} compiler, or
7934 other compilers that adhere to the local conventions.
7935 Best results are usually obtained from @sc{gnu} compilers; for example,
7936 using @code{@value{GCC}} you can generate debugging information for
7939 For most kinds of object files, with the exception of old SVR3 systems
7940 using COFF, the @code{symbol-file} command does not normally read the
7941 symbol table in full right away. Instead, it scans the symbol table
7942 quickly to find which source files and which symbols are present. The
7943 details are read later, one source file at a time, as they are needed.
7945 The purpose of this two-stage reading strategy is to make @value{GDBN}
7946 start up faster. For the most part, it is invisible except for
7947 occasional pauses while the symbol table details for a particular source
7948 file are being read. (The @code{set verbose} command can turn these
7949 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
7950 warnings and messages}.)
7952 We have not implemented the two-stage strategy for COFF yet. When the
7953 symbol table is stored in COFF format, @code{symbol-file} reads the
7954 symbol table data in full right away. Note that ``stabs-in-COFF''
7955 still does the two-stage strategy, since the debug info is actually
7959 @cindex reading symbols immediately
7960 @cindex symbols, reading immediately
7962 @cindex memory-mapped symbol file
7963 @cindex saving symbol table
7964 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7965 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7966 You can override the @value{GDBN} two-stage strategy for reading symbol
7967 tables by using the @samp{-readnow} option with any of the commands that
7968 load symbol table information, if you want to be sure @value{GDBN} has the
7969 entire symbol table available.
7971 If memory-mapped files are available on your system through the
7972 @code{mmap} system call, you can use another option, @samp{-mapped}, to
7973 cause @value{GDBN} to write the symbols for your program into a reusable
7974 file. Future @value{GDBN} debugging sessions map in symbol information
7975 from this auxiliary symbol file (if the program has not changed), rather
7976 than spending time reading the symbol table from the executable
7977 program. Using the @samp{-mapped} option has the same effect as
7978 starting @value{GDBN} with the @samp{-mapped} command-line option.
7980 You can use both options together, to make sure the auxiliary symbol
7981 file has all the symbol information for your program.
7983 The auxiliary symbol file for a program called @var{myprog} is called
7984 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
7985 than the corresponding executable), @value{GDBN} always attempts to use
7986 it when you debug @var{myprog}; no special options or commands are
7989 The @file{.syms} file is specific to the host machine where you run
7990 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
7991 symbol table. It cannot be shared across multiple host platforms.
7993 @c FIXME: for now no mention of directories, since this seems to be in
7994 @c flux. 13mar1992 status is that in theory GDB would look either in
7995 @c current dir or in same dir as myprog; but issues like competing
7996 @c GDB's, or clutter in system dirs, mean that in practice right now
7997 @c only current dir is used. FFish says maybe a special GDB hierarchy
7998 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
8003 @item core-file @r{[} @var{filename} @r{]}
8004 Specify the whereabouts of a core dump file to be used as the ``contents
8005 of memory''. Traditionally, core files contain only some parts of the
8006 address space of the process that generated them; @value{GDBN} can access the
8007 executable file itself for other parts.
8009 @code{core-file} with no argument specifies that no core file is
8012 Note that the core file is ignored when your program is actually running
8013 under @value{GDBN}. So, if you have been running your program and you
8014 wish to debug a core file instead, you must kill the subprocess in which
8015 the program is running. To do this, use the @code{kill} command
8016 (@pxref{Kill Process, ,Killing the child process}).
8018 @kindex add-symbol-file
8019 @cindex dynamic linking
8020 @item add-symbol-file @var{filename} @var{address}
8021 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
8022 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address}
8023 The @code{add-symbol-file} command reads additional symbol table
8024 information from the file @var{filename}. You would use this command
8025 when @var{filename} has been dynamically loaded (by some other means)
8026 into the program that is running. @var{address} should be the memory
8027 address at which the file has been loaded; @value{GDBN} cannot figure
8028 this out for itself. You can additionally specify an arbitrary number
8029 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
8030 section name and base address for that section. You can specify any
8031 @var{address} as an expression.
8033 The symbol table of the file @var{filename} is added to the symbol table
8034 originally read with the @code{symbol-file} command. You can use the
8035 @code{add-symbol-file} command any number of times; the new symbol data
8036 thus read keeps adding to the old. To discard all old symbol data
8037 instead, use the @code{symbol-file} command without any arguments.
8039 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
8041 You can use the @samp{-mapped} and @samp{-readnow} options just as with
8042 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
8043 table information for @var{filename}.
8045 @kindex add-shared-symbol-file
8046 @item add-shared-symbol-file
8047 The @code{add-shared-symbol-file} command can be used only under Harris' CXUX
8048 operating system for the Motorola 88k. @value{GDBN} automatically looks for
8049 shared libraries, however if @value{GDBN} does not find yours, you can run
8050 @code{add-shared-symbol-file}. It takes no arguments.
8054 The @code{section} command changes the base address of section SECTION of
8055 the exec file to ADDR. This can be used if the exec file does not contain
8056 section addresses, (such as in the a.out format), or when the addresses
8057 specified in the file itself are wrong. Each section must be changed
8058 separately. The @code{info files} command, described below, lists all
8059 the sections and their addresses.
8065 @code{info files} and @code{info target} are synonymous; both print the
8066 current target (@pxref{Targets, ,Specifying a Debugging Target}),
8067 including the names of the executable and core dump files currently in
8068 use by @value{GDBN}, and the files from which symbols were loaded. The
8069 command @code{help target} lists all possible targets rather than
8074 All file-specifying commands allow both absolute and relative file names
8075 as arguments. @value{GDBN} always converts the file name to an absolute file
8076 name and remembers it that way.
8078 @cindex shared libraries
8079 @value{GDBN} supports HP-UX, SunOS, SVr4, Irix 5, and IBM RS/6000 shared
8082 @value{GDBN} automatically loads symbol definitions from shared libraries
8083 when you use the @code{run} command, or when you examine a core file.
8084 (Before you issue the @code{run} command, @value{GDBN} does not understand
8085 references to a function in a shared library, however---unless you are
8086 debugging a core file).
8088 On HP-UX, if the program loads a library explicitly, @value{GDBN}
8089 automatically loads the symbols at the time of the @code{shl_load} call.
8091 @c FIXME: some @value{GDBN} release may permit some refs to undef
8092 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
8093 @c FIXME...lib; check this from time to time when updating manual
8096 @kindex info sharedlibrary
8099 @itemx info sharedlibrary
8100 Print the names of the shared libraries which are currently loaded.
8102 @kindex sharedlibrary
8104 @item sharedlibrary @var{regex}
8105 @itemx share @var{regex}
8106 Load shared object library symbols for files matching a
8107 Unix regular expression.
8108 As with files loaded automatically, it only loads shared libraries
8109 required by your program for a core file or after typing @code{run}. If
8110 @var{regex} is omitted all shared libraries required by your program are
8114 On HP-UX systems, @value{GDBN} detects the loading of a shared library
8115 and automatically reads in symbols from the newly loaded library, up to
8116 a threshold that is initially set but that you can modify if you wish.
8118 Beyond that threshold, symbols from shared libraries must be explicitly
8119 loaded. To load these symbols, use the command @code{sharedlibrary
8120 @var{filename}}. The base address of the shared library is determined
8121 automatically by @value{GDBN} and need not be specified.
8123 To display or set the threshold, use the commands:
8126 @kindex set auto-solib-add
8127 @item set auto-solib-add @var{threshold}
8128 Set the autoloading size threshold, in megabytes. If @var{threshold} is
8129 nonzero, symbols from all shared object libraries will be loaded
8130 automatically when the inferior begins execution or when the dynamic
8131 linker informs @value{GDBN} that a new library has been loaded, until
8132 the symbol table of the program and libraries exceeds this threshold.
8133 Otherwise, symbols must be loaded manually, using the
8134 @code{sharedlibrary} command. The default threshold is 100 megabytes.
8136 @kindex show auto-solib-add
8137 @item show auto-solib-add
8138 Display the current autoloading size threshold, in megabytes.
8142 @section Errors reading symbol files
8144 While reading a symbol file, @value{GDBN} occasionally encounters problems,
8145 such as symbol types it does not recognize, or known bugs in compiler
8146 output. By default, @value{GDBN} does not notify you of such problems, since
8147 they are relatively common and primarily of interest to people
8148 debugging compilers. If you are interested in seeing information
8149 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
8150 only one message about each such type of problem, no matter how many
8151 times the problem occurs; or you can ask @value{GDBN} to print more messages,
8152 to see how many times the problems occur, with the @code{set
8153 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
8156 The messages currently printed, and their meanings, include:
8159 @item inner block not inside outer block in @var{symbol}
8161 The symbol information shows where symbol scopes begin and end
8162 (such as at the start of a function or a block of statements). This
8163 error indicates that an inner scope block is not fully contained
8164 in its outer scope blocks.
8166 @value{GDBN} circumvents the problem by treating the inner block as if it had
8167 the same scope as the outer block. In the error message, @var{symbol}
8168 may be shown as ``@code{(don't know)}'' if the outer block is not a
8171 @item block at @var{address} out of order
8173 The symbol information for symbol scope blocks should occur in
8174 order of increasing addresses. This error indicates that it does not
8177 @value{GDBN} does not circumvent this problem, and has trouble
8178 locating symbols in the source file whose symbols it is reading. (You
8179 can often determine what source file is affected by specifying
8180 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
8183 @item bad block start address patched
8185 The symbol information for a symbol scope block has a start address
8186 smaller than the address of the preceding source line. This is known
8187 to occur in the SunOS 4.1.1 (and earlier) C compiler.
8189 @value{GDBN} circumvents the problem by treating the symbol scope block as
8190 starting on the previous source line.
8192 @item bad string table offset in symbol @var{n}
8195 Symbol number @var{n} contains a pointer into the string table which is
8196 larger than the size of the string table.
8198 @value{GDBN} circumvents the problem by considering the symbol to have the
8199 name @code{foo}, which may cause other problems if many symbols end up
8202 @item unknown symbol type @code{0x@var{nn}}
8204 The symbol information contains new data types that @value{GDBN} does
8205 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
8206 uncomprehended information, in hexadecimal.
8208 @value{GDBN} circumvents the error by ignoring this symbol information.
8209 This usually allows you to debug your program, though certain symbols
8210 are not accessible. If you encounter such a problem and feel like
8211 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
8212 on @code{complain}, then go up to the function @code{read_dbx_symtab}
8213 and examine @code{*bufp} to see the symbol.
8215 @item stub type has NULL name
8217 @value{GDBN} could not find the full definition for a struct or class.
8219 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
8220 The symbol information for a C++ member function is missing some
8221 information that recent versions of the compiler should have output for
8224 @item info mismatch between compiler and debugger
8226 @value{GDBN} could not parse a type specification output by the compiler.
8231 @chapter Specifying a Debugging Target
8233 @cindex debugging target
8236 A @dfn{target} is the execution environment occupied by your program.
8238 Often, @value{GDBN} runs in the same host environment as your program;
8239 in that case, the debugging target is specified as a side effect when
8240 you use the @code{file} or @code{core} commands. When you need more
8241 flexibility---for example, running @value{GDBN} on a physically separate
8242 host, or controlling a standalone system over a serial port or a
8243 realtime system over a TCP/IP connection---you can use the @code{target}
8244 command to specify one of the target types configured for @value{GDBN}
8245 (@pxref{Target Commands, ,Commands for managing targets}).
8248 * Active Targets:: Active targets
8249 * Target Commands:: Commands for managing targets
8250 * Byte Order:: Choosing target byte order
8251 * Remote:: Remote debugging
8252 * KOD:: Kernel Object Display
8256 @node Active Targets
8257 @section Active targets
8259 @cindex stacking targets
8260 @cindex active targets
8261 @cindex multiple targets
8263 There are three classes of targets: processes, core files, and
8264 executable files. @value{GDBN} can work concurrently on up to three
8265 active targets, one in each class. This allows you to (for example)
8266 start a process and inspect its activity without abandoning your work on
8269 For example, if you execute @samp{gdb a.out}, then the executable file
8270 @code{a.out} is the only active target. If you designate a core file as
8271 well---presumably from a prior run that crashed and coredumped---then
8272 @value{GDBN} has two active targets and uses them in tandem, looking
8273 first in the corefile target, then in the executable file, to satisfy
8274 requests for memory addresses. (Typically, these two classes of target
8275 are complementary, since core files contain only a program's
8276 read-write memory---variables and so on---plus machine status, while
8277 executable files contain only the program text and initialized data.)
8279 When you type @code{run}, your executable file becomes an active process
8280 target as well. When a process target is active, all @value{GDBN}
8281 commands requesting memory addresses refer to that target; addresses in
8282 an active core file or executable file target are obscured while the
8283 process target is active.
8285 Use the @code{core-file} and @code{exec-file} commands to select a new
8286 core file or executable target (@pxref{Files, ,Commands to specify
8287 files}). To specify as a target a process that is already running, use
8288 the @code{attach} command (@pxref{Attach, ,Debugging an already-running
8291 @node Target Commands
8292 @section Commands for managing targets
8295 @item target @var{type} @var{parameters}
8296 Connects the @value{GDBN} host environment to a target machine or
8297 process. A target is typically a protocol for talking to debugging
8298 facilities. You use the argument @var{type} to specify the type or
8299 protocol of the target machine.
8301 Further @var{parameters} are interpreted by the target protocol, but
8302 typically include things like device names or host names to connect
8303 with, process numbers, and baud rates.
8305 The @code{target} command does not repeat if you press @key{RET} again
8306 after executing the command.
8310 Displays the names of all targets available. To display targets
8311 currently selected, use either @code{info target} or @code{info files}
8312 (@pxref{Files, ,Commands to specify files}).
8314 @item help target @var{name}
8315 Describe a particular target, including any parameters necessary to
8318 @kindex set gnutarget
8319 @item set gnutarget @var{args}
8320 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
8321 knows whether it is reading an @dfn{executable},
8322 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
8323 with the @code{set gnutarget} command. Unlike most @code{target} commands,
8324 with @code{gnutarget} the @code{target} refers to a program, not a machine.
8327 @emph{Warning:} To specify a file format with @code{set gnutarget},
8328 you must know the actual BFD name.
8332 @xref{Files, , Commands to specify files}.
8334 @kindex show gnutarget
8335 @item show gnutarget
8336 Use the @code{show gnutarget} command to display what file format
8337 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
8338 @value{GDBN} will determine the file format for each file automatically,
8339 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
8342 Here are some common targets (available, or not, depending on the GDB
8347 @item target exec @var{program}
8348 An executable file. @samp{target exec @var{program}} is the same as
8349 @samp{exec-file @var{program}}.
8352 @item target core @var{filename}
8353 A core dump file. @samp{target core @var{filename}} is the same as
8354 @samp{core-file @var{filename}}.
8356 @kindex target remote
8357 @item target remote @var{dev}
8358 Remote serial target in GDB-specific protocol. The argument @var{dev}
8359 specifies what serial device to use for the connection (e.g.
8360 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}. @code{target remote}
8361 supports the @code{load} command. This is only useful if you have
8362 some other way of getting the stub to the target system, and you can put
8363 it somewhere in memory where it won't get clobbered by the download.
8367 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
8375 works; however, you cannot assume that a specific memory map, device
8376 drivers, or even basic I/O is available, although some simulators do
8377 provide these. For info about any processor-specific simulator details,
8378 see the appropriate section in @ref{Embedded Processors, ,Embedded
8383 Some configurations may include these targets as well:
8388 @item target nrom @var{dev}
8389 NetROM ROM emulator. This target only supports downloading.
8393 Different targets are available on different configurations of @value{GDBN};
8394 your configuration may have more or fewer targets.
8396 Many remote targets require you to download the executable's code
8397 once you've successfully established a connection.
8401 @kindex load @var{filename}
8402 @item load @var{filename}
8403 Depending on what remote debugging facilities are configured into
8404 @value{GDBN}, the @code{load} command may be available. Where it exists, it
8405 is meant to make @var{filename} (an executable) available for debugging
8406 on the remote system---by downloading, or dynamic linking, for example.
8407 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
8408 the @code{add-symbol-file} command.
8410 If your @value{GDBN} does not have a @code{load} command, attempting to
8411 execute it gets the error message ``@code{You can't do that when your
8412 target is @dots{}}''
8414 The file is loaded at whatever address is specified in the executable.
8415 For some object file formats, you can specify the load address when you
8416 link the program; for other formats, like a.out, the object file format
8417 specifies a fixed address.
8418 @c FIXME! This would be a good place for an xref to the GNU linker doc.
8420 @code{load} does not repeat if you press @key{RET} again after using it.
8424 @section Choosing target byte order
8426 @cindex choosing target byte order
8427 @cindex target byte order
8429 Some types of processors, such as the MIPS, PowerPC, and Hitachi SH,
8430 offer the ability to run either big-endian or little-endian byte
8431 orders. Usually the executable or symbol will include a bit to
8432 designate the endian-ness, and you will not need to worry about
8433 which to use. However, you may still find it useful to adjust
8434 @value{GDBN}'s idea of processor endian-ness manually.
8437 @kindex set endian big
8438 @item set endian big
8439 Instruct @value{GDBN} to assume the target is big-endian.
8441 @kindex set endian little
8442 @item set endian little
8443 Instruct @value{GDBN} to assume the target is little-endian.
8445 @kindex set endian auto
8446 @item set endian auto
8447 Instruct @value{GDBN} to use the byte order associated with the
8451 Display @value{GDBN}'s current idea of the target byte order.
8455 Note that these commands merely adjust interpretation of symbolic
8456 data on the host, and that they have absolutely no effect on the
8460 @section Remote debugging
8461 @cindex remote debugging
8463 If you are trying to debug a program running on a machine that cannot run
8464 @value{GDBN} in the usual way, it is often useful to use remote debugging.
8465 For example, you might use remote debugging on an operating system kernel,
8466 or on a small system which does not have a general purpose operating system
8467 powerful enough to run a full-featured debugger.
8469 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
8470 to make this work with particular debugging targets. In addition,
8471 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
8472 but not specific to any particular target system) which you can use if you
8473 write the remote stubs---the code that runs on the remote system to
8474 communicate with @value{GDBN}.
8476 Other remote targets may be available in your
8477 configuration of @value{GDBN}; use @code{help target} to list them.
8480 * Remote Serial:: @value{GDBN} remote serial protocol
8484 @subsection The @value{GDBN} remote serial protocol
8486 @cindex remote serial debugging, overview
8487 To debug a program running on another machine (the debugging
8488 @dfn{target} machine), you must first arrange for all the usual
8489 prerequisites for the program to run by itself. For example, for a C
8494 A startup routine to set up the C runtime environment; these usually
8495 have a name like @file{crt0}. The startup routine may be supplied by
8496 your hardware supplier, or you may have to write your own.
