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
27 @c !!set GDB manual's edition---not the same as GDB version!
30 @c !!set GDB manual's revision date
31 @set DATE February 1999
33 @c THIS MANUAL REQUIRES TEXINFO-2 macros and info-makers to format properly.
35 @c This is a dir.info fragment to support semi-automated addition of
36 @c manuals to an info tree. zoo@cygnus.com is developing this facility.
37 @dircategory Programming & development tools.
39 * Gdb: (gdb). The @sc{gnu} debugger.
43 This file documents the @sc{gnu} debugger @value{GDBN}.
46 This is the @value{EDITION} Edition, @value{DATE},
47 of @cite{Debugging with @value{GDBN}: the @sc{gnu} Source-Level Debugger}
48 for @value{GDBN} Version @value{GDBVN}.
50 Copyright (C) 1988-1999 Free Software Foundation, Inc.
52 Permission is granted to make and distribute verbatim copies of
53 this manual provided the copyright notice and this permission notice
54 are preserved on all copies.
57 Permission is granted to process this file through TeX and print the
58 results, provided the printed document carries copying permission
59 notice identical to this one except for the removal of this paragraph
60 (this paragraph not being relevant to the printed manual).
63 Permission is granted to copy and distribute modified versions of this
64 manual under the conditions for verbatim copying, provided also that the
65 entire resulting derived work is distributed under the terms of a
66 permission notice identical to this one.
68 Permission is granted to copy and distribute translations of this manual
69 into another language, under the above conditions for modified versions.
73 @title Debugging with @value{GDBN}
74 @subtitle The @sc{gnu} Source-Level Debugger
76 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
77 @subtitle @value{DATE}
78 @author Richard M. Stallman and Roland H. Pesch
82 \hfill (Send bugs and comments on @value{GDBN} to bug-gdb\@gnu.org.)\par
83 \hfill {\it Debugging with @value{GDBN}}\par
84 \hfill \TeX{}info \texinfoversion\par
88 @c ISBN seems to be wrong...
90 @vskip 0pt plus 1filll
91 Copyright @copyright{} 1988-1999 Free Software Foundation, Inc.
93 Published by the Free Software Foundation @*
94 59 Temple Place - Suite 330, @*
95 Boston, MA 02111-1307 USA @*
96 Printed copies are available for $20 each. @*
99 Permission is granted to make and distribute verbatim copies of
100 this manual provided the copyright notice and this permission notice
101 are preserved on all copies.
103 Permission is granted to copy and distribute modified versions of this
104 manual under the conditions for verbatim copying, provided also that the
105 entire resulting derived work is distributed under the terms of a
106 permission notice identical to this one.
108 Permission is granted to copy and distribute translations of this manual
109 into another language, under the above conditions for modified versions.
115 @top Debugging with @value{GDBN}
117 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
119 This is the @value{EDITION} Edition, @value{DATE}, for @value{GDBN} Version
122 Copyright (C) 1988-1999 Free Software Foundation, Inc.
124 * Summary:: Summary of @value{GDBN}
125 * Sample Session:: A sample @value{GDBN} session
127 * Invocation:: Getting in and out of @value{GDBN}
128 * Commands:: @value{GDBN} commands
129 * Running:: Running programs under @value{GDBN}
130 * Stopping:: Stopping and continuing
131 * Stack:: Examining the stack
132 * Source:: Examining source files
133 * Data:: Examining data
135 * Languages:: Using @value{GDBN} with different languages
137 * Symbols:: Examining the symbol table
138 * Altering:: Altering execution
139 * GDB Files:: @value{GDBN} files
140 * Targets:: Specifying a debugging target
141 * Configurations:: Configuration-specific information
142 * Controlling GDB:: Controlling @value{GDBN}
143 * Sequences:: Canned sequences of commands
144 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
145 * Annotations:: @value{GDBN}'s annotations interface.
147 * GDB Bugs:: Reporting bugs in @value{GDBN}
148 * Formatting Documentation:: How to format and print @value{GDBN} documentation
150 * Command Line Editing:: Command Line Editing
151 * Using History Interactively:: Using History Interactively
152 * Installing GDB:: Installing GDB
159 @unnumbered Summary of @value{GDBN}
161 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
162 going on ``inside'' another program while it executes---or what another
163 program was doing at the moment it crashed.
165 @value{GDBN} can do four main kinds of things (plus other things in support of
166 these) to help you catch bugs in the act:
170 Start your program, specifying anything that might affect its behavior.
173 Make your program stop on specified conditions.
176 Examine what has happened, when your program has stopped.
179 Change things in your program, so you can experiment with correcting the
180 effects of one bug and go on to learn about another.
183 You can use @value{GDBN} to debug programs written in C and C++.
184 For more information, see @ref{Support,,Supported languages}.
185 For more information, see @ref{C,,C and C++}.
189 Support for Modula-2 and Chill is partial. For information on Modula-2,
190 see @ref{Modula-2,,Modula-2}. For information on Chill, see @ref{Chill}.
193 Debugging Pascal programs which use sets, subranges, file variables, or
194 nested functions does not currently work. @value{GDBN} does not support
195 entering expressions, printing values, or similar features using Pascal
199 @value{GDBN} can be used to debug programs written in Fortran, although
200 it may be necessary to refer to some variables with a trailing
204 * Free Software:: Freely redistributable software
205 * Contributors:: Contributors to GDB
209 @unnumberedsec Free software
211 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
212 General Public License
213 (GPL). The GPL gives you the freedom to copy or adapt a licensed
214 program---but every person getting a copy also gets with it the
215 freedom to modify that copy (which means that they must get access to
216 the source code), and the freedom to distribute further copies.
217 Typical software companies use copyrights to limit your freedoms; the
218 Free Software Foundation uses the GPL to preserve these freedoms.
220 Fundamentally, the General Public License is a license which says that
221 you have these freedoms and that you cannot take these freedoms away
225 @unnumberedsec Contributors to @value{GDBN}
227 Richard Stallman was the original author of @value{GDBN}, and of many
228 other @sc{gnu} programs. Many others have contributed to its
229 development. This section attempts to credit major contributors. One
230 of the virtues of free software is that everyone is free to contribute
231 to it; with regret, we cannot actually acknowledge everyone here. The
232 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
233 blow-by-blow account.
235 Changes much prior to version 2.0 are lost in the mists of time.
238 @emph{Plea:} Additions to this section are particularly welcome. If you
239 or your friends (or enemies, to be evenhanded) have been unfairly
240 omitted from this list, we would like to add your names!
243 So that they may not regard their many labors as thankless, we
244 particularly thank those who shepherded @value{GDBN} through major
246 Jim Blandy (release 4.18);
247 Jason Molenda (release 4.17);
248 Stan Shebs (release 4.14);
249 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
250 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
251 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
252 Jim Kingdon (releases 3.5, 3.4, and 3.3);
253 and Randy Smith (releases 3.2, 3.1, and 3.0).
255 Richard Stallman, assisted at various times by Peter TerMaat, Chris
256 Hanson, and Richard Mlynarik, handled releases through 2.8.
258 Michael Tiemann is the author of most of the @sc{gnu} C++ support in
259 @value{GDBN}, with significant additional contributions from Per
260 Bothner. James Clark wrote the @sc{gnu} C++ demangler. Early work on
261 C++ was by Peter TerMaat (who also did much general update work leading
264 @value{GDBN} 4 uses the BFD subroutine library to examine multiple
265 object-file formats; BFD was a joint project of David V.
266 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
268 David Johnson wrote the original COFF support; Pace Willison did
269 the original support for encapsulated COFF.
271 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
273 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
274 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
276 Jean-Daniel Fekete contributed Sun 386i support.
277 Chris Hanson improved the HP9000 support.
278 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
279 David Johnson contributed Encore Umax support.
280 Jyrki Kuoppala contributed Altos 3068 support.
281 Jeff Law contributed HP PA and SOM support.
282 Keith Packard contributed NS32K support.
283 Doug Rabson contributed Acorn Risc Machine support.
284 Bob Rusk contributed Harris Nighthawk CX-UX support.
285 Chris Smith contributed Convex support (and Fortran debugging).
286 Jonathan Stone contributed Pyramid support.
287 Michael Tiemann contributed SPARC support.
288 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
289 Pace Willison contributed Intel 386 support.
290 Jay Vosburgh contributed Symmetry support.
292 Andreas Schwab contributed M68K Linux support.
294 Rich Schaefer and Peter Schauer helped with support of SunOS shared
297 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
298 about several machine instruction sets.
300 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
301 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
302 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
303 and RDI targets, respectively.
305 Brian Fox is the author of the readline libraries providing
306 command-line editing and command history.
308 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
309 Modula-2 support, and contributed the Languages chapter of this manual.
311 Fred Fish wrote most of the support for Unix System Vr4.
312 He also enhanced the command-completion support to cover C++ overloaded
315 Hitachi America, Ltd. sponsored the support for H8/300, H8/500, and
318 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
320 Mitsubishi sponsored the support for D10V, D30V, and M32R/D processors.
322 Toshiba sponsored the support for the TX39 Mips processor.
324 Matsushita sponsored the support for the MN10200 and MN10300 processors.
326 Fujitsu sponsored the support for SPARClite and FR30 processors.
328 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
331 Michael Snyder added support for tracepoints.
333 Stu Grossman wrote gdbserver.
335 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
336 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
338 The following people at the Hewlett-Packard Company contributed
339 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
340 (narrow mode), HP's implementation of kernel threads, HP's aC++
341 compiler, and the terminal user interface: Ben Krepp, Richard Title,
342 John Bishop, Susan Macchia, Kathy Mann, Satish Pai, India Paul, Steve
343 Rehrauer, and Elena Zannoni. Kim Haase provided HP-specific
344 information in this manual.
346 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
347 development since 1991. Cygnus engineers who have worked on @value{GDBN}
348 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
349 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
350 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
351 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
352 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
353 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
354 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
355 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
356 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
357 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
358 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
359 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
360 Zuhn have made contributions both large and small.
364 @chapter A Sample @value{GDBN} Session
366 You can use this manual at your leisure to read all about @value{GDBN}.
367 However, a handful of commands are enough to get started using the
368 debugger. This chapter illustrates those commands.
371 In this sample session, we emphasize user input like this: @b{input},
372 to make it easier to pick out from the surrounding output.
375 @c FIXME: this example may not be appropriate for some configs, where
376 @c FIXME...primary interest is in remote use.
378 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
379 processor) exhibits the following bug: sometimes, when we change its
380 quote strings from the default, the commands used to capture one macro
381 definition within another stop working. In the following short @code{m4}
382 session, we define a macro @code{foo} which expands to @code{0000}; we
383 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
384 same thing. However, when we change the open quote string to
385 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
386 procedure fails to define a new synonym @code{baz}:
395 @b{define(bar,defn(`foo'))}
399 @b{changequote(<QUOTE>,<UNQUOTE>)}
401 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
404 m4: End of input: 0: fatal error: EOF in string
408 Let us use @value{GDBN} to try to see what is going on.
411 $ @b{@value{GDBP} m4}
412 @c FIXME: this falsifies the exact text played out, to permit smallbook
413 @c FIXME... format to come out better.
414 @value{GDBN} is free software and you are welcome to distribute copies
415 of it under certain conditions; type "show copying" to see
417 There is absolutely no warranty for @value{GDBN}; type "show warranty"
420 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
425 @value{GDBN} reads only enough symbol data to know where to find the
426 rest when needed; as a result, the first prompt comes up very quickly.
427 We now tell @value{GDBN} to use a narrower display width than usual, so
428 that examples fit in this manual.
431 (@value{GDBP}) @b{set width 70}
435 We need to see how the @code{m4} built-in @code{changequote} works.
436 Having looked at the source, we know the relevant subroutine is
437 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
438 @code{break} command.
441 (@value{GDBP}) @b{break m4_changequote}
442 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
446 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
447 control; as long as control does not reach the @code{m4_changequote}
448 subroutine, the program runs as usual:
451 (@value{GDBP}) @b{run}
452 Starting program: /work/Editorial/gdb/gnu/m4/m4
460 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
461 suspends execution of @code{m4}, displaying information about the
462 context where it stops.
465 @b{changequote(<QUOTE>,<UNQUOTE>)}
467 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
469 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
473 Now we use the command @code{n} (@code{next}) to advance execution to
474 the next line of the current function.
478 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
483 @code{set_quotes} looks like a promising subroutine. We can go into it
484 by using the command @code{s} (@code{step}) instead of @code{next}.
485 @code{step} goes to the next line to be executed in @emph{any}
486 subroutine, so it steps into @code{set_quotes}.
490 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
492 530 if (lquote != def_lquote)
496 The display that shows the subroutine where @code{m4} is now
497 suspended (and its arguments) is called a stack frame display. It
498 shows a summary of the stack. We can use the @code{backtrace}
499 command (which can also be spelled @code{bt}), to see where we are
500 in the stack as a whole: the @code{backtrace} command displays a
501 stack frame for each active subroutine.
504 (@value{GDBP}) @b{bt}
505 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
507 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
509 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
510 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
512 #4 0x79dc in expand_input () at macro.c:40
513 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
517 We step through a few more lines to see what happens. The first two
518 times, we can use @samp{s}; the next two times we use @code{n} to avoid
519 falling into the @code{xstrdup} subroutine.
523 0x3b5c 532 if (rquote != def_rquote)
525 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
526 def_lquote : xstrdup(lq);
528 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
531 538 len_lquote = strlen(rquote);
535 The last line displayed looks a little odd; we can examine the variables
536 @code{lquote} and @code{rquote} to see if they are in fact the new left
537 and right quotes we specified. We use the command @code{p}
538 (@code{print}) to see their values.
541 (@value{GDBP}) @b{p lquote}
542 $1 = 0x35d40 "<QUOTE>"
543 (@value{GDBP}) @b{p rquote}
544 $2 = 0x35d50 "<UNQUOTE>"
548 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
549 To look at some context, we can display ten lines of source
550 surrounding the current line with the @code{l} (@code{list}) command.
556 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
558 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
561 538 len_lquote = strlen(rquote);
562 539 len_rquote = strlen(lquote);
569 Let us step past the two lines that set @code{len_lquote} and
570 @code{len_rquote}, and then examine the values of those variables.
574 539 len_rquote = strlen(lquote);
577 (@value{GDBP}) @b{p len_lquote}
579 (@value{GDBP}) @b{p len_rquote}
584 That certainly looks wrong, assuming @code{len_lquote} and
585 @code{len_rquote} are meant to be the lengths of @code{lquote} and
586 @code{rquote} respectively. We can set them to better values using
587 the @code{p} command, since it can print the value of
588 any expression---and that expression can include subroutine calls and
592 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
594 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
599 Is that enough to fix the problem of using the new quotes with the
600 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
601 executing with the @code{c} (@code{continue}) command, and then try the
602 example that caused trouble initially:
608 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
615 Success! The new quotes now work just as well as the default ones. The
616 problem seems to have been just the two typos defining the wrong
617 lengths. We allow @code{m4} exit by giving it an EOF as input:
621 Program exited normally.
625 The message @samp{Program exited normally.} is from @value{GDBN}; it
626 indicates @code{m4} has finished executing. We can end our @value{GDBN}
627 session with the @value{GDBN} @code{quit} command.
630 (@value{GDBP}) @b{quit}
634 @chapter Getting In and Out of @value{GDBN}
636 This chapter discusses how to start @value{GDBN}, and how to get out of it.
640 type @samp{@value{GDBP}} to start @value{GDBN}.
642 type @kbd{quit} or @kbd{C-d} to exit.
646 * Invoking GDB:: How to start @value{GDBN}
647 * Quitting GDB:: How to quit @value{GDBN}
648 * Shell Commands:: How to use shell commands inside @value{GDBN}
652 @section Invoking @value{GDBN}
654 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
655 @value{GDBN} reads commands from the terminal until you tell it to exit.
657 You can also run @code{@value{GDBP}} with a variety of arguments and options,
658 to specify more of your debugging environment at the outset.
660 The command-line options described here are designed
661 to cover a variety of situations; in some environments, some of these
662 options may effectively be unavailable.
664 The most usual way to start @value{GDBN} is with one argument,
665 specifying an executable program:
668 @value{GDBP} @var{program}
672 You can also start with both an executable program and a core file
676 @value{GDBP} @var{program} @var{core}
679 You can, instead, specify a process ID as a second argument, if you want
680 to debug a running process:
683 @value{GDBP} @var{program} 1234
687 would attach @value{GDBN} to process @code{1234} (unless you also have a file
688 named @file{1234}; @value{GDBN} does check for a core file first).
690 Taking advantage of the second command-line argument requires a fairly
691 complete operating system; when you use @value{GDBN} as a remote
692 debugger attached to a bare board, there may not be any notion of
693 ``process'', and there is often no way to get a core dump. @value{GDBN}
694 will warn you if it is unable to attach or to read core dumps.
696 You can run @code{@value{GDBP}} without printing the front material, which describes
697 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
704 You can further control how @value{GDBN} starts up by using command-line
705 options. @value{GDBN} itself can remind you of the options available.
715 to display all available options and briefly describe their use
716 (@samp{@value{GDBP} -h} is a shorter equivalent).
718 All options and command line arguments you give are processed
719 in sequential order. The order makes a difference when the
720 @samp{-x} option is used.
724 * File Options:: Choosing files
725 * Mode Options:: Choosing modes
729 @subsection Choosing files
731 When @value{GDBN} starts, it reads any arguments other than options as
732 specifying an executable file and core file (or process ID). This is
733 the same as if the arguments were specified by the @samp{-se} and
734 @samp{-c} options respectively. (@value{GDBN} reads the first argument
735 that does not have an associated option flag as equivalent to the
736 @samp{-se} option followed by that argument; and the second argument
737 that does not have an associated option flag, if any, as equivalent to
738 the @samp{-c} option followed by that argument.)
740 If @value{GDBN} has not been configured to included core file support,
741 such as for most embedded targets, then it will complain about a second
742 argument and ignore it.
744 Many options have both long and short forms; both are shown in the
745 following list. @value{GDBN} also recognizes the long forms if you truncate
746 them, so long as enough of the option is present to be unambiguous.
747 (If you prefer, you can flag option arguments with @samp{--} rather
748 than @samp{-}, though we illustrate the more usual convention.)
750 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
751 @c way, both those who look for -foo and --foo in the index, will find
755 @item -symbols @var{file}
757 @cindex @code{--symbols}
759 Read symbol table from file @var{file}.
761 @item -exec @var{file}
763 @cindex @code{--exec}
765 Use file @var{file} as the executable file to execute when appropriate,
766 and for examining pure data in conjunction with a core dump.
770 Read symbol table from file @var{file} and use it as the executable
773 @item -core @var{file}
775 @cindex @code{--core}
777 Use file @var{file} as a core dump to examine.
779 @item -c @var{number}
780 Connect to process ID @var{number}, as with the @code{attach} command
781 (unless there is a file in core-dump format named @var{number}, in which
782 case @samp{-c} specifies that file as a core dump to read).
784 @item -command @var{file}
786 @cindex @code{--command}
788 Execute @value{GDBN} commands from file @var{file}. @xref{Command
789 Files,, Command files}.
791 @item -directory @var{directory}
792 @itemx -d @var{directory}
793 @cindex @code{--directory}
795 Add @var{directory} to the path to search for source files.
799 @cindex @code{--mapped}
801 @emph{Warning: this option depends on operating system facilities that are not
802 supported on all systems.}@*
803 If memory-mapped files are available on your system through the @code{mmap}
804 system call, you can use this option
805 to have @value{GDBN} write the symbols from your
806 program into a reusable file in the current directory. If the program you are debugging is
807 called @file{/tmp/fred}, the mapped symbol file is @file{/tmp/fred.syms}.
808 Future @value{GDBN} debugging sessions notice the presence of this file,
809 and can quickly map in symbol information from it, rather than reading
810 the symbol table from the executable program.
812 The @file{.syms} file is specific to the host machine where @value{GDBN}
813 is run. It holds an exact image of the internal @value{GDBN} symbol
814 table. It cannot be shared across multiple host platforms.
818 @cindex @code{--readnow}
820 Read each symbol file's entire symbol table immediately, rather than
821 the default, which is to read it incrementally as it is needed.
822 This makes startup slower, but makes future operations faster.
826 You typically combine the @code{-mapped} and @code{-readnow} options in
827 order to build a @file{.syms} file that contains complete symbol
828 information. (@xref{Files,,Commands to specify files}, for information
829 on @file{.syms} files.) A simple @value{GDBN} invocation to do nothing
830 but build a @file{.syms} file for future use is:
833 gdb -batch -nx -mapped -readnow programname
837 @subsection Choosing modes
839 You can run @value{GDBN} in various alternative modes---for example, in
840 batch mode or quiet mode.
847 Do not execute commands found in any initialization files (normally
848 called @file{.gdbinit}, or @file{gdb.ini} on PCs). Normally,
849 @value{GDBN} executes the commands in these files after all the command
850 options and arguments have been processed. @xref{Command Files,,Command
856 @cindex @code{--quiet}
857 @cindex @code{--silent}
859 ``Quiet''. Do not print the introductory and copyright messages. These
860 messages are also suppressed in batch mode.
863 @cindex @code{--batch}
864 Run in batch mode. Exit with status @code{0} after processing all the
865 command files specified with @samp{-x} (and all commands from
866 initialization files, if not inhibited with @samp{-n}). Exit with
867 nonzero status if an error occurs in executing the @value{GDBN} commands
868 in the command files.
870 Batch mode may be useful for running @value{GDBN} as a filter, for
871 example to download and run a program on another computer; in order to
872 make this more useful, the message
875 Program exited normally.
879 (which is ordinarily issued whenever a program running under
880 @value{GDBN} control terminates) is not issued when running in batch
885 @cindex @code{--nowindows}
887 ``No windows''. If @value{GDBN} comes with a graphical user interface
888 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
889 interface. If no GUI is available, this option has no effect.
893 @cindex @code{--windows}
895 If @value{GDBN} includes a GUI, then this option requires it to be
898 @item -cd @var{directory}
900 Run @value{GDBN} using @var{directory} as its working directory,
901 instead of the current directory.
905 @cindex @code{--fullname}
907 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
908 subprocess. It tells @value{GDBN} to output the full file name and line
909 number in a standard, recognizable fashion each time a stack frame is
910 displayed (which includes each time your program stops). This
911 recognizable format looks like two @samp{\032} characters, followed by
912 the file name, line number and character position separated by colons,
913 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
914 @samp{\032} characters as a signal to display the source code for the
918 @cindex @code{--epoch}
919 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
920 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
921 routines so as to allow Epoch to display values of expressions in a
924 @item -annotate @var{level}
925 @cindex @code{--annotate}
926 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
927 effect is identical to using @samp{set annotate @var{level}}
928 (@pxref{Annotations}).
929 Annotation level controls how much information does @value{GDBN} print
930 together with its prompt, values of expressions, source lines, and other
931 types of output. Level 0 is the normal, level 1 is for use when
932 @value{GDBN} is run as a subprocess of @sc{gnu} Emacs, level 2 is the
933 maximum annotation suitable for programs that control @value{GDBN}.
936 @cindex @code{--async}
937 Use the asynchronous event loop for the command-line interface.
938 @value{GDBN} processes all events, such as user keyboard input, via a
939 special event loop. This allows @value{GDBN} to accept and process user
940 commands in parallel with the debugged process being
941 run@footnote{@value{GDBN} built with @sc{djgpp} tools for
942 MS-DOS/MS-Windows supports this mode of operation, but the event loop is
943 suspended when the debuggee runs.}, so you don't need to wait for
944 control to return to @value{GDBN} before you type the next command.
945 (@emph{Note:} as of version 5.0, the target side of the asynchronous
946 operation is not yet in place, so @samp{-async} does not work fully
948 @c FIXME: when the target side of the event loop is done, the above NOTE
949 @c should be removed.
951 When the standard input is connected to a terminal device, @value{GDBN}
952 uses the asynchronous event loop by default, unless disabled by the
953 @samp{-noasync} option.
956 @cindex @code{--noasync}
957 Disable the asynchronous event loop for the command-line interface.
959 @item -baud @var{bps}
961 @cindex @code{--baud}
963 Set the line speed (baud rate or bits per second) of any serial
964 interface used by @value{GDBN} for remote debugging.
966 @item -tty @var{device}
967 @itemx -t @var{device}
970 Run using @var{device} for your program's standard input and output.
971 @c FIXME: kingdon thinks there is more to -tty. Investigate.
