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 Stallman, Roland Pesch, Stan Shebs, et al.
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 Andrew Cagney (release 5.0);
301 Jim Blandy (release 4.18);
302 Jason Molenda (release 4.17);
303 Stan Shebs (release 4.14);
304 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
305 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
306 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
307 Jim Kingdon (releases 3.5, 3.4, and 3.3);
308 and Randy Smith (releases 3.2, 3.1, and 3.0).
310 Richard Stallman, assisted at various times by Peter TerMaat, Chris
311 Hanson, and Richard Mlynarik, handled releases through 2.8.
313 Michael Tiemann is the author of most of the @sc{gnu} C++ support in
314 @value{GDBN}, with significant additional contributions from Per
315 Bothner. James Clark wrote the @sc{gnu} C++ demangler. Early work on
316 C++ was by Peter TerMaat (who also did much general update work leading
319 @value{GDBN} 4 uses the BFD subroutine library to examine multiple
320 object-file formats; BFD was a joint project of David V.
321 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
323 David Johnson wrote the original COFF support; Pace Willison did
324 the original support for encapsulated COFF.
326 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
328 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
329 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
331 Jean-Daniel Fekete contributed Sun 386i support.
332 Chris Hanson improved the HP9000 support.
333 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
334 David Johnson contributed Encore Umax support.
335 Jyrki Kuoppala contributed Altos 3068 support.
336 Jeff Law contributed HP PA and SOM support.
337 Keith Packard contributed NS32K support.
338 Doug Rabson contributed Acorn Risc Machine support.
339 Bob Rusk contributed Harris Nighthawk CX-UX support.
340 Chris Smith contributed Convex support (and Fortran debugging).
341 Jonathan Stone contributed Pyramid support.
342 Michael Tiemann contributed SPARC support.
343 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
344 Pace Willison contributed Intel 386 support.
345 Jay Vosburgh contributed Symmetry support.
347 Andreas Schwab contributed M68K Linux support.
349 Rich Schaefer and Peter Schauer helped with support of SunOS shared
352 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
353 about several machine instruction sets.
355 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
356 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
357 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
358 and RDI targets, respectively.
360 Brian Fox is the author of the readline libraries providing
361 command-line editing and command history.
363 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
364 Modula-2 support, and contributed the Languages chapter of this manual.
366 Fred Fish wrote most of the support for Unix System Vr4.
367 He also enhanced the command-completion support to cover C++ overloaded
370 Hitachi America, Ltd. sponsored the support for H8/300, H8/500, and
373 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
375 Mitsubishi sponsored the support for D10V, D30V, and M32R/D processors.
377 Toshiba sponsored the support for the TX39 Mips processor.
379 Matsushita sponsored the support for the MN10200 and MN10300 processors.
381 Fujitsu sponsored the support for SPARClite and FR30 processors.
383 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
386 Michael Snyder added support for tracepoints.
388 Stu Grossman wrote gdbserver.
390 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
391 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
393 The following people at the Hewlett-Packard Company contributed
394 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
395 (narrow mode), HP's implementation of kernel threads, HP's aC++
396 compiler, and the terminal user interface: Ben Krepp, Richard Title,
397 John Bishop, Susan Macchia, Kathy Mann, Satish Pai, India Paul, Steve
398 Rehrauer, and Elena Zannoni. Kim Haase provided HP-specific
399 information in this manual.
401 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
402 development since 1991. Cygnus engineers who have worked on @value{GDBN}
403 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
404 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
405 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
406 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
407 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
408 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
409 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
410 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
411 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
412 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
413 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
414 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
415 Zuhn have made contributions both large and small.
419 @chapter A Sample @value{GDBN} Session
421 You can use this manual at your leisure to read all about @value{GDBN}.
422 However, a handful of commands are enough to get started using the
423 debugger. This chapter illustrates those commands.
426 In this sample session, we emphasize user input like this: @b{input},
427 to make it easier to pick out from the surrounding output.
430 @c FIXME: this example may not be appropriate for some configs, where
431 @c FIXME...primary interest is in remote use.
433 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
434 processor) exhibits the following bug: sometimes, when we change its
435 quote strings from the default, the commands used to capture one macro
436 definition within another stop working. In the following short @code{m4}
437 session, we define a macro @code{foo} which expands to @code{0000}; we
438 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
439 same thing. However, when we change the open quote string to
440 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
441 procedure fails to define a new synonym @code{baz}:
450 @b{define(bar,defn(`foo'))}
454 @b{changequote(<QUOTE>,<UNQUOTE>)}
456 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
459 m4: End of input: 0: fatal error: EOF in string
463 Let us use @value{GDBN} to try to see what is going on.
466 $ @b{@value{GDBP} m4}
467 @c FIXME: this falsifies the exact text played out, to permit smallbook
468 @c FIXME... format to come out better.
469 @value{GDBN} is free software and you are welcome to distribute copies
470 of it under certain conditions; type "show copying" to see
472 There is absolutely no warranty for @value{GDBN}; type "show warranty"
475 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
480 @value{GDBN} reads only enough symbol data to know where to find the
481 rest when needed; as a result, the first prompt comes up very quickly.
482 We now tell @value{GDBN} to use a narrower display width than usual, so
483 that examples fit in this manual.
486 (@value{GDBP}) @b{set width 70}
490 We need to see how the @code{m4} built-in @code{changequote} works.
491 Having looked at the source, we know the relevant subroutine is
492 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
493 @code{break} command.
496 (@value{GDBP}) @b{break m4_changequote}
497 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
501 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
502 control; as long as control does not reach the @code{m4_changequote}
503 subroutine, the program runs as usual:
506 (@value{GDBP}) @b{run}
507 Starting program: /work/Editorial/gdb/gnu/m4/m4
515 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
516 suspends execution of @code{m4}, displaying information about the
517 context where it stops.
520 @b{changequote(<QUOTE>,<UNQUOTE>)}
522 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
524 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
528 Now we use the command @code{n} (@code{next}) to advance execution to
529 the next line of the current function.
533 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
538 @code{set_quotes} looks like a promising subroutine. We can go into it
539 by using the command @code{s} (@code{step}) instead of @code{next}.
540 @code{step} goes to the next line to be executed in @emph{any}
541 subroutine, so it steps into @code{set_quotes}.
545 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
547 530 if (lquote != def_lquote)
551 The display that shows the subroutine where @code{m4} is now
552 suspended (and its arguments) is called a stack frame display. It
553 shows a summary of the stack. We can use the @code{backtrace}
554 command (which can also be spelled @code{bt}), to see where we are
555 in the stack as a whole: the @code{backtrace} command displays a
556 stack frame for each active subroutine.
559 (@value{GDBP}) @b{bt}
560 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
562 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
564 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
565 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
567 #4 0x79dc in expand_input () at macro.c:40
568 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
572 We step through a few more lines to see what happens. The first two
573 times, we can use @samp{s}; the next two times we use @code{n} to avoid
574 falling into the @code{xstrdup} subroutine.
578 0x3b5c 532 if (rquote != def_rquote)
580 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
581 def_lquote : xstrdup(lq);
583 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
586 538 len_lquote = strlen(rquote);
590 The last line displayed looks a little odd; we can examine the variables
591 @code{lquote} and @code{rquote} to see if they are in fact the new left
592 and right quotes we specified. We use the command @code{p}
593 (@code{print}) to see their values.
596 (@value{GDBP}) @b{p lquote}
597 $1 = 0x35d40 "<QUOTE>"
598 (@value{GDBP}) @b{p rquote}
599 $2 = 0x35d50 "<UNQUOTE>"
603 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
604 To look at some context, we can display ten lines of source
605 surrounding the current line with the @code{l} (@code{list}) command.
611 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
613 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
616 538 len_lquote = strlen(rquote);
617 539 len_rquote = strlen(lquote);
624 Let us step past the two lines that set @code{len_lquote} and
625 @code{len_rquote}, and then examine the values of those variables.
629 539 len_rquote = strlen(lquote);
632 (@value{GDBP}) @b{p len_lquote}
634 (@value{GDBP}) @b{p len_rquote}
639 That certainly looks wrong, assuming @code{len_lquote} and
640 @code{len_rquote} are meant to be the lengths of @code{lquote} and
641 @code{rquote} respectively. We can set them to better values using
642 the @code{p} command, since it can print the value of
643 any expression---and that expression can include subroutine calls and
647 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
649 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
654 Is that enough to fix the problem of using the new quotes with the
655 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
656 executing with the @code{c} (@code{continue}) command, and then try the
657 example that caused trouble initially:
663 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
670 Success! The new quotes now work just as well as the default ones. The
671 problem seems to have been just the two typos defining the wrong
672 lengths. We allow @code{m4} exit by giving it an EOF as input:
676 Program exited normally.
680 The message @samp{Program exited normally.} is from @value{GDBN}; it
681 indicates @code{m4} has finished executing. We can end our @value{GDBN}
682 session with the @value{GDBN} @code{quit} command.
685 (@value{GDBP}) @b{quit}
689 @chapter Getting In and Out of @value{GDBN}
691 This chapter discusses how to start @value{GDBN}, and how to get out of it.
695 type @samp{@value{GDBP}} to start @value{GDBN}.
697 type @kbd{quit} or @kbd{C-d} to exit.
701 * Invoking GDB:: How to start @value{GDBN}
702 * Quitting GDB:: How to quit @value{GDBN}
703 * Shell Commands:: How to use shell commands inside @value{GDBN}
707 @section Invoking @value{GDBN}
709 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
710 @value{GDBN} reads commands from the terminal until you tell it to exit.
712 You can also run @code{@value{GDBP}} with a variety of arguments and options,
713 to specify more of your debugging environment at the outset.
715 The command-line options described here are designed
716 to cover a variety of situations; in some environments, some of these
717 options may effectively be unavailable.
719 The most usual way to start @value{GDBN} is with one argument,
720 specifying an executable program:
723 @value{GDBP} @var{program}
727 You can also start with both an executable program and a core file
731 @value{GDBP} @var{program} @var{core}
734 You can, instead, specify a process ID as a second argument, if you want
735 to debug a running process:
738 @value{GDBP} @var{program} 1234
742 would attach @value{GDBN} to process @code{1234} (unless you also have a file
743 named @file{1234}; @value{GDBN} does check for a core file first).
745 Taking advantage of the second command-line argument requires a fairly
746 complete operating system; when you use @value{GDBN} as a remote
747 debugger attached to a bare board, there may not be any notion of
748 ``process'', and there is often no way to get a core dump. @value{GDBN}
749 will warn you if it is unable to attach or to read core dumps.
751 You can run @code{@value{GDBP}} without printing the front material, which describes
752 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
759 You can further control how @value{GDBN} starts up by using command-line
760 options. @value{GDBN} itself can remind you of the options available.
770 to display all available options and briefly describe their use
771 (@samp{@value{GDBP} -h} is a shorter equivalent).
773 All options and command line arguments you give are processed
774 in sequential order. The order makes a difference when the
775 @samp{-x} option is used.
779 * File Options:: Choosing files
780 * Mode Options:: Choosing modes
784 @subsection Choosing files
786 When @value{GDBN} starts, it reads any arguments other than options as
787 specifying an executable file and core file (or process ID). This is
788 the same as if the arguments were specified by the @samp{-se} and
789 @samp{-c} options respectively. (@value{GDBN} reads the first argument
790 that does not have an associated option flag as equivalent to the
791 @samp{-se} option followed by that argument; and the second argument
792 that does not have an associated option flag, if any, as equivalent to
793 the @samp{-c} option followed by that argument.)
795 If @value{GDBN} has not been configured to included core file support,
796 such as for most embedded targets, then it will complain about a second
797 argument and ignore it.
799 Many options have both long and short forms; both are shown in the
800 following list. @value{GDBN} also recognizes the long forms if you truncate
801 them, so long as enough of the option is present to be unambiguous.
802 (If you prefer, you can flag option arguments with @samp{--} rather
803 than @samp{-}, though we illustrate the more usual convention.)
805 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
806 @c way, both those who look for -foo and --foo in the index, will find
810 @item -symbols @var{file}
812 @cindex @code{--symbols}
814 Read symbol table from file @var{file}.
816 @item -exec @var{file}
818 @cindex @code{--exec}
820 Use file @var{file} as the executable file to execute when appropriate,
821 and for examining pure data in conjunction with a core dump.
825 Read symbol table from file @var{file} and use it as the executable
828 @item -core @var{file}
830 @cindex @code{--core}
832 Use file @var{file} as a core dump to examine.
834 @item -c @var{number}
835 Connect to process ID @var{number}, as with the @code{attach} command
836 (unless there is a file in core-dump format named @var{number}, in which
837 case @samp{-c} specifies that file as a core dump to read).
839 @item -command @var{file}
841 @cindex @code{--command}
843 Execute @value{GDBN} commands from file @var{file}. @xref{Command
844 Files,, Command files}.
846 @item -directory @var{directory}
847 @itemx -d @var{directory}
848 @cindex @code{--directory}
850 Add @var{directory} to the path to search for source files.
854 @cindex @code{--mapped}
856 @emph{Warning: this option depends on operating system facilities that are not
857 supported on all systems.}@*
858 If memory-mapped files are available on your system through the @code{mmap}
859 system call, you can use this option
860 to have @value{GDBN} write the symbols from your
861 program into a reusable file in the current directory. If the program you are debugging is
862 called @file{/tmp/fred}, the mapped symbol file is @file{/tmp/fred.syms}.
863 Future @value{GDBN} debugging sessions notice the presence of this file,
864 and can quickly map in symbol information from it, rather than reading
865 the symbol table from the executable program.
867 The @file{.syms} file is specific to the host machine where @value{GDBN}
868 is run. It holds an exact image of the internal @value{GDBN} symbol
869 table. It cannot be shared across multiple host platforms.
873 @cindex @code{--readnow}
875 Read each symbol file's entire symbol table immediately, rather than
876 the default, which is to read it incrementally as it is needed.
877 This makes startup slower, but makes future operations faster.
881 You typically combine the @code{-mapped} and @code{-readnow} options in
882 order to build a @file{.syms} file that contains complete symbol
883 information. (@xref{Files,,Commands to specify files}, for information
884 on @file{.syms} files.) A simple @value{GDBN} invocation to do nothing
885 but build a @file{.syms} file for future use is:
888 gdb -batch -nx -mapped -readnow programname
892 @subsection Choosing modes
894 You can run @value{GDBN} in various alternative modes---for example, in
895 batch mode or quiet mode.
902 Do not execute commands found in any initialization files (normally
903 called @file{.gdbinit}, or @file{gdb.ini} on PCs). Normally,
904 @value{GDBN} executes the commands in these files after all the command
905 options and arguments have been processed. @xref{Command Files,,Command
911 @cindex @code{--quiet}
912 @cindex @code{--silent}
914 ``Quiet''. Do not print the introductory and copyright messages. These
915 messages are also suppressed in batch mode.
918 @cindex @code{--batch}
919 Run in batch mode. Exit with status @code{0} after processing all the
920 command files specified with @samp{-x} (and all commands from
921 initialization files, if not inhibited with @samp{-n}). Exit with
922 nonzero status if an error occurs in executing the @value{GDBN} commands
923 in the command files.
925 Batch mode may be useful for running @value{GDBN} as a filter, for
926 example to download and run a program on another computer; in order to
927 make this more useful, the message
930 Program exited normally.
934 (which is ordinarily issued whenever a program running under
935 @value{GDBN} control terminates) is not issued when running in batch
940 @cindex @code{--nowindows}
942 ``No windows''. If @value{GDBN} comes with a graphical user interface
943 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
944 interface. If no GUI is available, this option has no effect.
948 @cindex @code{--windows}
950 If @value{GDBN} includes a GUI, then this option requires it to be
953 @item -cd @var{directory}
955 Run @value{GDBN} using @var{directory} as its working directory,
956 instead of the current directory.
960 @cindex @code{--fullname}
962 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
963 subprocess. It tells @value{GDBN} to output the full file name and line
964 number in a standard, recognizable fashion each time a stack frame is
965 displayed (which includes each time your program stops). This
966 recognizable format looks like two @samp{\032} characters, followed by
967 the file name, line number and character position separated by colons,
968 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
969 @samp{\032} characters as a signal to display the source code for the
973 @cindex @code{--epoch}
974 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
975 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
976 routines so as to allow Epoch to display values of expressions in a
979 @item -annotate @var{level}
980 @cindex @code{--annotate}
981 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
982 effect is identical to using @samp{set annotate @var{level}}
983 (@pxref{Annotations}).
984 Annotation level controls how much information does @value{GDBN} print
985 together with its prompt, values of expressions, source lines, and other
986 types of output. Level 0 is the normal, level 1 is for use when
987 @value{GDBN} is run as a subprocess of @sc{gnu} Emacs, level 2 is the
988 maximum annotation suitable for programs that control @value{GDBN}.
991 @cindex @code{--async}
992 Use the asynchronous event loop for the command-line interface.
993 @value{GDBN} processes all events, such as user keyboard input, via a
994 special event loop. This allows @value{GDBN} to accept and process user
995 commands in parallel with the debugged process being
996 run@footnote{@value{GDBN} built with @sc{djgpp} tools for
997 MS-DOS/MS-Windows supports this mode of operation, but the event loop is
998 suspended when the debuggee runs.}, so you don't need to wait for
999 control to return to @value{GDBN} before you type the next command.
1000 (@emph{Note:} as of version 5.0, the target side of the asynchronous
1001 operation is not yet in place, so @samp{-async} does not work fully
1003 @c FIXME: when the target side of the event loop is done, the above NOTE
1004 @c should be removed.
1006 When the standard input is connected to a terminal device, @value{GDBN}
1007 uses the asynchronous event loop by default, unless disabled by the
1008 @samp{-noasync} option.
1011 @cindex @code{--noasync}
1012 Disable the asynchronous event loop for the command-line interface.
1014 @item -baud @var{bps}
1016 @cindex @code{--baud}
1018 Set the line speed (baud rate or bits per second) of any serial
1019 interface used by @value{GDBN} for remote debugging.
1021 @item -tty @var{device}
1022 @itemx -t @var{device}
1023 @cindex @code{--tty}
1025 Run using @var{device} for your program's standard input and output.
1026 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1028 @c resolve the situation of these eventually
1030 @c @cindex @code{--tui}
1031 @c Use a Terminal User Interface. For information, use your Web browser to
1032 @c read the file @file{TUI.html}, which is usually installed in the
1033 @c directory @code{/opt/langtools/wdb/doc} on HP-UX systems. Do not use
1034 @c this option if you run @value{GDBN} from Emacs (see @pxref{Emacs, ,Using
1035 @c @value{GDBN} under @sc{gnu} Emacs}).
1038 @c @cindex @code{--xdb}
1039 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1040 @c For information, see the file @file{xdb_trans.html}, which is usually
1041 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1044 @item -interpreter @var{interp}
1045 @cindex @code{--interpreter}
1046 Use the interpreter @var{interp} for interface with the controlling
1047 program or device. This option is meant to be set by programs which
1048 communicate with @value{GDBN} using it as a back end. For example,
1049 @samp{--interpreter=mi} causes @value{GDBN} to use the @dfn{gdbmi
1050 interface} (@pxref{GDB/MI, , The @sc{gdb/mi} Interface}).
1053 @cindex @code{--write}
1054 Open the executable and core files for both reading and writing. This
1055 is equivalent to the @samp{set write on} command inside @value{GDBN}
1059 @cindex @code{--statistics}
1060 This option causes @value{GDBN} to print statistics about time and
1061 memory usage after it completes each command and returns to the prompt.
1064 @cindex @code{--version}
1065 This option causes @value{GDBN} to print its version number and
1066 no-warranty blurb, and exit.
1071 @section Quitting @value{GDBN}
1072 @cindex exiting @value{GDBN}
1073 @cindex leaving @value{GDBN}
1076 @kindex quit @r{[}@var{expression}@r{]}
1077 @kindex q @r{(@code{quit})}
1078 @item quit @r{[}@var{expression}@r{]}
1080 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1081 @code{q}), or type an end-of-file character (usually @kbd{C-d}). If you
1082 do not supply @var{expression}, @value{GDBN} will terminate normally;
1083 otherwise it will terminate using the result of @var{expression} as the
1088 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
1089 terminates the action of any @value{GDBN} command that is in progress and
1090 returns to @value{GDBN} command level. It is safe to type the interrupt
1091 character at any time because @value{GDBN} does not allow it to take effect
1092 until a time when it is safe.
1094 If you have been using @value{GDBN} to control an attached process or
1095 device, you can release it with the @code{detach} command
1096 (@pxref{Attach, ,Debugging an already-running process}).
1098 @node Shell Commands
1099 @section Shell commands
1101 If you need to execute occasional shell commands during your
1102 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1103 just use the @code{shell} command.
1107 @cindex shell escape
1108 @item shell @var{command string}
1109 Invoke a standard shell to execute @var{command string}.
1110 If it exists, the environment variable @code{SHELL} determines which
1111 shell to run. Otherwise @value{GDBN} uses the default shell
1112 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1115 The utility @code{make} is often needed in development environments.
1116 You do not have to use the @code{shell} command for this purpose in
1121 @cindex calling make
1122 @item make @var{make-args}
1123 Execute the @code{make} program with the specified
1124 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1128 @chapter @value{GDBN} Commands
1130 You can abbreviate a @value{GDBN} command to the first few letters of the command
1131 name, if that abbreviation is unambiguous; and you can repeat certain
1132 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1133 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1134 show you the alternatives available, if there is more than one possibility).
1137 * Command Syntax:: How to give commands to @value{GDBN}
1138 * Completion:: Command completion
1139 * Help:: How to ask @value{GDBN} for help
1142 @node Command Syntax
1143 @section Command syntax
1145 A @value{GDBN} command is a single line of input. There is no limit on
1146 how long it can be. It starts with a command name, which is followed by
1147 arguments whose meaning depends on the command name. For example, the
1148 command @code{step} accepts an argument which is the number of times to
1149 step, as in @samp{step 5}. You can also use the @code{step} command
1150 with no arguments. Some commands do not allow any arguments.
1152 @cindex abbreviation
1153 @value{GDBN} command names may always be truncated if that abbreviation is
1154 unambiguous. Other possible command abbreviations are listed in the
1155 documentation for individual commands. In some cases, even ambiguous
1156 abbreviations are allowed; for example, @code{s} is specially defined as
1157 equivalent to @code{step} even though there are other commands whose
1158 names start with @code{s}. You can test abbreviations by using them as
1159 arguments to the @code{help} command.
1161 @cindex repeating commands
1162 @kindex RET @r{(repeat last command)}
1163 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1164 repeat the previous command. Certain commands (for example, @code{run})
1165 will not repeat this way; these are commands whose unintentional
1166 repetition might cause trouble and which you are unlikely to want to
1169 The @code{list} and @code{x} commands, when you repeat them with
1170 @key{RET}, construct new arguments rather than repeating
1171 exactly as typed. This permits easy scanning of source or memory.
1173 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1174 output, in a way similar to the common utility @code{more}
1175 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1176 @key{RET} too many in this situation, @value{GDBN} disables command
1177 repetition after any command that generates this sort of display.
1179 @kindex # @r{(a comment)}
1181 Any text from a @kbd{#} to the end of the line is a comment; it does
1182 nothing. This is useful mainly in command files (@pxref{Command
1183 Files,,Command files}).
1186 @section Command completion
1189 @cindex word completion
1190 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1191 only one possibility; it can also show you what the valid possibilities
1192 are for the next word in a command, at any time. This works for @value{GDBN}
1193 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1195 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1196 of a word. If there is only one possibility, @value{GDBN} fills in the
1197 word, and waits for you to finish the command (or press @key{RET} to
1198 enter it). For example, if you type
1200 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1201 @c complete accuracy in these examples; space introduced for clarity.
1202 @c If texinfo enhancements make it unnecessary, it would be nice to
1203 @c replace " @key" by "@key" in the following...
1205 (@value{GDBP}) info bre @key{TAB}
1209 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1210 the only @code{info} subcommand beginning with @samp{bre}:
1213 (@value{GDBP}) info breakpoints
1217 You can either press @key{RET} at this point, to run the @code{info
1218 breakpoints} command, or backspace and enter something else, if
1219 @samp{breakpoints} does not look like the command you expected. (If you
1220 were sure you wanted @code{info breakpoints} in the first place, you
1221 might as well just type @key{RET} immediately after @samp{info bre},
1222 to exploit command abbreviations rather than command completion).
1224 If there is more than one possibility for the next word when you press
1225 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1226 characters and try again, or just press @key{TAB} a second time;
1227 @value{GDBN} displays all the possible completions for that word. For
1228 example, you might want to set a breakpoint on a subroutine whose name
1229 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1230 just sounds the bell. Typing @key{TAB} again displays all the
1231 function names in your program that begin with those characters, for
1235 (@value{GDBP}) b make_ @key{TAB}
1236 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1237 make_a_section_from_file make_environ
1238 make_abs_section make_function_type
1239 make_blockvector make_pointer_type
1240 make_cleanup make_reference_type
1241 make_command make_symbol_completion_list
1242 (@value{GDBP}) b make_
1246 After displaying the available possibilities, @value{GDBN} copies your
1247 partial input (@samp{b make_} in the example) so you can finish the
1250 If you just want to see the list of alternatives in the first place, you
1251 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1252 means @kbd{@key{META} ?}. You can type this either by holding down a
1253 key designated as the @key{META} shift on your keyboard (if there is
1254 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1256 @cindex quotes in commands
1257 @cindex completion of quoted strings
1258 Sometimes the string you need, while logically a ``word'', may contain
1259 parentheses or other characters that @value{GDBN} normally excludes from
1260 its notion of a word. To permit word completion to work in this
1261 situation, you may enclose words in @code{'} (single quote marks) in
1262 @value{GDBN} commands.
1264 The most likely situation where you might need this is in typing the
1265 name of a C++ function. This is because C++ allows function overloading
1266 (multiple definitions of the same function, distinguished by argument
1267 type). For example, when you want to set a breakpoint you may need to
1268 distinguish whether you mean the version of @code{name} that takes an
1269 @code{int} parameter, @code{name(int)}, or the version that takes a
1270 @code{float} parameter, @code{name(float)}. To use the word-completion
1271 facilities in this situation, type a single quote @code{'} at the
1272 beginning of the function name. This alerts @value{GDBN} that it may need to
1273 consider more information than usual when you press @key{TAB} or
1274 @kbd{M-?} to request word completion:
1277 (@value{GDBP}) b 'bubble( @kbd{M-?}
1278 bubble(double,double) bubble(int,int)
1279 (@value{GDBP}) b 'bubble(
1282 In some cases, @value{GDBN} can tell that completing a name requires using
1283 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1284 completing as much as it can) if you do not type the quote in the first
1288 (@value{GDBP}) b bub @key{TAB}
1289 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1290 (@value{GDBP}) b 'bubble(
1294 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1295 you have not yet started typing the argument list when you ask for
1296 completion on an overloaded symbol.
1298 For more information about overloaded functions, see @ref{C plus plus
1299 expressions, ,C++ expressions}. You can use the command @code{set
1300 overload-resolution off} to disable overload resolution;
1301 see @ref{Debugging C plus plus, ,@value{GDBN} features for C++}.
1305 @section Getting help
1306 @cindex online documentation
1309 You can always ask @value{GDBN} itself for information on its commands,
1310 using the command @code{help}.
1313 @kindex h @r{(@code{help})}
1316 You can use @code{help} (abbreviated @code{h}) with no arguments to
1317 display a short list of named classes of commands:
1321 List of classes of commands:
1323 aliases -- Aliases of other commands
1324 breakpoints -- Making program stop at certain points
1325 data -- Examining data
1326 files -- Specifying and examining files
1327 internals -- Maintenance commands
1328 obscure -- Obscure features
1329 running -- Running the program
1330 stack -- Examining the stack
1331 status -- Status inquiries
1332 support -- Support facilities
1333 tracepoints -- Tracing of program execution without@*
1334 stopping the program
1335 user-defined -- User-defined commands
1337 Type "help" followed by a class name for a list of
1338 commands in that class.
1339 Type "help" followed by command name for full
1341 Command name abbreviations are allowed if unambiguous.
1344 @c the above line break eliminates huge line overfull...
1346 @item help @var{class}
1347 Using one of the general help classes as an argument, you can get a
1348 list of the individual commands in that class. For example, here is the
1349 help display for the class @code{status}:
1352 (@value{GDBP}) help status
1357 @c Line break in "show" line falsifies real output, but needed
1358 @c to fit in smallbook page size.
1359 info -- Generic command for showing things
1360 about the program being debugged
1361 show -- Generic command for showing things
1364 Type "help" followed by command name for full
1366 Command name abbreviations are allowed if unambiguous.
1370 @item help @var{command}
1371 With a command name as @code{help} argument, @value{GDBN} displays a
1372 short paragraph on how to use that command.
1375 @item apropos @var{args}
1376 The @code{apropos @var{args}} command searches through all of the @value{GDBN}
1377 commands, and their documentation, for the regular expression specified in
1378 @var{args}. It prints out all matches found. For example:
1384 @noindent results in:
1388 set symbol-reloading -- Set dynamic symbol table reloading
1389 multiple times in one run
1390 show symbol-reloading -- Show dynamic symbol table reloading
1391 multiple times in one run
1396 @item complete @var{args}
1397 The @code{complete @var{args}} command lists all the possible completions
1398 for the beginning of a command. Use @var{args} to specify the beginning of the
1399 command you want completed. For example:
1405 @noindent results in:
1416 @noindent This is intended for use by @sc{gnu} Emacs.
1419 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1420 and @code{show} to inquire about the state of your program, or the state
1421 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1422 manual introduces each of them in the appropriate context. The listings
1423 under @code{info} and under @code{show} in the Index point to
1424 all the sub-commands. @xref{Index}.
1429 @kindex i @r{(@code{info})}
1431 This command (abbreviated @code{i}) is for describing the state of your
1432 program. For example, you can list the arguments given to your program
1433 with @code{info args}, list the registers currently in use with @code{info
1434 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1435 You can get a complete list of the @code{info} sub-commands with
1436 @w{@code{help info}}.
1440 You can assign the result of an expression to an environment variable with
1441 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1442 @code{set prompt $}.
1446 In contrast to @code{info}, @code{show} is for describing the state of
1447 @value{GDBN} itself.
1448 You can change most of the things you can @code{show}, by using the
1449 related command @code{set}; for example, you can control what number
1450 system is used for displays with @code{set radix}, or simply inquire
1451 which is currently in use with @code{show radix}.
1454 To display all the settable parameters and their current
1455 values, you can use @code{show} with no arguments; you may also use
1456 @code{info set}. Both commands produce the same display.
1457 @c FIXME: "info set" violates the rule that "info" is for state of
1458 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1459 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1463 Here are three miscellaneous @code{show} subcommands, all of which are
1464 exceptional in lacking corresponding @code{set} commands:
1467 @kindex show version
1468 @cindex version number
1470 Show what version of @value{GDBN} is running. You should include this
1471 information in @value{GDBN} bug-reports. If multiple versions of
1472 @value{GDBN} are in use at your site, you may need to determine which
1473 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1474 commands are introduced, and old ones may wither away. Also, many
1475 system vendors ship variant versions of @value{GDBN}, and there are
1476 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1477 The version number is the same as the one announced when you start
1480 @kindex show copying
1482 Display information about permission for copying @value{GDBN}.
1484 @kindex show warranty
1486 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1487 if your version of @value{GDBN} comes with one.
1492 @chapter Running Programs Under @value{GDBN}
1494 When you run a program under @value{GDBN}, you must first generate
1495 debugging information when you compile it.
1497 You may start @value{GDBN} with its arguments, if any, in an environment
1498 of your choice. If you are doing native debugging, you may redirect
1499 your program's input and output, debug an already running process, or
1500 kill a child process.
1503 * Compilation:: Compiling for debugging
1504 * Starting:: Starting your program
1505 * Arguments:: Your program's arguments
1506 * Environment:: Your program's environment
1508 * Working Directory:: Your program's working directory
1509 * Input/Output:: Your program's input and output
1510 * Attach:: Debugging an already-running process
1511 * Kill Process:: Killing the child process
1513 * Threads:: Debugging programs with multiple threads
1514 * Processes:: Debugging programs with multiple processes
1518 @section Compiling for debugging
1520 In order to debug a program effectively, you need to generate
1521 debugging information when you compile it. This debugging information
1522 is stored in the object file; it describes the data type of each
1523 variable or function and the correspondence between source line numbers
1524 and addresses in the executable code.
1526 To request debugging information, specify the @samp{-g} option when you run
1529 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1530 options together. Using those compilers, you cannot generate optimized
1531 executables containing debugging information.
1533 @value{NGCC}, the @sc{gnu} C compiler, supports @samp{-g} with or
1534 without @samp{-O}, making it possible to debug optimized code. We
1535 recommend that you @emph{always} use @samp{-g} whenever you compile a
1536 program. You may think your program is correct, but there is no sense
1537 in pushing your luck.
1539 @cindex optimized code, debugging
1540 @cindex debugging optimized code
1541 When you debug a program compiled with @samp{-g -O}, remember that the
1542 optimizer is rearranging your code; the debugger shows you what is
1543 really there. Do not be too surprised when the execution path does not
1544 exactly match your source file! An extreme example: if you define a
1545 variable, but never use it, @value{GDBN} never sees that
1546 variable---because the compiler optimizes it out of existence.
1548 Some things do not work as well with @samp{-g -O} as with just
1549 @samp{-g}, particularly on machines with instruction scheduling. If in
1550 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1551 please report it to us as a bug (including a test case!).
1553 Older versions of the @sc{gnu} C compiler permitted a variant option
1554 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1555 format; if your @sc{gnu} C compiler has this option, do not use it.
1559 @section Starting your program
1565 @kindex r @r{(@code{run})}
1568 Use the @code{run} command to start your program under @value{GDBN}.
1569 You must first specify the program name (except on VxWorks) with an
1570 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1571 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1572 (@pxref{Files, ,Commands to specify files}).
1576 If you are running your program in an execution environment that
1577 supports processes, @code{run} creates an inferior process and makes
1578 that process run your program. (In environments without processes,
1579 @code{run} jumps to the start of your program.)
1581 The execution of a program is affected by certain information it
1582 receives from its superior. @value{GDBN} provides ways to specify this
1583 information, which you must do @emph{before} starting your program. (You
1584 can change it after starting your program, but such changes only affect
1585 your program the next time you start it.) This information may be
1586 divided into four categories:
1589 @item The @emph{arguments.}
1590 Specify the arguments to give your program as the arguments of the
1591 @code{run} command. If a shell is available on your target, the shell
1592 is used to pass the arguments, so that you may use normal conventions
1593 (such as wildcard expansion or variable substitution) in describing
1595 In Unix systems, you can control which shell is used with the
1596 @code{SHELL} environment variable.
1597 @xref{Arguments, ,Your program's arguments}.
1599 @item The @emph{environment.}
1600 Your program normally inherits its environment from @value{GDBN}, but you can
1601 use the @value{GDBN} commands @code{set environment} and @code{unset
1602 environment} to change parts of the environment that affect
1603 your program. @xref{Environment, ,Your program's environment}.
1605 @item The @emph{working directory.}
1606 Your program inherits its working directory from @value{GDBN}. You can set
1607 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1608 @xref{Working Directory, ,Your program's working directory}.
1610 @item The @emph{standard input and output.}
1611 Your program normally uses the same device for standard input and
1612 standard output as @value{GDBN} is using. You can redirect input and output
1613 in the @code{run} command line, or you can use the @code{tty} command to
1614 set a different device for your program.
1615 @xref{Input/Output, ,Your program's input and output}.
1618 @emph{Warning:} While input and output redirection work, you cannot use
1619 pipes to pass the output of the program you are debugging to another
1620 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1624 When you issue the @code{run} command, your program begins to execute
1625 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1626 of how to arrange for your program to stop. Once your program has
1627 stopped, you may call functions in your program, using the @code{print}
1628 or @code{call} commands. @xref{Data, ,Examining Data}.
1630 If the modification time of your symbol file has changed since the last
1631 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1632 table, and reads it again. When it does this, @value{GDBN} tries to retain
1633 your current breakpoints.
1636 @section Your program's arguments
1638 @cindex arguments (to your program)
1639 The arguments to your program can be specified by the arguments of the
1641 They are passed to a shell, which expands wildcard characters and
1642 performs redirection of I/O, and thence to your program. Your
1643 @code{SHELL} environment variable (if it exists) specifies what shell
1644 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
1645 the default shell (@file{/bin/sh} on Unix).
1647 On non-Unix systems, the program is usually invoked directly by
1648 @value{GDBN}, which emulates I/O redirection via the appropriate system
1649 calls, and the wildcard characters are expanded by the startup code of
1650 the program, not by the shell.
1652 @code{run} with no arguments uses the same arguments used by the previous
1653 @code{run}, or those set by the @code{set args} command.
1658 Specify the arguments to be used the next time your program is run. If
1659 @code{set args} has no arguments, @code{run} executes your program
1660 with no arguments. Once you have run your program with arguments,
1661 using @code{set args} before the next @code{run} is the only way to run
1662 it again without arguments.
1666 Show the arguments to give your program when it is started.
1670 @section Your program's environment
1672 @cindex environment (of your program)
1673 The @dfn{environment} consists of a set of environment variables and
1674 their values. Environment variables conventionally record such things as
1675 your user name, your home directory, your terminal type, and your search
1676 path for programs to run. Usually you set up environment variables with
1677 the shell and they are inherited by all the other programs you run. When
1678 debugging, it can be useful to try running your program with a modified
1679 environment without having to start @value{GDBN} over again.
1683 @item path @var{directory}
1684 Add @var{directory} to the front of the @code{PATH} environment variable
1685 (the search path for executables), for both @value{GDBN} and your program.
1686 You may specify several directory names, separated by whitespace or by a
1687 system-dependent separator character (@samp{:} on Unix, @samp{;} on
1688 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
1689 is moved to the front, so it is searched sooner.
1691 You can use the string @samp{$cwd} to refer to whatever is the current
1692 working directory at the time @value{GDBN} searches the path. If you
1693 use @samp{.} instead, it refers to the directory where you executed the
1694 @code{path} command. @value{GDBN} replaces @samp{.} in the
1695 @var{directory} argument (with the current path) before adding
1696 @var{directory} to the search path.
1697 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1698 @c document that, since repeating it would be a no-op.
1702 Display the list of search paths for executables (the @code{PATH}
1703 environment variable).
1705 @kindex show environment
1706 @item show environment @r{[}@var{varname}@r{]}
1707 Print the value of environment variable @var{varname} to be given to
1708 your program when it starts. If you do not supply @var{varname},
1709 print the names and values of all environment variables to be given to
1710 your program. You can abbreviate @code{environment} as @code{env}.
1712 @kindex set environment
1713 @item set environment @var{varname} @r{[}=@var{value}@r{]}
1714 Set environment variable @var{varname} to @var{value}. The value
1715 changes for your program only, not for @value{GDBN} itself. @var{value} may
1716 be any string; the values of environment variables are just strings, and
1717 any interpretation is supplied by your program itself. The @var{value}
1718 parameter is optional; if it is eliminated, the variable is set to a
1720 @c "any string" here does not include leading, trailing
1721 @c blanks. Gnu asks: does anyone care?
1723 For example, this command:
1730 tells the debugged program, when subsequently run, that its user is named
1731 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1732 are not actually required.)
1734 @kindex unset environment
1735 @item unset environment @var{varname}
1736 Remove variable @var{varname} from the environment to be passed to your
1737 program. This is different from @samp{set env @var{varname} =};
1738 @code{unset environment} removes the variable from the environment,
1739 rather than assigning it an empty value.
1742 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
1744 by your @code{SHELL} environment variable if it exists (or
1745 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1746 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1747 @file{.bashrc} for BASH---any variables you set in that file affect
1748 your program. You may wish to move setting of environment variables to
1749 files that are only run when you sign on, such as @file{.login} or
1752 @node Working Directory
1753 @section Your program's working directory
1755 @cindex working directory (of your program)
1756 Each time you start your program with @code{run}, it inherits its
1757 working directory from the current working directory of @value{GDBN}.
1758 The @value{GDBN} working directory is initially whatever it inherited
1759 from its parent process (typically the shell), but you can specify a new
1760 working directory in @value{GDBN} with the @code{cd} command.
1762 The @value{GDBN} working directory also serves as a default for the commands
1763 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1768 @item cd @var{directory}
1769 Set the @value{GDBN} working directory to @var{directory}.
1773 Print the @value{GDBN} working directory.
1777 @section Your program's input and output
1782 By default, the program you run under @value{GDBN} does input and output to
1783 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
1784 to its own terminal modes to interact with you, but it records the terminal
1785 modes your program was using and switches back to them when you continue
1786 running your program.
1789 @kindex info terminal
1791 Displays information recorded by @value{GDBN} about the terminal modes your
1795 You can redirect your program's input and/or output using shell
1796 redirection with the @code{run} command. For example,
1803 starts your program, diverting its output to the file @file{outfile}.
1806 @cindex controlling terminal
1807 Another way to specify where your program should do input and output is
1808 with the @code{tty} command. This command accepts a file name as
1809 argument, and causes this file to be the default for future @code{run}
1810 commands. It also resets the controlling terminal for the child
1811 process, for future @code{run} commands. For example,
1818 directs that processes started with subsequent @code{run} commands
1819 default to do input and output on the terminal @file{/dev/ttyb} and have
1820 that as their controlling terminal.
1822 An explicit redirection in @code{run} overrides the @code{tty} command's
1823 effect on the input/output device, but not its effect on the controlling
1826 When you use the @code{tty} command or redirect input in the @code{run}
1827 command, only the input @emph{for your program} is affected. The input
1828 for @value{GDBN} still comes from your terminal.
1831 @section Debugging an already-running process
1836 @item attach @var{process-id}
1837 This command attaches to a running process---one that was started
1838 outside @value{GDBN}. (@code{info files} shows your active
1839 targets.) The command takes as argument a process ID. The usual way to
1840 find out the process-id of a Unix process is with the @code{ps} utility,
1841 or with the @samp{jobs -l} shell command.
1843 @code{attach} does not repeat if you press @key{RET} a second time after
1844 executing the command.
1847 To use @code{attach}, your program must be running in an environment
1848 which supports processes; for example, @code{attach} does not work for
1849 programs on bare-board targets that lack an operating system. You must
1850 also have permission to send the process a signal.
1852 When you use @code{attach}, the debugger finds the program running in
1853 the process first by looking in the current working directory, then (if
1854 the program is not found) by using the source file search path
1855 (@pxref{Source Path, ,Specifying source directories}). You can also use
1856 the @code{file} command to load the program. @xref{Files, ,Commands to
1859 The first thing @value{GDBN} does after arranging to debug the specified
1860 process is to stop it. You can examine and modify an attached process
1861 with all the @value{GDBN} commands that are ordinarily available when
1862 you start processes with @code{run}. You can insert breakpoints; you
1863 can step and continue; you can modify storage. If you would rather the
1864 process continue running, you may use the @code{continue} command after
1865 attaching @value{GDBN} to the process.
1870 When you have finished debugging the attached process, you can use the
1871 @code{detach} command to release it from @value{GDBN} control. Detaching
1872 the process continues its execution. After the @code{detach} command,
1873 that process and @value{GDBN} become completely independent once more, and you
1874 are ready to @code{attach} another process or start one with @code{run}.
1875 @code{detach} does not repeat if you press @key{RET} again after
1876 executing the command.
1879 If you exit @value{GDBN} or use the @code{run} command while you have an
1880 attached process, you kill that process. By default, @value{GDBN} asks
1881 for confirmation if you try to do either of these things; you can
1882 control whether or not you need to confirm by using the @code{set
1883 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
1887 @section Killing the child process
1892 Kill the child process in which your program is running under @value{GDBN}.
1895 This command is useful if you wish to debug a core dump instead of a
1896 running process. @value{GDBN} ignores any core dump file while your program
1899 On some operating systems, a program cannot be executed outside @value{GDBN}
1900 while you have breakpoints set on it inside @value{GDBN}. You can use the
1901 @code{kill} command in this situation to permit running your program
1902 outside the debugger.
1904 The @code{kill} command is also useful if you wish to recompile and
1905 relink your program, since on many systems it is impossible to modify an
1906 executable file while it is running in a process. In this case, when you
1907 next type @code{run}, @value{GDBN} notices that the file has changed, and
1908 reads the symbol table again (while trying to preserve your current
1909 breakpoint settings).
1912 @section Debugging programs with multiple threads
1914 @cindex threads of execution
1915 @cindex multiple threads
1916 @cindex switching threads
1917 In some operating systems, such as HP-UX and Solaris, a single program
1918 may have more than one @dfn{thread} of execution. The precise semantics
1919 of threads differ from one operating system to another, but in general
1920 the threads of a single program are akin to multiple processes---except
1921 that they share one address space (that is, they can all examine and
1922 modify the same variables). On the other hand, each thread has its own
1923 registers and execution stack, and perhaps private memory.
1925 @value{GDBN} provides these facilities for debugging multi-thread
1929 @item automatic notification of new threads
1930 @item @samp{thread @var{threadno}}, a command to switch among threads
1931 @item @samp{info threads}, a command to inquire about existing threads
1932 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
1933 a command to apply a command to a list of threads
1934 @item thread-specific breakpoints
1938 @emph{Warning:} These facilities are not yet available on every
1939 @value{GDBN} configuration where the operating system supports threads.
1940 If your @value{GDBN} does not support threads, these commands have no
1941 effect. For example, a system without thread support shows no output
1942 from @samp{info threads}, and always rejects the @code{thread} command,
1946 (@value{GDBP}) info threads
1947 (@value{GDBP}) thread 1
1948 Thread ID 1 not known. Use the "info threads" command to
1949 see the IDs of currently known threads.
1951 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
1952 @c doesn't support threads"?
1955 @cindex focus of debugging
1956 @cindex current thread
1957 The @value{GDBN} thread debugging facility allows you to observe all
1958 threads while your program runs---but whenever @value{GDBN} takes
1959 control, one thread in particular is always the focus of debugging.
1960 This thread is called the @dfn{current thread}. Debugging commands show
1961 program information from the perspective of the current thread.
1963 @cindex @code{New} @var{systag} message
1964 @cindex thread identifier (system)
1965 @c FIXME-implementors!! It would be more helpful if the [New...] message
1966 @c included GDB's numeric thread handle, so you could just go to that
1967 @c thread without first checking `info threads'.
1968 Whenever @value{GDBN} detects a new thread in your program, it displays
1969 the target system's identification for the thread with a message in the
1970 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
1971 whose form varies depending on the particular system. For example, on
1972 LynxOS, you might see
1975 [New process 35 thread 27]
1979 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
1980 the @var{systag} is simply something like @samp{process 368}, with no
1983 @c FIXME!! (1) Does the [New...] message appear even for the very first
1984 @c thread of a program, or does it only appear for the
1985 @c second---i.e., when it becomes obvious we have a multithread
1987 @c (2) *Is* there necessarily a first thread always? Or do some
1988 @c multithread systems permit starting a program with multiple
1989 @c threads ab initio?