8499 A C subroutine library to support your program's
8500 subroutine calls, notably managing input and output.
8503 A way of getting your program to the other machine---for example, a
8504 download program. These are often supplied by the hardware
8505 manufacturer, but you may have to write your own from hardware
8509 The next step is to arrange for your program to use a serial port to
8510 communicate with the machine where @value{GDBN} is running (the @dfn{host}
8511 machine). In general terms, the scheme looks like this:
8515 @value{GDBN} already understands how to use this protocol; when everything
8516 else is set up, you can simply use the @samp{target remote} command
8517 (@pxref{Targets,,Specifying a Debugging Target}).
8519 @item On the target,
8520 you must link with your program a few special-purpose subroutines that
8521 implement the @value{GDBN} remote serial protocol. The file containing these
8522 subroutines is called a @dfn{debugging stub}.
8524 On certain remote targets, you can use an auxiliary program
8525 @code{gdbserver} instead of linking a stub into your program.
8526 @xref{Server,,Using the @code{gdbserver} program}, for details.
8529 The debugging stub is specific to the architecture of the remote
8530 machine; for example, use @file{sparc-stub.c} to debug programs on
8533 @cindex remote serial stub list
8534 These working remote stubs are distributed with @value{GDBN}:
8539 @cindex @file{i386-stub.c}
8542 For Intel 386 and compatible architectures.
8545 @cindex @file{m68k-stub.c}
8546 @cindex Motorola 680x0
8548 For Motorola 680x0 architectures.
8551 @cindex @file{sh-stub.c}
8554 For Hitachi SH architectures.
8557 @cindex @file{sparc-stub.c}
8559 For @sc{sparc} architectures.
8562 @cindex @file{sparcl-stub.c}
8565 For Fujitsu @sc{sparclite} architectures.
8569 The @file{README} file in the @value{GDBN} distribution may list other
8570 recently added stubs.
8573 * Stub Contents:: What the stub can do for you
8574 * Bootstrapping:: What you must do for the stub
8575 * Debug Session:: Putting it all together
8576 * Protocol:: Definition of the communication protocol
8577 * Server:: Using the `gdbserver' program
8578 * NetWare:: Using the `gdbserve.nlm' program
8582 @subsubsection What the stub can do for you
8584 @cindex remote serial stub
8585 The debugging stub for your architecture supplies these three
8589 @item set_debug_traps
8590 @kindex set_debug_traps
8591 @cindex remote serial stub, initialization
8592 This routine arranges for @code{handle_exception} to run when your
8593 program stops. You must call this subroutine explicitly near the
8594 beginning of your program.
8596 @item handle_exception
8597 @kindex handle_exception
8598 @cindex remote serial stub, main routine
8599 This is the central workhorse, but your program never calls it
8600 explicitly---the setup code arranges for @code{handle_exception} to
8601 run when a trap is triggered.
8603 @code{handle_exception} takes control when your program stops during
8604 execution (for example, on a breakpoint), and mediates communications
8605 with @value{GDBN} on the host machine. This is where the communications
8606 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
8607 representative on the target machine. It begins by sending summary
8608 information on the state of your program, then continues to execute,
8609 retrieving and transmitting any information @value{GDBN} needs, until you
8610 execute a @value{GDBN} command that makes your program resume; at that point,
8611 @code{handle_exception} returns control to your own code on the target
8615 @cindex @code{breakpoint} subroutine, remote
8616 Use this auxiliary subroutine to make your program contain a
8617 breakpoint. Depending on the particular situation, this may be the only
8618 way for @value{GDBN} to get control. For instance, if your target
8619 machine has some sort of interrupt button, you won't need to call this;
8620 pressing the interrupt button transfers control to
8621 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
8622 simply receiving characters on the serial port may also trigger a trap;
8623 again, in that situation, you don't need to call @code{breakpoint} from
8624 your own program---simply running @samp{target remote} from the host
8625 @value{GDBN} session gets control.
8627 Call @code{breakpoint} if none of these is true, or if you simply want
8628 to make certain your program stops at a predetermined point for the
8629 start of your debugging session.
8633 @subsubsection What you must do for the stub
8635 @cindex remote stub, support routines
8636 The debugging stubs that come with @value{GDBN} are set up for a particular
8637 chip architecture, but they have no information about the rest of your
8638 debugging target machine.
8640 First of all you need to tell the stub how to communicate with the
8644 @item int getDebugChar()
8645 @kindex getDebugChar
8646 Write this subroutine to read a single character from the serial port.
8647 It may be identical to @code{getchar} for your target system; a
8648 different name is used to allow you to distinguish the two if you wish.
8650 @item void putDebugChar(int)
8651 @kindex putDebugChar
8652 Write this subroutine to write a single character to the serial port.
8653 It may be identical to @code{putchar} for your target system; a
8654 different name is used to allow you to distinguish the two if you wish.
8657 @cindex control C, and remote debugging
8658 @cindex interrupting remote targets
8659 If you want @value{GDBN} to be able to stop your program while it is
8660 running, you need to use an interrupt-driven serial driver, and arrange
8661 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
8662 character). That is the character which @value{GDBN} uses to tell the
8663 remote system to stop.
8665 Getting the debugging target to return the proper status to @value{GDBN}
8666 probably requires changes to the standard stub; one quick and dirty way
8667 is to just execute a breakpoint instruction (the ``dirty'' part is that
8668 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
8670 Other routines you need to supply are:
8673 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
8674 @kindex exceptionHandler
8675 Write this function to install @var{exception_address} in the exception
8676 handling tables. You need to do this because the stub does not have any
8677 way of knowing what the exception handling tables on your target system
8678 are like (for example, the processor's table might be in @sc{rom},
8679 containing entries which point to a table in @sc{ram}).
8680 @var{exception_number} is the exception number which should be changed;
8681 its meaning is architecture-dependent (for example, different numbers
8682 might represent divide by zero, misaligned access, etc). When this
8683 exception occurs, control should be transferred directly to
8684 @var{exception_address}, and the processor state (stack, registers,
8685 and so on) should be just as it is when a processor exception occurs. So if
8686 you want to use a jump instruction to reach @var{exception_address}, it
8687 should be a simple jump, not a jump to subroutine.
8689 For the 386, @var{exception_address} should be installed as an interrupt
8690 gate so that interrupts are masked while the handler runs. The gate
8691 should be at privilege level 0 (the most privileged level). The
8692 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
8693 help from @code{exceptionHandler}.
8695 @item void flush_i_cache()
8696 @kindex flush_i_cache
8697 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
8698 instruction cache, if any, on your target machine. If there is no
8699 instruction cache, this subroutine may be a no-op.
8701 On target machines that have instruction caches, @value{GDBN} requires this
8702 function to make certain that the state of your program is stable.
8706 You must also make sure this library routine is available:
8709 @item void *memset(void *, int, int)
8711 This is the standard library function @code{memset} that sets an area of
8712 memory to a known value. If you have one of the free versions of
8713 @code{libc.a}, @code{memset} can be found there; otherwise, you must
8714 either obtain it from your hardware manufacturer, or write your own.
8717 If you do not use the GNU C compiler, you may need other standard
8718 library subroutines as well; this varies from one stub to another,
8719 but in general the stubs are likely to use any of the common library
8720 subroutines which @code{@value{GCC}} generates as inline code.
8724 @subsubsection Putting it all together
8726 @cindex remote serial debugging summary
8727 In summary, when your program is ready to debug, you must follow these
8732 Make sure you have defined the supporting low-level routines
8733 (@pxref{Bootstrapping,,What you must do for the stub}):
8735 @code{getDebugChar}, @code{putDebugChar},
8736 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
8740 Insert these lines near the top of your program:
8748 For the 680x0 stub only, you need to provide a variable called
8749 @code{exceptionHook}. Normally you just use:
8752 void (*exceptionHook)() = 0;
8756 but if before calling @code{set_debug_traps}, you set it to point to a
8757 function in your program, that function is called when
8758 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
8759 error). The function indicated by @code{exceptionHook} is called with
8760 one parameter: an @code{int} which is the exception number.
8763 Compile and link together: your program, the @value{GDBN} debugging stub for
8764 your target architecture, and the supporting subroutines.
8767 Make sure you have a serial connection between your target machine and
8768 the @value{GDBN} host, and identify the serial port on the host.
8771 @c The "remote" target now provides a `load' command, so we should
8772 @c document that. FIXME.
8773 Download your program to your target machine (or get it there by
8774 whatever means the manufacturer provides), and start it.
8777 To start remote debugging, run @value{GDBN} on the host machine, and specify
8778 as an executable file the program that is running in the remote machine.
8779 This tells @value{GDBN} how to find your program's symbols and the contents
8783 @cindex serial line, @code{target remote}
8784 Establish communication using the @code{target remote} command.
8785 Its argument specifies how to communicate with the target
8786 machine---either via a devicename attached to a direct serial line, or a
8787 TCP port (usually to a terminal server which in turn has a serial line
8788 to the target). For example, to use a serial line connected to the
8789 device named @file{/dev/ttyb}:
8792 target remote /dev/ttyb
8795 @cindex TCP port, @code{target remote}
8796 To use a TCP connection, use an argument of the form
8797 @code{@var{host}:port}. For example, to connect to port 2828 on a
8798 terminal server named @code{manyfarms}:
8801 target remote manyfarms:2828
8805 Now you can use all the usual commands to examine and change data and to
8806 step and continue the remote program.
8808 To resume the remote program and stop debugging it, use the @code{detach}
8811 @cindex interrupting remote programs
8812 @cindex remote programs, interrupting
8813 Whenever @value{GDBN} is waiting for the remote program, if you type the
8814 interrupt character (often @key{C-C}), @value{GDBN} attempts to stop the
8815 program. This may or may not succeed, depending in part on the hardware
8816 and the serial drivers the remote system uses. If you type the
8817 interrupt character once again, @value{GDBN} displays this prompt:
8820 Interrupted while waiting for the program.
8821 Give up (and stop debugging it)? (y or n)
8824 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
8825 (If you decide you want to try again later, you can use @samp{target
8826 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
8827 goes back to waiting.
8830 @subsubsection Communication protocol
8832 @cindex debugging stub, example
8833 @cindex remote stub, example
8834 @cindex stub example, remote debugging
8835 The stub files provided with @value{GDBN} implement the target side of the
8836 communication protocol, and the @value{GDBN} side is implemented in the
8837 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
8838 these subroutines to communicate, and ignore the details. (If you're
8839 implementing your own stub file, you can still ignore the details: start
8840 with one of the existing stub files. @file{sparc-stub.c} is the best
8841 organized, and therefore the easiest to read.)
8843 However, there may be occasions when you need to know something about
8844 the protocol---for example, if there is only one serial port to your
8845 target machine, you might want your program to do something special if
8846 it recognizes a packet meant for @value{GDBN}.
8848 In the examples below, @samp{<-} and @samp{->} are used to indicate
8849 transmitted and received data respectfully.
8851 @cindex protocol, @value{GDBN} remote serial
8852 @cindex serial protocol, @value{GDBN} remote
8853 @cindex remote serial protocol
8854 All @value{GDBN} commands and responses (other than acknowledgments) are
8855 sent as a @var{packet}. A @var{packet} is introduced with the character
8856 @samp{$}, the actual @var{packet-data}, and the terminating character
8857 @samp{#} followed by a two-digit @var{checksum}:
8860 @code{$}@var{packet-data}@code{#}@var{checksum}
8864 @cindex checksum, for @value{GDBN} remote
8866 The two-digit @var{checksum} is computed as the modulo 256 sum of all
8867 characters between the leading @samp{$} and the trailing @samp{#} (an
8868 eight bit unsigned checksum).
8870 Implementors should note that prior to @value{GDBN} 5.0 the protocol
8871 specification also included an optional two-digit @var{sequence-id}:
8874 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
8877 @cindex sequence-id, for @value{GDBN} remote
8879 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
8880 has never output @var{sequence-id}s. Stubs that handle packets added
8881 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
8883 @cindex acknowledgment, for @value{GDBN} remote
8884 When either the host or the target machine receives a packet, the first
8885 response expected is an acknowledgment: either @samp{+} (to indicate
8886 the package was received correctly) or @samp{-} (to request
8890 <- @code{$}@var{packet-data}@code{#}@var{checksum}
8895 The host (@value{GDBN}) sends @var{command}s, and the target (the
8896 debugging stub incorporated in your program) sends a @var{response}. In
8897 the case of step and continue @var{command}s, the response is only sent
8898 when the operation has completed (the target has again stopped).
8900 @var{packet-data} consists of a sequence of characters with the
8901 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
8904 Fields within the packet should be separated using @samp{,} @samp{;} or
8905 @samp{:}. Except where otherwise noted all numbers are represented in
8906 HEX with leading zeros suppressed.
8908 Implementors should note that prior to @value{GDBN} 5.0, the character
8909 @samp{:} could not appear as the third character in a packet (as it
8910 would potentially conflict with the @var{sequence-id}).
8912 Response @var{data} can be run-length encoded to save space. A @samp{*}
8913 means that the next character is an @sc{ascii} encoding giving a repeat count
8914 which stands for that many repetitions of the character preceding the
8915 @samp{*}. The encoding is @code{n+29}, yielding a printable character
8916 where @code{n >=3} (which is where rle starts to win). The printable
8917 characters @samp{$}, @samp{#}, @samp{+} and @samp{-} or with a numeric
8918 value greater than 126 should not be used.
8920 Some remote systems have used a different run-length encoding mechanism
8921 loosely refered to as the cisco encoding. Following the @samp{*}
8922 character are two hex digits that indicate the size of the packet.
8929 means the same as "0000".
8931 The error response returned for some packets includes a two character
8932 error number. That number is not well defined.
8934 For any @var{command} not supported by the stub, an empty response
8935 (@samp{$#00}) should be returned. That way it is possible to extend the
8936 protocol. A newer @value{GDBN} can tell if a packet is supported based
8939 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
8940 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
8943 Below is a complete list of all currently defined @var{command}s and
8944 their corresponding response @var{data}:
8946 @multitable @columnfractions .30 .30 .40
8954 Use the extended remote protocol. Sticky---only needs to be set once.
8955 The extended remote protocol supports the @samp{R} packet.
8959 Stubs that support the extended remote protocol return @samp{} which,
8960 unfortunately, is identical to the response returned by stubs that do not
8961 support protocol extensions.
8966 Indicate the reason the target halted. The reply is the same as for step
8975 @tab Reserved for future use
8977 @item set program arguments @strong{(reserved)}
8978 @tab @code{A}@var{arglen}@code{,}@var{argnum}@code{,}@var{arg}@code{,...}
8983 Initialized @samp{argv[]} array passed into program. @var{arglen}
8984 specifies the number of bytes in the hex encoded byte stream @var{arg}.
8985 See @file{gdbserver} for more details.
8987 @tab reply @code{OK}
8989 @tab reply @code{E}@var{NN}
8991 @item set baud @strong{(deprecated)}
8992 @tab @code{b}@var{baud}
8994 Change the serial line speed to @var{baud}. JTC: @emph{When does the
8995 transport layer state change? When it's received, or after the ACK is
8996 transmitted. In either case, there are problems if the command or the
8997 acknowledgment packet is dropped.} Stan: @emph{If people really wanted
8998 to add something like this, and get it working for the first time, they
8999 ought to modify ser-unix.c to send some kind of out-of-band message to a
9000 specially-setup stub and have the switch happen "in between" packets, so
9001 that from remote protocol's point of view, nothing actually
9004 @item set breakpoint @strong{(deprecated)}
9005 @tab @code{B}@var{addr},@var{mode}
9007 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
9008 breakpoint at @var{addr}. @emph{This has been replaced by the @samp{Z} and
9012 @tab @code{c}@var{addr}
9014 @var{addr} is address to resume. If @var{addr} is omitted, resume at
9020 @item continue with signal
9021 @tab @code{C}@var{sig}@code{;}@var{addr}
9023 Continue with signal @var{sig} (hex signal number). If
9024 @code{;}@var{addr} is omitted, resume at same address.
9029 @item toggle debug @strong{(deprecated)}
9037 Detach @value{GDBN} from the remote system. Sent to the remote target before
9038 @value{GDBN} disconnects.
9040 @tab reply @emph{no response}
9042 @value{GDBN} does not check for any response after sending this packet.
9046 @tab Reserved for future use
9050 @tab Reserved for future use
9054 @tab Reserved for future use
9058 @tab Reserved for future use
9060 @item read registers
9062 @tab Read general registers.
9064 @tab reply @var{XX...}
9066 Each byte of register data is described by two hex digits. The bytes
9067 with the register are transmitted in target byte order. The size of
9068 each register and their position within the @samp{g} @var{packet} are
9069 determined by the @value{GDBN} internal macros @var{REGISTER_RAW_SIZE} and
9070 @var{REGISTER_NAME} macros. The specification of several standard
9071 @code{g} packets is specified below.
9073 @tab @code{E}@var{NN}
9077 @tab @code{G}@var{XX...}
9079 See @samp{g} for a description of the @var{XX...} data.
9081 @tab reply @code{OK}
9084 @tab reply @code{E}@var{NN}
9089 @tab Reserved for future use
9092 @tab @code{H}@var{c}@var{t...}
9094 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
9095 @samp{G}, et.al.). @var{c} = @samp{c} for thread used in step and
9096 continue; @var{t...} can be -1 for all threads. @var{c} = @samp{g} for
9097 thread used in other operations. If zero, pick a thread, any thread.
9099 @tab reply @code{OK}
9102 @tab reply @code{E}@var{NN}
9106 @c 'H': How restrictive (or permissive) is the thread model. If a
9107 @c thread is selected and stopped, are other threads allowed
9108 @c to continue to execute? As I mentioned above, I think the
9109 @c semantics of each command when a thread is selected must be
9110 @c described. For example:
9112 @c 'g': If the stub supports threads and a specific thread is
9113 @c selected, returns the register block from that thread;
9114 @c otherwise returns current registers.
9116 @c 'G' If the stub supports threads and a specific thread is
9117 @c selected, sets the registers of the register block of
9118 @c that thread; otherwise sets current registers.
9120 @item cycle step @strong{(draft)}
9121 @tab @code{i}@var{addr}@code{,}@var{nnn}
9123 Step the remote target by a single clock cycle. If @code{,}@var{nnn} is
9124 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
9125 step starting at that address.