973 @c resolve the situation of these eventually
975 @c @cindex @code{--tui}
976 @c Use a Terminal User Interface. For information, use your Web browser to
977 @c read the file @file{TUI.html}, which is usually installed in the
978 @c directory @code{/opt/langtools/wdb/doc} on HP-UX systems. Do not use
979 @c this option if you run @value{GDBN} from Emacs (see @pxref{Emacs, ,Using
980 @c @value{GDBN} under @sc{gnu} Emacs}).
983 @c @cindex @code{--xdb}
984 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
985 @c For information, see the file @file{xdb_trans.html}, which is usually
986 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
989 @item -interpreter @var{interp}
990 @cindex @code{--interpreter}
991 Use the interpreter @var{interp} for interface with the controlling
992 program or device. This option is meant to be set by programs which
993 communicate with @value{GDBN} using it as a back end. For example,
994 @samp{--interpreter=mi} causes @value{GDBN} to use the @dfn{gdbmi
996 @c FIXME: There should be an @xref here to the GDB/MI docs, but
997 @c gdbmi.texi doesn't have a single node to reference!
1000 @cindex @code{--write}
1001 Open the executable and core files for both reading and writing. This
1002 is equivalent to the @samp{set write on} command inside @value{GDBN}
1006 @cindex @code{--statistics}
1007 This option causes @value{GDBN} to print statistics about time and
1008 memory usage after it completes each command and returns to the prompt.
1011 @cindex @code{--version}
1012 This option causes @value{GDBN} to print its version number and
1013 no-warranty blurb, and exit.
1018 @section Quitting @value{GDBN}
1019 @cindex exiting @value{GDBN}
1020 @cindex leaving @value{GDBN}
1023 @kindex quit @r{[}@var{expression}@r{]}
1025 @item quit @r{[}@var{expression}@r{]}
1027 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1028 @code{q}), or type an end-of-file character (usually @kbd{C-d}). If you
1029 do not supply @var{expression}, @value{GDBN} will terminate normally;
1030 otherwise it will terminate using the result of @var{expression} as the
1035 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
1036 terminates the action of any @value{GDBN} command that is in progress and
1037 returns to @value{GDBN} command level. It is safe to type the interrupt
1038 character at any time because @value{GDBN} does not allow it to take effect
1039 until a time when it is safe.
1041 If you have been using @value{GDBN} to control an attached process or
1042 device, you can release it with the @code{detach} command
1043 (@pxref{Attach, ,Debugging an already-running process}).
1045 @node Shell Commands
1046 @section Shell commands
1048 If you need to execute occasional shell commands during your
1049 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1050 just use the @code{shell} command.
1054 @cindex shell escape
1055 @item shell @var{command string}
1056 Invoke a standard shell to execute @var{command string}.
1057 If it exists, the environment variable @code{SHELL} determines which
1058 shell to run. Otherwise @value{GDBN} uses the default shell
1059 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1062 The utility @code{make} is often needed in development environments.
1063 You do not have to use the @code{shell} command for this purpose in
1068 @cindex calling make
1069 @item make @var{make-args}
1070 Execute the @code{make} program with the specified
1071 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1075 @chapter @value{GDBN} Commands
1077 You can abbreviate a @value{GDBN} command to the first few letters of the command
1078 name, if that abbreviation is unambiguous; and you can repeat certain
1079 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1080 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1081 show you the alternatives available, if there is more than one possibility).
1084 * Command Syntax:: How to give commands to @value{GDBN}
1085 * Completion:: Command completion
1086 * Help:: How to ask @value{GDBN} for help
1089 @node Command Syntax
1090 @section Command syntax
1092 A @value{GDBN} command is a single line of input. There is no limit on
1093 how long it can be. It starts with a command name, which is followed by
1094 arguments whose meaning depends on the command name. For example, the
1095 command @code{step} accepts an argument which is the number of times to
1096 step, as in @samp{step 5}. You can also use the @code{step} command
1097 with no arguments. Some commands do not allow any arguments.
1099 @cindex abbreviation
1100 @value{GDBN} command names may always be truncated if that abbreviation is
1101 unambiguous. Other possible command abbreviations are listed in the
1102 documentation for individual commands. In some cases, even ambiguous
1103 abbreviations are allowed; for example, @code{s} is specially defined as
1104 equivalent to @code{step} even though there are other commands whose
1105 names start with @code{s}. You can test abbreviations by using them as
1106 arguments to the @code{help} command.
1108 @cindex repeating commands
1110 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1111 repeat the previous command. Certain commands (for example, @code{run})
1112 will not repeat this way; these are commands whose unintentional
1113 repetition might cause trouble and which you are unlikely to want to
1116 The @code{list} and @code{x} commands, when you repeat them with
1117 @key{RET}, construct new arguments rather than repeating
1118 exactly as typed. This permits easy scanning of source or memory.
1120 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1121 output, in a way similar to the common utility @code{more}
1122 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1123 @key{RET} too many in this situation, @value{GDBN} disables command
1124 repetition after any command that generates this sort of display.
1128 Any text from a @kbd{#} to the end of the line is a comment; it does
1129 nothing. This is useful mainly in command files (@pxref{Command
1130 Files,,Command files}).
1133 @section Command completion
1136 @cindex word completion
1137 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1138 only one possibility; it can also show you what the valid possibilities
1139 are for the next word in a command, at any time. This works for @value{GDBN}
1140 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1142 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1143 of a word. If there is only one possibility, @value{GDBN} fills in the
1144 word, and waits for you to finish the command (or press @key{RET} to
1145 enter it). For example, if you type
1147 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1148 @c complete accuracy in these examples; space introduced for clarity.
1149 @c If texinfo enhancements make it unnecessary, it would be nice to
1150 @c replace " @key" by "@key" in the following...
1152 (@value{GDBP}) info bre @key{TAB}
1156 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1157 the only @code{info} subcommand beginning with @samp{bre}:
1160 (@value{GDBP}) info breakpoints
1164 You can either press @key{RET} at this point, to run the @code{info
1165 breakpoints} command, or backspace and enter something else, if
1166 @samp{breakpoints} does not look like the command you expected. (If you
1167 were sure you wanted @code{info breakpoints} in the first place, you
1168 might as well just type @key{RET} immediately after @samp{info bre},
1169 to exploit command abbreviations rather than command completion).
1171 If there is more than one possibility for the next word when you press
1172 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1173 characters and try again, or just press @key{TAB} a second time;
1174 @value{GDBN} displays all the possible completions for that word. For
1175 example, you might want to set a breakpoint on a subroutine whose name
1176 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1177 just sounds the bell. Typing @key{TAB} again displays all the
1178 function names in your program that begin with those characters, for
1182 (@value{GDBP}) b make_ @key{TAB}
1183 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1184 make_a_section_from_file make_environ
1185 make_abs_section make_function_type
1186 make_blockvector make_pointer_type
1187 make_cleanup make_reference_type
1188 make_command make_symbol_completion_list
1189 (@value{GDBP}) b make_
1193 After displaying the available possibilities, @value{GDBN} copies your
1194 partial input (@samp{b make_} in the example) so you can finish the
1197 If you just want to see the list of alternatives in the first place, you
1198 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1199 means @kbd{@key{META} ?}. You can type this either by holding down a
1200 key designated as the @key{META} shift on your keyboard (if there is
1201 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1203 @cindex quotes in commands
1204 @cindex completion of quoted strings
1205 Sometimes the string you need, while logically a ``word'', may contain
1206 parentheses or other characters that @value{GDBN} normally excludes from
1207 its notion of a word. To permit word completion to work in this
1208 situation, you may enclose words in @code{'} (single quote marks) in
1209 @value{GDBN} commands.
1211 The most likely situation where you might need this is in typing the
1212 name of a C++ function. This is because C++ allows function overloading
1213 (multiple definitions of the same function, distinguished by argument
1214 type). For example, when you want to set a breakpoint you may need to
1215 distinguish whether you mean the version of @code{name} that takes an
1216 @code{int} parameter, @code{name(int)}, or the version that takes a
1217 @code{float} parameter, @code{name(float)}. To use the word-completion
1218 facilities in this situation, type a single quote @code{'} at the
1219 beginning of the function name. This alerts @value{GDBN} that it may need to
1220 consider more information than usual when you press @key{TAB} or
1221 @kbd{M-?} to request word completion:
1224 (@value{GDBP}) b 'bubble( @kbd{M-?}
1225 bubble(double,double) bubble(int,int)
1226 (@value{GDBP}) b 'bubble(
1229 In some cases, @value{GDBN} can tell that completing a name requires using
1230 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1231 completing as much as it can) if you do not type the quote in the first
1235 (@value{GDBP}) b bub @key{TAB}
1236 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1237 (@value{GDBP}) b 'bubble(
1241 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1242 you have not yet started typing the argument list when you ask for
1243 completion on an overloaded symbol.
1245 For more information about overloaded functions, see @ref{C plus plus
1246 expressions, ,C++ expressions}. You can use the command @code{set
1247 overload-resolution off} to disable overload resolution;
1248 see @ref{Debugging C plus plus, ,@value{GDBN} features for C++}.
1252 @section Getting help
1253 @cindex online documentation
1256 You can always ask @value{GDBN} itself for information on its commands,
1257 using the command @code{help}.
1263 You can use @code{help} (abbreviated @code{h}) with no arguments to
1264 display a short list of named classes of commands:
1268 List of classes of commands:
1270 aliases -- Aliases of other commands
1271 breakpoints -- Making program stop at certain points
1272 data -- Examining data
1273 files -- Specifying and examining files
1274 internals -- Maintenance commands
1275 obscure -- Obscure features
1276 running -- Running the program
1277 stack -- Examining the stack
1278 status -- Status inquiries
1279 support -- Support facilities
1280 tracepoints -- Tracing of program execution without@*
1281 stopping the program
1282 user-defined -- User-defined commands
1284 Type "help" followed by a class name for a list of
1285 commands in that class.
1286 Type "help" followed by command name for full
1288 Command name abbreviations are allowed if unambiguous.
1291 @c the above line break eliminates huge line overfull...
1293 @item help @var{class}
1294 Using one of the general help classes as an argument, you can get a
1295 list of the individual commands in that class. For example, here is the
1296 help display for the class @code{status}:
1299 (@value{GDBP}) help status
1304 @c Line break in "show" line falsifies real output, but needed
1305 @c to fit in smallbook page size.
1306 info -- Generic command for showing things
1307 about the program being debugged
1308 show -- Generic command for showing things
1311 Type "help" followed by command name for full
1313 Command name abbreviations are allowed if unambiguous.
1317 @item help @var{command}
1318 With a command name as @code{help} argument, @value{GDBN} displays a
1319 short paragraph on how to use that command.
1322 @item complete @var{args}
1323 The @code{complete @var{args}} command lists all the possible completions
1324 for the beginning of a command. Use @var{args} to specify the beginning of the
1325 command you want completed. For example:
1331 @noindent results in:
1342 @noindent This is intended for use by @sc{gnu} Emacs.
1345 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1346 and @code{show} to inquire about the state of your program, or the state
1347 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1348 manual introduces each of them in the appropriate context. The listings
1349 under @code{info} and under @code{show} in the Index point to
1350 all the sub-commands. @xref{Index}.
1357 This command (abbreviated @code{i}) is for describing the state of your
1358 program. For example, you can list the arguments given to your program
1359 with @code{info args}, list the registers currently in use with @code{info
1360 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1361 You can get a complete list of the @code{info} sub-commands with
1362 @w{@code{help info}}.
1366 You can assign the result of an expression to an environment variable with
1367 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1368 @code{set prompt $}.
1372 In contrast to @code{info}, @code{show} is for describing the state of
1373 @value{GDBN} itself.
1374 You can change most of the things you can @code{show}, by using the
1375 related command @code{set}; for example, you can control what number
1376 system is used for displays with @code{set radix}, or simply inquire
1377 which is currently in use with @code{show radix}.
1380 To display all the settable parameters and their current
1381 values, you can use @code{show} with no arguments; you may also use
1382 @code{info set}. Both commands produce the same display.
1383 @c FIXME: "info set" violates the rule that "info" is for state of
1384 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1385 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1389 Here are three miscellaneous @code{show} subcommands, all of which are
1390 exceptional in lacking corresponding @code{set} commands:
1393 @kindex show version
1394 @cindex version number
1396 Show what version of @value{GDBN} is running. You should include this
1397 information in @value{GDBN} bug-reports. If multiple versions of
1398 @value{GDBN} are in use at your site, you may need to determine which
1399 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1400 commands are introduced, and old ones may wither away. Also, many
1401 system vendors ship variant versions of @value{GDBN}, and there are
1402 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1403 The version number is the same as the one announced when you start
1406 @kindex show copying
1408 Display information about permission for copying @value{GDBN}.
1410 @kindex show warranty
1412 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1413 if your version of @value{GDBN} comes with one.
1418 @chapter Running Programs Under @value{GDBN}
1420 When you run a program under @value{GDBN}, you must first generate
1421 debugging information when you compile it.
1423 You may start @value{GDBN} with its arguments, if any, in an environment
1424 of your choice. If you are doing native debugging, you may redirect
1425 your program's input and output, debug an already running process, or
1426 kill a child process.
1429 * Compilation:: Compiling for debugging
1430 * Starting:: Starting your program
1431 * Arguments:: Your program's arguments
1432 * Environment:: Your program's environment
1434 * Working Directory:: Your program's working directory
1435 * Input/Output:: Your program's input and output
1436 * Attach:: Debugging an already-running process
1437 * Kill Process:: Killing the child process
1439 * Threads:: Debugging programs with multiple threads
1440 * Processes:: Debugging programs with multiple processes
1444 @section Compiling for debugging
1446 In order to debug a program effectively, you need to generate
1447 debugging information when you compile it. This debugging information
1448 is stored in the object file; it describes the data type of each
1449 variable or function and the correspondence between source line numbers
1450 and addresses in the executable code.
1452 To request debugging information, specify the @samp{-g} option when you run
1455 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1456 options together. Using those compilers, you cannot generate optimized
1457 executables containing debugging information.
1459 @value{NGCC}, the @sc{gnu} C compiler, supports @samp{-g} with or
1460 without @samp{-O}, making it possible to debug optimized code. We
1461 recommend that you @emph{always} use @samp{-g} whenever you compile a
1462 program. You may think your program is correct, but there is no sense
1463 in pushing your luck.
1465 @cindex optimized code, debugging
1466 @cindex debugging optimized code
1467 When you debug a program compiled with @samp{-g -O}, remember that the
1468 optimizer is rearranging your code; the debugger shows you what is
1469 really there. Do not be too surprised when the execution path does not
1470 exactly match your source file! An extreme example: if you define a
1471 variable, but never use it, @value{GDBN} never sees that
1472 variable---because the compiler optimizes it out of existence.
1474 Some things do not work as well with @samp{-g -O} as with just
1475 @samp{-g}, particularly on machines with instruction scheduling. If in
1476 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1477 please report it to us as a bug (including a test case!).
1479 Older versions of the @sc{gnu} C compiler permitted a variant option
1480 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1481 format; if your @sc{gnu} C compiler has this option, do not use it.
1485 @section Starting your program
1493 Use the @code{run} command to start your program under @value{GDBN}.
1494 You must first specify the program name (except on VxWorks) with an
1495 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1496 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1497 (@pxref{Files, ,Commands to specify files}).
1501 If you are running your program in an execution environment that
1502 supports processes, @code{run} creates an inferior process and makes
1503 that process run your program. (In environments without processes,
1504 @code{run} jumps to the start of your program.)
1506 The execution of a program is affected by certain information it
1507 receives from its superior. @value{GDBN} provides ways to specify this
1508 information, which you must do @emph{before} starting your program. (You
1509 can change it after starting your program, but such changes only affect
1510 your program the next time you start it.) This information may be
1511 divided into four categories:
1514 @item The @emph{arguments.}
1515 Specify the arguments to give your program as the arguments of the
1516 @code{run} command. If a shell is available on your target, the shell
1517 is used to pass the arguments, so that you may use normal conventions
1518 (such as wildcard expansion or variable substitution) in describing
1520 In Unix systems, you can control which shell is used with the
1521 @code{SHELL} environment variable.
1522 @xref{Arguments, ,Your program's arguments}.
1524 @item The @emph{environment.}
1525 Your program normally inherits its environment from @value{GDBN}, but you can
1526 use the @value{GDBN} commands @code{set environment} and @code{unset
1527 environment} to change parts of the environment that affect
1528 your program. @xref{Environment, ,Your program's environment}.
1530 @item The @emph{working directory.}
1531 Your program inherits its working directory from @value{GDBN}. You can set
1532 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1533 @xref{Working Directory, ,Your program's working directory}.
1535 @item The @emph{standard input and output.}
1536 Your program normally uses the same device for standard input and
1537 standard output as @value{GDBN} is using. You can redirect input and output
1538 in the @code{run} command line, or you can use the @code{tty} command to
1539 set a different device for your program.
1540 @xref{Input/Output, ,Your program's input and output}.
1543 @emph{Warning:} While input and output redirection work, you cannot use
1544 pipes to pass the output of the program you are debugging to another
1545 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1549 When you issue the @code{run} command, your program begins to execute
1550 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1551 of how to arrange for your program to stop. Once your program has
1552 stopped, you may call functions in your program, using the @code{print}
1553 or @code{call} commands. @xref{Data, ,Examining Data}.
1555 If the modification time of your symbol file has changed since the last
1556 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1557 table, and reads it again. When it does this, @value{GDBN} tries to retain
1558 your current breakpoints.
1561 @section Your program's arguments
1563 @cindex arguments (to your program)
1564 The arguments to your program can be specified by the arguments of the
1566 They are passed to a shell, which expands wildcard characters and
1567 performs redirection of I/O, and thence to your program. Your
1568 @code{SHELL} environment variable (if it exists) specifies what shell
1569 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
1570 the default shell (@file{/bin/sh} on Unix).
1572 On non-Unix systems, the program is usually invoked directly by
1573 @value{GDBN}, which emulates I/O redirection via the appropriate system
1574 calls, and the wildcard characters are expanded by the startup code of
1575 the program, not by the shell.
1577 @code{run} with no arguments uses the same arguments used by the previous
1578 @code{run}, or those set by the @code{set args} command.
1583 Specify the arguments to be used the next time your program is run. If
1584 @code{set args} has no arguments, @code{run} executes your program
1585 with no arguments. Once you have run your program with arguments,
1586 using @code{set args} before the next @code{run} is the only way to run
1587 it again without arguments.
1591 Show the arguments to give your program when it is started.
1595 @section Your program's environment
1597 @cindex environment (of your program)
1598 The @dfn{environment} consists of a set of environment variables and
1599 their values. Environment variables conventionally record such things as
1600 your user name, your home directory, your terminal type, and your search
1601 path for programs to run. Usually you set up environment variables with
1602 the shell and they are inherited by all the other programs you run. When
1603 debugging, it can be useful to try running your program with a modified
1604 environment without having to start @value{GDBN} over again.
1608 @item path @var{directory}
1609 Add @var{directory} to the front of the @code{PATH} environment variable
1610 (the search path for executables), for both @value{GDBN} and your program.
1611 You may specify several directory names, separated by whitespace or by a
1612 system-dependent separator character (@samp{:} on Unix, @samp{;} on
1613 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
1614 is moved to the front, so it is searched sooner.
1616 You can use the string @samp{$cwd} to refer to whatever is the current
1617 working directory at the time @value{GDBN} searches the path. If you
1618 use @samp{.} instead, it refers to the directory where you executed the
1619 @code{path} command. @value{GDBN} replaces @samp{.} in the
1620 @var{directory} argument (with the current path) before adding
1621 @var{directory} to the search path.
1622 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1623 @c document that, since repeating it would be a no-op.
1627 Display the list of search paths for executables (the @code{PATH}
1628 environment variable).
1630 @kindex show environment
1631 @item show environment @r{[}@var{varname}@r{]}
1632 Print the value of environment variable @var{varname} to be given to
1633 your program when it starts. If you do not supply @var{varname},
1634 print the names and values of all environment variables to be given to
1635 your program. You can abbreviate @code{environment} as @code{env}.
1637 @kindex set environment
1638 @item set environment @var{varname} @r{[}=@var{value}@r{]}
1639 Set environment variable @var{varname} to @var{value}. The value
1640 changes for your program only, not for @value{GDBN} itself. @var{value} may
1641 be any string; the values of environment variables are just strings, and
1642 any interpretation is supplied by your program itself. The @var{value}
1643 parameter is optional; if it is eliminated, the variable is set to a
1645 @c "any string" here does not include leading, trailing
1646 @c blanks. Gnu asks: does anyone care?
1648 For example, this command:
1655 tells the debugged program, when subsequently run, that its user is named
1656 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1657 are not actually required.)
1659 @kindex unset environment
1660 @item unset environment @var{varname}
1661 Remove variable @var{varname} from the environment to be passed to your
1662 program. This is different from @samp{set env @var{varname} =};
1663 @code{unset environment} removes the variable from the environment,
1664 rather than assigning it an empty value.
1667 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
1669 by your @code{SHELL} environment variable if it exists (or
1670 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1671 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1672 @file{.bashrc} for BASH---any variables you set in that file affect
1673 your program. You may wish to move setting of environment variables to
1674 files that are only run when you sign on, such as @file{.login} or
1677 @node Working Directory
1678 @section Your program's working directory
1680 @cindex working directory (of your program)
1681 Each time you start your program with @code{run}, it inherits its
1682 working directory from the current working directory of @value{GDBN}.
1683 The @value{GDBN} working directory is initially whatever it inherited
1684 from its parent process (typically the shell), but you can specify a new
1685 working directory in @value{GDBN} with the @code{cd} command.
1687 The @value{GDBN} working directory also serves as a default for the commands
1688 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1693 @item cd @var{directory}
1694 Set the @value{GDBN} working directory to @var{directory}.
1698 Print the @value{GDBN} working directory.
1702 @section Your program's input and output
1707 By default, the program you run under @value{GDBN} does input and output to
1708 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
1709 to its own terminal modes to interact with you, but it records the terminal
1710 modes your program was using and switches back to them when you continue
1711 running your program.
1714 @kindex info terminal
1716 Displays information recorded by @value{GDBN} about the terminal modes your
1720 You can redirect your program's input and/or output using shell
1721 redirection with the @code{run} command. For example,
1728 starts your program, diverting its output to the file @file{outfile}.
1731 @cindex controlling terminal
1732 Another way to specify where your program should do input and output is
1733 with the @code{tty} command. This command accepts a file name as
1734 argument, and causes this file to be the default for future @code{run}
1735 commands. It also resets the controlling terminal for the child
1736 process, for future @code{run} commands. For example,
1743 directs that processes started with subsequent @code{run} commands
1744 default to do input and output on the terminal @file{/dev/ttyb} and have
1745 that as their controlling terminal.
1747 An explicit redirection in @code{run} overrides the @code{tty} command's
1748 effect on the input/output device, but not its effect on the controlling
1751 When you use the @code{tty} command or redirect input in the @code{run}
1752 command, only the input @emph{for your program} is affected. The input
1753 for @value{GDBN} still comes from your terminal.
1756 @section Debugging an already-running process
1761 @item attach @var{process-id}
1762 This command attaches to a running process---one that was started
1763 outside @value{GDBN}. (@code{info files} shows your active
1764 targets.) The command takes as argument a process ID. The usual way to
1765 find out the process-id of a Unix process is with the @code{ps} utility,
1766 or with the @samp{jobs -l} shell command.
1768 @code{attach} does not repeat if you press @key{RET} a second time after
1769 executing the command.
1772 To use @code{attach}, your program must be running in an environment
1773 which supports processes; for example, @code{attach} does not work for
1774 programs on bare-board targets that lack an operating system. You must
1775 also have permission to send the process a signal.
1777 When you use @code{attach}, the debugger finds the program running in
1778 the process first by looking in the current working directory, then (if
1779 the program is not found) by using the source file search path
1780 (@pxref{Source Path, ,Specifying source directories}). You can also use
1781 the @code{file} command to load the program. @xref{Files, ,Commands to
1784 The first thing @value{GDBN} does after arranging to debug the specified
1785 process is to stop it. You can examine and modify an attached process
1786 with all the @value{GDBN} commands that are ordinarily available when
1787 you start processes with @code{run}. You can insert breakpoints; you
1788 can step and continue; you can modify storage. If you would rather the
1789 process continue running, you may use the @code{continue} command after
1790 attaching @value{GDBN} to the process.
1795 When you have finished debugging the attached process, you can use the
1796 @code{detach} command to release it from @value{GDBN} control. Detaching
1797 the process continues its execution. After the @code{detach} command,
1798 that process and @value{GDBN} become completely independent once more, and you
1799 are ready to @code{attach} another process or start one with @code{run}.