1991 @cindex thread number
1992 @cindex thread identifier (GDB)
1993 For debugging purposes, @value{GDBN} associates its own thread
1994 number---always a single integer---with each thread in your program.
1997 @kindex info threads
1999 Display a summary of all threads currently in your
2000 program. @value{GDBN} displays for each thread (in this order):
2003 @item the thread number assigned by @value{GDBN}
2005 @item the target system's thread identifier (@var{systag})
2007 @item the current stack frame summary for that thread
2011 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2012 indicates the current thread.
2016 @c end table here to get a little more width for example
2019 (@value{GDBP}) info threads
2020 3 process 35 thread 27 0x34e5 in sigpause ()
2021 2 process 35 thread 23 0x34e5 in sigpause ()
2022 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2028 @cindex thread number
2029 @cindex thread identifier (GDB)
2030 For debugging purposes, @value{GDBN} associates its own thread
2031 number---a small integer assigned in thread-creation order---with each
2032 thread in your program.
2034 @cindex @code{New} @var{systag} message, on HP-UX
2035 @cindex thread identifier (system), on HP-UX
2036 @c FIXME-implementors!! It would be more helpful if the [New...] message
2037 @c included GDB's numeric thread handle, so you could just go to that
2038 @c thread without first checking `info threads'.
2039 Whenever @value{GDBN} detects a new thread in your program, it displays
2040 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2041 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2042 whose form varies depending on the particular system. For example, on
2046 [New thread 2 (system thread 26594)]
2050 when @value{GDBN} notices a new thread.
2053 @kindex info threads
2055 Display a summary of all threads currently in your
2056 program. @value{GDBN} displays for each thread (in this order):
2059 @item the thread number assigned by @value{GDBN}
2061 @item the target system's thread identifier (@var{systag})
2063 @item the current stack frame summary for that thread
2067 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2068 indicates the current thread.
2072 @c end table here to get a little more width for example
2075 (@value{GDBP}) info threads
2076 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2078 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2079 from /usr/lib/libc.2
2080 1 system thread 27905 0x7b003498 in _brk () \@*
2081 from /usr/lib/libc.2
2085 @kindex thread @var{threadno}
2086 @item thread @var{threadno}
2087 Make thread number @var{threadno} the current thread. The command
2088 argument @var{threadno} is the internal @value{GDBN} thread number, as
2089 shown in the first field of the @samp{info threads} display.
2090 @value{GDBN} responds by displaying the system identifier of the thread
2091 you selected, and its current stack frame summary:
2094 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2095 (@value{GDBP}) thread 2
2096 [Switching to process 35 thread 23]
2097 0x34e5 in sigpause ()
2101 As with the @samp{[New @dots{}]} message, the form of the text after
2102 @samp{Switching to} depends on your system's conventions for identifying
2105 @kindex thread apply
2106 @item thread apply [@var{threadno}] [@var{all}] @var{args}
2107 The @code{thread apply} command allows you to apply a command to one or
2108 more threads. Specify the numbers of the threads that you want affected
2109 with the command argument @var{threadno}. @var{threadno} is the internal
2110 @value{GDBN} thread number, as shown in the first field of the @samp{info
2111 threads} display. To apply a command to all threads, use
2112 @code{thread apply all} @var{args}.
2115 @cindex automatic thread selection
2116 @cindex switching threads automatically
2117 @cindex threads, automatic switching
2118 Whenever @value{GDBN} stops your program, due to a breakpoint or a
2119 signal, it automatically selects the thread where that breakpoint or
2120 signal happened. @value{GDBN} alerts you to the context switch with a
2121 message of the form @samp{[Switching to @var{systag}]} to identify the
2124 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
2125 more information about how @value{GDBN} behaves when you stop and start
2126 programs with multiple threads.
2128 @xref{Set Watchpoints,,Setting watchpoints}, for information about
2129 watchpoints in programs with multiple threads.
2132 @section Debugging programs with multiple processes
2134 @cindex fork, debugging programs which call
2135 @cindex multiple processes
2136 @cindex processes, multiple
2137 On most systems, @value{GDBN} has no special support for debugging
2138 programs which create additional processes using the @code{fork}
2139 function. When a program forks, @value{GDBN} will continue to debug the
2140 parent process and the child process will run unimpeded. If you have
2141 set a breakpoint in any code which the child then executes, the child
2142 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2143 will cause it to terminate.
2145 However, if you want to debug the child process there is a workaround
2146 which isn't too painful. Put a call to @code{sleep} in the code which
2147 the child process executes after the fork. It may be useful to sleep
2148 only if a certain environment variable is set, or a certain file exists,
2149 so that the delay need not occur when you don't want to run @value{GDBN}
2150 on the child. While the child is sleeping, use the @code{ps} program to
2151 get its process ID. Then tell @value{GDBN} (a new invocation of
2152 @value{GDBN} if you are also debugging the parent process) to attach to
2153 the child process (@pxref{Attach}). From that point on you can debug
2154 the child process just like any other process which you attached to.
2156 On HP-UX (11.x and later only?), @value{GDBN} provides support for
2157 debugging programs that create additional processes using the
2158 @code{fork} or @code{vfork} function.
2160 By default, when a program forks, @value{GDBN} will continue to debug
2161 the parent process and the child process will run unimpeded.
2163 If you want to follow the child process instead of the parent process,
2164 use the command @w{@code{set follow-fork-mode}}.
2167 @kindex set follow-fork-mode
2168 @item set follow-fork-mode @var{mode}
2169 Set the debugger response to a program call of @code{fork} or
2170 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2171 process. The @var{mode} can be:
2175 The original process is debugged after a fork. The child process runs
2176 unimpeded. This is the default.
2179 The new process is debugged after a fork. The parent process runs
2183 The debugger will ask for one of the above choices.
2186 @item show follow-fork-mode
2187 Display the current debugger response to a @code{fork} or @code{vfork} call.
2190 If you ask to debug a child process and a @code{vfork} is followed by an
2191 @code{exec}, @value{GDBN} executes the new target up to the first
2192 breakpoint in the new target. If you have a breakpoint set on
2193 @code{main} in your original program, the breakpoint will also be set on
2194 the child process's @code{main}.
2196 When a child process is spawned by @code{vfork}, you cannot debug the
2197 child or parent until an @code{exec} call completes.
2199 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2200 call executes, the new target restarts. To restart the parent process,
2201 use the @code{file} command with the parent executable name as its
2204 You can use the @code{catch} command to make @value{GDBN} stop whenever
2205 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2206 Catchpoints, ,Setting catchpoints}.
2209 @chapter Stopping and Continuing
2211 The principal purposes of using a debugger are so that you can stop your
2212 program before it terminates; or so that, if your program runs into
2213 trouble, you can investigate and find out why.
2215 Inside @value{GDBN}, your program may stop for any of several reasons,
2216 such as a signal, a breakpoint, or reaching a new line after a
2217 @value{GDBN} command such as @code{step}. You may then examine and
2218 change variables, set new breakpoints or remove old ones, and then
2219 continue execution. Usually, the messages shown by @value{GDBN} provide
2220 ample explanation of the status of your program---but you can also
2221 explicitly request this information at any time.
2224 @kindex info program
2226 Display information about the status of your program: whether it is
2227 running or not, what process it is, and why it stopped.
2231 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2232 * Continuing and Stepping:: Resuming execution
2234 * Thread Stops:: Stopping and starting multi-thread programs
2238 @section Breakpoints, watchpoints, and catchpoints
2241 A @dfn{breakpoint} makes your program stop whenever a certain point in
2242 the program is reached. For each breakpoint, you can add conditions to
2243 control in finer detail whether your program stops. You can set
2244 breakpoints with the @code{break} command and its variants (@pxref{Set
2245 Breaks, ,Setting breakpoints}), to specify the place where your program
2246 should stop by line number, function name or exact address in the
2249 In HP-UX, SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can set
2250 breakpoints in shared libraries before the executable is run. There is
2251 a minor limitation on HP-UX systems: you must wait until the executable
2252 is run in order to set breakpoints in shared library routines that are
2253 not called directly by the program (for example, routines that are
2254 arguments in a @code{pthread_create} call).
2257 @cindex memory tracing
2258 @cindex breakpoint on memory address
2259 @cindex breakpoint on variable modification
2260 A @dfn{watchpoint} is a special breakpoint that stops your program
2261 when the value of an expression changes. You must use a different
2262 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2263 watchpoints}), but aside from that, you can manage a watchpoint like
2264 any other breakpoint: you enable, disable, and delete both breakpoints
2265 and watchpoints using the same commands.
2267 You can arrange to have values from your program displayed automatically
2268 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2272 @cindex breakpoint on events
2273 A @dfn{catchpoint} is another special breakpoint that stops your program
2274 when a certain kind of event occurs, such as the throwing of a C++
2275 exception or the loading of a library. As with watchpoints, you use a
2276 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2277 catchpoints}), but aside from that, you can manage a catchpoint like any
2278 other breakpoint. (To stop when your program receives a signal, use the
2279 @code{handle} command; see @ref{Signals, ,Signals}.)
2281 @cindex breakpoint numbers
2282 @cindex numbers for breakpoints
2283 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2284 catchpoint when you create it; these numbers are successive integers
2285 starting with one. In many of the commands for controlling various
2286 features of breakpoints you use the breakpoint number to say which
2287 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2288 @dfn{disabled}; if disabled, it has no effect on your program until you
2291 @cindex breakpoint ranges
2292 @cindex ranges of breakpoints
2293 Some @value{GDBN} commands accept a range of breakpoints on which to
2294 operate. A breakpoint range is either a single breakpoint number, like
2295 @samp{5}, or two such numbers, in increasing order, separated by a
2296 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
2297 all breakpoint in that range are operated on.
2300 * Set Breaks:: Setting breakpoints
2301 * Set Watchpoints:: Setting watchpoints
2302 * Set Catchpoints:: Setting catchpoints
2303 * Delete Breaks:: Deleting breakpoints
2304 * Disabling:: Disabling breakpoints
2305 * Conditions:: Break conditions
2306 * Break Commands:: Breakpoint command lists
2307 * Breakpoint Menus:: Breakpoint menus
2308 * Error in Breakpoints:: ``Cannot insert breakpoints''
2312 @subsection Setting breakpoints
2314 @c FIXME LMB what does GDB do if no code on line of breakpt?
2315 @c consider in particular declaration with/without initialization.
2317 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2320 @kindex b @r{(@code{break})}
2321 @vindex $bpnum@r{, convenience variable}
2322 @cindex latest breakpoint
2323 Breakpoints are set with the @code{break} command (abbreviated
2324 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2325 number of the breakpoints you've set most recently; see @ref{Convenience
2326 Vars,, Convenience variables}, for a discussion of what you can do with
2327 convenience variables.
2329 You have several ways to say where the breakpoint should go.
2332 @item break @var{function}
2333 Set a breakpoint at entry to function @var{function}.
2334 When using source languages that permit overloading of symbols, such as
2335 C++, @var{function} may refer to more than one possible place to break.
2336 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2338 @item break +@var{offset}
2339 @itemx break -@var{offset}
2340 Set a breakpoint some number of lines forward or back from the position
2341 at which execution stopped in the currently selected @dfn{stack frame}.
2342 (@xref{Frames, ,Frames}, for a description of stack frames.)
2344 @item break @var{linenum}
2345 Set a breakpoint at line @var{linenum} in the current source file.
2346 The current source file is the last file whose source text was printed.
2347 The breakpoint will stop your program just before it executes any of the
2350 @item break @var{filename}:@var{linenum}
2351 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2353 @item break @var{filename}:@var{function}
2354 Set a breakpoint at entry to function @var{function} found in file
2355 @var{filename}. Specifying a file name as well as a function name is
2356 superfluous except when multiple files contain similarly named
2359 @item break *@var{address}
2360 Set a breakpoint at address @var{address}. You can use this to set
2361 breakpoints in parts of your program which do not have debugging
2362 information or source files.
2365 When called without any arguments, @code{break} sets a breakpoint at
2366 the next instruction to be executed in the selected stack frame
2367 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2368 innermost, this makes your program stop as soon as control
2369 returns to that frame. This is similar to the effect of a
2370 @code{finish} command in the frame inside the selected frame---except
2371 that @code{finish} does not leave an active breakpoint. If you use
2372 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2373 the next time it reaches the current location; this may be useful
2376 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2377 least one instruction has been executed. If it did not do this, you
2378 would be unable to proceed past a breakpoint without first disabling the
2379 breakpoint. This rule applies whether or not the breakpoint already
2380 existed when your program stopped.
2382 @item break @dots{} if @var{cond}
2383 Set a breakpoint with condition @var{cond}; evaluate the expression
2384 @var{cond} each time the breakpoint is reached, and stop only if the
2385 value is nonzero---that is, if @var{cond} evaluates as true.
2386 @samp{@dots{}} stands for one of the possible arguments described
2387 above (or no argument) specifying where to break. @xref{Conditions,
2388 ,Break conditions}, for more information on breakpoint conditions.
2391 @item tbreak @var{args}
2392 Set a breakpoint enabled only for one stop. @var{args} are the
2393 same as for the @code{break} command, and the breakpoint is set in the same
2394 way, but the breakpoint is automatically deleted after the first time your
2395 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2398 @item hbreak @var{args}
2399 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2400 @code{break} command and the breakpoint is set in the same way, but the
2401 breakpoint requires hardware support and some target hardware may not
2402 have this support. The main purpose of this is EPROM/ROM code
2403 debugging, so you can set a breakpoint at an instruction without
2404 changing the instruction. This can be used with the new trap-generation
2405 provided by SPARClite DSU and some x86-based targets. These targets
2406 will generate traps when a program accesses some data or instruction
2407 address that is assigned to the debug registers. However the hardware
2408 breakpoint registers can take a limited number of breakpoints. For
2409 example, on the DSU, only two data breakpoints can be set at a time, and
2410 @value{GDBN} will reject this command if more than two are used. Delete
2411 or disable unused hardware breakpoints before setting new ones
2412 (@pxref{Disabling, ,Disabling}). @xref{Conditions, ,Break conditions}.
2415 @item thbreak @var{args}
2416 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2417 are the same as for the @code{hbreak} command and the breakpoint is set in
2418 the same way. However, like the @code{tbreak} command,
2419 the breakpoint is automatically deleted after the
2420 first time your program stops there. Also, like the @code{hbreak}
2421 command, the breakpoint requires hardware support and some target hardware
2422 may not have this support. @xref{Disabling, ,Disabling breakpoints}.
2423 See also @ref{Conditions, ,Break conditions}.
2426 @cindex regular expression
2427 @item rbreak @var{regex}
2428 Set breakpoints on all functions matching the regular expression
2429 @var{regex}. This command sets an unconditional breakpoint on all
2430 matches, printing a list of all breakpoints it set. Once these
2431 breakpoints are set, they are treated just like the breakpoints set with
2432 the @code{break} command. You can delete them, disable them, or make
2433 them conditional the same way as any other breakpoint.
2435 The syntax of the regular expression is the standard one used with tools
2436 like @file{grep}. Note that this is different from the syntax used by
2437 shells, so for instance @code{foo*} matches all functions that include
2438 an @code{fo} followed by zero or more @code{o}s. There is an implicit
2439 @code{.*} leading and trailing the regular expression you supply, so to
2440 match only functions that begin with @code{foo}, use @code{^foo}.
2442 When debugging C++ programs, @code{rbreak} is useful for setting
2443 breakpoints on overloaded functions that are not members of any special
2446 @kindex info breakpoints
2447 @cindex @code{$_} and @code{info breakpoints}
2448 @item info breakpoints @r{[}@var{n}@r{]}
2449 @itemx info break @r{[}@var{n}@r{]}
2450 @itemx info watchpoints @r{[}@var{n}@r{]}
2451 Print a table of all breakpoints, watchpoints, and catchpoints set and
2452 not deleted, with the following columns for each breakpoint:
2455 @item Breakpoint Numbers
2457 Breakpoint, watchpoint, or catchpoint.
2459 Whether the breakpoint is marked to be disabled or deleted when hit.
2460 @item Enabled or Disabled
2461 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2462 that are not enabled.
2464 Where the breakpoint is in your program, as a memory address.
2466 Where the breakpoint is in the source for your program, as a file and
2471 If a breakpoint is conditional, @code{info break} shows the condition on
2472 the line following the affected breakpoint; breakpoint commands, if any,
2473 are listed after that.
2476 @code{info break} with a breakpoint
2477 number @var{n} as argument lists only that breakpoint. The
2478 convenience variable @code{$_} and the default examining-address for
2479 the @code{x} command are set to the address of the last breakpoint
2480 listed (@pxref{Memory, ,Examining memory}).
2483 @code{info break} displays a count of the number of times the breakpoint
2484 has been hit. This is especially useful in conjunction with the
2485 @code{ignore} command. You can ignore a large number of breakpoint
2486 hits, look at the breakpoint info to see how many times the breakpoint
2487 was hit, and then run again, ignoring one less than that number. This
2488 will get you quickly to the last hit of that breakpoint.
2491 @value{GDBN} allows you to set any number of breakpoints at the same place in
2492 your program. There is nothing silly or meaningless about this. When
2493 the breakpoints are conditional, this is even useful
2494 (@pxref{Conditions, ,Break conditions}).
2496 @cindex negative breakpoint numbers
2497 @cindex internal @value{GDBN} breakpoints
2498 @value{GDBN} itself sometimes sets breakpoints in your program for special
2499 purposes, such as proper handling of @code{longjmp} (in C programs).
2500 These internal breakpoints are assigned negative numbers, starting with
2501 @code{-1}; @samp{info breakpoints} does not display them.
2503 You can see these breakpoints with the @value{GDBN} maintenance command
2504 @samp{maint info breakpoints}.
2507 @kindex maint info breakpoints
2508 @item maint info breakpoints
2509 Using the same format as @samp{info breakpoints}, display both the
2510 breakpoints you've set explicitly, and those @value{GDBN} is using for
2511 internal purposes. Internal breakpoints are shown with negative
2512 breakpoint numbers. The type column identifies what kind of breakpoint
2517 Normal, explicitly set breakpoint.
2520 Normal, explicitly set watchpoint.
2523 Internal breakpoint, used to handle correctly stepping through
2524 @code{longjmp} calls.
2526 @item longjmp resume
2527 Internal breakpoint at the target of a @code{longjmp}.
2530 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2533 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2536 Shared library events.
2543 @node Set Watchpoints
2544 @subsection Setting watchpoints
2546 @cindex setting watchpoints
2547 @cindex software watchpoints
2548 @cindex hardware watchpoints
2549 You can use a watchpoint to stop execution whenever the value of an
2550 expression changes, without having to predict a particular place where
2553 Depending on your system, watchpoints may be implemented in software or
2554 hardware. @value{GDBN} does software watchpointing by single-stepping your
2555 program and testing the variable's value each time, which is hundreds of
2556 times slower than normal execution. (But this may still be worth it, to
2557 catch errors where you have no clue what part of your program is the
2560 On some systems, such as HP-UX, Linux and some other x86-based targets,
2561 @value{GDBN} includes support for
2562 hardware watchpoints, which do not slow down the running of your
2567 @item watch @var{expr}
2568 Set a watchpoint for an expression. @value{GDBN} will break when @var{expr}
2569 is written into by the program and its value changes.
2572 @item rwatch @var{expr}
2573 Set a watchpoint that will break when watch @var{expr} is read by the program.
2576 @item awatch @var{expr}
2577 Set a watchpoint that will break when @var{expr} is either read or written into
2580 @kindex info watchpoints
2581 @item info watchpoints
2582 This command prints a list of watchpoints, breakpoints, and catchpoints;
2583 it is the same as @code{info break}.
2586 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
2587 watchpoints execute very quickly, and the debugger reports a change in
2588 value at the exact instruction where the change occurs. If @value{GDBN}
2589 cannot set a hardware watchpoint, it sets a software watchpoint, which
2590 executes more slowly and reports the change in value at the next
2591 statement, not the instruction, after the change occurs.
2593 When you issue the @code{watch} command, @value{GDBN} reports
2596 Hardware watchpoint @var{num}: @var{expr}
2600 if it was able to set a hardware watchpoint.
2602 Currently, the @code{awatch} and @code{rwatch} commands can only set
2603 hardware watchpoints, because accesses to data that don't change the
2604 value of the watched expression cannot be detected without examining
2605 every instruction as it is being executed, and @value{GDBN} does not do
2606 that currently. If @value{GDBN} finds that it is unable to set a
2607 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
2608 will print a message like this:
2611 Expression cannot be implemented with read/access watchpoint.
2614 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
2615 data type of the watched expression is wider than what a hardware
2616 watchpoint on the target machine can handle. For example, some systems
2617 can only watch regions that are up to 4 bytes wide; on such systems you
2618 cannot set hardware watchpoints for an expression that yields a
2619 double-precision floating-point number (which is typically 8 bytes
2620 wide). As a work-around, it might be possible to break the large region
2621 into a series of smaller ones and watch them with separate watchpoints.
2623 If you set too many hardware watchpoints, @value{GDBN} might be unable
2624 to insert all of them when you resume the execution of your program.
2625 Since the precise number of active watchpoints is unknown until such
2626 time as the program is about to be resumed, @value{GDBN} might not be
2627 able to warn you about this when you set the watchpoints, and the
2628 warning will be printed only when the program is resumed:
2631 Hardware watchpoint @var{num}: Could not insert watchpoint
2635 If this happens, delete or disable some of the watchpoints.
2637 The SPARClite DSU will generate traps when a program accesses some data
2638 or instruction address that is assigned to the debug registers. For the
2639 data addresses, DSU facilitates the @code{watch} command. However the
2640 hardware breakpoint registers can only take two data watchpoints, and
2641 both watchpoints must be the same kind. For example, you can set two
2642 watchpoints with @code{watch} commands, two with @code{rwatch} commands,
2643 @strong{or} two with @code{awatch} commands, but you cannot set one
2644 watchpoint with one command and the other with a different command.
2645 @value{GDBN} will reject the command if you try to mix watchpoints.
2646 Delete or disable unused watchpoint commands before setting new ones.
2648 If you call a function interactively using @code{print} or @code{call},
2649 any watchpoints you have set will be inactive until @value{GDBN} reaches another
2650 kind of breakpoint or the call completes.
2652 @value{GDBN} automatically deletes watchpoints that watch local
2653 (automatic) variables, or expressions that involve such variables, when
2654 they go out of scope, that is, when the execution leaves the block in
2655 which these variables were defined. In particular, when the program
2656 being debugged terminates, @emph{all} local variables go out of scope,
2657 and so only watchpoints that watch global variables remain set. If you
2658 rerun the program, you will need to set all such watchpoints again. One
2659 way of doing that would be to set a code breakpoint at the entry to the
2660 @code{main} function and when it breaks, set all the watchpoints.
2663 @cindex watchpoints and threads
2664 @cindex threads and watchpoints
2665 @emph{Warning:} In multi-thread programs, watchpoints have only limited
2666 usefulness. With the current watchpoint implementation, @value{GDBN}
2667 can only watch the value of an expression @emph{in a single thread}. If
2668 you are confident that the expression can only change due to the current
2669 thread's activity (and if you are also confident that no other thread
2670 can become current), then you can use watchpoints as usual. However,
2671 @value{GDBN} may not notice when a non-current thread's activity changes
2674 @c FIXME: this is almost identical to the previous paragraph.
2675 @emph{HP-UX Warning:} In multi-thread programs, software watchpoints
2676 have only limited usefulness. If @value{GDBN} creates a software
2677 watchpoint, it can only watch the value of an expression @emph{in a
2678 single thread}. If you are confident that the expression can only
2679 change due to the current thread's activity (and if you are also
2680 confident that no other thread can become current), then you can use
2681 software watchpoints as usual. However, @value{GDBN} may not notice
2682 when a non-current thread's activity changes the expression. (Hardware
2683 watchpoints, in contrast, watch an expression in all threads.)
2686 @node Set Catchpoints
2687 @subsection Setting catchpoints
2688 @cindex catchpoints, setting
2689 @cindex exception handlers
2690 @cindex event handling
2692 You can use @dfn{catchpoints} to cause the debugger to stop for certain
2693 kinds of program events, such as C++ exceptions or the loading of a
2694 shared library. Use the @code{catch} command to set a catchpoint.
2698 @item catch @var{event}
2699 Stop when @var{event} occurs. @var{event} can be any of the following:
2703 The throwing of a C++ exception.
2707 The catching of a C++ exception.
2711 A call to @code{exec}. This is currently only available for HP-UX.
2715 A call to @code{fork}. This is currently only available for HP-UX.
2719 A call to @code{vfork}. This is currently only available for HP-UX.
2722 @itemx load @var{libname}
2724 The dynamic loading of any shared library, or the loading of the library
2725 @var{libname}. This is currently only available for HP-UX.
2728 @itemx unload @var{libname}
2729 @kindex catch unload
2730 The unloading of any dynamically loaded shared library, or the unloading
2731 of the library @var{libname}. This is currently only available for HP-UX.
2734 @item tcatch @var{event}
2735 Set a catchpoint that is enabled only for one stop. The catchpoint is
2736 automatically deleted after the first time the event is caught.
2740 Use the @code{info break} command to list the current catchpoints.
2742 There are currently some limitations to C++ exception handling
2743 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
2747 If you call a function interactively, @value{GDBN} normally returns
2748 control to you when the function has finished executing. If the call
2749 raises an exception, however, the call may bypass the mechanism that
2750 returns control to you and cause your program either to abort or to
2751 simply continue running until it hits a breakpoint, catches a signal
2752 that @value{GDBN} is listening for, or exits. This is the case even if
2753 you set a catchpoint for the exception; catchpoints on exceptions are
2754 disabled within interactive calls.
2757 You cannot raise an exception interactively.
2760 You cannot install an exception handler interactively.
2763 @cindex raise exceptions
2764 Sometimes @code{catch} is not the best way to debug exception handling:
2765 if you need to know exactly where an exception is raised, it is better to
2766 stop @emph{before} the exception handler is called, since that way you
2767 can see the stack before any unwinding takes place. If you set a
2768 breakpoint in an exception handler instead, it may not be easy to find
2769 out where the exception was raised.
2771 To stop just before an exception handler is called, you need some
2772 knowledge of the implementation. In the case of @sc{gnu} C++, exceptions are
2773 raised by calling a library function named @code{__raise_exception}
2774 which has the following ANSI C interface:
2777 /* @var{addr} is where the exception identifier is stored.
2778 @var{id} is the exception identifier. */
2779 void __raise_exception (void **addr, void *id);
2783 To make the debugger catch all exceptions before any stack
2784 unwinding takes place, set a breakpoint on @code{__raise_exception}
2785 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2787 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2788 that depends on the value of @var{id}, you can stop your program when
2789 a specific exception is raised. You can use multiple conditional
2790 breakpoints to stop your program when any of a number of exceptions are
2795 @subsection Deleting breakpoints
2797 @cindex clearing breakpoints, watchpoints, catchpoints
2798 @cindex deleting breakpoints, watchpoints, catchpoints
2799 It is often necessary to eliminate a breakpoint, watchpoint, or
2800 catchpoint once it has done its job and you no longer want your program
2801 to stop there. This is called @dfn{deleting} the breakpoint. A
2802 breakpoint that has been deleted no longer exists; it is forgotten.
2804 With the @code{clear} command you can delete breakpoints according to
2805 where they are in your program. With the @code{delete} command you can
2806 delete individual breakpoints, watchpoints, or catchpoints by specifying
2807 their breakpoint numbers.
2809 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2810 automatically ignores breakpoints on the first instruction to be executed
2811 when you continue execution without changing the execution address.
2816 Delete any breakpoints at the next instruction to be executed in the
2817 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2818 the innermost frame is selected, this is a good way to delete a
2819 breakpoint where your program just stopped.
2821 @item clear @var{function}
2822 @itemx clear @var{filename}:@var{function}
2823 Delete any breakpoints set at entry to the function @var{function}.
2825 @item clear @var{linenum}
2826 @itemx clear @var{filename}:@var{linenum}
2827 Delete any breakpoints set at or within the code of the specified line.
2829 @cindex delete breakpoints
2831 @kindex d @r{(@code{delete})}
2832 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2833 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
2834 ranges specified as arguments. If no argument is specified, delete all
2835 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
2836 confirm off}). You can abbreviate this command as @code{d}.
2840 @subsection Disabling breakpoints
2842 @kindex disable breakpoints
2843 @kindex enable breakpoints
2844 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
2845 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
2846 it had been deleted, but remembers the information on the breakpoint so
2847 that you can @dfn{enable} it again later.
2849 You disable and enable breakpoints, watchpoints, and catchpoints with
2850 the @code{enable} and @code{disable} commands, optionally specifying one
2851 or more breakpoint numbers as arguments. Use @code{info break} or
2852 @code{info watch} to print a list of breakpoints, watchpoints, and
2853 catchpoints if you do not know which numbers to use.
2855 A breakpoint, watchpoint, or catchpoint can have any of four different
2856 states of enablement:
2860 Enabled. The breakpoint stops your program. A breakpoint set
2861 with the @code{break} command starts out in this state.
2863 Disabled. The breakpoint has no effect on your program.
2865 Enabled once. The breakpoint stops your program, but then becomes
2868 Enabled for deletion. The breakpoint stops your program, but
2869 immediately after it does so it is deleted permanently. A breakpoint
2870 set with the @code{tbreak} command starts out in this state.
2873 You can use the following commands to enable or disable breakpoints,
2874 watchpoints, and catchpoints:
2877 @kindex disable breakpoints
2879 @kindex dis @r{(@code{disable})}
2880 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2881 Disable the specified breakpoints---or all breakpoints, if none are
2882 listed. A disabled breakpoint has no effect but is not forgotten. All
2883 options such as ignore-counts, conditions and commands are remembered in
2884 case the breakpoint is enabled again later. You may abbreviate
2885 @code{disable} as @code{dis}.
2887 @kindex enable breakpoints
2889 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2890 Enable the specified breakpoints (or all defined breakpoints). They
2891 become effective once again in stopping your program.
2893 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
2894 Enable the specified breakpoints temporarily. @value{GDBN} disables any
2895 of these breakpoints immediately after stopping your program.
2897 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
2898 Enable the specified breakpoints to work once, then die. @value{GDBN}
2899 deletes any of these breakpoints as soon as your program stops there.
2902 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
2903 @c confusing: tbreak is also initially enabled.
2904 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2905 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2906 subsequently, they become disabled or enabled only when you use one of
2907 the commands above. (The command @code{until} can set and delete a
2908 breakpoint of its own, but it does not change the state of your other
2909 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2913 @subsection Break conditions
2914 @cindex conditional breakpoints
2915 @cindex breakpoint conditions
2917 @c FIXME what is scope of break condition expr? Context where wanted?
2918 @c in particular for a watchpoint?
2919 The simplest sort of breakpoint breaks every time your program reaches a
2920 specified place. You can also specify a @dfn{condition} for a
2921 breakpoint. A condition is just a Boolean expression in your
2922 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2923 a condition evaluates the expression each time your program reaches it,
2924 and your program stops only if the condition is @emph{true}.
2926 This is the converse of using assertions for program validation; in that
2927 situation, you want to stop when the assertion is violated---that is,
2928 when the condition is false. In C, if you want to test an assertion expressed
2929 by the condition @var{assert}, you should set the condition
2930 @samp{! @var{assert}} on the appropriate breakpoint.
2932 Conditions are also accepted for watchpoints; you may not need them,
2933 since a watchpoint is inspecting the value of an expression anyhow---but
2934 it might be simpler, say, to just set a watchpoint on a variable name,
2935 and specify a condition that tests whether the new value is an interesting
2938 Break conditions can have side effects, and may even call functions in
2939 your program. This can be useful, for example, to activate functions
2940 that log program progress, or to use your own print functions to
2941 format special data structures. The effects are completely predictable
2942 unless there is another enabled breakpoint at the same address. (In
2943 that case, @value{GDBN} might see the other breakpoint first and stop your
2944 program without checking the condition of this one.) Note that
2945 breakpoint commands are usually more convenient and flexible than break
2947 purpose of performing side effects when a breakpoint is reached
2948 (@pxref{Break Commands, ,Breakpoint command lists}).
2950 Break conditions can be specified when a breakpoint is set, by using
2951 @samp{if} in the arguments to the @code{break} command. @xref{Set
2952 Breaks, ,Setting breakpoints}. They can also be changed at any time
2953 with the @code{condition} command.
2955 You can also use the @code{if} keyword with the @code{watch} command.
2956 The @code{catch} command does not recognize the @code{if} keyword;
2957 @code{condition} is the only way to impose a further condition on a
2962 @item condition @var{bnum} @var{expression}
2963 Specify @var{expression} as the break condition for breakpoint,
2964 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
2965 breakpoint @var{bnum} stops your program only if the value of
2966 @var{expression} is true (nonzero, in C). When you use
2967 @code{condition}, @value{GDBN} checks @var{expression} immediately for
2968 syntactic correctness, and to determine whether symbols in it have
2969 referents in the context of your breakpoint. If @var{expression} uses
2970 symbols not referenced in the context of the breakpoint, @value{GDBN}
2971 prints an error message:
2974 No symbol "foo" in current context.
2979 not actually evaluate @var{expression} at the time the @code{condition}
2980 command (or a command that sets a breakpoint with a condition, like
2981 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
2983 @item condition @var{bnum}
2984 Remove the condition from breakpoint number @var{bnum}. It becomes
2985 an ordinary unconditional breakpoint.
2988 @cindex ignore count (of breakpoint)
2989 A special case of a breakpoint condition is to stop only when the
2990 breakpoint has been reached a certain number of times. This is so
2991 useful that there is a special way to do it, using the @dfn{ignore
2992 count} of the breakpoint. Every breakpoint has an ignore count, which
2993 is an integer. Most of the time, the ignore count is zero, and
2994 therefore has no effect. But if your program reaches a breakpoint whose
2995 ignore count is positive, then instead of stopping, it just decrements
2996 the ignore count by one and continues. As a result, if the ignore count
2997 value is @var{n}, the breakpoint does not stop the next @var{n} times
2998 your program reaches it.
3002 @item ignore @var{bnum} @var{count}
3003 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3004 The next @var{count} times the breakpoint is reached, your program's
3005 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3008 To make the breakpoint stop the next time it is reached, specify
3011 When you use @code{continue} to resume execution of your program from a
3012 breakpoint, you can specify an ignore count directly as an argument to
3013 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3014 Stepping,,Continuing and stepping}.
3016 If a breakpoint has a positive ignore count and a condition, the
3017 condition is not checked. Once the ignore count reaches zero,
3018 @value{GDBN} resumes checking the condition.
3020 You could achieve the effect of the ignore count with a condition such
3021 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3022 is decremented each time. @xref{Convenience Vars, ,Convenience
3026 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3029 @node Break Commands
3030 @subsection Breakpoint command lists
3032 @cindex breakpoint commands
3033 You can give any breakpoint (or watchpoint or catchpoint) a series of
3034 commands to execute when your program stops due to that breakpoint. For
3035 example, you might want to print the values of certain expressions, or
3036 enable other breakpoints.
3041 @item commands @r{[}@var{bnum}@r{]}
3042 @itemx @dots{} @var{command-list} @dots{}
3044 Specify a list of commands for breakpoint number @var{bnum}. The commands
3045 themselves appear on the following lines. Type a line containing just
3046 @code{end} to terminate the commands.
3048 To remove all commands from a breakpoint, type @code{commands} and
3049 follow it immediately with @code{end}; that is, give no commands.
3051 With no @var{bnum} argument, @code{commands} refers to the last
3052 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
3053 recently encountered).
3056 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
3057 disabled within a @var{command-list}.
3059 You can use breakpoint commands to start your program up again. Simply
3060 use the @code{continue} command, or @code{step}, or any other command
3061 that resumes execution.
3063 Any other commands in the command list, after a command that resumes
3064 execution, are ignored. This is because any time you resume execution
3065 (even with a simple @code{next} or @code{step}), you may encounter
3066 another breakpoint---which could have its own command list, leading to
3067 ambiguities about which list to execute.
3070 If the first command you specify in a command list is @code{silent}, the
3071 usual message about stopping at a breakpoint is not printed. This may
3072 be desirable for breakpoints that are to print a specific message and
3073 then continue. If none of the remaining commands print anything, you
3074 see no sign that the breakpoint was reached. @code{silent} is
3075 meaningful only at the beginning of a breakpoint command list.
3077 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3078 print precisely controlled output, and are often useful in silent
3079 breakpoints. @xref{Output, ,Commands for controlled output}.
3081 For example, here is how you could use breakpoint commands to print the
3082 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
3088 printf "x is %d\n",x
3093 One application for breakpoint commands is to compensate for one bug so
3094 you can test for another. Put a breakpoint just after the erroneous line
3095 of code, give it a condition to detect the case in which something
3096 erroneous has been done, and give it commands to assign correct values
3097 to any variables that need them. End with the @code{continue} command
3098 so that your program does not stop, and start with the @code{silent}
3099 command so that no output is produced. Here is an example:
3110 @node Breakpoint Menus
3111 @subsection Breakpoint menus
3113 @cindex symbol overloading
3115 Some programming languages (notably C++) permit a single function name
3116 to be defined several times, for application in different contexts.
3117 This is called @dfn{overloading}. When a function name is overloaded,
3118 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
3119 a breakpoint. If you realize this is a problem, you can use
3120 something like @samp{break @var{function}(@var{types})} to specify which
3121 particular version of the function you want. Otherwise, @value{GDBN} offers
3122 you a menu of numbered choices for different possible breakpoints, and
3123 waits for your selection with the prompt @samp{>}. The first two
3124 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
3125 sets a breakpoint at each definition of @var{function}, and typing
3126 @kbd{0} aborts the @code{break} command without setting any new
3129 For example, the following session excerpt shows an attempt to set a
3130 breakpoint at the overloaded symbol @code{String::after}.
3131 We choose three particular definitions of that function name:
3133 @c FIXME! This is likely to change to show arg type lists, at least
3136 (@value{GDBP}) b String::after
3139 [2] file:String.cc; line number:867
3140 [3] file:String.cc; line number:860
3141 [4] file:String.cc; line number:875
3142 [5] file:String.cc; line number:853
3143 [6] file:String.cc; line number:846
3144 [7] file:String.cc; line number:735
3146 Breakpoint 1 at 0xb26c: file String.cc, line 867.
3147 Breakpoint 2 at 0xb344: file String.cc, line 875.
3148 Breakpoint 3 at 0xafcc: file String.cc, line 846.
3149 Multiple breakpoints were set.
3150 Use the "delete" command to delete unwanted
3156 @c @ifclear BARETARGET
3157 @node Error in Breakpoints
3158 @subsection ``Cannot insert breakpoints''
3160 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
3162 Under some operating systems, breakpoints cannot be used in a program if
3163 any other process is running that program. In this situation,
3164 attempting to run or continue a program with a breakpoint causes
3165 @value{GDBN} to print an error message:
3168 Cannot insert breakpoints.
3169 The same program may be running in another process.
3172 When this happens, you have three ways to proceed:
3176 Remove or disable the breakpoints, then continue.
3179 Suspend @value{GDBN}, and copy the file containing your program to a new
3180 name. Resume @value{GDBN} and use the @code{exec-file} command to specify
3181 that @value{GDBN} should run your program under that name.
3182 Then start your program again.
3185 Relink your program so that the text segment is nonsharable, using the
3186 linker option @samp{-N}. The operating system limitation may not apply
3187 to nonsharable executables.
3191 A similar message can be printed if you request too many active
3192 hardware-assisted breakpoints and watchpoints:
3194 @c FIXME: the precise wording of this message may change; the relevant
3195 @c source change is not committed yet (Sep 3, 1999).
3197 Stopped; cannot insert breakpoints.
3198 You may have requested too many hardware breakpoints and watchpoints.
3202 This message is printed when you attempt to resume the program, since
3203 only then @value{GDBN} knows exactly how many hardware breakpoints and
3204 watchpoints it needs to insert.
3206 When this message is printed, you need to disable or remove some of the
3207 hardware-assisted breakpoints and watchpoints, and then continue.
3210 @node Continuing and Stepping
3211 @section Continuing and stepping
3215 @cindex resuming execution
3216 @dfn{Continuing} means resuming program execution until your program
3217 completes normally. In contrast, @dfn{stepping} means executing just
3218 one more ``step'' of your program, where ``step'' may mean either one
3219 line of source code, or one machine instruction (depending on what
3220 particular command you use). Either when continuing or when stepping,
3221 your program may stop even sooner, due to a breakpoint or a signal. (If
3222 it stops due to a signal, you may want to use @code{handle}, or use
3223 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
3227 @kindex c @r{(@code{continue})}
3228 @kindex fg @r{(resume foreground execution)}
3229 @item continue @r{[}@var{ignore-count}@r{]}
3230 @itemx c @r{[}@var{ignore-count}@r{]}
3231 @itemx fg @r{[}@var{ignore-count}@r{]}
3232 Resume program execution, at the address where your program last stopped;
3233 any breakpoints set at that address are bypassed. The optional argument
3234 @var{ignore-count} allows you to specify a further number of times to
3235 ignore a breakpoint at this location; its effect is like that of
3236 @code{ignore} (@pxref{Conditions, ,Break conditions}).
3238 The argument @var{ignore-count} is meaningful only when your program
3239 stopped due to a breakpoint. At other times, the argument to
3240 @code{continue} is ignored.
3242 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
3243 debugged program is deemed to be the foreground program) are provided
3244 purely for convenience, and have exactly the same behavior as
3248 To resume execution at a different place, you can use @code{return}
3249 (@pxref{Returning, ,Returning from a function}) to go back to the
3250 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
3251 different address}) to go to an arbitrary location in your program.
3253 A typical technique for using stepping is to set a breakpoint
3254 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and catchpoints}) at the
3255 beginning of the function or the section of your program where a problem
3256 is believed to lie, run your program until it stops at that breakpoint,
3257 and then step through the suspect area, examining the variables that are
3258 interesting, until you see the problem happen.
3262 @kindex s @r{(@code{step})}
3264 Continue running your program until control reaches a different source
3265 line, then stop it and return control to @value{GDBN}. This command is
3266 abbreviated @code{s}.
3269 @c "without debugging information" is imprecise; actually "without line
3270 @c numbers in the debugging information". (gcc -g1 has debugging info but
3271 @c not line numbers). But it seems complex to try to make that
3272 @c distinction here.
3273 @emph{Warning:} If you use the @code{step} command while control is
3274 within a function that was compiled without debugging information,
3275 execution proceeds until control reaches a function that does have
3276 debugging information. Likewise, it will not step into a function which
3277 is compiled without debugging information. To step through functions
3278 without debugging information, use the @code{stepi} command, described
3282 The @code{step} command only stops at the first instruction of a
3283 source line. This prevents the multiple stops that could otherwise occur in
3284 switch statements, for loops, etc. @code{step} continues to stop if a
3285 function that has debugging information is called within the line.
3286 In other words, @code{step} @emph{steps inside} any functions called
3289 Also, the @code{step} command only enters a function if there is line
3290 number information for the function. Otherwise it acts like the
3291 @code{next} command. This avoids problems when using @code{cc -gl}
3292 on MIPS machines. Previously, @code{step} entered subroutines if there
3293 was any debugging information about the routine.
3295 @item step @var{count}
3296 Continue running as in @code{step}, but do so @var{count} times. If a
3297 breakpoint is reached, or a signal not related to stepping occurs before
3298 @var{count} steps, stepping stops right away.
3301 @kindex n @r{(@code{next})}
3302 @item next @r{[}@var{count}@r{]}
3303 Continue to the next source line in the current (innermost) stack frame.
3304 This is similar to @code{step}, but function calls that appear within
3305 the line of code are executed without stopping. Execution stops when
3306 control reaches a different line of code at the original stack level
3307 that was executing when you gave the @code{next} command. This command
3308 is abbreviated @code{n}.
3310 An argument @var{count} is a repeat count, as for @code{step}.
3313 @c FIX ME!! Do we delete this, or is there a way it fits in with
3314 @c the following paragraph? --- Vctoria
3316 @c @code{next} within a function that lacks debugging information acts like
3317 @c @code{step}, but any function calls appearing within the code of the
3318 @c function are executed without stopping.
3320 The @code{next} command only stops at the first instruction of a
3321 source line. This prevents multiple stops that could otherwise occur in
3322 switch statements, for loops, etc.
3326 Continue running until just after function in the selected stack frame
3327 returns. Print the returned value (if any).
3329 Contrast this with the @code{return} command (@pxref{Returning,
3330 ,Returning from a function}).
3333 @kindex u @r{(@code{until})}
3336 Continue running until a source line past the current line, in the
3337 current stack frame, is reached. This command is used to avoid single
3338 stepping through a loop more than once. It is like the @code{next}
3339 command, except that when @code{until} encounters a jump, it
3340 automatically continues execution until the program counter is greater
3341 than the address of the jump.
3343 This means that when you reach the end of a loop after single stepping
3344 though it, @code{until} makes your program continue execution until it
3345 exits the loop. In contrast, a @code{next} command at the end of a loop
3346 simply steps back to the beginning of the loop, which forces you to step
3347 through the next iteration.
3349 @code{until} always stops your program if it attempts to exit the current
3352 @code{until} may produce somewhat counterintuitive results if the order
3353 of machine code does not match the order of the source lines. For
3354 example, in the following excerpt from a debugging session, the @code{f}
3355 (@code{frame}) command shows that execution is stopped at line
3356 @code{206}; yet when we use @code{until}, we get to line @code{195}:
3360 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
3362 (@value{GDBP}) until
3363 195 for ( ; argc > 0; NEXTARG) @{
3366 This happened because, for execution efficiency, the compiler had
3367 generated code for the loop closure test at the end, rather than the
3368 start, of the loop---even though the test in a C @code{for}-loop is
3369 written before the body of the loop. The @code{until} command appeared
3370 to step back to the beginning of the loop when it advanced to this
3371 expression; however, it has not really gone to an earlier
3372 statement---not in terms of the actual machine code.
3374 @code{until} with no argument works by means of single
3375 instruction stepping, and hence is slower than @code{until} with an
3378 @item until @var{location}
3379 @itemx u @var{location}
3380 Continue running your program until either the specified location is
3381 reached, or the current stack frame returns. @var{location} is any of
3382 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
3383 ,Setting breakpoints}). This form of the command uses breakpoints,
3384 and hence is quicker than @code{until} without an argument.
3387 @kindex si @r{(@code{stepi})}
3389 @itemx stepi @var{arg}
3391 Execute one machine instruction, then stop and return to the debugger.
3393 It is often useful to do @samp{display/i $pc} when stepping by machine
3394 instructions. This makes @value{GDBN} automatically display the next
3395 instruction to be executed, each time your program stops. @xref{Auto
3396 Display,, Automatic display}.
3398 An argument is a repeat count, as in @code{step}.
3402 @kindex ni @r{(@code{nexti})}
3404 @itemx nexti @var{arg}
3406 Execute one machine instruction, but if it is a function call,
3407 proceed until the function returns.