9127 @item signal then cycle step @strong{(reserved)}
9130 See @samp{i} and @samp{S} for likely syntax and semantics.
9134 @tab Reserved for future use
9138 @tab Reserved for future use
9143 FIXME: @emph{There is no description of how operate when a specific
9144 thread context has been selected (ie. does 'k' kill only that thread?)}.
9148 @tab Reserved for future use
9152 @tab Reserved for future use
9155 @tab @code{m}@var{addr}@code{,}@var{length}
9157 Read @var{length} bytes of memory starting at address @var{addr}.
9158 Neither @value{GDBN} nor the stub assume that sized memory transfers are assumed
9159 using word alligned accesses. FIXME: @emph{A word aligned memory
9160 transfer mechanism is needed.}
9162 @tab reply @var{XX...}
9164 @var{XX...} is mem contents. Can be fewer bytes than requested if able
9165 to read only part of the data. Neither @value{GDBN} nor the stub assume that
9166 sized memory transfers are assumed using word alligned accesses. FIXME:
9167 @emph{A word aligned memory transfer mechanism is needed.}
9169 @tab reply @code{E}@var{NN}
9170 @tab @var{NN} is errno
9173 @tab @code{M}@var{addr},@var{length}@code{:}@var{XX...}
9175 Write @var{length} bytes of memory starting at address @var{addr}.
9176 @var{XX...} is the data.
9178 @tab reply @code{OK}
9181 @tab reply @code{E}@var{NN}
9183 for an error (this includes the case where only part of the data was
9188 @tab Reserved for future use
9192 @tab Reserved for future use
9196 @tab Reserved for future use
9200 @tab Reserved for future use
9202 @item read reg @strong{(reserved)}
9203 @tab @code{p}@var{n...}
9207 @tab return @var{r....}
9208 @tab The hex encoded value of the register in target byte order.
9211 @tab @code{P}@var{n...}@code{=}@var{r...}
9213 Write register @var{n...} with value @var{r...}, which contains two hex
9214 digits for each byte in the register (target byte order).
9216 @tab reply @code{OK}
9219 @tab reply @code{E}@var{NN}
9223 @tab @code{q}@var{query}
9225 Request info about @var{query}. In general @value{GDBN} queries
9226 have a leading upper case letter. Custom vendor queries should use a
9227 company prefix (in lower case) ex: @samp{qfsf.var}. @var{query} may
9228 optionally be followed by a @samp{,} or @samp{;} separated list. Stubs
9229 must ensure that they match the full @var{query} name.
9231 @tab reply @code{XX...}
9232 @tab Hex encoded data from query. The reply can not be empty.
9234 @tab reply @code{E}@var{NN}
9238 @tab Indicating an unrecognized @var{query}.
9241 @tab @code{Q}@var{var}@code{=}@var{val}
9243 Set value of @var{var} to @var{val}. See @samp{q} for a discussing of
9246 @item reset @strong{(deprecated)}
9249 Reset the entire system.
9251 @item remote restart
9252 @tab @code{R}@var{XX}
9254 Restart the remote server. @var{XX} while needed has no clear
9255 definition. FIXME: @emph{An example interaction explaining how this
9256 packet is used in extended-remote mode is needed}.
9259 @tab @code{s}@var{addr}
9261 @var{addr} is address to resume. If @var{addr} is omitted, resume at
9267 @item step with signal
9268 @tab @code{S}@var{sig}@code{;}@var{addr}
9270 Like @samp{C} but step not continue.
9276 @tab @code{t}@var{addr}@code{:}@var{PP}@code{,}@var{MM}
9278 Search backwards starting at address @var{addr} for a match with pattern
9279 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4
9280 bytes. @var{addr} must be at least 3 digits.
9283 @tab @code{T}@var{XX}
9284 @tab Find out if the thread XX is alive.
9286 @tab reply @code{OK}
9287 @tab thread is still alive
9289 @tab reply @code{E}@var{NN}
9294 @tab Reserved for future use
9298 @tab Reserved for future use
9302 @tab Reserved for future use
9306 @tab Reserved for future use
9310 @tab Reserved for future use
9314 @tab Reserved for future use
9318 @tab Reserved for future use
9320 @item write mem (binary)
9321 @tab @code{X}@var{addr}@code{,}@var{length}@var{:}@var{XX...}
9323 @var{addr} is address, @var{length} is number of bytes, @var{XX...} is
9324 binary data. The characters @code{$}, @code{#}, and @code{0x7d} are
9325 escaped using @code{0x7d}.
9327 @tab reply @code{OK}
9330 @tab reply @code{E}@var{NN}
9335 @tab Reserved for future use
9339 @tab Reserved for future use
9341 @item remove break or watchpoint @strong{(draft)}
9342 @tab @code{z}@var{t}@code{,}@var{addr}@code{,}@var{length}
9346 @item insert break or watchpoint @strong{(draft)}
9347 @tab @code{Z}@var{t}@code{,}@var{addr}@code{,}@var{length}
9349 @var{t} is type: @samp{0} - software breakpoint, @samp{1} - hardware
9350 breakpoint, @samp{2} - write watchpoint, @samp{3} - read watchpoint,
9351 @samp{4} - access watchpoint; @var{addr} is address; @var{length} is in
9352 bytes. For a software breakpoint, @var{length} specifies the size of
9353 the instruction to be patched. For hardware breakpoints and watchpoints
9354 @var{length} specifies the memory region to be monitored. To avoid
9355 potential problems with duplicate packets, the operations should be
9356 implemented in an idempotent way.
9358 @tab reply @code{E}@var{NN}
9361 @tab reply @code{OK}
9365 @tab If not supported.
9369 @tab Reserved for future use
9373 The @samp{C}, @samp{c}, @samp{S}, @samp{s} and @samp{?} packets can
9374 receive any of the below as a reply. In the case of the @samp{C},
9375 @samp{c}, @samp{S} and @samp{s} packets, that reply is only returned
9376 when the target halts. In the below the exact meaning of @samp{signal
9377 number} is poorly defined. In general one of the UNIX signal numbering
9378 conventions is used.
9380 @multitable @columnfractions .4 .6
9382 @item @code{S}@var{AA}
9383 @tab @var{AA} is the signal number
9385 @item @code{T}@var{AA}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}
9387 @var{AA} = two hex digit signal number; @var{n...} = register number
9388 (hex), @var{r...} = target byte ordered register contents, size defined
9389 by @code{REGISTER_RAW_SIZE}; @var{n...} = @samp{thread}, @var{r...} =
9390 thread process ID, this is a hex integer; @var{n...} = other string not
9391 starting with valid hex digit. @value{GDBN} should ignore this
9392 @var{n...}, @var{r...} pair and go on to the next. This way we can
9393 extend the protocol.
9395 @item @code{W}@var{AA}
9397 The process exited, and @var{AA} is the exit status. This is only
9398 applicable for certains sorts of targets.
9400 @item @code{X}@var{AA}
9402 The process terminated with signal @var{AA}.
9404 @item @code{N}@var{AA}@code{;}@var{t...}@code{;}@var{d...}@code{;}@var{b...} @strong{(obsolete)}
9406 @var{AA} = signal number; @var{t...} = address of symbol "_start";
9407 @var{d...} = base of data section; @var{b...} = base of bss section.
9408 @emph{Note: only used by Cisco Systems targets. The difference between
9409 this reply and the "qOffsets" query is that the 'N' packet may arrive
9410 spontaneously whereas the 'qOffsets' is a query initiated by the host
9413 @item @code{O}@var{XX...}
9415 @var{XX...} is hex encoding of @sc{ascii} data. This can happen at any time
9416 while the program is running and the debugger should continue to wait
9421 The following set and query packets have already been defined.
9423 @multitable @columnfractions .2 .2 .6
9425 @item current thread
9426 @tab @code{q}@code{C}
9427 @tab Return the current thread id.
9429 @tab reply @code{QC}@var{pid}
9431 Where @var{pid} is a HEX encoded 16 bit process id.
9434 @tab Any other reply implies the old pid.
9436 @item all thread ids
9437 @tab @code{q}@code{fThreadInfo}
9439 @tab @code{q}@code{sThreadInfo}
9441 Obtain a list of active thread ids from the target (OS). Since there
9442 may be too many active threads to fit into one reply packet, this query
9443 works iteratively: it may require more than one query/reply sequence to
9444 obtain the entire list of threads. The first query of the sequence will
9445 be the @code{qf}@code{ThreadInfo} query; subsequent queries in the
9446 sequence will be the @code{qs}@code{ThreadInfo} query.
9449 @tab NOTE: replaces the @code{qL} query (see below).
9451 @tab reply @code{m}@var{<id>}
9452 @tab A single thread id
9454 @tab reply @code{m}@var{<id>},@var{<id>...}
9455 @tab a comma-separated list of thread ids
9458 @tab (lower case 'el') denotes end of list.
9462 In response to each query, the target will reply with a list of one
9463 or more thread ids, in big-endian hex, separated by commas. GDB will
9464 respond to each reply with a request for more thread ids (using the
9465 @code{qs} form of the query), until the target responds with @code{l}
9466 (lower-case el, for @code{'last'}).
9468 @item extra thread info
9469 @tab @code{q}@code{ThreadExtraInfo}@code{,}@var{id}
9474 Where @var{<id>} is a thread-id in big-endian hex.
9475 Obtain a printable string description of a thread's attributes from
9476 the target OS. This string may contain anything that the target OS
9477 thinks is interesting for @value{GDBN} to tell the user about the thread.
9478 The string is displayed in @value{GDBN}'s @samp{info threads} display.
9479 Some examples of possible thread extra info strings are "Runnable", or
9482 @tab reply @var{XX...}
9484 Where @var{XX...} is a hex encoding of @sc{ascii} data, comprising the
9485 printable string containing the extra information about the thread's
9488 @item query @var{LIST} or @var{threadLIST} @strong{(deprecated)}
9489 @tab @code{q}@code{L}@var{startflag}@var{threadcount}@var{nextthread}
9494 Obtain thread information from RTOS. Where: @var{startflag} (one hex
9495 digit) is one to indicate the first query and zero to indicate a
9496 subsequent query; @var{threadcount} (two hex digits) is the maximum
9497 number of threads the response packet can contain; and @var{nextthread}
9498 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
9499 returned in the response as @var{argthread}.
9502 @tab NOTE: this query is replaced by the @code{q}@code{fThreadInfo}
9505 @tab reply @code{q}@code{M}@var{count}@var{done}@var{argthread}@var{thread...}
9510 Where: @var{count} (two hex digits) is the number of threads being
9511 returned; @var{done} (one hex digit) is zero to indicate more threads
9512 and one indicates no further threads; @var{argthreadid} (eight hex
9513 digits) is @var{nextthread} from the request packet; @var{thread...} is
9514 a sequence of thread IDs from the target. @var{threadid} (eight hex
9515 digits). See @code{remote.c:parse_threadlist_response()}.
9517 @item compute CRC of memory block
9518 @tab @code{q}@code{CRC:}@var{addr}@code{,}@var{length}
9521 @tab reply @code{E}@var{NN}
9522 @tab An error (such as memory fault)
9524 @tab reply @code{C}@var{CRC32}
9525 @tab A 32 bit cyclic redundancy check of the specified memory region.
9527 @item query sect offs
9528 @tab @code{q}@code{Offsets}
9530 Get section offsets that the target used when re-locating the downloaded
9531 image. @emph{Note: while a @code{Bss} offset is included in the
9532 response, @value{GDBN} ignores this and instead applies the @code{Data}
9533 offset to the @code{Bss} section.}
9535 @tab reply @code{Text=}@var{xxx}@code{;Data=}@var{yyy}@code{;Bss=}@var{zzz}
9537 @item thread info request
9538 @tab @code{q}@code{P}@var{mode}@var{threadid}
9543 Returns information on @var{threadid}. Where: @var{mode} is a hex
9544 encoded 32 bit mode; @var{threadid} is a hex encoded 64 bit thread ID.
9548 See @code{remote.c:remote_unpack_thread_info_response()}.
9550 @item remote command
9551 @tab @code{q}@code{Rcmd,}@var{COMMAND}
9556 @var{COMMAND} (hex encoded) is passed to the local interpreter for
9557 execution. Invalid commands should be reported using the output string.
9558 Before the final result packet, the target may also respond with a
9559 number of intermediate @code{O}@var{OUTPUT} console output
9560 packets. @emph{Implementors should note that providing access to a
9561 stubs's interpreter may have security implications}.
9563 @tab reply @code{OK}
9565 A command response with no output.
9567 @tab reply @var{OUTPUT}
9569 A command response with the hex encoded output string @var{OUTPUT}.
9571 @tab reply @code{E}@var{NN}
9573 Indicate a badly formed request.
9578 When @samp{q}@samp{Rcmd} is not recognized.
9582 The following @samp{g}/@samp{G} packets have previously been defined.
9583 In the below, some thirty-two bit registers are transferred as sixty-four
9584 bits. Those registers should be zero/sign extended (which?) to fill the
9585 space allocated. Register bytes are transfered in target byte order.
9586 The two nibbles within a register byte are transfered most-significant -
9589 @multitable @columnfractions .5 .5
9593 All registers are transfered as thirty-two bit quantities in the order:
9594 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
9595 registers; fsr; fir; fp.
9599 All registers are transfered as sixty-four bit quantities (including
9600 thirty-two bit registers such as @code{sr}). The ordering is the same
9605 Example sequence of a target being re-started. Notice how the restart
9606 does not get any direct output:
9611 @emph{target restarts}
9614 -> @code{T001:1234123412341234}
9618 Example sequence of a target being stepped by a single instruction:
9626 -> @code{T001:1234123412341234}
9635 @subsubsection Using the @code{gdbserver} program
9638 @cindex remote connection without stubs
9639 @code{gdbserver} is a control program for Unix-like systems, which
9640 allows you to connect your program with a remote @value{GDBN} via
9641 @code{target remote}---but without linking in the usual debugging stub.
9643 @code{gdbserver} is not a complete replacement for the debugging stubs,
9644 because it requires essentially the same operating-system facilities
9645 that @value{GDBN} itself does. In fact, a system that can run
9646 @code{gdbserver} to connect to a remote @value{GDBN} could also run
9647 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
9648 because it is a much smaller program than @value{GDBN} itself. It is
9649 also easier to port than all of @value{GDBN}, so you may be able to get
9650 started more quickly on a new system by using @code{gdbserver}.
9651 Finally, if you develop code for real-time systems, you may find that
9652 the tradeoffs involved in real-time operation make it more convenient to
9653 do as much development work as possible on another system, for example
9654 by cross-compiling. You can use @code{gdbserver} to make a similar
9655 choice for debugging.
9657 @value{GDBN} and @code{gdbserver} communicate via either a serial line
9658 or a TCP connection, using the standard @value{GDBN} remote serial
9662 @item On the target machine,
9663 you need to have a copy of the program you want to debug.
9664 @code{gdbserver} does not need your program's symbol table, so you can
9665 strip the program if necessary to save space. @value{GDBN} on the host
9666 system does all the symbol handling.
9668 To use the server, you must tell it how to communicate with @value{GDBN};
9669 the name of your program; and the arguments for your program. The
9673 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
9676 @var{comm} is either a device name (to use a serial line) or a TCP
9677 hostname and portnumber. For example, to debug Emacs with the argument
9678 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
9682 target> gdbserver /dev/com1 emacs foo.txt
9685 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
9688 To use a TCP connection instead of a serial line:
9691 target> gdbserver host:2345 emacs foo.txt
9694 The only difference from the previous example is the first argument,
9695 specifying that you are communicating with the host @value{GDBN} via
9696 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
9697 expect a TCP connection from machine @samp{host} to local TCP port 2345.
9698 (Currently, the @samp{host} part is ignored.) You can choose any number
9699 you want for the port number as long as it does not conflict with any
9700 TCP ports already in use on the target system (for example, @code{23} is
9701 reserved for @code{telnet}).@footnote{If you choose a port number that
9702 conflicts with another service, @code{gdbserver} prints an error message
9703 and exits.} You must use the same port number with the host @value{GDBN}
9704 @code{target remote} command.
9706 @item On the @value{GDBN} host machine,
9707 you need an unstripped copy of your program, since @value{GDBN} needs
9708 symbols and debugging information. Start up @value{GDBN} as usual,
9709 using the name of the local copy of your program as the first argument.
9710 (You may also need the @w{@samp{--baud}} option if the serial line is
9711 running at anything other than 9600@dmn{bps}.) After that, use @code{target
9712 remote} to establish communications with @code{gdbserver}. Its argument
9713 is either a device name (usually a serial device, like
9714 @file{/dev/ttyb}), or a TCP port descriptor in the form
9715 @code{@var{host}:@var{PORT}}. For example:
9718 (@value{GDBP}) target remote /dev/ttyb
9722 communicates with the server via serial line @file{/dev/ttyb}, and
9725 (@value{GDBP}) target remote the-target:2345
9729 communicates via a TCP connection to port 2345 on host @w{@file{the-target}}.
9730 For TCP connections, you must start up @code{gdbserver} prior to using
9731 the @code{target remote} command. Otherwise you may get an error whose
9732 text depends on the host system, but which usually looks something like
9733 @samp{Connection refused}.
9737 @subsubsection Using the @code{gdbserve.nlm} program
9739 @kindex gdbserve.nlm
9740 @code{gdbserve.nlm} is a control program for NetWare systems, which
9741 allows you to connect your program with a remote @value{GDBN} via
9742 @code{target remote}.
9744 @value{GDBN} and @code{gdbserve.nlm} communicate via a serial line,
9745 using the standard @value{GDBN} remote serial protocol.
9748 @item On the target machine,
9749 you need to have a copy of the program you want to debug.
9750 @code{gdbserve.nlm} does not need your program's symbol table, so you
9751 can strip the program if necessary to save space. @value{GDBN} on the
9752 host system does all the symbol handling.
9754 To use the server, you must tell it how to communicate with
9755 @value{GDBN}; the name of your program; and the arguments for your
9756 program. The syntax is:
9759 load gdbserve [ BOARD=@var{board} ] [ PORT=@var{port} ]
9760 [ BAUD=@var{baud} ] @var{program} [ @var{args} @dots{} ]
9763 @var{board} and @var{port} specify the serial line; @var{baud} specifies
9764 the baud rate used by the connection. @var{port} and @var{node} default
9765 to 0, @var{baud} defaults to 9600@dmn{bps}.
9767 For example, to debug Emacs with the argument @samp{foo.txt}and
9768 communicate with @value{GDBN} over serial port number 2 or board 1
9769 using a 19200@dmn{bps} connection:
9772 load gdbserve BOARD=1 PORT=2 BAUD=19200 emacs foo.txt
9775 @item On the @value{GDBN} host machine,
9776 you need an unstripped copy of your program, since @value{GDBN} needs
9777 symbols and debugging information. Start up @value{GDBN} as usual,
9778 using the name of the local copy of your program as the first argument.