1800 @code{detach} does not repeat if you press @key{RET} again after
1801 executing the command.
1804 If you exit @value{GDBN} or use the @code{run} command while you have an
1805 attached process, you kill that process. By default, @value{GDBN} asks
1806 for confirmation if you try to do either of these things; you can
1807 control whether or not you need to confirm by using the @code{set
1808 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
1812 @section Killing the child process
1817 Kill the child process in which your program is running under @value{GDBN}.
1820 This command is useful if you wish to debug a core dump instead of a
1821 running process. @value{GDBN} ignores any core dump file while your program
1824 On some operating systems, a program cannot be executed outside @value{GDBN}
1825 while you have breakpoints set on it inside @value{GDBN}. You can use the
1826 @code{kill} command in this situation to permit running your program
1827 outside the debugger.
1829 The @code{kill} command is also useful if you wish to recompile and
1830 relink your program, since on many systems it is impossible to modify an
1831 executable file while it is running in a process. In this case, when you
1832 next type @code{run}, @value{GDBN} notices that the file has changed, and
1833 reads the symbol table again (while trying to preserve your current
1834 breakpoint settings).
1837 @section Debugging programs with multiple threads
1839 @cindex threads of execution
1840 @cindex multiple threads
1841 @cindex switching threads
1842 In some operating systems, such as HP-UX and Solaris, a single program
1843 may have more than one @dfn{thread} of execution. The precise semantics
1844 of threads differ from one operating system to another, but in general
1845 the threads of a single program are akin to multiple processes---except
1846 that they share one address space (that is, they can all examine and
1847 modify the same variables). On the other hand, each thread has its own
1848 registers and execution stack, and perhaps private memory.
1850 @value{GDBN} provides these facilities for debugging multi-thread
1854 @item automatic notification of new threads
1855 @item @samp{thread @var{threadno}}, a command to switch among threads
1856 @item @samp{info threads}, a command to inquire about existing threads
1857 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
1858 a command to apply a command to a list of threads
1859 @item thread-specific breakpoints
1863 @emph{Warning:} These facilities are not yet available on every
1864 @value{GDBN} configuration where the operating system supports threads.
1865 If your @value{GDBN} does not support threads, these commands have no
1866 effect. For example, a system without thread support shows no output
1867 from @samp{info threads}, and always rejects the @code{thread} command,
1871 (@value{GDBP}) info threads
1872 (@value{GDBP}) thread 1
1873 Thread ID 1 not known. Use the "info threads" command to
1874 see the IDs of currently known threads.
1876 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
1877 @c doesn't support threads"?
1880 @cindex focus of debugging
1881 @cindex current thread
1882 The @value{GDBN} thread debugging facility allows you to observe all
1883 threads while your program runs---but whenever @value{GDBN} takes
1884 control, one thread in particular is always the focus of debugging.
1885 This thread is called the @dfn{current thread}. Debugging commands show
1886 program information from the perspective of the current thread.
1888 @kindex New @var{systag}
1889 @cindex thread identifier (system)
1890 @c FIXME-implementors!! It would be more helpful if the [New...] message
1891 @c included GDB's numeric thread handle, so you could just go to that
1892 @c thread without first checking `info threads'.
1893 Whenever @value{GDBN} detects a new thread in your program, it displays
1894 the target system's identification for the thread with a message in the
1895 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
1896 whose form varies depending on the particular system. For example, on
1897 LynxOS, you might see
1900 [New process 35 thread 27]
1904 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
1905 the @var{systag} is simply something like @samp{process 368}, with no
1908 @c FIXME!! (1) Does the [New...] message appear even for the very first
1909 @c thread of a program, or does it only appear for the
1910 @c second---i.e., when it becomes obvious we have a multithread
1912 @c (2) *Is* there necessarily a first thread always? Or do some
1913 @c multithread systems permit starting a program with multiple
1914 @c threads ab initio?
1916 @cindex thread number
1917 @cindex thread identifier (GDB)
1918 For debugging purposes, @value{GDBN} associates its own thread
1919 number---always a single integer---with each thread in your program.
1922 @kindex info threads
1924 Display a summary of all threads currently in your
1925 program. @value{GDBN} displays for each thread (in this order):
1928 @item the thread number assigned by @value{GDBN}
1930 @item the target system's thread identifier (@var{systag})
1932 @item the current stack frame summary for that thread
1936 An asterisk @samp{*} to the left of the @value{GDBN} thread number
1937 indicates the current thread.
1941 @c end table here to get a little more width for example
1944 (@value{GDBP}) info threads
1945 3 process 35 thread 27 0x34e5 in sigpause ()
1946 2 process 35 thread 23 0x34e5 in sigpause ()
1947 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
1953 @cindex thread number
1954 @cindex thread identifier (GDB)
1955 For debugging purposes, @value{GDBN} associates its own thread
1956 number---a small integer assigned in thread-creation order---with each
1957 thread in your program.
1959 @kindex New @var{systag}
1960 @cindex thread identifier (system)
1961 @c FIXME-implementors!! It would be more helpful if the [New...] message
1962 @c included GDB's numeric thread handle, so you could just go to that
1963 @c thread without first checking `info threads'.
1964 Whenever @value{GDBN} detects a new thread in your program, it displays
1965 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
1966 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
1967 whose form varies depending on the particular system. For example, on
1971 [New thread 2 (system thread 26594)]
1975 when @value{GDBN} notices a new thread.
1978 @kindex info threads
1980 Display a summary of all threads currently in your
1981 program. @value{GDBN} displays for each thread (in this order):
1984 @item the thread number assigned by @value{GDBN}
1986 @item the target system's thread identifier (@var{systag})
1988 @item the current stack frame summary for that thread
1992 An asterisk @samp{*} to the left of the @value{GDBN} thread number
1993 indicates the current thread.
1997 @c end table here to get a little more width for example
2000 (@value{GDBP}) info threads
2001 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") at quicksort.c:137
2002 2 system thread 26606 0x7b0030d8 in __ksleep () from /usr/lib/libc.2
2003 1 system thread 27905 0x7b003498 in _brk () from /usr/lib/libc.2
2007 @kindex thread @var{threadno}
2008 @item thread @var{threadno}
2009 Make thread number @var{threadno} the current thread. The command
2010 argument @var{threadno} is the internal @value{GDBN} thread number, as
2011 shown in the first field of the @samp{info threads} display.
2012 @value{GDBN} responds by displaying the system identifier of the thread
2013 you selected, and its current stack frame summary:
2016 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2017 (@value{GDBP}) thread 2
2018 [Switching to process 35 thread 23]
2019 0x34e5 in sigpause ()
2023 As with the @samp{[New @dots{}]} message, the form of the text after
2024 @samp{Switching to} depends on your system's conventions for identifying
2027 @kindex thread apply
2028 @item thread apply [@var{threadno}] [@var{all}] @var{args}
2029 The @code{thread apply} command allows you to apply a command to one or
2030 more threads. Specify the numbers of the threads that you want affected
2031 with the command argument @var{threadno}. @var{threadno} is the internal
2032 @value{GDBN} thread number, as shown in the first field of the @samp{info
2033 threads} display. To apply a command to all threads, use
2034 @code{thread apply all} @var{args}.
2037 @cindex automatic thread selection
2038 @cindex switching threads automatically
2039 @cindex threads, automatic switching
2040 Whenever @value{GDBN} stops your program, due to a breakpoint or a
2041 signal, it automatically selects the thread where that breakpoint or
2042 signal happened. @value{GDBN} alerts you to the context switch with a
2043 message of the form @samp{[Switching to @var{systag}]} to identify the
2046 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
2047 more information about how @value{GDBN} behaves when you stop and start
2048 programs with multiple threads.
2050 @xref{Set Watchpoints,,Setting watchpoints}, for information about
2051 watchpoints in programs with multiple threads.
2054 @section Debugging programs with multiple processes
2056 @cindex fork, debugging programs which call
2057 @cindex multiple processes
2058 @cindex processes, multiple
2059 On most systems, @value{GDBN} has no special support for debugging
2060 programs which create additional processes using the @code{fork}
2061 function. When a program forks, @value{GDBN} will continue to debug the
2062 parent process and the child process will run unimpeded. If you have
2063 set a breakpoint in any code which the child then executes, the child
2064 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2065 will cause it to terminate.
2067 However, if you want to debug the child process there is a workaround
2068 which isn't too painful. Put a call to @code{sleep} in the code which
2069 the child process executes after the fork. It may be useful to sleep
2070 only if a certain environment variable is set, or a certain file exists,
2071 so that the delay need not occur when you don't want to run @value{GDBN}
2072 on the child. While the child is sleeping, use the @code{ps} program to
2073 get its process ID. Then tell @value{GDBN} (a new invocation of
2074 @value{GDBN} if you are also debugging the parent process) to attach to
2075 the child process (@pxref{Attach}). From that point on you can debug
2076 the child process just like any other process which you attached to.
2078 On HP-UX (11.x and later only?), @value{GDBN} provides support for
2079 debugging programs that create additional processes using the
2080 @code{fork} or @code{vfork} function.
2082 By default, when a program forks, @value{GDBN} will continue to debug
2083 the parent process and the child process will run unimpeded.
2085 If you want to follow the child process instead of the parent process,
2086 use the command @w{@code{set follow-fork-mode}}.
2089 @kindex set follow-fork-mode
2090 @item set follow-fork-mode @var{mode}
2091 Set the debugger response to a program call of @code{fork} or
2092 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2093 process. The @var{mode} can be:
2097 The original process is debugged after a fork. The child process runs
2098 unimpeded. This is the default.
2101 The new process is debugged after a fork. The parent process runs
2105 The debugger will ask for one of the above choices.
2108 @item show follow-fork-mode
2109 Display the current debugger response to a @code{fork} or @code{vfork} call.
2112 If you ask to debug a child process and a @code{vfork} is followed by an
2113 @code{exec}, @value{GDBN} executes the new target up to the first
2114 breakpoint in the new target. If you have a breakpoint set on
2115 @code{main} in your original program, the breakpoint will also be set on
2116 the child process's @code{main}.
2118 When a child process is spawned by @code{vfork}, you cannot debug the
2119 child or parent until an @code{exec} call completes.
2121 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2122 call executes, the new target restarts. To restart the parent process,
2123 use the @code{file} command with the parent executable name as its
2126 You can use the @code{catch} command to make @value{GDBN} stop whenever
2127 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2128 Catchpoints, ,Setting catchpoints}.
2131 @chapter Stopping and Continuing
2133 The principal purposes of using a debugger are so that you can stop your
2134 program before it terminates; or so that, if your program runs into
2135 trouble, you can investigate and find out why.
2137 Inside @value{GDBN}, your program may stop for any of several reasons,
2138 such as a signal, a breakpoint, or reaching a new line after a
2139 @value{GDBN} command such as @code{step}. You may then examine and
2140 change variables, set new breakpoints or remove old ones, and then
2141 continue execution. Usually, the messages shown by @value{GDBN} provide
2142 ample explanation of the status of your program---but you can also
2143 explicitly request this information at any time.
2146 @kindex info program
2148 Display information about the status of your program: whether it is
2149 running or not, what process it is, and why it stopped.
2153 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2154 * Continuing and Stepping:: Resuming execution
2156 * Thread Stops:: Stopping and starting multi-thread programs
2160 @section Breakpoints, watchpoints, and catchpoints
2163 A @dfn{breakpoint} makes your program stop whenever a certain point in
2164 the program is reached. For each breakpoint, you can add conditions to
2165 control in finer detail whether your program stops. You can set
2166 breakpoints with the @code{break} command and its variants (@pxref{Set
2167 Breaks, ,Setting breakpoints}), to specify the place where your program
2168 should stop by line number, function name or exact address in the
2171 In HP-UX, SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can set
2172 breakpoints in shared libraries before the executable is run. There is
2173 a minor limitation on HP-UX systems: you must wait until the executable
2174 is run in order to set breakpoints in shared library routines that are
2175 not called directly by the program (for example, routines that are
2176 arguments in a @code{pthread_create} call).
2179 @cindex memory tracing
2180 @cindex breakpoint on memory address
2181 @cindex breakpoint on variable modification
2182 A @dfn{watchpoint} is a special breakpoint that stops your program
2183 when the value of an expression changes. You must use a different
2184 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2185 watchpoints}), but aside from that, you can manage a watchpoint like
2186 any other breakpoint: you enable, disable, and delete both breakpoints
2187 and watchpoints using the same commands.
2189 You can arrange to have values from your program displayed automatically
2190 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2194 @cindex breakpoint on events
2195 A @dfn{catchpoint} is another special breakpoint that stops your program
2196 when a certain kind of event occurs, such as the throwing of a C++
2197 exception or the loading of a library. As with watchpoints, you use a
2198 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2199 catchpoints}), but aside from that, you can manage a catchpoint like any
2200 other breakpoint. (To stop when your program receives a signal, use the
2201 @code{handle} command; see @ref{Signals, ,Signals}.)
2203 @cindex breakpoint numbers
2204 @cindex numbers for breakpoints
2205 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2206 catchpoint when you create it; these numbers are successive integers
2207 starting with one. In many of the commands for controlling various
2208 features of breakpoints you use the breakpoint number to say which
2209 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2210 @dfn{disabled}; if disabled, it has no effect on your program until you
2213 @cindex breakpoint ranges
2214 @cindex ranges of breakpoints
2215 Some @value{GDBN} commands accept a range of breakpoints on which to
2216 operate. A breakpoint range is either a single breakpoint number, like
2217 @samp{5}, or two such numbers, in increasing order, separated by a
2218 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
2219 all breakpoint in that range are operated on.
2222 * Set Breaks:: Setting breakpoints
2223 * Set Watchpoints:: Setting watchpoints
2224 * Set Catchpoints:: Setting catchpoints
2225 * Delete Breaks:: Deleting breakpoints
2226 * Disabling:: Disabling breakpoints
2227 * Conditions:: Break conditions
2228 * Break Commands:: Breakpoint command lists
2229 * Breakpoint Menus:: Breakpoint menus
2230 * Error in Breakpoints:: ``Cannot insert breakpoints''
2234 @subsection Setting breakpoints
2236 @c FIXME LMB what does GDB do if no code on line of breakpt?
2237 @c consider in particular declaration with/without initialization.
2239 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2244 @cindex latest breakpoint
2245 Breakpoints are set with the @code{break} command (abbreviated
2246 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2247 number of the breakpoints you've set most recently; see @ref{Convenience
2248 Vars,, Convenience variables}, for a discussion of what you can do with
2249 convenience variables.
2251 You have several ways to say where the breakpoint should go.
2254 @item break @var{function}
2255 Set a breakpoint at entry to function @var{function}.
2256 When using source languages that permit overloading of symbols, such as
2257 C++, @var{function} may refer to more than one possible place to break.
2258 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2260 @item break +@var{offset}
2261 @itemx break -@var{offset}
2262 Set a breakpoint some number of lines forward or back from the position
2263 at which execution stopped in the currently selected @dfn{stack frame}.
2264 (@xref{Frames, ,Frames}, for a description of stack frames.)
2266 @item break @var{linenum}
2267 Set a breakpoint at line @var{linenum} in the current source file.
2268 The current source file is the last file whose source text was printed.
2269 The breakpoint will stop your program just before it executes any of the
2272 @item break @var{filename}:@var{linenum}
2273 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2275 @item break @var{filename}:@var{function}
2276 Set a breakpoint at entry to function @var{function} found in file
2277 @var{filename}. Specifying a file name as well as a function name is
2278 superfluous except when multiple files contain similarly named
2281 @item break *@var{address}
2282 Set a breakpoint at address @var{address}. You can use this to set
2283 breakpoints in parts of your program which do not have debugging
2284 information or source files.
2287 When called without any arguments, @code{break} sets a breakpoint at
2288 the next instruction to be executed in the selected stack frame
2289 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2290 innermost, this makes your program stop as soon as control
2291 returns to that frame. This is similar to the effect of a
2292 @code{finish} command in the frame inside the selected frame---except
2293 that @code{finish} does not leave an active breakpoint. If you use
2294 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2295 the next time it reaches the current location; this may be useful
2298 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2299 least one instruction has been executed. If it did not do this, you
2300 would be unable to proceed past a breakpoint without first disabling the
2301 breakpoint. This rule applies whether or not the breakpoint already
2302 existed when your program stopped.
2304 @item break @dots{} if @var{cond}
2305 Set a breakpoint with condition @var{cond}; evaluate the expression
2306 @var{cond} each time the breakpoint is reached, and stop only if the
2307 value is nonzero---that is, if @var{cond} evaluates as true.
2308 @samp{@dots{}} stands for one of the possible arguments described
2309 above (or no argument) specifying where to break. @xref{Conditions,
2310 ,Break conditions}, for more information on breakpoint conditions.
2313 @item tbreak @var{args}
2314 Set a breakpoint enabled only for one stop. @var{args} are the
2315 same as for the @code{break} command, and the breakpoint is set in the same
2316 way, but the breakpoint is automatically deleted after the first time your
2317 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2320 @item hbreak @var{args}
2321 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2322 @code{break} command and the breakpoint is set in the same way, but the
2323 breakpoint requires hardware support and some target hardware may not
2324 have this support. The main purpose of this is EPROM/ROM code
2325 debugging, so you can set a breakpoint at an instruction without
2326 changing the instruction. This can be used with the new trap-generation
2327 provided by SPARClite DSU and some x86-based targets. These targets
2328 will generate traps when a program accesses some data or instruction
2329 address that is assigned to the debug registers. However the hardware
2330 breakpoint registers can take a limited number of breakpoints. For
2331 example, on the DSU, only two data breakpoints can be set at a time, and
2332 @value{GDBN} will reject this command if more than two are used. Delete
2333 or disable unused hardware breakpoints before setting new ones
2334 (@pxref{Disabling, ,Disabling}). @xref{Conditions, ,Break conditions}.
2337 @item thbreak @var{args}
2338 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2339 are the same as for the @code{hbreak} command and the breakpoint is set in
2340 the same way. However, like the @code{tbreak} command,
2341 the breakpoint is automatically deleted after the
2342 first time your program stops there. Also, like the @code{hbreak}
2343 command, the breakpoint requires hardware support and some target hardware
2344 may not have this support. @xref{Disabling, ,Disabling breakpoints}.
2345 See also @ref{Conditions, ,Break conditions}.
2348 @cindex regular expression
2349 @item rbreak @var{regex}
2350 Set breakpoints on all functions matching the regular expression
2351 @var{regex}. This command sets an unconditional breakpoint on all
2352 matches, printing a list of all breakpoints it set. Once these
2353 breakpoints are set, they are treated just like the breakpoints set with
2354 the @code{break} command. You can delete them, disable them, or make
2355 them conditional the same way as any other breakpoint.
2357 The syntax of the regular expression is the standard one used with tools
2358 like @file{grep}. Note that this is different from the syntax used by
2359 shells, so for instance @code{foo*} matches all functions that include
2360 an @code{fo} followed by zero or more @code{o}s. There is an implicit
2361 @code{.*} leading and trailing the regular expression you supply, so to
2362 match only functions that begin with @code{foo}, use @code{^foo}.
2364 When debugging C++ programs, @code{rbreak} is useful for setting
2365 breakpoints on overloaded functions that are not members of any special
2368 @kindex info breakpoints
2369 @cindex @code{$_} and @code{info breakpoints}
2370 @item info breakpoints @r{[}@var{n}@r{]}
2371 @itemx info break @r{[}@var{n}@r{]}
2372 @itemx info watchpoints @r{[}@var{n}@r{]}
2373 Print a table of all breakpoints, watchpoints, and catchpoints set and
2374 not deleted, with the following columns for each breakpoint:
2377 @item Breakpoint Numbers
2379 Breakpoint, watchpoint, or catchpoint.
2381 Whether the breakpoint is marked to be disabled or deleted when hit.
2382 @item Enabled or Disabled
2383 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2384 that are not enabled.
2386 Where the breakpoint is in your program, as a memory address.
2388 Where the breakpoint is in the source for your program, as a file and
2393 If a breakpoint is conditional, @code{info break} shows the condition on
2394 the line following the affected breakpoint; breakpoint commands, if any,
2395 are listed after that.
2398 @code{info break} with a breakpoint
2399 number @var{n} as argument lists only that breakpoint. The
2400 convenience variable @code{$_} and the default examining-address for
2401 the @code{x} command are set to the address of the last breakpoint
2402 listed (@pxref{Memory, ,Examining memory}).
2405 @code{info break} displays a count of the number of times the breakpoint
2406 has been hit. This is especially useful in conjunction with the
2407 @code{ignore} command. You can ignore a large number of breakpoint
2408 hits, look at the breakpoint info to see how many times the breakpoint
2409 was hit, and then run again, ignoring one less than that number. This
2410 will get you quickly to the last hit of that breakpoint.
2413 @value{GDBN} allows you to set any number of breakpoints at the same place in
2414 your program. There is nothing silly or meaningless about this. When
2415 the breakpoints are conditional, this is even useful
2416 (@pxref{Conditions, ,Break conditions}).
2418 @cindex negative breakpoint numbers
2419 @cindex internal @value{GDBN} breakpoints
2420 @value{GDBN} itself sometimes sets breakpoints in your program for special
2421 purposes, such as proper handling of @code{longjmp} (in C programs).
2422 These internal breakpoints are assigned negative numbers, starting with
2423 @code{-1}; @samp{info breakpoints} does not display them.
2425 You can see these breakpoints with the @value{GDBN} maintenance command
2426 @samp{maint info breakpoints}.
2429 @kindex maint info breakpoints
2430 @item maint info breakpoints
2431 Using the same format as @samp{info breakpoints}, display both the
2432 breakpoints you've set explicitly, and those @value{GDBN} is using for
2433 internal purposes. Internal breakpoints are shown with negative
2434 breakpoint numbers. The type column identifies what kind of breakpoint
2439 Normal, explicitly set breakpoint.
2442 Normal, explicitly set watchpoint.
2445 Internal breakpoint, used to handle correctly stepping through
2446 @code{longjmp} calls.
2448 @item longjmp resume
2449 Internal breakpoint at the target of a @code{longjmp}.
2452 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2455 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2458 Shared library events.
2465 @node Set Watchpoints
2466 @subsection Setting watchpoints
2468 @cindex setting watchpoints
2469 @cindex software watchpoints
2470 @cindex hardware watchpoints
2471 You can use a watchpoint to stop execution whenever the value of an
2472 expression changes, without having to predict a particular place where
2475 Depending on your system, watchpoints may be implemented in software or
2476 hardware. @value{GDBN} does software watchpointing by single-stepping your
2477 program and testing the variable's value each time, which is hundreds of
2478 times slower than normal execution. (But this may still be worth it, to
2479 catch errors where you have no clue what part of your program is the
2482 On some systems, such as HP-UX, Linux and some other x86-based targets,
2483 @value{GDBN} includes support for
2484 hardware watchpoints, which do not slow down the running of your
2489 @item watch @var{expr}
2490 Set a watchpoint for an expression. @value{GDBN} will break when @var{expr}
2491 is written into by the program and its value changes.
2494 @item rwatch @var{expr}
2495 Set a watchpoint that will break when watch @var{expr} is read by the program.
2498 @item awatch @var{expr}
2499 Set a watchpoint that will break when @var{expr} is either read or written into
2502 @kindex info watchpoints
2503 @item info watchpoints
2504 This command prints a list of watchpoints, breakpoints, and catchpoints;
2505 it is the same as @code{info break}.
2508 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
2509 watchpoints execute very quickly, and the debugger reports a change in
2510 value at the exact instruction where the change occurs. If @value{GDBN}
2511 cannot set a hardware watchpoint, it sets a software watchpoint, which
2512 executes more slowly and reports the change in value at the next
2513 statement, not the instruction, after the change occurs.
2515 When you issue the @code{watch} command, @value{GDBN} reports
2518 Hardware watchpoint @var{num}: @var{expr}
2522 if it was able to set a hardware watchpoint.
2524 Currently, the @code{awatch} and @code{rwatch} commands can only set
2525 hardware watchpoints, because accesses to data that don't change the
2526 value of the watched expression cannot be detected without examining
2527 every instruction as it is being executed, and @value{GDBN} does not do
2528 that currently. If @value{GDBN} finds that it is unable to set a
2529 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
2530 will print a message like this:
2533 Expression cannot be implemented with read/access watchpoint.
2536 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
2537 data type of the watched expression is wider than what a hardware
2538 watchpoint on the target machine can handle. For example, some systems
2539 can only watch regions that are up to 4 bytes wide; on such systems you
2540 cannot set hardware watchpoints for an expression that yields a
2541 double-precision floating-point number (which is typically 8 bytes
2542 wide). As a work-around, it might be possible to break the large region
2543 into a series of smaller ones and watch them with separate watchpoints.