3409 An argument is a repeat count, as in @code{next}.
3416 A signal is an asynchronous event that can happen in a program. The
3417 operating system defines the possible kinds of signals, and gives each
3418 kind a name and a number. For example, in Unix @code{SIGINT} is the
3419 signal a program gets when you type an interrupt character (often @kbd{C-c});
3420 @code{SIGSEGV} is the signal a program gets from referencing a place in
3421 memory far away from all the areas in use; @code{SIGALRM} occurs when
3422 the alarm clock timer goes off (which happens only if your program has
3423 requested an alarm).
3425 @cindex fatal signals
3426 Some signals, including @code{SIGALRM}, are a normal part of the
3427 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3428 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
3429 program has not specified in advance some other way to handle the signal.
3430 @code{SIGINT} does not indicate an error in your program, but it is normally
3431 fatal so it can carry out the purpose of the interrupt: to kill the program.
3433 @value{GDBN} has the ability to detect any occurrence of a signal in your
3434 program. You can tell @value{GDBN} in advance what to do for each kind of
3437 @cindex handling signals
3438 Normally, @value{GDBN} is set up to ignore non-erroneous signals like @code{SIGALRM}
3439 (so as not to interfere with their role in the functioning of your program)
3440 but to stop your program immediately whenever an error signal happens.
3441 You can change these settings with the @code{handle} command.
3444 @kindex info signals
3447 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3448 handle each one. You can use this to see the signal numbers of all
3449 the defined types of signals.
3451 @code{info handle} is an alias for @code{info signals}.
3454 @item handle @var{signal} @var{keywords}@dots{}
3455 Change the way @value{GDBN} handles signal @var{signal}. @var{signal} can
3456 be the number of a signal or its name (with or without the @samp{SIG} at the
3457 beginning). The @var{keywords} say what change to make.
3461 The keywords allowed by the @code{handle} command can be abbreviated.
3462 Their full names are:
3466 @value{GDBN} should not stop your program when this signal happens. It may
3467 still print a message telling you that the signal has come in.
3470 @value{GDBN} should stop your program when this signal happens. This implies
3471 the @code{print} keyword as well.
3474 @value{GDBN} should print a message when this signal happens.
3477 @value{GDBN} should not mention the occurrence of the signal at all. This
3478 implies the @code{nostop} keyword as well.
3481 @value{GDBN} should allow your program to see this signal; your program
3482 can handle the signal, or else it may terminate if the signal is fatal
3486 @value{GDBN} should not allow your program to see this signal.
3490 When a signal stops your program, the signal is not visible to the
3492 continue. Your program sees the signal then, if @code{pass} is in
3493 effect for the signal in question @emph{at that time}. In other words,
3494 after @value{GDBN} reports a signal, you can use the @code{handle}
3495 command with @code{pass} or @code{nopass} to control whether your
3496 program sees that signal when you continue.
3498 You can also use the @code{signal} command to prevent your program from
3499 seeing a signal, or cause it to see a signal it normally would not see,
3500 or to give it any signal at any time. For example, if your program stopped
3501 due to some sort of memory reference error, you might store correct
3502 values into the erroneous variables and continue, hoping to see more
3503 execution; but your program would probably terminate immediately as
3504 a result of the fatal signal once it saw the signal. To prevent this,
3505 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3509 @section Stopping and starting multi-thread programs
3511 When your program has multiple threads (@pxref{Threads,, Debugging
3512 programs with multiple threads}), you can choose whether to set
3513 breakpoints on all threads, or on a particular thread.
3516 @cindex breakpoints and threads
3517 @cindex thread breakpoints
3518 @kindex break @dots{} thread @var{threadno}
3519 @item break @var{linespec} thread @var{threadno}
3520 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3521 @var{linespec} specifies source lines; there are several ways of
3522 writing them, but the effect is always to specify some source line.
3524 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3525 to specify that you only want @value{GDBN} to stop the program when a
3526 particular thread reaches this breakpoint. @var{threadno} is one of the
3527 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3528 column of the @samp{info threads} display.
3530 If you do not specify @samp{thread @var{threadno}} when you set a
3531 breakpoint, the breakpoint applies to @emph{all} threads of your
3534 You can use the @code{thread} qualifier on conditional breakpoints as
3535 well; in this case, place @samp{thread @var{threadno}} before the
3536 breakpoint condition, like this:
3539 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
3544 @cindex stopped threads
3545 @cindex threads, stopped
3546 Whenever your program stops under @value{GDBN} for any reason,
3547 @emph{all} threads of execution stop, not just the current thread. This
3548 allows you to examine the overall state of the program, including
3549 switching between threads, without worrying that things may change
3552 @cindex continuing threads
3553 @cindex threads, continuing
3554 Conversely, whenever you restart the program, @emph{all} threads start
3555 executing. @emph{This is true even when single-stepping} with commands
3556 like @code{step} or @code{next}.
3558 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3559 Since thread scheduling is up to your debugging target's operating
3560 system (not controlled by @value{GDBN}), other threads may
3561 execute more than one statement while the current thread completes a
3562 single step. Moreover, in general other threads stop in the middle of a
3563 statement, rather than at a clean statement boundary, when the program
3566 You might even find your program stopped in another thread after
3567 continuing or even single-stepping. This happens whenever some other
3568 thread runs into a breakpoint, a signal, or an exception before the
3569 first thread completes whatever you requested.
3571 On some OSes, you can lock the OS scheduler and thus allow only a single
3575 @item set scheduler-locking @var{mode}
3576 Set the scheduler locking mode. If it is @code{off}, then there is no
3577 locking and any thread may run at any time. If @code{on}, then only the
3578 current thread may run when the inferior is resumed. The @code{step}
3579 mode optimizes for single-stepping. It stops other threads from
3580 ``seizing the prompt'' by preempting the current thread while you are
3581 stepping. Other threads will only rarely (or never) get a chance to run
3582 when you step. They are more likely to run when you @samp{next} over a
3583 function call, and they are completely free to run when you use commands
3584 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
3585 thread hits a breakpoint during its timeslice, they will never steal the
3586 @value{GDBN} prompt away from the thread that you are debugging.
3588 @item show scheduler-locking
3589 Display the current scheduler locking mode.
3594 @chapter Examining the Stack
3596 When your program has stopped, the first thing you need to know is where it
3597 stopped and how it got there.
3600 Each time your program performs a function call, information about the call
3602 That information includes the location of the call in your program,
3603 the arguments of the call,
3604 and the local variables of the function being called.
3605 The information is saved in a block of data called a @dfn{stack frame}.
3606 The stack frames are allocated in a region of memory called the @dfn{call
3609 When your program stops, the @value{GDBN} commands for examining the
3610 stack allow you to see all of this information.
3612 @cindex selected frame
3613 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3614 @value{GDBN} commands refer implicitly to the selected frame. In
3615 particular, whenever you ask @value{GDBN} for the value of a variable in
3616 your program, the value is found in the selected frame. There are
3617 special @value{GDBN} commands to select whichever frame you are
3618 interested in. @xref{Selection, ,Selecting a frame}.
3620 When your program stops, @value{GDBN} automatically selects the
3621 currently executing frame and describes it briefly, similar to the
3622 @code{frame} command (@pxref{Frame Info, ,Information about a frame}).
3625 * Frames:: Stack frames
3626 * Backtrace:: Backtraces
3627 * Selection:: Selecting a frame
3628 * Frame Info:: Information on a frame
3633 @section Stack frames
3635 @cindex frame, definition
3637 The call stack is divided up into contiguous pieces called @dfn{stack
3638 frames}, or @dfn{frames} for short; each frame is the data associated
3639 with one call to one function. The frame contains the arguments given
3640 to the function, the function's local variables, and the address at
3641 which the function is executing.
3643 @cindex initial frame
3644 @cindex outermost frame
3645 @cindex innermost frame
3646 When your program is started, the stack has only one frame, that of the
3647 function @code{main}. This is called the @dfn{initial} frame or the
3648 @dfn{outermost} frame. Each time a function is called, a new frame is
3649 made. Each time a function returns, the frame for that function invocation
3650 is eliminated. If a function is recursive, there can be many frames for
3651 the same function. The frame for the function in which execution is
3652 actually occurring is called the @dfn{innermost} frame. This is the most
3653 recently created of all the stack frames that still exist.
3655 @cindex frame pointer
3656 Inside your program, stack frames are identified by their addresses. A
3657 stack frame consists of many bytes, each of which has its own address; each
3658 kind of computer has a convention for choosing one byte whose
3659 address serves as the address of the frame. Usually this address is kept
3660 in a register called the @dfn{frame pointer register} while execution is
3661 going on in that frame.
3663 @cindex frame number
3664 @value{GDBN} assigns numbers to all existing stack frames, starting with
3665 zero for the innermost frame, one for the frame that called it,
3666 and so on upward. These numbers do not really exist in your program;
3667 they are assigned by @value{GDBN} to give you a way of designating stack
3668 frames in @value{GDBN} commands.
3670 @c The -fomit-frame-pointer below perennially causes hbox overflow
3671 @c underflow problems.
3672 @cindex frameless execution
3673 Some compilers provide a way to compile functions so that they operate
3674 without stack frames. (For example, the @value{GCC} option
3676 @samp{-fomit-frame-pointer}
3678 generates functions without a frame.)
3679 This is occasionally done with heavily used library functions to save
3680 the frame setup time. @value{GDBN} has limited facilities for dealing
3681 with these function invocations. If the innermost function invocation
3682 has no stack frame, @value{GDBN} nevertheless regards it as though
3683 it had a separate frame, which is numbered zero as usual, allowing
3684 correct tracing of the function call chain. However, @value{GDBN} has
3685 no provision for frameless functions elsewhere in the stack.
3688 @kindex frame@r{, command}
3689 @cindex current stack frame
3690 @item frame @var{args}
3691 The @code{frame} command allows you to move from one stack frame to another,
3692 and to print the stack frame you select. @var{args} may be either the
3693 address of the frame or the stack frame number. Without an argument,
3694 @code{frame} prints the current stack frame.
3696 @kindex select-frame
3697 @cindex selecting frame silently
3699 The @code{select-frame} command allows you to move from one stack frame
3700 to another without printing the frame. This is the silent version of
3709 @cindex stack traces
3710 A backtrace is a summary of how your program got where it is. It shows one
3711 line per frame, for many frames, starting with the currently executing
3712 frame (frame zero), followed by its caller (frame one), and on up the
3717 @kindex bt @r{(@code{backtrace})}
3720 Print a backtrace of the entire stack: one line per frame for all
3721 frames in the stack.
3723 You can stop the backtrace at any time by typing the system interrupt
3724 character, normally @kbd{C-c}.
3726 @item backtrace @var{n}
3728 Similar, but print only the innermost @var{n} frames.
3730 @item backtrace -@var{n}
3732 Similar, but print only the outermost @var{n} frames.
3737 @kindex info s @r{(@code{info stack})}
3738 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3739 are additional aliases for @code{backtrace}.
3741 Each line in the backtrace shows the frame number and the function name.
3742 The program counter value is also shown---unless you use @code{set
3743 print address off}. The backtrace also shows the source file name and
3744 line number, as well as the arguments to the function. The program
3745 counter value is omitted if it is at the beginning of the code for that
3748 Here is an example of a backtrace. It was made with the command
3749 @samp{bt 3}, so it shows the innermost three frames.
3753 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3755 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3756 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3758 (More stack frames follow...)
3763 The display for frame zero does not begin with a program counter
3764 value, indicating that your program has stopped at the beginning of the
3765 code for line @code{993} of @code{builtin.c}.
3768 @section Selecting a frame
3770 Most commands for examining the stack and other data in your program work on
3771 whichever stack frame is selected at the moment. Here are the commands for
3772 selecting a stack frame; all of them finish by printing a brief description
3773 of the stack frame just selected.
3776 @kindex frame@r{, selecting}
3777 @kindex f @r{(@code{frame})}
3780 Select frame number @var{n}. Recall that frame zero is the innermost
3781 (currently executing) frame, frame one is the frame that called the
3782 innermost one, and so on. The highest-numbered frame is the one for
3785 @item frame @var{addr}
3787 Select the frame at address @var{addr}. This is useful mainly if the
3788 chaining of stack frames has been damaged by a bug, making it
3789 impossible for @value{GDBN} to assign numbers properly to all frames. In
3790 addition, this can be useful when your program has multiple stacks and
3791 switches between them.
3793 On the SPARC architecture, @code{frame} needs two addresses to
3794 select an arbitrary frame: a frame pointer and a stack pointer.
3796 On the MIPS and Alpha architecture, it needs two addresses: a stack
3797 pointer and a program counter.
3799 On the 29k architecture, it needs three addresses: a register stack
3800 pointer, a program counter, and a memory stack pointer.
3801 @c note to future updaters: this is conditioned on a flag
3802 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3803 @c as of 27 Jan 1994.
3807 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3808 advances toward the outermost frame, to higher frame numbers, to frames
3809 that have existed longer. @var{n} defaults to one.
3812 @kindex do @r{(@code{down})}
3814 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3815 advances toward the innermost frame, to lower frame numbers, to frames
3816 that were created more recently. @var{n} defaults to one. You may
3817 abbreviate @code{down} as @code{do}.
3820 All of these commands end by printing two lines of output describing the
3821 frame. The first line shows the frame number, the function name, the
3822 arguments, and the source file and line number of execution in that
3823 frame. The second line shows the text of that source line.
3831 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3833 10 read_input_file (argv[i]);
3837 After such a printout, the @code{list} command with no arguments
3838 prints ten lines centered on the point of execution in the frame.
3839 @xref{List, ,Printing source lines}.
3842 @kindex down-silently
3844 @item up-silently @var{n}
3845 @itemx down-silently @var{n}
3846 These two commands are variants of @code{up} and @code{down},
3847 respectively; they differ in that they do their work silently, without
3848 causing display of the new frame. They are intended primarily for use
3849 in @value{GDBN} command scripts, where the output might be unnecessary and
3854 @section Information about a frame
3856 There are several other commands to print information about the selected
3862 When used without any argument, this command does not change which
3863 frame is selected, but prints a brief description of the currently
3864 selected stack frame. It can be abbreviated @code{f}. With an
3865 argument, this command is used to select a stack frame.
3866 @xref{Selection, ,Selecting a frame}.
3869 @kindex info f @r{(@code{info frame})}
3872 This command prints a verbose description of the selected stack frame,
3877 the address of the frame
3879 the address of the next frame down (called by this frame)
3881 the address of the next frame up (caller of this frame)
3883 the language in which the source code corresponding to this frame is written
3885 the address of the frame's arguments
3887 the address of the frame's local variables
3889 the program counter saved in it (the address of execution in the caller frame)
3891 which registers were saved in the frame
3894 @noindent The verbose description is useful when
3895 something has gone wrong that has made the stack format fail to fit
3896 the usual conventions.
3898 @item info frame @var{addr}
3899 @itemx info f @var{addr}
3900 Print a verbose description of the frame at address @var{addr}, without
3901 selecting that frame. The selected frame remains unchanged by this
3902 command. This requires the same kind of address (more than one for some
3903 architectures) that you specify in the @code{frame} command.
3904 @xref{Selection, ,Selecting a frame}.
3908 Print the arguments of the selected frame, each on a separate line.
3912 Print the local variables of the selected frame, each on a separate
3913 line. These are all variables (declared either static or automatic)
3914 accessible at the point of execution of the selected frame.
3917 @cindex catch exceptions, list active handlers
3918 @cindex exception handlers, how to list
3920 Print a list of all the exception handlers that are active in the
3921 current stack frame at the current point of execution. To see other
3922 exception handlers, visit the associated frame (using the @code{up},
3923 @code{down}, or @code{frame} commands); then type @code{info catch}.
3924 @xref{Set Catchpoints, , Setting catchpoints}.
3930 @chapter Examining Source Files
3932 @value{GDBN} can print parts of your program's source, since the debugging
3933 information recorded in the program tells @value{GDBN} what source files were
3934 used to build it. When your program stops, @value{GDBN} spontaneously prints
3935 the line where it stopped. Likewise, when you select a stack frame
3936 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
3937 execution in that frame has stopped. You can print other portions of
3938 source files by explicit command.
3940 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
3941 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
3942 @value{GDBN} under @sc{gnu} Emacs}.
3945 * List:: Printing source lines
3946 * Search:: Searching source files
3947 * Source Path:: Specifying source directories
3948 * Machine Code:: Source and machine code
3952 @section Printing source lines
3955 @kindex l @r{(@code{list})}
3956 To print lines from a source file, use the @code{list} command
3957 (abbreviated @code{l}). By default, ten lines are printed.
3958 There are several ways to specify what part of the file you want to print.
3960 Here are the forms of the @code{list} command most commonly used:
3963 @item list @var{linenum}
3964 Print lines centered around line number @var{linenum} in the
3965 current source file.
3967 @item list @var{function}
3968 Print lines centered around the beginning of function
3972 Print more lines. If the last lines printed were printed with a
3973 @code{list} command, this prints lines following the last lines
3974 printed; however, if the last line printed was a solitary line printed
3975 as part of displaying a stack frame (@pxref{Stack, ,Examining the
3976 Stack}), this prints lines centered around that line.
3979 Print lines just before the lines last printed.
3982 By default, @value{GDBN} prints ten source lines with any of these forms of
3983 the @code{list} command. You can change this using @code{set listsize}:
3986 @kindex set listsize
3987 @item set listsize @var{count}
3988 Make the @code{list} command display @var{count} source lines (unless
3989 the @code{list} argument explicitly specifies some other number).
3991 @kindex show listsize
3993 Display the number of lines that @code{list} prints.
3996 Repeating a @code{list} command with @key{RET} discards the argument,
3997 so it is equivalent to typing just @code{list}. This is more useful
3998 than listing the same lines again. An exception is made for an
3999 argument of @samp{-}; that argument is preserved in repetition so that
4000 each repetition moves up in the source file.
4003 In general, the @code{list} command expects you to supply zero, one or two
4004 @dfn{linespecs}. Linespecs specify source lines; there are several ways
4005 of writing them, but the effect is always to specify some source line.
4006 Here is a complete description of the possible arguments for @code{list}:
4009 @item list @var{linespec}
4010 Print lines centered around the line specified by @var{linespec}.
4012 @item list @var{first},@var{last}
4013 Print lines from @var{first} to @var{last}. Both arguments are
4016 @item list ,@var{last}
4017 Print lines ending with @var{last}.
4019 @item list @var{first},
4020 Print lines starting with @var{first}.
4023 Print lines just after the lines last printed.
4026 Print lines just before the lines last printed.
4029 As described in the preceding table.
4032 Here are the ways of specifying a single source line---all the
4037 Specifies line @var{number} of the current source file.
4038 When a @code{list} command has two linespecs, this refers to
4039 the same source file as the first linespec.
4042 Specifies the line @var{offset} lines after the last line printed.
4043 When used as the second linespec in a @code{list} command that has
4044 two, this specifies the line @var{offset} lines down from the
4048 Specifies the line @var{offset} lines before the last line printed.
4050 @item @var{filename}:@var{number}
4051 Specifies line @var{number} in the source file @var{filename}.
4053 @item @var{function}
4054 Specifies the line that begins the body of the function @var{function}.
4055 For example: in C, this is the line with the open brace.
4057 @item @var{filename}:@var{function}
4058 Specifies the line of the open-brace that begins the body of the
4059 function @var{function} in the file @var{filename}. You only need the
4060 file name with a function name to avoid ambiguity when there are
4061 identically named functions in different source files.
4063 @item *@var{address}
4064 Specifies the line containing the program address @var{address}.
4065 @var{address} may be any expression.
4069 @section Searching source files
4071 @kindex reverse-search
4073 There are two commands for searching through the current source file for a
4078 @kindex forward-search
4079 @item forward-search @var{regexp}
4080 @itemx search @var{regexp}
4081 The command @samp{forward-search @var{regexp}} checks each line,
4082 starting with the one following the last line listed, for a match for
4083 @var{regexp}. It lists the line that is found. You can use the
4084 synonym @samp{search @var{regexp}} or abbreviate the command name as
4087 @item reverse-search @var{regexp}
4088 The command @samp{reverse-search @var{regexp}} checks each line, starting
4089 with the one before the last line listed and going backward, for a match
4090 for @var{regexp}. It lists the line that is found. You can abbreviate
4091 this command as @code{rev}.
4095 @section Specifying source directories
4098 @cindex directories for source files
4099 Executable programs sometimes do not record the directories of the source
4100 files from which they were compiled, just the names. Even when they do,
4101 the directories could be moved between the compilation and your debugging
4102 session. @value{GDBN} has a list of directories to search for source files;
4103 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
4104 it tries all the directories in the list, in the order they are present
4105 in the list, until it finds a file with the desired name. Note that
4106 the executable search path is @emph{not} used for this purpose. Neither is
4107 the current working directory, unless it happens to be in the source
4110 If @value{GDBN} cannot find a source file in the source path, and the
4111 object program records a directory, @value{GDBN} tries that directory
4112 too. If the source path is empty, and there is no record of the
4113 compilation directory, @value{GDBN} looks in the current directory as a
4116 Whenever you reset or rearrange the source path, @value{GDBN} clears out
4117 any information it has cached about where source files are found and where
4118 each line is in the file.
4122 When you start @value{GDBN}, its source path includes only @samp{cdir}
4123 and @samp{cwd}, in that order.
4124 To add other directories, use the @code{directory} command.
4127 @item directory @var{dirname} @dots{}
4128 @item dir @var{dirname} @dots{}
4129 Add directory @var{dirname} to the front of the source path. Several
4130 directory names may be given to this command, separated by @samp{:}
4131 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
4132 part of absolute file names) or
4133 whitespace. You may specify a directory that is already in the source
4134 path; this moves it forward, so @value{GDBN} searches it sooner.
4138 @vindex $cdir@r{, convenience variable}
4139 @vindex $cwdr@r{, convenience variable}
4140 @cindex compilation directory
4141 @cindex current directory
4142 @cindex working directory
4143 @cindex directory, current
4144 @cindex directory, compilation
4145 You can use the string @samp{$cdir} to refer to the compilation
4146 directory (if one is recorded), and @samp{$cwd} to refer to the current
4147 working directory. @samp{$cwd} is not the same as @samp{.}---the former
4148 tracks the current working directory as it changes during your @value{GDBN}
4149 session, while the latter is immediately expanded to the current
4150 directory at the time you add an entry to the source path.
4153 Reset the source path to empty again. This requires confirmation.
4155 @c RET-repeat for @code{directory} is explicitly disabled, but since
4156 @c repeating it would be a no-op we do not say that. (thanks to RMS)
4158 @item show directories
4159 @kindex show directories
4160 Print the source path: show which directories it contains.
4163 If your source path is cluttered with directories that are no longer of
4164 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
4165 versions of source. You can correct the situation as follows:
4169 Use @code{directory} with no argument to reset the source path to empty.
4172 Use @code{directory} with suitable arguments to reinstall the
4173 directories you want in the source path. You can add all the
4174 directories in one command.
4178 @section Source and machine code
4180 You can use the command @code{info line} to map source lines to program
4181 addresses (and vice versa), and the command @code{disassemble} to display
4182 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
4183 mode, the @code{info line} command causes the arrow to point to the
4184 line specified. Also, @code{info line} prints addresses in symbolic form as
4189 @item info line @var{linespec}
4190 Print the starting and ending addresses of the compiled code for
4191 source line @var{linespec}. You can specify source lines in any of
4192 the ways understood by the @code{list} command (@pxref{List, ,Printing
4196 For example, we can use @code{info line} to discover the location of
4197 the object code for the first line of function
4198 @code{m4_changequote}:
4200 @c FIXME: I think this example should also show the addresses in
4201 @c symbolic form, as they usually would be displayed.
4203 (@value{GDBP}) info line m4_changequote
4204 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
4208 We can also inquire (using @code{*@var{addr}} as the form for
4209 @var{linespec}) what source line covers a particular address:
4211 (@value{GDBP}) info line *0x63ff
4212 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
4215 @cindex @code{$_} and @code{info line}
4216 @kindex x@r{(examine), and} info line
4217 After @code{info line}, the default address for the @code{x} command
4218 is changed to the starting address of the line, so that @samp{x/i} is
4219 sufficient to begin examining the machine code (@pxref{Memory,
4220 ,Examining memory}). Also, this address is saved as the value of the
4221 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
4226 @cindex assembly instructions
4227 @cindex instructions, assembly
4228 @cindex machine instructions
4229 @cindex listing machine instructions
4231 This specialized command dumps a range of memory as machine
4232 instructions. The default memory range is the function surrounding the
4233 program counter of the selected frame. A single argument to this
4234 command is a program counter value; @value{GDBN} dumps the function
4235 surrounding this value. Two arguments specify a range of addresses
4236 (first inclusive, second exclusive) to dump.
4239 The following example shows the disassembly of a range of addresses of
4240 HP PA-RISC 2.0 code:
4243 (@value{GDBP}) disas 0x32c4 0x32e4
4244 Dump of assembler code from 0x32c4 to 0x32e4:
4245 0x32c4 <main+204>: addil 0,dp
4246 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
4247 0x32cc <main+212>: ldil 0x3000,r31
4248 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
4249 0x32d4 <main+220>: ldo 0(r31),rp
4250 0x32d8 <main+224>: addil -0x800,dp
4251 0x32dc <main+228>: ldo 0x588(r1),r26
4252 0x32e0 <main+232>: ldil 0x3000,r31
4253 End of assembler dump.
4256 Some architectures have more than one commonly-used set of instruction
4257 mnemonics or other syntax.
4260 @kindex set disassembly-flavor
4261 @cindex assembly instructions
4262 @cindex instructions, assembly
4263 @cindex machine instructions
4264 @cindex listing machine instructions
4265 @cindex Intel disassembly flavor
4266 @cindex AT&T disassembly flavor
4267 @item set disassembly-flavor @var{instruction-set}
4268 Select the instruction set to use when disassembling the
4269 program via the @code{disassemble} or @code{x/i} commands.
4271 Currently this command is only defined for the Intel x86 family. You
4272 can set @var{instruction-set} to either @code{intel} or @code{att}.
4273 The default is @code{att}, the AT&T flavor used by default by Unix
4274 assemblers for x86-based targets.
4279 @chapter Examining Data
4281 @cindex printing data
4282 @cindex examining data
4285 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
4286 @c document because it is nonstandard... Under Epoch it displays in a
4287 @c different window or something like that.
4288 The usual way to examine data in your program is with the @code{print}
4289 command (abbreviated @code{p}), or its synonym @code{inspect}. It
4290 evaluates and prints the value of an expression of the language your
4291 program is written in (@pxref{Languages, ,Using @value{GDBN} with
4292 Different Languages}).
4295 @item print @var{expr}
4296 @itemx print /@var{f} @var{expr}
4297 @var{expr} is an expression (in the source language). By default the
4298 value of @var{expr} is printed in a format appropriate to its data type;
4299 you can choose a different format by specifying @samp{/@var{f}}, where
4300 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
4304 @itemx print /@var{f}
4305 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
4306 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
4307 conveniently inspect the same value in an alternative format.
4310 A more low-level way of examining data is with the @code{x} command.
4311 It examines data in memory at a specified address and prints it in a
4312 specified format. @xref{Memory, ,Examining memory}.
4314 If you are interested in information about types, or about how the
4315 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
4316 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
4320 * Expressions:: Expressions
4321 * Variables:: Program variables
4322 * Arrays:: Artificial arrays
4323 * Output Formats:: Output formats
4324 * Memory:: Examining memory
4325 * Auto Display:: Automatic display
4326 * Print Settings:: Print settings
4327 * Value History:: Value history
4328 * Convenience Vars:: Convenience variables
4329 * Registers:: Registers
4330 * Floating Point Hardware:: Floating point hardware
4334 @section Expressions
4337 @code{print} and many other @value{GDBN} commands accept an expression and
4338 compute its value. Any kind of constant, variable or operator defined
4339 by the programming language you are using is valid in an expression in
4340 @value{GDBN}. This includes conditional expressions, function calls, casts
4341 and string constants. It unfortunately does not include symbols defined
4342 by preprocessor @code{#define} commands.
4344 @value{GDBN} supports array constants in expressions input by
4345 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
4346 you can use the command @code{print @{1, 2, 3@}} to build up an array in
4347 memory that is @code{malloc}ed in the target program.
4349 Because C is so widespread, most of the expressions shown in examples in
4350 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
4351 Languages}, for information on how to use expressions in other
4354 In this section, we discuss operators that you can use in @value{GDBN}
4355 expressions regardless of your programming language.
4357 Casts are supported in all languages, not just in C, because it is so
4358 useful to cast a number into a pointer in order to examine a structure
4359 at that address in memory.
4360 @c FIXME: casts supported---Mod2 true?
4362 @value{GDBN} supports these operators, in addition to those common
4363 to programming languages:
4367 @samp{@@} is a binary operator for treating parts of memory as arrays.
4368 @xref{Arrays, ,Artificial arrays}, for more information.
4371 @samp{::} allows you to specify a variable in terms of the file or
4372 function where it is defined. @xref{Variables, ,Program variables}.
4374 @cindex @{@var{type}@}
4375 @cindex type casting memory
4376 @cindex memory, viewing as typed object
4377 @cindex casts, to view memory
4378 @item @{@var{type}@} @var{addr}
4379 Refers to an object of type @var{type} stored at address @var{addr} in
4380 memory. @var{addr} may be any expression whose value is an integer or
4381 pointer (but parentheses are required around binary operators, just as in
4382 a cast). This construct is allowed regardless of what kind of data is
4383 normally supposed to reside at @var{addr}.
4387 @section Program variables
4389 The most common kind of expression to use is the name of a variable
4392 Variables in expressions are understood in the selected stack frame
4393 (@pxref{Selection, ,Selecting a frame}); they must be either:
4397 global (or file-static)
4404 visible according to the scope rules of the
4405 programming language from the point of execution in that frame
4408 @noindent This means that in the function
4423 you can examine and use the variable @code{a} whenever your program is
4424 executing within the function @code{foo}, but you can only use or
4425 examine the variable @code{b} while your program is executing inside
4426 the block where @code{b} is declared.
4428 @cindex variable name conflict
4429 There is an exception: you can refer to a variable or function whose
4430 scope is a single source file even if the current execution point is not
4431 in this file. But it is possible to have more than one such variable or
4432 function with the same name (in different source files). If that
4433 happens, referring to that name has unpredictable effects. If you wish,
4434 you can specify a static variable in a particular function or file,
4435 using the colon-colon notation:
4437 @cindex colon-colon, context for variables/functions
4439 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4440 @cindex @code{::}, context for variables/functions
4443 @var{file}::@var{variable}
4444 @var{function}::@var{variable}
4448 Here @var{file} or @var{function} is the name of the context for the
4449 static @var{variable}. In the case of file names, you can use quotes to
4450 make sure @value{GDBN} parses the file name as a single word---for example,
4451 to print a global value of @code{x} defined in @file{f2.c}:
4454 (@value{GDBP}) p 'f2.c'::x
4457 @cindex C++ scope resolution
4458 This use of @samp{::} is very rarely in conflict with the very similar
4459 use of the same notation in C++. @value{GDBN} also supports use of the C++
4460 scope resolution operator in @value{GDBN} expressions.
4461 @c FIXME: Um, so what happens in one of those rare cases where it's in
4464 @cindex wrong values
4465 @cindex variable values, wrong
4467 @emph{Warning:} Occasionally, a local variable may appear to have the
4468 wrong value at certain points in a function---just after entry to a new
4469 scope, and just before exit.
4471 You may see this problem when you are stepping by machine instructions.
4472 This is because, on most machines, it takes more than one instruction to
4473 set up a stack frame (including local variable definitions); if you are
4474 stepping by machine instructions, variables may appear to have the wrong
4475 values until the stack frame is completely built. On exit, it usually
4476 also takes more than one machine instruction to destroy a stack frame;
4477 after you begin stepping through that group of instructions, local
4478 variable definitions may be gone.
4480 This may also happen when the compiler does significant optimizations.
4481 To be sure of always seeing accurate values, turn off all optimization
4484 @cindex ``No symbol "foo" in current context''
4485 Another possible effect of compiler optimizations is to optimize
4486 unused variables out of existence, or assign variables to registers (as
4487 opposed to memory addresses). Depending on the support for such cases
4488 offered by the debug info format used by the compiler, @value{GDBN}
4489 might not be able to display values for such local variables. If that
4490 happens, @value{GDBN} will print a message like this:
4493 No symbol "foo" in current context.
4496 To solve such problems, either recompile without optimizations, or use a
4497 different debug info format, if the compiler supports several such
4498 formats. For example, @value{NGCC}, the @sc{gnu} C/C++ compiler usually
4499 supports the @samp{-gstabs} option. @samp{-gstabs} produces debug info
4500 in a format that is superior to formats such as COFF. You may be able
4501 to use DWARF-2 (@samp{-gdwarf-2}), which is also an effective form for
4502 debug info. See @ref{Debugging Options,,Options for Debugging Your
4503 Program or @sc{gnu} CC, gcc.info, Using @sc{gnu} CC}, for more
4508 @section Artificial arrays
4510 @cindex artificial array
4511 @kindex @@@r{, referencing memory as an array}
4512 It is often useful to print out several successive objects of the
4513 same type in memory; a section of an array, or an array of
4514 dynamically determined size for which only a pointer exists in the
4517 You can do this by referring to a contiguous span of memory as an
4518 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4519 operand of @samp{@@} should be the first element of the desired array
4520 and be an individual object. The right operand should be the desired length
4521 of the array. The result is an array value whose elements are all of
4522 the type of the left argument. The first element is actually the left
4523 argument; the second element comes from bytes of memory immediately
4524 following those that hold the first element, and so on. Here is an
4525 example. If a program says
4528 int *array = (int *) malloc (len * sizeof (int));
4532 you can print the contents of @code{array} with
4538 The left operand of @samp{@@} must reside in memory. Array values made
4539 with @samp{@@} in this way behave just like other arrays in terms of
4540 subscripting, and are coerced to pointers when used in expressions.
4541 Artificial arrays most often appear in expressions via the value history
4542 (@pxref{Value History, ,Value history}), after printing one out.
4544 Another way to create an artificial array is to use a cast.
4545 This re-interprets a value as if it were an array.
4546 The value need not be in memory:
4548 (@value{GDBP}) p/x (short[2])0x12345678
4549 $1 = @{0x1234, 0x5678@}
4552 As a convenience, if you leave the array length out (as in
4553 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
4554 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4556 (@value{GDBP}) p/x (short[])0x12345678
4557 $2 = @{0x1234, 0x5678@}
4560 Sometimes the artificial array mechanism is not quite enough; in
4561 moderately complex data structures, the elements of interest may not
4562 actually be adjacent---for example, if you are interested in the values
4563 of pointers in an array. One useful work-around in this situation is
4564 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4565 variables}) as a counter in an expression that prints the first
4566 interesting value, and then repeat that expression via @key{RET}. For
4567 instance, suppose you have an array @code{dtab} of pointers to
4568 structures, and you are interested in the values of a field @code{fv}
4569 in each structure. Here is an example of what you might type:
4579 @node Output Formats
4580 @section Output formats
4582 @cindex formatted output
4583 @cindex output formats
4584 By default, @value{GDBN} prints a value according to its data type. Sometimes
4585 this is not what you want. For example, you might want to print a number
4586 in hex, or a pointer in decimal. Or you might want to view data in memory
4587 at a certain address as a character string or as an instruction. To do
4588 these things, specify an @dfn{output format} when you print a value.
4590 The simplest use of output formats is to say how to print a value
4591 already computed. This is done by starting the arguments of the
4592 @code{print} command with a slash and a format letter. The format
4593 letters supported are:
4597 Regard the bits of the value as an integer, and print the integer in
4601 Print as integer in signed decimal.
4604 Print as integer in unsigned decimal.
4607 Print as integer in octal.
4610 Print as integer in binary. The letter @samp{t} stands for ``two''.
4611 @footnote{@samp{b} cannot be used because these format letters are also
4612 used with the @code{x} command, where @samp{b} stands for ``byte'';
4613 see @ref{Memory,,Examining memory}.}
4616 @cindex unknown address, locating
4617 Print as an address, both absolute in hexadecimal and as an offset from
4618 the nearest preceding symbol. You can use this format used to discover
4619 where (in what function) an unknown address is located:
4622 (@value{GDBP}) p/a 0x54320
4623 $3 = 0x54320 <_initialize_vx+396>
4627 Regard as an integer and print it as a character constant.
4630 Regard the bits of the value as a floating point number and print
4631 using typical floating point syntax.
4634 For example, to print the program counter in hex (@pxref{Registers}), type
4641 Note that no space is required before the slash; this is because command
4642 names in @value{GDBN} cannot contain a slash.
4644 To reprint the last value in the value history with a different format,
4645 you can use the @code{print} command with just a format and no
4646 expression. For example, @samp{p/x} reprints the last value in hex.
4649 @section Examining memory
4651 You can use the command @code{x} (for ``examine'') to examine memory in
4652 any of several formats, independently of your program's data types.
4654 @cindex examining memory
4656 @kindex x @r{(examine memory)}
4657 @item x/@var{nfu} @var{addr}
4660 Use the @code{x} command to examine memory.
4663 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4664 much memory to display and how to format it; @var{addr} is an
4665 expression giving the address where you want to start displaying memory.
4666 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4667 Several commands set convenient defaults for @var{addr}.
4670 @item @var{n}, the repeat count
4671 The repeat count is a decimal integer; the default is 1. It specifies
4672 how much memory (counting by units @var{u}) to display.
4673 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4676 @item @var{f}, the display format
4677 The display format is one of the formats used by @code{print},
4678 @samp{s} (null-terminated string), or @samp{i} (machine instruction).
4679 The default is @samp{x} (hexadecimal) initially.
4680 The default changes each time you use either @code{x} or @code{print}.
4682 @item @var{u}, the unit size
4683 The unit size is any of
4689 Halfwords (two bytes).
4691 Words (four bytes). This is the initial default.
4693 Giant words (eight bytes).
4696 Each time you specify a unit size with @code{x}, that size becomes the
4697 default unit the next time you use @code{x}. (For the @samp{s} and
4698 @samp{i} formats, the unit size is ignored and is normally not written.)
4700 @item @var{addr}, starting display address
4701 @var{addr} is the address where you want @value{GDBN} to begin displaying
4702 memory. The expression need not have a pointer value (though it may);
4703 it is always interpreted as an integer address of a byte of memory.
4704 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4705 @var{addr} is usually just after the last address examined---but several
4706 other commands also set the default address: @code{info breakpoints} (to
4707 the address of the last breakpoint listed), @code{info line} (to the
4708 starting address of a line), and @code{print} (if you use it to display
4709 a value from memory).
4712 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4713 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4714 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4715 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4716 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
4718 Since the letters indicating unit sizes are all distinct from the
4719 letters specifying output formats, you do not have to remember whether
4720 unit size or format comes first; either order works. The output
4721 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4722 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4724 Even though the unit size @var{u} is ignored for the formats @samp{s}
4725 and @samp{i}, you might still want to use a count @var{n}; for example,
4726 @samp{3i} specifies that you want to see three machine instructions,
4727 including any operands. The command @code{disassemble} gives an
4728 alternative way of inspecting machine instructions; see @ref{Machine
4729 Code,,Source and machine code}.
4731 All the defaults for the arguments to @code{x} are designed to make it
4732 easy to continue scanning memory with minimal specifications each time
4733 you use @code{x}. For example, after you have inspected three machine
4734 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4735 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4736 the repeat count @var{n} is used again; the other arguments default as
4737 for successive uses of @code{x}.
4739 @cindex @code{$_}, @code{$__}, and value history
4740 The addresses and contents printed by the @code{x} command are not saved
4741 in the value history because there is often too much of them and they
4742 would get in the way. Instead, @value{GDBN} makes these values available for
4743 subsequent use in expressions as values of the convenience variables
4744 @code{$_} and @code{$__}. After an @code{x} command, the last address
4745 examined is available for use in expressions in the convenience variable
4746 @code{$_}. The contents of that address, as examined, are available in
4747 the convenience variable @code{$__}.
4749 If the @code{x} command has a repeat count, the address and contents saved
4750 are from the last memory unit printed; this is not the same as the last
4751 address printed if several units were printed on the last line of output.
4754 @section Automatic display
4755 @cindex automatic display
4756 @cindex display of expressions
4758 If you find that you want to print the value of an expression frequently
4759 (to see how it changes), you might want to add it to the @dfn{automatic
4760 display list} so that @value{GDBN} prints its value each time your program stops.
4761 Each expression added to the list is given a number to identify it;
4762 to remove an expression from the list, you specify that number.
4763 The automatic display looks like this:
4767 3: bar[5] = (struct hack *) 0x3804
4771 This display shows item numbers, expressions and their current values. As with
4772 displays you request manually using @code{x} or @code{print}, you can
4773 specify the output format you prefer; in fact, @code{display} decides
4774 whether to use @code{print} or @code{x} depending on how elaborate your
4775 format specification is---it uses @code{x} if you specify a unit size,
4776 or one of the two formats (@samp{i} and @samp{s}) that are only
4777 supported by @code{x}; otherwise it uses @code{print}.
4781 @item display @var{expr}
4782 Add the expression @var{expr} to the list of expressions to display
4783 each time your program stops. @xref{Expressions, ,Expressions}.
4785 @code{display} does not repeat if you press @key{RET} again after using it.
4787 @item display/@var{fmt} @var{expr}
4788 For @var{fmt} specifying only a display format and not a size or
4789 count, add the expression @var{expr} to the auto-display list but
4790 arrange to display it each time in the specified format @var{fmt}.
4791 @xref{Output Formats,,Output formats}.
4793 @item display/@var{fmt} @var{addr}
4794 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4795 number of units, add the expression @var{addr} as a memory address to
4796 be examined each time your program stops. Examining means in effect
4797 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4800 For example, @samp{display/i $pc} can be helpful, to see the machine
4801 instruction about to be executed each time execution stops (@samp{$pc}
4802 is a common name for the program counter; @pxref{Registers, ,Registers}).
4805 @kindex delete display
4807 @item undisplay @var{dnums}@dots{}
4808 @itemx delete display @var{dnums}@dots{}
4809 Remove item numbers @var{dnums} from the list of expressions to display.
4811 @code{undisplay} does not repeat if you press @key{RET} after using it.
4812 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4814 @kindex disable display
4815 @item disable display @var{dnums}@dots{}
4816 Disable the display of item numbers @var{dnums}. A disabled display
4817 item is not printed automatically, but is not forgotten. It may be
4818 enabled again later.
4820 @kindex enable display
4821 @item enable display @var{dnums}@dots{}
4822 Enable display of item numbers @var{dnums}. It becomes effective once
4823 again in auto display of its expression, until you specify otherwise.
4826 Display the current values of the expressions on the list, just as is
4827 done when your program stops.
4829 @kindex info display
4831 Print the list of expressions previously set up to display
4832 automatically, each one with its item number, but without showing the
4833 values. This includes disabled expressions, which are marked as such.
4834 It also includes expressions which would not be displayed right now
4835 because they refer to automatic variables not currently available.
4838 If a display expression refers to local variables, then it does not make
4839 sense outside the lexical context for which it was set up. Such an
4840 expression is disabled when execution enters a context where one of its
4841 variables is not defined. For example, if you give the command
4842 @code{display last_char} while inside a function with an argument
4843 @code{last_char}, @value{GDBN} displays this argument while your program
4844 continues to stop inside that function. When it stops elsewhere---where
4845 there is no variable @code{last_char}---the display is disabled
4846 automatically. The next time your program stops where @code{last_char}
4847 is meaningful, you can enable the display expression once again.
4849 @node Print Settings
4850 @section Print settings
4852 @cindex format options
4853 @cindex print settings
4854 @value{GDBN} provides the following ways to control how arrays, structures,
4855 and symbols are printed.
4858 These settings are useful for debugging programs in any language:
4861 @kindex set print address
4862 @item set print address
4863 @itemx set print address on
4864 @value{GDBN} prints memory addresses showing the location of stack
4865 traces, structure values, pointer values, breakpoints, and so forth,
4866 even when it also displays the contents of those addresses. The default
4867 is @code{on}. For example, this is what a stack frame display looks like with
4868 @code{set print address on}:
4873 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
4875 530 if (lquote != def_lquote)
4879 @item set print address off
4880 Do not print addresses when displaying their contents. For example,
4881 this is the same stack frame displayed with @code{set print address off}:
4885 (@value{GDBP}) set print addr off
4887 #0 set_quotes (lq="<<", rq=">>") at input.c:530
4888 530 if (lquote != def_lquote)
4892 You can use @samp{set print address off} to eliminate all machine
4893 dependent displays from the @value{GDBN} interface. For example, with
4894 @code{print address off}, you should get the same text for backtraces on
4895 all machines---whether or not they involve pointer arguments.
4897 @kindex show print address
4898 @item show print address
4899 Show whether or not addresses are to be printed.
4902 When @value{GDBN} prints a symbolic address, it normally prints the
4903 closest earlier symbol plus an offset. If that symbol does not uniquely
4904 identify the address (for example, it is a name whose scope is a single
4905 source file), you may need to clarify. One way to do this is with
4906 @code{info line}, for example @samp{info line *0x4537}. Alternately,
4907 you can set @value{GDBN} to print the source file and line number when
4908 it prints a symbolic address:
4911 @kindex set print symbol-filename
4912 @item set print symbol-filename on
4913 Tell @value{GDBN} to print the source file name and line number of a
4914 symbol in the symbolic form of an address.
4916 @item set print symbol-filename off
4917 Do not print source file name and line number of a symbol. This is the
4920 @kindex show print symbol-filename
4921 @item show print symbol-filename
4922 Show whether or not @value{GDBN} will print the source file name and
4923 line number of a symbol in the symbolic form of an address.
4926 Another situation where it is helpful to show symbol filenames and line
4927 numbers is when disassembling code; @value{GDBN} shows you the line
4928 number and source file that corresponds to each instruction.
4930 Also, you may wish to see the symbolic form only if the address being
4931 printed is reasonably close to the closest earlier symbol:
4934 @kindex set print max-symbolic-offset
4935 @item set print max-symbolic-offset @var{max-offset}
4936 Tell @value{GDBN} to only display the symbolic form of an address if the
4937 offset between the closest earlier symbol and the address is less than
4938 @var{max-offset}. The default is 0, which tells @value{GDBN}
4939 to always print the symbolic form of an address if any symbol precedes it.
4941 @kindex show print max-symbolic-offset
4942 @item show print max-symbolic-offset
4943 Ask how large the maximum offset is that @value{GDBN} prints in a
4947 @cindex wild pointer, interpreting
4948 @cindex pointer, finding referent
4949 If you have a pointer and you are not sure where it points, try
4950 @samp{set print symbol-filename on}. Then you can determine the name
4951 and source file location of the variable where it points, using
4952 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
4953 For example, here @value{GDBN} shows that a variable @code{ptt} points
4954 at another variable @code{t}, defined in @file{hi2.c}:
4957 (@value{GDBP}) set print symbol-filename on
4958 (@value{GDBP}) p/a ptt
4959 $4 = 0xe008 <t in hi2.c>
4963 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
4964 does not show the symbol name and filename of the referent, even with
4965 the appropriate @code{set print} options turned on.