9779 (You may also need the @w{@samp{--baud}} option if the serial line is
9780 running at anything other than 9600@dmn{bps}. After that, use @code{target
9781 remote} to establish communications with @code{gdbserve.nlm}. Its
9782 argument is a device name (usually a serial device, like
9783 @file{/dev/ttyb}). For example:
9786 (@value{GDBP}) target remote /dev/ttyb
9790 communications with the server via serial line @file{/dev/ttyb}.
9794 @section Kernel Object Display
9796 @cindex kernel object display
9797 @cindex kernel object
9800 Some targets support kernel object display. Using this facility,
9801 @value{GDBN} communicates specially with the underlying operating system
9802 and can display information about operating system-level objects such as
9803 mutexes and other synchronization objects. Exactly which objects can be
9804 displayed is determined on a per-OS basis.
9806 Use the @code{set os} command to set the operating system. This tells
9807 @value{GDBN} which kernel object display module to initialize:
9810 (@value{GDBP}) set os cisco
9813 If @code{set os} succeeds, @value{GDBN} will display some information
9814 about the operating system, and will create a new @code{info} command
9815 which can be used to query the target. The @code{info} command is named
9816 after the operating system:
9819 (@value{GDBP}) info cisco
9820 List of Cisco Kernel Objects
9822 any Any and all objects
9825 Further subcommands can be used to query about particular objects known
9828 There is currently no way to determine whether a given operating system
9829 is supported other than to try it.
9832 @node Configurations
9833 @chapter Configuration-Specific Information
9835 While nearly all @value{GDBN} commands are available for all native and
9836 cross versions of the debugger, there are some exceptions. This chapter
9837 describes things that are only available in certain configurations.
9839 There are three major categories of configurations: native
9840 configurations, where the host and target are the same, embedded
9841 operating system configurations, which are usually the same for several
9842 different processor architectures, and bare embedded processors, which
9843 are quite different from each other.
9848 * Embedded Processors::
9855 This section describes details specific to particular native
9860 * SVR4 Process Information:: SVR4 process information
9866 On HP-UX systems, if you refer to a function or variable name that
9867 begins with a dollar sign, @value{GDBN} searches for a user or system
9868 name first, before it searches for a convenience variable.
9870 @node SVR4 Process Information
9871 @subsection SVR4 process information
9874 @cindex process image
9876 Many versions of SVR4 provide a facility called @samp{/proc} that can be
9877 used to examine the image of a running process using file-system
9878 subroutines. If @value{GDBN} is configured for an operating system with
9879 this facility, the command @code{info proc} is available to report on
9880 several kinds of information about the process running your program.
9881 @code{info proc} works only on SVR4 systems that include the
9882 @code{procfs} code. This includes OSF/1 (Digital Unix), Solaris, Irix,
9883 and Unixware, but not HP-UX or Linux, for example.
9888 Summarize available information about the process.
9890 @kindex info proc mappings
9891 @item info proc mappings
9892 Report on the address ranges accessible in the program, with information
9893 on whether your program may read, write, or execute each range.
9895 @kindex info proc times
9896 @item info proc times
9897 Starting time, user CPU time, and system CPU time for your program and
9900 @kindex info proc id
9902 Report on the process IDs related to your program: its own process ID,
9903 the ID of its parent, the process group ID, and the session ID.
9905 @kindex info proc status
9906 @item info proc status
9907 General information on the state of the process. If the process is
9908 stopped, this report includes the reason for stopping, and any signal
9912 Show all the above information about the process.
9916 @section Embedded Operating Systems
9918 This section describes configurations involving the debugging of
9919 embedded operating systems that are available for several different
9923 * VxWorks:: Using @value{GDBN} with VxWorks
9926 @value{GDBN} includes the ability to debug programs running on
9927 various real-time operating systems.
9930 @subsection Using @value{GDBN} with VxWorks
9936 @kindex target vxworks
9937 @item target vxworks @var{machinename}
9938 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
9939 is the target system's machine name or IP address.
9943 On VxWorks, @code{load} links @var{filename} dynamically on the
9944 current target system as well as adding its symbols in @value{GDBN}.
9946 @value{GDBN} enables developers to spawn and debug tasks running on networked
9947 VxWorks targets from a Unix host. Already-running tasks spawned from
9948 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
9949 both the Unix host and on the VxWorks target. The program
9950 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
9951 installed with the name @code{vxgdb}, to distinguish it from a
9952 @value{GDBN} for debugging programs on the host itself.)
9955 @item VxWorks-timeout @var{args}
9956 @kindex vxworks-timeout
9957 All VxWorks-based targets now support the option @code{vxworks-timeout}.
9958 This option is set by the user, and @var{args} represents the number of
9959 seconds @value{GDBN} waits for responses to rpc's. You might use this if
9960 your VxWorks target is a slow software simulator or is on the far side
9961 of a thin network line.
9964 The following information on connecting to VxWorks was current when
9965 this manual was produced; newer releases of VxWorks may use revised
9969 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
9970 to include the remote debugging interface routines in the VxWorks
9971 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
9972 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
9973 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
9974 source debugging task @code{tRdbTask} when VxWorks is booted. For more
9975 information on configuring and remaking VxWorks, see the manufacturer's
9977 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
9979 Once you have included @file{rdb.a} in your VxWorks system image and set
9980 your Unix execution search path to find @value{GDBN}, you are ready to
9981 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
9982 @code{vxgdb}, depending on your installation).
9984 @value{GDBN} comes up showing the prompt:
9991 * VxWorks Connection:: Connecting to VxWorks
9992 * VxWorks Download:: VxWorks download
9993 * VxWorks Attach:: Running tasks
9996 @node VxWorks Connection
9997 @subsubsection Connecting to VxWorks
9999 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
10000 network. To connect to a target whose host name is ``@code{tt}'', type:
10003 (vxgdb) target vxworks tt
10007 @value{GDBN} displays messages like these:
10010 Attaching remote machine across net...
10015 @value{GDBN} then attempts to read the symbol tables of any object modules
10016 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
10017 these files by searching the directories listed in the command search
10018 path (@pxref{Environment, ,Your program's environment}); if it fails
10019 to find an object file, it displays a message such as:
10022 prog.o: No such file or directory.
10025 When this happens, add the appropriate directory to the search path with
10026 the @value{GDBN} command @code{path}, and execute the @code{target}
10029 @node VxWorks Download
10030 @subsubsection VxWorks download
10032 @cindex download to VxWorks
10033 If you have connected to the VxWorks target and you want to debug an
10034 object that has not yet been loaded, you can use the @value{GDBN}
10035 @code{load} command to download a file from Unix to VxWorks
10036 incrementally. The object file given as an argument to the @code{load}
10037 command is actually opened twice: first by the VxWorks target in order
10038 to download the code, then by @value{GDBN} in order to read the symbol
10039 table. This can lead to problems if the current working directories on
10040 the two systems differ. If both systems have NFS mounted the same
10041 filesystems, you can avoid these problems by using absolute paths.
10042 Otherwise, it is simplest to set the working directory on both systems
10043 to the directory in which the object file resides, and then to reference
10044 the file by its name, without any path. For instance, a program
10045 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
10046 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
10047 program, type this on VxWorks:
10050 -> cd "@var{vxpath}/vw/demo/rdb"
10054 Then, in @value{GDBN}, type:
10057 (vxgdb) cd @var{hostpath}/vw/demo/rdb
10058 (vxgdb) load prog.o
10061 @value{GDBN} displays a response similar to this:
10064 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
10067 You can also use the @code{load} command to reload an object module
10068 after editing and recompiling the corresponding source file. Note that
10069 this makes @value{GDBN} delete all currently-defined breakpoints,
10070 auto-displays, and convenience variables, and to clear the value
10071 history. (This is necessary in order to preserve the integrity of
10072 debugger's data structures that reference the target system's symbol
10075 @node VxWorks Attach
10076 @subsubsection Running tasks
10078 @cindex running VxWorks tasks
10079 You can also attach to an existing task using the @code{attach} command as
10083 (vxgdb) attach @var{task}
10087 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
10088 or suspended when you attach to it. Running tasks are suspended at
10089 the time of attachment.
10091 @node Embedded Processors
10092 @section Embedded Processors
10094 This section goes into details specific to particular embedded
10098 * A29K Embedded:: AMD A29K Embedded
10100 * H8/300:: Hitachi H8/300
10101 * H8/500:: Hitachi H8/500
10102 * i960:: Intel i960
10103 * M32R/D:: Mitsubishi M32R/D
10104 * M68K:: Motorola M68K
10105 * M88K:: Motorola M88K
10106 * MIPS Embedded:: MIPS Embedded
10107 * PA:: HP PA Embedded
10110 * Sparclet:: Tsqware Sparclet
10111 * Sparclite:: Fujitsu Sparclite
10112 * ST2000:: Tandem ST2000
10113 * Z8000:: Zilog Z8000
10116 @node A29K Embedded
10117 @subsection AMD A29K Embedded
10122 * Comms (EB29K):: Communications setup
10123 * gdb-EB29K:: EB29K cross-debugging
10124 * Remote Log:: Remote log
10129 @kindex target adapt
10130 @item target adapt @var{dev}
10131 Adapt monitor for A29K.
10133 @kindex target amd-eb
10134 @item target amd-eb @var{dev} @var{speed} @var{PROG}
10136 Remote PC-resident AMD EB29K board, attached over serial lines.
10137 @var{dev} is the serial device, as for @code{target remote};
10138 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
10139 name of the program to be debugged, as it appears to DOS on the PC.
10140 @xref{A29K EB29K, ,EBMON protocol for AMD29K}.
10145 @subsubsection A29K UDI
10148 @cindex AMD29K via UDI
10150 @value{GDBN} supports AMD's UDI (``Universal Debugger Interface'')
10151 protocol for debugging the a29k processor family. To use this
10152 configuration with AMD targets running the MiniMON monitor, you need the
10153 program @code{MONTIP}, available from AMD at no charge. You can also
10154 use @value{GDBN} with the UDI-conformant a29k simulator program
10155 @code{ISSTIP}, also available from AMD.
10158 @item target udi @var{keyword}
10160 Select the UDI interface to a remote a29k board or simulator, where
10161 @var{keyword} is an entry in the AMD configuration file @file{udi_soc}.
10162 This file contains keyword entries which specify parameters used to
10163 connect to a29k targets. If the @file{udi_soc} file is not in your
10164 working directory, you must set the environment variable @samp{UDICONF}
10169 @subsubsection EBMON protocol for AMD29K
10171 @cindex EB29K board
10172 @cindex running 29K programs
10174 AMD distributes a 29K development board meant to fit in a PC, together
10175 with a DOS-hosted monitor program called @code{EBMON}. As a shorthand
10176 term, this development system is called the ``EB29K''. To use
10177 @value{GDBN} from a Unix system to run programs on the EB29K board, you
10178 must first connect a serial cable between the PC (which hosts the EB29K
10179 board) and a serial port on the Unix system. In the following, we
10180 assume you've hooked the cable between the PC's @file{COM1} port and
10181 @file{/dev/ttya} on the Unix system.
10183 @node Comms (EB29K)
10184 @subsubsection Communications setup
10186 The next step is to set up the PC's port, by doing something like this
10190 C:\> MODE com1:9600,n,8,1,none
10194 This example---run on an MS DOS 4.0 system---sets the PC port to 9600
10195 bps, no parity, eight data bits, one stop bit, and no ``retry'' action;
10196 you must match the communications parameters when establishing the Unix
10197 end of the connection as well.
10198 @c FIXME: Who knows what this "no retry action" crud from the DOS manual may
10199 @c mean? It's optional; leave it out? ---doc@cygnus.com, 25feb91
10201 @c It's optional, but it's unwise to omit it: who knows what is the
10202 @c default value set when the DOS machines boots? "No retry" means that
10203 @c the DOS serial device driver won't retry the operation if it fails;
10204 @c I understand that this is needed because the GDB serial protocol
10205 @c handles any errors and retransmissions itself. ---Eli Zaretskii, 3sep99
10207 To give control of the PC to the Unix side of the serial line, type
10208 the following at the DOS console:
10215 (Later, if you wish to return control to the DOS console, you can use
10216 the command @code{CTTY con}---but you must send it over the device that
10217 had control, in our example over the @file{COM1} serial line.)
10219 From the Unix host, use a communications program such as @code{tip} or
10220 @code{cu} to communicate with the PC; for example,
10223 cu -s 9600 -l /dev/ttya
10227 The @code{cu} options shown specify, respectively, the linespeed and the
10228 serial port to use. If you use @code{tip} instead, your command line
10229 may look something like the following:
10232 tip -9600 /dev/ttya
10236 Your system may require a different name where we show
10237 @file{/dev/ttya} as the argument to @code{tip}. The communications
10238 parameters, including which port to use, are associated with the
10239 @code{tip} argument in the ``remote'' descriptions file---normally the
10240 system table @file{/etc/remote}.
10241 @c FIXME: What if anything needs doing to match the "n,8,1,none" part of
10242 @c the DOS side's comms setup? cu can support -o (odd
10243 @c parity), -e (even parity)---apparently no settings for no parity or
10244 @c for character size. Taken from stty maybe...? John points out tip
10245 @c can set these as internal variables, eg ~s parity=none; man stty
10246 @c suggests that it *might* work to stty these options with stdin or
10247 @c stdout redirected... ---doc@cygnus.com, 25feb91
10249 @c There's nothing to be done for the "none" part of the DOS MODE
10250 @c command. The rest of the parameters should be matched by the
10251 @c baudrate, bits, and parity used by the Unix side. ---Eli Zaretskii, 3Sep99
10254 Using the @code{tip} or @code{cu} connection, change the DOS working
10255 directory to the directory containing a copy of your 29K program, then
10256 start the PC program @code{EBMON} (an EB29K control program supplied
10257 with your board by AMD). You should see an initial display from
10258 @code{EBMON} similar to the one that follows, ending with the
10259 @code{EBMON} prompt @samp{#}---
10264 G:\> CD \usr\joe\work29k
10266 G:\USR\JOE\WORK29K> EBMON
10267 Am29000 PC Coprocessor Board Monitor, version 3.0-18
10268 Copyright 1990 Advanced Micro Devices, Inc.
10269 Written by Gibbons and Associates, Inc.
10271 Enter '?' or 'H' for help
10273 PC Coprocessor Type = EB29K
10275 Memory Base = 0xd0000
10277 Data Memory Size = 2048KB
10278 Available I-RAM Range = 0x8000 to 0x1fffff
10279 Available D-RAM Range = 0x80002000 to 0x801fffff
10282 Register Stack Size = 0x800
10283 Memory Stack Size = 0x1800
10286 Am29027 Available = No
10287 Byte Write Available = Yes
10292 Then exit the @code{cu} or @code{tip} program (done in the example by
10293 typing @code{~.} at the @code{EBMON} prompt). @code{EBMON} keeps
10294 running, ready for @value{GDBN} to take over.
10296 For this example, we've assumed what is probably the most convenient
10297 way to make sure the same 29K program is on both the PC and the Unix
10298 system: a PC/NFS connection that establishes ``drive @file{G:}'' on the
10299 PC as a file system on the Unix host. If you do not have PC/NFS or
10300 something similar connecting the two systems, you must arrange some
10301 other way---perhaps floppy-disk transfer---of getting the 29K program
10302 from the Unix system to the PC; @value{GDBN} does @emph{not} download it over the
10306 @subsubsection EB29K cross-debugging
10308 Finally, @code{cd} to the directory containing an image of your 29K
10309 program on the Unix system, and start @value{GDBN}---specifying as argument the
10310 name of your 29K program:
10313 cd /usr/joe/work29k
10318 Now you can use the @code{target} command:
10321 target amd-eb /dev/ttya 9600 MYFOO
10322 @c FIXME: test above 'target amd-eb' as spelled, with caps! caps are meant to
10323 @c emphasize that this is the name as seen by DOS (since I think DOS is
10324 @c single-minded about case of letters). ---doc@cygnus.com, 25feb91
10328 In this example, we've assumed your program is in a file called
10329 @file{myfoo}. Note that the filename given as the last argument to
10330 @code{target amd-eb} should be the name of the program as it appears to DOS.
10331 In our example this is simply @code{MYFOO}, but in general it can include
10332 a DOS path, and depending on your transfer mechanism may not resemble
10333 the name on the Unix side.
10335 At this point, you can set any breakpoints you wish; when you are ready
10336 to see your program run on the 29K board, use the @value{GDBN} command
10339 To stop debugging the remote program, use the @value{GDBN} @code{detach}
10342 To return control of the PC to its console, use @code{tip} or @code{cu}
10343 once again, after your @value{GDBN} session has concluded, to attach to
10344 @code{EBMON}. You can then type the command @code{q} to shut down
10345 @code{EBMON}, returning control to the DOS command-line interpreter.
10346 Type @kbd{CTTY con} to return command input to the main DOS console,
10347 and type @kbd{~.} to leave @code{tip} or @code{cu}.
10350 @subsubsection Remote log
10351 @cindex @file{eb.log}, a log file for EB29K
10352 @cindex log file for EB29K
10354 The @code{target amd-eb} command creates a file @file{eb.log} in the
10355 current working directory, to help debug problems with the connection.
10356 @file{eb.log} records all the output from @code{EBMON}, including echoes
10357 of the commands sent to it. Running @samp{tail -f} on this file in
10358 another window often helps to understand trouble with @code{EBMON}, or
10359 unexpected events on the PC side of the connection.
10367 @item target rdi @var{dev}
10368 ARM Angel monitor, via RDI library interface to ADP protocol. You may
10369 use this target to communicate with both boards running the Angel
10370 monitor, or with the EmbeddedICE JTAG debug device.
10373 @item target rdp @var{dev}
10379 @subsection Hitachi H8/300
10383 @kindex target hms@r{, with H8/300}
10384 @item target hms @var{dev}
10385 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
10386 Use special commands @code{device} and @code{speed} to control the serial
10387 line and the communications speed used.
10389 @kindex target e7000@r{, with H8/300}
10390 @item target e7000 @var{dev}
10391 E7000 emulator for Hitachi H8 and SH.
10393 @kindex target sh3@r{, with H8/300}
10394 @kindex target sh3e@r{, with H8/300}
10395 @item target sh3 @var{dev}
10396 @itemx target sh3e @var{dev}
10397 Hitachi SH-3 and SH-3E target systems.
10401 @cindex download to H8/300 or H8/500
10402 @cindex H8/300 or H8/500 download
10403 @cindex download to Hitachi SH
10404 @cindex Hitachi SH download
10405 When you select remote debugging to a Hitachi SH, H8/300, or H8/500
10406 board, the @code{load} command downloads your program to the Hitachi
10407 board and also opens it as the current executable target for
10408 @value{GDBN} on your host (like the @code{file} command).