2545 If you set too many hardware watchpoints, @value{GDBN} might be unable
2546 to insert all of them when you resume the execution of your program.
2547 Since the precise number of active watchpoints is unknown until such
2548 time as the program is about to be resumed, @value{GDBN} might not be
2549 able to warn you about this when you set the watchpoints, and the
2550 warning will be printed only when the program is resumed:
2553 Hardware watchpoint @var{num}: Could not insert watchpoint
2557 If this happens, delete or disable some of the watchpoints.
2559 The SPARClite DSU will generate traps when a program accesses some data
2560 or instruction address that is assigned to the debug registers. For the
2561 data addresses, DSU facilitates the @code{watch} command. However the
2562 hardware breakpoint registers can only take two data watchpoints, and
2563 both watchpoints must be the same kind. For example, you can set two
2564 watchpoints with @code{watch} commands, two with @code{rwatch} commands,
2565 @strong{or} two with @code{awatch} commands, but you cannot set one
2566 watchpoint with one command and the other with a different command.
2567 @value{GDBN} will reject the command if you try to mix watchpoints.
2568 Delete or disable unused watchpoint commands before setting new ones.
2570 If you call a function interactively using @code{print} or @code{call},
2571 any watchpoints you have set will be inactive until @value{GDBN} reaches another
2572 kind of breakpoint or the call completes.
2574 @value{GDBN} automatically deletes watchpoints that watch local
2575 (automatic) variables, or expressions that involve such variables, when
2576 they go out of scope, that is, when the execution leaves the block in
2577 which these variables were defined. In particular, when the program
2578 being debugged terminates, @emph{all} local variables go out of scope,
2579 and so only watchpoints that watch global variables remain set. If you
2580 rerun the program, you will need to set all such watchpoints again. One
2581 way of doing that would be to set a code breakpoint at the entry to the
2582 @code{main} function and when it breaks, set all the watchpoints.
2585 @cindex watchpoints and threads
2586 @cindex threads and watchpoints
2587 @emph{Warning:} In multi-thread programs, watchpoints have only limited
2588 usefulness. With the current watchpoint implementation, @value{GDBN}
2589 can only watch the value of an expression @emph{in a single thread}. If
2590 you are confident that the expression can only change due to the current
2591 thread's activity (and if you are also confident that no other thread
2592 can become current), then you can use watchpoints as usual. However,
2593 @value{GDBN} may not notice when a non-current thread's activity changes
2596 @c FIXME: this is almost identical to the previous paragraph.
2597 @emph{HP-UX Warning:} In multi-thread programs, software watchpoints
2598 have only limited usefulness. If @value{GDBN} creates a software
2599 watchpoint, it can only watch the value of an expression @emph{in a
2600 single thread}. If you are confident that the expression can only
2601 change due to the current thread's activity (and if you are also
2602 confident that no other thread can become current), then you can use
2603 software watchpoints as usual. However, @value{GDBN} may not notice
2604 when a non-current thread's activity changes the expression. (Hardware
2605 watchpoints, in contrast, watch an expression in all threads.)
2608 @node Set Catchpoints
2609 @subsection Setting catchpoints
2610 @cindex catchpoints, setting
2611 @cindex exception handlers
2612 @cindex event handling
2614 You can use @dfn{catchpoints} to cause the debugger to stop for certain
2615 kinds of program events, such as C++ exceptions or the loading of a
2616 shared library. Use the @code{catch} command to set a catchpoint.
2620 @item catch @var{event}
2621 Stop when @var{event} occurs. @var{event} can be any of the following:
2625 The throwing of a C++ exception.
2629 The catching of a C++ exception.
2633 A call to @code{exec}. This is currently only available for HP-UX.
2637 A call to @code{fork}. This is currently only available for HP-UX.
2641 A call to @code{vfork}. This is currently only available for HP-UX.
2644 @itemx load @var{libname}
2646 The dynamic loading of any shared library, or the loading of the library
2647 @var{libname}. This is currently only available for HP-UX.
2650 @itemx unload @var{libname}
2651 @kindex catch unload
2652 The unloading of any dynamically loaded shared library, or the unloading
2653 of the library @var{libname}. This is currently only available for HP-UX.
2656 @item tcatch @var{event}
2657 Set a catchpoint that is enabled only for one stop. The catchpoint is
2658 automatically deleted after the first time the event is caught.
2662 Use the @code{info break} command to list the current catchpoints.
2664 There are currently some limitations to C++ exception handling
2665 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
2669 If you call a function interactively, @value{GDBN} normally returns
2670 control to you when the function has finished executing. If the call
2671 raises an exception, however, the call may bypass the mechanism that
2672 returns control to you and cause your program either to abort or to
2673 simply continue running until it hits a breakpoint, catches a signal
2674 that @value{GDBN} is listening for, or exits. This is the case even if
2675 you set a catchpoint for the exception; catchpoints on exceptions are
2676 disabled within interactive calls.
2679 You cannot raise an exception interactively.
2682 You cannot install an exception handler interactively.
2685 @cindex raise exceptions
2686 Sometimes @code{catch} is not the best way to debug exception handling:
2687 if you need to know exactly where an exception is raised, it is better to
2688 stop @emph{before} the exception handler is called, since that way you
2689 can see the stack before any unwinding takes place. If you set a
2690 breakpoint in an exception handler instead, it may not be easy to find
2691 out where the exception was raised.
2693 To stop just before an exception handler is called, you need some
2694 knowledge of the implementation. In the case of @sc{gnu} C++, exceptions are
2695 raised by calling a library function named @code{__raise_exception}
2696 which has the following ANSI C interface:
2699 /* @var{addr} is where the exception identifier is stored.
2700 @var{id} is the exception identifier. */
2701 void __raise_exception (void **addr, void *id);
2705 To make the debugger catch all exceptions before any stack
2706 unwinding takes place, set a breakpoint on @code{__raise_exception}
2707 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2709 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2710 that depends on the value of @var{id}, you can stop your program when
2711 a specific exception is raised. You can use multiple conditional
2712 breakpoints to stop your program when any of a number of exceptions are
2717 @subsection Deleting breakpoints
2719 @cindex clearing breakpoints, watchpoints, catchpoints
2720 @cindex deleting breakpoints, watchpoints, catchpoints
2721 It is often necessary to eliminate a breakpoint, watchpoint, or
2722 catchpoint once it has done its job and you no longer want your program
2723 to stop there. This is called @dfn{deleting} the breakpoint. A
2724 breakpoint that has been deleted no longer exists; it is forgotten.
2726 With the @code{clear} command you can delete breakpoints according to
2727 where they are in your program. With the @code{delete} command you can
2728 delete individual breakpoints, watchpoints, or catchpoints by specifying
2729 their breakpoint numbers.
2731 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2732 automatically ignores breakpoints on the first instruction to be executed
2733 when you continue execution without changing the execution address.
2738 Delete any breakpoints at the next instruction to be executed in the
2739 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2740 the innermost frame is selected, this is a good way to delete a
2741 breakpoint where your program just stopped.
2743 @item clear @var{function}
2744 @itemx clear @var{filename}:@var{function}
2745 Delete any breakpoints set at entry to the function @var{function}.
2747 @item clear @var{linenum}
2748 @itemx clear @var{filename}:@var{linenum}
2749 Delete any breakpoints set at or within the code of the specified line.
2751 @cindex delete breakpoints
2754 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2755 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
2756 ranges specified as arguments. If no argument is specified, delete all
2757 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
2758 confirm off}). You can abbreviate this command as @code{d}.
2762 @subsection Disabling breakpoints
2764 @kindex disable breakpoints
2765 @kindex enable breakpoints
2766 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
2767 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
2768 it had been deleted, but remembers the information on the breakpoint so
2769 that you can @dfn{enable} it again later.
2771 You disable and enable breakpoints, watchpoints, and catchpoints with
2772 the @code{enable} and @code{disable} commands, optionally specifying one
2773 or more breakpoint numbers as arguments. Use @code{info break} or
2774 @code{info watch} to print a list of breakpoints, watchpoints, and
2775 catchpoints if you do not know which numbers to use.
2777 A breakpoint, watchpoint, or catchpoint can have any of four different
2778 states of enablement:
2782 Enabled. The breakpoint stops your program. A breakpoint set
2783 with the @code{break} command starts out in this state.
2785 Disabled. The breakpoint has no effect on your program.
2787 Enabled once. The breakpoint stops your program, but then becomes
2790 Enabled for deletion. The breakpoint stops your program, but
2791 immediately after it does so it is deleted permanently. A breakpoint
2792 set with the @code{tbreak} command starts out in this state.
2795 You can use the following commands to enable or disable breakpoints,
2796 watchpoints, and catchpoints:
2799 @kindex disable breakpoints
2802 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2803 Disable the specified breakpoints---or all breakpoints, if none are
2804 listed. A disabled breakpoint has no effect but is not forgotten. All
2805 options such as ignore-counts, conditions and commands are remembered in
2806 case the breakpoint is enabled again later. You may abbreviate
2807 @code{disable} as @code{dis}.
2809 @kindex enable breakpoints
2811 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2812 Enable the specified breakpoints (or all defined breakpoints). They
2813 become effective once again in stopping your program.
2815 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
2816 Enable the specified breakpoints temporarily. @value{GDBN} disables any
2817 of these breakpoints immediately after stopping your program.
2819 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
2820 Enable the specified breakpoints to work once, then die. @value{GDBN}
2821 deletes any of these breakpoints as soon as your program stops there.
2824 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
2825 @c confusing: tbreak is also initially enabled.
2826 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2827 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2828 subsequently, they become disabled or enabled only when you use one of
2829 the commands above. (The command @code{until} can set and delete a
2830 breakpoint of its own, but it does not change the state of your other
2831 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2835 @subsection Break conditions
2836 @cindex conditional breakpoints
2837 @cindex breakpoint conditions
2839 @c FIXME what is scope of break condition expr? Context where wanted?
2840 @c in particular for a watchpoint?
2841 The simplest sort of breakpoint breaks every time your program reaches a
2842 specified place. You can also specify a @dfn{condition} for a
2843 breakpoint. A condition is just a Boolean expression in your
2844 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2845 a condition evaluates the expression each time your program reaches it,
2846 and your program stops only if the condition is @emph{true}.
2848 This is the converse of using assertions for program validation; in that
2849 situation, you want to stop when the assertion is violated---that is,
2850 when the condition is false. In C, if you want to test an assertion expressed
2851 by the condition @var{assert}, you should set the condition
2852 @samp{! @var{assert}} on the appropriate breakpoint.
2854 Conditions are also accepted for watchpoints; you may not need them,
2855 since a watchpoint is inspecting the value of an expression anyhow---but
2856 it might be simpler, say, to just set a watchpoint on a variable name,
2857 and specify a condition that tests whether the new value is an interesting
2860 Break conditions can have side effects, and may even call functions in
2861 your program. This can be useful, for example, to activate functions
2862 that log program progress, or to use your own print functions to
2863 format special data structures. The effects are completely predictable
2864 unless there is another enabled breakpoint at the same address. (In
2865 that case, @value{GDBN} might see the other breakpoint first and stop your
2866 program without checking the condition of this one.) Note that
2867 breakpoint commands are usually more convenient and flexible than break
2869 purpose of performing side effects when a breakpoint is reached
2870 (@pxref{Break Commands, ,Breakpoint command lists}).
2872 Break conditions can be specified when a breakpoint is set, by using
2873 @samp{if} in the arguments to the @code{break} command. @xref{Set
2874 Breaks, ,Setting breakpoints}. They can also be changed at any time
2875 with the @code{condition} command.
2877 You can also use the @code{if} keyword with the @code{watch} command.
2878 The @code{catch} command does not recognize the @code{if} keyword;
2879 @code{condition} is the only way to impose a further condition on a
2884 @item condition @var{bnum} @var{expression}
2885 Specify @var{expression} as the break condition for breakpoint,
2886 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
2887 breakpoint @var{bnum} stops your program only if the value of
2888 @var{expression} is true (nonzero, in C). When you use
2889 @code{condition}, @value{GDBN} checks @var{expression} immediately for
2890 syntactic correctness, and to determine whether symbols in it have
2891 referents in the context of your breakpoint. If @var{expression} uses
2892 symbols not referenced in the context of the breakpoint, @value{GDBN}
2893 prints an error message:
2896 No symbol "foo" in current context.
2901 not actually evaluate @var{expression} at the time the @code{condition}
2902 command (or a command that sets a breakpoint with a condition, like
2903 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
2905 @item condition @var{bnum}
2906 Remove the condition from breakpoint number @var{bnum}. It becomes
2907 an ordinary unconditional breakpoint.
2910 @cindex ignore count (of breakpoint)
2911 A special case of a breakpoint condition is to stop only when the
2912 breakpoint has been reached a certain number of times. This is so
2913 useful that there is a special way to do it, using the @dfn{ignore
2914 count} of the breakpoint. Every breakpoint has an ignore count, which
2915 is an integer. Most of the time, the ignore count is zero, and
2916 therefore has no effect. But if your program reaches a breakpoint whose
2917 ignore count is positive, then instead of stopping, it just decrements
2918 the ignore count by one and continues. As a result, if the ignore count
2919 value is @var{n}, the breakpoint does not stop the next @var{n} times
2920 your program reaches it.
2924 @item ignore @var{bnum} @var{count}
2925 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
2926 The next @var{count} times the breakpoint is reached, your program's
2927 execution does not stop; other than to decrement the ignore count, @value{GDBN}
2930 To make the breakpoint stop the next time it is reached, specify
2933 When you use @code{continue} to resume execution of your program from a
2934 breakpoint, you can specify an ignore count directly as an argument to
2935 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
2936 Stepping,,Continuing and stepping}.
2938 If a breakpoint has a positive ignore count and a condition, the
2939 condition is not checked. Once the ignore count reaches zero,
2940 @value{GDBN} resumes checking the condition.
2942 You could achieve the effect of the ignore count with a condition such
2943 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
2944 is decremented each time. @xref{Convenience Vars, ,Convenience
2948 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
2951 @node Break Commands
2952 @subsection Breakpoint command lists
2954 @cindex breakpoint commands
2955 You can give any breakpoint (or watchpoint or catchpoint) a series of
2956 commands to execute when your program stops due to that breakpoint. For
2957 example, you might want to print the values of certain expressions, or
2958 enable other breakpoints.
2963 @item commands @r{[}@var{bnum}@r{]}
2964 @itemx @dots{} @var{command-list} @dots{}
2966 Specify a list of commands for breakpoint number @var{bnum}. The commands
2967 themselves appear on the following lines. Type a line containing just
2968 @code{end} to terminate the commands.
2970 To remove all commands from a breakpoint, type @code{commands} and
2971 follow it immediately with @code{end}; that is, give no commands.
2973 With no @var{bnum} argument, @code{commands} refers to the last
2974 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
2975 recently encountered).
2978 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
2979 disabled within a @var{command-list}.
2981 You can use breakpoint commands to start your program up again. Simply
2982 use the @code{continue} command, or @code{step}, or any other command
2983 that resumes execution.
2985 Any other commands in the command list, after a command that resumes
2986 execution, are ignored. This is because any time you resume execution
2987 (even with a simple @code{next} or @code{step}), you may encounter
2988 another breakpoint---which could have its own command list, leading to
2989 ambiguities about which list to execute.
2992 If the first command you specify in a command list is @code{silent}, the
2993 usual message about stopping at a breakpoint is not printed. This may
2994 be desirable for breakpoints that are to print a specific message and
2995 then continue. If none of the remaining commands print anything, you
2996 see no sign that the breakpoint was reached. @code{silent} is
2997 meaningful only at the beginning of a breakpoint command list.
2999 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3000 print precisely controlled output, and are often useful in silent
3001 breakpoints. @xref{Output, ,Commands for controlled output}.
3003 For example, here is how you could use breakpoint commands to print the
3004 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
3010 printf "x is %d\n",x
3015 One application for breakpoint commands is to compensate for one bug so
3016 you can test for another. Put a breakpoint just after the erroneous line
3017 of code, give it a condition to detect the case in which something
3018 erroneous has been done, and give it commands to assign correct values
3019 to any variables that need them. End with the @code{continue} command
3020 so that your program does not stop, and start with the @code{silent}
3021 command so that no output is produced. Here is an example:
3032 @node Breakpoint Menus
3033 @subsection Breakpoint menus
3035 @cindex symbol overloading
3037 Some programming languages (notably C++) permit a single function name
3038 to be defined several times, for application in different contexts.
3039 This is called @dfn{overloading}. When a function name is overloaded,
3040 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
3041 a breakpoint. If you realize this is a problem, you can use
3042 something like @samp{break @var{function}(@var{types})} to specify which
3043 particular version of the function you want. Otherwise, @value{GDBN} offers
3044 you a menu of numbered choices for different possible breakpoints, and
3045 waits for your selection with the prompt @samp{>}. The first two
3046 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
3047 sets a breakpoint at each definition of @var{function}, and typing
3048 @kbd{0} aborts the @code{break} command without setting any new
3051 For example, the following session excerpt shows an attempt to set a
3052 breakpoint at the overloaded symbol @code{String::after}.
3053 We choose three particular definitions of that function name:
3055 @c FIXME! This is likely to change to show arg type lists, at least
3058 (@value{GDBP}) b String::after
3061 [2] file:String.cc; line number:867
3062 [3] file:String.cc; line number:860
3063 [4] file:String.cc; line number:875
3064 [5] file:String.cc; line number:853
3065 [6] file:String.cc; line number:846
3066 [7] file:String.cc; line number:735
3068 Breakpoint 1 at 0xb26c: file String.cc, line 867.
3069 Breakpoint 2 at 0xb344: file String.cc, line 875.
3070 Breakpoint 3 at 0xafcc: file String.cc, line 846.
3071 Multiple breakpoints were set.
3072 Use the "delete" command to delete unwanted
3078 @c @ifclear BARETARGET
3079 @node Error in Breakpoints
3080 @subsection ``Cannot insert breakpoints''
3082 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
3084 Under some operating systems, breakpoints cannot be used in a program if
3085 any other process is running that program. In this situation,
3086 attempting to run or continue a program with a breakpoint causes
3087 @value{GDBN} to print an error message:
3090 Cannot insert breakpoints.
3091 The same program may be running in another process.
3094 When this happens, you have three ways to proceed:
3098 Remove or disable the breakpoints, then continue.
3101 Suspend @value{GDBN}, and copy the file containing your program to a new
3102 name. Resume @value{GDBN} and use the @code{exec-file} command to specify
3103 that @value{GDBN} should run your program under that name.
3104 Then start your program again.
3107 Relink your program so that the text segment is nonsharable, using the
3108 linker option @samp{-N}. The operating system limitation may not apply
3109 to nonsharable executables.
3113 A similar message can be printed if you request too many active
3114 hardware-assisted breakpoints and watchpoints:
3116 @c FIXME: the precise wording of this message may change; the relevant
3117 @c source change is not committed yet (Sep 3, 1999).
3119 Stopped; cannot insert breakpoints.
3120 You may have requested too many hardware breakpoints and watchpoints.
3124 This message is printed when you attempt to resume the program, since
3125 only then @value{GDBN} knows exactly how many hardware breakpoints and
3126 watchpoints it needs to insert.
3128 When this message is printed, you need to disable or remove some of the
3129 hardware-assisted breakpoints and watchpoints, and then continue.
3132 @node Continuing and Stepping
3133 @section Continuing and stepping
3137 @cindex resuming execution
3138 @dfn{Continuing} means resuming program execution until your program
3139 completes normally. In contrast, @dfn{stepping} means executing just
3140 one more ``step'' of your program, where ``step'' may mean either one
3141 line of source code, or one machine instruction (depending on what
3142 particular command you use). Either when continuing or when stepping,
3143 your program may stop even sooner, due to a breakpoint or a signal. (If
3144 it stops due to a signal, you may want to use @code{handle}, or use
3145 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
3151 @item continue @r{[}@var{ignore-count}@r{]}
3152 @itemx c @r{[}@var{ignore-count}@r{]}
3153 @itemx fg @r{[}@var{ignore-count}@r{]}
3154 Resume program execution, at the address where your program last stopped;
3155 any breakpoints set at that address are bypassed. The optional argument
3156 @var{ignore-count} allows you to specify a further number of times to
3157 ignore a breakpoint at this location; its effect is like that of
3158 @code{ignore} (@pxref{Conditions, ,Break conditions}).
3160 The argument @var{ignore-count} is meaningful only when your program
3161 stopped due to a breakpoint. At other times, the argument to
3162 @code{continue} is ignored.
3164 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
3165 debugged program is deemed to be the foreground program) are provided
3166 purely for convenience, and have exactly the same behavior as
3170 To resume execution at a different place, you can use @code{return}
3171 (@pxref{Returning, ,Returning from a function}) to go back to the
3172 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
3173 different address}) to go to an arbitrary location in your program.
3175 A typical technique for using stepping is to set a breakpoint
3176 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and catchpoints}) at the
3177 beginning of the function or the section of your program where a problem
3178 is believed to lie, run your program until it stops at that breakpoint,
3179 and then step through the suspect area, examining the variables that are
3180 interesting, until you see the problem happen.
3186 Continue running your program until control reaches a different source
3187 line, then stop it and return control to @value{GDBN}. This command is
3188 abbreviated @code{s}.
3191 @c "without debugging information" is imprecise; actually "without line
3192 @c numbers in the debugging information". (gcc -g1 has debugging info but
3193 @c not line numbers). But it seems complex to try to make that
3194 @c distinction here.
3195 @emph{Warning:} If you use the @code{step} command while control is
3196 within a function that was compiled without debugging information,
3197 execution proceeds until control reaches a function that does have
3198 debugging information. Likewise, it will not step into a function which
3199 is compiled without debugging information. To step through functions
3200 without debugging information, use the @code{stepi} command, described
3204 The @code{step} command only stops at the first instruction of a
3205 source line. This prevents the multiple stops that could otherwise occur in
3206 switch statements, for loops, etc. @code{step} continues to stop if a
3207 function that has debugging information is called within the line.
3208 In other words, @code{step} @emph{steps inside} any functions called
3211 Also, the @code{step} command only enters a function if there is line
3212 number information for the function. Otherwise it acts like the
3213 @code{next} command. This avoids problems when using @code{cc -gl}
3214 on MIPS machines. Previously, @code{step} entered subroutines if there
3215 was any debugging information about the routine.
3217 @item step @var{count}
3218 Continue running as in @code{step}, but do so @var{count} times. If a
3219 breakpoint is reached, or a signal not related to stepping occurs before
3220 @var{count} steps, stepping stops right away.
3224 @item next @r{[}@var{count}@r{]}
3225 Continue to the next source line in the current (innermost) stack frame.
3226 This is similar to @code{step}, but function calls that appear within
3227 the line of code are executed without stopping. Execution stops when
3228 control reaches a different line of code at the original stack level
3229 that was executing when you gave the @code{next} command. This command
3230 is abbreviated @code{n}.
3232 An argument @var{count} is a repeat count, as for @code{step}.
3235 @c FIX ME!! Do we delete this, or is there a way it fits in with
3236 @c the following paragraph? --- Vctoria
3238 @c @code{next} within a function that lacks debugging information acts like
3239 @c @code{step}, but any function calls appearing within the code of the
3240 @c function are executed without stopping.
3242 The @code{next} command only stops at the first instruction of a
3243 source line. This prevents multiple stops that could otherwise occur in
3244 switch statements, for loops, etc.
3248 Continue running until just after function in the selected stack frame
3249 returns. Print the returned value (if any).
3251 Contrast this with the @code{return} command (@pxref{Returning,
3252 ,Returning from a function}).
3258 Continue running until a source line past the current line, in the
3259 current stack frame, is reached. This command is used to avoid single
3260 stepping through a loop more than once. It is like the @code{next}
3261 command, except that when @code{until} encounters a jump, it
3262 automatically continues execution until the program counter is greater
3263 than the address of the jump.
3265 This means that when you reach the end of a loop after single stepping
3266 though it, @code{until} makes your program continue execution until it
3267 exits the loop. In contrast, a @code{next} command at the end of a loop
3268 simply steps back to the beginning of the loop, which forces you to step
3269 through the next iteration.