4968 Other settings control how different kinds of objects are printed:
4971 @kindex set print array
4972 @item set print array
4973 @itemx set print array on
4974 Pretty print arrays. This format is more convenient to read,
4975 but uses more space. The default is off.
4977 @item set print array off
4978 Return to compressed format for arrays.
4980 @kindex show print array
4981 @item show print array
4982 Show whether compressed or pretty format is selected for displaying
4985 @kindex set print elements
4986 @item set print elements @var{number-of-elements}
4987 Set a limit on how many elements of an array @value{GDBN} will print.
4988 If @value{GDBN} is printing a large array, it stops printing after it has
4989 printed the number of elements set by the @code{set print elements} command.
4990 This limit also applies to the display of strings.
4991 When @value{GDBN} starts, this limit is set to 200.
4992 Setting @var{number-of-elements} to zero means that the printing is unlimited.
4994 @kindex show print elements
4995 @item show print elements
4996 Display the number of elements of a large array that @value{GDBN} will print.
4997 If the number is 0, then the printing is unlimited.
4999 @kindex set print null-stop
5000 @item set print null-stop
5001 Cause @value{GDBN} to stop printing the characters of an array when the first
5002 @sc{null} is encountered. This is useful when large arrays actually
5003 contain only short strings.
5006 @kindex set print pretty
5007 @item set print pretty on
5008 Cause @value{GDBN} to print structures in an indented format with one member
5009 per line, like this:
5024 @item set print pretty off
5025 Cause @value{GDBN} to print structures in a compact format, like this:
5029 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
5030 meat = 0x54 "Pork"@}
5035 This is the default format.
5037 @kindex show print pretty
5038 @item show print pretty
5039 Show which format @value{GDBN} is using to print structures.
5041 @kindex set print sevenbit-strings
5042 @item set print sevenbit-strings on
5043 Print using only seven-bit characters; if this option is set,
5044 @value{GDBN} displays any eight-bit characters (in strings or
5045 character values) using the notation @code{\}@var{nnn}. This setting is
5046 best if you are working in English (@sc{ascii}) and you use the
5047 high-order bit of characters as a marker or ``meta'' bit.
5049 @item set print sevenbit-strings off
5050 Print full eight-bit characters. This allows the use of more
5051 international character sets, and is the default.
5053 @kindex show print sevenbit-strings
5054 @item show print sevenbit-strings
5055 Show whether or not @value{GDBN} is printing only seven-bit characters.
5057 @kindex set print union
5058 @item set print union on
5059 Tell @value{GDBN} to print unions which are contained in structures. This
5060 is the default setting.
5062 @item set print union off
5063 Tell @value{GDBN} not to print unions which are contained in structures.
5065 @kindex show print union
5066 @item show print union
5067 Ask @value{GDBN} whether or not it will print unions which are contained in
5070 For example, given the declarations
5073 typedef enum @{Tree, Bug@} Species;
5074 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
5075 typedef enum @{Caterpillar, Cocoon, Butterfly@}
5086 struct thing foo = @{Tree, @{Acorn@}@};
5090 with @code{set print union on} in effect @samp{p foo} would print
5093 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
5097 and with @code{set print union off} in effect it would print
5100 $1 = @{it = Tree, form = @{...@}@}
5106 These settings are of interest when debugging C++ programs:
5110 @kindex set print demangle
5111 @item set print demangle
5112 @itemx set print demangle on
5113 Print C++ names in their source form rather than in the encoded
5114 (``mangled'') form passed to the assembler and linker for type-safe
5115 linkage. The default is on.
5117 @kindex show print demangle
5118 @item show print demangle
5119 Show whether C++ names are printed in mangled or demangled form.
5121 @kindex set print asm-demangle
5122 @item set print asm-demangle
5123 @itemx set print asm-demangle on
5124 Print C++ names in their source form rather than their mangled form, even
5125 in assembler code printouts such as instruction disassemblies.
5128 @kindex show print asm-demangle
5129 @item show print asm-demangle
5130 Show whether C++ names in assembly listings are printed in mangled
5133 @kindex set demangle-style
5134 @cindex C++ symbol decoding style
5135 @cindex symbol decoding style, C++
5136 @item set demangle-style @var{style}
5137 Choose among several encoding schemes used by different compilers to
5138 represent C++ names. The choices for @var{style} are currently:
5142 Allow @value{GDBN} to choose a decoding style by inspecting your program.
5145 Decode based on the @sc{gnu} C++ compiler (@code{g++}) encoding algorithm.
5146 This is the default.
5149 Decode based on the HP ANSI C++ (@code{aCC}) encoding algorithm.
5152 Decode based on the Lucid C++ compiler (@code{lcc}) encoding algorithm.
5155 Decode using the algorithm in the @cite{C++ Annotated Reference Manual}.
5156 @strong{Warning:} this setting alone is not sufficient to allow
5157 debugging @code{cfront}-generated executables. @value{GDBN} would
5158 require further enhancement to permit that.
5161 If you omit @var{style}, you will see a list of possible formats.
5163 @kindex show demangle-style
5164 @item show demangle-style
5165 Display the encoding style currently in use for decoding C++ symbols.
5167 @kindex set print object
5168 @item set print object
5169 @itemx set print object on
5170 When displaying a pointer to an object, identify the @emph{actual}
5171 (derived) type of the object rather than the @emph{declared} type, using
5172 the virtual function table.
5174 @item set print object off
5175 Display only the declared type of objects, without reference to the
5176 virtual function table. This is the default setting.
5178 @kindex show print object
5179 @item show print object
5180 Show whether actual, or declared, object types are displayed.
5182 @kindex set print static-members
5183 @item set print static-members
5184 @itemx set print static-members on
5185 Print static members when displaying a C++ object. The default is on.
5187 @item set print static-members off
5188 Do not print static members when displaying a C++ object.
5190 @kindex show print static-members
5191 @item show print static-members
5192 Show whether C++ static members are printed, or not.
5194 @c These don't work with HP ANSI C++ yet.
5195 @kindex set print vtbl
5196 @item set print vtbl
5197 @itemx set print vtbl on
5198 Pretty print C++ virtual function tables. The default is off.
5199 (The @code{vtbl} commands do not work on programs compiled with the HP
5200 ANSI C++ compiler (@code{aCC}).)
5202 @item set print vtbl off
5203 Do not pretty print C++ virtual function tables.
5205 @kindex show print vtbl
5206 @item show print vtbl
5207 Show whether C++ virtual function tables are pretty printed, or not.
5211 @section Value history
5213 @cindex value history
5214 Values printed by the @code{print} command are saved in the @value{GDBN}
5215 @dfn{value history}. This allows you to refer to them in other expressions.
5216 Values are kept until the symbol table is re-read or discarded
5217 (for example with the @code{file} or @code{symbol-file} commands).
5218 When the symbol table changes, the value history is discarded,
5219 since the values may contain pointers back to the types defined in the
5224 @cindex history number
5225 The values printed are given @dfn{history numbers} by which you can
5226 refer to them. These are successive integers starting with one.
5227 @code{print} shows you the history number assigned to a value by
5228 printing @samp{$@var{num} = } before the value; here @var{num} is the
5231 To refer to any previous value, use @samp{$} followed by the value's
5232 history number. The way @code{print} labels its output is designed to
5233 remind you of this. Just @code{$} refers to the most recent value in
5234 the history, and @code{$$} refers to the value before that.
5235 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
5236 is the value just prior to @code{$$}, @code{$$1} is equivalent to
5237 @code{$$}, and @code{$$0} is equivalent to @code{$}.
5239 For example, suppose you have just printed a pointer to a structure and
5240 want to see the contents of the structure. It suffices to type
5246 If you have a chain of structures where the component @code{next} points
5247 to the next one, you can print the contents of the next one with this:
5254 You can print successive links in the chain by repeating this
5255 command---which you can do by just typing @key{RET}.
5257 Note that the history records values, not expressions. If the value of
5258 @code{x} is 4 and you type these commands:
5266 then the value recorded in the value history by the @code{print} command
5267 remains 4 even though the value of @code{x} has changed.
5272 Print the last ten values in the value history, with their item numbers.
5273 This is like @samp{p@ $$9} repeated ten times, except that @code{show
5274 values} does not change the history.
5276 @item show values @var{n}
5277 Print ten history values centered on history item number @var{n}.
5280 Print ten history values just after the values last printed. If no more
5281 values are available, @code{show values +} produces no display.
5284 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
5285 same effect as @samp{show values +}.
5287 @node Convenience Vars
5288 @section Convenience variables
5290 @cindex convenience variables
5291 @value{GDBN} provides @dfn{convenience variables} that you can use within
5292 @value{GDBN} to hold on to a value and refer to it later. These variables
5293 exist entirely within @value{GDBN}; they are not part of your program, and
5294 setting a convenience variable has no direct effect on further execution
5295 of your program. That is why you can use them freely.
5297 Convenience variables are prefixed with @samp{$}. Any name preceded by
5298 @samp{$} can be used for a convenience variable, unless it is one of
5299 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
5300 (Value history references, in contrast, are @emph{numbers} preceded
5301 by @samp{$}. @xref{Value History, ,Value history}.)
5303 You can save a value in a convenience variable with an assignment
5304 expression, just as you would set a variable in your program.
5308 set $foo = *object_ptr
5312 would save in @code{$foo} the value contained in the object pointed to by
5315 Using a convenience variable for the first time creates it, but its
5316 value is @code{void} until you assign a new value. You can alter the
5317 value with another assignment at any time.
5319 Convenience variables have no fixed types. You can assign a convenience
5320 variable any type of value, including structures and arrays, even if
5321 that variable already has a value of a different type. The convenience
5322 variable, when used as an expression, has the type of its current value.
5325 @kindex show convenience
5326 @item show convenience
5327 Print a list of convenience variables used so far, and their values.
5328 Abbreviated @code{show conv}.
5331 One of the ways to use a convenience variable is as a counter to be
5332 incremented or a pointer to be advanced. For example, to print
5333 a field from successive elements of an array of structures:
5337 print bar[$i++]->contents
5341 Repeat that command by typing @key{RET}.
5343 Some convenience variables are created automatically by @value{GDBN} and given
5344 values likely to be useful.
5347 @vindex $_@r{, convenience variable}
5349 The variable @code{$_} is automatically set by the @code{x} command to
5350 the last address examined (@pxref{Memory, ,Examining memory}). Other
5351 commands which provide a default address for @code{x} to examine also
5352 set @code{$_} to that address; these commands include @code{info line}
5353 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
5354 except when set by the @code{x} command, in which case it is a pointer
5355 to the type of @code{$__}.
5357 @vindex $__@r{, convenience variable}
5359 The variable @code{$__} is automatically set by the @code{x} command
5360 to the value found in the last address examined. Its type is chosen
5361 to match the format in which the data was printed.
5364 @vindex $_exitcode@r{, convenience variable}
5365 The variable @code{$_exitcode} is automatically set to the exit code when
5366 the program being debugged terminates.
5369 On HP-UX systems, if you refer to a function or variable name that
5370 begins with a dollar sign, @value{GDBN} searches for a user or system
5371 name first, before it searches for a convenience variable.
5377 You can refer to machine register contents, in expressions, as variables
5378 with names starting with @samp{$}. The names of registers are different
5379 for each machine; use @code{info registers} to see the names used on
5383 @kindex info registers
5384 @item info registers
5385 Print the names and values of all registers except floating-point
5386 registers (in the selected stack frame).
5388 @kindex info all-registers
5389 @cindex floating point registers
5390 @item info all-registers
5391 Print the names and values of all registers, including floating-point
5394 @item info registers @var{regname} @dots{}
5395 Print the @dfn{relativized} value of each specified register @var{regname}.
5396 As discussed in detail below, register values are normally relative to
5397 the selected stack frame. @var{regname} may be any register name valid on
5398 the machine you are using, with or without the initial @samp{$}.
5401 @value{GDBN} has four ``standard'' register names that are available (in
5402 expressions) on most machines---whenever they do not conflict with an
5403 architecture's canonical mnemonics for registers. The register names
5404 @code{$pc} and @code{$sp} are used for the program counter register and
5405 the stack pointer. @code{$fp} is used for a register that contains a
5406 pointer to the current stack frame, and @code{$ps} is used for a
5407 register that contains the processor status. For example,
5408 you could print the program counter in hex with
5415 or print the instruction to be executed next with
5422 or add four to the stack pointer@footnote{This is a way of removing
5423 one word from the stack, on machines where stacks grow downward in
5424 memory (most machines, nowadays). This assumes that the innermost
5425 stack frame is selected; setting @code{$sp} is not allowed when other
5426 stack frames are selected. To pop entire frames off the stack,
5427 regardless of machine architecture, use @code{return};
5428 see @ref{Returning, ,Returning from a function}.} with
5434 Whenever possible, these four standard register names are available on
5435 your machine even though the machine has different canonical mnemonics,
5436 so long as there is no conflict. The @code{info registers} command
5437 shows the canonical names. For example, on the SPARC, @code{info
5438 registers} displays the processor status register as @code{$psr} but you
5439 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
5440 is an alias for the @sc{eflags} register.
5442 @value{GDBN} always considers the contents of an ordinary register as an
5443 integer when the register is examined in this way. Some machines have
5444 special registers which can hold nothing but floating point; these
5445 registers are considered to have floating point values. There is no way
5446 to refer to the contents of an ordinary register as floating point value
5447 (although you can @emph{print} it as a floating point value with
5448 @samp{print/f $@var{regname}}).
5450 Some registers have distinct ``raw'' and ``virtual'' data formats. This
5451 means that the data format in which the register contents are saved by
5452 the operating system is not the same one that your program normally
5453 sees. For example, the registers of the 68881 floating point
5454 coprocessor are always saved in ``extended'' (raw) format, but all C
5455 programs expect to work with ``double'' (virtual) format. In such
5456 cases, @value{GDBN} normally works with the virtual format only (the format
5457 that makes sense for your program), but the @code{info registers} command
5458 prints the data in both formats.
5460 Normally, register values are relative to the selected stack frame
5461 (@pxref{Selection, ,Selecting a frame}). This means that you get the
5462 value that the register would contain if all stack frames farther in
5463 were exited and their saved registers restored. In order to see the
5464 true contents of hardware registers, you must select the innermost
5465 frame (with @samp{frame 0}).
5467 However, @value{GDBN} must deduce where registers are saved, from the machine
5468 code generated by your compiler. If some registers are not saved, or if
5469 @value{GDBN} is unable to locate the saved registers, the selected stack
5470 frame makes no difference.
5472 @node Floating Point Hardware
5473 @section Floating point hardware
5474 @cindex floating point
5476 Depending on the configuration, @value{GDBN} may be able to give
5477 you more information about the status of the floating point hardware.
5482 Display hardware-dependent information about the floating
5483 point unit. The exact contents and layout vary depending on the
5484 floating point chip. Currently, @samp{info float} is supported on
5485 the ARM and x86 machines.
5489 @chapter Using @value{GDBN} with Different Languages
5492 Although programming languages generally have common aspects, they are
5493 rarely expressed in the same manner. For instance, in ANSI C,
5494 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
5495 Modula-2, it is accomplished by @code{p^}. Values can also be
5496 represented (and displayed) differently. Hex numbers in C appear as
5497 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
5499 @cindex working language
5500 Language-specific information is built into @value{GDBN} for some languages,
5501 allowing you to express operations like the above in your program's
5502 native language, and allowing @value{GDBN} to output values in a manner
5503 consistent with the syntax of your program's native language. The
5504 language you use to build expressions is called the @dfn{working
5508 * Setting:: Switching between source languages
5509 * Show:: Displaying the language
5510 * Checks:: Type and range checks
5511 * Support:: Supported languages
5515 @section Switching between source languages
5517 There are two ways to control the working language---either have @value{GDBN}
5518 set it automatically, or select it manually yourself. You can use the
5519 @code{set language} command for either purpose. On startup, @value{GDBN}
5520 defaults to setting the language automatically. The working language is
5521 used to determine how expressions you type are interpreted, how values
5524 In addition to the working language, every source file that
5525 @value{GDBN} knows about has its own working language. For some object
5526 file formats, the compiler might indicate which language a particular
5527 source file is in. However, most of the time @value{GDBN} infers the
5528 language from the name of the file. The language of a source file
5529 controls whether C++ names are demangled---this way @code{backtrace} can
5530 show each frame appropriately for its own language. There is no way to
5531 set the language of a source file from within @value{GDBN}, but you can
5532 set the language associated with a filename extension. @xref{Show, ,
5533 Displaying the language}.
5535 This is most commonly a problem when you use a program, such
5536 as @code{cfront} or @code{f2c}, that generates C but is written in
5537 another language. In that case, make the
5538 program use @code{#line} directives in its C output; that way
5539 @value{GDBN} will know the correct language of the source code of the original
5540 program, and will display that source code, not the generated C code.
5543 * Filenames:: Filename extensions and languages.
5544 * Manually:: Setting the working language manually
5545 * Automatically:: Having @value{GDBN} infer the source language
5549 @subsection List of filename extensions and languages
5551 If a source file name ends in one of the following extensions, then
5552 @value{GDBN} infers that its language is the one indicated.
5577 Modula-2 source file
5581 Assembler source file. This actually behaves almost like C, but
5582 @value{GDBN} does not skip over function prologues when stepping.
5585 In addition, you may set the language associated with a filename
5586 extension. @xref{Show, , Displaying the language}.
5589 @subsection Setting the working language
5591 If you allow @value{GDBN} to set the language automatically,
5592 expressions are interpreted the same way in your debugging session and
5595 @kindex set language
5596 If you wish, you may set the language manually. To do this, issue the
5597 command @samp{set language @var{lang}}, where @var{lang} is the name of
5599 @code{c} or @code{modula-2}.
5600 For a list of the supported languages, type @samp{set language}.
5602 Setting the language manually prevents @value{GDBN} from updating the working
5603 language automatically. This can lead to confusion if you try
5604 to debug a program when the working language is not the same as the
5605 source language, when an expression is acceptable to both
5606 languages---but means different things. For instance, if the current
5607 source file were written in C, and @value{GDBN} was parsing Modula-2, a
5615 might not have the effect you intended. In C, this means to add
5616 @code{b} and @code{c} and place the result in @code{a}. The result
5617 printed would be the value of @code{a}. In Modula-2, this means to compare
5618 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
5621 @subsection Having @value{GDBN} infer the source language
5623 To have @value{GDBN} set the working language automatically, use
5624 @samp{set language local} or @samp{set language auto}. @value{GDBN}
5625 then infers the working language. That is, when your program stops in a
5626 frame (usually by encountering a breakpoint), @value{GDBN} sets the
5627 working language to the language recorded for the function in that
5628 frame. If the language for a frame is unknown (that is, if the function
5629 or block corresponding to the frame was defined in a source file that
5630 does not have a recognized extension), the current working language is
5631 not changed, and @value{GDBN} issues a warning.
5633 This may not seem necessary for most programs, which are written
5634 entirely in one source language. However, program modules and libraries
5635 written in one source language can be used by a main program written in
5636 a different source language. Using @samp{set language auto} in this
5637 case frees you from having to set the working language manually.
5640 @section Displaying the language
5642 The following commands help you find out which language is the
5643 working language, and also what language source files were written in.
5645 @kindex show language
5646 @kindex info frame@r{, show the source language}
5647 @kindex info source@r{, show the source language}
5650 Display the current working language. This is the
5651 language you can use with commands such as @code{print} to
5652 build and compute expressions that may involve variables in your program.
5655 Display the source language for this frame. This language becomes the
5656 working language if you use an identifier from this frame.
5657 @xref{Frame Info, ,Information about a frame}, to identify the other
5658 information listed here.
5661 Display the source language of this source file.
5662 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
5663 information listed here.
5666 In unusual circumstances, you may have source files with extensions
5667 not in the standard list. You can then set the extension associated
5668 with a language explicitly:
5670 @kindex set extension-language
5671 @kindex info extensions
5673 @item set extension-language @var{.ext} @var{language}
5674 Set source files with extension @var{.ext} to be assumed to be in
5675 the source language @var{language}.
5677 @item info extensions
5678 List all the filename extensions and the associated languages.
5682 @section Type and range checking
5685 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
5686 checking are included, but they do not yet have any effect. This
5687 section documents the intended facilities.
5689 @c FIXME remove warning when type/range code added
5691 Some languages are designed to guard you against making seemingly common
5692 errors through a series of compile- and run-time checks. These include
5693 checking the type of arguments to functions and operators, and making
5694 sure mathematical overflows are caught at run time. Checks such as
5695 these help to ensure a program's correctness once it has been compiled
5696 by eliminating type mismatches, and providing active checks for range
5697 errors when your program is running.
5699 @value{GDBN} can check for conditions like the above if you wish.
5700 Although @value{GDBN} does not check the statements in your program, it
5701 can check expressions entered directly into @value{GDBN} for evaluation via
5702 the @code{print} command, for example. As with the working language,
5703 @value{GDBN} can also decide whether or not to check automatically based on
5704 your program's source language. @xref{Support, ,Supported languages},
5705 for the default settings of supported languages.
5708 * Type Checking:: An overview of type checking
5709 * Range Checking:: An overview of range checking
5712 @cindex type checking
5713 @cindex checks, type
5715 @subsection An overview of type checking
5717 Some languages, such as Modula-2, are strongly typed, meaning that the
5718 arguments to operators and functions have to be of the correct type,
5719 otherwise an error occurs. These checks prevent type mismatch
5720 errors from ever causing any run-time problems. For example,
5728 The second example fails because the @code{CARDINAL} 1 is not
5729 type-compatible with the @code{REAL} 2.3.
5731 For the expressions you use in @value{GDBN} commands, you can tell the
5732 @value{GDBN} type checker to skip checking;
5733 to treat any mismatches as errors and abandon the expression;
5734 or to only issue warnings when type mismatches occur,
5735 but evaluate the expression anyway. When you choose the last of
5736 these, @value{GDBN} evaluates expressions like the second example above, but
5737 also issues a warning.
5739 Even if you turn type checking off, there may be other reasons
5740 related to type that prevent @value{GDBN} from evaluating an expression.
5741 For instance, @value{GDBN} does not know how to add an @code{int} and
5742 a @code{struct foo}. These particular type errors have nothing to do
5743 with the language in use, and usually arise from expressions, such as
5744 the one described above, which make little sense to evaluate anyway.
5746 Each language defines to what degree it is strict about type. For
5747 instance, both Modula-2 and C require the arguments to arithmetical
5748 operators to be numbers. In C, enumerated types and pointers can be
5749 represented as numbers, so that they are valid arguments to mathematical
5750 operators. @xref{Support, ,Supported languages}, for further
5751 details on specific languages.
5753 @value{GDBN} provides some additional commands for controlling the type checker:
5755 @kindex set check@r{, type}
5756 @kindex set check type
5757 @kindex show check type
5759 @item set check type auto
5760 Set type checking on or off based on the current working language.
5761 @xref{Support, ,Supported languages}, for the default settings for
5764 @item set check type on
5765 @itemx set check type off
5766 Set type checking on or off, overriding the default setting for the
5767 current working language. Issue a warning if the setting does not
5768 match the language default. If any type mismatches occur in
5769 evaluating an expression while type checking is on, @value{GDBN} prints a
5770 message and aborts evaluation of the expression.
5772 @item set check type warn
5773 Cause the type checker to issue warnings, but to always attempt to
5774 evaluate the expression. Evaluating the expression may still
5775 be impossible for other reasons. For example, @value{GDBN} cannot add
5776 numbers and structures.
5779 Show the current setting of the type checker, and whether or not @value{GDBN}
5780 is setting it automatically.
5783 @cindex range checking
5784 @cindex checks, range
5785 @node Range Checking
5786 @subsection An overview of range checking
5788 In some languages (such as Modula-2), it is an error to exceed the
5789 bounds of a type; this is enforced with run-time checks. Such range
5790 checking is meant to ensure program correctness by making sure
5791 computations do not overflow, or indices on an array element access do
5792 not exceed the bounds of the array.
5794 For expressions you use in @value{GDBN} commands, you can tell
5795 @value{GDBN} to treat range errors in one of three ways: ignore them,
5796 always treat them as errors and abandon the expression, or issue
5797 warnings but evaluate the expression anyway.
5799 A range error can result from numerical overflow, from exceeding an
5800 array index bound, or when you type a constant that is not a member
5801 of any type. Some languages, however, do not treat overflows as an
5802 error. In many implementations of C, mathematical overflow causes the
5803 result to ``wrap around'' to lower values---for example, if @var{m} is
5804 the largest integer value, and @var{s} is the smallest, then
5807 @var{m} + 1 @result{} @var{s}
5810 This, too, is specific to individual languages, and in some cases
5811 specific to individual compilers or machines. @xref{Support, ,
5812 Supported languages}, for further details on specific languages.
5814 @value{GDBN} provides some additional commands for controlling the range checker:
5816 @kindex set check@r{, range}
5817 @kindex set check range
5818 @kindex show check range
5820 @item set check range auto
5821 Set range checking on or off based on the current working language.
5822 @xref{Support, ,Supported languages}, for the default settings for
5825 @item set check range on
5826 @itemx set check range off
5827 Set range checking on or off, overriding the default setting for the
5828 current working language. A warning is issued if the setting does not
5829 match the language default. If a range error occurs and range checking is on,
5830 then a message is printed and evaluation of the expression is aborted.
5832 @item set check range warn
5833 Output messages when the @value{GDBN} range checker detects a range error,
5834 but attempt to evaluate the expression anyway. Evaluating the
5835 expression may still be impossible for other reasons, such as accessing
5836 memory that the process does not own (a typical example from many Unix
5840 Show the current setting of the range checker, and whether or not it is
5841 being set automatically by @value{GDBN}.
5845 @section Supported languages
5847 @value{GDBN} supports C, C++, Fortran, Java, Chill, assembly, and Modula-2.
5848 @c This is false ...
5849 Some @value{GDBN} features may be used in expressions regardless of the
5850 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
5851 and the @samp{@{type@}addr} construct (@pxref{Expressions,
5852 ,Expressions}) can be used with the constructs of any supported
5855 The following sections detail to what degree each source language is
5856 supported by @value{GDBN}. These sections are not meant to be language
5857 tutorials or references, but serve only as a reference guide to what the
5858 @value{GDBN} expression parser accepts, and what input and output
5859 formats should look like for different languages. There are many good
5860 books written on each of these languages; please look to these for a
5861 language reference or tutorial.
5865 * Modula-2:: Modula-2
5870 @subsection C and C++
5873 @cindex expressions in C or C++
5875 Since C and C++ are so closely related, many features of @value{GDBN} apply
5876 to both languages. Whenever this is the case, we discuss those languages
5880 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
5881 @cindex @sc{gnu} C++
5882 The C++ debugging facilities are jointly implemented by the C++
5883 compiler and @value{GDBN}. Therefore, to debug your C++ code
5884 effectively, you must compile your C++ programs with a supported
5885 C++ compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C++
5886 compiler (@code{aCC}).
5888 For best results when using @sc{gnu} C++, use the stabs debugging
5889 format. You can select that format explicitly with the @code{g++}
5890 command-line options @samp{-gstabs} or @samp{-gstabs+}. See
5891 @ref{Debugging Options,,Options for Debugging Your Program or @sc{gnu}
5892 CC, gcc.info, Using @sc{gnu} CC}, for more information.
5895 * C Operators:: C and C++ operators
5896 * C Constants:: C and C++ constants
5897 * C plus plus expressions:: C++ expressions
5898 * C Defaults:: Default settings for C and C++
5899 * C Checks:: C and C++ type and range checks
5900 * Debugging C:: @value{GDBN} and C
5901 * Debugging C plus plus:: @value{GDBN} features for C++
5905 @subsubsection C and C++ operators
5907 @cindex C and C++ operators
5909 Operators must be defined on values of specific types. For instance,
5910 @code{+} is defined on numbers, but not on structures. Operators are
5911 often defined on groups of types.
5913 For the purposes of C and C++, the following definitions hold:
5918 @emph{Integral types} include @code{int} with any of its storage-class
5919 specifiers; @code{char}; @code{enum}; and, for C++, @code{bool}.
5922 @emph{Floating-point types} include @code{float}, @code{double}, and
5923 @code{long double} (if supported by the target platform).
5926 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
5929 @emph{Scalar types} include all of the above.
5934 The following operators are supported. They are listed here
5935 in order of increasing precedence:
5939 The comma or sequencing operator. Expressions in a comma-separated list
5940 are evaluated from left to right, with the result of the entire
5941 expression being the last expression evaluated.
5944 Assignment. The value of an assignment expression is the value
5945 assigned. Defined on scalar types.
5948 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
5949 and translated to @w{@code{@var{a} = @var{a op b}}}.
5950 @w{@code{@var{op}=}} and @code{=} have the same precedence.
5951 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
5952 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
5955 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
5956 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
5960 Logical @sc{or}. Defined on integral types.
5963 Logical @sc{and}. Defined on integral types.
5966 Bitwise @sc{or}. Defined on integral types.
5969 Bitwise exclusive-@sc{or}. Defined on integral types.
5972 Bitwise @sc{and}. Defined on integral types.
5975 Equality and inequality. Defined on scalar types. The value of these
5976 expressions is 0 for false and non-zero for true.
5978 @item <@r{, }>@r{, }<=@r{, }>=
5979 Less than, greater than, less than or equal, greater than or equal.
5980 Defined on scalar types. The value of these expressions is 0 for false
5981 and non-zero for true.
5984 left shift, and right shift. Defined on integral types.
5987 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5990 Addition and subtraction. Defined on integral types, floating-point types and
5993 @item *@r{, }/@r{, }%
5994 Multiplication, division, and modulus. Multiplication and division are
5995 defined on integral and floating-point types. Modulus is defined on
5999 Increment and decrement. When appearing before a variable, the
6000 operation is performed before the variable is used in an expression;
6001 when appearing after it, the variable's value is used before the
6002 operation takes place.
6005 Pointer dereferencing. Defined on pointer types. Same precedence as
6009 Address operator. Defined on variables. Same precedence as @code{++}.
6011 For debugging C++, @value{GDBN} implements a use of @samp{&} beyond what is
6012 allowed in the C++ language itself: you can use @samp{&(&@var{ref})}
6013 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
6014 where a C++ reference variable (declared with @samp{&@var{ref}}) is
6018 Negative. Defined on integral and floating-point types. Same
6019 precedence as @code{++}.
6022 Logical negation. Defined on integral types. Same precedence as
6026 Bitwise complement operator. Defined on integral types. Same precedence as
6031 Structure member, and pointer-to-structure member. For convenience,
6032 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
6033 pointer based on the stored type information.
6034 Defined on @code{struct} and @code{union} data.
6037 Dereferences of pointers to members.
6040 Array indexing. @code{@var{a}[@var{i}]} is defined as
6041 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
6044 Function parameter list. Same precedence as @code{->}.
6047 C++ scope resolution operator. Defined on @code{struct}, @code{union},
6048 and @code{class} types.
6051 Doubled colons also represent the @value{GDBN} scope operator
6052 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
6056 If an operator is redefined in the user code, @value{GDBN} usually
6057 attempts to invoke the redefined version instead of using the operator's
6065 @subsubsection C and C++ constants
6067 @cindex C and C++ constants
6069 @value{GDBN} allows you to express the constants of C and C++ in the
6074 Integer constants are a sequence of digits. Octal constants are
6075 specified by a leading @samp{0} (i.e. zero), and hexadecimal constants by
6076 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
6077 @samp{l}, specifying that the constant should be treated as a
6081 Floating point constants are a sequence of digits, followed by a decimal
6082 point, followed by a sequence of digits, and optionally followed by an
6083 exponent. An exponent is of the form:
6084 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
6085 sequence of digits. The @samp{+} is optional for positive exponents.
6086 A floating-point constant may also end with a letter @samp{f} or
6087 @samp{F}, specifying that the constant should be treated as being of
6088 the @code{float} (as opposed to the default @code{double}) type; or with
6089 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
6093 Enumerated constants consist of enumerated identifiers, or their
6094 integral equivalents.
6097 Character constants are a single character surrounded by single quotes
6098 (@code{'}), or a number---the ordinal value of the corresponding character
6099 (usually its @sc{ascii} value). Within quotes, the single character may
6100 be represented by a letter or by @dfn{escape sequences}, which are of
6101 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
6102 of the character's ordinal value; or of the form @samp{\@var{x}}, where
6103 @samp{@var{x}} is a predefined special character---for example,
6104 @samp{\n} for newline.
6107 String constants are a sequence of character constants surrounded by
6108 double quotes (@code{"}). Any valid character constant (as described
6109 above) may appear. Double quotes within the string must be preceded by
6110 a backslash, so for instance @samp{"a\"b'c"} is a string of five
6114 Pointer constants are an integral value. You can also write pointers
6115 to constants using the C operator @samp{&}.
6118 Array constants are comma-separated lists surrounded by braces @samp{@{}
6119 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
6120 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
6121 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
6125 * C plus plus expressions::
6132 @node C plus plus expressions
6133 @subsubsection C++ expressions
6135 @cindex expressions in C++
6136 @value{GDBN} expression handling can interpret most C++ expressions.
6138 @cindex C++ support, not in @sc{coff}
6139 @cindex @sc{coff} versus C++
6140 @cindex C++ and object formats
6141 @cindex object formats and C++
6142 @cindex a.out and C++
6143 @cindex @sc{ecoff} and C++
6144 @cindex @sc{xcoff} and C++
6145 @cindex @sc{elf}/stabs and C++
6146 @cindex @sc{elf}/@sc{dwarf} and C++
6147 @c FIXME!! GDB may eventually be able to debug C++ using DWARF; check
6148 @c periodically whether this has happened...
6150 @emph{Warning:} @value{GDBN} can only debug C++ code if you use the
6151 proper compiler. Typically, C++ debugging depends on the use of
6152 additional debugging information in the symbol table, and thus requires
6153 special support. In particular, if your compiler generates a.out, MIPS
6154 @sc{ecoff}, RS/6000 @sc{xcoff}, or @sc{elf} with stabs extensions to the
6155 symbol table, these facilities are all available. (With @sc{gnu} CC,
6156 you can use the @samp{-gstabs} option to request stabs debugging
6157 extensions explicitly.) Where the object code format is standard
6158 @sc{coff} or @sc{dwarf} in @sc{elf}, on the other hand, most of the C++
6159 support in @value{GDBN} does @emph{not} work.
6164 @cindex member functions
6166 Member function calls are allowed; you can use expressions like
6169 count = aml->GetOriginal(x, y)
6172 @vindex this@r{, inside C@t{++} member functions}
6173 @cindex namespace in C++
6175 While a member function is active (in the selected stack frame), your
6176 expressions have the same namespace available as the member function;
6177 that is, @value{GDBN} allows implicit references to the class instance
6178 pointer @code{this} following the same rules as C++.
6180 @cindex call overloaded functions
6181 @cindex overloaded functions, calling
6182 @cindex type conversions in C++
6184 You can call overloaded functions; @value{GDBN} resolves the function
6185 call to the right definition, with some restrictions. @value{GDBN} does not
6186 perform overload resolution involving user-defined type conversions,
6187 calls to constructors, or instantiations of templates that do not exist
6188 in the program. It also cannot handle ellipsis argument lists or
6191 It does perform integral conversions and promotions, floating-point
6192 promotions, arithmetic conversions, pointer conversions, conversions of
6193 class objects to base classes, and standard conversions such as those of
6194 functions or arrays to pointers; it requires an exact match on the
6195 number of function arguments.
6197 Overload resolution is always performed, unless you have specified
6198 @code{set overload-resolution off}. @xref{Debugging C plus plus,
6199 ,@value{GDBN} features for C++}.
6201 You must specify @code{set overload-resolution off} in order to use an
6202 explicit function signature to call an overloaded function, as in
6204 p 'foo(char,int)'('x', 13)
6207 The @value{GDBN} command-completion facility can simplify this;
6208 see @ref{Completion, ,Command completion}.
6210 @cindex reference declarations
6212 @value{GDBN} understands variables declared as C++ references; you can use
6213 them in expressions just as you do in C++ source---they are automatically
6216 In the parameter list shown when @value{GDBN} displays a frame, the values of
6217 reference variables are not displayed (unlike other variables); this
6218 avoids clutter, since references are often used for large structures.
6219 The @emph{address} of a reference variable is always shown, unless
6220 you have specified @samp{set print address off}.
6223 @value{GDBN} supports the C++ name resolution operator @code{::}---your
6224 expressions can use it just as expressions in your program do. Since
6225 one scope may be defined in another, you can use @code{::} repeatedly if
6226 necessary, for example in an expression like
6227 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
6228 resolving name scope by reference to source files, in both C and C++
6229 debugging (@pxref{Variables, ,Program variables}).
6232 In addition, when used with HP's C++ compiler, @value{GDBN} supports
6233 calling virtual functions correctly, printing out virtual bases of
6234 objects, calling functions in a base subobject, casting objects, and
6235 invoking user-defined operators.
6238 @subsubsection C and C++ defaults
6240 @cindex C and C++ defaults
6242 If you allow @value{GDBN} to set type and range checking automatically, they
6243 both default to @code{off} whenever the working language changes to
6244 C or C++. This happens regardless of whether you or @value{GDBN}
6245 selects the working language.
6247 If you allow @value{GDBN} to set the language automatically, it
6248 recognizes source files whose names end with @file{.c}, @file{.C}, or
6249 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
6250 these files, it sets the working language to C or C++.
6251 @xref{Automatically, ,Having @value{GDBN} infer the source language},
6252 for further details.
6254 @c Type checking is (a) primarily motivated by Modula-2, and (b)
6255 @c unimplemented. If (b) changes, it might make sense to let this node
6256 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
6259 @subsubsection C and C++ type and range checks
6261 @cindex C and C++ checks
6263 By default, when @value{GDBN} parses C or C++ expressions, type checking
6264 is not used. However, if you turn type checking on, @value{GDBN}
6265 considers two variables type equivalent if:
6269 The two variables are structured and have the same structure, union, or
6273 The two variables have the same type name, or types that have been
6274 declared equivalent through @code{typedef}.
6277 @c leaving this out because neither J Gilmore nor R Pesch understand it.
6280 The two @code{struct}, @code{union}, or @code{enum} variables are
6281 declared in the same declaration. (Note: this may not be true for all C
6286 Range checking, if turned on, is done on mathematical operations. Array
6287 indices are not checked, since they are often used to index a pointer
6288 that is not itself an array.
6291 @subsubsection @value{GDBN} and C
6293 The @code{set print union} and @code{show print union} commands apply to
6294 the @code{union} type. When set to @samp{on}, any @code{union} that is
6295 inside a @code{struct} or @code{class} is also printed. Otherwise, it
6296 appears as @samp{@{...@}}.
6298 The @code{@@} operator aids in the debugging of dynamic arrays, formed
6299 with pointers and a memory allocation function. @xref{Expressions,
6303 * Debugging C plus plus::
6306 @node Debugging C plus plus
6307 @subsubsection @value{GDBN} features for C++
6309 @cindex commands for C++
6311 Some @value{GDBN} commands are particularly useful with C++, and some are
6312 designed specifically for use with C++. Here is a summary:
6315 @cindex break in overloaded functions
6316 @item @r{breakpoint menus}
6317 When you want a breakpoint in a function whose name is overloaded,
6318 @value{GDBN} breakpoint menus help you specify which function definition
6319 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
6321 @cindex overloading in C++
6322 @item rbreak @var{regex}
6323 Setting breakpoints using regular expressions is helpful for setting
6324 breakpoints on overloaded functions that are not members of any special
6326 @xref{Set Breaks, ,Setting breakpoints}.
6328 @cindex C++ exception handling
6331 Debug C++ exception handling using these commands. @xref{Set
6332 Catchpoints, , Setting catchpoints}.
6335 @item ptype @var{typename}
6336 Print inheritance relationships as well as other information for type
6338 @xref{Symbols, ,Examining the Symbol Table}.
6340 @cindex C++ symbol display
6341 @item set print demangle
6342 @itemx show print demangle
6343 @itemx set print asm-demangle
6344 @itemx show print asm-demangle
6345 Control whether C++ symbols display in their source form, both when
6346 displaying code as C++ source and when displaying disassemblies.
6347 @xref{Print Settings, ,Print settings}.
6349 @item set print object
6350 @itemx show print object
6351 Choose whether to print derived (actual) or declared types of objects.
6352 @xref{Print Settings, ,Print settings}.
6354 @item set print vtbl
6355 @itemx show print vtbl
6356 Control the format for printing virtual function tables.
6357 @xref{Print Settings, ,Print settings}.
6358 (The @code{vtbl} commands do not work on programs compiled with the HP
6359 ANSI C++ compiler (@code{aCC}).)
6361 @kindex set overload-resolution
6362 @cindex overloaded functions, overload resolution
6363 @item set overload-resolution on
6364 Enable overload resolution for C++ expression evaluation. The default
6365 is on. For overloaded functions, @value{GDBN} evaluates the arguments
6366 and searches for a function whose signature matches the argument types,
6367 using the standard C++ conversion rules (see @ref{C plus plus expressions, ,C++
6368 expressions}, for details). If it cannot find a match, it emits a
6371 @item set overload-resolution off
6372 Disable overload resolution for C++ expression evaluation. For
6373 overloaded functions that are not class member functions, @value{GDBN}
6374 chooses the first function of the specified name that it finds in the
6375 symbol table, whether or not its arguments are of the correct type. For
6376 overloaded functions that are class member functions, @value{GDBN}
6377 searches for a function whose signature @emph{exactly} matches the
6380 @item @r{Overloaded symbol names}
6381 You can specify a particular definition of an overloaded symbol, using
6382 the same notation that is used to declare such symbols in C++: type
6383 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
6384 also use the @value{GDBN} command-line word completion facilities to list the
6385 available choices, or to finish the type list for you.
6386 @xref{Completion,, Command completion}, for details on how to do this.
6390 @subsection Modula-2
6392 @cindex Modula-2, @value{GDBN} support
6394 The extensions made to @value{GDBN} to support Modula-2 only support
6395 output from the @sc{gnu} Modula-2 compiler (which is currently being
6396 developed). Other Modula-2 compilers are not currently supported, and
6397 attempting to debug executables produced by them is most likely
6398 to give an error as @value{GDBN} reads in the executable's symbol
6401 @cindex expressions in Modula-2
6403 * M2 Operators:: Built-in operators
6404 * Built-In Func/Proc:: Built-in functions and procedures
6405 * M2 Constants:: Modula-2 constants
6406 * M2 Defaults:: Default settings for Modula-2
6407 * Deviations:: Deviations from standard Modula-2
6408 * M2 Checks:: Modula-2 type and range checks
6409 * M2 Scope:: The scope operators @code{::} and @code{.}
6410 * GDB/M2:: @value{GDBN} and Modula-2
6414 @subsubsection Operators
6415 @cindex Modula-2 operators
6417 Operators must be defined on values of specific types. For instance,
6418 @code{+} is defined on numbers, but not on structures. Operators are
6419 often defined on groups of types. For the purposes of Modula-2, the
6420 following definitions hold:
6425 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
6429 @emph{Character types} consist of @code{CHAR} and its subranges.
6432 @emph{Floating-point types} consist of @code{REAL}.
6435 @emph{Pointer types} consist of anything declared as @code{POINTER TO
6439 @emph{Scalar types} consist of all of the above.
6442 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
6445 @emph{Boolean types} consist of @code{BOOLEAN}.
6449 The following operators are supported, and appear in order of
6450 increasing precedence:
6454 Function argument or array index separator.
6457 Assignment. The value of @var{var} @code{:=} @var{value} is
6461 Less than, greater than on integral, floating-point, or enumerated
6465 Less than or equal to, greater than or equal to
6466 on integral, floating-point and enumerated types, or set inclusion on
6467 set types. Same precedence as @code{<}.
6469 @item =@r{, }<>@r{, }#
6470 Equality and two ways of expressing inequality, valid on scalar types.
6471 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
6472 available for inequality, since @code{#} conflicts with the script
6476 Set membership. Defined on set types and the types of their members.
6477 Same precedence as @code{<}.
6480 Boolean disjunction. Defined on boolean types.
6483 Boolean conjunction. Defined on boolean types.
6486 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
6489 Addition and subtraction on integral and floating-point types, or union
6490 and difference on set types.
6493 Multiplication on integral and floating-point types, or set intersection
6497 Division on floating-point types, or symmetric set difference on set
6498 types. Same precedence as @code{*}.
6501 Integer division and remainder. Defined on integral types. Same
6502 precedence as @code{*}.
6505 Negative. Defined on @code{INTEGER} and @code{REAL} data.
6508 Pointer dereferencing. Defined on pointer types.
6511 Boolean negation. Defined on boolean types. Same precedence as
6515 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
6516 precedence as @code{^}.
6519 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
6522 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
6526 @value{GDBN} and Modula-2 scope operators.
6530 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
6531 treats the use of the operator @code{IN}, or the use of operators
6532 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
6533 @code{<=}, and @code{>=} on sets as an error.
6536 @cindex Modula-2 built-ins
6537 @node Built-In Func/Proc
6538 @subsubsection Built-in functions and procedures
6540 Modula-2 also makes available several built-in procedures and functions.
6541 In describing these, the following metavariables are used:
6546 represents an @code{ARRAY} variable.
6549 represents a @code{CHAR} constant or variable.
6552 represents a variable or constant of integral type.
6555 represents an identifier that belongs to a set. Generally used in the
6556 same function with the metavariable @var{s}. The type of @var{s} should
6557 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
6560 represents a variable or constant of integral or floating-point type.
6563 represents a variable or constant of floating-point type.
6569 represents a variable.
6572 represents a variable or constant of one of many types. See the
6573 explanation of the function for details.
6576 All Modula-2 built-in procedures also return a result, described below.
6580 Returns the absolute value of @var{n}.
6583 If @var{c} is a lower case letter, it returns its upper case
6584 equivalent, otherwise it returns its argument.
6587 Returns the character whose ordinal value is @var{i}.
6590 Decrements the value in the variable @var{v} by one. Returns the new value.
6592 @item DEC(@var{v},@var{i})
6593 Decrements the value in the variable @var{v} by @var{i}. Returns the
6596 @item EXCL(@var{m},@var{s})
6597 Removes the element @var{m} from the set @var{s}. Returns the new
6600 @item FLOAT(@var{i})
6601 Returns the floating point equivalent of the integer @var{i}.
6604 Returns the index of the last member of @var{a}.
6607 Increments the value in the variable @var{v} by one. Returns the new value.
6609 @item INC(@var{v},@var{i})
6610 Increments the value in the variable @var{v} by @var{i}. Returns the
6613 @item INCL(@var{m},@var{s})
6614 Adds the element @var{m} to the set @var{s} if it is not already
6615 there. Returns the new set.