10410 @value{GDBN} needs to know these things to talk to your
10411 Hitachi SH, H8/300, or H8/500:
10415 that you want to use @samp{target hms}, the remote debugging interface
10416 for Hitachi microprocessors, or @samp{target e7000}, the in-circuit
10417 emulator for the Hitachi SH and the Hitachi 300H. (@samp{target hms} is
10418 the default when @value{GDBN} is configured specifically for the Hitachi SH,
10419 H8/300, or H8/500.)
10422 what serial device connects your host to your Hitachi board (the first
10423 serial device available on your host is the default).
10426 what speed to use over the serial device.
10430 * Hitachi Boards:: Connecting to Hitachi boards.
10431 * Hitachi ICE:: Using the E7000 In-Circuit Emulator.
10432 * Hitachi Special:: Special @value{GDBN} commands for Hitachi micros.
10435 @node Hitachi Boards
10436 @subsubsection Connecting to Hitachi boards
10438 @c only for Unix hosts
10440 @cindex serial device, Hitachi micros
10441 Use the special @code{@value{GDBN}} command @samp{device @var{port}} if you
10442 need to explicitly set the serial device. The default @var{port} is the
10443 first available port on your host. This is only necessary on Unix
10444 hosts, where it is typically something like @file{/dev/ttya}.
10447 @cindex serial line speed, Hitachi micros
10448 @code{@value{GDBN}} has another special command to set the communications
10449 speed: @samp{speed @var{bps}}. This command also is only used from Unix
10450 hosts; on DOS hosts, set the line speed as usual from outside @value{GDBN} with
10451 the DOS @code{mode} command (for instance,
10452 @w{@kbd{mode com2:9600,n,8,1,p}} for a 9600@dmn{bps} connection).
10454 The @samp{device} and @samp{speed} commands are available only when you
10455 use a Unix host to debug your Hitachi microprocessor programs. If you
10457 @value{GDBN} depends on an auxiliary terminate-and-stay-resident program
10458 called @code{asynctsr} to communicate with the development board
10459 through a PC serial port. You must also use the DOS @code{mode} command
10460 to set up the serial port on the DOS side.
10462 The following sample session illustrates the steps needed to start a
10463 program under @value{GDBN} control on an H8/300. The example uses a
10464 sample H8/300 program called @file{t.x}. The procedure is the same for
10465 the Hitachi SH and the H8/500.
10467 First hook up your development board. In this example, we use a
10468 board attached to serial port @code{COM2}; if you use a different serial
10469 port, substitute its name in the argument of the @code{mode} command.
10470 When you call @code{asynctsr}, the auxiliary comms program used by the
10471 debugger, you give it just the numeric part of the serial port's name;
10472 for example, @samp{asyncstr 2} below runs @code{asyncstr} on
10476 C:\H8300\TEST> asynctsr 2
10477 C:\H8300\TEST> mode com2:9600,n,8,1,p
10479 Resident portion of MODE loaded
10481 COM2: 9600, n, 8, 1, p
10486 @emph{Warning:} We have noticed a bug in PC-NFS that conflicts with
10487 @code{asynctsr}. If you also run PC-NFS on your DOS host, you may need to
10488 disable it, or even boot without it, to use @code{asynctsr} to control
10489 your development board.
10492 @kindex target hms@r{, and serial protocol}
10493 Now that serial communications are set up, and the development board is
10494 connected, you can start up @value{GDBN}. Call @code{@value{GDBP}} with
10495 the name of your program as the argument. @code{@value{GDBN}} prompts
10496 you, as usual, with the prompt @samp{(@value{GDBP})}. Use two special
10497 commands to begin your debugging session: @samp{target hms} to specify
10498 cross-debugging to the Hitachi board, and the @code{load} command to
10499 download your program to the board. @code{load} displays the names of
10500 the program's sections, and a @samp{*} for each 2K of data downloaded.
10501 (If you want to refresh @value{GDBN} data on symbols or on the
10502 executable file without downloading, use the @value{GDBN} commands
10503 @code{file} or @code{symbol-file}. These commands, and @code{load}
10504 itself, are described in @ref{Files,,Commands to specify files}.)
10507 (eg-C:\H8300\TEST) @value{GDBP} t.x
10508 @value{GDBN} is free software and you are welcome to distribute copies
10509 of it under certain conditions; type "show copying" to see
10511 There is absolutely no warranty for @value{GDBN}; type "show warranty"
10513 @value{GDBN} @value{GDBVN}, Copyright 1992 Free Software Foundation, Inc...
10514 (@value{GDBP}) target hms
10515 Connected to remote H8/300 HMS system.
10516 (@value{GDBP}) load t.x
10517 .text : 0x8000 .. 0xabde ***********
10518 .data : 0xabde .. 0xad30 *
10519 .stack : 0xf000 .. 0xf014 *
10522 At this point, you're ready to run or debug your program. From here on,
10523 you can use all the usual @value{GDBN} commands. The @code{break} command
10524 sets breakpoints; the @code{run} command starts your program;
10525 @code{print} or @code{x} display data; the @code{continue} command
10526 resumes execution after stopping at a breakpoint. You can use the
10527 @code{help} command at any time to find out more about @value{GDBN} commands.
10529 Remember, however, that @emph{operating system} facilities aren't
10530 available on your development board; for example, if your program hangs,
10531 you can't send an interrupt---but you can press the @sc{reset} switch!
10533 Use the @sc{reset} button on the development board
10536 to interrupt your program (don't use @kbd{ctl-C} on the DOS host---it has
10537 no way to pass an interrupt signal to the development board); and
10540 to return to the @value{GDBN} command prompt after your program finishes
10541 normally. The communications protocol provides no other way for @value{GDBN}
10542 to detect program completion.
10545 In either case, @value{GDBN} sees the effect of a @sc{reset} on the
10546 development board as a ``normal exit'' of your program.
10549 @subsubsection Using the E7000 in-circuit emulator
10551 @kindex target e7000@r{, with Hitachi ICE}
10552 You can use the E7000 in-circuit emulator to develop code for either the
10553 Hitachi SH or the H8/300H. Use one of these forms of the @samp{target
10554 e7000} command to connect @value{GDBN} to your E7000:
10557 @item target e7000 @var{port} @var{speed}
10558 Use this form if your E7000 is connected to a serial port. The
10559 @var{port} argument identifies what serial port to use (for example,
10560 @samp{com2}). The third argument is the line speed in bits per second
10561 (for example, @samp{9600}).
10563 @item target e7000 @var{hostname}
10564 If your E7000 is installed as a host on a TCP/IP network, you can just
10565 specify its hostname; @value{GDBN} uses @code{telnet} to connect.
10568 @node Hitachi Special
10569 @subsubsection Special @value{GDBN} commands for Hitachi micros
10571 Some @value{GDBN} commands are available only for the H8/300:
10575 @kindex set machine
10576 @kindex show machine
10577 @item set machine h8300
10578 @itemx set machine h8300h
10579 Condition @value{GDBN} for one of the two variants of the H8/300
10580 architecture with @samp{set machine}. You can use @samp{show machine}
10581 to check which variant is currently in effect.
10590 @kindex set memory @var{mod}
10591 @cindex memory models, H8/500
10592 @item set memory @var{mod}
10594 Specify which H8/500 memory model (@var{mod}) you are using with
10595 @samp{set memory}; check which memory model is in effect with @samp{show
10596 memory}. The accepted values for @var{mod} are @code{small},
10597 @code{big}, @code{medium}, and @code{compact}.
10602 @subsection Intel i960
10606 @kindex target mon960
10607 @item target mon960 @var{dev}
10608 MON960 monitor for Intel i960.
10610 @kindex target nindy
10611 @item target nindy @var{devicename}
10612 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
10613 the name of the serial device to use for the connection, e.g.
10620 @dfn{Nindy} is a ROM Monitor program for Intel 960 target systems. When
10621 @value{GDBN} is configured to control a remote Intel 960 using Nindy, you can
10622 tell @value{GDBN} how to connect to the 960 in several ways:
10626 Through command line options specifying serial port, version of the
10627 Nindy protocol, and communications speed;
10630 By responding to a prompt on startup;
10633 By using the @code{target} command at any point during your @value{GDBN}
10634 session. @xref{Target Commands, ,Commands for managing targets}.
10638 @cindex download to Nindy-960
10639 With the Nindy interface to an Intel 960 board, @code{load}
10640 downloads @var{filename} to the 960 as well as adding its symbols in
10644 * Nindy Startup:: Startup with Nindy
10645 * Nindy Options:: Options for Nindy
10646 * Nindy Reset:: Nindy reset command
10649 @node Nindy Startup
10650 @subsubsection Startup with Nindy
10652 If you simply start @code{@value{GDBP}} without using any command-line
10653 options, you are prompted for what serial port to use, @emph{before} you
10654 reach the ordinary @value{GDBN} prompt:
10657 Attach /dev/ttyNN -- specify NN, or "quit" to quit:
10661 Respond to the prompt with whatever suffix (after @samp{/dev/tty})
10662 identifies the serial port you want to use. You can, if you choose,
10663 simply start up with no Nindy connection by responding to the prompt
10664 with an empty line. If you do this and later wish to attach to Nindy,
10665 use @code{target} (@pxref{Target Commands, ,Commands for managing targets}).
10667 @node Nindy Options
10668 @subsubsection Options for Nindy
10670 These are the startup options for beginning your @value{GDBN} session with a
10671 Nindy-960 board attached:
10674 @item -r @var{port}
10675 Specify the serial port name of a serial interface to be used to connect
10676 to the target system. This option is only available when @value{GDBN} is
10677 configured for the Intel 960 target architecture. You may specify
10678 @var{port} as any of: a full pathname (e.g. @samp{-r /dev/ttya}), a
10679 device name in @file{/dev} (e.g. @samp{-r ttya}), or simply the unique
10680 suffix for a specific @code{tty} (e.g. @samp{-r a}).
10683 (An uppercase letter ``O'', not a zero.) Specify that @value{GDBN} should use
10684 the ``old'' Nindy monitor protocol to connect to the target system.
10685 This option is only available when @value{GDBN} is configured for the Intel 960
10686 target architecture.
10689 @emph{Warning:} if you specify @samp{-O}, but are actually trying to
10690 connect to a target system that expects the newer protocol, the connection
10691 fails, appearing to be a speed mismatch. @value{GDBN} repeatedly
10692 attempts to reconnect at several different line speeds. You can abort
10693 this process with an interrupt.
10697 Specify that @value{GDBN} should first send a @code{BREAK} signal to the target
10698 system, in an attempt to reset it, before connecting to a Nindy target.
10701 @emph{Warning:} Many target systems do not have the hardware that this
10702 requires; it only works with a few boards.
10706 The standard @samp{-b} option controls the line speed used on the serial
10711 @subsubsection Nindy reset command
10716 For a Nindy target, this command sends a ``break'' to the remote target
10717 system; this is only useful if the target has been equipped with a
10718 circuit to perform a hard reset (or some other interesting action) when
10719 a break is detected.
10724 @subsection Mitsubishi M32R/D
10728 @kindex target m32r
10729 @item target m32r @var{dev}
10730 Mitsubishi M32R/D ROM monitor.
10737 The Motorola m68k configuration includes ColdFire support, and
10738 target command for the following ROM monitors.
10742 @kindex target abug
10743 @item target abug @var{dev}
10744 ABug ROM monitor for M68K.
10746 @kindex target cpu32bug
10747 @item target cpu32bug @var{dev}
10748 CPU32BUG monitor, running on a CPU32 (M68K) board.
10750 @kindex target dbug
10751 @item target dbug @var{dev}
10752 dBUG ROM monitor for Motorola ColdFire.
10755 @item target est @var{dev}
10756 EST-300 ICE monitor, running on a CPU32 (M68K) board.
10758 @kindex target rom68k
10759 @item target rom68k @var{dev}
10760 ROM 68K monitor, running on an M68K IDP board.
10764 If @value{GDBN} is configured with @code{m68*-ericsson-*}, it will
10765 instead have only a single special target command:
10769 @kindex target es1800
10770 @item target es1800 @var{dev}
10771 ES-1800 emulator for M68K.
10779 @kindex target rombug
10780 @item target rombug @var{dev}
10781 ROMBUG ROM monitor for OS/9000.
10791 @item target bug @var{dev}
10792 BUG monitor, running on a MVME187 (m88k) board.
10796 @node MIPS Embedded
10797 @subsection MIPS Embedded
10799 @cindex MIPS boards
10800 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
10801 MIPS board attached to a serial line. This is available when
10802 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
10805 Use these @value{GDBN} commands to specify the connection to your target board:
10808 @item target mips @var{port}
10809 @kindex target mips @var{port}
10810 To run a program on the board, start up @code{@value{GDBP}} with the
10811 name of your program as the argument. To connect to the board, use the
10812 command @samp{target mips @var{port}}, where @var{port} is the name of
10813 the serial port connected to the board. If the program has not already
10814 been downloaded to the board, you may use the @code{load} command to
10815 download it. You can then use all the usual @value{GDBN} commands.
10817 For example, this sequence connects to the target board through a serial
10818 port, and loads and runs a program called @var{prog} through the
10822 host$ @value{GDBP} @var{prog}
10823 @value{GDBN} is free software and @dots{}
10824 (@value{GDBP}) target mips /dev/ttyb
10825 (@value{GDBP}) load @var{prog}
10829 @item target mips @var{hostname}:@var{portnumber}
10830 On some @value{GDBN} host configurations, you can specify a TCP
10831 connection (for instance, to a serial line managed by a terminal
10832 concentrator) instead of a serial port, using the syntax
10833 @samp{@var{hostname}:@var{portnumber}}.
10835 @item target pmon @var{port}
10836 @kindex target pmon @var{port}
10839 @item target ddb @var{port}
10840 @kindex target ddb @var{port}
10841 NEC's DDB variant of PMON for Vr4300.
10843 @item target lsi @var{port}
10844 @kindex target lsi @var{port}
10845 LSI variant of PMON.
10847 @kindex target r3900
10848 @item target r3900 @var{dev}
10849 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
10851 @kindex target array
10852 @item target array @var{dev}
10853 Array Tech LSI33K RAID controller board.
10859 @value{GDBN} also supports these special commands for MIPS targets:
10862 @item set processor @var{args}
10863 @itemx show processor
10864 @kindex set processor @var{args}
10865 @kindex show processor
10866 Use the @code{set processor} command to set the type of MIPS
10867 processor when you want to access processor-type-specific registers.
10868 For example, @code{set processor @var{r3041}} tells @value{GDBN}
10869 to use the CPO registers appropriate for the 3041 chip.
10870 Use the @code{show processor} command to see what MIPS processor @value{GDBN}
10871 is using. Use the @code{info reg} command to see what registers
10872 @value{GDBN} is using.
10874 @item set mipsfpu double
10875 @itemx set mipsfpu single
10876 @itemx set mipsfpu none
10877 @itemx show mipsfpu
10878 @kindex set mipsfpu
10879 @kindex show mipsfpu
10880 @cindex MIPS remote floating point
10881 @cindex floating point, MIPS remote
10882 If your target board does not support the MIPS floating point
10883 coprocessor, you should use the command @samp{set mipsfpu none} (if you
10884 need this, you may wish to put the command in your @value{GDBN} init
10885 file). This tells @value{GDBN} how to find the return value of
10886 functions which return floating point values. It also allows
10887 @value{GDBN} to avoid saving the floating point registers when calling
10888 functions on the board. If you are using a floating point coprocessor
10889 with only single precision floating point support, as on the @sc{r4650}
10890 processor, use the command @samp{set mipsfpu single}. The default
10891 double precision floating point coprocessor may be selected using
10892 @samp{set mipsfpu double}.
10894 In previous versions the only choices were double precision or no
10895 floating point, so @samp{set mipsfpu on} will select double precision
10896 and @samp{set mipsfpu off} will select no floating point.
10898 As usual, you can inquire about the @code{mipsfpu} variable with
10899 @samp{show mipsfpu}.
10901 @item set remotedebug @var{n}
10902 @itemx show remotedebug
10903 @kindex set remotedebug@r{, MIPS protocol}
10904 @kindex show remotedebug@r{, MIPS protocol}
10905 @cindex @code{remotedebug}, MIPS protocol
10906 @cindex MIPS @code{remotedebug} protocol
10907 @c FIXME! For this to be useful, you must know something about the MIPS
10908 @c FIXME...protocol. Where is it described?
10909 You can see some debugging information about communications with the board
10910 by setting the @code{remotedebug} variable. If you set it to @code{1} using
10911 @samp{set remotedebug 1}, every packet is displayed. If you set it
10912 to @code{2}, every character is displayed. You can check the current value
10913 at any time with the command @samp{show remotedebug}.
10915 @item set timeout @var{seconds}
10916 @itemx set retransmit-timeout @var{seconds}
10917 @itemx show timeout
10918 @itemx show retransmit-timeout
10919 @cindex @code{timeout}, MIPS protocol
10920 @cindex @code{retransmit-timeout}, MIPS protocol
10921 @kindex set timeout
10922 @kindex show timeout
10923 @kindex set retransmit-timeout
10924 @kindex show retransmit-timeout
10925 You can control the timeout used while waiting for a packet, in the MIPS
10926 remote protocol, with the @code{set timeout @var{seconds}} command. The
10927 default is 5 seconds. Similarly, you can control the timeout used while
10928 waiting for an acknowledgement of a packet with the @code{set
10929 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
10930 You can inspect both values with @code{show timeout} and @code{show
10931 retransmit-timeout}. (These commands are @emph{only} available when
10932 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
10934 The timeout set by @code{set timeout} does not apply when @value{GDBN}
10935 is waiting for your program to stop. In that case, @value{GDBN} waits
10936 forever because it has no way of knowing how long the program is going
10937 to run before stopping.
10941 @subsection PowerPC
10945 @kindex target dink32
10946 @item target dink32 @var{dev}
10947 DINK32 ROM monitor.
10949 @kindex target ppcbug
10950 @item target ppcbug @var{dev}
10951 @kindex target ppcbug1
10952 @item target ppcbug1 @var{dev}
10953 PPCBUG ROM monitor for PowerPC.
10956 @item target sds @var{dev}
10957 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
10962 @subsection HP PA Embedded
10966 @kindex target op50n
10967 @item target op50n @var{dev}
10968 OP50N monitor, running on an OKI HPPA board.
10970 @kindex target w89k
10971 @item target w89k @var{dev}
10972 W89K monitor, running on a Winbond HPPA board.
10977 @subsection Hitachi SH
10981 @kindex target hms@r{, with Hitachi SH}
10982 @item target hms @var{dev}
10983 A Hitachi SH board attached via serial line to your host. Use special
10984 commands @code{device} and @code{speed} to control the serial line and
10985 the communications speed used.