3271 @code{until} always stops your program if it attempts to exit the current
3274 @code{until} may produce somewhat counterintuitive results if the order
3275 of machine code does not match the order of the source lines. For
3276 example, in the following excerpt from a debugging session, the @code{f}
3277 (@code{frame}) command shows that execution is stopped at line
3278 @code{206}; yet when we use @code{until}, we get to line @code{195}:
3282 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
3284 (@value{GDBP}) until
3285 195 for ( ; argc > 0; NEXTARG) @{
3288 This happened because, for execution efficiency, the compiler had
3289 generated code for the loop closure test at the end, rather than the
3290 start, of the loop---even though the test in a C @code{for}-loop is
3291 written before the body of the loop. The @code{until} command appeared
3292 to step back to the beginning of the loop when it advanced to this
3293 expression; however, it has not really gone to an earlier
3294 statement---not in terms of the actual machine code.
3296 @code{until} with no argument works by means of single
3297 instruction stepping, and hence is slower than @code{until} with an
3300 @item until @var{location}
3301 @itemx u @var{location}
3302 Continue running your program until either the specified location is
3303 reached, or the current stack frame returns. @var{location} is any of
3304 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
3305 ,Setting breakpoints}). This form of the command uses breakpoints,
3306 and hence is quicker than @code{until} without an argument.
3311 @itemx stepi @var{arg}
3313 Execute one machine instruction, then stop and return to the debugger.
3315 It is often useful to do @samp{display/i $pc} when stepping by machine
3316 instructions. This makes @value{GDBN} automatically display the next
3317 instruction to be executed, each time your program stops. @xref{Auto
3318 Display,, Automatic display}.
3320 An argument is a repeat count, as in @code{step}.
3326 @itemx nexti @var{arg}
3328 Execute one machine instruction, but if it is a function call,
3329 proceed until the function returns.
3331 An argument is a repeat count, as in @code{next}.
3338 A signal is an asynchronous event that can happen in a program. The
3339 operating system defines the possible kinds of signals, and gives each
3340 kind a name and a number. For example, in Unix @code{SIGINT} is the
3341 signal a program gets when you type an interrupt character (often @kbd{C-c});
3342 @code{SIGSEGV} is the signal a program gets from referencing a place in
3343 memory far away from all the areas in use; @code{SIGALRM} occurs when
3344 the alarm clock timer goes off (which happens only if your program has
3345 requested an alarm).
3347 @cindex fatal signals
3348 Some signals, including @code{SIGALRM}, are a normal part of the
3349 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3350 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
3351 program has not specified in advance some other way to handle the signal.
3352 @code{SIGINT} does not indicate an error in your program, but it is normally
3353 fatal so it can carry out the purpose of the interrupt: to kill the program.
3355 @value{GDBN} has the ability to detect any occurrence of a signal in your
3356 program. You can tell @value{GDBN} in advance what to do for each kind of
3359 @cindex handling signals
3360 Normally, @value{GDBN} is set up to ignore non-erroneous signals like @code{SIGALRM}
3361 (so as not to interfere with their role in the functioning of your program)
3362 but to stop your program immediately whenever an error signal happens.
3363 You can change these settings with the @code{handle} command.
3366 @kindex info signals
3369 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3370 handle each one. You can use this to see the signal numbers of all
3371 the defined types of signals.
3373 @code{info handle} is an alias for @code{info signals}.
3376 @item handle @var{signal} @var{keywords}@dots{}
3377 Change the way @value{GDBN} handles signal @var{signal}. @var{signal} can
3378 be the number of a signal or its name (with or without the @samp{SIG} at the
3379 beginning). The @var{keywords} say what change to make.
3383 The keywords allowed by the @code{handle} command can be abbreviated.
3384 Their full names are:
3388 @value{GDBN} should not stop your program when this signal happens. It may
3389 still print a message telling you that the signal has come in.
3392 @value{GDBN} should stop your program when this signal happens. This implies
3393 the @code{print} keyword as well.
3396 @value{GDBN} should print a message when this signal happens.
3399 @value{GDBN} should not mention the occurrence of the signal at all. This
3400 implies the @code{nostop} keyword as well.
3403 @value{GDBN} should allow your program to see this signal; your program
3404 can handle the signal, or else it may terminate if the signal is fatal
3408 @value{GDBN} should not allow your program to see this signal.
3412 When a signal stops your program, the signal is not visible to the
3414 continue. Your program sees the signal then, if @code{pass} is in
3415 effect for the signal in question @emph{at that time}. In other words,
3416 after @value{GDBN} reports a signal, you can use the @code{handle}
3417 command with @code{pass} or @code{nopass} to control whether your
3418 program sees that signal when you continue.
3420 You can also use the @code{signal} command to prevent your program from
3421 seeing a signal, or cause it to see a signal it normally would not see,
3422 or to give it any signal at any time. For example, if your program stopped
3423 due to some sort of memory reference error, you might store correct
3424 values into the erroneous variables and continue, hoping to see more
3425 execution; but your program would probably terminate immediately as
3426 a result of the fatal signal once it saw the signal. To prevent this,
3427 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3431 @section Stopping and starting multi-thread programs
3433 When your program has multiple threads (@pxref{Threads,, Debugging
3434 programs with multiple threads}), you can choose whether to set
3435 breakpoints on all threads, or on a particular thread.
3438 @cindex breakpoints and threads
3439 @cindex thread breakpoints
3440 @kindex break @dots{} thread @var{threadno}
3441 @item break @var{linespec} thread @var{threadno}
3442 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3443 @var{linespec} specifies source lines; there are several ways of
3444 writing them, but the effect is always to specify some source line.
3446 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3447 to specify that you only want @value{GDBN} to stop the program when a
3448 particular thread reaches this breakpoint. @var{threadno} is one of the
3449 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3450 column of the @samp{info threads} display.
3452 If you do not specify @samp{thread @var{threadno}} when you set a
3453 breakpoint, the breakpoint applies to @emph{all} threads of your
3456 You can use the @code{thread} qualifier on conditional breakpoints as
3457 well; in this case, place @samp{thread @var{threadno}} before the
3458 breakpoint condition, like this:
3461 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
3466 @cindex stopped threads
3467 @cindex threads, stopped
3468 Whenever your program stops under @value{GDBN} for any reason,
3469 @emph{all} threads of execution stop, not just the current thread. This
3470 allows you to examine the overall state of the program, including
3471 switching between threads, without worrying that things may change
3474 @cindex continuing threads
3475 @cindex threads, continuing
3476 Conversely, whenever you restart the program, @emph{all} threads start
3477 executing. @emph{This is true even when single-stepping} with commands
3478 like @code{step} or @code{next}.
3480 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3481 Since thread scheduling is up to your debugging target's operating
3482 system (not controlled by @value{GDBN}), other threads may
3483 execute more than one statement while the current thread completes a
3484 single step. Moreover, in general other threads stop in the middle of a
3485 statement, rather than at a clean statement boundary, when the program
3488 You might even find your program stopped in another thread after
3489 continuing or even single-stepping. This happens whenever some other
3490 thread runs into a breakpoint, a signal, or an exception before the
3491 first thread completes whatever you requested.
3493 On some OSes, you can lock the OS scheduler and thus allow only a single
3497 @item set scheduler-locking @var{mode}
3498 Set the scheduler locking mode. If it is @code{off}, then there is no
3499 locking and any thread may run at any time. If @code{on}, then only the
3500 current thread may run when the inferior is resumed. The @code{step}
3501 mode optimizes for single-stepping. It stops other threads from
3502 ``seizing the prompt'' by preempting the current thread while you are
3503 stepping. Other threads will only rarely (or never) get a chance to run
3504 when you step. They are more likely to run when you @samp{next} over a
3505 function call, and they are completely free to run when you use commands
3506 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
3507 thread hits a breakpoint during its timeslice, they will never steal the
3508 @value{GDBN} prompt away from the thread that you are debugging.
3510 @item show scheduler-locking
3511 Display the current scheduler locking mode.
3516 @chapter Examining the Stack
3518 When your program has stopped, the first thing you need to know is where it
3519 stopped and how it got there.
3522 Each time your program performs a function call, information about the call
3524 That information includes the location of the call in your program,
3525 the arguments of the call,
3526 and the local variables of the function being called.
3527 The information is saved in a block of data called a @dfn{stack frame}.
3528 The stack frames are allocated in a region of memory called the @dfn{call
3531 When your program stops, the @value{GDBN} commands for examining the
3532 stack allow you to see all of this information.
3534 @cindex selected frame
3535 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3536 @value{GDBN} commands refer implicitly to the selected frame. In
3537 particular, whenever you ask @value{GDBN} for the value of a variable in
3538 your program, the value is found in the selected frame. There are
3539 special @value{GDBN} commands to select whichever frame you are
3540 interested in. @xref{Selection, ,Selecting a frame}.
3542 When your program stops, @value{GDBN} automatically selects the
3543 currently executing frame and describes it briefly, similar to the
3544 @code{frame} command (@pxref{Frame Info, ,Information about a frame}).
3547 * Frames:: Stack frames
3548 * Backtrace:: Backtraces
3549 * Selection:: Selecting a frame
3550 * Frame Info:: Information on a frame
3555 @section Stack frames
3557 @cindex frame, definition
3559 The call stack is divided up into contiguous pieces called @dfn{stack
3560 frames}, or @dfn{frames} for short; each frame is the data associated
3561 with one call to one function. The frame contains the arguments given
3562 to the function, the function's local variables, and the address at
3563 which the function is executing.
3565 @cindex initial frame
3566 @cindex outermost frame
3567 @cindex innermost frame
3568 When your program is started, the stack has only one frame, that of the
3569 function @code{main}. This is called the @dfn{initial} frame or the
3570 @dfn{outermost} frame. Each time a function is called, a new frame is
3571 made. Each time a function returns, the frame for that function invocation
3572 is eliminated. If a function is recursive, there can be many frames for
3573 the same function. The frame for the function in which execution is
3574 actually occurring is called the @dfn{innermost} frame. This is the most
3575 recently created of all the stack frames that still exist.
3577 @cindex frame pointer
3578 Inside your program, stack frames are identified by their addresses. A
3579 stack frame consists of many bytes, each of which has its own address; each
3580 kind of computer has a convention for choosing one byte whose
3581 address serves as the address of the frame. Usually this address is kept
3582 in a register called the @dfn{frame pointer register} while execution is
3583 going on in that frame.
3585 @cindex frame number
3586 @value{GDBN} assigns numbers to all existing stack frames, starting with
3587 zero for the innermost frame, one for the frame that called it,
3588 and so on upward. These numbers do not really exist in your program;
3589 they are assigned by @value{GDBN} to give you a way of designating stack
3590 frames in @value{GDBN} commands.
3592 @c below produces an acceptable overful hbox. --mew 13aug1993
3593 @cindex frameless execution
3594 Some compilers provide a way to compile functions so that they operate
3595 without stack frames. (For example, the @code{@value{GCC}} option
3596 @samp{-fomit-frame-pointer} generates functions without a frame.)
3597 This is occasionally done with heavily used library functions to save
3598 the frame setup time. @value{GDBN} has limited facilities for dealing
3599 with these function invocations. If the innermost function invocation
3600 has no stack frame, @value{GDBN} nevertheless regards it as though
3601 it had a separate frame, which is numbered zero as usual, allowing
3602 correct tracing of the function call chain. However, @value{GDBN} has
3603 no provision for frameless functions elsewhere in the stack.
3606 @kindex frame@r{, command}
3607 @item frame @var{args}
3608 The @code{frame} command allows you to move from one stack frame to another,
3609 and to print the stack frame you select. @var{args} may be either the
3610 address of the frame or the stack frame number. Without an argument,
3611 @code{frame} prints the current stack frame.
3613 @kindex select-frame
3615 The @code{select-frame} command allows you to move from one stack frame
3616 to another without printing the frame. This is the silent version of
3625 @cindex stack traces
3626 A backtrace is a summary of how your program got where it is. It shows one
3627 line per frame, for many frames, starting with the currently executing
3628 frame (frame zero), followed by its caller (frame one), and on up the
3636 Print a backtrace of the entire stack: one line per frame for all
3637 frames in the stack.
3639 You can stop the backtrace at any time by typing the system interrupt
3640 character, normally @kbd{C-c}.
3642 @item backtrace @var{n}
3644 Similar, but print only the innermost @var{n} frames.
3646 @item backtrace -@var{n}
3648 Similar, but print only the outermost @var{n} frames.
3654 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3655 are additional aliases for @code{backtrace}.
3657 Each line in the backtrace shows the frame number and the function name.
3658 The program counter value is also shown---unless you use @code{set
3659 print address off}. The backtrace also shows the source file name and
3660 line number, as well as the arguments to the function. The program
3661 counter value is omitted if it is at the beginning of the code for that
3664 Here is an example of a backtrace. It was made with the command
3665 @samp{bt 3}, so it shows the innermost three frames.
3669 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3671 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3672 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3674 (More stack frames follow...)
3679 The display for frame zero does not begin with a program counter
3680 value, indicating that your program has stopped at the beginning of the
3681 code for line @code{993} of @code{builtin.c}.
3684 @section Selecting a frame
3686 Most commands for examining the stack and other data in your program work on
3687 whichever stack frame is selected at the moment. Here are the commands for
3688 selecting a stack frame; all of them finish by printing a brief description
3689 of the stack frame just selected.
3692 @kindex frame@r{, selecting}
3696 Select frame number @var{n}. Recall that frame zero is the innermost
3697 (currently executing) frame, frame one is the frame that called the
3698 innermost one, and so on. The highest-numbered frame is the one for
3701 @item frame @var{addr}
3703 Select the frame at address @var{addr}. This is useful mainly if the
3704 chaining of stack frames has been damaged by a bug, making it
3705 impossible for @value{GDBN} to assign numbers properly to all frames. In
3706 addition, this can be useful when your program has multiple stacks and
3707 switches between them.
3709 On the SPARC architecture, @code{frame} needs two addresses to
3710 select an arbitrary frame: a frame pointer and a stack pointer.
3712 On the MIPS and Alpha architecture, it needs two addresses: a stack
3713 pointer and a program counter.
3715 On the 29k architecture, it needs three addresses: a register stack
3716 pointer, a program counter, and a memory stack pointer.
3717 @c note to future updaters: this is conditioned on a flag
3718 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3719 @c as of 27 Jan 1994.
3723 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3724 advances toward the outermost frame, to higher frame numbers, to frames
3725 that have existed longer. @var{n} defaults to one.
3730 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3731 advances toward the innermost frame, to lower frame numbers, to frames
3732 that were created more recently. @var{n} defaults to one. You may
3733 abbreviate @code{down} as @code{do}.
3736 All of these commands end by printing two lines of output describing the
3737 frame. The first line shows the frame number, the function name, the
3738 arguments, and the source file and line number of execution in that
3739 frame. The second line shows the text of that source line.
3747 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3749 10 read_input_file (argv[i]);
3753 After such a printout, the @code{list} command with no arguments
3754 prints ten lines centered on the point of execution in the frame.
3755 @xref{List, ,Printing source lines}.
3758 @kindex down-silently
3760 @item up-silently @var{n}
3761 @itemx down-silently @var{n}
3762 These two commands are variants of @code{up} and @code{down},
3763 respectively; they differ in that they do their work silently, without
3764 causing display of the new frame. They are intended primarily for use
3765 in @value{GDBN} command scripts, where the output might be unnecessary and
3770 @section Information about a frame
3772 There are several other commands to print information about the selected
3778 When used without any argument, this command does not change which
3779 frame is selected, but prints a brief description of the currently
3780 selected stack frame. It can be abbreviated @code{f}. With an
3781 argument, this command is used to select a stack frame.
3782 @xref{Selection, ,Selecting a frame}.
3788 This command prints a verbose description of the selected stack frame,
3793 the address of the frame
3795 the address of the next frame down (called by this frame)
3797 the address of the next frame up (caller of this frame)
3799 the language in which the source code corresponding to this frame is written
3801 the address of the frame's arguments
3803 the address of the frame's local variables
3805 the program counter saved in it (the address of execution in the caller frame)
3807 which registers were saved in the frame
3810 @noindent The verbose description is useful when
3811 something has gone wrong that has made the stack format fail to fit
3812 the usual conventions.
3814 @item info frame @var{addr}
3815 @itemx info f @var{addr}
3816 Print a verbose description of the frame at address @var{addr}, without
3817 selecting that frame. The selected frame remains unchanged by this
3818 command. This requires the same kind of address (more than one for some
3819 architectures) that you specify in the @code{frame} command.
3820 @xref{Selection, ,Selecting a frame}.
3824 Print the arguments of the selected frame, each on a separate line.
3828 Print the local variables of the selected frame, each on a separate
3829 line. These are all variables (declared either static or automatic)
3830 accessible at the point of execution of the selected frame.
3833 @cindex catch exceptions, list active handlers
3834 @cindex exception handlers, how to list
3836 Print a list of all the exception handlers that are active in the
3837 current stack frame at the current point of execution. To see other
3838 exception handlers, visit the associated frame (using the @code{up},
3839 @code{down}, or @code{frame} commands); then type @code{info catch}.
3840 @xref{Set Catchpoints, , Setting catchpoints}.
3846 @chapter Examining Source Files
3848 @value{GDBN} can print parts of your program's source, since the debugging
3849 information recorded in the program tells @value{GDBN} what source files were
3850 used to build it. When your program stops, @value{GDBN} spontaneously prints
3851 the line where it stopped. Likewise, when you select a stack frame
3852 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
3853 execution in that frame has stopped. You can print other portions of
3854 source files by explicit command.
3856 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
3857 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
3858 @value{GDBN} under @sc{gnu} Emacs}.
3861 * List:: Printing source lines
3862 * Search:: Searching source files
3863 * Source Path:: Specifying source directories
3864 * Machine Code:: Source and machine code
3868 @section Printing source lines
3872 To print lines from a source file, use the @code{list} command
3873 (abbreviated @code{l}). By default, ten lines are printed.
3874 There are several ways to specify what part of the file you want to print.
3876 Here are the forms of the @code{list} command most commonly used:
3879 @item list @var{linenum}
3880 Print lines centered around line number @var{linenum} in the
3881 current source file.
3883 @item list @var{function}
3884 Print lines centered around the beginning of function
3888 Print more lines. If the last lines printed were printed with a
3889 @code{list} command, this prints lines following the last lines
3890 printed; however, if the last line printed was a solitary line printed
3891 as part of displaying a stack frame (@pxref{Stack, ,Examining the
3892 Stack}), this prints lines centered around that line.
3895 Print lines just before the lines last printed.
3898 By default, @value{GDBN} prints ten source lines with any of these forms of
3899 the @code{list} command. You can change this using @code{set listsize}:
3902 @kindex set listsize
3903 @item set listsize @var{count}
3904 Make the @code{list} command display @var{count} source lines (unless
3905 the @code{list} argument explicitly specifies some other number).
3907 @kindex show listsize
3909 Display the number of lines that @code{list} prints.
3912 Repeating a @code{list} command with @key{RET} discards the argument,
3913 so it is equivalent to typing just @code{list}. This is more useful
3914 than listing the same lines again. An exception is made for an
3915 argument of @samp{-}; that argument is preserved in repetition so that
3916 each repetition moves up in the source file.
3919 In general, the @code{list} command expects you to supply zero, one or two
3920 @dfn{linespecs}. Linespecs specify source lines; there are several ways
3921 of writing them, but the effect is always to specify some source line.
3922 Here is a complete description of the possible arguments for @code{list}:
3925 @item list @var{linespec}
3926 Print lines centered around the line specified by @var{linespec}.
3928 @item list @var{first},@var{last}
3929 Print lines from @var{first} to @var{last}. Both arguments are
3932 @item list ,@var{last}
3933 Print lines ending with @var{last}.
3935 @item list @var{first},
3936 Print lines starting with @var{first}.
3939 Print lines just after the lines last printed.
3942 Print lines just before the lines last printed.
3945 As described in the preceding table.
3948 Here are the ways of specifying a single source line---all the
3953 Specifies line @var{number} of the current source file.
3954 When a @code{list} command has two linespecs, this refers to
3955 the same source file as the first linespec.
3958 Specifies the line @var{offset} lines after the last line printed.
3959 When used as the second linespec in a @code{list} command that has
3960 two, this specifies the line @var{offset} lines down from the
3964 Specifies the line @var{offset} lines before the last line printed.
3966 @item @var{filename}:@var{number}
3967 Specifies line @var{number} in the source file @var{filename}.
3969 @item @var{function}
3970 Specifies the line that begins the body of the function @var{function}.
3971 For example: in C, this is the line with the open brace.
3973 @item @var{filename}:@var{function}
3974 Specifies the line of the open-brace that begins the body of the
3975 function @var{function} in the file @var{filename}. You only need the
3976 file name with a function name to avoid ambiguity when there are
3977 identically named functions in different source files.
3979 @item *@var{address}
3980 Specifies the line containing the program address @var{address}.
3981 @var{address} may be any expression.
3985 @section Searching source files
3987 @kindex reverse-search
3989 There are two commands for searching through the current source file for a
3994 @kindex forward-search
3995 @item forward-search @var{regexp}
3996 @itemx search @var{regexp}
3997 The command @samp{forward-search @var{regexp}} checks each line,
3998 starting with the one following the last line listed, for a match for
3999 @var{regexp}. It lists the line that is found. You can use the
4000 synonym @samp{search @var{regexp}} or abbreviate the command name as
4003 @item reverse-search @var{regexp}
4004 The command @samp{reverse-search @var{regexp}} checks each line, starting
4005 with the one before the last line listed and going backward, for a match
4006 for @var{regexp}. It lists the line that is found. You can abbreviate
4007 this command as @code{rev}.
4011 @section Specifying source directories
4014 @cindex directories for source files
4015 Executable programs sometimes do not record the directories of the source
4016 files from which they were compiled, just the names. Even when they do,
4017 the directories could be moved between the compilation and your debugging
4018 session. @value{GDBN} has a list of directories to search for source files;
4019 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
4020 it tries all the directories in the list, in the order they are present
4021 in the list, until it finds a file with the desired name. Note that
4022 the executable search path is @emph{not} used for this purpose. Neither is
4023 the current working directory, unless it happens to be in the source
4026 If @value{GDBN} cannot find a source file in the source path, and the
4027 object program records a directory, @value{GDBN} tries that directory
4028 too. If the source path is empty, and there is no record of the
4029 compilation directory, @value{GDBN} looks in the current directory as a
4032 Whenever you reset or rearrange the source path, @value{GDBN} clears out
4033 any information it has cached about where source files are found and where
4034 each line is in the file.
4038 When you start @value{GDBN}, its source path includes only @samp{cdir}
4039 and @samp{cwd}, in that order.
4040 To add other directories, use the @code{directory} command.
4043 @item directory @var{dirname} @dots{}
4044 @item dir @var{dirname} @dots{}
4045 Add directory @var{dirname} to the front of the source path. Several
4046 directory names may be given to this command, separated by @samp{:}
4047 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
4048 part of absolute file names) or
4049 whitespace. You may specify a directory that is already in the source
4050 path; this moves it forward, so @value{GDBN} searches it sooner.
4056 @cindex compilation directory
4057 @cindex current directory
4058 @cindex working directory
4059 @cindex directory, current
4060 @cindex directory, compilation
4061 You can use the string @samp{$cdir} to refer to the compilation
4062 directory (if one is recorded), and @samp{$cwd} to refer to the current
4063 working directory. @samp{$cwd} is not the same as @samp{.}---the former
4064 tracks the current working directory as it changes during your @value{GDBN}
4065 session, while the latter is immediately expanded to the current
4066 directory at the time you add an entry to the source path.
4069 Reset the source path to empty again. This requires confirmation.
4071 @c RET-repeat for @code{directory} is explicitly disabled, but since
4072 @c repeating it would be a no-op we do not say that. (thanks to RMS)
4074 @item show directories
4075 @kindex show directories
4076 Print the source path: show which directories it contains.
4079 If your source path is cluttered with directories that are no longer of
4080 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
4081 versions of source. You can correct the situation as follows:
4085 Use @code{directory} with no argument to reset the source path to empty.
4088 Use @code{directory} with suitable arguments to reinstall the
4089 directories you want in the source path. You can add all the
4090 directories in one command.
4094 @section Source and machine code
4096 You can use the command @code{info line} to map source lines to program
4097 addresses (and vice versa), and the command @code{disassemble} to display
4098 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
4099 mode, the @code{info line} command causes the arrow to point to the
4100 line specified. Also, @code{info line} prints addresses in symbolic form as
4105 @item info line @var{linespec}
4106 Print the starting and ending addresses of the compiled code for
4107 source line @var{linespec}. You can specify source lines in any of
4108 the ways understood by the @code{list} command (@pxref{List, ,Printing
4112 For example, we can use @code{info line} to discover the location of
4113 the object code for the first line of function
4114 @code{m4_changequote}:
4116 @c FIXME: I think this example should also show the addresses in
4117 @c symbolic form, as they usually would be displayed.