6618 Returns the maximum value of the type @var{t}.
6621 Returns the minimum value of the type @var{t}.
6624 Returns boolean TRUE if @var{i} is an odd number.
6627 Returns the ordinal value of its argument. For example, the ordinal
6628 value of a character is its @sc{ascii} value (on machines supporting the
6629 @sc{ascii} character set). @var{x} must be of an ordered type, which include
6630 integral, character and enumerated types.
6633 Returns the size of its argument. @var{x} can be a variable or a type.
6635 @item TRUNC(@var{r})
6636 Returns the integral part of @var{r}.
6638 @item VAL(@var{t},@var{i})
6639 Returns the member of the type @var{t} whose ordinal value is @var{i}.
6643 @emph{Warning:} Sets and their operations are not yet supported, so
6644 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
6648 @cindex Modula-2 constants
6650 @subsubsection Constants
6652 @value{GDBN} allows you to express the constants of Modula-2 in the following
6658 Integer constants are simply a sequence of digits. When used in an
6659 expression, a constant is interpreted to be type-compatible with the
6660 rest of the expression. Hexadecimal integers are specified by a
6661 trailing @samp{H}, and octal integers by a trailing @samp{B}.
6664 Floating point constants appear as a sequence of digits, followed by a
6665 decimal point and another sequence of digits. An optional exponent can
6666 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
6667 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
6668 digits of the floating point constant must be valid decimal (base 10)
6672 Character constants consist of a single character enclosed by a pair of
6673 like quotes, either single (@code{'}) or double (@code{"}). They may
6674 also be expressed by their ordinal value (their @sc{ascii} value, usually)
6675 followed by a @samp{C}.
6678 String constants consist of a sequence of characters enclosed by a
6679 pair of like quotes, either single (@code{'}) or double (@code{"}).
6680 Escape sequences in the style of C are also allowed. @xref{C
6681 Constants, ,C and C++ constants}, for a brief explanation of escape
6685 Enumerated constants consist of an enumerated identifier.
6688 Boolean constants consist of the identifiers @code{TRUE} and
6692 Pointer constants consist of integral values only.
6695 Set constants are not yet supported.
6699 @subsubsection Modula-2 defaults
6700 @cindex Modula-2 defaults
6702 If type and range checking are set automatically by @value{GDBN}, they
6703 both default to @code{on} whenever the working language changes to
6704 Modula-2. This happens regardless of whether you or @value{GDBN}
6705 selected the working language.
6707 If you allow @value{GDBN} to set the language automatically, then entering
6708 code compiled from a file whose name ends with @file{.mod} sets the
6709 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
6710 the language automatically}, for further details.
6713 @subsubsection Deviations from standard Modula-2
6714 @cindex Modula-2, deviations from
6716 A few changes have been made to make Modula-2 programs easier to debug.
6717 This is done primarily via loosening its type strictness:
6721 Unlike in standard Modula-2, pointer constants can be formed by
6722 integers. This allows you to modify pointer variables during
6723 debugging. (In standard Modula-2, the actual address contained in a
6724 pointer variable is hidden from you; it can only be modified
6725 through direct assignment to another pointer variable or expression that
6726 returned a pointer.)
6729 C escape sequences can be used in strings and characters to represent
6730 non-printable characters. @value{GDBN} prints out strings with these
6731 escape sequences embedded. Single non-printable characters are
6732 printed using the @samp{CHR(@var{nnn})} format.
6735 The assignment operator (@code{:=}) returns the value of its right-hand
6739 All built-in procedures both modify @emph{and} return their argument.
6743 @subsubsection Modula-2 type and range checks
6744 @cindex Modula-2 checks
6747 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
6750 @c FIXME remove warning when type/range checks added
6752 @value{GDBN} considers two Modula-2 variables type equivalent if:
6756 They are of types that have been declared equivalent via a @code{TYPE
6757 @var{t1} = @var{t2}} statement
6760 They have been declared on the same line. (Note: This is true of the
6761 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
6764 As long as type checking is enabled, any attempt to combine variables
6765 whose types are not equivalent is an error.
6767 Range checking is done on all mathematical operations, assignment, array
6768 index bounds, and all built-in functions and procedures.
6771 @subsubsection The scope operators @code{::} and @code{.}
6773 @cindex @code{.}, Modula-2 scope operator
6774 @cindex colon, doubled as scope operator
6776 @vindex colon-colon@r{, in Modula-2}
6777 @c Info cannot handle :: but TeX can.
6780 @vindex ::@r{, in Modula-2}
6783 There are a few subtle differences between the Modula-2 scope operator
6784 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
6789 @var{module} . @var{id}
6790 @var{scope} :: @var{id}
6794 where @var{scope} is the name of a module or a procedure,
6795 @var{module} the name of a module, and @var{id} is any declared
6796 identifier within your program, except another module.
6798 Using the @code{::} operator makes @value{GDBN} search the scope
6799 specified by @var{scope} for the identifier @var{id}. If it is not
6800 found in the specified scope, then @value{GDBN} searches all scopes
6801 enclosing the one specified by @var{scope}.
6803 Using the @code{.} operator makes @value{GDBN} search the current scope for
6804 the identifier specified by @var{id} that was imported from the
6805 definition module specified by @var{module}. With this operator, it is
6806 an error if the identifier @var{id} was not imported from definition
6807 module @var{module}, or if @var{id} is not an identifier in
6811 @subsubsection @value{GDBN} and Modula-2
6813 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
6814 Five subcommands of @code{set print} and @code{show print} apply
6815 specifically to C and C++: @samp{vtbl}, @samp{demangle},
6816 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
6817 apply to C++, and the last to the C @code{union} type, which has no direct
6818 analogue in Modula-2.
6820 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
6821 with any language, is not useful with Modula-2. Its
6822 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
6823 created in Modula-2 as they can in C or C++. However, because an
6824 address can be specified by an integral constant, the construct
6825 @samp{@{@var{type}@}@var{adrexp}} is still useful.
6827 @cindex @code{#} in Modula-2
6828 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
6829 interpreted as the beginning of a comment. Use @code{<>} instead.
6834 The extensions made to @value{GDBN} to support Chill only support output
6835 from the @sc{gnu} Chill compiler. Other Chill compilers are not currently
6836 supported, and attempting to debug executables produced by them is most
6837 likely to give an error as @value{GDBN} reads in the executable's symbol
6840 @c This used to say "... following Chill related topics ...", but since
6841 @c menus are not shown in the printed manual, it would look awkward.
6842 This section covers the Chill related topics and the features
6843 of @value{GDBN} which support these topics.
6846 * How modes are displayed:: How modes are displayed
6847 * Locations:: Locations and their accesses
6848 * Values and their Operations:: Values and their Operations
6849 * Chill type and range checks::
6853 @node How modes are displayed
6854 @subsubsection How modes are displayed
6856 The Chill Datatype- (Mode) support of @value{GDBN} is directly related
6857 with the functionality of the @sc{gnu} Chill compiler, and therefore deviates
6858 slightly from the standard specification of the Chill language. The
6861 @c FIXME: this @table's contents effectively disable @code by using @r
6862 @c on every @item. So why does it need @code?
6864 @item @r{@emph{Discrete modes:}}
6867 @emph{Integer Modes} which are predefined by @code{BYTE, UBYTE, INT,
6870 @emph{Boolean Mode} which is predefined by @code{BOOL},
6872 @emph{Character Mode} which is predefined by @code{CHAR},
6874 @emph{Set Mode} which is displayed by the keyword @code{SET}.
6876 (@value{GDBP}) ptype x
6877 type = SET (karli = 10, susi = 20, fritzi = 100)
6879 If the type is an unnumbered set the set element values are omitted.
6881 @emph{Range Mode} which is displayed by
6883 @code{type = <basemode>(<lower bound> : <upper bound>)}
6885 where @code{<lower bound>, <upper bound>} can be of any discrete literal
6886 expression (e.g. set element names).
6889 @item @r{@emph{Powerset Mode:}}
6890 A Powerset Mode is displayed by the keyword @code{POWERSET} followed by
6891 the member mode of the powerset. The member mode can be any discrete mode.
6893 (@value{GDBP}) ptype x
6894 type = POWERSET SET (egon, hugo, otto)
6897 @item @r{@emph{Reference Modes:}}
6900 @emph{Bound Reference Mode} which is displayed by the keyword @code{REF}
6901 followed by the mode name to which the reference is bound.
6903 @emph{Free Reference Mode} which is displayed by the keyword @code{PTR}.
6906 @item @r{@emph{Procedure mode}}
6907 The procedure mode is displayed by @code{type = PROC(<parameter list>)
6908 <return mode> EXCEPTIONS (<exception list>)}. The @code{<parameter
6909 list>} is a list of the parameter modes. @code{<return mode>} indicates
6910 the mode of the result of the procedure if any. The exceptionlist lists
6911 all possible exceptions which can be raised by the procedure.
6914 @item @r{@emph{Instance mode}}
6915 The instance mode is represented by a structure, which has a static
6916 type, and is therefore not really of interest.
6919 @item @r{@emph{Synchronization Modes:}}
6922 @emph{Event Mode} which is displayed by
6924 @code{EVENT (<event length>)}
6926 where @code{(<event length>)} is optional.
6928 @emph{Buffer Mode} which is displayed by
6930 @code{BUFFER (<buffer length>)<buffer element mode>}
6932 where @code{(<buffer length>)} is optional.
6935 @item @r{@emph{Timing Modes:}}
6938 @emph{Duration Mode} which is predefined by @code{DURATION}
6940 @emph{Absolute Time Mode} which is predefined by @code{TIME}
6943 @item @r{@emph{Real Modes:}}
6944 Real Modes are predefined with @code{REAL} and @code{LONG_REAL}.
6946 @item @r{@emph{String Modes:}}
6949 @emph{Character String Mode} which is displayed by
6951 @code{CHARS(<string length>)}
6953 followed by the keyword @code{VARYING} if the String Mode is a varying
6956 @emph{Bit String Mode} which is displayed by
6963 @item @r{@emph{Array Mode:}}
6964 The Array Mode is displayed by the keyword @code{ARRAY(<range>)}
6965 followed by the element mode (which may in turn be an array mode).
6967 (@value{GDBP}) ptype x
6970 SET (karli = 10, susi = 20, fritzi = 100)
6973 @item @r{@emph{Structure Mode}}
6974 The Structure mode is displayed by the keyword @code{STRUCT(<field
6975 list>)}. The @code{<field list>} consists of names and modes of fields
6976 of the structure. Variant structures have the keyword @code{CASE <field>
6977 OF <variant fields> ESAC} in their field list. Since the current version
6978 of the GNU Chill compiler doesn't implement tag processing (no runtime
6979 checks of variant fields, and therefore no debugging info), the output
6980 always displays all variant fields.
6982 (@value{GDBP}) ptype str
6997 @subsubsection Locations and their accesses
6999 A location in Chill is an object which can contain values.
7001 A value of a location is generally accessed by the (declared) name of
7002 the location. The output conforms to the specification of values in
7003 Chill programs. How values are specified
7004 is the topic of the next section, @ref{Values and their Operations}.
7006 The pseudo-location @code{RESULT} (or @code{result}) can be used to
7007 display or change the result of a currently-active procedure:
7014 This does the same as the Chill action @code{RESULT EXPR} (which
7015 is not available in @value{GDBN}).
7017 Values of reference mode locations are printed by @code{PTR(<hex
7018 value>)} in case of a free reference mode, and by @code{(REF <reference
7019 mode>) (<hex-value>)} in case of a bound reference. @code{<hex value>}
7020 represents the address where the reference points to. To access the
7021 value of the location referenced by the pointer, use the dereference
7024 Values of procedure mode locations are displayed by
7027 (<argument modes> ) <return mode> @} <address> <name of procedure
7030 @code{<argument modes>} is a list of modes according to the parameter
7031 specification of the procedure and @code{<address>} shows the address of
7035 Locations of instance modes are displayed just like a structure with two
7036 fields specifying the @emph{process type} and the @emph{copy number} of
7037 the investigated instance location@footnote{This comes from the current
7038 implementation of instances. They are implemented as a structure (no
7039 na). The output should be something like @code{[<name of the process>;
7040 <instance number>]}.}. The field names are @code{__proc_type} and
7043 Locations of synchronization modes are displayed like a structure with
7044 the field name @code{__event_data} in case of a event mode location, and
7045 like a structure with the field @code{__buffer_data} in case of a buffer
7046 mode location (refer to previous paragraph).
7048 Structure Mode locations are printed by @code{[.<field name>: <value>,
7049 ...]}. The @code{<field name>} corresponds to the structure mode
7050 definition and the layout of @code{<value>} varies depending of the mode
7051 of the field. If the investigated structure mode location is of variant
7052 structure mode, the variant parts of the structure are enclosed in curled
7053 braces (@samp{@{@}}). Fields enclosed by @samp{@{,@}} are residing
7054 on the same memory location and represent the current values of the
7055 memory location in their specific modes. Since no tag processing is done
7056 all variants are displayed. A variant field is printed by
7057 @code{(<variant name>) = .<field name>: <value>}. (who implements the
7060 (@value{GDBP}) print str1 $4 = [.as: 0, .bs: karli, .<TAG>: { (karli) =
7061 [.cs: []], (susi) = [.ds: susi]}]
7065 Substructures of string mode-, array mode- or structure mode-values
7066 (e.g. array slices, fields of structure locations) are accessed using
7067 certain operations which are described in the next section, @ref{Values
7068 and their Operations}.
7070 A location value may be interpreted as having a different mode using the
7071 location conversion. This mode conversion is written as @code{<mode
7072 name>(<location>)}. The user has to consider that the sizes of the modes
7073 have to be equal otherwise an error occurs. Furthermore, no range
7074 checking of the location against the destination mode is performed, and
7075 therefore the result can be quite confusing.
7078 (@value{GDBP}) print int (s(3 up 4)) XXX TO be filled in !! XXX
7081 @node Values and their Operations
7082 @subsubsection Values and their Operations
7084 Values are used to alter locations, to investigate complex structures in
7085 more detail or to filter relevant information out of a large amount of
7086 data. There are several (mode dependent) operations defined which enable
7087 such investigations. These operations are not only applicable to
7088 constant values but also to locations, which can become quite useful
7089 when debugging complex structures. During parsing the command line
7090 (e.g. evaluating an expression) @value{GDBN} treats location names as
7091 the values behind these locations.
7093 This section describes how values have to be specified and which
7094 operations are legal to be used with such values.
7097 @item Literal Values
7098 Literal values are specified in the same manner as in @sc{gnu} Chill programs.
7099 For detailed specification refer to the @sc{gnu} Chill implementation Manual
7101 @c FIXME: if the Chill Manual is a Texinfo documents, the above should
7102 @c be converted to a @ref.
7107 @emph{Integer Literals} are specified in the same manner as in Chill
7108 programs (refer to the Chill Standard z200/88 chpt 5.2.4.2)
7110 @emph{Boolean Literals} are defined by @code{TRUE} and @code{FALSE}.
7112 @emph{Character Literals} are defined by @code{'<character>'}. (e.g.
7115 @emph{Set Literals} are defined by a name which was specified in a set
7116 mode. The value delivered by a Set Literal is the set value. This is
7117 comparable to an enumeration in C/C++ language.
7119 @emph{Emptiness Literal} is predefined by @code{NULL}. The value of the
7120 emptiness literal delivers either the empty reference value, the empty
7121 procedure value or the empty instance value.
7124 @emph{Character String Literals} are defined by a sequence of characters
7125 enclosed in single- or double quotes. If a single- or double quote has
7126 to be part of the string literal it has to be stuffed (specified twice).
7128 @emph{Bitstring Literals} are specified in the same manner as in Chill
7129 programs (refer z200/88 chpt 5.2.4.8).
7131 @emph{Floating point literals} are specified in the same manner as in
7132 (gnu-)Chill programs (refer @sc{gnu} Chill implementation Manual chapter 1.5).
7137 A tuple is specified by @code{<mode name>[<tuple>]}, where @code{<mode
7138 name>} can be omitted if the mode of the tuple is unambiguous. This
7139 unambiguity is derived from the context of a evaluated expression.
7140 @code{<tuple>} can be one of the following:
7143 @item @emph{Powerset Tuple}
7144 @item @emph{Array Tuple}
7145 @item @emph{Structure Tuple}
7146 Powerset tuples, array tuples and structure tuples are specified in the
7147 same manner as in Chill programs refer to z200/88 chpt 5.2.5.
7150 @item String Element Value
7151 A string element value is specified by
7153 @code{<string value>(<index>)}
7155 where @code{<index>} is a integer expression. It delivers a character
7156 value which is equivalent to the character indexed by @code{<index>} in
7159 @item String Slice Value
7160 A string slice value is specified by @code{<string value>(<slice
7161 spec>)}, where @code{<slice spec>} can be either a range of integer
7162 expressions or specified by @code{<start expr> up <size>}.
7163 @code{<size>} denotes the number of elements which the slice contains.
7164 The delivered value is a string value, which is part of the specified
7167 @item Array Element Values
7168 An array element value is specified by @code{<array value>(<expr>)} and
7169 delivers a array element value of the mode of the specified array.
7171 @item Array Slice Values
7172 An array slice is specified by @code{<array value>(<slice spec>)}, where
7173 @code{<slice spec>} can be either a range specified by expressions or by
7174 @code{<start expr> up <size>}. @code{<size>} denotes the number of
7175 arrayelements the slice contains. The delivered value is an array value
7176 which is part of the specified array.
7178 @item Structure Field Values
7179 A structure field value is derived by @code{<structure value>.<field
7180 name>}, where @code{<field name>} indicates the name of a field specified
7181 in the mode definition of the structure. The mode of the delivered value
7182 corresponds to this mode definition in the structure definition.
7184 @item Procedure Call Value
7185 The procedure call value is derived from the return value of the
7186 procedure@footnote{If a procedure call is used for instance in an
7187 expression, then this procedure is called with all its side
7188 effects. This can lead to confusing results if used carelessly.}.
7190 Values of duration mode locations are represented by @code{ULONG} literals.
7192 Values of time mode locations appear as
7194 @code{TIME(<secs>:<nsecs>)}
7199 This is not implemented yet:
7200 @item Built-in Value
7202 The following built in functions are provided:
7214 @item @code{UPPER()}
7215 @item @code{LOWER()}
7216 @item @code{LENGTH()}
7220 @item @code{ARCSIN()}
7221 @item @code{ARCCOS()}
7222 @item @code{ARCTAN()}
7229 For a detailed description refer to the GNU Chill implementation manual
7233 @item Zero-adic Operator Value
7234 The zero-adic operator value is derived from the instance value for the
7235 current active process.
7237 @item Expression Values
7238 The value delivered by an expression is the result of the evaluation of
7239 the specified expression. If there are error conditions (mode
7240 incompatibility, etc.) the evaluation of expressions is aborted with a
7241 corresponding error message. Expressions may be parenthesised which
7242 causes the evaluation of this expression before any other expression
7243 which uses the result of the parenthesised expression. The following
7244 operators are supported by @value{GDBN}:
7247 @item @code{OR, ORIF, XOR}
7248 @itemx @code{AND, ANDIF}
7250 Logical operators defined over operands of boolean mode.
7253 Equality and inequality operators defined over all modes.
7257 Relational operators defined over predefined modes.
7260 @itemx @code{*, /, MOD, REM}
7261 Arithmetic operators defined over predefined modes.
7264 Change sign operator.
7267 String concatenation operator.
7270 String repetition operator.
7273 Referenced location operator which can be used either to take the
7274 address of a location (@code{->loc}), or to dereference a reference
7275 location (@code{loc->}).
7277 @item @code{OR, XOR}
7280 Powerset and bitstring operators.
7284 Powerset inclusion operators.
7287 Membership operator.
7291 @node Chill type and range checks
7292 @subsubsection Chill type and range checks
7294 @value{GDBN} considers two Chill variables mode equivalent if the sizes
7295 of the two modes are equal. This rule applies recursively to more
7296 complex datatypes which means that complex modes are treated
7297 equivalent if all element modes (which also can be complex modes like
7298 structures, arrays, etc.) have the same size.
7300 Range checking is done on all mathematical operations, assignment, array
7301 index bounds and all built in procedures.
7303 Strong type checks are forced using the @value{GDBN} command @code{set
7304 check strong}. This enforces strong type and range checks on all
7305 operations where Chill constructs are used (expressions, built in
7306 functions, etc.) in respect to the semantics as defined in the z.200
7307 language specification.
7309 All checks can be disabled by the @value{GDBN} command @code{set check
7313 @c Deviations from the Chill Standard Z200/88
7314 see last paragraph ?
7317 @node Chill defaults
7318 @subsubsection Chill defaults
7320 If type and range checking are set automatically by @value{GDBN}, they
7321 both default to @code{on} whenever the working language changes to
7322 Chill. This happens regardless of whether you or @value{GDBN}
7323 selected the working language.
7325 If you allow @value{GDBN} to set the language automatically, then entering
7326 code compiled from a file whose name ends with @file{.ch} sets the
7327 working language to Chill. @xref{Automatically, ,Having @value{GDBN} set
7328 the language automatically}, for further details.
7331 @chapter Examining the Symbol Table
7333 The commands described in this chapter allow you to inquire about the
7334 symbols (names of variables, functions and types) defined in your
7335 program. This information is inherent in the text of your program and
7336 does not change as your program executes. @value{GDBN} finds it in your
7337 program's symbol table, in the file indicated when you started @value{GDBN}
7338 (@pxref{File Options, ,Choosing files}), or by one of the
7339 file-management commands (@pxref{Files, ,Commands to specify files}).
7341 @cindex symbol names
7342 @cindex names of symbols
7343 @cindex quoting names
7344 Occasionally, you may need to refer to symbols that contain unusual
7345 characters, which @value{GDBN} ordinarily treats as word delimiters. The
7346 most frequent case is in referring to static variables in other
7347 source files (@pxref{Variables,,Program variables}). File names
7348 are recorded in object files as debugging symbols, but @value{GDBN} would
7349 ordinarily parse a typical file name, like @file{foo.c}, as the three words
7350 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
7351 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
7358 looks up the value of @code{x} in the scope of the file @file{foo.c}.
7361 @kindex info address
7362 @item info address @var{symbol}
7363 Describe where the data for @var{symbol} is stored. For a register
7364 variable, this says which register it is kept in. For a non-register
7365 local variable, this prints the stack-frame offset at which the variable
7368 Note the contrast with @samp{print &@var{symbol}}, which does not work
7369 at all for a register variable, and for a stack local variable prints
7370 the exact address of the current instantiation of the variable.
7373 @item whatis @var{expr}
7374 Print the data type of expression @var{expr}. @var{expr} is not
7375 actually evaluated, and any side-effecting operations (such as
7376 assignments or function calls) inside it do not take place.
7377 @xref{Expressions, ,Expressions}.
7380 Print the data type of @code{$}, the last value in the value history.
7383 @item ptype @var{typename}
7384 Print a description of data type @var{typename}. @var{typename} may be
7385 the name of a type, or for C code it may have the form @samp{class
7386 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
7387 @var{union-tag}} or @samp{enum @var{enum-tag}}.
7389 @item ptype @var{expr}
7391 Print a description of the type of expression @var{expr}. @code{ptype}
7392 differs from @code{whatis} by printing a detailed description, instead
7393 of just the name of the type.
7395 For example, for this variable declaration:
7398 struct complex @{double real; double imag;@} v;
7402 the two commands give this output:
7406 (@value{GDBP}) whatis v
7407 type = struct complex
7408 (@value{GDBP}) ptype v
7409 type = struct complex @{
7417 As with @code{whatis}, using @code{ptype} without an argument refers to
7418 the type of @code{$}, the last value in the value history.
7421 @item info types @var{regexp}
7423 Print a brief description of all types whose names match @var{regexp}
7424 (or all types in your program, if you supply no argument). Each
7425 complete typename is matched as though it were a complete line; thus,
7426 @samp{i type value} gives information on all types in your program whose
7427 names include the string @code{value}, but @samp{i type ^value$} gives
7428 information only on types whose complete name is @code{value}.
7430 This command differs from @code{ptype} in two ways: first, like
7431 @code{whatis}, it does not print a detailed description; second, it
7432 lists all source files where a type is defined.
7436 Show the name of the current source file---that is, the source file for
7437 the function containing the current point of execution---and the language
7440 @kindex info sources
7442 Print the names of all source files in your program for which there is
7443 debugging information, organized into two lists: files whose symbols
7444 have already been read, and files whose symbols will be read when needed.
7446 @kindex info functions
7447 @item info functions
7448 Print the names and data types of all defined functions.
7450 @item info functions @var{regexp}
7451 Print the names and data types of all defined functions
7452 whose names contain a match for regular expression @var{regexp}.
7453 Thus, @samp{info fun step} finds all functions whose names
7454 include @code{step}; @samp{info fun ^step} finds those whose names
7455 start with @code{step}.
7457 @kindex info variables
7458 @item info variables
7459 Print the names and data types of all variables that are declared
7460 outside of functions (i.e., excluding local variables).
7462 @item info variables @var{regexp}
7463 Print the names and data types of all variables (except for local
7464 variables) whose names contain a match for regular expression
7468 This was never implemented.
7469 @kindex info methods
7471 @itemx info methods @var{regexp}
7472 The @code{info methods} command permits the user to examine all defined
7473 methods within C++ program, or (with the @var{regexp} argument) a
7474 specific set of methods found in the various C++ classes. Many
7475 C++ classes provide a large number of methods. Thus, the output
7476 from the @code{ptype} command can be overwhelming and hard to use. The
7477 @code{info-methods} command filters the methods, printing only those
7478 which match the regular-expression @var{regexp}.
7481 @cindex reloading symbols
7482 Some systems allow individual object files that make up your program to
7483 be replaced without stopping and restarting your program. For example,
7484 in VxWorks you can simply recompile a defective object file and keep on
7485 running. If you are running on one of these systems, you can allow
7486 @value{GDBN} to reload the symbols for automatically relinked modules:
7489 @kindex set symbol-reloading
7490 @item set symbol-reloading on
7491 Replace symbol definitions for the corresponding source file when an
7492 object file with a particular name is seen again.
7494 @item set symbol-reloading off
7495 Do not replace symbol definitions when encountering object files of the
7496 same name more than once. This is the default state; if you are not
7497 running on a system that permits automatic relinking of modules, you
7498 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
7499 may discard symbols when linking large programs, that may contain
7500 several modules (from different directories or libraries) with the same
7503 @kindex show symbol-reloading
7504 @item show symbol-reloading
7505 Show the current @code{on} or @code{off} setting.
7508 @kindex set opaque-type-resolution
7509 @item set opaque-type-resolution on
7510 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
7511 declared as a pointer to a @code{struct}, @code{class}, or
7512 @code{union}---for example, @code{struct MyType *}---that is used in one
7513 source file although the full declaration of @code{struct MyType} is in
7514 another source file. The default is on.
7516 A change in the setting of this subcommand will not take effect until
7517 the next time symbols for a file are loaded.
7519 @item set opaque-type-resolution off
7520 Tell @value{GDBN} not to resolve opaque types. In this case, the type
7521 is printed as follows:
7523 @{<no data fields>@}
7526 @kindex show opaque-type-resolution
7527 @item show opaque-type-resolution
7528 Show whether opaque types are resolved or not.
7530 @kindex maint print symbols
7532 @kindex maint print psymbols
7533 @cindex partial symbol dump
7534 @item maint print symbols @var{filename}
7535 @itemx maint print psymbols @var{filename}
7536 @itemx maint print msymbols @var{filename}
7537 Write a dump of debugging symbol data into the file @var{filename}.
7538 These commands are used to debug the @value{GDBN} symbol-reading code. Only
7539 symbols with debugging data are included. If you use @samp{maint print
7540 symbols}, @value{GDBN} includes all the symbols for which it has already
7541 collected full details: that is, @var{filename} reflects symbols for
7542 only those files whose symbols @value{GDBN} has read. You can use the
7543 command @code{info sources} to find out which files these are. If you
7544 use @samp{maint print psymbols} instead, the dump shows information about
7545 symbols that @value{GDBN} only knows partially---that is, symbols defined in
7546 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
7547 @samp{maint print msymbols} dumps just the minimal symbol information
7548 required for each object file from which @value{GDBN} has read some symbols.
7549 @xref{Files, ,Commands to specify files}, for a discussion of how
7550 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
7554 @chapter Altering Execution
7556 Once you think you have found an error in your program, you might want to
7557 find out for certain whether correcting the apparent error would lead to
7558 correct results in the rest of the run. You can find the answer by
7559 experiment, using the @value{GDBN} features for altering execution of the
7562 For example, you can store new values into variables or memory
7563 locations, give your program a signal, restart it at a different
7564 address, or even return prematurely from a function.
7567 * Assignment:: Assignment to variables
7568 * Jumping:: Continuing at a different address
7569 * Signaling:: Giving your program a signal
7570 * Returning:: Returning from a function
7571 * Calling:: Calling your program's functions
7572 * Patching:: Patching your program
7576 @section Assignment to variables
7579 @cindex setting variables
7580 To alter the value of a variable, evaluate an assignment expression.
7581 @xref{Expressions, ,Expressions}. For example,
7588 stores the value 4 into the variable @code{x}, and then prints the
7589 value of the assignment expression (which is 4).
7590 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
7591 information on operators in supported languages.
7593 @kindex set variable
7594 @cindex variables, setting
7595 If you are not interested in seeing the value of the assignment, use the
7596 @code{set} command instead of the @code{print} command. @code{set} is
7597 really the same as @code{print} except that the expression's value is
7598 not printed and is not put in the value history (@pxref{Value History,
7599 ,Value history}). The expression is evaluated only for its effects.
7601 If the beginning of the argument string of the @code{set} command
7602 appears identical to a @code{set} subcommand, use the @code{set
7603 variable} command instead of just @code{set}. This command is identical
7604 to @code{set} except for its lack of subcommands. For example, if your
7605 program has a variable @code{width}, you get an error if you try to set
7606 a new value with just @samp{set width=13}, because @value{GDBN} has the
7607 command @code{set width}:
7610 (@value{GDBP}) whatis width
7612 (@value{GDBP}) p width
7614 (@value{GDBP}) set width=47
7615 Invalid syntax in expression.
7619 The invalid expression, of course, is @samp{=47}. In
7620 order to actually set the program's variable @code{width}, use
7623 (@value{GDBP}) set var width=47
7626 Because the @code{set} command has many subcommands that can conflict
7627 with the names of program variables, it is a good idea to use the
7628 @code{set variable} command instead of just @code{set}. For example, if
7629 your program has a variable @code{g}, you run into problems if you try
7630 to set a new value with just @samp{set g=4}, because @value{GDBN} has
7631 the command @code{set gnutarget}, abbreviated @code{set g}:
7635 (@value{GDBP}) whatis g
7639 (@value{GDBP}) set g=4
7643 The program being debugged has been started already.
7644 Start it from the beginning? (y or n) y
7645 Starting program: /home/smith/cc_progs/a.out
7646 "/home/smith/cc_progs/a.out": can't open to read symbols:
7648 (@value{GDBP}) show g
7649 The current BFD target is "=4".
7654 The program variable @code{g} did not change, and you silently set the
7655 @code{gnutarget} to an invalid value. In order to set the variable
7659 (@value{GDBP}) set var g=4
7662 @value{GDBN} allows more implicit conversions in assignments than C; you can
7663 freely store an integer value into a pointer variable or vice versa,
7664 and you can convert any structure to any other structure that is the
7665 same length or shorter.
7666 @comment FIXME: how do structs align/pad in these conversions?
7667 @comment /doc@cygnus.com 18dec1990
7669 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
7670 construct to generate a value of specified type at a specified address
7671 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
7672 to memory location @code{0x83040} as an integer (which implies a certain size
7673 and representation in memory), and
7676 set @{int@}0x83040 = 4
7680 stores the value 4 into that memory location.
7683 @section Continuing at a different address
7685 Ordinarily, when you continue your program, you do so at the place where
7686 it stopped, with the @code{continue} command. You can instead continue at
7687 an address of your own choosing, with the following commands:
7691 @item jump @var{linespec}
7692 Resume execution at line @var{linespec}. Execution stops again
7693 immediately if there is a breakpoint there. @xref{List, ,Printing
7694 source lines}, for a description of the different forms of
7695 @var{linespec}. It is common practice to use the @code{tbreak} command
7696 in conjunction with @code{jump}. @xref{Set Breaks, ,Setting
7699 The @code{jump} command does not change the current stack frame, or
7700 the stack pointer, or the contents of any memory location or any
7701 register other than the program counter. If line @var{linespec} is in
7702 a different function from the one currently executing, the results may
7703 be bizarre if the two functions expect different patterns of arguments or
7704 of local variables. For this reason, the @code{jump} command requests
7705 confirmation if the specified line is not in the function currently
7706 executing. However, even bizarre results are predictable if you are
7707 well acquainted with the machine-language code of your program.
7709 @item jump *@var{address}
7710 Resume execution at the instruction at address @var{address}.
7713 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
7714 On many systems, you can get much the same effect as the @code{jump}
7715 command by storing a new value into the register @code{$pc}. The
7716 difference is that this does not start your program running; it only
7717 changes the address of where it @emph{will} run when you continue. For
7725 makes the next @code{continue} command or stepping command execute at
7726 address @code{0x485}, rather than at the address where your program stopped.
7727 @xref{Continuing and Stepping, ,Continuing and stepping}.
7729 The most common occasion to use the @code{jump} command is to back
7730 up---perhaps with more breakpoints set---over a portion of a program
7731 that has already executed, in order to examine its execution in more
7736 @section Giving your program a signal
7740 @item signal @var{signal}
7741 Resume execution where your program stopped, but immediately give it the
7742 signal @var{signal}. @var{signal} can be the name or the number of a
7743 signal. For example, on many systems @code{signal 2} and @code{signal
7744 SIGINT} are both ways of sending an interrupt signal.
7746 Alternatively, if @var{signal} is zero, continue execution without
7747 giving a signal. This is useful when your program stopped on account of
7748 a signal and would ordinary see the signal when resumed with the
7749 @code{continue} command; @samp{signal 0} causes it to resume without a
7752 @code{signal} does not repeat when you press @key{RET} a second time
7753 after executing the command.
7757 Invoking the @code{signal} command is not the same as invoking the
7758 @code{kill} utility from the shell. Sending a signal with @code{kill}
7759 causes @value{GDBN} to decide what to do with the signal depending on
7760 the signal handling tables (@pxref{Signals}). The @code{signal} command
7761 passes the signal directly to your program.
7765 @section Returning from a function
7768 @cindex returning from a function
7771 @itemx return @var{expression}
7772 You can cancel execution of a function call with the @code{return}
7773 command. If you give an
7774 @var{expression} argument, its value is used as the function's return
7778 When you use @code{return}, @value{GDBN} discards the selected stack frame
7779 (and all frames within it). You can think of this as making the
7780 discarded frame return prematurely. If you wish to specify a value to
7781 be returned, give that value as the argument to @code{return}.
7783 This pops the selected stack frame (@pxref{Selection, ,Selecting a
7784 frame}), and any other frames inside of it, leaving its caller as the
7785 innermost remaining frame. That frame becomes selected. The
7786 specified value is stored in the registers used for returning values
7789 The @code{return} command does not resume execution; it leaves the
7790 program stopped in the state that would exist if the function had just
7791 returned. In contrast, the @code{finish} command (@pxref{Continuing
7792 and Stepping, ,Continuing and stepping}) resumes execution until the
7793 selected stack frame returns naturally.
7796 @section Calling program functions
7798 @cindex calling functions
7801 @item call @var{expr}
7802 Evaluate the expression @var{expr} without displaying @code{void}
7806 You can use this variant of the @code{print} command if you want to
7807 execute a function from your program, but without cluttering the output
7808 with @code{void} returned values. If the result is not void, it
7809 is printed and saved in the value history.
7811 For the A29K, a user-controlled variable @code{call_scratch_address},
7812 specifies the location of a scratch area to be used when @value{GDBN}
7813 calls a function in the target. This is necessary because the usual
7814 method of putting the scratch area on the stack does not work in systems
7815 that have separate instruction and data spaces.
7818 @section Patching programs
7820 @cindex patching binaries
7821 @cindex writing into executables
7822 @cindex writing into corefiles
7824 By default, @value{GDBN} opens the file containing your program's
7825 executable code (or the corefile) read-only. This prevents accidental
7826 alterations to machine code; but it also prevents you from intentionally
7827 patching your program's binary.
7829 If you'd like to be able to patch the binary, you can specify that
7830 explicitly with the @code{set write} command. For example, you might
7831 want to turn on internal debugging flags, or even to make emergency
7837 @itemx set write off
7838 If you specify @samp{set write on}, @value{GDBN} opens executable and
7839 core files for both reading and writing; if you specify @samp{set write
7840 off} (the default), @value{GDBN} opens them read-only.
7842 If you have already loaded a file, you must load it again (using the
7843 @code{exec-file} or @code{core-file} command) after changing @code{set
7844 write}, for your new setting to take effect.
7848 Display whether executable files and core files are opened for writing
7853 @chapter @value{GDBN} Files
7855 @value{GDBN} needs to know the file name of the program to be debugged,
7856 both in order to read its symbol table and in order to start your
7857 program. To debug a core dump of a previous run, you must also tell
7858 @value{GDBN} the name of the core dump file.
7861 * Files:: Commands to specify files
7862 * Symbol Errors:: Errors reading symbol files
7866 @section Commands to specify files
7868 @cindex symbol table
7869 @cindex core dump file
7871 You may want to specify executable and core dump file names. The usual
7872 way to do this is at start-up time, using the arguments to
7873 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
7874 Out of @value{GDBN}}).
7876 Occasionally it is necessary to change to a different file during a
7877 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
7878 a file you want to use. In these situations the @value{GDBN} commands
7879 to specify new files are useful.
7882 @cindex executable file
7884 @item file @var{filename}
7885 Use @var{filename} as the program to be debugged. It is read for its
7886 symbols and for the contents of pure memory. It is also the program
7887 executed when you use the @code{run} command. If you do not specify a
7888 directory and the file is not found in the @value{GDBN} working directory,
7889 @value{GDBN} uses the environment variable @code{PATH} as a list of
7890 directories to search, just as the shell does when looking for a program
7891 to run. You can change the value of this variable, for both @value{GDBN}
7892 and your program, using the @code{path} command.
7894 On systems with memory-mapped files, an auxiliary file named
7895 @file{@var{filename}.syms} may hold symbol table information for
7896 @var{filename}. If so, @value{GDBN} maps in the symbol table from
7897 @file{@var{filename}.syms}, starting up more quickly. See the
7898 descriptions of the file options @samp{-mapped} and @samp{-readnow}
7899 (available on the command line, and with the commands @code{file},
7900 @code{symbol-file}, or @code{add-symbol-file}, described below),
7901 for more information.
7904 @code{file} with no argument makes @value{GDBN} discard any information it
7905 has on both executable file and the symbol table.
7908 @item exec-file @r{[} @var{filename} @r{]}
7909 Specify that the program to be run (but not the symbol table) is found
7910 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
7911 if necessary to locate your program. Omitting @var{filename} means to
7912 discard information on the executable file.
7915 @item symbol-file @r{[} @var{filename} @r{]}
7916 Read symbol table information from file @var{filename}. @code{PATH} is
7917 searched when necessary. Use the @code{file} command to get both symbol
7918 table and program to run from the same file.
7920 @code{symbol-file} with no argument clears out @value{GDBN} information on your
7921 program's symbol table.
7923 The @code{symbol-file} command causes @value{GDBN} to forget the contents
7924 of its convenience variables, the value history, and all breakpoints and
7925 auto-display expressions. This is because they may contain pointers to
7926 the internal data recording symbols and data types, which are part of
7927 the old symbol table data being discarded inside @value{GDBN}.
7929 @code{symbol-file} does not repeat if you press @key{RET} again after
7932 When @value{GDBN} is configured for a particular environment, it
7933 understands debugging information in whatever format is the standard
7934 generated for that environment; you may use either a @sc{gnu} compiler, or
7935 other compilers that adhere to the local conventions.
7936 Best results are usually obtained from @sc{gnu} compilers; for example,
7937 using @code{@value{GCC}} you can generate debugging information for
7940 For most kinds of object files, with the exception of old SVR3 systems
7941 using COFF, the @code{symbol-file} command does not normally read the
7942 symbol table in full right away. Instead, it scans the symbol table
7943 quickly to find which source files and which symbols are present. The
7944 details are read later, one source file at a time, as they are needed.
7946 The purpose of this two-stage reading strategy is to make @value{GDBN}
7947 start up faster. For the most part, it is invisible except for
7948 occasional pauses while the symbol table details for a particular source
7949 file are being read. (The @code{set verbose} command can turn these
7950 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
7951 warnings and messages}.)
7953 We have not implemented the two-stage strategy for COFF yet. When the
7954 symbol table is stored in COFF format, @code{symbol-file} reads the
7955 symbol table data in full right away. Note that ``stabs-in-COFF''
7956 still does the two-stage strategy, since the debug info is actually
7960 @cindex reading symbols immediately
7961 @cindex symbols, reading immediately
7963 @cindex memory-mapped symbol file
7964 @cindex saving symbol table
7965 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7966 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7967 You can override the @value{GDBN} two-stage strategy for reading symbol
7968 tables by using the @samp{-readnow} option with any of the commands that
7969 load symbol table information, if you want to be sure @value{GDBN} has the
7970 entire symbol table available.
7972 If memory-mapped files are available on your system through the
7973 @code{mmap} system call, you can use another option, @samp{-mapped}, to
7974 cause @value{GDBN} to write the symbols for your program into a reusable
7975 file. Future @value{GDBN} debugging sessions map in symbol information
7976 from this auxiliary symbol file (if the program has not changed), rather
7977 than spending time reading the symbol table from the executable
7978 program. Using the @samp{-mapped} option has the same effect as
7979 starting @value{GDBN} with the @samp{-mapped} command-line option.
7981 You can use both options together, to make sure the auxiliary symbol
7982 file has all the symbol information for your program.
7984 The auxiliary symbol file for a program called @var{myprog} is called
7985 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
7986 than the corresponding executable), @value{GDBN} always attempts to use
7987 it when you debug @var{myprog}; no special options or commands are
7990 The @file{.syms} file is specific to the host machine where you run
7991 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
7992 symbol table. It cannot be shared across multiple host platforms.
7994 @c FIXME: for now no mention of directories, since this seems to be in
7995 @c flux. 13mar1992 status is that in theory GDB would look either in
7996 @c current dir or in same dir as myprog; but issues like competing
7997 @c GDB's, or clutter in system dirs, mean that in practice right now
7998 @c only current dir is used. FFish says maybe a special GDB hierarchy
7999 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
8004 @item core-file @r{[} @var{filename} @r{]}
8005 Specify the whereabouts of a core dump file to be used as the ``contents
8006 of memory''. Traditionally, core files contain only some parts of the
8007 address space of the process that generated them; @value{GDBN} can access the
8008 executable file itself for other parts.
8010 @code{core-file} with no argument specifies that no core file is
8013 Note that the core file is ignored when your program is actually running
8014 under @value{GDBN}. So, if you have been running your program and you
8015 wish to debug a core file instead, you must kill the subprocess in which
8016 the program is running. To do this, use the @code{kill} command
8017 (@pxref{Kill Process, ,Killing the child process}).
8019 @kindex add-symbol-file
8020 @cindex dynamic linking
8021 @item add-symbol-file @var{filename} @var{address}
8022 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
8023 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address}
8024 The @code{add-symbol-file} command reads additional symbol table
8025 information from the file @var{filename}. You would use this command
8026 when @var{filename} has been dynamically loaded (by some other means)
8027 into the program that is running. @var{address} should be the memory
8028 address at which the file has been loaded; @value{GDBN} cannot figure
8029 this out for itself. You can additionally specify an arbitrary number
8030 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
8031 section name and base address for that section. You can specify any
8032 @var{address} as an expression.
8034 The symbol table of the file @var{filename} is added to the symbol table
8035 originally read with the @code{symbol-file} command. You can use the
8036 @code{add-symbol-file} command any number of times; the new symbol data
8037 thus read keeps adding to the old. To discard all old symbol data
8038 instead, use the @code{symbol-file} command without any arguments.
8040 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
8042 You can use the @samp{-mapped} and @samp{-readnow} options just as with
8043 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
8044 table information for @var{filename}.
8046 @kindex add-shared-symbol-file
8047 @item add-shared-symbol-file
8048 The @code{add-shared-symbol-file} command can be used only under Harris' CXUX
8049 operating system for the Motorola 88k. @value{GDBN} automatically looks for
8050 shared libraries, however if @value{GDBN} does not find yours, you can run
8051 @code{add-shared-symbol-file}. It takes no arguments.
8055 The @code{section} command changes the base address of section SECTION of
8056 the exec file to ADDR. This can be used if the exec file does not contain
8057 section addresses, (such as in the a.out format), or when the addresses
8058 specified in the file itself are wrong. Each section must be changed
8059 separately. The @code{info files} command, described below, lists all
8060 the sections and their addresses.
8066 @code{info files} and @code{info target} are synonymous; both print the
8067 current target (@pxref{Targets, ,Specifying a Debugging Target}),
8068 including the names of the executable and core dump files currently in
8069 use by @value{GDBN}, and the files from which symbols were loaded. The
8070 command @code{help target} lists all possible targets rather than
8075 All file-specifying commands allow both absolute and relative file names
8076 as arguments. @value{GDBN} always converts the file name to an absolute file
8077 name and remembers it that way.
8079 @cindex shared libraries
8080 @value{GDBN} supports HP-UX, SunOS, SVr4, Irix 5, and IBM RS/6000 shared
8083 @value{GDBN} automatically loads symbol definitions from shared libraries
8084 when you use the @code{run} command, or when you examine a core file.
8085 (Before you issue the @code{run} command, @value{GDBN} does not understand
8086 references to a function in a shared library, however---unless you are
8087 debugging a core file).
8089 On HP-UX, if the program loads a library explicitly, @value{GDBN}
8090 automatically loads the symbols at the time of the @code{shl_load} call.
8092 @c FIXME: some @value{GDBN} release may permit some refs to undef
8093 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
8094 @c FIXME...lib; check this from time to time when updating manual
8097 @kindex info sharedlibrary
8100 @itemx info sharedlibrary
8101 Print the names of the shared libraries which are currently loaded.
8103 @kindex sharedlibrary
8105 @item sharedlibrary @var{regex}
8106 @itemx share @var{regex}
8107 Load shared object library symbols for files matching a
8108 Unix regular expression.
8109 As with files loaded automatically, it only loads shared libraries
8110 required by your program for a core file or after typing @code{run}. If
8111 @var{regex} is omitted all shared libraries required by your program are
8115 On HP-UX systems, @value{GDBN} detects the loading of a shared library
8116 and automatically reads in symbols from the newly loaded library, up to
8117 a threshold that is initially set but that you can modify if you wish.