10987 @kindex target e7000@r{, with Hitachi SH}
10988 @item target e7000 @var{dev}
10989 E7000 emulator for Hitachi SH.
10991 @kindex target sh3@r{, with SH}
10992 @kindex target sh3e@r{, with SH}
10993 @item target sh3 @var{dev}
10994 @item target sh3e @var{dev}
10995 Hitachi SH-3 and SH-3E target systems.
11000 @subsection Tsqware Sparclet
11004 @value{GDBN} enables developers to debug tasks running on
11005 Sparclet targets from a Unix host.
11006 @value{GDBN} uses code that runs on
11007 both the Unix host and on the Sparclet target. The program
11008 @code{@value{GDBP}} is installed and executed on the Unix host.
11011 @item remotetimeout @var{args}
11012 @kindex remotetimeout
11013 @value{GDBN} supports the option @code{remotetimeout}.
11014 This option is set by the user, and @var{args} represents the number of
11015 seconds @value{GDBN} waits for responses.
11018 @cindex compiling, on Sparclet
11019 When compiling for debugging, include the options @samp{-g} to get debug
11020 information and @samp{-Ttext} to relocate the program to where you wish to
11021 load it on the target. You may also want to add the options @samp{-n} or
11022 @samp{-N} in order to reduce the size of the sections. Example:
11025 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
11028 You can use @code{objdump} to verify that the addresses are what you intended:
11031 sparclet-aout-objdump --headers --syms prog
11034 @cindex running, on Sparclet
11036 your Unix execution search path to find @value{GDBN}, you are ready to
11037 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
11038 (or @code{sparclet-aout-gdb}, depending on your installation).
11040 @value{GDBN} comes up showing the prompt:
11047 * Sparclet File:: Setting the file to debug
11048 * Sparclet Connection:: Connecting to Sparclet
11049 * Sparclet Download:: Sparclet download
11050 * Sparclet Execution:: Running and debugging
11053 @node Sparclet File
11054 @subsubsection Setting file to debug
11056 The @value{GDBN} command @code{file} lets you choose with program to debug.
11059 (gdbslet) file prog
11063 @value{GDBN} then attempts to read the symbol table of @file{prog}.
11064 @value{GDBN} locates
11065 the file by searching the directories listed in the command search
11067 If the file was compiled with debug information (option "-g"), source
11068 files will be searched as well.
11069 @value{GDBN} locates
11070 the source files by searching the directories listed in the directory search
11071 path (@pxref{Environment, ,Your program's environment}).
11073 to find a file, it displays a message such as:
11076 prog: No such file or directory.
11079 When this happens, add the appropriate directories to the search paths with
11080 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
11081 @code{target} command again.
11083 @node Sparclet Connection
11084 @subsubsection Connecting to Sparclet
11086 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
11087 To connect to a target on serial port ``@code{ttya}'', type:
11090 (gdbslet) target sparclet /dev/ttya
11091 Remote target sparclet connected to /dev/ttya
11092 main () at ../prog.c:3
11096 @value{GDBN} displays messages like these:
11102 @node Sparclet Download
11103 @subsubsection Sparclet download
11105 @cindex download to Sparclet
11106 Once connected to the Sparclet target,
11107 you can use the @value{GDBN}
11108 @code{load} command to download the file from the host to the target.
11109 The file name and load offset should be given as arguments to the @code{load}
11111 Since the file format is aout, the program must be loaded to the starting
11112 address. You can use @code{objdump} to find out what this value is. The load
11113 offset is an offset which is added to the VMA (virtual memory address)
11114 of each of the file's sections.
11115 For instance, if the program
11116 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
11117 and bss at 0x12010170, in @value{GDBN}, type:
11120 (gdbslet) load prog 0x12010000
11121 Loading section .text, size 0xdb0 vma 0x12010000
11124 If the code is loaded at a different address then what the program was linked
11125 to, you may need to use the @code{section} and @code{add-symbol-file} commands
11126 to tell @value{GDBN} where to map the symbol table.
11128 @node Sparclet Execution
11129 @subsubsection Running and debugging
11131 @cindex running and debugging Sparclet programs
11132 You can now begin debugging the task using @value{GDBN}'s execution control
11133 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
11134 manual for the list of commands.
11138 Breakpoint 1 at 0x12010000: file prog.c, line 3.
11140 Starting program: prog
11141 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
11142 3 char *symarg = 0;
11144 4 char *execarg = "hello!";
11149 @subsection Fujitsu Sparclite
11153 @kindex target sparclite
11154 @item target sparclite @var{dev}
11155 Fujitsu sparclite boards, used only for the purpose of loading.
11156 You must use an additional command to debug the program.
11157 For example: target remote @var{dev} using @value{GDBN} standard
11163 @subsection Tandem ST2000
11165 @value{GDBN} may be used with a Tandem ST2000 phone switch, running Tandem's
11168 To connect your ST2000 to the host system, see the manufacturer's
11169 manual. Once the ST2000 is physically attached, you can run:
11172 target st2000 @var{dev} @var{speed}
11176 to establish it as your debugging environment. @var{dev} is normally
11177 the name of a serial device, such as @file{/dev/ttya}, connected to the
11178 ST2000 via a serial line. You can instead specify @var{dev} as a TCP
11179 connection (for example, to a serial line attached via a terminal
11180 concentrator) using the syntax @code{@var{hostname}:@var{portnumber}}.
11182 The @code{load} and @code{attach} commands are @emph{not} defined for
11183 this target; you must load your program into the ST2000 as you normally
11184 would for standalone operation. @value{GDBN} reads debugging information
11185 (such as symbols) from a separate, debugging version of the program
11186 available on your host computer.
11187 @c FIXME!! This is terribly vague; what little content is here is
11188 @c basically hearsay.
11190 @cindex ST2000 auxiliary commands
11191 These auxiliary @value{GDBN} commands are available to help you with the ST2000
11195 @item st2000 @var{command}
11196 @kindex st2000 @var{cmd}
11197 @cindex STDBUG commands (ST2000)
11198 @cindex commands to STDBUG (ST2000)
11199 Send a @var{command} to the STDBUG monitor. See the manufacturer's
11200 manual for available commands.
11203 @cindex connect (to STDBUG)
11204 Connect the controlling terminal to the STDBUG command monitor. When
11205 you are done interacting with STDBUG, typing either of two character
11206 sequences gets you back to the @value{GDBN} command prompt:
11207 @kbd{@key{RET}~.} (Return, followed by tilde and period) or
11208 @kbd{@key{RET}~@key{C-d}} (Return, followed by tilde and control-D).
11212 @subsection Zilog Z8000
11215 @cindex simulator, Z8000
11216 @cindex Zilog Z8000 simulator
11218 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
11221 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
11222 unsegmented variant of the Z8000 architecture) or the Z8001 (the
11223 segmented variant). The simulator recognizes which architecture is
11224 appropriate by inspecting the object code.
11227 @item target sim @var{args}
11229 @kindex target sim@r{, with Z8000}
11230 Debug programs on a simulated CPU. If the simulator supports setup
11231 options, specify them via @var{args}.
11235 After specifying this target, you can debug programs for the simulated
11236 CPU in the same style as programs for your host computer; use the
11237 @code{file} command to load a new program image, the @code{run} command
11238 to run your program, and so on.
11240 As well as making available all the usual machine registers
11241 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
11242 additional items of information as specially named registers:
11247 Counts clock-ticks in the simulator.
11250 Counts instructions run in the simulator.
11253 Execution time in 60ths of a second.
11257 You can refer to these values in @value{GDBN} expressions with the usual
11258 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
11259 conditional breakpoint that suspends only after at least 5000
11260 simulated clock ticks.
11262 @node Architectures
11263 @section Architectures
11265 This section describes characteristics of architectures that affect
11266 all uses of @value{GDBN} with the architecture, both native and cross.
11279 @kindex set rstack_high_address
11280 @cindex AMD 29K register stack
11281 @cindex register stack, AMD29K
11282 @item set rstack_high_address @var{address}
11283 On AMD 29000 family processors, registers are saved in a separate
11284 @dfn{register stack}. There is no way for @value{GDBN} to determine the
11285 extent of this stack. Normally, @value{GDBN} just assumes that the
11286 stack is ``large enough''. This may result in @value{GDBN} referencing
11287 memory locations that do not exist. If necessary, you can get around
11288 this problem by specifying the ending address of the register stack with
11289 the @code{set rstack_high_address} command. The argument should be an
11290 address, which you probably want to precede with @samp{0x} to specify in
11293 @kindex show rstack_high_address
11294 @item show rstack_high_address
11295 Display the current limit of the register stack, on AMD 29000 family
11303 See the following section.
11308 @cindex stack on Alpha
11309 @cindex stack on MIPS
11310 @cindex Alpha stack
11312 Alpha- and MIPS-based computers use an unusual stack frame, which
11313 sometimes requires @value{GDBN} to search backward in the object code to
11314 find the beginning of a function.
11316 @cindex response time, MIPS debugging
11317 To improve response time (especially for embedded applications, where
11318 @value{GDBN} may be restricted to a slow serial line for this search)
11319 you may want to limit the size of this search, using one of these
11323 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
11324 @item set heuristic-fence-post @var{limit}
11325 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
11326 search for the beginning of a function. A value of @var{0} (the
11327 default) means there is no limit. However, except for @var{0}, the
11328 larger the limit the more bytes @code{heuristic-fence-post} must search
11329 and therefore the longer it takes to run.
11331 @item show heuristic-fence-post
11332 Display the current limit.
11336 These commands are available @emph{only} when @value{GDBN} is configured
11337 for debugging programs on Alpha or MIPS processors.
11340 @node Controlling GDB
11341 @chapter Controlling @value{GDBN}
11343 You can alter the way @value{GDBN} interacts with you by using the
11344 @code{set} command. For commands controlling how @value{GDBN} displays
11345 data, see @ref{Print Settings, ,Print settings}. Other settings are
11350 * Editing:: Command editing
11351 * History:: Command history
11352 * Screen Size:: Screen size
11353 * Numbers:: Numbers
11354 * Messages/Warnings:: Optional warnings and messages
11355 * Debugging Output:: Optional messages about internal happenings
11363 @value{GDBN} indicates its readiness to read a command by printing a string
11364 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
11365 can change the prompt string with the @code{set prompt} command. For
11366 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
11367 the prompt in one of the @value{GDBN} sessions so that you can always tell
11368 which one you are talking to.
11370 @emph{Note:} @code{set prompt} does not add a space for you after the
11371 prompt you set. This allows you to set a prompt which ends in a space
11372 or a prompt that does not.
11376 @item set prompt @var{newprompt}
11377 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
11379 @kindex show prompt
11381 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
11385 @section Command editing
11387 @cindex command line editing
11389 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
11390 @sc{gnu} library provides consistent behavior for programs which provide a
11391 command line interface to the user. Advantages are @sc{gnu} Emacs-style
11392 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
11393 substitution, and a storage and recall of command history across
11394 debugging sessions.
11396 You may control the behavior of command line editing in @value{GDBN} with the
11397 command @code{set}.
11400 @kindex set editing
11403 @itemx set editing on
11404 Enable command line editing (enabled by default).
11406 @item set editing off
11407 Disable command line editing.
11409 @kindex show editing
11411 Show whether command line editing is enabled.
11415 @section Command history
11417 @value{GDBN} can keep track of the commands you type during your
11418 debugging sessions, so that you can be certain of precisely what
11419 happened. Use these commands to manage the @value{GDBN} command
11423 @cindex history substitution
11424 @cindex history file
11425 @kindex set history filename
11426 @kindex GDBHISTFILE
11427 @item set history filename @var{fname}
11428 Set the name of the @value{GDBN} command history file to @var{fname}.
11429 This is the file where @value{GDBN} reads an initial command history
11430 list, and where it writes the command history from this session when it
11431 exits. You can access this list through history expansion or through
11432 the history command editing characters listed below. This file defaults
11433 to the value of the environment variable @code{GDBHISTFILE}, or to
11434 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
11437 @cindex history save
11438 @kindex set history save
11439 @item set history save
11440 @itemx set history save on
11441 Record command history in a file, whose name may be specified with the
11442 @code{set history filename} command. By default, this option is disabled.
11444 @item set history save off
11445 Stop recording command history in a file.
11447 @cindex history size
11448 @kindex set history size
11449 @item set history size @var{size}
11450 Set the number of commands which @value{GDBN} keeps in its history list.
11451 This defaults to the value of the environment variable
11452 @code{HISTSIZE}, or to 256 if this variable is not set.
11455 @cindex history expansion
11456 History expansion assigns special meaning to the character @kbd{!}.
11457 @ifset have-readline-appendices
11458 @xref{Event Designators}.
11461 Since @kbd{!} is also the logical not operator in C, history expansion
11462 is off by default. If you decide to enable history expansion with the
11463 @code{set history expansion on} command, you may sometimes need to
11464 follow @kbd{!} (when it is used as logical not, in an expression) with
11465 a space or a tab to prevent it from being expanded. The readline
11466 history facilities do not attempt substitution on the strings
11467 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
11469 The commands to control history expansion are:
11472 @kindex set history expansion
11473 @item set history expansion on
11474 @itemx set history expansion
11475 Enable history expansion. History expansion is off by default.
11477 @item set history expansion off
11478 Disable history expansion.
11480 The readline code comes with more complete documentation of
11481 editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs
11482 or @code{vi} may wish to read it.
11483 @ifset have-readline-appendices
11484 @xref{Command Line Editing}.
11488 @kindex show history
11490 @itemx show history filename
11491 @itemx show history save
11492 @itemx show history size
11493 @itemx show history expansion
11494 These commands display the state of the @value{GDBN} history parameters.
11495 @code{show history} by itself displays all four states.
11501 @item show commands
11502 Display the last ten commands in the command history.
11504 @item show commands @var{n}
11505 Print ten commands centered on command number @var{n}.
11507 @item show commands +
11508 Print ten commands just after the commands last printed.
11512 @section Screen size
11513 @cindex size of screen
11514 @cindex pauses in output
11516 Certain commands to @value{GDBN} may produce large amounts of
11517 information output to the screen. To help you read all of it,
11518 @value{GDBN} pauses and asks you for input at the end of each page of
11519 output. Type @key{RET} when you want to continue the output, or @kbd{q}
11520 to discard the remaining output. Also, the screen width setting
11521 determines when to wrap lines of output. Depending on what is being
11522 printed, @value{GDBN} tries to break the line at a readable place,
11523 rather than simply letting it overflow onto the following line.
11525 Normally @value{GDBN} knows the size of the screen from the terminal
11526 driver software. For example, on Unix @value{GDBN} uses the termcap data base
11527 together with the value of the @code{TERM} environment variable and the
11528 @code{stty rows} and @code{stty cols} settings. If this is not correct,
11529 you can override it with the @code{set height} and @code{set
11536 @kindex show height
11537 @item set height @var{lpp}
11539 @itemx set width @var{cpl}
11541 These @code{set} commands specify a screen height of @var{lpp} lines and
11542 a screen width of @var{cpl} characters. The associated @code{show}
11543 commands display the current settings.
11545 If you specify a height of zero lines, @value{GDBN} does not pause during
11546 output no matter how long the output is. This is useful if output is to a
11547 file or to an editor buffer.
11549 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
11550 from wrapping its output.
11555 @cindex number representation
11556 @cindex entering numbers
11558 You can always enter numbers in octal, decimal, or hexadecimal in
11559 @value{GDBN} by the usual conventions: octal numbers begin with
11560 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
11561 begin with @samp{0x}. Numbers that begin with none of these are, by
11562 default, entered in base 10; likewise, the default display for
11563 numbers---when no particular format is specified---is base 10. You can
11564 change the default base for both input and output with the @code{set
11568 @kindex set input-radix
11569 @item set input-radix @var{base}
11570 Set the default base for numeric input. Supported choices
11571 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
11572 specified either unambiguously or using the current default radix; for
11582 sets the base to decimal. On the other hand, @samp{set radix 10}
11583 leaves the radix unchanged no matter what it was.
11585 @kindex set output-radix
11586 @item set output-radix @var{base}
11587 Set the default base for numeric display. Supported choices
11588 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
11589 specified either unambiguously or using the current default radix.
11591 @kindex show input-radix
11592 @item show input-radix
11593 Display the current default base for numeric input.
11595 @kindex show output-radix
11596 @item show output-radix
11597 Display the current default base for numeric display.
11600 @node Messages/Warnings
11601 @section Optional warnings and messages
11603 By default, @value{GDBN} is silent about its inner workings. If you are
11604 running on a slow machine, you may want to use the @code{set verbose}
11605 command. This makes @value{GDBN} tell you when it does a lengthy
11606 internal operation, so you will not think it has crashed.
11608 Currently, the messages controlled by @code{set verbose} are those
11609 which announce that the symbol table for a source file is being read;
11610 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
11613 @kindex set verbose
11614 @item set verbose on
11615 Enables @value{GDBN} output of certain informational messages.
11617 @item set verbose off
11618 Disables @value{GDBN} output of certain informational messages.
11620 @kindex show verbose
11622 Displays whether @code{set verbose} is on or off.
11625 By default, if @value{GDBN} encounters bugs in the symbol table of an
11626 object file, it is silent; but if you are debugging a compiler, you may
11627 find this information useful (@pxref{Symbol Errors, ,Errors reading
11632 @kindex set complaints
11633 @item set complaints @var{limit}
11634 Permits @value{GDBN} to output @var{limit} complaints about each type of
11635 unusual symbols before becoming silent about the problem. Set
11636 @var{limit} to zero to suppress all complaints; set it to a large number
11637 to prevent complaints from being suppressed.