4119 (@value{GDBP}) info line m4_changequote
4120 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
4124 We can also inquire (using @code{*@var{addr}} as the form for
4125 @var{linespec}) what source line covers a particular address:
4127 (@value{GDBP}) info line *0x63ff
4128 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
4131 @cindex @code{$_} and @code{info line}
4132 @kindex x@r{, and }@code{info line}
4133 After @code{info line}, the default address for the @code{x} command
4134 is changed to the starting address of the line, so that @samp{x/i} is
4135 sufficient to begin examining the machine code (@pxref{Memory,
4136 ,Examining memory}). Also, this address is saved as the value of the
4137 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
4142 @cindex assembly instructions
4143 @cindex instructions, assembly
4144 @cindex machine instructions
4145 @cindex listing machine instructions
4147 This specialized command dumps a range of memory as machine
4148 instructions. The default memory range is the function surrounding the
4149 program counter of the selected frame. A single argument to this
4150 command is a program counter value; @value{GDBN} dumps the function
4151 surrounding this value. Two arguments specify a range of addresses
4152 (first inclusive, second exclusive) to dump.
4155 The following example shows the disassembly of a range of addresses of
4156 HP PA-RISC 2.0 code:
4159 (@value{GDBP}) disas 0x32c4 0x32e4
4160 Dump of assembler code from 0x32c4 to 0x32e4:
4161 0x32c4 <main+204>: addil 0,dp
4162 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
4163 0x32cc <main+212>: ldil 0x3000,r31
4164 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
4165 0x32d4 <main+220>: ldo 0(r31),rp
4166 0x32d8 <main+224>: addil -0x800,dp
4167 0x32dc <main+228>: ldo 0x588(r1),r26
4168 0x32e0 <main+232>: ldil 0x3000,r31
4169 End of assembler dump.
4172 Some architectures have more than one commonly-used set of instruction
4173 mnemonics or other syntax.
4176 @kindex set disassembly-flavor
4177 @cindex assembly instructions
4178 @cindex instructions, assembly
4179 @cindex machine instructions
4180 @cindex listing machine instructions
4181 @cindex Intel disassembly flavor
4182 @cindex AT&T disassembly flavor
4183 @item set disassembly-flavor @var{instruction-set}
4184 Select the instruction set to use when disassembling the
4185 program via the @code{disassemble} or @code{x/i} commands.
4187 Currently this command is only defined for the Intel x86 family. You
4188 can set @var{instruction-set} to either @code{intel} or @code{att}.
4189 The default is @code{att}, the AT&T flavor used by default by Unix
4190 assemblers for x86-based targets.
4195 @chapter Examining Data
4197 @cindex printing data
4198 @cindex examining data
4201 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
4202 @c document because it is nonstandard... Under Epoch it displays in a
4203 @c different window or something like that.
4204 The usual way to examine data in your program is with the @code{print}
4205 command (abbreviated @code{p}), or its synonym @code{inspect}. It
4206 evaluates and prints the value of an expression of the language your
4207 program is written in (@pxref{Languages, ,Using @value{GDBN} with
4208 Different Languages}).
4211 @item print @var{expr}
4212 @itemx print /@var{f} @var{expr}
4213 @var{expr} is an expression (in the source language). By default the
4214 value of @var{expr} is printed in a format appropriate to its data type;
4215 you can choose a different format by specifying @samp{/@var{f}}, where
4216 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
4220 @itemx print /@var{f}
4221 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
4222 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
4223 conveniently inspect the same value in an alternative format.
4226 A more low-level way of examining data is with the @code{x} command.
4227 It examines data in memory at a specified address and prints it in a
4228 specified format. @xref{Memory, ,Examining memory}.
4230 If you are interested in information about types, or about how the
4231 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
4232 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
4236 * Expressions:: Expressions
4237 * Variables:: Program variables
4238 * Arrays:: Artificial arrays
4239 * Output Formats:: Output formats
4240 * Memory:: Examining memory
4241 * Auto Display:: Automatic display
4242 * Print Settings:: Print settings
4243 * Value History:: Value history
4244 * Convenience Vars:: Convenience variables
4245 * Registers:: Registers
4246 * Floating Point Hardware:: Floating point hardware
4250 @section Expressions
4253 @code{print} and many other @value{GDBN} commands accept an expression and
4254 compute its value. Any kind of constant, variable or operator defined
4255 by the programming language you are using is valid in an expression in
4256 @value{GDBN}. This includes conditional expressions, function calls, casts
4257 and string constants. It unfortunately does not include symbols defined
4258 by preprocessor @code{#define} commands.
4260 @value{GDBN} supports array constants in expressions input by
4261 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
4262 you can use the command @code{print @{1, 2, 3@}} to build up an array in
4263 memory that is @code{malloc}ed in the target program.
4265 Because C is so widespread, most of the expressions shown in examples in
4266 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
4267 Languages}, for information on how to use expressions in other
4270 In this section, we discuss operators that you can use in @value{GDBN}
4271 expressions regardless of your programming language.
4273 Casts are supported in all languages, not just in C, because it is so
4274 useful to cast a number into a pointer in order to examine a structure
4275 at that address in memory.
4276 @c FIXME: casts supported---Mod2 true?
4278 @value{GDBN} supports these operators, in addition to those common
4279 to programming languages:
4283 @samp{@@} is a binary operator for treating parts of memory as arrays.
4284 @xref{Arrays, ,Artificial arrays}, for more information.
4287 @samp{::} allows you to specify a variable in terms of the file or
4288 function where it is defined. @xref{Variables, ,Program variables}.
4290 @cindex @{@var{type}@}
4291 @cindex type casting memory
4292 @cindex memory, viewing as typed object
4293 @cindex casts, to view memory
4294 @item @{@var{type}@} @var{addr}
4295 Refers to an object of type @var{type} stored at address @var{addr} in
4296 memory. @var{addr} may be any expression whose value is an integer or
4297 pointer (but parentheses are required around binary operators, just as in
4298 a cast). This construct is allowed regardless of what kind of data is
4299 normally supposed to reside at @var{addr}.
4303 @section Program variables
4305 The most common kind of expression to use is the name of a variable
4308 Variables in expressions are understood in the selected stack frame
4309 (@pxref{Selection, ,Selecting a frame}); they must be either:
4313 global (or file-static)
4320 visible according to the scope rules of the
4321 programming language from the point of execution in that frame
4324 @noindent This means that in the function
4339 you can examine and use the variable @code{a} whenever your program is
4340 executing within the function @code{foo}, but you can only use or
4341 examine the variable @code{b} while your program is executing inside
4342 the block where @code{b} is declared.
4344 @cindex variable name conflict
4345 There is an exception: you can refer to a variable or function whose
4346 scope is a single source file even if the current execution point is not
4347 in this file. But it is possible to have more than one such variable or
4348 function with the same name (in different source files). If that
4349 happens, referring to that name has unpredictable effects. If you wish,
4350 you can specify a static variable in a particular function or file,
4351 using the colon-colon notation:
4353 @cindex colon-colon, context for variables/functions
4355 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4359 @var{file}::@var{variable}
4360 @var{function}::@var{variable}
4364 Here @var{file} or @var{function} is the name of the context for the
4365 static @var{variable}. In the case of file names, you can use quotes to
4366 make sure @value{GDBN} parses the file name as a single word---for example,
4367 to print a global value of @code{x} defined in @file{f2.c}:
4370 (@value{GDBP}) p 'f2.c'::x
4373 @cindex C++ scope resolution
4374 This use of @samp{::} is very rarely in conflict with the very similar
4375 use of the same notation in C++. @value{GDBN} also supports use of the C++
4376 scope resolution operator in @value{GDBN} expressions.
4377 @c FIXME: Um, so what happens in one of those rare cases where it's in
4380 @cindex wrong values
4381 @cindex variable values, wrong
4383 @emph{Warning:} Occasionally, a local variable may appear to have the
4384 wrong value at certain points in a function---just after entry to a new
4385 scope, and just before exit.
4387 You may see this problem when you are stepping by machine instructions.
4388 This is because, on most machines, it takes more than one instruction to
4389 set up a stack frame (including local variable definitions); if you are
4390 stepping by machine instructions, variables may appear to have the wrong
4391 values until the stack frame is completely built. On exit, it usually
4392 also takes more than one machine instruction to destroy a stack frame;
4393 after you begin stepping through that group of instructions, local
4394 variable definitions may be gone.
4396 This may also happen when the compiler does significant optimizations.
4397 To be sure of always seeing accurate values, turn off all optimization
4400 @cindex ``No symbol "foo" in current context''
4401 Another possible effect of compiler optimizations is to optimize
4402 unused variables out of existence, or assign variables to registers (as
4403 opposed to memory addresses). Depending on the support for such cases
4404 offered by the debug info format used by the compiler, @value{GDBN}
4405 might not be able to display values for such local variables. If that
4406 happens, @value{GDBN} will print a message like this:
4409 No symbol "foo" in current context.
4412 To solve such problems, either recompile without optimizations, or use a
4413 different debug info format, if the compiler supports several such
4414 formats. For example, @value{NGCC}, the @sc{gnu} C/C++ compiler usually
4415 supports the @samp{-gstabs} option. @samp{-gstabs} produces debug info
4416 in a format that is superior to formats such as COFF. You may be able
4417 to use DWARF-2 (@samp{-gdwarf-2}), which is also an effective form for
4418 debug info. See @ref{Debugging Options,,Options for Debugging Your
4419 Program or @sc{gnu} CC, gcc.info, Using @sc{gnu} CC}, for more
4424 @section Artificial arrays
4426 @cindex artificial array
4428 It is often useful to print out several successive objects of the
4429 same type in memory; a section of an array, or an array of
4430 dynamically determined size for which only a pointer exists in the
4433 You can do this by referring to a contiguous span of memory as an
4434 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4435 operand of @samp{@@} should be the first element of the desired array
4436 and be an individual object. The right operand should be the desired length
4437 of the array. The result is an array value whose elements are all of
4438 the type of the left argument. The first element is actually the left
4439 argument; the second element comes from bytes of memory immediately
4440 following those that hold the first element, and so on. Here is an
4441 example. If a program says
4444 int *array = (int *) malloc (len * sizeof (int));
4448 you can print the contents of @code{array} with
4454 The left operand of @samp{@@} must reside in memory. Array values made
4455 with @samp{@@} in this way behave just like other arrays in terms of
4456 subscripting, and are coerced to pointers when used in expressions.
4457 Artificial arrays most often appear in expressions via the value history
4458 (@pxref{Value History, ,Value history}), after printing one out.
4460 Another way to create an artificial array is to use a cast.
4461 This re-interprets a value as if it were an array.
4462 The value need not be in memory:
4464 (@value{GDBP}) p/x (short[2])0x12345678
4465 $1 = @{0x1234, 0x5678@}
4468 As a convenience, if you leave the array length out (as in
4469 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
4470 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4472 (@value{GDBP}) p/x (short[])0x12345678
4473 $2 = @{0x1234, 0x5678@}
4476 Sometimes the artificial array mechanism is not quite enough; in
4477 moderately complex data structures, the elements of interest may not
4478 actually be adjacent---for example, if you are interested in the values
4479 of pointers in an array. One useful work-around in this situation is
4480 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4481 variables}) as a counter in an expression that prints the first
4482 interesting value, and then repeat that expression via @key{RET}. For
4483 instance, suppose you have an array @code{dtab} of pointers to
4484 structures, and you are interested in the values of a field @code{fv}
4485 in each structure. Here is an example of what you might type:
4495 @node Output Formats
4496 @section Output formats
4498 @cindex formatted output
4499 @cindex output formats
4500 By default, @value{GDBN} prints a value according to its data type. Sometimes
4501 this is not what you want. For example, you might want to print a number
4502 in hex, or a pointer in decimal. Or you might want to view data in memory
4503 at a certain address as a character string or as an instruction. To do
4504 these things, specify an @dfn{output format} when you print a value.
4506 The simplest use of output formats is to say how to print a value
4507 already computed. This is done by starting the arguments of the
4508 @code{print} command with a slash and a format letter. The format
4509 letters supported are:
4513 Regard the bits of the value as an integer, and print the integer in
4517 Print as integer in signed decimal.
4520 Print as integer in unsigned decimal.
4523 Print as integer in octal.
4526 Print as integer in binary. The letter @samp{t} stands for ``two''.
4527 @footnote{@samp{b} cannot be used because these format letters are also
4528 used with the @code{x} command, where @samp{b} stands for ``byte'';
4529 see @ref{Memory,,Examining memory}.}
4532 @cindex unknown address, locating
4533 Print as an address, both absolute in hexadecimal and as an offset from
4534 the nearest preceding symbol. You can use this format used to discover
4535 where (in what function) an unknown address is located:
4538 (@value{GDBP}) p/a 0x54320
4539 $3 = 0x54320 <_initialize_vx+396>
4543 Regard as an integer and print it as a character constant.
4546 Regard the bits of the value as a floating point number and print
4547 using typical floating point syntax.
4550 For example, to print the program counter in hex (@pxref{Registers}), type
4557 Note that no space is required before the slash; this is because command
4558 names in @value{GDBN} cannot contain a slash.
4560 To reprint the last value in the value history with a different format,
4561 you can use the @code{print} command with just a format and no
4562 expression. For example, @samp{p/x} reprints the last value in hex.
4565 @section Examining memory
4567 You can use the command @code{x} (for ``examine'') to examine memory in
4568 any of several formats, independently of your program's data types.
4570 @cindex examining memory
4573 @item x/@var{nfu} @var{addr}
4576 Use the @code{x} command to examine memory.
4579 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4580 much memory to display and how to format it; @var{addr} is an
4581 expression giving the address where you want to start displaying memory.
4582 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4583 Several commands set convenient defaults for @var{addr}.
4586 @item @var{n}, the repeat count
4587 The repeat count is a decimal integer; the default is 1. It specifies
4588 how much memory (counting by units @var{u}) to display.
4589 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4592 @item @var{f}, the display format
4593 The display format is one of the formats used by @code{print},
4594 @samp{s} (null-terminated string), or @samp{i} (machine instruction).
4595 The default is @samp{x} (hexadecimal) initially.
4596 The default changes each time you use either @code{x} or @code{print}.
4598 @item @var{u}, the unit size
4599 The unit size is any of
4605 Halfwords (two bytes).
4607 Words (four bytes). This is the initial default.
4609 Giant words (eight bytes).
4612 Each time you specify a unit size with @code{x}, that size becomes the
4613 default unit the next time you use @code{x}. (For the @samp{s} and
4614 @samp{i} formats, the unit size is ignored and is normally not written.)
4616 @item @var{addr}, starting display address
4617 @var{addr} is the address where you want @value{GDBN} to begin displaying
4618 memory. The expression need not have a pointer value (though it may);
4619 it is always interpreted as an integer address of a byte of memory.
4620 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4621 @var{addr} is usually just after the last address examined---but several
4622 other commands also set the default address: @code{info breakpoints} (to
4623 the address of the last breakpoint listed), @code{info line} (to the
4624 starting address of a line), and @code{print} (if you use it to display
4625 a value from memory).
4628 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4629 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4630 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4631 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4632 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
4634 Since the letters indicating unit sizes are all distinct from the
4635 letters specifying output formats, you do not have to remember whether
4636 unit size or format comes first; either order works. The output
4637 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4638 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4640 Even though the unit size @var{u} is ignored for the formats @samp{s}
4641 and @samp{i}, you might still want to use a count @var{n}; for example,
4642 @samp{3i} specifies that you want to see three machine instructions,
4643 including any operands. The command @code{disassemble} gives an
4644 alternative way of inspecting machine instructions; see @ref{Machine
4645 Code,,Source and machine code}.
4647 All the defaults for the arguments to @code{x} are designed to make it
4648 easy to continue scanning memory with minimal specifications each time
4649 you use @code{x}. For example, after you have inspected three machine
4650 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4651 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4652 the repeat count @var{n} is used again; the other arguments default as
4653 for successive uses of @code{x}.
4655 @cindex @code{$_}, @code{$__}, and value history
4656 The addresses and contents printed by the @code{x} command are not saved
4657 in the value history because there is often too much of them and they
4658 would get in the way. Instead, @value{GDBN} makes these values available for
4659 subsequent use in expressions as values of the convenience variables
4660 @code{$_} and @code{$__}. After an @code{x} command, the last address
4661 examined is available for use in expressions in the convenience variable
4662 @code{$_}. The contents of that address, as examined, are available in
4663 the convenience variable @code{$__}.
4665 If the @code{x} command has a repeat count, the address and contents saved
4666 are from the last memory unit printed; this is not the same as the last
4667 address printed if several units were printed on the last line of output.
4670 @section Automatic display
4671 @cindex automatic display
4672 @cindex display of expressions
4674 If you find that you want to print the value of an expression frequently
4675 (to see how it changes), you might want to add it to the @dfn{automatic
4676 display list} so that @value{GDBN} prints its value each time your program stops.
4677 Each expression added to the list is given a number to identify it;
4678 to remove an expression from the list, you specify that number.
4679 The automatic display looks like this:
4683 3: bar[5] = (struct hack *) 0x3804
4687 This display shows item numbers, expressions and their current values. As with
4688 displays you request manually using @code{x} or @code{print}, you can
4689 specify the output format you prefer; in fact, @code{display} decides
4690 whether to use @code{print} or @code{x} depending on how elaborate your
4691 format specification is---it uses @code{x} if you specify a unit size,
4692 or one of the two formats (@samp{i} and @samp{s}) that are only
4693 supported by @code{x}; otherwise it uses @code{print}.
4697 @item display @var{expr}
4698 Add the expression @var{expr} to the list of expressions to display
4699 each time your program stops. @xref{Expressions, ,Expressions}.
4701 @code{display} does not repeat if you press @key{RET} again after using it.
4703 @item display/@var{fmt} @var{expr}
4704 For @var{fmt} specifying only a display format and not a size or
4705 count, add the expression @var{expr} to the auto-display list but
4706 arrange to display it each time in the specified format @var{fmt}.
4707 @xref{Output Formats,,Output formats}.
4709 @item display/@var{fmt} @var{addr}
4710 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4711 number of units, add the expression @var{addr} as a memory address to
4712 be examined each time your program stops. Examining means in effect
4713 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4716 For example, @samp{display/i $pc} can be helpful, to see the machine
4717 instruction about to be executed each time execution stops (@samp{$pc}
4718 is a common name for the program counter; @pxref{Registers, ,Registers}).
4721 @kindex delete display
4723 @item undisplay @var{dnums}@dots{}
4724 @itemx delete display @var{dnums}@dots{}
4725 Remove item numbers @var{dnums} from the list of expressions to display.
4727 @code{undisplay} does not repeat if you press @key{RET} after using it.
4728 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4730 @kindex disable display
4731 @item disable display @var{dnums}@dots{}
4732 Disable the display of item numbers @var{dnums}. A disabled display
4733 item is not printed automatically, but is not forgotten. It may be
4734 enabled again later.
4736 @kindex enable display
4737 @item enable display @var{dnums}@dots{}
4738 Enable display of item numbers @var{dnums}. It becomes effective once
4739 again in auto display of its expression, until you specify otherwise.
4742 Display the current values of the expressions on the list, just as is
4743 done when your program stops.
4745 @kindex info display
4747 Print the list of expressions previously set up to display
4748 automatically, each one with its item number, but without showing the
4749 values. This includes disabled expressions, which are marked as such.
4750 It also includes expressions which would not be displayed right now
4751 because they refer to automatic variables not currently available.
4754 If a display expression refers to local variables, then it does not make
4755 sense outside the lexical context for which it was set up. Such an
4756 expression is disabled when execution enters a context where one of its
4757 variables is not defined. For example, if you give the command
4758 @code{display last_char} while inside a function with an argument
4759 @code{last_char}, @value{GDBN} displays this argument while your program
4760 continues to stop inside that function. When it stops elsewhere---where
4761 there is no variable @code{last_char}---the display is disabled
4762 automatically. The next time your program stops where @code{last_char}
4763 is meaningful, you can enable the display expression once again.
4765 @node Print Settings
4766 @section Print settings
4768 @cindex format options
4769 @cindex print settings
4770 @value{GDBN} provides the following ways to control how arrays, structures,
4771 and symbols are printed.
4774 These settings are useful for debugging programs in any language:
4777 @kindex set print address
4778 @item set print address
4779 @itemx set print address on
4780 @value{GDBN} prints memory addresses showing the location of stack
4781 traces, structure values, pointer values, breakpoints, and so forth,
4782 even when it also displays the contents of those addresses. The default
4783 is @code{on}. For example, this is what a stack frame display looks like with
4784 @code{set print address on}:
4789 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
4791 530 if (lquote != def_lquote)
4795 @item set print address off
4796 Do not print addresses when displaying their contents. For example,
4797 this is the same stack frame displayed with @code{set print address off}:
4801 (@value{GDBP}) set print addr off
4803 #0 set_quotes (lq="<<", rq=">>") at input.c:530
4804 530 if (lquote != def_lquote)
4808 You can use @samp{set print address off} to eliminate all machine
4809 dependent displays from the @value{GDBN} interface. For example, with
4810 @code{print address off}, you should get the same text for backtraces on
4811 all machines---whether or not they involve pointer arguments.
4813 @kindex show print address
4814 @item show print address
4815 Show whether or not addresses are to be printed.
4818 When @value{GDBN} prints a symbolic address, it normally prints the
4819 closest earlier symbol plus an offset. If that symbol does not uniquely
4820 identify the address (for example, it is a name whose scope is a single
4821 source file), you may need to clarify. One way to do this is with
4822 @code{info line}, for example @samp{info line *0x4537}. Alternately,
4823 you can set @value{GDBN} to print the source file and line number when
4824 it prints a symbolic address:
4827 @kindex set print symbol-filename
4828 @item set print symbol-filename on
4829 Tell @value{GDBN} to print the source file name and line number of a
4830 symbol in the symbolic form of an address.
4832 @item set print symbol-filename off
4833 Do not print source file name and line number of a symbol. This is the
4836 @kindex show print symbol-filename
4837 @item show print symbol-filename
4838 Show whether or not @value{GDBN} will print the source file name and
4839 line number of a symbol in the symbolic form of an address.
4842 Another situation where it is helpful to show symbol filenames and line
4843 numbers is when disassembling code; @value{GDBN} shows you the line
4844 number and source file that corresponds to each instruction.
4846 Also, you may wish to see the symbolic form only if the address being
4847 printed is reasonably close to the closest earlier symbol:
4850 @kindex set print max-symbolic-offset
4851 @item set print max-symbolic-offset @var{max-offset}
4852 Tell @value{GDBN} to only display the symbolic form of an address if the
4853 offset between the closest earlier symbol and the address is less than
4854 @var{max-offset}. The default is 0, which tells @value{GDBN}
4855 to always print the symbolic form of an address if any symbol precedes it.
4857 @kindex show print max-symbolic-offset
4858 @item show print max-symbolic-offset
4859 Ask how large the maximum offset is that @value{GDBN} prints in a
4863 @cindex wild pointer, interpreting
4864 @cindex pointer, finding referent
4865 If you have a pointer and you are not sure where it points, try
4866 @samp{set print symbol-filename on}. Then you can determine the name
4867 and source file location of the variable where it points, using
4868 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
4869 For example, here @value{GDBN} shows that a variable @code{ptt} points
4870 at another variable @code{t}, defined in @file{hi2.c}:
4873 (@value{GDBP}) set print symbol-filename on
4874 (@value{GDBP}) p/a ptt
4875 $4 = 0xe008 <t in hi2.c>
4879 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
4880 does not show the symbol name and filename of the referent, even with
4881 the appropriate @code{set print} options turned on.
4884 Other settings control how different kinds of objects are printed:
4887 @kindex set print array
4888 @item set print array
4889 @itemx set print array on
4890 Pretty print arrays. This format is more convenient to read,
4891 but uses more space. The default is off.
4893 @item set print array off
4894 Return to compressed format for arrays.
4896 @kindex show print array
4897 @item show print array
4898 Show whether compressed or pretty format is selected for displaying
4901 @kindex set print elements
4902 @item set print elements @var{number-of-elements}
4903 Set a limit on how many elements of an array @value{GDBN} will print.
4904 If @value{GDBN} is printing a large array, it stops printing after it has
4905 printed the number of elements set by the @code{set print elements} command.