8119 Beyond that threshold, symbols from shared libraries must be explicitly
8120 loaded. To load these symbols, use the command @code{sharedlibrary
8121 @var{filename}}. The base address of the shared library is determined
8122 automatically by @value{GDBN} and need not be specified.
8124 To display or set the threshold, use the commands:
8127 @kindex set auto-solib-add
8128 @item set auto-solib-add @var{threshold}
8129 Set the autoloading size threshold, in megabytes. If @var{threshold} is
8130 nonzero, symbols from all shared object libraries will be loaded
8131 automatically when the inferior begins execution or when the dynamic
8132 linker informs @value{GDBN} that a new library has been loaded, until
8133 the symbol table of the program and libraries exceeds this threshold.
8134 Otherwise, symbols must be loaded manually, using the
8135 @code{sharedlibrary} command. The default threshold is 100 megabytes.
8137 @kindex show auto-solib-add
8138 @item show auto-solib-add
8139 Display the current autoloading size threshold, in megabytes.
8143 @section Errors reading symbol files
8145 While reading a symbol file, @value{GDBN} occasionally encounters problems,
8146 such as symbol types it does not recognize, or known bugs in compiler
8147 output. By default, @value{GDBN} does not notify you of such problems, since
8148 they are relatively common and primarily of interest to people
8149 debugging compilers. If you are interested in seeing information
8150 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
8151 only one message about each such type of problem, no matter how many
8152 times the problem occurs; or you can ask @value{GDBN} to print more messages,
8153 to see how many times the problems occur, with the @code{set
8154 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
8157 The messages currently printed, and their meanings, include:
8160 @item inner block not inside outer block in @var{symbol}
8162 The symbol information shows where symbol scopes begin and end
8163 (such as at the start of a function or a block of statements). This
8164 error indicates that an inner scope block is not fully contained
8165 in its outer scope blocks.
8167 @value{GDBN} circumvents the problem by treating the inner block as if it had
8168 the same scope as the outer block. In the error message, @var{symbol}
8169 may be shown as ``@code{(don't know)}'' if the outer block is not a
8172 @item block at @var{address} out of order
8174 The symbol information for symbol scope blocks should occur in
8175 order of increasing addresses. This error indicates that it does not
8178 @value{GDBN} does not circumvent this problem, and has trouble
8179 locating symbols in the source file whose symbols it is reading. (You
8180 can often determine what source file is affected by specifying
8181 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
8184 @item bad block start address patched
8186 The symbol information for a symbol scope block has a start address
8187 smaller than the address of the preceding source line. This is known
8188 to occur in the SunOS 4.1.1 (and earlier) C compiler.
8190 @value{GDBN} circumvents the problem by treating the symbol scope block as
8191 starting on the previous source line.
8193 @item bad string table offset in symbol @var{n}
8196 Symbol number @var{n} contains a pointer into the string table which is
8197 larger than the size of the string table.
8199 @value{GDBN} circumvents the problem by considering the symbol to have the
8200 name @code{foo}, which may cause other problems if many symbols end up
8203 @item unknown symbol type @code{0x@var{nn}}
8205 The symbol information contains new data types that @value{GDBN} does
8206 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
8207 uncomprehended information, in hexadecimal.
8209 @value{GDBN} circumvents the error by ignoring this symbol information.
8210 This usually allows you to debug your program, though certain symbols
8211 are not accessible. If you encounter such a problem and feel like
8212 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
8213 on @code{complain}, then go up to the function @code{read_dbx_symtab}
8214 and examine @code{*bufp} to see the symbol.
8216 @item stub type has NULL name
8218 @value{GDBN} could not find the full definition for a struct or class.
8220 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
8221 The symbol information for a C++ member function is missing some
8222 information that recent versions of the compiler should have output for
8225 @item info mismatch between compiler and debugger
8227 @value{GDBN} could not parse a type specification output by the compiler.
8232 @chapter Specifying a Debugging Target
8234 @cindex debugging target
8237 A @dfn{target} is the execution environment occupied by your program.
8239 Often, @value{GDBN} runs in the same host environment as your program;
8240 in that case, the debugging target is specified as a side effect when
8241 you use the @code{file} or @code{core} commands. When you need more
8242 flexibility---for example, running @value{GDBN} on a physically separate
8243 host, or controlling a standalone system over a serial port or a
8244 realtime system over a TCP/IP connection---you can use the @code{target}
8245 command to specify one of the target types configured for @value{GDBN}
8246 (@pxref{Target Commands, ,Commands for managing targets}).
8249 * Active Targets:: Active targets
8250 * Target Commands:: Commands for managing targets
8251 * Byte Order:: Choosing target byte order
8252 * Remote:: Remote debugging
8253 * KOD:: Kernel Object Display
8257 @node Active Targets
8258 @section Active targets
8260 @cindex stacking targets
8261 @cindex active targets
8262 @cindex multiple targets
8264 There are three classes of targets: processes, core files, and
8265 executable files. @value{GDBN} can work concurrently on up to three
8266 active targets, one in each class. This allows you to (for example)
8267 start a process and inspect its activity without abandoning your work on
8270 For example, if you execute @samp{gdb a.out}, then the executable file
8271 @code{a.out} is the only active target. If you designate a core file as
8272 well---presumably from a prior run that crashed and coredumped---then
8273 @value{GDBN} has two active targets and uses them in tandem, looking
8274 first in the corefile target, then in the executable file, to satisfy
8275 requests for memory addresses. (Typically, these two classes of target
8276 are complementary, since core files contain only a program's
8277 read-write memory---variables and so on---plus machine status, while
8278 executable files contain only the program text and initialized data.)
8280 When you type @code{run}, your executable file becomes an active process
8281 target as well. When a process target is active, all @value{GDBN}
8282 commands requesting memory addresses refer to that target; addresses in
8283 an active core file or executable file target are obscured while the
8284 process target is active.
8286 Use the @code{core-file} and @code{exec-file} commands to select a new
8287 core file or executable target (@pxref{Files, ,Commands to specify
8288 files}). To specify as a target a process that is already running, use
8289 the @code{attach} command (@pxref{Attach, ,Debugging an already-running
8292 @node Target Commands
8293 @section Commands for managing targets
8296 @item target @var{type} @var{parameters}
8297 Connects the @value{GDBN} host environment to a target machine or
8298 process. A target is typically a protocol for talking to debugging
8299 facilities. You use the argument @var{type} to specify the type or
8300 protocol of the target machine.
8302 Further @var{parameters} are interpreted by the target protocol, but
8303 typically include things like device names or host names to connect
8304 with, process numbers, and baud rates.
8306 The @code{target} command does not repeat if you press @key{RET} again
8307 after executing the command.
8311 Displays the names of all targets available. To display targets
8312 currently selected, use either @code{info target} or @code{info files}
8313 (@pxref{Files, ,Commands to specify files}).
8315 @item help target @var{name}
8316 Describe a particular target, including any parameters necessary to
8319 @kindex set gnutarget
8320 @item set gnutarget @var{args}
8321 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
8322 knows whether it is reading an @dfn{executable},
8323 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
8324 with the @code{set gnutarget} command. Unlike most @code{target} commands,
8325 with @code{gnutarget} the @code{target} refers to a program, not a machine.
8328 @emph{Warning:} To specify a file format with @code{set gnutarget},
8329 you must know the actual BFD name.
8333 @xref{Files, , Commands to specify files}.
8335 @kindex show gnutarget
8336 @item show gnutarget
8337 Use the @code{show gnutarget} command to display what file format
8338 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
8339 @value{GDBN} will determine the file format for each file automatically,
8340 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
8343 Here are some common targets (available, or not, depending on the GDB
8348 @item target exec @var{program}
8349 An executable file. @samp{target exec @var{program}} is the same as
8350 @samp{exec-file @var{program}}.
8353 @item target core @var{filename}
8354 A core dump file. @samp{target core @var{filename}} is the same as
8355 @samp{core-file @var{filename}}.
8357 @kindex target remote
8358 @item target remote @var{dev}
8359 Remote serial target in GDB-specific protocol. The argument @var{dev}
8360 specifies what serial device to use for the connection (e.g.
8361 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}. @code{target remote}
8362 supports the @code{load} command. This is only useful if you have
8363 some other way of getting the stub to the target system, and you can put
8364 it somewhere in memory where it won't get clobbered by the download.
8368 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
8376 works; however, you cannot assume that a specific memory map, device
8377 drivers, or even basic I/O is available, although some simulators do
8378 provide these. For info about any processor-specific simulator details,
8379 see the appropriate section in @ref{Embedded Processors, ,Embedded
8384 Some configurations may include these targets as well:
8389 @item target nrom @var{dev}
8390 NetROM ROM emulator. This target only supports downloading.
8394 Different targets are available on different configurations of @value{GDBN};
8395 your configuration may have more or fewer targets.
8397 Many remote targets require you to download the executable's code
8398 once you've successfully established a connection.
8402 @kindex load @var{filename}
8403 @item load @var{filename}
8404 Depending on what remote debugging facilities are configured into
8405 @value{GDBN}, the @code{load} command may be available. Where it exists, it
8406 is meant to make @var{filename} (an executable) available for debugging
8407 on the remote system---by downloading, or dynamic linking, for example.
8408 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
8409 the @code{add-symbol-file} command.
8411 If your @value{GDBN} does not have a @code{load} command, attempting to
8412 execute it gets the error message ``@code{You can't do that when your
8413 target is @dots{}}''
8415 The file is loaded at whatever address is specified in the executable.
8416 For some object file formats, you can specify the load address when you
8417 link the program; for other formats, like a.out, the object file format
8418 specifies a fixed address.
8419 @c FIXME! This would be a good place for an xref to the GNU linker doc.
8421 @code{load} does not repeat if you press @key{RET} again after using it.
8425 @section Choosing target byte order
8427 @cindex choosing target byte order
8428 @cindex target byte order
8430 Some types of processors, such as the MIPS, PowerPC, and Hitachi SH,
8431 offer the ability to run either big-endian or little-endian byte
8432 orders. Usually the executable or symbol will include a bit to
8433 designate the endian-ness, and you will not need to worry about
8434 which to use. However, you may still find it useful to adjust
8435 @value{GDBN}'s idea of processor endian-ness manually.
8438 @kindex set endian big
8439 @item set endian big
8440 Instruct @value{GDBN} to assume the target is big-endian.
8442 @kindex set endian little
8443 @item set endian little
8444 Instruct @value{GDBN} to assume the target is little-endian.
8446 @kindex set endian auto
8447 @item set endian auto
8448 Instruct @value{GDBN} to use the byte order associated with the
8452 Display @value{GDBN}'s current idea of the target byte order.
8456 Note that these commands merely adjust interpretation of symbolic
8457 data on the host, and that they have absolutely no effect on the
8461 @section Remote debugging
8462 @cindex remote debugging
8464 If you are trying to debug a program running on a machine that cannot run
8465 @value{GDBN} in the usual way, it is often useful to use remote debugging.
8466 For example, you might use remote debugging on an operating system kernel,
8467 or on a small system which does not have a general purpose operating system
8468 powerful enough to run a full-featured debugger.
8470 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
8471 to make this work with particular debugging targets. In addition,
8472 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
8473 but not specific to any particular target system) which you can use if you
8474 write the remote stubs---the code that runs on the remote system to
8475 communicate with @value{GDBN}.
8477 Other remote targets may be available in your
8478 configuration of @value{GDBN}; use @code{help target} to list them.
8481 * Remote Serial:: @value{GDBN} remote serial protocol
8485 @subsection The @value{GDBN} remote serial protocol
8487 @cindex remote serial debugging, overview
8488 To debug a program running on another machine (the debugging
8489 @dfn{target} machine), you must first arrange for all the usual
8490 prerequisites for the program to run by itself. For example, for a C
8495 A startup routine to set up the C runtime environment; these usually
8496 have a name like @file{crt0}. The startup routine may be supplied by
8497 your hardware supplier, or you may have to write your own.
8500 A C subroutine library to support your program's
8501 subroutine calls, notably managing input and output.
8504 A way of getting your program to the other machine---for example, a
8505 download program. These are often supplied by the hardware
8506 manufacturer, but you may have to write your own from hardware
8510 The next step is to arrange for your program to use a serial port to
8511 communicate with the machine where @value{GDBN} is running (the @dfn{host}
8512 machine). In general terms, the scheme looks like this:
8516 @value{GDBN} already understands how to use this protocol; when everything
8517 else is set up, you can simply use the @samp{target remote} command
8518 (@pxref{Targets,,Specifying a Debugging Target}).
8520 @item On the target,
8521 you must link with your program a few special-purpose subroutines that
8522 implement the @value{GDBN} remote serial protocol. The file containing these
8523 subroutines is called a @dfn{debugging stub}.
8525 On certain remote targets, you can use an auxiliary program
8526 @code{gdbserver} instead of linking a stub into your program.
8527 @xref{Server,,Using the @code{gdbserver} program}, for details.
8530 The debugging stub is specific to the architecture of the remote
8531 machine; for example, use @file{sparc-stub.c} to debug programs on
8534 @cindex remote serial stub list
8535 These working remote stubs are distributed with @value{GDBN}:
8540 @cindex @file{i386-stub.c}
8543 For Intel 386 and compatible architectures.
8546 @cindex @file{m68k-stub.c}
8547 @cindex Motorola 680x0
8549 For Motorola 680x0 architectures.
8552 @cindex @file{sh-stub.c}
8555 For Hitachi SH architectures.
8558 @cindex @file{sparc-stub.c}
8560 For @sc{sparc} architectures.
8563 @cindex @file{sparcl-stub.c}
8566 For Fujitsu @sc{sparclite} architectures.
8570 The @file{README} file in the @value{GDBN} distribution may list other
8571 recently added stubs.
8574 * Stub Contents:: What the stub can do for you
8575 * Bootstrapping:: What you must do for the stub
8576 * Debug Session:: Putting it all together
8577 * Protocol:: Definition of the communication protocol
8578 * Server:: Using the `gdbserver' program
8579 * NetWare:: Using the `gdbserve.nlm' program
8583 @subsubsection What the stub can do for you
8585 @cindex remote serial stub
8586 The debugging stub for your architecture supplies these three
8590 @item set_debug_traps
8591 @kindex set_debug_traps
8592 @cindex remote serial stub, initialization
8593 This routine arranges for @code{handle_exception} to run when your
8594 program stops. You must call this subroutine explicitly near the
8595 beginning of your program.
8597 @item handle_exception
8598 @kindex handle_exception
8599 @cindex remote serial stub, main routine
8600 This is the central workhorse, but your program never calls it
8601 explicitly---the setup code arranges for @code{handle_exception} to
8602 run when a trap is triggered.
8604 @code{handle_exception} takes control when your program stops during
8605 execution (for example, on a breakpoint), and mediates communications
8606 with @value{GDBN} on the host machine. This is where the communications
8607 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
8608 representative on the target machine. It begins by sending summary
8609 information on the state of your program, then continues to execute,
8610 retrieving and transmitting any information @value{GDBN} needs, until you
8611 execute a @value{GDBN} command that makes your program resume; at that point,
8612 @code{handle_exception} returns control to your own code on the target
8616 @cindex @code{breakpoint} subroutine, remote
8617 Use this auxiliary subroutine to make your program contain a
8618 breakpoint. Depending on the particular situation, this may be the only
8619 way for @value{GDBN} to get control. For instance, if your target
8620 machine has some sort of interrupt button, you won't need to call this;
8621 pressing the interrupt button transfers control to
8622 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
8623 simply receiving characters on the serial port may also trigger a trap;
8624 again, in that situation, you don't need to call @code{breakpoint} from
8625 your own program---simply running @samp{target remote} from the host
8626 @value{GDBN} session gets control.
8628 Call @code{breakpoint} if none of these is true, or if you simply want
8629 to make certain your program stops at a predetermined point for the
8630 start of your debugging session.
8634 @subsubsection What you must do for the stub
8636 @cindex remote stub, support routines
8637 The debugging stubs that come with @value{GDBN} are set up for a particular
8638 chip architecture, but they have no information about the rest of your
8639 debugging target machine.
8641 First of all you need to tell the stub how to communicate with the
8645 @item int getDebugChar()
8646 @kindex getDebugChar
8647 Write this subroutine to read a single character from the serial port.
8648 It may be identical to @code{getchar} for your target system; a
8649 different name is used to allow you to distinguish the two if you wish.
8651 @item void putDebugChar(int)
8652 @kindex putDebugChar
8653 Write this subroutine to write a single character to the serial port.
8654 It may be identical to @code{putchar} for your target system; a
8655 different name is used to allow you to distinguish the two if you wish.
8658 @cindex control C, and remote debugging
8659 @cindex interrupting remote targets
8660 If you want @value{GDBN} to be able to stop your program while it is
8661 running, you need to use an interrupt-driven serial driver, and arrange
8662 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
8663 character). That is the character which @value{GDBN} uses to tell the
8664 remote system to stop.
8666 Getting the debugging target to return the proper status to @value{GDBN}
8667 probably requires changes to the standard stub; one quick and dirty way
8668 is to just execute a breakpoint instruction (the ``dirty'' part is that
8669 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
8671 Other routines you need to supply are:
8674 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
8675 @kindex exceptionHandler
8676 Write this function to install @var{exception_address} in the exception
8677 handling tables. You need to do this because the stub does not have any
8678 way of knowing what the exception handling tables on your target system
8679 are like (for example, the processor's table might be in @sc{rom},
8680 containing entries which point to a table in @sc{ram}).
8681 @var{exception_number} is the exception number which should be changed;
8682 its meaning is architecture-dependent (for example, different numbers
8683 might represent divide by zero, misaligned access, etc). When this
8684 exception occurs, control should be transferred directly to
8685 @var{exception_address}, and the processor state (stack, registers,
8686 and so on) should be just as it is when a processor exception occurs. So if
8687 you want to use a jump instruction to reach @var{exception_address}, it
8688 should be a simple jump, not a jump to subroutine.
8690 For the 386, @var{exception_address} should be installed as an interrupt
8691 gate so that interrupts are masked while the handler runs. The gate
8692 should be at privilege level 0 (the most privileged level). The
8693 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
8694 help from @code{exceptionHandler}.
8696 @item void flush_i_cache()
8697 @kindex flush_i_cache
8698 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
8699 instruction cache, if any, on your target machine. If there is no
8700 instruction cache, this subroutine may be a no-op.
8702 On target machines that have instruction caches, @value{GDBN} requires this
8703 function to make certain that the state of your program is stable.
8707 You must also make sure this library routine is available:
8710 @item void *memset(void *, int, int)
8712 This is the standard library function @code{memset} that sets an area of
8713 memory to a known value. If you have one of the free versions of
8714 @code{libc.a}, @code{memset} can be found there; otherwise, you must
8715 either obtain it from your hardware manufacturer, or write your own.
8718 If you do not use the GNU C compiler, you may need other standard
8719 library subroutines as well; this varies from one stub to another,
8720 but in general the stubs are likely to use any of the common library
8721 subroutines which @code{@value{GCC}} generates as inline code.
8725 @subsubsection Putting it all together
8727 @cindex remote serial debugging summary
8728 In summary, when your program is ready to debug, you must follow these
8733 Make sure you have defined the supporting low-level routines
8734 (@pxref{Bootstrapping,,What you must do for the stub}):
8736 @code{getDebugChar}, @code{putDebugChar},
8737 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
8741 Insert these lines near the top of your program:
8749 For the 680x0 stub only, you need to provide a variable called
8750 @code{exceptionHook}. Normally you just use:
8753 void (*exceptionHook)() = 0;
8757 but if before calling @code{set_debug_traps}, you set it to point to a
8758 function in your program, that function is called when
8759 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
8760 error). The function indicated by @code{exceptionHook} is called with
8761 one parameter: an @code{int} which is the exception number.
8764 Compile and link together: your program, the @value{GDBN} debugging stub for
8765 your target architecture, and the supporting subroutines.
8768 Make sure you have a serial connection between your target machine and
8769 the @value{GDBN} host, and identify the serial port on the host.
8772 @c The "remote" target now provides a `load' command, so we should
8773 @c document that. FIXME.
8774 Download your program to your target machine (or get it there by
8775 whatever means the manufacturer provides), and start it.
8778 To start remote debugging, run @value{GDBN} on the host machine, and specify
8779 as an executable file the program that is running in the remote machine.
8780 This tells @value{GDBN} how to find your program's symbols and the contents
8784 @cindex serial line, @code{target remote}
8785 Establish communication using the @code{target remote} command.
8786 Its argument specifies how to communicate with the target
8787 machine---either via a devicename attached to a direct serial line, or a
8788 TCP port (usually to a terminal server which in turn has a serial line
8789 to the target). For example, to use a serial line connected to the
8790 device named @file{/dev/ttyb}:
8793 target remote /dev/ttyb
8796 @cindex TCP port, @code{target remote}
8797 To use a TCP connection, use an argument of the form
8798 @code{@var{host}:port}. For example, to connect to port 2828 on a
8799 terminal server named @code{manyfarms}:
8802 target remote manyfarms:2828
8806 Now you can use all the usual commands to examine and change data and to
8807 step and continue the remote program.
8809 To resume the remote program and stop debugging it, use the @code{detach}
8812 @cindex interrupting remote programs
8813 @cindex remote programs, interrupting
8814 Whenever @value{GDBN} is waiting for the remote program, if you type the
8815 interrupt character (often @key{C-C}), @value{GDBN} attempts to stop the
8816 program. This may or may not succeed, depending in part on the hardware
8817 and the serial drivers the remote system uses. If you type the
8818 interrupt character once again, @value{GDBN} displays this prompt:
8821 Interrupted while waiting for the program.
8822 Give up (and stop debugging it)? (y or n)
8825 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
8826 (If you decide you want to try again later, you can use @samp{target
8827 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
8828 goes back to waiting.
8831 @subsubsection Communication protocol
8833 @cindex debugging stub, example
8834 @cindex remote stub, example
8835 @cindex stub example, remote debugging
8836 The stub files provided with @value{GDBN} implement the target side of the
8837 communication protocol, and the @value{GDBN} side is implemented in the
8838 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
8839 these subroutines to communicate, and ignore the details. (If you're
8840 implementing your own stub file, you can still ignore the details: start
8841 with one of the existing stub files. @file{sparc-stub.c} is the best
8842 organized, and therefore the easiest to read.)
8844 However, there may be occasions when you need to know something about
8845 the protocol---for example, if there is only one serial port to your
8846 target machine, you might want your program to do something special if
8847 it recognizes a packet meant for @value{GDBN}.
8849 In the examples below, @samp{<-} and @samp{->} are used to indicate
8850 transmitted and received data respectfully.
8852 @cindex protocol, @value{GDBN} remote serial
8853 @cindex serial protocol, @value{GDBN} remote
8854 @cindex remote serial protocol
8855 All @value{GDBN} commands and responses (other than acknowledgments) are
8856 sent as a @var{packet}. A @var{packet} is introduced with the character
8857 @samp{$}, the actual @var{packet-data}, and the terminating character
8858 @samp{#} followed by a two-digit @var{checksum}:
8861 @code{$}@var{packet-data}@code{#}@var{checksum}
8865 @cindex checksum, for @value{GDBN} remote
8867 The two-digit @var{checksum} is computed as the modulo 256 sum of all
8868 characters between the leading @samp{$} and the trailing @samp{#} (an
8869 eight bit unsigned checksum).
8871 Implementors should note that prior to @value{GDBN} 5.0 the protocol
8872 specification also included an optional two-digit @var{sequence-id}:
8875 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
8878 @cindex sequence-id, for @value{GDBN} remote
8880 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
8881 has never output @var{sequence-id}s. Stubs that handle packets added
8882 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
8884 @cindex acknowledgment, for @value{GDBN} remote
8885 When either the host or the target machine receives a packet, the first
8886 response expected is an acknowledgment: either @samp{+} (to indicate
8887 the package was received correctly) or @samp{-} (to request
8891 <- @code{$}@var{packet-data}@code{#}@var{checksum}
8896 The host (@value{GDBN}) sends @var{command}s, and the target (the
8897 debugging stub incorporated in your program) sends a @var{response}. In
8898 the case of step and continue @var{command}s, the response is only sent
8899 when the operation has completed (the target has again stopped).
8901 @var{packet-data} consists of a sequence of characters with the
8902 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
8905 Fields within the packet should be separated using @samp{,} @samp{;} or
8906 @samp{:}. Except where otherwise noted all numbers are represented in
8907 HEX with leading zeros suppressed.
8909 Implementors should note that prior to @value{GDBN} 5.0, the character
8910 @samp{:} could not appear as the third character in a packet (as it
8911 would potentially conflict with the @var{sequence-id}).
8913 Response @var{data} can be run-length encoded to save space. A @samp{*}
8914 means that the next character is an @sc{ascii} encoding giving a repeat count
8915 which stands for that many repetitions of the character preceding the
8916 @samp{*}. The encoding is @code{n+29}, yielding a printable character
8917 where @code{n >=3} (which is where rle starts to win). The printable
8918 characters @samp{$}, @samp{#}, @samp{+} and @samp{-} or with a numeric
8919 value greater than 126 should not be used.
8921 Some remote systems have used a different run-length encoding mechanism
8922 loosely refered to as the cisco encoding. Following the @samp{*}
8923 character are two hex digits that indicate the size of the packet.
8930 means the same as "0000".
8932 The error response returned for some packets includes a two character
8933 error number. That number is not well defined.
8935 For any @var{command} not supported by the stub, an empty response
8936 (@samp{$#00}) should be returned. That way it is possible to extend the
8937 protocol. A newer @value{GDBN} can tell if a packet is supported based
8940 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
8941 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
8944 Below is a complete list of all currently defined @var{command}s and
8945 their corresponding response @var{data}:
8947 @multitable @columnfractions .30 .30 .40
8955 Use the extended remote protocol. Sticky---only needs to be set once.
8956 The extended remote protocol supports the @samp{R} packet.
8960 Stubs that support the extended remote protocol return @samp{} which,
8961 unfortunately, is identical to the response returned by stubs that do not
8962 support protocol extensions.
8967 Indicate the reason the target halted. The reply is the same as for step
8976 @tab Reserved for future use
8978 @item set program arguments @strong{(reserved)}
8979 @tab @code{A}@var{arglen}@code{,}@var{argnum}@code{,}@var{arg}@code{,...}
8984 Initialized @samp{argv[]} array passed into program. @var{arglen}
8985 specifies the number of bytes in the hex encoded byte stream @var{arg}.
8986 See @file{gdbserver} for more details.
8988 @tab reply @code{OK}
8990 @tab reply @code{E}@var{NN}
8992 @item set baud @strong{(deprecated)}
8993 @tab @code{b}@var{baud}
8995 Change the serial line speed to @var{baud}. JTC: @emph{When does the
8996 transport layer state change? When it's received, or after the ACK is
8997 transmitted. In either case, there are problems if the command or the
8998 acknowledgment packet is dropped.} Stan: @emph{If people really wanted
8999 to add something like this, and get it working for the first time, they
9000 ought to modify ser-unix.c to send some kind of out-of-band message to a
9001 specially-setup stub and have the switch happen "in between" packets, so
9002 that from remote protocol's point of view, nothing actually
9005 @item set breakpoint @strong{(deprecated)}
9006 @tab @code{B}@var{addr},@var{mode}
9008 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
9009 breakpoint at @var{addr}. @emph{This has been replaced by the @samp{Z} and
9013 @tab @code{c}@var{addr}
9015 @var{addr} is address to resume. If @var{addr} is omitted, resume at
9021 @item continue with signal
9022 @tab @code{C}@var{sig}@code{;}@var{addr}
9024 Continue with signal @var{sig} (hex signal number). If
9025 @code{;}@var{addr} is omitted, resume at same address.
9030 @item toggle debug @strong{(deprecated)}
9038 Detach @value{GDBN} from the remote system. Sent to the remote target before
9039 @value{GDBN} disconnects.
9041 @tab reply @emph{no response}
9043 @value{GDBN} does not check for any response after sending this packet.
9047 @tab Reserved for future use
9051 @tab Reserved for future use
9055 @tab Reserved for future use
9059 @tab Reserved for future use
9061 @item read registers
9063 @tab Read general registers.
9065 @tab reply @var{XX...}
9067 Each byte of register data is described by two hex digits. The bytes
9068 with the register are transmitted in target byte order. The size of
9069 each register and their position within the @samp{g} @var{packet} are
9070 determined by the @value{GDBN} internal macros @var{REGISTER_RAW_SIZE} and
9071 @var{REGISTER_NAME} macros. The specification of several standard
9072 @code{g} packets is specified below.
9074 @tab @code{E}@var{NN}
9078 @tab @code{G}@var{XX...}
9080 See @samp{g} for a description of the @var{XX...} data.
9082 @tab reply @code{OK}
9085 @tab reply @code{E}@var{NN}
9090 @tab Reserved for future use
9093 @tab @code{H}@var{c}@var{t...}
9095 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
9096 @samp{G}, et.al.). @var{c} = @samp{c} for thread used in step and
9097 continue; @var{t...} can be -1 for all threads. @var{c} = @samp{g} for
9098 thread used in other operations. If zero, pick a thread, any thread.
9100 @tab reply @code{OK}
9103 @tab reply @code{E}@var{NN}
9107 @c 'H': How restrictive (or permissive) is the thread model. If a
9108 @c thread is selected and stopped, are other threads allowed
9109 @c to continue to execute? As I mentioned above, I think the
9110 @c semantics of each command when a thread is selected must be
9111 @c described. For example:
9113 @c 'g': If the stub supports threads and a specific thread is
9114 @c selected, returns the register block from that thread;
9115 @c otherwise returns current registers.
9117 @c 'G' If the stub supports threads and a specific thread is
9118 @c selected, sets the registers of the register block of
9119 @c that thread; otherwise sets current registers.
9121 @item cycle step @strong{(draft)}
9122 @tab @code{i}@var{addr}@code{,}@var{nnn}
9124 Step the remote target by a single clock cycle. If @code{,}@var{nnn} is
9125 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
9126 step starting at that address.
9128 @item signal then cycle step @strong{(reserved)}
9131 See @samp{i} and @samp{S} for likely syntax and semantics.
9135 @tab Reserved for future use
9139 @tab Reserved for future use
9144 FIXME: @emph{There is no description of how operate when a specific
9145 thread context has been selected (ie. does 'k' kill only that thread?)}.
9149 @tab Reserved for future use
9153 @tab Reserved for future use
9156 @tab @code{m}@var{addr}@code{,}@var{length}
9158 Read @var{length} bytes of memory starting at address @var{addr}.
9159 Neither @value{GDBN} nor the stub assume that sized memory transfers are assumed
9160 using word alligned accesses. FIXME: @emph{A word aligned memory
9161 transfer mechanism is needed.}
9163 @tab reply @var{XX...}
9165 @var{XX...} is mem contents. Can be fewer bytes than requested if able
9166 to read only part of the data. Neither @value{GDBN} nor the stub assume that
9167 sized memory transfers are assumed using word alligned accesses. FIXME:
9168 @emph{A word aligned memory transfer mechanism is needed.}
9170 @tab reply @code{E}@var{NN}
9171 @tab @var{NN} is errno
9174 @tab @code{M}@var{addr},@var{length}@code{:}@var{XX...}
9176 Write @var{length} bytes of memory starting at address @var{addr}.
9177 @var{XX...} is the data.
9179 @tab reply @code{OK}
9182 @tab reply @code{E}@var{NN}
9184 for an error (this includes the case where only part of the data was
9189 @tab Reserved for future use
9193 @tab Reserved for future use
9197 @tab Reserved for future use
9201 @tab Reserved for future use
9203 @item read reg @strong{(reserved)}
9204 @tab @code{p}@var{n...}
9208 @tab return @var{r....}
9209 @tab The hex encoded value of the register in target byte order.
9212 @tab @code{P}@var{n...}@code{=}@var{r...}
9214 Write register @var{n...} with value @var{r...}, which contains two hex
9215 digits for each byte in the register (target byte order).
9217 @tab reply @code{OK}
9220 @tab reply @code{E}@var{NN}
9224 @tab @code{q}@var{query}
9226 Request info about @var{query}. In general @value{GDBN} queries
9227 have a leading upper case letter. Custom vendor queries should use a
9228 company prefix (in lower case) ex: @samp{qfsf.var}. @var{query} may
9229 optionally be followed by a @samp{,} or @samp{;} separated list. Stubs
9230 must ensure that they match the full @var{query} name.
9232 @tab reply @code{XX...}
9233 @tab Hex encoded data from query. The reply can not be empty.
9235 @tab reply @code{E}@var{NN}
9239 @tab Indicating an unrecognized @var{query}.
9242 @tab @code{Q}@var{var}@code{=}@var{val}
9244 Set value of @var{var} to @var{val}. See @samp{q} for a discussing of
9247 @item reset @strong{(deprecated)}
9250 Reset the entire system.
9252 @item remote restart
9253 @tab @code{R}@var{XX}
9255 Restart the remote server. @var{XX} while needed has no clear
9256 definition. FIXME: @emph{An example interaction explaining how this
9257 packet is used in extended-remote mode is needed}.
9260 @tab @code{s}@var{addr}
9262 @var{addr} is address to resume. If @var{addr} is omitted, resume at
9268 @item step with signal
9269 @tab @code{S}@var{sig}@code{;}@var{addr}
9271 Like @samp{C} but step not continue.
9277 @tab @code{t}@var{addr}@code{:}@var{PP}@code{,}@var{MM}
9279 Search backwards starting at address @var{addr} for a match with pattern
9280 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4
9281 bytes. @var{addr} must be at least 3 digits.
9284 @tab @code{T}@var{XX}
9285 @tab Find out if the thread XX is alive.
9287 @tab reply @code{OK}
9288 @tab thread is still alive
9290 @tab reply @code{E}@var{NN}
9295 @tab Reserved for future use
9299 @tab Reserved for future use
9303 @tab Reserved for future use
9307 @tab Reserved for future use
9311 @tab Reserved for future use
9315 @tab Reserved for future use
9319 @tab Reserved for future use
9321 @item write mem (binary)
9322 @tab @code{X}@var{addr}@code{,}@var{length}@var{:}@var{XX...}
9324 @var{addr} is address, @var{length} is number of bytes, @var{XX...} is
9325 binary data. The characters @code{$}, @code{#}, and @code{0x7d} are
9326 escaped using @code{0x7d}.
9328 @tab reply @code{OK}
9331 @tab reply @code{E}@var{NN}
9336 @tab Reserved for future use
9340 @tab Reserved for future use
9342 @item remove break or watchpoint @strong{(draft)}
9343 @tab @code{z}@var{t}@code{,}@var{addr}@code{,}@var{length}
9347 @item insert break or watchpoint @strong{(draft)}
9348 @tab @code{Z}@var{t}@code{,}@var{addr}@code{,}@var{length}
9350 @var{t} is type: @samp{0} - software breakpoint, @samp{1} - hardware
9351 breakpoint, @samp{2} - write watchpoint, @samp{3} - read watchpoint,
9352 @samp{4} - access watchpoint; @var{addr} is address; @var{length} is in
9353 bytes. For a software breakpoint, @var{length} specifies the size of
9354 the instruction to be patched. For hardware breakpoints and watchpoints
9355 @var{length} specifies the memory region to be monitored. To avoid
9356 potential problems with duplicate packets, the operations should be
9357 implemented in an idempotent way.
9359 @tab reply @code{E}@var{NN}
9362 @tab reply @code{OK}
9366 @tab If not supported.
9370 @tab Reserved for future use
9374 The @samp{C}, @samp{c}, @samp{S}, @samp{s} and @samp{?} packets can
9375 receive any of the below as a reply. In the case of the @samp{C},
9376 @samp{c}, @samp{S} and @samp{s} packets, that reply is only returned
9377 when the target halts. In the below the exact meaning of @samp{signal
9378 number} is poorly defined. In general one of the UNIX signal numbering
9379 conventions is used.
9381 @multitable @columnfractions .4 .6
9383 @item @code{S}@var{AA}
9384 @tab @var{AA} is the signal number
9386 @item @code{T}@var{AA}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}
9388 @var{AA} = two hex digit signal number; @var{n...} = register number
9389 (hex), @var{r...} = target byte ordered register contents, size defined
9390 by @code{REGISTER_RAW_SIZE}; @var{n...} = @samp{thread}, @var{r...} =
9391 thread process ID, this is a hex integer; @var{n...} = other string not
9392 starting with valid hex digit. @value{GDBN} should ignore this
9393 @var{n...}, @var{r...} pair and go on to the next. This way we can
9394 extend the protocol.
9396 @item @code{W}@var{AA}
9398 The process exited, and @var{AA} is the exit status. This is only
9399 applicable for certains sorts of targets.
9401 @item @code{X}@var{AA}
9403 The process terminated with signal @var{AA}.
9405 @item @code{N}@var{AA}@code{;}@var{t...}@code{;}@var{d...}@code{;}@var{b...} @strong{(obsolete)}
9407 @var{AA} = signal number; @var{t...} = address of symbol "_start";
9408 @var{d...} = base of data section; @var{b...} = base of bss section.
9409 @emph{Note: only used by Cisco Systems targets. The difference between
9410 this reply and the "qOffsets" query is that the 'N' packet may arrive
9411 spontaneously whereas the 'qOffsets' is a query initiated by the host
9414 @item @code{O}@var{XX...}
9416 @var{XX...} is hex encoding of @sc{ascii} data. This can happen at any time
9417 while the program is running and the debugger should continue to wait
9422 The following set and query packets have already been defined.
9424 @multitable @columnfractions .2 .2 .6
9426 @item current thread
9427 @tab @code{q}@code{C}
9428 @tab Return the current thread id.
9430 @tab reply @code{QC}@var{pid}
9432 Where @var{pid} is a HEX encoded 16 bit process id.
9435 @tab Any other reply implies the old pid.
9437 @item all thread ids
9438 @tab @code{q}@code{fThreadInfo}
9440 @tab @code{q}@code{sThreadInfo}
9442 Obtain a list of active thread ids from the target (OS). Since there
9443 may be too many active threads to fit into one reply packet, this query
9444 works iteratively: it may require more than one query/reply sequence to
9445 obtain the entire list of threads. The first query of the sequence will
9446 be the @code{qf}@code{ThreadInfo} query; subsequent queries in the
9447 sequence will be the @code{qs}@code{ThreadInfo} query.
9450 @tab NOTE: replaces the @code{qL} query (see below).
9452 @tab reply @code{m}@var{<id>}
9453 @tab A single thread id
9455 @tab reply @code{m}@var{<id>},@var{<id>...}
9456 @tab a comma-separated list of thread ids
9459 @tab (lower case 'el') denotes end of list.
9463 In response to each query, the target will reply with a list of one
9464 or more thread ids, in big-endian hex, separated by commas. GDB will
9465 respond to each reply with a request for more thread ids (using the
9466 @code{qs} form of the query), until the target responds with @code{l}
9467 (lower-case el, for @code{'last'}).
9469 @item extra thread info
9470 @tab @code{q}@code{ThreadExtraInfo}@code{,}@var{id}
9475 Where @var{<id>} is a thread-id in big-endian hex.
9476 Obtain a printable string description of a thread's attributes from
9477 the target OS. This string may contain anything that the target OS
9478 thinks is interesting for @value{GDBN} to tell the user about the thread.
9479 The string is displayed in @value{GDBN}'s @samp{info threads} display.
9480 Some examples of possible thread extra info strings are "Runnable", or
9483 @tab reply @var{XX...}
9485 Where @var{XX...} is a hex encoding of @sc{ascii} data, comprising the
9486 printable string containing the extra information about the thread's
9489 @item query @var{LIST} or @var{threadLIST} @strong{(deprecated)}
9490 @tab @code{q}@code{L}@var{startflag}@var{threadcount}@var{nextthread}
9495 Obtain thread information from RTOS. Where: @var{startflag} (one hex
9496 digit) is one to indicate the first query and zero to indicate a
9497 subsequent query; @var{threadcount} (two hex digits) is the maximum
9498 number of threads the response packet can contain; and @var{nextthread}
9499 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
9500 returned in the response as @var{argthread}.
9503 @tab NOTE: this query is replaced by the @code{q}@code{fThreadInfo}
9506 @tab reply @code{q}@code{M}@var{count}@var{done}@var{argthread}@var{thread...}
9511 Where: @var{count} (two hex digits) is the number of threads being
9512 returned; @var{done} (one hex digit) is zero to indicate more threads
9513 and one indicates no further threads; @var{argthreadid} (eight hex
9514 digits) is @var{nextthread} from the request packet; @var{thread...} is
9515 a sequence of thread IDs from the target. @var{threadid} (eight hex
9516 digits). See @code{remote.c:parse_threadlist_response()}.
9518 @item compute CRC of memory block
9519 @tab @code{q}@code{CRC:}@var{addr}@code{,}@var{length}
9522 @tab reply @code{E}@var{NN}
9523 @tab An error (such as memory fault)
9525 @tab reply @code{C}@var{CRC32}
9526 @tab A 32 bit cyclic redundancy check of the specified memory region.
9528 @item query sect offs
9529 @tab @code{q}@code{Offsets}
9531 Get section offsets that the target used when re-locating the downloaded
9532 image. @emph{Note: while a @code{Bss} offset is included in the
9533 response, @value{GDBN} ignores this and instead applies the @code{Data}
9534 offset to the @code{Bss} section.}
9536 @tab reply @code{Text=}@var{xxx}@code{;Data=}@var{yyy}@code{;Bss=}@var{zzz}
9538 @item thread info request
9539 @tab @code{q}@code{P}@var{mode}@var{threadid}
9544 Returns information on @var{threadid}. Where: @var{mode} is a hex
9545 encoded 32 bit mode; @var{threadid} is a hex encoded 64 bit thread ID.
9549 See @code{remote.c:remote_unpack_thread_info_response()}.
9551 @item remote command
9552 @tab @code{q}@code{Rcmd,}@var{COMMAND}
9557 @var{COMMAND} (hex encoded) is passed to the local interpreter for
9558 execution. Invalid commands should be reported using the output string.
9559 Before the final result packet, the target may also respond with a
9560 number of intermediate @code{O}@var{OUTPUT} console output
9561 packets. @emph{Implementors should note that providing access to a
9562 stubs's interpreter may have security implications}.
9564 @tab reply @code{OK}
9566 A command response with no output.
9568 @tab reply @var{OUTPUT}
9570 A command response with the hex encoded output string @var{OUTPUT}.
9572 @tab reply @code{E}@var{NN}
9574 Indicate a badly formed request.
9579 When @samp{q}@samp{Rcmd} is not recognized.
9583 The following @samp{g}/@samp{G} packets have previously been defined.
9584 In the below, some thirty-two bit registers are transferred as sixty-four
9585 bits. Those registers should be zero/sign extended (which?) to fill the
9586 space allocated. Register bytes are transfered in target byte order.
9587 The two nibbles within a register byte are transfered most-significant -
9590 @multitable @columnfractions .5 .5
9594 All registers are transfered as thirty-two bit quantities in the order:
9595 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
9596 registers; fsr; fir; fp.
9600 All registers are transfered as sixty-four bit quantities (including
9601 thirty-two bit registers such as @code{sr}). The ordering is the same
9606 Example sequence of a target being re-started. Notice how the restart
9607 does not get any direct output:
9612 @emph{target restarts}
9615 -> @code{T001:1234123412341234}
9619 Example sequence of a target being stepped by a single instruction:
9627 -> @code{T001:1234123412341234}
9636 @subsubsection Using the @code{gdbserver} program
9639 @cindex remote connection without stubs
9640 @code{gdbserver} is a control program for Unix-like systems, which
9641 allows you to connect your program with a remote @value{GDBN} via
9642 @code{target remote}---but without linking in the usual debugging stub.
9644 @code{gdbserver} is not a complete replacement for the debugging stubs,
9645 because it requires essentially the same operating-system facilities
9646 that @value{GDBN} itself does. In fact, a system that can run
9647 @code{gdbserver} to connect to a remote @value{GDBN} could also run
9648 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
9649 because it is a much smaller program than @value{GDBN} itself. It is
9650 also easier to port than all of @value{GDBN}, so you may be able to get
9651 started more quickly on a new system by using @code{gdbserver}.
9652 Finally, if you develop code for real-time systems, you may find that
9653 the tradeoffs involved in real-time operation make it more convenient to
9654 do as much development work as possible on another system, for example
9655 by cross-compiling. You can use @code{gdbserver} to make a similar
9656 choice for debugging.
9658 @value{GDBN} and @code{gdbserver} communicate via either a serial line
9659 or a TCP connection, using the standard @value{GDBN} remote serial
9663 @item On the target machine,
9664 you need to have a copy of the program you want to debug.
9665 @code{gdbserver} does not need your program's symbol table, so you can
9666 strip the program if necessary to save space. @value{GDBN} on the host
9667 system does all the symbol handling.
9669 To use the server, you must tell it how to communicate with @value{GDBN};
9670 the name of your program; and the arguments for your program. The
9674 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
9677 @var{comm} is either a device name (to use a serial line) or a TCP
9678 hostname and portnumber. For example, to debug Emacs with the argument
9679 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
9683 target> gdbserver /dev/com1 emacs foo.txt
9686 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
9689 To use a TCP connection instead of a serial line:
9692 target> gdbserver host:2345 emacs foo.txt
9695 The only difference from the previous example is the first argument,
9696 specifying that you are communicating with the host @value{GDBN} via
9697 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
9698 expect a TCP connection from machine @samp{host} to local TCP port 2345.
9699 (Currently, the @samp{host} part is ignored.) You can choose any number
9700 you want for the port number as long as it does not conflict with any
9701 TCP ports already in use on the target system (for example, @code{23} is
9702 reserved for @code{telnet}).@footnote{If you choose a port number that
9703 conflicts with another service, @code{gdbserver} prints an error message
9704 and exits.} You must use the same port number with the host @value{GDBN}
9705 @code{target remote} command.
9707 @item On the @value{GDBN} host machine,
9708 you need an unstripped copy of your program, since @value{GDBN} needs
9709 symbols and debugging information. Start up @value{GDBN} as usual,
9710 using the name of the local copy of your program as the first argument.
9711 (You may also need the @w{@samp{--baud}} option if the serial line is
9712 running at anything other than 9600@dmn{bps}.) After that, use @code{target
9713 remote} to establish communications with @code{gdbserver}. Its argument
9714 is either a device name (usually a serial device, like
9715 @file{/dev/ttyb}), or a TCP port descriptor in the form
9716 @code{@var{host}:@var{PORT}}. For example:
9719 (@value{GDBP}) target remote /dev/ttyb
9723 communicates with the server via serial line @file{/dev/ttyb}, and
9726 (@value{GDBP}) target remote the-target:2345
9730 communicates via a TCP connection to port 2345 on host @w{@file{the-target}}.
9731 For TCP connections, you must start up @code{gdbserver} prior to using
9732 the @code{target remote} command. Otherwise you may get an error whose
9733 text depends on the host system, but which usually looks something like
9734 @samp{Connection refused}.