11639 @kindex show complaints
11640 @item show complaints
11641 Displays how many symbol complaints @value{GDBN} is permitted to produce.
11645 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
11646 lot of stupid questions to confirm certain commands. For example, if
11647 you try to run a program which is already running:
11651 The program being debugged has been started already.
11652 Start it from the beginning? (y or n)
11655 If you are willing to unflinchingly face the consequences of your own
11656 commands, you can disable this ``feature'':
11660 @kindex set confirm
11662 @cindex confirmation
11663 @cindex stupid questions
11664 @item set confirm off
11665 Disables confirmation requests.
11667 @item set confirm on
11668 Enables confirmation requests (the default).
11670 @kindex show confirm
11672 Displays state of confirmation requests.
11676 @node Debugging Output
11677 @section Optional messages about internal happenings
11679 @kindex set debug arch
11680 @item set debug arch
11681 Turns on or off display of gdbarch debugging info. The default is off
11682 @kindex show debug arch
11683 @item show debug arch
11684 Displays the current state of displaying gdbarch debugging info.
11685 @kindex set debug event
11686 @item set debug event
11687 Turns on or off display of @value{GDBN} event debugging info. The
11689 @kindex show debug event
11690 @item show debug event
11691 Displays the current state of displaying @value{GDBN} event debugging
11693 @kindex set debug expression
11694 @item set debug expression
11695 Turns on or off display of @value{GDBN} expression debugging info. The
11697 @kindex show debug expression
11698 @item show debug expression
11699 Displays the current state of displaying @value{GDBN} expression
11701 @kindex set debug overload
11702 @item set debug overload
11703 Turns on or off display of @value{GDBN} C++ overload debugging
11704 info. This includes info such as ranking of functions, etc. The default
11706 @kindex show debug overload
11707 @item show debug overload
11708 Displays the current state of displaying @value{GDBN} C++ overload
11710 @kindex set debug remote
11711 @cindex packets, reporting on stdout
11712 @cindex serial connections, debugging
11713 @item set debug remote
11714 Turns on or off display of reports on all packets sent back and forth across
11715 the serial line to the remote machine. The info is printed on the
11716 @value{GDBN} standard output stream. The default is off.
11717 @kindex show debug remote
11718 @item show debug remote
11719 Displays the state of display of remote packets.
11720 @kindex set debug serial
11721 @item set debug serial
11722 Turns on or off display of @value{GDBN} serial debugging info. The
11724 @kindex show debug serial
11725 @item show debug serial
11726 Displays the current state of displaying @value{GDBN} serial debugging
11728 @kindex set debug target
11729 @item set debug target
11730 Turns on or off display of @value{GDBN} target debugging info. This info
11731 includes what is going on at the target level of GDB, as it happens. The
11733 @kindex show debug target
11734 @item show debug target
11735 Displays the current state of displaying @value{GDBN} target debugging
11737 @kindex set debug varobj
11738 @item set debug varobj
11739 Turns on or off display of @value{GDBN} variable object debugging
11740 info. The default is off.
11741 @kindex show debug varobj
11742 @item show debug varobj
11743 Displays the current state of displaying @value{GDBN} variable object
11748 @chapter Canned Sequences of Commands
11750 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
11751 command lists}), @value{GDBN} provides two ways to store sequences of
11752 commands for execution as a unit: user-defined commands and command
11756 * Define:: User-defined commands
11757 * Hooks:: User-defined command hooks
11758 * Command Files:: Command files
11759 * Output:: Commands for controlled output
11763 @section User-defined commands
11765 @cindex user-defined command
11766 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
11767 which you assign a new name as a command. This is done with the
11768 @code{define} command. User commands may accept up to 10 arguments
11769 separated by whitespace. Arguments are accessed within the user command
11770 via @var{$arg0@dots{}$arg9}. A trivial example:
11774 print $arg0 + $arg1 + $arg2
11778 To execute the command use:
11785 This defines the command @code{adder}, which prints the sum of
11786 its three arguments. Note the arguments are text substitutions, so they may
11787 reference variables, use complex expressions, or even perform inferior
11793 @item define @var{commandname}
11794 Define a command named @var{commandname}. If there is already a command
11795 by that name, you are asked to confirm that you want to redefine it.
11797 The definition of the command is made up of other @value{GDBN} command lines,
11798 which are given following the @code{define} command. The end of these
11799 commands is marked by a line containing @code{end}.
11804 Takes a single argument, which is an expression to evaluate.
11805 It is followed by a series of commands that are executed
11806 only if the expression is true (nonzero).
11807 There can then optionally be a line @code{else}, followed
11808 by a series of commands that are only executed if the expression
11809 was false. The end of the list is marked by a line containing @code{end}.
11813 The syntax is similar to @code{if}: the command takes a single argument,
11814 which is an expression to evaluate, and must be followed by the commands to
11815 execute, one per line, terminated by an @code{end}.
11816 The commands are executed repeatedly as long as the expression
11820 @item document @var{commandname}
11821 Document the user-defined command @var{commandname}, so that it can be
11822 accessed by @code{help}. The command @var{commandname} must already be
11823 defined. This command reads lines of documentation just as @code{define}
11824 reads the lines of the command definition, ending with @code{end}.
11825 After the @code{document} command is finished, @code{help} on command
11826 @var{commandname} displays the documentation you have written.
11828 You may use the @code{document} command again to change the
11829 documentation of a command. Redefining the command with @code{define}
11830 does not change the documentation.
11832 @kindex help user-defined
11833 @item help user-defined
11834 List all user-defined commands, with the first line of the documentation
11839 @itemx show user @var{commandname}
11840 Display the @value{GDBN} commands used to define @var{commandname} (but
11841 not its documentation). If no @var{commandname} is given, display the
11842 definitions for all user-defined commands.
11846 When user-defined commands are executed, the
11847 commands of the definition are not printed. An error in any command
11848 stops execution of the user-defined command.
11850 If used interactively, commands that would ask for confirmation proceed
11851 without asking when used inside a user-defined command. Many @value{GDBN}
11852 commands that normally print messages to say what they are doing omit the
11853 messages when used in a user-defined command.
11856 @section User-defined command hooks
11857 @cindex command hooks
11858 @cindex hooks, for commands
11860 You may define @emph{hooks}, which are a special kind of user-defined
11861 command. Whenever you run the command @samp{foo}, if the user-defined
11862 command @samp{hook-foo} exists, it is executed (with no arguments)
11863 before that command.
11865 @kindex stop@r{, a pseudo-command}
11866 In addition, a pseudo-command, @samp{stop} exists. Defining
11867 (@samp{hook-stop}) makes the associated commands execute every time
11868 execution stops in your program: before breakpoint commands are run,
11869 displays are printed, or the stack frame is printed.
11871 For example, to ignore @code{SIGALRM} signals while
11872 single-stepping, but treat them normally during normal execution,
11877 handle SIGALRM nopass
11881 handle SIGALRM pass
11884 define hook-continue
11885 handle SIGLARM pass
11889 You can define a hook for any single-word command in @value{GDBN}, but
11890 not for command aliases; you should define a hook for the basic command
11891 name, e.g. @code{backtrace} rather than @code{bt}.
11892 @c FIXME! So how does Joe User discover whether a command is an alias
11894 If an error occurs during the execution of your hook, execution of
11895 @value{GDBN} commands stops and @value{GDBN} issues a prompt
11896 (before the command that you actually typed had a chance to run).
11898 If you try to define a hook which does not match any known command, you
11899 get a warning from the @code{define} command.
11901 @node Command Files
11902 @section Command files
11904 @cindex command files
11905 A command file for @value{GDBN} is a file of lines that are @value{GDBN}
11906 commands. Comments (lines starting with @kbd{#}) may also be included.
11907 An empty line in a command file does nothing; it does not mean to repeat
11908 the last command, as it would from the terminal.
11911 @cindex @file{.gdbinit}
11912 @cindex @file{gdb.ini}
11913 When you start @value{GDBN}, it automatically executes commands from its
11914 @dfn{init files}. These are files named @file{.gdbinit} on Unix and
11915 @file{gdb.ini} on DOS/Windows. During startup, @value{GDBN} does the
11920 Reads the init file (if any) in your home directory@footnote{On
11921 DOS/Windows systems, the home directory is the one pointed to by the
11922 @code{HOME} environment variable.}.
11925 Processes command line options and operands.
11928 Reads the init file (if any) in the current working directory.
11931 Reads command files specified by the @samp{-x} option.
11934 The init file in your home directory can set options (such as @samp{set
11935 complaints}) that affect subsequent processing of command line options
11936 and operands. Init files are not executed if you use the @samp{-nx}
11937 option (@pxref{Mode Options, ,Choosing modes}).
11939 @cindex init file name
11940 On some configurations of @value{GDBN}, the init file is known by a
11941 different name (these are typically environments where a specialized
11942 form of @value{GDBN} may need to coexist with other forms, hence a
11943 different name for the specialized version's init file). These are the
11944 environments with special init file names:
11946 @cindex @file{.vxgdbinit}
11949 VxWorks (Wind River Systems real-time OS): @file{.vxgdbinit}
11951 @cindex @file{.os68gdbinit}
11953 OS68K (Enea Data Systems real-time OS): @file{.os68gdbinit}
11955 @cindex @file{.esgdbinit}
11957 ES-1800 (Ericsson Telecom AB M68000 emulator): @file{.esgdbinit}
11960 You can also request the execution of a command file with the
11961 @code{source} command:
11965 @item source @var{filename}
11966 Execute the command file @var{filename}.
11969 The lines in a command file are executed sequentially. They are not
11970 printed as they are executed. An error in any command terminates execution
11971 of the command file.
11973 Commands that would ask for confirmation if used interactively proceed
11974 without asking when used in a command file. Many @value{GDBN} commands that
11975 normally print messages to say what they are doing omit the messages
11976 when called from command files.
11979 @section Commands for controlled output
11981 During the execution of a command file or a user-defined command, normal
11982 @value{GDBN} output is suppressed; the only output that appears is what is
11983 explicitly printed by the commands in the definition. This section
11984 describes three commands useful for generating exactly the output you
11989 @item echo @var{text}
11990 @c I do not consider backslash-space a standard C escape sequence
11991 @c because it is not in ANSI.
11992 Print @var{text}. Nonprinting characters can be included in
11993 @var{text} using C escape sequences, such as @samp{\n} to print a
11994 newline. @strong{No newline is printed unless you specify one.}
11995 In addition to the standard C escape sequences, a backslash followed
11996 by a space stands for a space. This is useful for displaying a
11997 string with spaces at the beginning or the end, since leading and
11998 trailing spaces are otherwise trimmed from all arguments.
11999 To print @samp{@w{ }and foo =@w{ }}, use the command
12000 @samp{echo \@w{ }and foo = \@w{ }}.
12002 A backslash at the end of @var{text} can be used, as in C, to continue
12003 the command onto subsequent lines. For example,
12006 echo This is some text\n\
12007 which is continued\n\
12008 onto several lines.\n
12011 produces the same output as
12014 echo This is some text\n
12015 echo which is continued\n
12016 echo onto several lines.\n
12020 @item output @var{expression}
12021 Print the value of @var{expression} and nothing but that value: no
12022 newlines, no @samp{$@var{nn} = }. The value is not entered in the
12023 value history either. @xref{Expressions, ,Expressions}, for more information
12026 @item output/@var{fmt} @var{expression}
12027 Print the value of @var{expression} in format @var{fmt}. You can use
12028 the same formats as for @code{print}. @xref{Output Formats,,Output
12029 formats}, for more information.
12032 @item printf @var{string}, @var{expressions}@dots{}
12033 Print the values of the @var{expressions} under the control of
12034 @var{string}. The @var{expressions} are separated by commas and may be
12035 either numbers or pointers. Their values are printed as specified by
12036 @var{string}, exactly as if your program were to execute the C
12038 @c FIXME: the above implies that at least all ANSI C formats are
12039 @c supported, but it isn't true: %E and %G don't work (or so it seems).
12040 @c Either this is a bug, or the manual should document what formats are
12044 printf (@var{string}, @var{expressions}@dots{});
12047 For example, you can print two values in hex like this:
12050 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
12053 The only backslash-escape sequences that you can use in the format
12054 string are the simple ones that consist of backslash followed by a
12059 @chapter Using @value{GDBN} under @sc{gnu} Emacs
12062 @cindex @sc{gnu} Emacs
12063 A special interface allows you to use @sc{gnu} Emacs to view (and
12064 edit) the source files for the program you are debugging with
12067 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
12068 executable file you want to debug as an argument. This command starts
12069 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
12070 created Emacs buffer.
12071 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
12073 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
12078 All ``terminal'' input and output goes through the Emacs buffer.
12081 This applies both to @value{GDBN} commands and their output, and to the input
12082 and output done by the program you are debugging.
12084 This is useful because it means that you can copy the text of previous
12085 commands and input them again; you can even use parts of the output
12088 All the facilities of Emacs' Shell mode are available for interacting
12089 with your program. In particular, you can send signals the usual
12090 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
12095 @value{GDBN} displays source code through Emacs.
12098 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
12099 source file for that frame and puts an arrow (@samp{=>}) at the
12100 left margin of the current line. Emacs uses a separate buffer for
12101 source display, and splits the screen to show both your @value{GDBN} session
12104 Explicit @value{GDBN} @code{list} or search commands still produce output as
12105 usual, but you probably have no reason to use them from Emacs.
12108 @emph{Warning:} If the directory where your program resides is not your
12109 current directory, it can be easy to confuse Emacs about the location of
12110 the source files, in which case the auxiliary display buffer does not
12111 appear to show your source. @value{GDBN} can find programs by searching your
12112 environment's @code{PATH} variable, so the @value{GDBN} input and output
12113 session proceeds normally; but Emacs does not get enough information
12114 back from @value{GDBN} to locate the source files in this situation. To
12115 avoid this problem, either start @value{GDBN} mode from the directory where
12116 your program resides, or specify an absolute file name when prompted for the
12117 @kbd{M-x gdb} argument.
12119 A similar confusion can result if you use the @value{GDBN} @code{file} command to
12120 switch to debugging a program in some other location, from an existing
12121 @value{GDBN} buffer in Emacs.
12124 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
12125 you need to call @value{GDBN} by a different name (for example, if you keep
12126 several configurations around, with different names) you can set the
12127 Emacs variable @code{gdb-command-name}; for example,
12130 (setq gdb-command-name "mygdb")
12134 (preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or
12135 in your @file{.emacs} file) makes Emacs call the program named
12136 ``@code{mygdb}'' instead.
12138 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
12139 addition to the standard Shell mode commands:
12143 Describe the features of Emacs' @value{GDBN} Mode.
12146 Execute to another source line, like the @value{GDBN} @code{step} command; also
12147 update the display window to show the current file and location.
12150 Execute to next source line in this function, skipping all function
12151 calls, like the @value{GDBN} @code{next} command. Then update the display window
12152 to show the current file and location.
12155 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
12156 display window accordingly.
12158 @item M-x gdb-nexti
12159 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
12160 display window accordingly.
12163 Execute until exit from the selected stack frame, like the @value{GDBN}
12164 @code{finish} command.
12167 Continue execution of your program, like the @value{GDBN} @code{continue}
12170 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
12173 Go up the number of frames indicated by the numeric argument
12174 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
12175 like the @value{GDBN} @code{up} command.
12177 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
12180 Go down the number of frames indicated by the numeric argument, like the
12181 @value{GDBN} @code{down} command.
12183 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
12186 Read the number where the cursor is positioned, and insert it at the end
12187 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
12188 around an address that was displayed earlier, type @kbd{disassemble};
12189 then move the cursor to the address display, and pick up the
12190 argument for @code{disassemble} by typing @kbd{C-x &}.
12192 You can customize this further by defining elements of the list
12193 @code{gdb-print-command}; once it is defined, you can format or
12194 otherwise process numbers picked up by @kbd{C-x &} before they are
12195 inserted. A numeric argument to @kbd{C-x &} indicates that you
12196 wish special formatting, and also acts as an index to pick an element of the
12197 list. If the list element is a string, the number to be inserted is
12198 formatted using the Emacs function @code{format}; otherwise the number
12199 is passed as an argument to the corresponding list element.
12202 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
12203 tells @value{GDBN} to set a breakpoint on the source line point is on.
12205 If you accidentally delete the source-display buffer, an easy way to get
12206 it back is to type the command @code{f} in the @value{GDBN} buffer, to
12207 request a frame display; when you run under Emacs, this recreates
12208 the source buffer if necessary to show you the context of the current
12211 The source files displayed in Emacs are in ordinary Emacs buffers
12212 which are visiting the source files in the usual way. You can edit
12213 the files with these buffers if you wish; but keep in mind that @value{GDBN}
12214 communicates with Emacs in terms of line numbers. If you add or
12215 delete lines from the text, the line numbers that @value{GDBN} knows cease
12216 to correspond properly with the code.
12218 @c The following dropped because Epoch is nonstandard. Reactivate
12219 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
12221 @kindex Emacs Epoch environment
12225 Version 18 of @sc{gnu} Emacs has a built-in window system
12226 called the @code{epoch}
12227 environment. Users of this environment can use a new command,
12228 @code{inspect} which performs identically to @code{print} except that
12229 each value is printed in its own window.
12232 @include annotate.texi
12233 @include gdbmi.texinfo
12236 @chapter Reporting Bugs in @value{GDBN}
12237 @cindex bugs in @value{GDBN}
12238 @cindex reporting bugs in @value{GDBN}
12240 Your bug reports play an essential role in making @value{GDBN} reliable.
12242 Reporting a bug may help you by bringing a solution to your problem, or it
12243 may not. But in any case the principal function of a bug report is to help
12244 the entire community by making the next version of @value{GDBN} work better. Bug
12245 reports are your contribution to the maintenance of @value{GDBN}.
12247 In order for a bug report to serve its purpose, you must include the
12248 information that enables us to fix the bug.
12251 * Bug Criteria:: Have you found a bug?
12252 * Bug Reporting:: How to report bugs
12256 @section Have you found a bug?
12257 @cindex bug criteria
12259 If you are not sure whether you have found a bug, here are some guidelines:
12262 @cindex fatal signal
12263 @cindex debugger crash
12264 @cindex crash of debugger
12266 If the debugger gets a fatal signal, for any input whatever, that is a
12267 @value{GDBN} bug. Reliable debuggers never crash.
12269 @cindex error on valid input
12271 If @value{GDBN} produces an error message for valid input, that is a
12272 bug. (Note that if you're cross debugging, the problem may also be
12273 somewhere in the connection to the target.)
12275 @cindex invalid input
12277 If @value{GDBN} does not produce an error message for invalid input,
12278 that is a bug. However, you should note that your idea of
12279 ``invalid input'' might be our idea of ``an extension'' or ``support
12280 for traditional practice''.
12283 If you are an experienced user of debugging tools, your suggestions
12284 for improvement of @value{GDBN} are welcome in any case.
12287 @node Bug Reporting
12288 @section How to report bugs
12289 @cindex bug reports
12290 @cindex @value{GDBN} bugs, reporting
12292 A number of companies and individuals offer support for @sc{gnu} products.
12293 If you obtained @value{GDBN} from a support organization, we recommend you
12294 contact that organization first.
12296 You can find contact information for many support companies and
12297 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
12299 @c should add a web page ref...