4906 This limit also applies to the display of strings.
4907 When @value{GDBN} starts, this limit is set to 200.
4908 Setting @var{number-of-elements} to zero means that the printing is unlimited.
4910 @kindex show print elements
4911 @item show print elements
4912 Display the number of elements of a large array that @value{GDBN} will print.
4913 If the number is 0, then the printing is unlimited.
4915 @kindex set print null-stop
4916 @item set print null-stop
4917 Cause @value{GDBN} to stop printing the characters of an array when the first
4918 @sc{null} is encountered. This is useful when large arrays actually
4919 contain only short strings.
4922 @kindex set print pretty
4923 @item set print pretty on
4924 Cause @value{GDBN} to print structures in an indented format with one member
4925 per line, like this:
4940 @item set print pretty off
4941 Cause @value{GDBN} to print structures in a compact format, like this:
4945 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
4946 meat = 0x54 "Pork"@}
4951 This is the default format.
4953 @kindex show print pretty
4954 @item show print pretty
4955 Show which format @value{GDBN} is using to print structures.
4957 @kindex set print sevenbit-strings
4958 @item set print sevenbit-strings on
4959 Print using only seven-bit characters; if this option is set,
4960 @value{GDBN} displays any eight-bit characters (in strings or
4961 character values) using the notation @code{\}@var{nnn}. This setting is
4962 best if you are working in English (@sc{ascii}) and you use the
4963 high-order bit of characters as a marker or ``meta'' bit.
4965 @item set print sevenbit-strings off
4966 Print full eight-bit characters. This allows the use of more
4967 international character sets, and is the default.
4969 @kindex show print sevenbit-strings
4970 @item show print sevenbit-strings
4971 Show whether or not @value{GDBN} is printing only seven-bit characters.
4973 @kindex set print union
4974 @item set print union on
4975 Tell @value{GDBN} to print unions which are contained in structures. This
4976 is the default setting.
4978 @item set print union off
4979 Tell @value{GDBN} not to print unions which are contained in structures.
4981 @kindex show print union
4982 @item show print union
4983 Ask @value{GDBN} whether or not it will print unions which are contained in
4986 For example, given the declarations
4989 typedef enum @{Tree, Bug@} Species;
4990 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
4991 typedef enum @{Caterpillar, Cocoon, Butterfly@}
5002 struct thing foo = @{Tree, @{Acorn@}@};
5006 with @code{set print union on} in effect @samp{p foo} would print
5009 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
5013 and with @code{set print union off} in effect it would print
5016 $1 = @{it = Tree, form = @{...@}@}
5022 These settings are of interest when debugging C++ programs:
5026 @kindex set print demangle
5027 @item set print demangle
5028 @itemx set print demangle on
5029 Print C++ names in their source form rather than in the encoded
5030 (``mangled'') form passed to the assembler and linker for type-safe
5031 linkage. The default is on.
5033 @kindex show print demangle
5034 @item show print demangle
5035 Show whether C++ names are printed in mangled or demangled form.
5037 @kindex set print asm-demangle
5038 @item set print asm-demangle
5039 @itemx set print asm-demangle on
5040 Print C++ names in their source form rather than their mangled form, even
5041 in assembler code printouts such as instruction disassemblies.
5044 @kindex show print asm-demangle
5045 @item show print asm-demangle
5046 Show whether C++ names in assembly listings are printed in mangled
5049 @kindex set demangle-style
5050 @cindex C++ symbol decoding style
5051 @cindex symbol decoding style, C++
5052 @item set demangle-style @var{style}
5053 Choose among several encoding schemes used by different compilers to
5054 represent C++ names. The choices for @var{style} are currently:
5058 Allow @value{GDBN} to choose a decoding style by inspecting your program.
5061 Decode based on the @sc{gnu} C++ compiler (@code{g++}) encoding algorithm.
5062 This is the default.
5065 Decode based on the HP ANSI C++ (@code{aCC}) encoding algorithm.
5068 Decode based on the Lucid C++ compiler (@code{lcc}) encoding algorithm.
5071 Decode using the algorithm in the @cite{C++ Annotated Reference Manual}.
5072 @strong{Warning:} this setting alone is not sufficient to allow
5073 debugging @code{cfront}-generated executables. @value{GDBN} would
5074 require further enhancement to permit that.
5077 If you omit @var{style}, you will see a list of possible formats.
5079 @kindex show demangle-style
5080 @item show demangle-style
5081 Display the encoding style currently in use for decoding C++ symbols.
5083 @kindex set print object
5084 @item set print object
5085 @itemx set print object on
5086 When displaying a pointer to an object, identify the @emph{actual}
5087 (derived) type of the object rather than the @emph{declared} type, using
5088 the virtual function table.
5090 @item set print object off
5091 Display only the declared type of objects, without reference to the
5092 virtual function table. This is the default setting.
5094 @kindex show print object
5095 @item show print object
5096 Show whether actual, or declared, object types are displayed.
5098 @kindex set print static-members
5099 @item set print static-members
5100 @itemx set print static-members on
5101 Print static members when displaying a C++ object. The default is on.
5103 @item set print static-members off
5104 Do not print static members when displaying a C++ object.
5106 @kindex show print static-members
5107 @item show print static-members
5108 Show whether C++ static members are printed, or not.
5110 @c These don't work with HP ANSI C++ yet.
5111 @kindex set print vtbl
5112 @item set print vtbl
5113 @itemx set print vtbl on
5114 Pretty print C++ virtual function tables. The default is off.
5115 (The @code{vtbl} commands do not work on programs compiled with the HP
5116 ANSI C++ compiler (@code{aCC}).)
5118 @item set print vtbl off
5119 Do not pretty print C++ virtual function tables.
5121 @kindex show print vtbl
5122 @item show print vtbl
5123 Show whether C++ virtual function tables are pretty printed, or not.
5127 @section Value history
5129 @cindex value history
5130 Values printed by the @code{print} command are saved in the @value{GDBN}
5131 @dfn{value history}. This allows you to refer to them in other expressions.
5132 Values are kept until the symbol table is re-read or discarded
5133 (for example with the @code{file} or @code{symbol-file} commands).
5134 When the symbol table changes, the value history is discarded,
5135 since the values may contain pointers back to the types defined in the
5140 @cindex history number
5141 The values printed are given @dfn{history numbers} by which you can
5142 refer to them. These are successive integers starting with one.
5143 @code{print} shows you the history number assigned to a value by
5144 printing @samp{$@var{num} = } before the value; here @var{num} is the
5147 To refer to any previous value, use @samp{$} followed by the value's
5148 history number. The way @code{print} labels its output is designed to
5149 remind you of this. Just @code{$} refers to the most recent value in
5150 the history, and @code{$$} refers to the value before that.
5151 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
5152 is the value just prior to @code{$$}, @code{$$1} is equivalent to
5153 @code{$$}, and @code{$$0} is equivalent to @code{$}.
5155 For example, suppose you have just printed a pointer to a structure and
5156 want to see the contents of the structure. It suffices to type
5162 If you have a chain of structures where the component @code{next} points
5163 to the next one, you can print the contents of the next one with this:
5170 You can print successive links in the chain by repeating this
5171 command---which you can do by just typing @key{RET}.
5173 Note that the history records values, not expressions. If the value of
5174 @code{x} is 4 and you type these commands:
5182 then the value recorded in the value history by the @code{print} command
5183 remains 4 even though the value of @code{x} has changed.
5188 Print the last ten values in the value history, with their item numbers.
5189 This is like @samp{p@ $$9} repeated ten times, except that @code{show
5190 values} does not change the history.
5192 @item show values @var{n}
5193 Print ten history values centered on history item number @var{n}.
5196 Print ten history values just after the values last printed. If no more
5197 values are available, @code{show values +} produces no display.
5200 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
5201 same effect as @samp{show values +}.
5203 @node Convenience Vars
5204 @section Convenience variables
5206 @cindex convenience variables
5207 @value{GDBN} provides @dfn{convenience variables} that you can use within
5208 @value{GDBN} to hold on to a value and refer to it later. These variables
5209 exist entirely within @value{GDBN}; they are not part of your program, and
5210 setting a convenience variable has no direct effect on further execution
5211 of your program. That is why you can use them freely.
5213 Convenience variables are prefixed with @samp{$}. Any name preceded by
5214 @samp{$} can be used for a convenience variable, unless it is one of
5215 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
5216 (Value history references, in contrast, are @emph{numbers} preceded
5217 by @samp{$}. @xref{Value History, ,Value history}.)
5219 You can save a value in a convenience variable with an assignment
5220 expression, just as you would set a variable in your program.
5224 set $foo = *object_ptr
5228 would save in @code{$foo} the value contained in the object pointed to by
5231 Using a convenience variable for the first time creates it, but its
5232 value is @code{void} until you assign a new value. You can alter the
5233 value with another assignment at any time.
5235 Convenience variables have no fixed types. You can assign a convenience
5236 variable any type of value, including structures and arrays, even if
5237 that variable already has a value of a different type. The convenience
5238 variable, when used as an expression, has the type of its current value.
5241 @kindex show convenience
5242 @item show convenience
5243 Print a list of convenience variables used so far, and their values.
5244 Abbreviated @code{show conv}.
5247 One of the ways to use a convenience variable is as a counter to be
5248 incremented or a pointer to be advanced. For example, to print
5249 a field from successive elements of an array of structures:
5253 print bar[$i++]->contents
5257 Repeat that command by typing @key{RET}.
5259 Some convenience variables are created automatically by @value{GDBN} and given
5260 values likely to be useful.
5265 The variable @code{$_} is automatically set by the @code{x} command to
5266 the last address examined (@pxref{Memory, ,Examining memory}). Other
5267 commands which provide a default address for @code{x} to examine also
5268 set @code{$_} to that address; these commands include @code{info line}
5269 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
5270 except when set by the @code{x} command, in which case it is a pointer
5271 to the type of @code{$__}.
5275 The variable @code{$__} is automatically set by the @code{x} command
5276 to the value found in the last address examined. Its type is chosen
5277 to match the format in which the data was printed.
5281 The variable @code{$_exitcode} is automatically set to the exit code when
5282 the program being debugged terminates.
5285 On HP-UX systems, if you refer to a function or variable name that
5286 begins with a dollar sign, @value{GDBN} searches for a user or system
5287 name first, before it searches for a convenience variable.
5293 You can refer to machine register contents, in expressions, as variables
5294 with names starting with @samp{$}. The names of registers are different
5295 for each machine; use @code{info registers} to see the names used on
5299 @kindex info registers
5300 @item info registers
5301 Print the names and values of all registers except floating-point
5302 registers (in the selected stack frame).
5304 @kindex info all-registers
5305 @cindex floating point registers
5306 @item info all-registers
5307 Print the names and values of all registers, including floating-point
5310 @item info registers @var{regname} @dots{}
5311 Print the @dfn{relativized} value of each specified register @var{regname}.
5312 As discussed in detail below, register values are normally relative to
5313 the selected stack frame. @var{regname} may be any register name valid on
5314 the machine you are using, with or without the initial @samp{$}.
5317 @value{GDBN} has four ``standard'' register names that are available (in
5318 expressions) on most machines---whenever they do not conflict with an
5319 architecture's canonical mnemonics for registers. The register names
5320 @code{$pc} and @code{$sp} are used for the program counter register and
5321 the stack pointer. @code{$fp} is used for a register that contains a
5322 pointer to the current stack frame, and @code{$ps} is used for a
5323 register that contains the processor status. For example,
5324 you could print the program counter in hex with
5331 or print the instruction to be executed next with
5338 or add four to the stack pointer@footnote{This is a way of removing
5339 one word from the stack, on machines where stacks grow downward in
5340 memory (most machines, nowadays). This assumes that the innermost
5341 stack frame is selected; setting @code{$sp} is not allowed when other
5342 stack frames are selected. To pop entire frames off the stack,
5343 regardless of machine architecture, use @code{return};
5344 see @ref{Returning, ,Returning from a function}.} with
5350 Whenever possible, these four standard register names are available on
5351 your machine even though the machine has different canonical mnemonics,
5352 so long as there is no conflict. The @code{info registers} command
5353 shows the canonical names. For example, on the SPARC, @code{info
5354 registers} displays the processor status register as @code{$psr} but you
5355 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
5356 is an alias for the @sc{eflags} register.
5358 @value{GDBN} always considers the contents of an ordinary register as an
5359 integer when the register is examined in this way. Some machines have
5360 special registers which can hold nothing but floating point; these
5361 registers are considered to have floating point values. There is no way
5362 to refer to the contents of an ordinary register as floating point value
5363 (although you can @emph{print} it as a floating point value with
5364 @samp{print/f $@var{regname}}).
5366 Some registers have distinct ``raw'' and ``virtual'' data formats. This
5367 means that the data format in which the register contents are saved by
5368 the operating system is not the same one that your program normally
5369 sees. For example, the registers of the 68881 floating point
5370 coprocessor are always saved in ``extended'' (raw) format, but all C
5371 programs expect to work with ``double'' (virtual) format. In such
5372 cases, @value{GDBN} normally works with the virtual format only (the format
5373 that makes sense for your program), but the @code{info registers} command
5374 prints the data in both formats.
5376 Normally, register values are relative to the selected stack frame
5377 (@pxref{Selection, ,Selecting a frame}). This means that you get the
5378 value that the register would contain if all stack frames farther in
5379 were exited and their saved registers restored. In order to see the
5380 true contents of hardware registers, you must select the innermost
5381 frame (with @samp{frame 0}).
5383 However, @value{GDBN} must deduce where registers are saved, from the machine
5384 code generated by your compiler. If some registers are not saved, or if
5385 @value{GDBN} is unable to locate the saved registers, the selected stack
5386 frame makes no difference.
5388 @node Floating Point Hardware
5389 @section Floating point hardware
5390 @cindex floating point
5392 Depending on the configuration, @value{GDBN} may be able to give
5393 you more information about the status of the floating point hardware.
5398 Display hardware-dependent information about the floating
5399 point unit. The exact contents and layout vary depending on the
5400 floating point chip. Currently, @samp{info float} is supported on
5401 the ARM and x86 machines.
5405 @chapter Using @value{GDBN} with Different Languages
5408 Although programming languages generally have common aspects, they are
5409 rarely expressed in the same manner. For instance, in ANSI C,
5410 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
5411 Modula-2, it is accomplished by @code{p^}. Values can also be
5412 represented (and displayed) differently. Hex numbers in C appear as
5413 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
5415 @cindex working language
5416 Language-specific information is built into @value{GDBN} for some languages,
5417 allowing you to express operations like the above in your program's
5418 native language, and allowing @value{GDBN} to output values in a manner
5419 consistent with the syntax of your program's native language. The
5420 language you use to build expressions is called the @dfn{working
5424 * Setting:: Switching between source languages
5425 * Show:: Displaying the language
5426 * Checks:: Type and range checks
5427 * Support:: Supported languages
5431 @section Switching between source languages
5433 There are two ways to control the working language---either have @value{GDBN}
5434 set it automatically, or select it manually yourself. You can use the
5435 @code{set language} command for either purpose. On startup, @value{GDBN}
5436 defaults to setting the language automatically. The working language is
5437 used to determine how expressions you type are interpreted, how values
5440 In addition to the working language, every source file that
5441 @value{GDBN} knows about has its own working language. For some object
5442 file formats, the compiler might indicate which language a particular
5443 source file is in. However, most of the time @value{GDBN} infers the
5444 language from the name of the file. The language of a source file
5445 controls whether C++ names are demangled---this way @code{backtrace} can
5446 show each frame appropriately for its own language. There is no way to
5447 set the language of a source file from within @value{GDBN}, but you can
5448 set the language associated with a filename extension. @xref{Show, ,
5449 Displaying the language}.
5451 This is most commonly a problem when you use a program, such
5452 as @code{cfront} or @code{f2c}, that generates C but is written in
5453 another language. In that case, make the
5454 program use @code{#line} directives in its C output; that way
5455 @value{GDBN} will know the correct language of the source code of the original
5456 program, and will display that source code, not the generated C code.
5459 * Filenames:: Filename extensions and languages.
5460 * Manually:: Setting the working language manually
5461 * Automatically:: Having @value{GDBN} infer the source language
5465 @subsection List of filename extensions and languages
5467 If a source file name ends in one of the following extensions, then
5468 @value{GDBN} infers that its language is the one indicated.
5493 Modula-2 source file
5497 Assembler source file. This actually behaves almost like C, but
5498 @value{GDBN} does not skip over function prologues when stepping.
5501 In addition, you may set the language associated with a filename
5502 extension. @xref{Show, , Displaying the language}.
5505 @subsection Setting the working language
5507 If you allow @value{GDBN} to set the language automatically,
5508 expressions are interpreted the same way in your debugging session and
5511 @kindex set language
5512 If you wish, you may set the language manually. To do this, issue the
5513 command @samp{set language @var{lang}}, where @var{lang} is the name of
5515 @code{c} or @code{modula-2}.
5516 For a list of the supported languages, type @samp{set language}.
5518 Setting the language manually prevents @value{GDBN} from updating the working
5519 language automatically. This can lead to confusion if you try
5520 to debug a program when the working language is not the same as the
5521 source language, when an expression is acceptable to both
5522 languages---but means different things. For instance, if the current
5523 source file were written in C, and @value{GDBN} was parsing Modula-2, a
5531 might not have the effect you intended. In C, this means to add
5532 @code{b} and @code{c} and place the result in @code{a}. The result
5533 printed would be the value of @code{a}. In Modula-2, this means to compare
5534 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
5537 @subsection Having @value{GDBN} infer the source language
5539 To have @value{GDBN} set the working language automatically, use
5540 @samp{set language local} or @samp{set language auto}. @value{GDBN}
5541 then infers the working language. That is, when your program stops in a
5542 frame (usually by encountering a breakpoint), @value{GDBN} sets the
5543 working language to the language recorded for the function in that
5544 frame. If the language for a frame is unknown (that is, if the function
5545 or block corresponding to the frame was defined in a source file that
5546 does not have a recognized extension), the current working language is
5547 not changed, and @value{GDBN} issues a warning.
5549 This may not seem necessary for most programs, which are written
5550 entirely in one source language. However, program modules and libraries
5551 written in one source language can be used by a main program written in
5552 a different source language. Using @samp{set language auto} in this
5553 case frees you from having to set the working language manually.
5556 @section Displaying the language
5558 The following commands help you find out which language is the
5559 working language, and also what language source files were written in.
5561 @kindex show language
5562 @kindex info frame@r{, show the source language}
5563 @kindex info source@r{, show the source language}
5566 Display the current working language. This is the
5567 language you can use with commands such as @code{print} to
5568 build and compute expressions that may involve variables in your program.
5571 Display the source language for this frame. This language becomes the
5572 working language if you use an identifier from this frame.
5573 @xref{Frame Info, ,Information about a frame}, to identify the other
5574 information listed here.
5577 Display the source language of this source file.
5578 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
5579 information listed here.
5582 In unusual circumstances, you may have source files with extensions
5583 not in the standard list. You can then set the extension associated
5584 with a language explicitly:
5586 @kindex set extension-language
5587 @kindex info extensions
5589 @item set extension-language @var{.ext} @var{language}
5590 Set source files with extension @var{.ext} to be assumed to be in
5591 the source language @var{language}.
5593 @item info extensions
5594 List all the filename extensions and the associated languages.
5598 @section Type and range checking
5601 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
5602 checking are included, but they do not yet have any effect. This
5603 section documents the intended facilities.
5605 @c FIXME remove warning when type/range code added
5607 Some languages are designed to guard you against making seemingly common
5608 errors through a series of compile- and run-time checks. These include
5609 checking the type of arguments to functions and operators, and making
5610 sure mathematical overflows are caught at run time. Checks such as
5611 these help to ensure a program's correctness once it has been compiled
5612 by eliminating type mismatches, and providing active checks for range
5613 errors when your program is running.
5615 @value{GDBN} can check for conditions like the above if you wish.
5616 Although @value{GDBN} does not check the statements in your program, it
5617 can check expressions entered directly into @value{GDBN} for evaluation via
5618 the @code{print} command, for example. As with the working language,
5619 @value{GDBN} can also decide whether or not to check automatically based on
5620 your program's source language. @xref{Support, ,Supported languages},
5621 for the default settings of supported languages.
5624 * Type Checking:: An overview of type checking
5625 * Range Checking:: An overview of range checking
5628 @cindex type checking
5629 @cindex checks, type
5631 @subsection An overview of type checking
5633 Some languages, such as Modula-2, are strongly typed, meaning that the
5634 arguments to operators and functions have to be of the correct type,
5635 otherwise an error occurs. These checks prevent type mismatch
5636 errors from ever causing any run-time problems. For example,
5644 The second example fails because the @code{CARDINAL} 1 is not
5645 type-compatible with the @code{REAL} 2.3.
5647 For the expressions you use in @value{GDBN} commands, you can tell the
5648 @value{GDBN} type checker to skip checking;
5649 to treat any mismatches as errors and abandon the expression;
5650 or to only issue warnings when type mismatches occur,
5651 but evaluate the expression anyway. When you choose the last of
5652 these, @value{GDBN} evaluates expressions like the second example above, but
5653 also issues a warning.
5655 Even if you turn type checking off, there may be other reasons
5656 related to type that prevent @value{GDBN} from evaluating an expression.
5657 For instance, @value{GDBN} does not know how to add an @code{int} and
5658 a @code{struct foo}. These particular type errors have nothing to do
5659 with the language in use, and usually arise from expressions, such as
5660 the one described above, which make little sense to evaluate anyway.
5662 Each language defines to what degree it is strict about type. For
5663 instance, both Modula-2 and C require the arguments to arithmetical
5664 operators to be numbers. In C, enumerated types and pointers can be
5665 represented as numbers, so that they are valid arguments to mathematical
5666 operators. @xref{Support, ,Supported languages}, for further
5667 details on specific languages.
5669 @value{GDBN} provides some additional commands for controlling the type checker:
5671 @kindex set check@r{, type}
5672 @kindex set check type
5673 @kindex show check type
5675 @item set check type auto
5676 Set type checking on or off based on the current working language.
5677 @xref{Support, ,Supported languages}, for the default settings for
5680 @item set check type on
5681 @itemx set check type off
5682 Set type checking on or off, overriding the default setting for the
5683 current working language. Issue a warning if the setting does not
5684 match the language default. If any type mismatches occur in
5685 evaluating an expression while type checking is on, @value{GDBN} prints a
5686 message and aborts evaluation of the expression.
5688 @item set check type warn
5689 Cause the type checker to issue warnings, but to always attempt to
5690 evaluate the expression. Evaluating the expression may still
5691 be impossible for other reasons. For example, @value{GDBN} cannot add
5692 numbers and structures.
5695 Show the current setting of the type checker, and whether or not @value{GDBN}
5696 is setting it automatically.
5699 @cindex range checking
5700 @cindex checks, range
5701 @node Range Checking
5702 @subsection An overview of range checking
5704 In some languages (such as Modula-2), it is an error to exceed the
5705 bounds of a type; this is enforced with run-time checks. Such range
5706 checking is meant to ensure program correctness by making sure
5707 computations do not overflow, or indices on an array element access do
5708 not exceed the bounds of the array.
5710 For expressions you use in @value{GDBN} commands, you can tell
5711 @value{GDBN} to treat range errors in one of three ways: ignore them,
5712 always treat them as errors and abandon the expression, or issue
5713 warnings but evaluate the expression anyway.
5715 A range error can result from numerical overflow, from exceeding an
5716 array index bound, or when you type a constant that is not a member
5717 of any type. Some languages, however, do not treat overflows as an
5718 error. In many implementations of C, mathematical overflow causes the
5719 result to ``wrap around'' to lower values---for example, if @var{m} is
5720 the largest integer value, and @var{s} is the smallest, then
5723 @var{m} + 1 @result{} @var{s}
5726 This, too, is specific to individual languages, and in some cases
5727 specific to individual compilers or machines. @xref{Support, ,
5728 Supported languages}, for further details on specific languages.
5730 @value{GDBN} provides some additional commands for controlling the range checker:
5732 @kindex set check@r{, range}
5733 @kindex set check range
5734 @kindex show check range
5736 @item set check range auto
5737 Set range checking on or off based on the current working language.
5738 @xref{Support, ,Supported languages}, for the default settings for
5741 @item set check range on
5742 @itemx set check range off
5743 Set range checking on or off, overriding the default setting for the
5744 current working language. A warning is issued if the setting does not
5745 match the language default. If a range error occurs and range checking is on,
5746 then a message is printed and evaluation of the expression is aborted.