9738 @subsubsection Using the @code{gdbserve.nlm} program
9740 @kindex gdbserve.nlm
9741 @code{gdbserve.nlm} is a control program for NetWare systems, which
9742 allows you to connect your program with a remote @value{GDBN} via
9743 @code{target remote}.
9745 @value{GDBN} and @code{gdbserve.nlm} communicate via a serial line,
9746 using the standard @value{GDBN} remote serial protocol.
9749 @item On the target machine,
9750 you need to have a copy of the program you want to debug.
9751 @code{gdbserve.nlm} does not need your program's symbol table, so you
9752 can strip the program if necessary to save space. @value{GDBN} on the
9753 host system does all the symbol handling.
9755 To use the server, you must tell it how to communicate with
9756 @value{GDBN}; the name of your program; and the arguments for your
9757 program. The syntax is:
9760 load gdbserve [ BOARD=@var{board} ] [ PORT=@var{port} ]
9761 [ BAUD=@var{baud} ] @var{program} [ @var{args} @dots{} ]
9764 @var{board} and @var{port} specify the serial line; @var{baud} specifies
9765 the baud rate used by the connection. @var{port} and @var{node} default
9766 to 0, @var{baud} defaults to 9600@dmn{bps}.
9768 For example, to debug Emacs with the argument @samp{foo.txt}and
9769 communicate with @value{GDBN} over serial port number 2 or board 1
9770 using a 19200@dmn{bps} connection:
9773 load gdbserve BOARD=1 PORT=2 BAUD=19200 emacs foo.txt
9776 @item On the @value{GDBN} host machine,
9777 you need an unstripped copy of your program, since @value{GDBN} needs
9778 symbols and debugging information. Start up @value{GDBN} as usual,
9779 using the name of the local copy of your program as the first argument.
9780 (You may also need the @w{@samp{--baud}} option if the serial line is
9781 running at anything other than 9600@dmn{bps}. After that, use @code{target
9782 remote} to establish communications with @code{gdbserve.nlm}. Its
9783 argument is a device name (usually a serial device, like
9784 @file{/dev/ttyb}). For example:
9787 (@value{GDBP}) target remote /dev/ttyb
9791 communications with the server via serial line @file{/dev/ttyb}.
9795 @section Kernel Object Display
9797 @cindex kernel object display
9798 @cindex kernel object
9801 Some targets support kernel object display. Using this facility,
9802 @value{GDBN} communicates specially with the underlying operating system
9803 and can display information about operating system-level objects such as
9804 mutexes and other synchronization objects. Exactly which objects can be
9805 displayed is determined on a per-OS basis.
9807 Use the @code{set os} command to set the operating system. This tells
9808 @value{GDBN} which kernel object display module to initialize:
9811 (@value{GDBP}) set os cisco
9814 If @code{set os} succeeds, @value{GDBN} will display some information
9815 about the operating system, and will create a new @code{info} command
9816 which can be used to query the target. The @code{info} command is named
9817 after the operating system:
9820 (@value{GDBP}) info cisco
9821 List of Cisco Kernel Objects
9823 any Any and all objects
9826 Further subcommands can be used to query about particular objects known
9829 There is currently no way to determine whether a given operating system
9830 is supported other than to try it.
9833 @node Configurations
9834 @chapter Configuration-Specific Information
9836 While nearly all @value{GDBN} commands are available for all native and
9837 cross versions of the debugger, there are some exceptions. This chapter
9838 describes things that are only available in certain configurations.
9840 There are three major categories of configurations: native
9841 configurations, where the host and target are the same, embedded
9842 operating system configurations, which are usually the same for several
9843 different processor architectures, and bare embedded processors, which
9844 are quite different from each other.
9849 * Embedded Processors::
9856 This section describes details specific to particular native
9861 * SVR4 Process Information:: SVR4 process information
9867 On HP-UX systems, if you refer to a function or variable name that
9868 begins with a dollar sign, @value{GDBN} searches for a user or system
9869 name first, before it searches for a convenience variable.
9871 @node SVR4 Process Information
9872 @subsection SVR4 process information
9875 @cindex process image
9877 Many versions of SVR4 provide a facility called @samp{/proc} that can be
9878 used to examine the image of a running process using file-system
9879 subroutines. If @value{GDBN} is configured for an operating system with
9880 this facility, the command @code{info proc} is available to report on
9881 several kinds of information about the process running your program.
9882 @code{info proc} works only on SVR4 systems that include the
9883 @code{procfs} code. This includes OSF/1 (Digital Unix), Solaris, Irix,
9884 and Unixware, but not HP-UX or Linux, for example.
9889 Summarize available information about the process.
9891 @kindex info proc mappings
9892 @item info proc mappings
9893 Report on the address ranges accessible in the program, with information
9894 on whether your program may read, write, or execute each range.
9896 @kindex info proc times
9897 @item info proc times
9898 Starting time, user CPU time, and system CPU time for your program and
9901 @kindex info proc id
9903 Report on the process IDs related to your program: its own process ID,
9904 the ID of its parent, the process group ID, and the session ID.
9906 @kindex info proc status
9907 @item info proc status
9908 General information on the state of the process. If the process is
9909 stopped, this report includes the reason for stopping, and any signal
9913 Show all the above information about the process.
9917 @section Embedded Operating Systems
9919 This section describes configurations involving the debugging of
9920 embedded operating systems that are available for several different
9924 * VxWorks:: Using @value{GDBN} with VxWorks
9927 @value{GDBN} includes the ability to debug programs running on
9928 various real-time operating systems.
9931 @subsection Using @value{GDBN} with VxWorks
9937 @kindex target vxworks
9938 @item target vxworks @var{machinename}
9939 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
9940 is the target system's machine name or IP address.
9944 On VxWorks, @code{load} links @var{filename} dynamically on the
9945 current target system as well as adding its symbols in @value{GDBN}.
9947 @value{GDBN} enables developers to spawn and debug tasks running on networked
9948 VxWorks targets from a Unix host. Already-running tasks spawned from
9949 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
9950 both the Unix host and on the VxWorks target. The program
9951 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
9952 installed with the name @code{vxgdb}, to distinguish it from a
9953 @value{GDBN} for debugging programs on the host itself.)
9956 @item VxWorks-timeout @var{args}
9957 @kindex vxworks-timeout
9958 All VxWorks-based targets now support the option @code{vxworks-timeout}.
9959 This option is set by the user, and @var{args} represents the number of
9960 seconds @value{GDBN} waits for responses to rpc's. You might use this if
9961 your VxWorks target is a slow software simulator or is on the far side
9962 of a thin network line.
9965 The following information on connecting to VxWorks was current when
9966 this manual was produced; newer releases of VxWorks may use revised
9970 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
9971 to include the remote debugging interface routines in the VxWorks
9972 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
9973 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
9974 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
9975 source debugging task @code{tRdbTask} when VxWorks is booted. For more
9976 information on configuring and remaking VxWorks, see the manufacturer's
9978 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
9980 Once you have included @file{rdb.a} in your VxWorks system image and set
9981 your Unix execution search path to find @value{GDBN}, you are ready to
9982 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
9983 @code{vxgdb}, depending on your installation).
9985 @value{GDBN} comes up showing the prompt:
9992 * VxWorks Connection:: Connecting to VxWorks
9993 * VxWorks Download:: VxWorks download
9994 * VxWorks Attach:: Running tasks
9997 @node VxWorks Connection
9998 @subsubsection Connecting to VxWorks
10000 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
10001 network. To connect to a target whose host name is ``@code{tt}'', type:
10004 (vxgdb) target vxworks tt
10008 @value{GDBN} displays messages like these:
10011 Attaching remote machine across net...
10016 @value{GDBN} then attempts to read the symbol tables of any object modules
10017 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
10018 these files by searching the directories listed in the command search
10019 path (@pxref{Environment, ,Your program's environment}); if it fails
10020 to find an object file, it displays a message such as:
10023 prog.o: No such file or directory.
10026 When this happens, add the appropriate directory to the search path with
10027 the @value{GDBN} command @code{path}, and execute the @code{target}
10030 @node VxWorks Download
10031 @subsubsection VxWorks download
10033 @cindex download to VxWorks
10034 If you have connected to the VxWorks target and you want to debug an
10035 object that has not yet been loaded, you can use the @value{GDBN}
10036 @code{load} command to download a file from Unix to VxWorks
10037 incrementally. The object file given as an argument to the @code{load}
10038 command is actually opened twice: first by the VxWorks target in order
10039 to download the code, then by @value{GDBN} in order to read the symbol
10040 table. This can lead to problems if the current working directories on
10041 the two systems differ. If both systems have NFS mounted the same
10042 filesystems, you can avoid these problems by using absolute paths.
10043 Otherwise, it is simplest to set the working directory on both systems
10044 to the directory in which the object file resides, and then to reference
10045 the file by its name, without any path. For instance, a program
10046 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
10047 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
10048 program, type this on VxWorks:
10051 -> cd "@var{vxpath}/vw/demo/rdb"
10055 Then, in @value{GDBN}, type:
10058 (vxgdb) cd @var{hostpath}/vw/demo/rdb
10059 (vxgdb) load prog.o
10062 @value{GDBN} displays a response similar to this:
10065 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
10068 You can also use the @code{load} command to reload an object module
10069 after editing and recompiling the corresponding source file. Note that
10070 this makes @value{GDBN} delete all currently-defined breakpoints,
10071 auto-displays, and convenience variables, and to clear the value
10072 history. (This is necessary in order to preserve the integrity of
10073 debugger's data structures that reference the target system's symbol
10076 @node VxWorks Attach
10077 @subsubsection Running tasks
10079 @cindex running VxWorks tasks
10080 You can also attach to an existing task using the @code{attach} command as
10084 (vxgdb) attach @var{task}
10088 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
10089 or suspended when you attach to it. Running tasks are suspended at
10090 the time of attachment.
10092 @node Embedded Processors
10093 @section Embedded Processors
10095 This section goes into details specific to particular embedded
10099 * A29K Embedded:: AMD A29K Embedded
10101 * H8/300:: Hitachi H8/300
10102 * H8/500:: Hitachi H8/500
10103 * i960:: Intel i960
10104 * M32R/D:: Mitsubishi M32R/D
10105 * M68K:: Motorola M68K
10106 * M88K:: Motorola M88K
10107 * MIPS Embedded:: MIPS Embedded
10108 * PA:: HP PA Embedded
10111 * Sparclet:: Tsqware Sparclet
10112 * Sparclite:: Fujitsu Sparclite
10113 * ST2000:: Tandem ST2000
10114 * Z8000:: Zilog Z8000
10117 @node A29K Embedded
10118 @subsection AMD A29K Embedded
10123 * Comms (EB29K):: Communications setup
10124 * gdb-EB29K:: EB29K cross-debugging
10125 * Remote Log:: Remote log
10130 @kindex target adapt
10131 @item target adapt @var{dev}
10132 Adapt monitor for A29K.
10134 @kindex target amd-eb
10135 @item target amd-eb @var{dev} @var{speed} @var{PROG}
10137 Remote PC-resident AMD EB29K board, attached over serial lines.
10138 @var{dev} is the serial device, as for @code{target remote};
10139 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
10140 name of the program to be debugged, as it appears to DOS on the PC.
10141 @xref{A29K EB29K, ,EBMON protocol for AMD29K}.
10146 @subsubsection A29K UDI
10149 @cindex AMD29K via UDI
10151 @value{GDBN} supports AMD's UDI (``Universal Debugger Interface'')
10152 protocol for debugging the a29k processor family. To use this
10153 configuration with AMD targets running the MiniMON monitor, you need the
10154 program @code{MONTIP}, available from AMD at no charge. You can also
10155 use @value{GDBN} with the UDI-conformant a29k simulator program
10156 @code{ISSTIP}, also available from AMD.
10159 @item target udi @var{keyword}
10161 Select the UDI interface to a remote a29k board or simulator, where
10162 @var{keyword} is an entry in the AMD configuration file @file{udi_soc}.
10163 This file contains keyword entries which specify parameters used to
10164 connect to a29k targets. If the @file{udi_soc} file is not in your
10165 working directory, you must set the environment variable @samp{UDICONF}
10170 @subsubsection EBMON protocol for AMD29K
10172 @cindex EB29K board
10173 @cindex running 29K programs
10175 AMD distributes a 29K development board meant to fit in a PC, together
10176 with a DOS-hosted monitor program called @code{EBMON}. As a shorthand
10177 term, this development system is called the ``EB29K''. To use
10178 @value{GDBN} from a Unix system to run programs on the EB29K board, you
10179 must first connect a serial cable between the PC (which hosts the EB29K
10180 board) and a serial port on the Unix system. In the following, we
10181 assume you've hooked the cable between the PC's @file{COM1} port and
10182 @file{/dev/ttya} on the Unix system.
10184 @node Comms (EB29K)
10185 @subsubsection Communications setup
10187 The next step is to set up the PC's port, by doing something like this
10191 C:\> MODE com1:9600,n,8,1,none
10195 This example---run on an MS DOS 4.0 system---sets the PC port to 9600
10196 bps, no parity, eight data bits, one stop bit, and no ``retry'' action;
10197 you must match the communications parameters when establishing the Unix
10198 end of the connection as well.
10199 @c FIXME: Who knows what this "no retry action" crud from the DOS manual may
10200 @c mean? It's optional; leave it out? ---doc@cygnus.com, 25feb91
10202 @c It's optional, but it's unwise to omit it: who knows what is the
10203 @c default value set when the DOS machines boots? "No retry" means that
10204 @c the DOS serial device driver won't retry the operation if it fails;
10205 @c I understand that this is needed because the GDB serial protocol
10206 @c handles any errors and retransmissions itself. ---Eli Zaretskii, 3sep99
10208 To give control of the PC to the Unix side of the serial line, type
10209 the following at the DOS console:
10216 (Later, if you wish to return control to the DOS console, you can use
10217 the command @code{CTTY con}---but you must send it over the device that
10218 had control, in our example over the @file{COM1} serial line.)
10220 From the Unix host, use a communications program such as @code{tip} or
10221 @code{cu} to communicate with the PC; for example,
10224 cu -s 9600 -l /dev/ttya
10228 The @code{cu} options shown specify, respectively, the linespeed and the
10229 serial port to use. If you use @code{tip} instead, your command line
10230 may look something like the following:
10233 tip -9600 /dev/ttya
10237 Your system may require a different name where we show
10238 @file{/dev/ttya} as the argument to @code{tip}. The communications
10239 parameters, including which port to use, are associated with the
10240 @code{tip} argument in the ``remote'' descriptions file---normally the
10241 system table @file{/etc/remote}.
10242 @c FIXME: What if anything needs doing to match the "n,8,1,none" part of
10243 @c the DOS side's comms setup? cu can support -o (odd
10244 @c parity), -e (even parity)---apparently no settings for no parity or
10245 @c for character size. Taken from stty maybe...? John points out tip
10246 @c can set these as internal variables, eg ~s parity=none; man stty
10247 @c suggests that it *might* work to stty these options with stdin or
10248 @c stdout redirected... ---doc@cygnus.com, 25feb91
10250 @c There's nothing to be done for the "none" part of the DOS MODE
10251 @c command. The rest of the parameters should be matched by the
10252 @c baudrate, bits, and parity used by the Unix side. ---Eli Zaretskii, 3Sep99
10255 Using the @code{tip} or @code{cu} connection, change the DOS working
10256 directory to the directory containing a copy of your 29K program, then
10257 start the PC program @code{EBMON} (an EB29K control program supplied
10258 with your board by AMD). You should see an initial display from
10259 @code{EBMON} similar to the one that follows, ending with the
10260 @code{EBMON} prompt @samp{#}---
10265 G:\> CD \usr\joe\work29k
10267 G:\USR\JOE\WORK29K> EBMON
10268 Am29000 PC Coprocessor Board Monitor, version 3.0-18
10269 Copyright 1990 Advanced Micro Devices, Inc.
10270 Written by Gibbons and Associates, Inc.
10272 Enter '?' or 'H' for help
10274 PC Coprocessor Type = EB29K
10276 Memory Base = 0xd0000
10278 Data Memory Size = 2048KB
10279 Available I-RAM Range = 0x8000 to 0x1fffff
10280 Available D-RAM Range = 0x80002000 to 0x801fffff
10283 Register Stack Size = 0x800
10284 Memory Stack Size = 0x1800
10287 Am29027 Available = No
10288 Byte Write Available = Yes
10293 Then exit the @code{cu} or @code{tip} program (done in the example by
10294 typing @code{~.} at the @code{EBMON} prompt). @code{EBMON} keeps
10295 running, ready for @value{GDBN} to take over.
10297 For this example, we've assumed what is probably the most convenient
10298 way to make sure the same 29K program is on both the PC and the Unix
10299 system: a PC/NFS connection that establishes ``drive @file{G:}'' on the
10300 PC as a file system on the Unix host. If you do not have PC/NFS or
10301 something similar connecting the two systems, you must arrange some
10302 other way---perhaps floppy-disk transfer---of getting the 29K program
10303 from the Unix system to the PC; @value{GDBN} does @emph{not} download it over the
10307 @subsubsection EB29K cross-debugging
10309 Finally, @code{cd} to the directory containing an image of your 29K
10310 program on the Unix system, and start @value{GDBN}---specifying as argument the
10311 name of your 29K program:
10314 cd /usr/joe/work29k
10319 Now you can use the @code{target} command:
10322 target amd-eb /dev/ttya 9600 MYFOO
10323 @c FIXME: test above 'target amd-eb' as spelled, with caps! caps are meant to
10324 @c emphasize that this is the name as seen by DOS (since I think DOS is
10325 @c single-minded about case of letters). ---doc@cygnus.com, 25feb91
10329 In this example, we've assumed your program is in a file called
10330 @file{myfoo}. Note that the filename given as the last argument to
10331 @code{target amd-eb} should be the name of the program as it appears to DOS.
10332 In our example this is simply @code{MYFOO}, but in general it can include
10333 a DOS path, and depending on your transfer mechanism may not resemble
10334 the name on the Unix side.
10336 At this point, you can set any breakpoints you wish; when you are ready
10337 to see your program run on the 29K board, use the @value{GDBN} command
10340 To stop debugging the remote program, use the @value{GDBN} @code{detach}
10343 To return control of the PC to its console, use @code{tip} or @code{cu}
10344 once again, after your @value{GDBN} session has concluded, to attach to
10345 @code{EBMON}. You can then type the command @code{q} to shut down
10346 @code{EBMON}, returning control to the DOS command-line interpreter.
10347 Type @kbd{CTTY con} to return command input to the main DOS console,
10348 and type @kbd{~.} to leave @code{tip} or @code{cu}.
10351 @subsubsection Remote log
10352 @cindex @file{eb.log}, a log file for EB29K
10353 @cindex log file for EB29K
10355 The @code{target amd-eb} command creates a file @file{eb.log} in the
10356 current working directory, to help debug problems with the connection.
10357 @file{eb.log} records all the output from @code{EBMON}, including echoes
10358 of the commands sent to it. Running @samp{tail -f} on this file in
10359 another window often helps to understand trouble with @code{EBMON}, or
10360 unexpected events on the PC side of the connection.
10368 @item target rdi @var{dev}
10369 ARM Angel monitor, via RDI library interface to ADP protocol. You may
10370 use this target to communicate with both boards running the Angel
10371 monitor, or with the EmbeddedICE JTAG debug device.
10374 @item target rdp @var{dev}
10380 @subsection Hitachi H8/300
10384 @kindex target hms@r{, with H8/300}
10385 @item target hms @var{dev}
10386 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
10387 Use special commands @code{device} and @code{speed} to control the serial
10388 line and the communications speed used.
10390 @kindex target e7000@r{, with H8/300}
10391 @item target e7000 @var{dev}
10392 E7000 emulator for Hitachi H8 and SH.
10394 @kindex target sh3@r{, with H8/300}
10395 @kindex target sh3e@r{, with H8/300}
10396 @item target sh3 @var{dev}
10397 @itemx target sh3e @var{dev}
10398 Hitachi SH-3 and SH-3E target systems.
10402 @cindex download to H8/300 or H8/500
10403 @cindex H8/300 or H8/500 download
10404 @cindex download to Hitachi SH
10405 @cindex Hitachi SH download
10406 When you select remote debugging to a Hitachi SH, H8/300, or H8/500
10407 board, the @code{load} command downloads your program to the Hitachi
10408 board and also opens it as the current executable target for
10409 @value{GDBN} on your host (like the @code{file} command).
10411 @value{GDBN} needs to know these things to talk to your
10412 Hitachi SH, H8/300, or H8/500:
10416 that you want to use @samp{target hms}, the remote debugging interface
10417 for Hitachi microprocessors, or @samp{target e7000}, the in-circuit
10418 emulator for the Hitachi SH and the Hitachi 300H. (@samp{target hms} is
10419 the default when @value{GDBN} is configured specifically for the Hitachi SH,
10420 H8/300, or H8/500.)
10423 what serial device connects your host to your Hitachi board (the first
10424 serial device available on your host is the default).
10427 what speed to use over the serial device.
10431 * Hitachi Boards:: Connecting to Hitachi boards.
10432 * Hitachi ICE:: Using the E7000 In-Circuit Emulator.
10433 * Hitachi Special:: Special @value{GDBN} commands for Hitachi micros.
10436 @node Hitachi Boards
10437 @subsubsection Connecting to Hitachi boards
10439 @c only for Unix hosts
10441 @cindex serial device, Hitachi micros
10442 Use the special @code{@value{GDBN}} command @samp{device @var{port}} if you
10443 need to explicitly set the serial device. The default @var{port} is the
10444 first available port on your host. This is only necessary on Unix
10445 hosts, where it is typically something like @file{/dev/ttya}.
10448 @cindex serial line speed, Hitachi micros
10449 @code{@value{GDBN}} has another special command to set the communications
10450 speed: @samp{speed @var{bps}}. This command also is only used from Unix
10451 hosts; on DOS hosts, set the line speed as usual from outside @value{GDBN} with
10452 the DOS @code{mode} command (for instance,
10453 @w{@kbd{mode com2:9600,n,8,1,p}} for a 9600@dmn{bps} connection).
10455 The @samp{device} and @samp{speed} commands are available only when you
10456 use a Unix host to debug your Hitachi microprocessor programs. If you
10458 @value{GDBN} depends on an auxiliary terminate-and-stay-resident program
10459 called @code{asynctsr} to communicate with the development board
10460 through a PC serial port. You must also use the DOS @code{mode} command
10461 to set up the serial port on the DOS side.
10463 The following sample session illustrates the steps needed to start a
10464 program under @value{GDBN} control on an H8/300. The example uses a
10465 sample H8/300 program called @file{t.x}. The procedure is the same for
10466 the Hitachi SH and the H8/500.
10468 First hook up your development board. In this example, we use a
10469 board attached to serial port @code{COM2}; if you use a different serial
10470 port, substitute its name in the argument of the @code{mode} command.
10471 When you call @code{asynctsr}, the auxiliary comms program used by the
10472 debugger, you give it just the numeric part of the serial port's name;
10473 for example, @samp{asyncstr 2} below runs @code{asyncstr} on
10477 C:\H8300\TEST> asynctsr 2
10478 C:\H8300\TEST> mode com2:9600,n,8,1,p
10480 Resident portion of MODE loaded
10482 COM2: 9600, n, 8, 1, p
10487 @emph{Warning:} We have noticed a bug in PC-NFS that conflicts with
10488 @code{asynctsr}. If you also run PC-NFS on your DOS host, you may need to
10489 disable it, or even boot without it, to use @code{asynctsr} to control
10490 your development board.
10493 @kindex target hms@r{, and serial protocol}
10494 Now that serial communications are set up, and the development board is
10495 connected, you can start up @value{GDBN}. Call @code{@value{GDBP}} with
10496 the name of your program as the argument. @code{@value{GDBN}} prompts
10497 you, as usual, with the prompt @samp{(@value{GDBP})}. Use two special
10498 commands to begin your debugging session: @samp{target hms} to specify
10499 cross-debugging to the Hitachi board, and the @code{load} command to
10500 download your program to the board. @code{load} displays the names of
10501 the program's sections, and a @samp{*} for each 2K of data downloaded.
10502 (If you want to refresh @value{GDBN} data on symbols or on the
10503 executable file without downloading, use the @value{GDBN} commands
10504 @code{file} or @code{symbol-file}. These commands, and @code{load}
10505 itself, are described in @ref{Files,,Commands to specify files}.)
10508 (eg-C:\H8300\TEST) @value{GDBP} t.x
10509 @value{GDBN} is free software and you are welcome to distribute copies
10510 of it under certain conditions; type "show copying" to see
10512 There is absolutely no warranty for @value{GDBN}; type "show warranty"
10514 @value{GDBN} @value{GDBVN}, Copyright 1992 Free Software Foundation, Inc...
10515 (@value{GDBP}) target hms
10516 Connected to remote H8/300 HMS system.
10517 (@value{GDBP}) load t.x
10518 .text : 0x8000 .. 0xabde ***********
10519 .data : 0xabde .. 0xad30 *
10520 .stack : 0xf000 .. 0xf014 *
10523 At this point, you're ready to run or debug your program. From here on,
10524 you can use all the usual @value{GDBN} commands. The @code{break} command
10525 sets breakpoints; the @code{run} command starts your program;
10526 @code{print} or @code{x} display data; the @code{continue} command
10527 resumes execution after stopping at a breakpoint. You can use the
10528 @code{help} command at any time to find out more about @value{GDBN} commands.
10530 Remember, however, that @emph{operating system} facilities aren't
10531 available on your development board; for example, if your program hangs,
10532 you can't send an interrupt---but you can press the @sc{reset} switch!
10534 Use the @sc{reset} button on the development board
10537 to interrupt your program (don't use @kbd{ctl-C} on the DOS host---it has
10538 no way to pass an interrupt signal to the development board); and
10541 to return to the @value{GDBN} command prompt after your program finishes
10542 normally. The communications protocol provides no other way for @value{GDBN}
10543 to detect program completion.
10546 In either case, @value{GDBN} sees the effect of a @sc{reset} on the
10547 development board as a ``normal exit'' of your program.
10550 @subsubsection Using the E7000 in-circuit emulator
10552 @kindex target e7000@r{, with Hitachi ICE}
10553 You can use the E7000 in-circuit emulator to develop code for either the
10554 Hitachi SH or the H8/300H. Use one of these forms of the @samp{target
10555 e7000} command to connect @value{GDBN} to your E7000:
10558 @item target e7000 @var{port} @var{speed}
10559 Use this form if your E7000 is connected to a serial port. The
10560 @var{port} argument identifies what serial port to use (for example,
10561 @samp{com2}). The third argument is the line speed in bits per second
10562 (for example, @samp{9600}).
10564 @item target e7000 @var{hostname}
10565 If your E7000 is installed as a host on a TCP/IP network, you can just
10566 specify its hostname; @value{GDBN} uses @code{telnet} to connect.
10569 @node Hitachi Special
10570 @subsubsection Special @value{GDBN} commands for Hitachi micros
10572 Some @value{GDBN} commands are available only for the H8/300:
10576 @kindex set machine
10577 @kindex show machine
10578 @item set machine h8300
10579 @itemx set machine h8300h
10580 Condition @value{GDBN} for one of the two variants of the H8/300
10581 architecture with @samp{set machine}. You can use @samp{show machine}
10582 to check which variant is currently in effect.
10591 @kindex set memory @var{mod}
10592 @cindex memory models, H8/500
10593 @item set memory @var{mod}
10595 Specify which H8/500 memory model (@var{mod}) you are using with
10596 @samp{set memory}; check which memory model is in effect with @samp{show
10597 memory}. The accepted values for @var{mod} are @code{small},
10598 @code{big}, @code{medium}, and @code{compact}.
10603 @subsection Intel i960
10607 @kindex target mon960
10608 @item target mon960 @var{dev}
10609 MON960 monitor for Intel i960.
10611 @kindex target nindy
10612 @item target nindy @var{devicename}
10613 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
10614 the name of the serial device to use for the connection, e.g.
10621 @dfn{Nindy} is a ROM Monitor program for Intel 960 target systems. When
10622 @value{GDBN} is configured to control a remote Intel 960 using Nindy, you can
10623 tell @value{GDBN} how to connect to the 960 in several ways:
10627 Through command line options specifying serial port, version of the
10628 Nindy protocol, and communications speed;
10631 By responding to a prompt on startup;
10634 By using the @code{target} command at any point during your @value{GDBN}
10635 session. @xref{Target Commands, ,Commands for managing targets}.
10639 @cindex download to Nindy-960
10640 With the Nindy interface to an Intel 960 board, @code{load}
10641 downloads @var{filename} to the 960 as well as adding its symbols in
10645 * Nindy Startup:: Startup with Nindy
10646 * Nindy Options:: Options for Nindy
10647 * Nindy Reset:: Nindy reset command
10650 @node Nindy Startup
10651 @subsubsection Startup with Nindy
10653 If you simply start @code{@value{GDBP}} without using any command-line
10654 options, you are prompted for what serial port to use, @emph{before} you
10655 reach the ordinary @value{GDBN} prompt:
10658 Attach /dev/ttyNN -- specify NN, or "quit" to quit:
10662 Respond to the prompt with whatever suffix (after @samp{/dev/tty})
10663 identifies the serial port you want to use. You can, if you choose,
10664 simply start up with no Nindy connection by responding to the prompt
10665 with an empty line. If you do this and later wish to attach to Nindy,
10666 use @code{target} (@pxref{Target Commands, ,Commands for managing targets}).
10668 @node Nindy Options
10669 @subsubsection Options for Nindy
10671 These are the startup options for beginning your @value{GDBN} session with a
10672 Nindy-960 board attached:
10675 @item -r @var{port}
10676 Specify the serial port name of a serial interface to be used to connect
10677 to the target system. This option is only available when @value{GDBN} is
10678 configured for the Intel 960 target architecture. You may specify
10679 @var{port} as any of: a full pathname (e.g. @samp{-r /dev/ttya}), a
10680 device name in @file{/dev} (e.g. @samp{-r ttya}), or simply the unique
10681 suffix for a specific @code{tty} (e.g. @samp{-r a}).
10684 (An uppercase letter ``O'', not a zero.) Specify that @value{GDBN} should use
10685 the ``old'' Nindy monitor protocol to connect to the target system.
10686 This option is only available when @value{GDBN} is configured for the Intel 960
10687 target architecture.
10690 @emph{Warning:} if you specify @samp{-O}, but are actually trying to
10691 connect to a target system that expects the newer protocol, the connection
10692 fails, appearing to be a speed mismatch. @value{GDBN} repeatedly
10693 attempts to reconnect at several different line speeds. You can abort
10694 this process with an interrupt.
10698 Specify that @value{GDBN} should first send a @code{BREAK} signal to the target
10699 system, in an attempt to reset it, before connecting to a Nindy target.
10702 @emph{Warning:} Many target systems do not have the hardware that this
10703 requires; it only works with a few boards.
10707 The standard @samp{-b} option controls the line speed used on the serial
10712 @subsubsection Nindy reset command
10717 For a Nindy target, this command sends a ``break'' to the remote target
10718 system; this is only useful if the target has been equipped with a
10719 circuit to perform a hard reset (or some other interesting action) when
10720 a break is detected.
10725 @subsection Mitsubishi M32R/D
10729 @kindex target m32r
10730 @item target m32r @var{dev}
10731 Mitsubishi M32R/D ROM monitor.
10738 The Motorola m68k configuration includes ColdFire support, and
10739 target command for the following ROM monitors.
10743 @kindex target abug
10744 @item target abug @var{dev}
10745 ABug ROM monitor for M68K.
10747 @kindex target cpu32bug
10748 @item target cpu32bug @var{dev}
10749 CPU32BUG monitor, running on a CPU32 (M68K) board.
10751 @kindex target dbug
10752 @item target dbug @var{dev}
10753 dBUG ROM monitor for Motorola ColdFire.
10756 @item target est @var{dev}
10757 EST-300 ICE monitor, running on a CPU32 (M68K) board.
10759 @kindex target rom68k
10760 @item target rom68k @var{dev}
10761 ROM 68K monitor, running on an M68K IDP board.
10765 If @value{GDBN} is configured with @code{m68*-ericsson-*}, it will
10766 instead have only a single special target command:
10770 @kindex target es1800
10771 @item target es1800 @var{dev}
10772 ES-1800 emulator for M68K.
10780 @kindex target rombug
10781 @item target rombug @var{dev}
10782 ROMBUG ROM monitor for OS/9000.
10792 @item target bug @var{dev}
10793 BUG monitor, running on a MVME187 (m88k) board.
10797 @node MIPS Embedded
10798 @subsection MIPS Embedded
10800 @cindex MIPS boards
10801 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
10802 MIPS board attached to a serial line. This is available when
10803 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
10806 Use these @value{GDBN} commands to specify the connection to your target board:
10809 @item target mips @var{port}
10810 @kindex target mips @var{port}
10811 To run a program on the board, start up @code{@value{GDBP}} with the
10812 name of your program as the argument. To connect to the board, use the
10813 command @samp{target mips @var{port}}, where @var{port} is the name of
10814 the serial port connected to the board. If the program has not already
10815 been downloaded to the board, you may use the @code{load} command to
10816 download it. You can then use all the usual @value{GDBN} commands.
10818 For example, this sequence connects to the target board through a serial
10819 port, and loads and runs a program called @var{prog} through the
10823 host$ @value{GDBP} @var{prog}
10824 @value{GDBN} is free software and @dots{}
10825 (@value{GDBP}) target mips /dev/ttyb
10826 (@value{GDBP}) load @var{prog}
10830 @item target mips @var{hostname}:@var{portnumber}
10831 On some @value{GDBN} host configurations, you can specify a TCP
10832 connection (for instance, to a serial line managed by a terminal
10833 concentrator) instead of a serial port, using the syntax
10834 @samp{@var{hostname}:@var{portnumber}}.
10836 @item target pmon @var{port}
10837 @kindex target pmon @var{port}
10840 @item target ddb @var{port}
10841 @kindex target ddb @var{port}
10842 NEC's DDB variant of PMON for Vr4300.
10844 @item target lsi @var{port}
10845 @kindex target lsi @var{port}
10846 LSI variant of PMON.
10848 @kindex target r3900
10849 @item target r3900 @var{dev}
10850 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
10852 @kindex target array
10853 @item target array @var{dev}
10854 Array Tech LSI33K RAID controller board.
10860 @value{GDBN} also supports these special commands for MIPS targets:
10863 @item set processor @var{args}
10864 @itemx show processor
10865 @kindex set processor @var{args}
10866 @kindex show processor
10867 Use the @code{set processor} command to set the type of MIPS
10868 processor when you want to access processor-type-specific registers.
10869 For example, @code{set processor @var{r3041}} tells @value{GDBN}
10870 to use the CPO registers appropriate for the 3041 chip.
10871 Use the @code{show processor} command to see what MIPS processor @value{GDBN}
10872 is using. Use the @code{info reg} command to see what registers
10873 @value{GDBN} is using.
10875 @item set mipsfpu double
10876 @itemx set mipsfpu single
10877 @itemx set mipsfpu none
10878 @itemx show mipsfpu
10879 @kindex set mipsfpu
10880 @kindex show mipsfpu
10881 @cindex MIPS remote floating point
10882 @cindex floating point, MIPS remote
10883 If your target board does not support the MIPS floating point
10884 coprocessor, you should use the command @samp{set mipsfpu none} (if you
10885 need this, you may wish to put the command in your @value{GDBN} init
10886 file). This tells @value{GDBN} how to find the return value of
10887 functions which return floating point values. It also allows
10888 @value{GDBN} to avoid saving the floating point registers when calling
10889 functions on the board. If you are using a floating point coprocessor
10890 with only single precision floating point support, as on the @sc{r4650}
10891 processor, use the command @samp{set mipsfpu single}. The default
10892 double precision floating point coprocessor may be selected using
10893 @samp{set mipsfpu double}.
10895 In previous versions the only choices were double precision or no
10896 floating point, so @samp{set mipsfpu on} will select double precision
10897 and @samp{set mipsfpu off} will select no floating point.
10899 As usual, you can inquire about the @code{mipsfpu} variable with
10900 @samp{show mipsfpu}.
10902 @item set remotedebug @var{n}
10903 @itemx show remotedebug
10904 @kindex set remotedebug@r{, MIPS protocol}
10905 @kindex show remotedebug@r{, MIPS protocol}
10906 @cindex @code{remotedebug}, MIPS protocol
10907 @cindex MIPS @code{remotedebug} protocol
10908 @c FIXME! For this to be useful, you must know something about the MIPS
10909 @c FIXME...protocol. Where is it described?
10910 You can see some debugging information about communications with the board
10911 by setting the @code{remotedebug} variable. If you set it to @code{1} using
10912 @samp{set remotedebug 1}, every packet is displayed. If you set it
10913 to @code{2}, every character is displayed. You can check the current value
10914 at any time with the command @samp{show remotedebug}.
10916 @item set timeout @var{seconds}
10917 @itemx set retransmit-timeout @var{seconds}
10918 @itemx show timeout
10919 @itemx show retransmit-timeout
10920 @cindex @code{timeout}, MIPS protocol
10921 @cindex @code{retransmit-timeout}, MIPS protocol
10922 @kindex set timeout
10923 @kindex show timeout
10924 @kindex set retransmit-timeout
10925 @kindex show retransmit-timeout
10926 You can control the timeout used while waiting for a packet, in the MIPS
10927 remote protocol, with the @code{set timeout @var{seconds}} command. The
10928 default is 5 seconds. Similarly, you can control the timeout used while
10929 waiting for an acknowledgement of a packet with the @code{set
10930 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
10931 You can inspect both values with @code{show timeout} and @code{show
10932 retransmit-timeout}. (These commands are @emph{only} available when
10933 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
10935 The timeout set by @code{set timeout} does not apply when @value{GDBN}
10936 is waiting for your program to stop. In that case, @value{GDBN} waits
10937 forever because it has no way of knowing how long the program is going
10938 to run before stopping.
10942 @subsection PowerPC
10946 @kindex target dink32
10947 @item target dink32 @var{dev}
10948 DINK32 ROM monitor.
10950 @kindex target ppcbug
10951 @item target ppcbug @var{dev}
10952 @kindex target ppcbug1
10953 @item target ppcbug1 @var{dev}
10954 PPCBUG ROM monitor for PowerPC.
10957 @item target sds @var{dev}
10958 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
10963 @subsection HP PA Embedded
10967 @kindex target op50n
10968 @item target op50n @var{dev}
10969 OP50N monitor, running on an OKI HPPA board.
10971 @kindex target w89k
10972 @item target w89k @var{dev}
10973 W89K monitor, running on a Winbond HPPA board.
10978 @subsection Hitachi SH
10982 @kindex target hms@r{, with Hitachi SH}
10983 @item target hms @var{dev}
10984 A Hitachi SH board attached via serial line to your host. Use special
10985 commands @code{device} and @code{speed} to control the serial line and
10986 the communications speed used.
10988 @kindex target e7000@r{, with Hitachi SH}
10989 @item target e7000 @var{dev}
10990 E7000 emulator for Hitachi SH.
10992 @kindex target sh3@r{, with SH}
10993 @kindex target sh3e@r{, with SH}
10994 @item target sh3 @var{dev}
10995 @item target sh3e @var{dev}
10996 Hitachi SH-3 and SH-3E target systems.
11001 @subsection Tsqware Sparclet
11005 @value{GDBN} enables developers to debug tasks running on
11006 Sparclet targets from a Unix host.
11007 @value{GDBN} uses code that runs on
11008 both the Unix host and on the Sparclet target. The program
11009 @code{@value{GDBP}} is installed and executed on the Unix host.
11012 @item remotetimeout @var{args}
11013 @kindex remotetimeout
11014 @value{GDBN} supports the option @code{remotetimeout}.
11015 This option is set by the user, and @var{args} represents the number of
11016 seconds @value{GDBN} waits for responses.
11019 @cindex compiling, on Sparclet
11020 When compiling for debugging, include the options @samp{-g} to get debug
11021 information and @samp{-Ttext} to relocate the program to where you wish to
11022 load it on the target. You may also want to add the options @samp{-n} or
11023 @samp{-N} in order to reduce the size of the sections. Example:
11026 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
11029 You can use @code{objdump} to verify that the addresses are what you intended:
11032 sparclet-aout-objdump --headers --syms prog
11035 @cindex running, on Sparclet
11037 your Unix execution search path to find @value{GDBN}, you are ready to
11038 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
11039 (or @code{sparclet-aout-gdb}, depending on your installation).
11041 @value{GDBN} comes up showing the prompt:
11048 * Sparclet File:: Setting the file to debug
11049 * Sparclet Connection:: Connecting to Sparclet
11050 * Sparclet Download:: Sparclet download
11051 * Sparclet Execution:: Running and debugging
11054 @node Sparclet File
11055 @subsubsection Setting file to debug
11057 The @value{GDBN} command @code{file} lets you choose with program to debug.
11060 (gdbslet) file prog
11064 @value{GDBN} then attempts to read the symbol table of @file{prog}.
11065 @value{GDBN} locates
11066 the file by searching the directories listed in the command search
11068 If the file was compiled with debug information (option "-g"), source
11069 files will be searched as well.
11070 @value{GDBN} locates
11071 the source files by searching the directories listed in the directory search
11072 path (@pxref{Environment, ,Your program's environment}).
11074 to find a file, it displays a message such as:
11077 prog: No such file or directory.
11080 When this happens, add the appropriate directories to the search paths with
11081 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
11082 @code{target} command again.
11084 @node Sparclet Connection
11085 @subsubsection Connecting to Sparclet
11087 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
11088 To connect to a target on serial port ``@code{ttya}'', type:
11091 (gdbslet) target sparclet /dev/ttya
11092 Remote target sparclet connected to /dev/ttya
11093 main () at ../prog.c:3
11097 @value{GDBN} displays messages like these:
11103 @node Sparclet Download
11104 @subsubsection Sparclet download
11106 @cindex download to Sparclet
11107 Once connected to the Sparclet target,
11108 you can use the @value{GDBN}
11109 @code{load} command to download the file from the host to the target.
11110 The file name and load offset should be given as arguments to the @code{load}
11112 Since the file format is aout, the program must be loaded to the starting
11113 address. You can use @code{objdump} to find out what this value is. The load
11114 offset is an offset which is added to the VMA (virtual memory address)
11115 of each of the file's sections.
11116 For instance, if the program
11117 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
11118 and bss at 0x12010170, in @value{GDBN}, type:
11121 (gdbslet) load prog 0x12010000
11122 Loading section .text, size 0xdb0 vma 0x12010000
11125 If the code is loaded at a different address then what the program was linked
11126 to, you may need to use the @code{section} and @code{add-symbol-file} commands
11127 to tell @value{GDBN} where to map the symbol table.
11129 @node Sparclet Execution
11130 @subsubsection Running and debugging
11132 @cindex running and debugging Sparclet programs
11133 You can now begin debugging the task using @value{GDBN}'s execution control
11134 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
11135 manual for the list of commands.
11139 Breakpoint 1 at 0x12010000: file prog.c, line 3.
11141 Starting program: prog
11142 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
11143 3 char *symarg = 0;
11145 4 char *execarg = "hello!";
11150 @subsection Fujitsu Sparclite
11154 @kindex target sparclite
11155 @item target sparclite @var{dev}
11156 Fujitsu sparclite boards, used only for the purpose of loading.
11157 You must use an additional command to debug the program.
11158 For example: target remote @var{dev} using @value{GDBN} standard
11164 @subsection Tandem ST2000
11166 @value{GDBN} may be used with a Tandem ST2000 phone switch, running Tandem's
11169 To connect your ST2000 to the host system, see the manufacturer's
11170 manual. Once the ST2000 is physically attached, you can run:
11173 target st2000 @var{dev} @var{speed}
11177 to establish it as your debugging environment. @var{dev} is normally
11178 the name of a serial device, such as @file{/dev/ttya}, connected to the
11179 ST2000 via a serial line. You can instead specify @var{dev} as a TCP
11180 connection (for example, to a serial line attached via a terminal
11181 concentrator) using the syntax @code{@var{hostname}:@var{portnumber}}.
11183 The @code{load} and @code{attach} commands are @emph{not} defined for
11184 this target; you must load your program into the ST2000 as you normally
11185 would for standalone operation. @value{GDBN} reads debugging information
11186 (such as symbols) from a separate, debugging version of the program
11187 available on your host computer.
11188 @c FIXME!! This is terribly vague; what little content is here is
11189 @c basically hearsay.
11191 @cindex ST2000 auxiliary commands
11192 These auxiliary @value{GDBN} commands are available to help you with the ST2000
11196 @item st2000 @var{command}
11197 @kindex st2000 @var{cmd}
11198 @cindex STDBUG commands (ST2000)
11199 @cindex commands to STDBUG (ST2000)
11200 Send a @var{command} to the STDBUG monitor. See the manufacturer's
11201 manual for available commands.
11204 @cindex connect (to STDBUG)
11205 Connect the controlling terminal to the STDBUG command monitor. When
11206 you are done interacting with STDBUG, typing either of two character
11207 sequences gets you back to the @value{GDBN} command prompt:
11208 @kbd{@key{RET}~.} (Return, followed by tilde and period) or
11209 @kbd{@key{RET}~@key{C-d}} (Return, followed by tilde and control-D).
11213 @subsection Zilog Z8000
11216 @cindex simulator, Z8000
11217 @cindex Zilog Z8000 simulator
11219 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
11222 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
11223 unsegmented variant of the Z8000 architecture) or the Z8001 (the
11224 segmented variant). The simulator recognizes which architecture is
11225 appropriate by inspecting the object code.
11228 @item target sim @var{args}
11230 @kindex target sim@r{, with Z8000}
11231 Debug programs on a simulated CPU. If the simulator supports setup
11232 options, specify them via @var{args}.
11236 After specifying this target, you can debug programs for the simulated
11237 CPU in the same style as programs for your host computer; use the
11238 @code{file} command to load a new program image, the @code{run} command
11239 to run your program, and so on.
11241 As well as making available all the usual machine registers
11242 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
11243 additional items of information as specially named registers:
11248 Counts clock-ticks in the simulator.
11251 Counts instructions run in the simulator.
11254 Execution time in 60ths of a second.
11258 You can refer to these values in @value{GDBN} expressions with the usual
11259 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
11260 conditional breakpoint that suspends only after at least 5000
11261 simulated clock ticks.
11263 @node Architectures
11264 @section Architectures
11266 This section describes characteristics of architectures that affect
11267 all uses of @value{GDBN} with the architecture, both native and cross.