12301 In any event, we also recommend that you send bug reports for
12302 @value{GDBN} to this addresses:
12308 @strong{Do not send bug reports to @samp{info-gdb}, or to
12309 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
12310 not want to receive bug reports. Those that do have arranged to receive
12313 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
12314 serves as a repeater. The mailing list and the newsgroup carry exactly
12315 the same messages. Often people think of posting bug reports to the
12316 newsgroup instead of mailing them. This appears to work, but it has one
12317 problem which can be crucial: a newsgroup posting often lacks a mail
12318 path back to the sender. Thus, if we need to ask for more information,
12319 we may be unable to reach you. For this reason, it is better to send
12320 bug reports to the mailing list.
12322 As a last resort, send bug reports on paper to:
12325 @sc{gnu} Debugger Bugs
12326 Free Software Foundation Inc.
12327 59 Temple Place - Suite 330
12328 Boston, MA 02111-1307
12332 The fundamental principle of reporting bugs usefully is this:
12333 @strong{report all the facts}. If you are not sure whether to state a
12334 fact or leave it out, state it!
12336 Often people omit facts because they think they know what causes the
12337 problem and assume that some details do not matter. Thus, you might
12338 assume that the name of the variable you use in an example does not matter.
12339 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
12340 stray memory reference which happens to fetch from the location where that
12341 name is stored in memory; perhaps, if the name were different, the contents
12342 of that location would fool the debugger into doing the right thing despite
12343 the bug. Play it safe and give a specific, complete example. That is the
12344 easiest thing for you to do, and the most helpful.
12346 Keep in mind that the purpose of a bug report is to enable us to fix the
12347 bug. It may be that the bug has been reported previously, but neither
12348 you nor we can know that unless your bug report is complete and
12351 Sometimes people give a few sketchy facts and ask, ``Does this ring a
12352 bell?'' Those bug reports are useless, and we urge everyone to
12353 @emph{refuse to respond to them} except to chide the sender to report
12356 To enable us to fix the bug, you should include all these things:
12360 The version of @value{GDBN}. @value{GDBN} announces it if you start
12361 with no arguments; you can also print it at any time using @code{show
12364 Without this, we will not know whether there is any point in looking for
12365 the bug in the current version of @value{GDBN}.
12368 The type of machine you are using, and the operating system name and
12372 What compiler (and its version) was used to compile @value{GDBN}---e.g.
12373 ``@value{GCC}--2.8.1''.
12376 What compiler (and its version) was used to compile the program you are
12377 debugging---e.g. ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
12378 C Compiler''. For GCC, you can say @code{gcc --version} to get this
12379 information; for other compilers, see the documentation for those
12383 The command arguments you gave the compiler to compile your example and
12384 observe the bug. For example, did you use @samp{-O}? To guarantee
12385 you will not omit something important, list them all. A copy of the
12386 Makefile (or the output from make) is sufficient.
12388 If we were to try to guess the arguments, we would probably guess wrong
12389 and then we might not encounter the bug.
12392 A complete input script, and all necessary source files, that will
12396 A description of what behavior you observe that you believe is
12397 incorrect. For example, ``It gets a fatal signal.''
12399 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
12400 will certainly notice it. But if the bug is incorrect output, we might
12401 not notice unless it is glaringly wrong. You might as well not give us
12402 a chance to make a mistake.
12404 Even if the problem you experience is a fatal signal, you should still
12405 say so explicitly. Suppose something strange is going on, such as, your
12406 copy of @value{GDBN} is out of synch, or you have encountered a bug in
12407 the C library on your system. (This has happened!) Your copy might
12408 crash and ours would not. If you told us to expect a crash, then when
12409 ours fails to crash, we would know that the bug was not happening for
12410 us. If you had not told us to expect a crash, then we would not be able
12411 to draw any conclusion from our observations.
12414 If you wish to suggest changes to the @value{GDBN} source, send us context
12415 diffs. If you even discuss something in the @value{GDBN} source, refer to
12416 it by context, not by line number.
12418 The line numbers in our development sources will not match those in your
12419 sources. Your line numbers would convey no useful information to us.
12423 Here are some things that are not necessary:
12427 A description of the envelope of the bug.
12429 Often people who encounter a bug spend a lot of time investigating
12430 which changes to the input file will make the bug go away and which
12431 changes will not affect it.
12433 This is often time consuming and not very useful, because the way we
12434 will find the bug is by running a single example under the debugger
12435 with breakpoints, not by pure deduction from a series of examples.
12436 We recommend that you save your time for something else.
12438 Of course, if you can find a simpler example to report @emph{instead}
12439 of the original one, that is a convenience for us. Errors in the
12440 output will be easier to spot, running under the debugger will take
12441 less time, and so on.
12443 However, simplification is not vital; if you do not want to do this,
12444 report the bug anyway and send us the entire test case you used.
12447 A patch for the bug.
12449 A patch for the bug does help us if it is a good one. But do not omit
12450 the necessary information, such as the test case, on the assumption that
12451 a patch is all we need. We might see problems with your patch and decide
12452 to fix the problem another way, or we might not understand it at all.
12454 Sometimes with a program as complicated as @value{GDBN} it is very hard to
12455 construct an example that will make the program follow a certain path
12456 through the code. If you do not send us the example, we will not be able
12457 to construct one, so we will not be able to verify that the bug is fixed.
12459 And if we cannot understand what bug you are trying to fix, or why your
12460 patch should be an improvement, we will not install it. A test case will
12461 help us to understand.
12464 A guess about what the bug is or what it depends on.
12466 Such guesses are usually wrong. Even we cannot guess right about such
12467 things without first using the debugger to find the facts.
12470 @c The readline documentation is distributed with the readline code
12471 @c and consists of the two following files:
12473 @c inc-hist.texinfo
12474 @c Use -I with makeinfo to point to the appropriate directory,
12475 @c environment var TEXINPUTS with TeX.
12476 @include rluser.texinfo
12477 @include inc-hist.texinfo
12480 @node Formatting Documentation
12481 @appendix Formatting Documentation
12483 @cindex @value{GDBN} reference card
12484 @cindex reference card
12485 The @value{GDBN} 4 release includes an already-formatted reference card, ready
12486 for printing with PostScript or Ghostscript, in the @file{gdb}
12487 subdirectory of the main source directory@footnote{In
12488 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
12489 release.}. If you can use PostScript or Ghostscript with your printer,
12490 you can print the reference card immediately with @file{refcard.ps}.
12492 The release also includes the source for the reference card. You
12493 can format it, using @TeX{}, by typing:
12499 The @value{GDBN} reference card is designed to print in @dfn{landscape}
12500 mode on US ``letter'' size paper;
12501 that is, on a sheet 11 inches wide by 8.5 inches
12502 high. You will need to specify this form of printing as an option to
12503 your @sc{dvi} output program.
12505 @cindex documentation
12507 All the documentation for @value{GDBN} comes as part of the machine-readable
12508 distribution. The documentation is written in Texinfo format, which is
12509 a documentation system that uses a single source file to produce both
12510 on-line information and a printed manual. You can use one of the Info
12511 formatting commands to create the on-line version of the documentation
12512 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
12514 @value{GDBN} includes an already formatted copy of the on-line Info
12515 version of this manual in the @file{gdb} subdirectory. The main Info
12516 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
12517 subordinate files matching @samp{gdb.info*} in the same directory. If
12518 necessary, you can print out these files, or read them with any editor;
12519 but they are easier to read using the @code{info} subsystem in @sc{gnu}
12520 Emacs or the standalone @code{info} program, available as part of the
12521 @sc{gnu} Texinfo distribution.
12523 If you want to format these Info files yourself, you need one of the
12524 Info formatting programs, such as @code{texinfo-format-buffer} or
12527 If you have @code{makeinfo} installed, and are in the top level
12528 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
12529 version @value{GDBVN}), you can make the Info file by typing:
12536 If you want to typeset and print copies of this manual, you need @TeX{},
12537 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
12538 Texinfo definitions file.
12540 @TeX{} is a typesetting program; it does not print files directly, but
12541 produces output files called @sc{dvi} files. To print a typeset
12542 document, you need a program to print @sc{dvi} files. If your system
12543 has @TeX{} installed, chances are it has such a program. The precise
12544 command to use depends on your system; @kbd{lpr -d} is common; another
12545 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
12546 require a file name without any extension or a @samp{.dvi} extension.
12548 @TeX{} also requires a macro definitions file called
12549 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
12550 written in Texinfo format. On its own, @TeX{} cannot either read or
12551 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
12552 and is located in the @file{gdb-@var{version-number}/texinfo}
12555 If you have @TeX{} and a @sc{dvi} printer program installed, you can
12556 typeset and print this manual. First switch to the the @file{gdb}
12557 subdirectory of the main source directory (for example, to
12558 @file{gdb-@value{GDBVN}/gdb}) and type:
12564 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
12566 @node Installing GDB
12567 @appendix Installing @value{GDBN}
12568 @cindex configuring @value{GDBN}
12569 @cindex installation
12571 @value{GDBN} comes with a @code{configure} script that automates the process
12572 of preparing @value{GDBN} for installation; you can then use @code{make} to
12573 build the @code{gdb} program.
12575 @c irrelevant in info file; it's as current as the code it lives with.
12576 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
12577 look at the @file{README} file in the sources; we may have improved the
12578 installation procedures since publishing this manual.}
12581 The @value{GDBN} distribution includes all the source code you need for
12582 @value{GDBN} in a single directory, whose name is usually composed by
12583 appending the version number to @samp{gdb}.
12585 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
12586 @file{gdb-@value{GDBVN}} directory. That directory contains:
12589 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
12590 script for configuring @value{GDBN} and all its supporting libraries
12592 @item gdb-@value{GDBVN}/gdb
12593 the source specific to @value{GDBN} itself
12595 @item gdb-@value{GDBVN}/bfd
12596 source for the Binary File Descriptor library
12598 @item gdb-@value{GDBVN}/include
12599 @sc{gnu} include files
12601 @item gdb-@value{GDBVN}/libiberty
12602 source for the @samp{-liberty} free software library
12604 @item gdb-@value{GDBVN}/opcodes
12605 source for the library of opcode tables and disassemblers
12607 @item gdb-@value{GDBVN}/readline
12608 source for the @sc{gnu} command-line interface
12610 @item gdb-@value{GDBVN}/glob
12611 source for the @sc{gnu} filename pattern-matching subroutine
12613 @item gdb-@value{GDBVN}/mmalloc
12614 source for the @sc{gnu} memory-mapped malloc package
12617 The simplest way to configure and build @value{GDBN} is to run @code{configure}
12618 from the @file{gdb-@var{version-number}} source directory, which in
12619 this example is the @file{gdb-@value{GDBVN}} directory.
12621 First switch to the @file{gdb-@var{version-number}} source directory
12622 if you are not already in it; then run @code{configure}. Pass the
12623 identifier for the platform on which @value{GDBN} will run as an
12629 cd gdb-@value{GDBVN}
12630 ./configure @var{host}
12635 where @var{host} is an identifier such as @samp{sun4} or
12636 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
12637 (You can often leave off @var{host}; @code{configure} tries to guess the
12638 correct value by examining your system.)
12640 Running @samp{configure @var{host}} and then running @code{make} builds the
12641 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
12642 libraries, then @code{gdb} itself. The configured source files, and the
12643 binaries, are left in the corresponding source directories.
12646 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
12647 system does not recognize this automatically when you run a different
12648 shell, you may need to run @code{sh} on it explicitly:
12651 sh configure @var{host}
12654 If you run @code{configure} from a directory that contains source
12655 directories for multiple libraries or programs, such as the
12656 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
12657 creates configuration files for every directory level underneath (unless
12658 you tell it not to, with the @samp{--norecursion} option).
12660 You can run the @code{configure} script from any of the
12661 subordinate directories in the @value{GDBN} distribution if you only want to
12662 configure that subdirectory, but be sure to specify a path to it.
12664 For example, with version @value{GDBVN}, type the following to configure only
12665 the @code{bfd} subdirectory:
12669 cd gdb-@value{GDBVN}/bfd
12670 ../configure @var{host}
12674 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
12675 However, you should make sure that the shell on your path (named by
12676 the @samp{SHELL} environment variable) is publicly readable. Remember
12677 that @value{GDBN} uses the shell to start your program---some systems refuse to
12678 let @value{GDBN} debug child processes whose programs are not readable.
12681 * Separate Objdir:: Compiling @value{GDBN} in another directory
12682 * Config Names:: Specifying names for hosts and targets
12683 * Configure Options:: Summary of options for configure
12686 @node Separate Objdir
12687 @section Compiling @value{GDBN} in another directory
12689 If you want to run @value{GDBN} versions for several host or target machines,
12690 you need a different @code{gdb} compiled for each combination of
12691 host and target. @code{configure} is designed to make this easy by
12692 allowing you to generate each configuration in a separate subdirectory,
12693 rather than in the source directory. If your @code{make} program
12694 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
12695 @code{make} in each of these directories builds the @code{gdb}
12696 program specified there.
12698 To build @code{gdb} in a separate directory, run @code{configure}
12699 with the @samp{--srcdir} option to specify where to find the source.
12700 (You also need to specify a path to find @code{configure}
12701 itself from your working directory. If the path to @code{configure}
12702 would be the same as the argument to @samp{--srcdir}, you can leave out
12703 the @samp{--srcdir} option; it is assumed.)
12705 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
12706 separate directory for a Sun 4 like this:
12710 cd gdb-@value{GDBVN}
12713 ../gdb-@value{GDBVN}/configure sun4
12718 When @code{configure} builds a configuration using a remote source
12719 directory, it creates a tree for the binaries with the same structure
12720 (and using the same names) as the tree under the source directory. In
12721 the example, you'd find the Sun 4 library @file{libiberty.a} in the
12722 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
12723 @file{gdb-sun4/gdb}.
12725 One popular reason to build several @value{GDBN} configurations in separate
12726 directories is to configure @value{GDBN} for cross-compiling (where
12727 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
12728 programs that run on another machine---the @dfn{target}).
12729 You specify a cross-debugging target by
12730 giving the @samp{--target=@var{target}} option to @code{configure}.
12732 When you run @code{make} to build a program or library, you must run
12733 it in a configured directory---whatever directory you were in when you
12734 called @code{configure} (or one of its subdirectories).
12736 The @code{Makefile} that @code{configure} generates in each source
12737 directory also runs recursively. If you type @code{make} in a source
12738 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
12739 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
12740 will build all the required libraries, and then build GDB.
12742 When you have multiple hosts or targets configured in separate
12743 directories, you can run @code{make} on them in parallel (for example,
12744 if they are NFS-mounted on each of the hosts); they will not interfere
12748 @section Specifying names for hosts and targets
12750 The specifications used for hosts and targets in the @code{configure}
12751 script are based on a three-part naming scheme, but some short predefined
12752 aliases are also supported. The full naming scheme encodes three pieces
12753 of information in the following pattern:
12756 @var{architecture}-@var{vendor}-@var{os}
12759 For example, you can use the alias @code{sun4} as a @var{host} argument,
12760 or as the value for @var{target} in a @code{--target=@var{target}}
12761 option. The equivalent full name is @samp{sparc-sun-sunos4}.
12763 The @code{configure} script accompanying @value{GDBN} does not provide
12764 any query facility to list all supported host and target names or
12765 aliases. @code{configure} calls the Bourne shell script
12766 @code{config.sub} to map abbreviations to full names; you can read the
12767 script, if you wish, or you can use it to test your guesses on
12768 abbreviations---for example:
12771 % sh config.sub i386-linux
12773 % sh config.sub alpha-linux
12774 alpha-unknown-linux-gnu
12775 % sh config.sub hp9k700
12777 % sh config.sub sun4
12778 sparc-sun-sunos4.1.1
12779 % sh config.sub sun3
12780 m68k-sun-sunos4.1.1
12781 % sh config.sub i986v
12782 Invalid configuration `i986v': machine `i986v' not recognized
12786 @code{config.sub} is also distributed in the @value{GDBN} source
12787 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
12789 @node Configure Options
12790 @section @code{configure} options
12792 Here is a summary of the @code{configure} options and arguments that
12793 are most often useful for building @value{GDBN}. @code{configure} also has
12794 several other options not listed here. @inforef{What Configure
12795 Does,,configure.info}, for a full explanation of @code{configure}.
12798 configure @r{[}--help@r{]}
12799 @r{[}--prefix=@var{dir}@r{]}
12800 @r{[}--exec-prefix=@var{dir}@r{]}
12801 @r{[}--srcdir=@var{dirname}@r{]}
12802 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
12803 @r{[}--target=@var{target}@r{]}
12808 You may introduce options with a single @samp{-} rather than
12809 @samp{--} if you prefer; but you may abbreviate option names if you use
12814 Display a quick summary of how to invoke @code{configure}.
12816 @item --prefix=@var{dir}
12817 Configure the source to install programs and files under directory
12820 @item --exec-prefix=@var{dir}
12821 Configure the source to install programs under directory
12824 @c avoid splitting the warning from the explanation:
12826 @item --srcdir=@var{dirname}
12827 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
12828 @code{make} that implements the @code{VPATH} feature.}@*
12829 Use this option to make configurations in directories separate from the
12830 @value{GDBN} source directories. Among other things, you can use this to
12831 build (or maintain) several configurations simultaneously, in separate
12832 directories. @code{configure} writes configuration specific files in
12833 the current directory, but arranges for them to use the source in the
12834 directory @var{dirname}. @code{configure} creates directories under
12835 the working directory in parallel to the source directories below
12838 @item --norecursion
12839 Configure only the directory level where @code{configure} is executed; do not
12840 propagate configuration to subdirectories.
12842 @item --target=@var{target}
12843 Configure @value{GDBN} for cross-debugging programs running on the specified
12844 @var{target}. Without this option, @value{GDBN} is configured to debug
12845 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
12847 There is no convenient way to generate a list of all available targets.
12849 @item @var{host} @dots{}
12850 Configure @value{GDBN} to run on the specified @var{host}.
12852 There is no convenient way to generate a list of all available hosts.
12855 There are many other options available as well, but they are generally
12856 needed for special purposes only.
12864 % I think something like @colophon should be in texinfo. In the
12866 \long\def\colophon{\hbox to0pt{}\vfill
12867 \centerline{The body of this manual is set in}
12868 \centerline{\fontname\tenrm,}
12869 \centerline{with headings in {\bf\fontname\tenbf}}
12870 \centerline{and examples in {\tt\fontname\tentt}.}
12871 \centerline{{\it\fontname\tenit\/},}
12872 \centerline{{\bf\fontname\tenbf}, and}
12873 \centerline{{\sl\fontname\tensl\/}}
12874 \centerline{are used for emphasis.}\vfill}
12876 % Blame: doc@cygnus.com, 1991.