11280 @kindex set rstack_high_address
11281 @cindex AMD 29K register stack
11282 @cindex register stack, AMD29K
11283 @item set rstack_high_address @var{address}
11284 On AMD 29000 family processors, registers are saved in a separate
11285 @dfn{register stack}. There is no way for @value{GDBN} to determine the
11286 extent of this stack. Normally, @value{GDBN} just assumes that the
11287 stack is ``large enough''. This may result in @value{GDBN} referencing
11288 memory locations that do not exist. If necessary, you can get around
11289 this problem by specifying the ending address of the register stack with
11290 the @code{set rstack_high_address} command. The argument should be an
11291 address, which you probably want to precede with @samp{0x} to specify in
11294 @kindex show rstack_high_address
11295 @item show rstack_high_address
11296 Display the current limit of the register stack, on AMD 29000 family
11304 See the following section.
11309 @cindex stack on Alpha
11310 @cindex stack on MIPS
11311 @cindex Alpha stack
11313 Alpha- and MIPS-based computers use an unusual stack frame, which
11314 sometimes requires @value{GDBN} to search backward in the object code to
11315 find the beginning of a function.
11317 @cindex response time, MIPS debugging
11318 To improve response time (especially for embedded applications, where
11319 @value{GDBN} may be restricted to a slow serial line for this search)
11320 you may want to limit the size of this search, using one of these
11324 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
11325 @item set heuristic-fence-post @var{limit}
11326 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
11327 search for the beginning of a function. A value of @var{0} (the
11328 default) means there is no limit. However, except for @var{0}, the
11329 larger the limit the more bytes @code{heuristic-fence-post} must search
11330 and therefore the longer it takes to run.
11332 @item show heuristic-fence-post
11333 Display the current limit.
11337 These commands are available @emph{only} when @value{GDBN} is configured
11338 for debugging programs on Alpha or MIPS processors.
11341 @node Controlling GDB
11342 @chapter Controlling @value{GDBN}
11344 You can alter the way @value{GDBN} interacts with you by using the
11345 @code{set} command. For commands controlling how @value{GDBN} displays
11346 data, see @ref{Print Settings, ,Print settings}. Other settings are
11351 * Editing:: Command editing
11352 * History:: Command history
11353 * Screen Size:: Screen size
11354 * Numbers:: Numbers
11355 * Messages/Warnings:: Optional warnings and messages
11356 * Debugging Output:: Optional messages about internal happenings
11364 @value{GDBN} indicates its readiness to read a command by printing a string
11365 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
11366 can change the prompt string with the @code{set prompt} command. For
11367 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
11368 the prompt in one of the @value{GDBN} sessions so that you can always tell
11369 which one you are talking to.
11371 @emph{Note:} @code{set prompt} does not add a space for you after the
11372 prompt you set. This allows you to set a prompt which ends in a space
11373 or a prompt that does not.
11377 @item set prompt @var{newprompt}
11378 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
11380 @kindex show prompt
11382 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
11386 @section Command editing
11388 @cindex command line editing
11390 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
11391 @sc{gnu} library provides consistent behavior for programs which provide a
11392 command line interface to the user. Advantages are @sc{gnu} Emacs-style
11393 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
11394 substitution, and a storage and recall of command history across
11395 debugging sessions.
11397 You may control the behavior of command line editing in @value{GDBN} with the
11398 command @code{set}.
11401 @kindex set editing
11404 @itemx set editing on
11405 Enable command line editing (enabled by default).
11407 @item set editing off
11408 Disable command line editing.
11410 @kindex show editing
11412 Show whether command line editing is enabled.
11416 @section Command history
11418 @value{GDBN} can keep track of the commands you type during your
11419 debugging sessions, so that you can be certain of precisely what
11420 happened. Use these commands to manage the @value{GDBN} command
11424 @cindex history substitution
11425 @cindex history file
11426 @kindex set history filename
11427 @kindex GDBHISTFILE
11428 @item set history filename @var{fname}
11429 Set the name of the @value{GDBN} command history file to @var{fname}.
11430 This is the file where @value{GDBN} reads an initial command history
11431 list, and where it writes the command history from this session when it
11432 exits. You can access this list through history expansion or through
11433 the history command editing characters listed below. This file defaults
11434 to the value of the environment variable @code{GDBHISTFILE}, or to
11435 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
11438 @cindex history save
11439 @kindex set history save
11440 @item set history save
11441 @itemx set history save on
11442 Record command history in a file, whose name may be specified with the
11443 @code{set history filename} command. By default, this option is disabled.
11445 @item set history save off
11446 Stop recording command history in a file.
11448 @cindex history size
11449 @kindex set history size
11450 @item set history size @var{size}
11451 Set the number of commands which @value{GDBN} keeps in its history list.
11452 This defaults to the value of the environment variable
11453 @code{HISTSIZE}, or to 256 if this variable is not set.
11456 @cindex history expansion
11457 History expansion assigns special meaning to the character @kbd{!}.
11458 @ifset have-readline-appendices
11459 @xref{Event Designators}.
11462 Since @kbd{!} is also the logical not operator in C, history expansion
11463 is off by default. If you decide to enable history expansion with the
11464 @code{set history expansion on} command, you may sometimes need to
11465 follow @kbd{!} (when it is used as logical not, in an expression) with
11466 a space or a tab to prevent it from being expanded. The readline
11467 history facilities do not attempt substitution on the strings
11468 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
11470 The commands to control history expansion are:
11473 @kindex set history expansion
11474 @item set history expansion on
11475 @itemx set history expansion
11476 Enable history expansion. History expansion is off by default.
11478 @item set history expansion off
11479 Disable history expansion.
11481 The readline code comes with more complete documentation of
11482 editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs
11483 or @code{vi} may wish to read it.
11484 @ifset have-readline-appendices
11485 @xref{Command Line Editing}.
11489 @kindex show history
11491 @itemx show history filename
11492 @itemx show history save
11493 @itemx show history size
11494 @itemx show history expansion
11495 These commands display the state of the @value{GDBN} history parameters.
11496 @code{show history} by itself displays all four states.
11502 @item show commands
11503 Display the last ten commands in the command history.
11505 @item show commands @var{n}
11506 Print ten commands centered on command number @var{n}.
11508 @item show commands +
11509 Print ten commands just after the commands last printed.
11513 @section Screen size
11514 @cindex size of screen
11515 @cindex pauses in output
11517 Certain commands to @value{GDBN} may produce large amounts of
11518 information output to the screen. To help you read all of it,
11519 @value{GDBN} pauses and asks you for input at the end of each page of
11520 output. Type @key{RET} when you want to continue the output, or @kbd{q}
11521 to discard the remaining output. Also, the screen width setting
11522 determines when to wrap lines of output. Depending on what is being
11523 printed, @value{GDBN} tries to break the line at a readable place,
11524 rather than simply letting it overflow onto the following line.
11526 Normally @value{GDBN} knows the size of the screen from the terminal
11527 driver software. For example, on Unix @value{GDBN} uses the termcap data base
11528 together with the value of the @code{TERM} environment variable and the
11529 @code{stty rows} and @code{stty cols} settings. If this is not correct,
11530 you can override it with the @code{set height} and @code{set
11537 @kindex show height
11538 @item set height @var{lpp}
11540 @itemx set width @var{cpl}
11542 These @code{set} commands specify a screen height of @var{lpp} lines and
11543 a screen width of @var{cpl} characters. The associated @code{show}
11544 commands display the current settings.
11546 If you specify a height of zero lines, @value{GDBN} does not pause during
11547 output no matter how long the output is. This is useful if output is to a
11548 file or to an editor buffer.
11550 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
11551 from wrapping its output.
11556 @cindex number representation
11557 @cindex entering numbers
11559 You can always enter numbers in octal, decimal, or hexadecimal in
11560 @value{GDBN} by the usual conventions: octal numbers begin with
11561 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
11562 begin with @samp{0x}. Numbers that begin with none of these are, by
11563 default, entered in base 10; likewise, the default display for
11564 numbers---when no particular format is specified---is base 10. You can
11565 change the default base for both input and output with the @code{set
11569 @kindex set input-radix
11570 @item set input-radix @var{base}
11571 Set the default base for numeric input. Supported choices
11572 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
11573 specified either unambiguously or using the current default radix; for
11583 sets the base to decimal. On the other hand, @samp{set radix 10}
11584 leaves the radix unchanged no matter what it was.
11586 @kindex set output-radix
11587 @item set output-radix @var{base}
11588 Set the default base for numeric display. Supported choices
11589 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
11590 specified either unambiguously or using the current default radix.
11592 @kindex show input-radix
11593 @item show input-radix
11594 Display the current default base for numeric input.
11596 @kindex show output-radix
11597 @item show output-radix
11598 Display the current default base for numeric display.
11601 @node Messages/Warnings
11602 @section Optional warnings and messages
11604 By default, @value{GDBN} is silent about its inner workings. If you are
11605 running on a slow machine, you may want to use the @code{set verbose}
11606 command. This makes @value{GDBN} tell you when it does a lengthy
11607 internal operation, so you will not think it has crashed.
11609 Currently, the messages controlled by @code{set verbose} are those
11610 which announce that the symbol table for a source file is being read;
11611 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
11614 @kindex set verbose
11615 @item set verbose on
11616 Enables @value{GDBN} output of certain informational messages.
11618 @item set verbose off
11619 Disables @value{GDBN} output of certain informational messages.
11621 @kindex show verbose
11623 Displays whether @code{set verbose} is on or off.
11626 By default, if @value{GDBN} encounters bugs in the symbol table of an
11627 object file, it is silent; but if you are debugging a compiler, you may
11628 find this information useful (@pxref{Symbol Errors, ,Errors reading
11633 @kindex set complaints
11634 @item set complaints @var{limit}
11635 Permits @value{GDBN} to output @var{limit} complaints about each type of
11636 unusual symbols before becoming silent about the problem. Set
11637 @var{limit} to zero to suppress all complaints; set it to a large number
11638 to prevent complaints from being suppressed.
11640 @kindex show complaints
11641 @item show complaints
11642 Displays how many symbol complaints @value{GDBN} is permitted to produce.
11646 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
11647 lot of stupid questions to confirm certain commands. For example, if
11648 you try to run a program which is already running:
11652 The program being debugged has been started already.
11653 Start it from the beginning? (y or n)
11656 If you are willing to unflinchingly face the consequences of your own
11657 commands, you can disable this ``feature'':
11661 @kindex set confirm
11663 @cindex confirmation
11664 @cindex stupid questions
11665 @item set confirm off
11666 Disables confirmation requests.
11668 @item set confirm on
11669 Enables confirmation requests (the default).
11671 @kindex show confirm
11673 Displays state of confirmation requests.
11677 @node Debugging Output
11678 @section Optional messages about internal happenings
11680 @kindex set debug arch
11681 @item set debug arch
11682 Turns on or off display of gdbarch debugging info. The default is off
11683 @kindex show debug arch
11684 @item show debug arch
11685 Displays the current state of displaying gdbarch debugging info.
11686 @kindex set debug event
11687 @item set debug event
11688 Turns on or off display of @value{GDBN} event debugging info. The
11690 @kindex show debug event
11691 @item show debug event
11692 Displays the current state of displaying @value{GDBN} event debugging
11694 @kindex set debug expression
11695 @item set debug expression
11696 Turns on or off display of @value{GDBN} expression debugging info. The
11698 @kindex show debug expression
11699 @item show debug expression
11700 Displays the current state of displaying @value{GDBN} expression
11702 @kindex set debug overload
11703 @item set debug overload
11704 Turns on or off display of @value{GDBN} C++ overload debugging
11705 info. This includes info such as ranking of functions, etc. The default
11707 @kindex show debug overload
11708 @item show debug overload
11709 Displays the current state of displaying @value{GDBN} C++ overload
11711 @kindex set debug remote
11712 @cindex packets, reporting on stdout
11713 @cindex serial connections, debugging
11714 @item set debug remote
11715 Turns on or off display of reports on all packets sent back and forth across
11716 the serial line to the remote machine. The info is printed on the
11717 @value{GDBN} standard output stream. The default is off.
11718 @kindex show debug remote
11719 @item show debug remote
11720 Displays the state of display of remote packets.
11721 @kindex set debug serial
11722 @item set debug serial
11723 Turns on or off display of @value{GDBN} serial debugging info. The
11725 @kindex show debug serial
11726 @item show debug serial
11727 Displays the current state of displaying @value{GDBN} serial debugging
11729 @kindex set debug target
11730 @item set debug target
11731 Turns on or off display of @value{GDBN} target debugging info. This info
11732 includes what is going on at the target level of GDB, as it happens. The
11734 @kindex show debug target
11735 @item show debug target
11736 Displays the current state of displaying @value{GDBN} target debugging
11738 @kindex set debug varobj
11739 @item set debug varobj
11740 Turns on or off display of @value{GDBN} variable object debugging
11741 info. The default is off.
11742 @kindex show debug varobj
11743 @item show debug varobj
11744 Displays the current state of displaying @value{GDBN} variable object
11749 @chapter Canned Sequences of Commands
11751 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
11752 command lists}), @value{GDBN} provides two ways to store sequences of
11753 commands for execution as a unit: user-defined commands and command
11757 * Define:: User-defined commands
11758 * Hooks:: User-defined command hooks
11759 * Command Files:: Command files
11760 * Output:: Commands for controlled output
11764 @section User-defined commands
11766 @cindex user-defined command
11767 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
11768 which you assign a new name as a command. This is done with the
11769 @code{define} command. User commands may accept up to 10 arguments
11770 separated by whitespace. Arguments are accessed within the user command
11771 via @var{$arg0@dots{}$arg9}. A trivial example:
11775 print $arg0 + $arg1 + $arg2
11779 To execute the command use:
11786 This defines the command @code{adder}, which prints the sum of
11787 its three arguments. Note the arguments are text substitutions, so they may
11788 reference variables, use complex expressions, or even perform inferior
11794 @item define @var{commandname}
11795 Define a command named @var{commandname}. If there is already a command
11796 by that name, you are asked to confirm that you want to redefine it.
11798 The definition of the command is made up of other @value{GDBN} command lines,
11799 which are given following the @code{define} command. The end of these
11800 commands is marked by a line containing @code{end}.
11805 Takes a single argument, which is an expression to evaluate.
11806 It is followed by a series of commands that are executed
11807 only if the expression is true (nonzero).
11808 There can then optionally be a line @code{else}, followed
11809 by a series of commands that are only executed if the expression
11810 was false. The end of the list is marked by a line containing @code{end}.
11814 The syntax is similar to @code{if}: the command takes a single argument,
11815 which is an expression to evaluate, and must be followed by the commands to
11816 execute, one per line, terminated by an @code{end}.
11817 The commands are executed repeatedly as long as the expression
11821 @item document @var{commandname}
11822 Document the user-defined command @var{commandname}, so that it can be
11823 accessed by @code{help}. The command @var{commandname} must already be
11824 defined. This command reads lines of documentation just as @code{define}
11825 reads the lines of the command definition, ending with @code{end}.
11826 After the @code{document} command is finished, @code{help} on command
11827 @var{commandname} displays the documentation you have written.
11829 You may use the @code{document} command again to change the
11830 documentation of a command. Redefining the command with @code{define}
11831 does not change the documentation.
11833 @kindex help user-defined
11834 @item help user-defined
11835 List all user-defined commands, with the first line of the documentation
11840 @itemx show user @var{commandname}
11841 Display the @value{GDBN} commands used to define @var{commandname} (but
11842 not its documentation). If no @var{commandname} is given, display the
11843 definitions for all user-defined commands.
11847 When user-defined commands are executed, the
11848 commands of the definition are not printed. An error in any command
11849 stops execution of the user-defined command.
11851 If used interactively, commands that would ask for confirmation proceed
11852 without asking when used inside a user-defined command. Many @value{GDBN}
11853 commands that normally print messages to say what they are doing omit the
11854 messages when used in a user-defined command.
11857 @section User-defined command hooks
11858 @cindex command hooks
11859 @cindex hooks, for commands
11861 You may define @emph{hooks}, which are a special kind of user-defined
11862 command. Whenever you run the command @samp{foo}, if the user-defined
11863 command @samp{hook-foo} exists, it is executed (with no arguments)
11864 before that command.
11866 @kindex stop@r{, a pseudo-command}
11867 In addition, a pseudo-command, @samp{stop} exists. Defining
11868 (@samp{hook-stop}) makes the associated commands execute every time
11869 execution stops in your program: before breakpoint commands are run,
11870 displays are printed, or the stack frame is printed.
11872 For example, to ignore @code{SIGALRM} signals while
11873 single-stepping, but treat them normally during normal execution,
11878 handle SIGALRM nopass
11882 handle SIGALRM pass
11885 define hook-continue
11886 handle SIGLARM pass
11890 You can define a hook for any single-word command in @value{GDBN}, but
11891 not for command aliases; you should define a hook for the basic command
11892 name, e.g. @code{backtrace} rather than @code{bt}.
11893 @c FIXME! So how does Joe User discover whether a command is an alias
11895 If an error occurs during the execution of your hook, execution of
11896 @value{GDBN} commands stops and @value{GDBN} issues a prompt
11897 (before the command that you actually typed had a chance to run).
11899 If you try to define a hook which does not match any known command, you
11900 get a warning from the @code{define} command.
11902 @node Command Files
11903 @section Command files
11905 @cindex command files
11906 A command file for @value{GDBN} is a file of lines that are @value{GDBN}
11907 commands. Comments (lines starting with @kbd{#}) may also be included.
11908 An empty line in a command file does nothing; it does not mean to repeat
11909 the last command, as it would from the terminal.
11912 @cindex @file{.gdbinit}
11913 @cindex @file{gdb.ini}
11914 When you start @value{GDBN}, it automatically executes commands from its
11915 @dfn{init files}. These are files named @file{.gdbinit} on Unix and
11916 @file{gdb.ini} on DOS/Windows. During startup, @value{GDBN} does the
11921 Reads the init file (if any) in your home directory@footnote{On
11922 DOS/Windows systems, the home directory is the one pointed to by the
11923 @code{HOME} environment variable.}.
11926 Processes command line options and operands.
11929 Reads the init file (if any) in the current working directory.
11932 Reads command files specified by the @samp{-x} option.
11935 The init file in your home directory can set options (such as @samp{set
11936 complaints}) that affect subsequent processing of command line options
11937 and operands. Init files are not executed if you use the @samp{-nx}
11938 option (@pxref{Mode Options, ,Choosing modes}).
11940 @cindex init file name
11941 On some configurations of @value{GDBN}, the init file is known by a
11942 different name (these are typically environments where a specialized
11943 form of @value{GDBN} may need to coexist with other forms, hence a
11944 different name for the specialized version's init file). These are the
11945 environments with special init file names:
11947 @cindex @file{.vxgdbinit}
11950 VxWorks (Wind River Systems real-time OS): @file{.vxgdbinit}
11952 @cindex @file{.os68gdbinit}
11954 OS68K (Enea Data Systems real-time OS): @file{.os68gdbinit}
11956 @cindex @file{.esgdbinit}
11958 ES-1800 (Ericsson Telecom AB M68000 emulator): @file{.esgdbinit}
11961 You can also request the execution of a command file with the
11962 @code{source} command:
11966 @item source @var{filename}
11967 Execute the command file @var{filename}.
11970 The lines in a command file are executed sequentially. They are not
11971 printed as they are executed. An error in any command terminates execution
11972 of the command file.
11974 Commands that would ask for confirmation if used interactively proceed
11975 without asking when used in a command file. Many @value{GDBN} commands that
11976 normally print messages to say what they are doing omit the messages
11977 when called from command files.
11980 @section Commands for controlled output
11982 During the execution of a command file or a user-defined command, normal
11983 @value{GDBN} output is suppressed; the only output that appears is what is
11984 explicitly printed by the commands in the definition. This section
11985 describes three commands useful for generating exactly the output you
11990 @item echo @var{text}
11991 @c I do not consider backslash-space a standard C escape sequence
11992 @c because it is not in ANSI.
11993 Print @var{text}. Nonprinting characters can be included in
11994 @var{text} using C escape sequences, such as @samp{\n} to print a
11995 newline. @strong{No newline is printed unless you specify one.}
11996 In addition to the standard C escape sequences, a backslash followed
11997 by a space stands for a space. This is useful for displaying a
11998 string with spaces at the beginning or the end, since leading and
11999 trailing spaces are otherwise trimmed from all arguments.
12000 To print @samp{@w{ }and foo =@w{ }}, use the command
12001 @samp{echo \@w{ }and foo = \@w{ }}.
12003 A backslash at the end of @var{text} can be used, as in C, to continue
12004 the command onto subsequent lines. For example,
12007 echo This is some text\n\
12008 which is continued\n\
12009 onto several lines.\n
12012 produces the same output as
12015 echo This is some text\n
12016 echo which is continued\n
12017 echo onto several lines.\n
12021 @item output @var{expression}
12022 Print the value of @var{expression} and nothing but that value: no
12023 newlines, no @samp{$@var{nn} = }. The value is not entered in the
12024 value history either. @xref{Expressions, ,Expressions}, for more information
12027 @item output/@var{fmt} @var{expression}
12028 Print the value of @var{expression} in format @var{fmt}. You can use
12029 the same formats as for @code{print}. @xref{Output Formats,,Output
12030 formats}, for more information.
12033 @item printf @var{string}, @var{expressions}@dots{}
12034 Print the values of the @var{expressions} under the control of
12035 @var{string}. The @var{expressions} are separated by commas and may be
12036 either numbers or pointers. Their values are printed as specified by
12037 @var{string}, exactly as if your program were to execute the C
12039 @c FIXME: the above implies that at least all ANSI C formats are
12040 @c supported, but it isn't true: %E and %G don't work (or so it seems).
12041 @c Either this is a bug, or the manual should document what formats are
12045 printf (@var{string}, @var{expressions}@dots{});
12048 For example, you can print two values in hex like this:
12051 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
12054 The only backslash-escape sequences that you can use in the format
12055 string are the simple ones that consist of backslash followed by a
12060 @chapter Using @value{GDBN} under @sc{gnu} Emacs
12063 @cindex @sc{gnu} Emacs
12064 A special interface allows you to use @sc{gnu} Emacs to view (and
12065 edit) the source files for the program you are debugging with
12068 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
12069 executable file you want to debug as an argument. This command starts
12070 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
12071 created Emacs buffer.
12072 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
12074 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
12079 All ``terminal'' input and output goes through the Emacs buffer.
12082 This applies both to @value{GDBN} commands and their output, and to the input
12083 and output done by the program you are debugging.
12085 This is useful because it means that you can copy the text of previous
12086 commands and input them again; you can even use parts of the output
12089 All the facilities of Emacs' Shell mode are available for interacting
12090 with your program. In particular, you can send signals the usual
12091 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
12096 @value{GDBN} displays source code through Emacs.
12099 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
12100 source file for that frame and puts an arrow (@samp{=>}) at the
12101 left margin of the current line. Emacs uses a separate buffer for
12102 source display, and splits the screen to show both your @value{GDBN} session
12105 Explicit @value{GDBN} @code{list} or search commands still produce output as
12106 usual, but you probably have no reason to use them from Emacs.
12109 @emph{Warning:} If the directory where your program resides is not your
12110 current directory, it can be easy to confuse Emacs about the location of
12111 the source files, in which case the auxiliary display buffer does not
12112 appear to show your source. @value{GDBN} can find programs by searching your
12113 environment's @code{PATH} variable, so the @value{GDBN} input and output
12114 session proceeds normally; but Emacs does not get enough information
12115 back from @value{GDBN} to locate the source files in this situation. To
12116 avoid this problem, either start @value{GDBN} mode from the directory where
12117 your program resides, or specify an absolute file name when prompted for the
12118 @kbd{M-x gdb} argument.
12120 A similar confusion can result if you use the @value{GDBN} @code{file} command to
12121 switch to debugging a program in some other location, from an existing
12122 @value{GDBN} buffer in Emacs.
12125 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
12126 you need to call @value{GDBN} by a different name (for example, if you keep
12127 several configurations around, with different names) you can set the
12128 Emacs variable @code{gdb-command-name}; for example,
12131 (setq gdb-command-name "mygdb")
12135 (preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or
12136 in your @file{.emacs} file) makes Emacs call the program named
12137 ``@code{mygdb}'' instead.
12139 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
12140 addition to the standard Shell mode commands:
12144 Describe the features of Emacs' @value{GDBN} Mode.
12147 Execute to another source line, like the @value{GDBN} @code{step} command; also
12148 update the display window to show the current file and location.
12151 Execute to next source line in this function, skipping all function
12152 calls, like the @value{GDBN} @code{next} command. Then update the display window
12153 to show the current file and location.
12156 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
12157 display window accordingly.
12159 @item M-x gdb-nexti
12160 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
12161 display window accordingly.
12164 Execute until exit from the selected stack frame, like the @value{GDBN}
12165 @code{finish} command.
12168 Continue execution of your program, like the @value{GDBN} @code{continue}
12171 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
12174 Go up the number of frames indicated by the numeric argument
12175 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
12176 like the @value{GDBN} @code{up} command.
12178 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
12181 Go down the number of frames indicated by the numeric argument, like the
12182 @value{GDBN} @code{down} command.
12184 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
12187 Read the number where the cursor is positioned, and insert it at the end
12188 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
12189 around an address that was displayed earlier, type @kbd{disassemble};
12190 then move the cursor to the address display, and pick up the
12191 argument for @code{disassemble} by typing @kbd{C-x &}.
12193 You can customize this further by defining elements of the list
12194 @code{gdb-print-command}; once it is defined, you can format or
12195 otherwise process numbers picked up by @kbd{C-x &} before they are
12196 inserted. A numeric argument to @kbd{C-x &} indicates that you
12197 wish special formatting, and also acts as an index to pick an element of the
12198 list. If the list element is a string, the number to be inserted is
12199 formatted using the Emacs function @code{format}; otherwise the number
12200 is passed as an argument to the corresponding list element.
12203 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
12204 tells @value{GDBN} to set a breakpoint on the source line point is on.
12206 If you accidentally delete the source-display buffer, an easy way to get
12207 it back is to type the command @code{f} in the @value{GDBN} buffer, to
12208 request a frame display; when you run under Emacs, this recreates
12209 the source buffer if necessary to show you the context of the current
12212 The source files displayed in Emacs are in ordinary Emacs buffers
12213 which are visiting the source files in the usual way. You can edit
12214 the files with these buffers if you wish; but keep in mind that @value{GDBN}
12215 communicates with Emacs in terms of line numbers. If you add or
12216 delete lines from the text, the line numbers that @value{GDBN} knows cease
12217 to correspond properly with the code.
12219 @c The following dropped because Epoch is nonstandard. Reactivate
12220 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
12222 @kindex Emacs Epoch environment
12226 Version 18 of @sc{gnu} Emacs has a built-in window system
12227 called the @code{epoch}
12228 environment. Users of this environment can use a new command,
12229 @code{inspect} which performs identically to @code{print} except that
12230 each value is printed in its own window.
12233 @include annotate.texi
12234 @include gdbmi.texinfo
12237 @chapter Reporting Bugs in @value{GDBN}
12238 @cindex bugs in @value{GDBN}
12239 @cindex reporting bugs in @value{GDBN}
12241 Your bug reports play an essential role in making @value{GDBN} reliable.
12243 Reporting a bug may help you by bringing a solution to your problem, or it
12244 may not. But in any case the principal function of a bug report is to help
12245 the entire community by making the next version of @value{GDBN} work better. Bug
12246 reports are your contribution to the maintenance of @value{GDBN}.
12248 In order for a bug report to serve its purpose, you must include the
12249 information that enables us to fix the bug.
12252 * Bug Criteria:: Have you found a bug?
12253 * Bug Reporting:: How to report bugs
12257 @section Have you found a bug?
12258 @cindex bug criteria
12260 If you are not sure whether you have found a bug, here are some guidelines:
12263 @cindex fatal signal
12264 @cindex debugger crash
12265 @cindex crash of debugger
12267 If the debugger gets a fatal signal, for any input whatever, that is a
12268 @value{GDBN} bug. Reliable debuggers never crash.
12270 @cindex error on valid input
12272 If @value{GDBN} produces an error message for valid input, that is a
12273 bug. (Note that if you're cross debugging, the problem may also be
12274 somewhere in the connection to the target.)
12276 @cindex invalid input
12278 If @value{GDBN} does not produce an error message for invalid input,
12279 that is a bug. However, you should note that your idea of
12280 ``invalid input'' might be our idea of ``an extension'' or ``support
12281 for traditional practice''.
12284 If you are an experienced user of debugging tools, your suggestions
12285 for improvement of @value{GDBN} are welcome in any case.
12288 @node Bug Reporting
12289 @section How to report bugs
12290 @cindex bug reports
12291 @cindex @value{GDBN} bugs, reporting
12293 A number of companies and individuals offer support for @sc{gnu} products.
12294 If you obtained @value{GDBN} from a support organization, we recommend you
12295 contact that organization first.
12297 You can find contact information for many support companies and
12298 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
12300 @c should add a web page ref...
12302 In any event, we also recommend that you send bug reports for
12303 @value{GDBN} to this addresses:
12309 @strong{Do not send bug reports to @samp{info-gdb}, or to
12310 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
12311 not want to receive bug reports. Those that do have arranged to receive
12314 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
12315 serves as a repeater. The mailing list and the newsgroup carry exactly
12316 the same messages. Often people think of posting bug reports to the
12317 newsgroup instead of mailing them. This appears to work, but it has one
12318 problem which can be crucial: a newsgroup posting often lacks a mail
12319 path back to the sender. Thus, if we need to ask for more information,
12320 we may be unable to reach you. For this reason, it is better to send
12321 bug reports to the mailing list.
12323 As a last resort, send bug reports on paper to:
12326 @sc{gnu} Debugger Bugs
12327 Free Software Foundation Inc.
12328 59 Temple Place - Suite 330
12329 Boston, MA 02111-1307
12333 The fundamental principle of reporting bugs usefully is this:
12334 @strong{report all the facts}. If you are not sure whether to state a
12335 fact or leave it out, state it!
12337 Often people omit facts because they think they know what causes the
12338 problem and assume that some details do not matter. Thus, you might
12339 assume that the name of the variable you use in an example does not matter.
12340 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
12341 stray memory reference which happens to fetch from the location where that
12342 name is stored in memory; perhaps, if the name were different, the contents
12343 of that location would fool the debugger into doing the right thing despite
12344 the bug. Play it safe and give a specific, complete example. That is the
12345 easiest thing for you to do, and the most helpful.
12347 Keep in mind that the purpose of a bug report is to enable us to fix the
12348 bug. It may be that the bug has been reported previously, but neither
12349 you nor we can know that unless your bug report is complete and
12352 Sometimes people give a few sketchy facts and ask, ``Does this ring a
12353 bell?'' Those bug reports are useless, and we urge everyone to
12354 @emph{refuse to respond to them} except to chide the sender to report
12357 To enable us to fix the bug, you should include all these things:
12361 The version of @value{GDBN}. @value{GDBN} announces it if you start
12362 with no arguments; you can also print it at any time using @code{show
12365 Without this, we will not know whether there is any point in looking for
12366 the bug in the current version of @value{GDBN}.
12369 The type of machine you are using, and the operating system name and
12373 What compiler (and its version) was used to compile @value{GDBN}---e.g.
12374 ``@value{GCC}--2.8.1''.
12377 What compiler (and its version) was used to compile the program you are
12378 debugging---e.g. ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
12379 C Compiler''. For GCC, you can say @code{gcc --version} to get this
12380 information; for other compilers, see the documentation for those
12384 The command arguments you gave the compiler to compile your example and
12385 observe the bug. For example, did you use @samp{-O}? To guarantee
12386 you will not omit something important, list them all. A copy of the
12387 Makefile (or the output from make) is sufficient.
12389 If we were to try to guess the arguments, we would probably guess wrong
12390 and then we might not encounter the bug.
12393 A complete input script, and all necessary source files, that will
12397 A description of what behavior you observe that you believe is
12398 incorrect. For example, ``It gets a fatal signal.''
12400 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
12401 will certainly notice it. But if the bug is incorrect output, we might
12402 not notice unless it is glaringly wrong. You might as well not give us
12403 a chance to make a mistake.
12405 Even if the problem you experience is a fatal signal, you should still
12406 say so explicitly. Suppose something strange is going on, such as, your
12407 copy of @value{GDBN} is out of synch, or you have encountered a bug in
12408 the C library on your system. (This has happened!) Your copy might
12409 crash and ours would not. If you told us to expect a crash, then when
12410 ours fails to crash, we would know that the bug was not happening for
12411 us. If you had not told us to expect a crash, then we would not be able
12412 to draw any conclusion from our observations.
12415 If you wish to suggest changes to the @value{GDBN} source, send us context
12416 diffs. If you even discuss something in the @value{GDBN} source, refer to
12417 it by context, not by line number.
12419 The line numbers in our development sources will not match those in your
12420 sources. Your line numbers would convey no useful information to us.
12424 Here are some things that are not necessary:
12428 A description of the envelope of the bug.
12430 Often people who encounter a bug spend a lot of time investigating
12431 which changes to the input file will make the bug go away and which
12432 changes will not affect it.
12434 This is often time consuming and not very useful, because the way we
12435 will find the bug is by running a single example under the debugger
12436 with breakpoints, not by pure deduction from a series of examples.
12437 We recommend that you save your time for something else.
12439 Of course, if you can find a simpler example to report @emph{instead}
12440 of the original one, that is a convenience for us. Errors in the
12441 output will be easier to spot, running under the debugger will take
12442 less time, and so on.
12444 However, simplification is not vital; if you do not want to do this,
12445 report the bug anyway and send us the entire test case you used.
12448 A patch for the bug.
12450 A patch for the bug does help us if it is a good one. But do not omit
12451 the necessary information, such as the test case, on the assumption that
12452 a patch is all we need. We might see problems with your patch and decide
12453 to fix the problem another way, or we might not understand it at all.
12455 Sometimes with a program as complicated as @value{GDBN} it is very hard to
12456 construct an example that will make the program follow a certain path
12457 through the code. If you do not send us the example, we will not be able
12458 to construct one, so we will not be able to verify that the bug is fixed.
12460 And if we cannot understand what bug you are trying to fix, or why your
12461 patch should be an improvement, we will not install it. A test case will
12462 help us to understand.
12465 A guess about what the bug is or what it depends on.
12467 Such guesses are usually wrong. Even we cannot guess right about such
12468 things without first using the debugger to find the facts.
12471 @c The readline documentation is distributed with the readline code
12472 @c and consists of the two following files:
12474 @c inc-hist.texinfo
12475 @c Use -I with makeinfo to point to the appropriate directory,
12476 @c environment var TEXINPUTS with TeX.
12477 @include rluser.texinfo
12478 @include inc-hist.texinfo
12481 @node Formatting Documentation
12482 @appendix Formatting Documentation
12484 @cindex @value{GDBN} reference card
12485 @cindex reference card
12486 The @value{GDBN} 4 release includes an already-formatted reference card, ready
12487 for printing with PostScript or Ghostscript, in the @file{gdb}
12488 subdirectory of the main source directory@footnote{In
12489 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
12490 release.}. If you can use PostScript or Ghostscript with your printer,
12491 you can print the reference card immediately with @file{refcard.ps}.
12493 The release also includes the source for the reference card. You
12494 can format it, using @TeX{}, by typing:
12500 The @value{GDBN} reference card is designed to print in @dfn{landscape}
12501 mode on US ``letter'' size paper;
12502 that is, on a sheet 11 inches wide by 8.5 inches
12503 high. You will need to specify this form of printing as an option to
12504 your @sc{dvi} output program.
12506 @cindex documentation
12508 All the documentation for @value{GDBN} comes as part of the machine-readable
12509 distribution. The documentation is written in Texinfo format, which is
12510 a documentation system that uses a single source file to produce both
12511 on-line information and a printed manual. You can use one of the Info
12512 formatting commands to create the on-line version of the documentation
12513 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
12515 @value{GDBN} includes an already formatted copy of the on-line Info
12516 version of this manual in the @file{gdb} subdirectory. The main Info
12517 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
12518 subordinate files matching @samp{gdb.info*} in the same directory. If
12519 necessary, you can print out these files, or read them with any editor;
12520 but they are easier to read using the @code{info} subsystem in @sc{gnu}
12521 Emacs or the standalone @code{info} program, available as part of the
12522 @sc{gnu} Texinfo distribution.
12524 If you want to format these Info files yourself, you need one of the
12525 Info formatting programs, such as @code{texinfo-format-buffer} or
12528 If you have @code{makeinfo} installed, and are in the top level
12529 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
12530 version @value{GDBVN}), you can make the Info file by typing:
12537 If you want to typeset and print copies of this manual, you need @TeX{},
12538 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
12539 Texinfo definitions file.
12541 @TeX{} is a typesetting program; it does not print files directly, but
12542 produces output files called @sc{dvi} files. To print a typeset
12543 document, you need a program to print @sc{dvi} files. If your system
12544 has @TeX{} installed, chances are it has such a program. The precise
12545 command to use depends on your system; @kbd{lpr -d} is common; another
12546 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
12547 require a file name without any extension or a @samp{.dvi} extension.
12549 @TeX{} also requires a macro definitions file called
12550 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
12551 written in Texinfo format. On its own, @TeX{} cannot either read or
12552 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
12553 and is located in the @file{gdb-@var{version-number}/texinfo}
12556 If you have @TeX{} and a @sc{dvi} printer program installed, you can
12557 typeset and print this manual. First switch to the the @file{gdb}
12558 subdirectory of the main source directory (for example, to
12559 @file{gdb-@value{GDBVN}/gdb}) and type:
12565 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
12567 @node Installing GDB
12568 @appendix Installing @value{GDBN}
12569 @cindex configuring @value{GDBN}
12570 @cindex installation
12572 @value{GDBN} comes with a @code{configure} script that automates the process
12573 of preparing @value{GDBN} for installation; you can then use @code{make} to
12574 build the @code{gdb} program.
12576 @c irrelevant in info file; it's as current as the code it lives with.
12577 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
12578 look at the @file{README} file in the sources; we may have improved the
12579 installation procedures since publishing this manual.}
12582 The @value{GDBN} distribution includes all the source code you need for
12583 @value{GDBN} in a single directory, whose name is usually composed by
12584 appending the version number to @samp{gdb}.
12586 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
12587 @file{gdb-@value{GDBVN}} directory. That directory contains:
12590 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
12591 script for configuring @value{GDBN} and all its supporting libraries
12593 @item gdb-@value{GDBVN}/gdb
12594 the source specific to @value{GDBN} itself
12596 @item gdb-@value{GDBVN}/bfd
12597 source for the Binary File Descriptor library
12599 @item gdb-@value{GDBVN}/include
12600 @sc{gnu} include files
12602 @item gdb-@value{GDBVN}/libiberty
12603 source for the @samp{-liberty} free software library
12605 @item gdb-@value{GDBVN}/opcodes
12606 source for the library of opcode tables and disassemblers
12608 @item gdb-@value{GDBVN}/readline
12609 source for the @sc{gnu} command-line interface
12611 @item gdb-@value{GDBVN}/glob
12612 source for the @sc{gnu} filename pattern-matching subroutine
12614 @item gdb-@value{GDBVN}/mmalloc
12615 source for the @sc{gnu} memory-mapped malloc package
12618 The simplest way to configure and build @value{GDBN} is to run @code{configure}
12619 from the @file{gdb-@var{version-number}} source directory, which in
12620 this example is the @file{gdb-@value{GDBVN}} directory.
12622 First switch to the @file{gdb-@var{version-number}} source directory
12623 if you are not already in it; then run @code{configure}. Pass the
12624 identifier for the platform on which @value{GDBN} will run as an
12630 cd gdb-@value{GDBVN}
12631 ./configure @var{host}
12636 where @var{host} is an identifier such as @samp{sun4} or
12637 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
12638 (You can often leave off @var{host}; @code{configure} tries to guess the
12639 correct value by examining your system.)
12641 Running @samp{configure @var{host}} and then running @code{make} builds the
12642 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
12643 libraries, then @code{gdb} itself. The configured source files, and the
12644 binaries, are left in the corresponding source directories.
12647 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
12648 system does not recognize this automatically when you run a different
12649 shell, you may need to run @code{sh} on it explicitly:
12652 sh configure @var{host}
12655 If you run @code{configure} from a directory that contains source
12656 directories for multiple libraries or programs, such as the
12657 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
12658 creates configuration files for every directory level underneath (unless
12659 you tell it not to, with the @samp{--norecursion} option).
12661 You can run the @code{configure} script from any of the
12662 subordinate directories in the @value{GDBN} distribution if you only want to
12663 configure that subdirectory, but be sure to specify a path to it.
12665 For example, with version @value{GDBVN}, type the following to configure only
12666 the @code{bfd} subdirectory:
12670 cd gdb-@value{GDBVN}/bfd
12671 ../configure @var{host}
12675 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
12676 However, you should make sure that the shell on your path (named by
12677 the @samp{SHELL} environment variable) is publicly readable. Remember
12678 that @value{GDBN} uses the shell to start your program---some systems refuse to
12679 let @value{GDBN} debug child processes whose programs are not readable.
12682 * Separate Objdir:: Compiling @value{GDBN} in another directory
12683 * Config Names:: Specifying names for hosts and targets
12684 * Configure Options:: Summary of options for configure
12687 @node Separate Objdir
12688 @section Compiling @value{GDBN} in another directory
12690 If you want to run @value{GDBN} versions for several host or target machines,
12691 you need a different @code{gdb} compiled for each combination of
12692 host and target. @code{configure} is designed to make this easy by
12693 allowing you to generate each configuration in a separate subdirectory,
12694 rather than in the source directory. If your @code{make} program
12695 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
12696 @code{make} in each of these directories builds the @code{gdb}
12697 program specified there.
12699 To build @code{gdb} in a separate directory, run @code{configure}
12700 with the @samp{--srcdir} option to specify where to find the source.
12701 (You also need to specify a path to find @code{configure}
12702 itself from your working directory. If the path to @code{configure}
12703 would be the same as the argument to @samp{--srcdir}, you can leave out
12704 the @samp{--srcdir} option; it is assumed.)
12706 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
12707 separate directory for a Sun 4 like this:
12711 cd gdb-@value{GDBVN}
12714 ../gdb-@value{GDBVN}/configure sun4
12719 When @code{configure} builds a configuration using a remote source
12720 directory, it creates a tree for the binaries with the same structure
12721 (and using the same names) as the tree under the source directory. In
12722 the example, you'd find the Sun 4 library @file{libiberty.a} in the
12723 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
12724 @file{gdb-sun4/gdb}.
12726 One popular reason to build several @value{GDBN} configurations in separate
12727 directories is to configure @value{GDBN} for cross-compiling (where
12728 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
12729 programs that run on another machine---the @dfn{target}).
12730 You specify a cross-debugging target by
12731 giving the @samp{--target=@var{target}} option to @code{configure}.
12733 When you run @code{make} to build a program or library, you must run
12734 it in a configured directory---whatever directory you were in when you
12735 called @code{configure} (or one of its subdirectories).
12737 The @code{Makefile} that @code{configure} generates in each source
12738 directory also runs recursively. If you type @code{make} in a source
12739 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
12740 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
12741 will build all the required libraries, and then build GDB.
12743 When you have multiple hosts or targets configured in separate
12744 directories, you can run @code{make} on them in parallel (for example,
12745 if they are NFS-mounted on each of the hosts); they will not interfere
12749 @section Specifying names for hosts and targets
12751 The specifications used for hosts and targets in the @code{configure}
12752 script are based on a three-part naming scheme, but some short predefined
12753 aliases are also supported. The full naming scheme encodes three pieces
12754 of information in the following pattern:
12757 @var{architecture}-@var{vendor}-@var{os}
12760 For example, you can use the alias @code{sun4} as a @var{host} argument,
12761 or as the value for @var{target} in a @code{--target=@var{target}}
12762 option. The equivalent full name is @samp{sparc-sun-sunos4}.
12764 The @code{configure} script accompanying @value{GDBN} does not provide
12765 any query facility to list all supported host and target names or
12766 aliases. @code{configure} calls the Bourne shell script
12767 @code{config.sub} to map abbreviations to full names; you can read the
12768 script, if you wish, or you can use it to test your guesses on
12769 abbreviations---for example:
12772 % sh config.sub i386-linux
12774 % sh config.sub alpha-linux
12775 alpha-unknown-linux-gnu
12776 % sh config.sub hp9k700
12778 % sh config.sub sun4
12779 sparc-sun-sunos4.1.1
12780 % sh config.sub sun3
12781 m68k-sun-sunos4.1.1
12782 % sh config.sub i986v
12783 Invalid configuration `i986v': machine `i986v' not recognized
12787 @code{config.sub} is also distributed in the @value{GDBN} source
12788 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
12790 @node Configure Options
12791 @section @code{configure} options
12793 Here is a summary of the @code{configure} options and arguments that
12794 are most often useful for building @value{GDBN}. @code{configure} also has
12795 several other options not listed here. @inforef{What Configure
12796 Does,,configure.info}, for a full explanation of @code{configure}.
12799 configure @r{[}--help@r{]}
12800 @r{[}--prefix=@var{dir}@r{]}
12801 @r{[}--exec-prefix=@var{dir}@r{]}
12802 @r{[}--srcdir=@var{dirname}@r{]}
12803 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
12804 @r{[}--target=@var{target}@r{]}
12809 You may introduce options with a single @samp{-} rather than
12810 @samp{--} if you prefer; but you may abbreviate option names if you use
12815 Display a quick summary of how to invoke @code{configure}.
12817 @item --prefix=@var{dir}
12818 Configure the source to install programs and files under directory
12821 @item --exec-prefix=@var{dir}
12822 Configure the source to install programs under directory
12825 @c avoid splitting the warning from the explanation:
12827 @item --srcdir=@var{dirname}
12828 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
12829 @code{make} that implements the @code{VPATH} feature.}@*
12830 Use this option to make configurations in directories separate from the
12831 @value{GDBN} source directories. Among other things, you can use this to
12832 build (or maintain) several configurations simultaneously, in separate
12833 directories. @code{configure} writes configuration specific files in
12834 the current directory, but arranges for them to use the source in the
12835 directory @var{dirname}. @code{configure} creates directories under
12836 the working directory in parallel to the source directories below
12839 @item --norecursion
12840 Configure only the directory level where @code{configure} is executed; do not
12841 propagate configuration to subdirectories.
12843 @item --target=@var{target}
12844 Configure @value{GDBN} for cross-debugging programs running on the specified
12845 @var{target}. Without this option, @value{GDBN} is configured to debug
12846 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
12848 There is no convenient way to generate a list of all available targets.
12850 @item @var{host} @dots{}
12851 Configure @value{GDBN} to run on the specified @var{host}.
12853 There is no convenient way to generate a list of all available hosts.
12856 There are many other options available as well, but they are generally
12857 needed for special purposes only.
12865 % I think something like @colophon should be in texinfo. In the
12867 \long\def\colophon{\hbox to0pt{}\vfill
12868 \centerline{The body of this manual is set in}
12869 \centerline{\fontname\tenrm,}
12870 \centerline{with headings in {\bf\fontname\tenbf}}
12871 \centerline{and examples in {\tt\fontname\tentt}.}
12872 \centerline{{\it\fontname\tenit\/},}
12873 \centerline{{\bf\fontname\tenbf}, and}
12874 \centerline{{\sl\fontname\tensl\/}}
12875 \centerline{are used for emphasis.}\vfill}
12877 % Blame: doc@cygnus.com, 1991.