5748 @item set check range warn
5749 Output messages when the @value{GDBN} range checker detects a range error,
5750 but attempt to evaluate the expression anyway. Evaluating the
5751 expression may still be impossible for other reasons, such as accessing
5752 memory that the process does not own (a typical example from many Unix
5756 Show the current setting of the range checker, and whether or not it is
5757 being set automatically by @value{GDBN}.
5761 @section Supported languages
5763 @value{GDBN} supports C, C++, Fortran, Java, Chill, assembly, and Modula-2.
5764 @c This is false ...
5765 Some @value{GDBN} features may be used in expressions regardless of the
5766 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
5767 and the @samp{@{type@}addr} construct (@pxref{Expressions,
5768 ,Expressions}) can be used with the constructs of any supported
5771 The following sections detail to what degree each source language is
5772 supported by @value{GDBN}. These sections are not meant to be language
5773 tutorials or references, but serve only as a reference guide to what the
5774 @value{GDBN} expression parser accepts, and what input and output
5775 formats should look like for different languages. There are many good
5776 books written on each of these languages; please look to these for a
5777 language reference or tutorial.
5781 * Modula-2:: Modula-2
5786 @subsection C and C++
5789 @cindex expressions in C or C++
5791 Since C and C++ are so closely related, many features of @value{GDBN} apply
5792 to both languages. Whenever this is the case, we discuss those languages
5797 @cindex @sc{gnu} C++
5798 The C++ debugging facilities are jointly implemented by the C++
5799 compiler and @value{GDBN}. Therefore, to debug your C++ code
5800 effectively, you must compile your C++ programs with a supported
5801 C++ compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C++
5802 compiler (@code{aCC}).
5804 For best results when using @sc{gnu} C++, use the stabs debugging
5805 format. You can select that format explicitly with the @code{g++}
5806 command-line options @samp{-gstabs} or @samp{-gstabs+}. See
5807 @ref{Debugging Options,,Options for Debugging Your Program or @sc{gnu}
5808 CC, gcc.info, Using @sc{gnu} CC}, for more information.
5811 * C Operators:: C and C++ operators
5812 * C Constants:: C and C++ constants
5813 * C plus plus expressions:: C++ expressions
5814 * C Defaults:: Default settings for C and C++
5815 * C Checks:: C and C++ type and range checks
5816 * Debugging C:: @value{GDBN} and C
5817 * Debugging C plus plus:: @value{GDBN} features for C++
5821 @subsubsection C and C++ operators
5823 @cindex C and C++ operators
5825 Operators must be defined on values of specific types. For instance,
5826 @code{+} is defined on numbers, but not on structures. Operators are
5827 often defined on groups of types.
5829 For the purposes of C and C++, the following definitions hold:
5834 @emph{Integral types} include @code{int} with any of its storage-class
5835 specifiers; @code{char}; @code{enum}; and, for C++, @code{bool}.
5838 @emph{Floating-point types} include @code{float}, @code{double}, and
5839 @code{long double} (if supported by the target platform).
5842 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
5845 @emph{Scalar types} include all of the above.
5850 The following operators are supported. They are listed here
5851 in order of increasing precedence:
5855 The comma or sequencing operator. Expressions in a comma-separated list
5856 are evaluated from left to right, with the result of the entire
5857 expression being the last expression evaluated.
5860 Assignment. The value of an assignment expression is the value
5861 assigned. Defined on scalar types.
5864 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
5865 and translated to @w{@code{@var{a} = @var{a op b}}}.
5866 @w{@code{@var{op}=}} and @code{=} have the same precedence.
5867 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
5868 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
5871 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
5872 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
5876 Logical @sc{or}. Defined on integral types.
5879 Logical @sc{and}. Defined on integral types.
5882 Bitwise @sc{or}. Defined on integral types.
5885 Bitwise exclusive-@sc{or}. Defined on integral types.
5888 Bitwise @sc{and}. Defined on integral types.
5891 Equality and inequality. Defined on scalar types. The value of these
5892 expressions is 0 for false and non-zero for true.
5894 @item <@r{, }>@r{, }<=@r{, }>=
5895 Less than, greater than, less than or equal, greater than or equal.
5896 Defined on scalar types. The value of these expressions is 0 for false
5897 and non-zero for true.
5900 left shift, and right shift. Defined on integral types.
5903 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5906 Addition and subtraction. Defined on integral types, floating-point types and
5909 @item *@r{, }/@r{, }%
5910 Multiplication, division, and modulus. Multiplication and division are
5911 defined on integral and floating-point types. Modulus is defined on
5915 Increment and decrement. When appearing before a variable, the
5916 operation is performed before the variable is used in an expression;
5917 when appearing after it, the variable's value is used before the
5918 operation takes place.
5921 Pointer dereferencing. Defined on pointer types. Same precedence as
5925 Address operator. Defined on variables. Same precedence as @code{++}.
5927 For debugging C++, @value{GDBN} implements a use of @samp{&} beyond what is
5928 allowed in the C++ language itself: you can use @samp{&(&@var{ref})}
5929 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
5930 where a C++ reference variable (declared with @samp{&@var{ref}}) is
5934 Negative. Defined on integral and floating-point types. Same
5935 precedence as @code{++}.
5938 Logical negation. Defined on integral types. Same precedence as
5942 Bitwise complement operator. Defined on integral types. Same precedence as
5947 Structure member, and pointer-to-structure member. For convenience,
5948 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
5949 pointer based on the stored type information.
5950 Defined on @code{struct} and @code{union} data.
5953 Dereferences of pointers to members.
5956 Array indexing. @code{@var{a}[@var{i}]} is defined as
5957 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
5960 Function parameter list. Same precedence as @code{->}.
5963 C++ scope resolution operator. Defined on @code{struct}, @code{union},
5964 and @code{class} types.
5967 Doubled colons also represent the @value{GDBN} scope operator
5968 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
5972 If an operator is redefined in the user code, @value{GDBN} usually
5973 attempts to invoke the redefined version instead of using the operator's
5981 @subsubsection C and C++ constants
5983 @cindex C and C++ constants
5985 @value{GDBN} allows you to express the constants of C and C++ in the
5990 Integer constants are a sequence of digits. Octal constants are
5991 specified by a leading @samp{0} (i.e. zero), and hexadecimal constants by
5992 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
5993 @samp{l}, specifying that the constant should be treated as a
5997 Floating point constants are a sequence of digits, followed by a decimal
5998 point, followed by a sequence of digits, and optionally followed by an
5999 exponent. An exponent is of the form:
6000 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
6001 sequence of digits. The @samp{+} is optional for positive exponents.
6002 A floating-point constant may also end with a letter @samp{f} or
6003 @samp{F}, specifying that the constant should be treated as being of
6004 the @code{float} (as opposed to the default @code{double}) type; or with
6005 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
6009 Enumerated constants consist of enumerated identifiers, or their
6010 integral equivalents.
6013 Character constants are a single character surrounded by single quotes
6014 (@code{'}), or a number---the ordinal value of the corresponding character
6015 (usually its @sc{ascii} value). Within quotes, the single character may
6016 be represented by a letter or by @dfn{escape sequences}, which are of
6017 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
6018 of the character's ordinal value; or of the form @samp{\@var{x}}, where
6019 @samp{@var{x}} is a predefined special character---for example,
6020 @samp{\n} for newline.
6023 String constants are a sequence of character constants surrounded by
6024 double quotes (@code{"}). Any valid character constant (as described
6025 above) may appear. Double quotes within the string must be preceded by
6026 a backslash, so for instance @samp{"a\"b'c"} is a string of five
6030 Pointer constants are an integral value. You can also write pointers
6031 to constants using the C operator @samp{&}.
6034 Array constants are comma-separated lists surrounded by braces @samp{@{}
6035 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
6036 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
6037 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
6041 * C plus plus expressions::
6048 @node C plus plus expressions
6049 @subsubsection C++ expressions
6051 @cindex expressions in C++
6052 @value{GDBN} expression handling can interpret most C++ expressions.
6054 @cindex C++ support, not in @sc{coff}
6055 @cindex @sc{coff} versus C++
6056 @cindex C++ and object formats
6057 @cindex object formats and C++
6058 @cindex a.out and C++
6059 @cindex @sc{ecoff} and C++
6060 @cindex @sc{xcoff} and C++
6061 @cindex @sc{elf}/stabs and C++
6062 @cindex @sc{elf}/@sc{dwarf} and C++
6063 @c FIXME!! GDB may eventually be able to debug C++ using DWARF; check
6064 @c periodically whether this has happened...
6066 @emph{Warning:} @value{GDBN} can only debug C++ code if you use the
6067 proper compiler. Typically, C++ debugging depends on the use of
6068 additional debugging information in the symbol table, and thus requires
6069 special support. In particular, if your compiler generates a.out, MIPS
6070 @sc{ecoff}, RS/6000 @sc{xcoff}, or @sc{elf} with stabs extensions to the
6071 symbol table, these facilities are all available. (With @sc{gnu} CC,
6072 you can use the @samp{-gstabs} option to request stabs debugging
6073 extensions explicitly.) Where the object code format is standard
6074 @sc{coff} or @sc{dwarf} in @sc{elf}, on the other hand, most of the C++
6075 support in @value{GDBN} does @emph{not} work.
6080 @cindex member functions
6082 Member function calls are allowed; you can use expressions like
6085 count = aml->GetOriginal(x, y)
6089 @cindex namespace in C++
6091 While a member function is active (in the selected stack frame), your
6092 expressions have the same namespace available as the member function;
6093 that is, @value{GDBN} allows implicit references to the class instance
6094 pointer @code{this} following the same rules as C++.
6096 @cindex call overloaded functions
6097 @cindex overloaded functions, calling
6098 @cindex type conversions in C++
6100 You can call overloaded functions; @value{GDBN} resolves the function
6101 call to the right definition, with some restrictions. @value{GDBN} does not
6102 perform overload resolution involving user-defined type conversions,
6103 calls to constructors, or instantiations of templates that do not exist
6104 in the program. It also cannot handle ellipsis argument lists or
6107 It does perform integral conversions and promotions, floating-point
6108 promotions, arithmetic conversions, pointer conversions, conversions of
6109 class objects to base classes, and standard conversions such as those of
6110 functions or arrays to pointers; it requires an exact match on the
6111 number of function arguments.
6113 Overload resolution is always performed, unless you have specified
6114 @code{set overload-resolution off}. @xref{Debugging C plus plus,
6115 ,@value{GDBN} features for C++}.
6117 You must specify @code{set overload-resolution off} in order to use an
6118 explicit function signature to call an overloaded function, as in
6120 p 'foo(char,int)'('x', 13)
6123 The @value{GDBN} command-completion facility can simplify this;
6124 see @ref{Completion, ,Command completion}.
6126 @cindex reference declarations
6128 @value{GDBN} understands variables declared as C++ references; you can use
6129 them in expressions just as you do in C++ source---they are automatically
6132 In the parameter list shown when @value{GDBN} displays a frame, the values of
6133 reference variables are not displayed (unlike other variables); this
6134 avoids clutter, since references are often used for large structures.
6135 The @emph{address} of a reference variable is always shown, unless
6136 you have specified @samp{set print address off}.
6139 @value{GDBN} supports the C++ name resolution operator @code{::}---your
6140 expressions can use it just as expressions in your program do. Since
6141 one scope may be defined in another, you can use @code{::} repeatedly if
6142 necessary, for example in an expression like
6143 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
6144 resolving name scope by reference to source files, in both C and C++
6145 debugging (@pxref{Variables, ,Program variables}).
6148 In addition, when used with HP's C++ compiler, @value{GDBN} supports
6149 calling virtual functions correctly, printing out virtual bases of
6150 objects, calling functions in a base subobject, casting objects, and
6151 invoking user-defined operators.
6154 @subsubsection C and C++ defaults
6156 @cindex C and C++ defaults
6158 If you allow @value{GDBN} to set type and range checking automatically, they
6159 both default to @code{off} whenever the working language changes to
6160 C or C++. This happens regardless of whether you or @value{GDBN}
6161 selects the working language.
6163 If you allow @value{GDBN} to set the language automatically, it
6164 recognizes source files whose names end with @file{.c}, @file{.C}, or
6165 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
6166 these files, it sets the working language to C or C++.
6167 @xref{Automatically, ,Having @value{GDBN} infer the source language},
6168 for further details.
6170 @c Type checking is (a) primarily motivated by Modula-2, and (b)
6171 @c unimplemented. If (b) changes, it might make sense to let this node
6172 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
6175 @subsubsection C and C++ type and range checks
6177 @cindex C and C++ checks
6179 By default, when @value{GDBN} parses C or C++ expressions, type checking
6180 is not used. However, if you turn type checking on, @value{GDBN}
6181 considers two variables type equivalent if:
6185 The two variables are structured and have the same structure, union, or
6189 The two variables have the same type name, or types that have been
6190 declared equivalent through @code{typedef}.
6193 @c leaving this out because neither J Gilmore nor R Pesch understand it.
6196 The two @code{struct}, @code{union}, or @code{enum} variables are
6197 declared in the same declaration. (Note: this may not be true for all C
6202 Range checking, if turned on, is done on mathematical operations. Array
6203 indices are not checked, since they are often used to index a pointer
6204 that is not itself an array.
6207 @subsubsection @value{GDBN} and C
6209 The @code{set print union} and @code{show print union} commands apply to
6210 the @code{union} type. When set to @samp{on}, any @code{union} that is
6211 inside a @code{struct} or @code{class} is also printed. Otherwise, it
6212 appears as @samp{@{...@}}.
6214 The @code{@@} operator aids in the debugging of dynamic arrays, formed
6215 with pointers and a memory allocation function. @xref{Expressions,
6219 * Debugging C plus plus::
6222 @node Debugging C plus plus
6223 @subsubsection @value{GDBN} features for C++
6225 @cindex commands for C++
6227 Some @value{GDBN} commands are particularly useful with C++, and some are
6228 designed specifically for use with C++. Here is a summary:
6231 @cindex break in overloaded functions
6232 @item @r{breakpoint menus}
6233 When you want a breakpoint in a function whose name is overloaded,
6234 @value{GDBN} breakpoint menus help you specify which function definition
6235 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
6237 @cindex overloading in C++
6238 @item rbreak @var{regex}
6239 Setting breakpoints using regular expressions is helpful for setting
6240 breakpoints on overloaded functions that are not members of any special
6242 @xref{Set Breaks, ,Setting breakpoints}.
6244 @cindex C++ exception handling
6247 Debug C++ exception handling using these commands. @xref{Set
6248 Catchpoints, , Setting catchpoints}.
6251 @item ptype @var{typename}
6252 Print inheritance relationships as well as other information for type
6254 @xref{Symbols, ,Examining the Symbol Table}.
6256 @cindex C++ symbol display
6257 @item set print demangle
6258 @itemx show print demangle
6259 @itemx set print asm-demangle
6260 @itemx show print asm-demangle
6261 Control whether C++ symbols display in their source form, both when
6262 displaying code as C++ source and when displaying disassemblies.
6263 @xref{Print Settings, ,Print settings}.
6265 @item set print object
6266 @itemx show print object
6267 Choose whether to print derived (actual) or declared types of objects.
6268 @xref{Print Settings, ,Print settings}.
6270 @item set print vtbl
6271 @itemx show print vtbl
6272 Control the format for printing virtual function tables.
6273 @xref{Print Settings, ,Print settings}.
6274 (The @code{vtbl} commands do not work on programs compiled with the HP
6275 ANSI C++ compiler (@code{aCC}).)
6277 @kindex set overload-resolution
6278 @cindex overloaded functions, overload resolution
6279 @item set overload-resolution on
6280 Enable overload resolution for C++ expression evaluation. The default
6281 is on. For overloaded functions, @value{GDBN} evaluates the arguments
6282 and searches for a function whose signature matches the argument types,
6283 using the standard C++ conversion rules (see @ref{C plus plus expressions, ,C++
6284 expressions}, for details). If it cannot find a match, it emits a
6287 @item set overload-resolution off
6288 Disable overload resolution for C++ expression evaluation. For
6289 overloaded functions that are not class member functions, @value{GDBN}
6290 chooses the first function of the specified name that it finds in the
6291 symbol table, whether or not its arguments are of the correct type. For
6292 overloaded functions that are class member functions, @value{GDBN}
6293 searches for a function whose signature @emph{exactly} matches the
6296 @item @r{Overloaded symbol names}
6297 You can specify a particular definition of an overloaded symbol, using
6298 the same notation that is used to declare such symbols in C++: type
6299 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
6300 also use the @value{GDBN} command-line word completion facilities to list the
6301 available choices, or to finish the type list for you.
6302 @xref{Completion,, Command completion}, for details on how to do this.
6306 @subsection Modula-2
6308 @cindex Modula-2, @value{GDBN} support
6310 The extensions made to @value{GDBN} to support Modula-2 only support
6311 output from the @sc{gnu} Modula-2 compiler (which is currently being
6312 developed). Other Modula-2 compilers are not currently supported, and
6313 attempting to debug executables produced by them is most likely
6314 to give an error as @value{GDBN} reads in the executable's symbol
6317 @cindex expressions in Modula-2
6319 * M2 Operators:: Built-in operators
6320 * Built-In Func/Proc:: Built-in functions and procedures
6321 * M2 Constants:: Modula-2 constants
6322 * M2 Defaults:: Default settings for Modula-2
6323 * Deviations:: Deviations from standard Modula-2
6324 * M2 Checks:: Modula-2 type and range checks
6325 * M2 Scope:: The scope operators @code{::} and @code{.}
6326 * GDB/M2:: @value{GDBN} and Modula-2
6330 @subsubsection Operators
6331 @cindex Modula-2 operators
6333 Operators must be defined on values of specific types. For instance,
6334 @code{+} is defined on numbers, but not on structures. Operators are
6335 often defined on groups of types. For the purposes of Modula-2, the
6336 following definitions hold:
6341 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
6345 @emph{Character types} consist of @code{CHAR} and its subranges.
6348 @emph{Floating-point types} consist of @code{REAL}.
6351 @emph{Pointer types} consist of anything declared as @code{POINTER TO
6355 @emph{Scalar types} consist of all of the above.
6358 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
6361 @emph{Boolean types} consist of @code{BOOLEAN}.
6365 The following operators are supported, and appear in order of
6366 increasing precedence:
6370 Function argument or array index separator.
6373 Assignment. The value of @var{var} @code{:=} @var{value} is
6377 Less than, greater than on integral, floating-point, or enumerated
6381 Less than or equal to, greater than or equal to
6382 on integral, floating-point and enumerated types, or set inclusion on
6383 set types. Same precedence as @code{<}.
6385 @item =@r{, }<>@r{, }#
6386 Equality and two ways of expressing inequality, valid on scalar types.
6387 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
6388 available for inequality, since @code{#} conflicts with the script
6392 Set membership. Defined on set types and the types of their members.
6393 Same precedence as @code{<}.
6396 Boolean disjunction. Defined on boolean types.
6399 Boolean conjunction. Defined on boolean types.
6402 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
6405 Addition and subtraction on integral and floating-point types, or union
6406 and difference on set types.
6409 Multiplication on integral and floating-point types, or set intersection
6413 Division on floating-point types, or symmetric set difference on set
6414 types. Same precedence as @code{*}.
6417 Integer division and remainder. Defined on integral types. Same
6418 precedence as @code{*}.
6421 Negative. Defined on @code{INTEGER} and @code{REAL} data.
6424 Pointer dereferencing. Defined on pointer types.
6427 Boolean negation. Defined on boolean types. Same precedence as
6431 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
6432 precedence as @code{^}.
6435 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
6438 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
6442 @value{GDBN} and Modula-2 scope operators.
6446 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
6447 treats the use of the operator @code{IN}, or the use of operators
6448 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
6449 @code{<=}, and @code{>=} on sets as an error.
6452 @cindex Modula-2 built-ins
6453 @node Built-In Func/Proc
6454 @subsubsection Built-in functions and procedures
6456 Modula-2 also makes available several built-in procedures and functions.
6457 In describing these, the following metavariables are used:
6462 represents an @code{ARRAY} variable.
6465 represents a @code{CHAR} constant or variable.
6468 represents a variable or constant of integral type.
6471 represents an identifier that belongs to a set. Generally used in the
6472 same function with the metavariable @var{s}. The type of @var{s} should
6473 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
6476 represents a variable or constant of integral or floating-point type.
6479 represents a variable or constant of floating-point type.
6485 represents a variable.
6488 represents a variable or constant of one of many types. See the
6489 explanation of the function for details.
6492 All Modula-2 built-in procedures also return a result, described below.
6496 Returns the absolute value of @var{n}.
6499 If @var{c} is a lower case letter, it returns its upper case
6500 equivalent, otherwise it returns its argument.
6503 Returns the character whose ordinal value is @var{i}.
6506 Decrements the value in the variable @var{v} by one. Returns the new value.
6508 @item DEC(@var{v},@var{i})
6509 Decrements the value in the variable @var{v} by @var{i}. Returns the
6512 @item EXCL(@var{m},@var{s})
6513 Removes the element @var{m} from the set @var{s}. Returns the new
6516 @item FLOAT(@var{i})
6517 Returns the floating point equivalent of the integer @var{i}.
6520 Returns the index of the last member of @var{a}.
6523 Increments the value in the variable @var{v} by one. Returns the new value.
6525 @item INC(@var{v},@var{i})
6526 Increments the value in the variable @var{v} by @var{i}. Returns the
6529 @item INCL(@var{m},@var{s})
6530 Adds the element @var{m} to the set @var{s} if it is not already
6531 there. Returns the new set.
6534 Returns the maximum value of the type @var{t}.
6537 Returns the minimum value of the type @var{t}.
6540 Returns boolean TRUE if @var{i} is an odd number.
6543 Returns the ordinal value of its argument. For example, the ordinal
6544 value of a character is its @sc{ascii} value (on machines supporting the
6545 @sc{ascii} character set). @var{x} must be of an ordered type, which include
6546 integral, character and enumerated types.
6549 Returns the size of its argument. @var{x} can be a variable or a type.
6551 @item TRUNC(@var{r})
6552 Returns the integral part of @var{r}.
6554 @item VAL(@var{t},@var{i})
6555 Returns the member of the type @var{t} whose ordinal value is @var{i}.
6559 @emph{Warning:} Sets and their operations are not yet supported, so
6560 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
6564 @cindex Modula-2 constants
6566 @subsubsection Constants
6568 @value{GDBN} allows you to express the constants of Modula-2 in the following
6574 Integer constants are simply a sequence of digits. When used in an
6575 expression, a constant is interpreted to be type-compatible with the
6576 rest of the expression. Hexadecimal integers are specified by a
6577 trailing @samp{H}, and octal integers by a trailing @samp{B}.
6580 Floating point constants appear as a sequence of digits, followed by a
6581 decimal point and another sequence of digits. An optional exponent can
6582 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
6583 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
6584 digits of the floating point constant must be valid decimal (base 10)
6588 Character constants consist of a single character enclosed by a pair of
6589 like quotes, either single (@code{'}) or double (@code{"}). They may
6590 also be expressed by their ordinal value (their @sc{ascii} value, usually)
6591 followed by a @samp{C}.
6594 String constants consist of a sequence of characters enclosed by a
6595 pair of like quotes, either single (@code{'}) or double (@code{"}).
6596 Escape sequences in the style of C are also allowed. @xref{C
6597 Constants, ,C and C++ constants}, for a brief explanation of escape
6601 Enumerated constants consist of an enumerated identifier.
6604 Boolean constants consist of the identifiers @code{TRUE} and
6608 Pointer constants consist of integral values only.
6611 Set constants are not yet supported.
6615 @subsubsection Modula-2 defaults
6616 @cindex Modula-2 defaults
6618 If type and range checking are set automatically by @value{GDBN}, they
6619 both default to @code{on} whenever the working language changes to
6620 Modula-2. This happens regardless of whether you or @value{GDBN}
6621 selected the working language.
6623 If you allow @value{GDBN} to set the language automatically, then entering
6624 code compiled from a file whose name ends with @file{.mod} sets the
6625 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
6626 the language automatically}, for further details.
6629 @subsubsection Deviations from standard Modula-2
6630 @cindex Modula-2, deviations from
6632 A few changes have been made to make Modula-2 programs easier to debug.
6633 This is done primarily via loosening its type strictness:
6637 Unlike in standard Modula-2, pointer constants can be formed by
6638 integers. This allows you to modify pointer variables during
6639 debugging. (In standard Modula-2, the actual address contained in a
6640 pointer variable is hidden from you; it can only be modified
6641 through direct assignment to another pointer variable or expression that
6642 returned a pointer.)
6645 C escape sequences can be used in strings and characters to represent
6646 non-printable characters. @value{GDBN} prints out strings with these
6647 escape sequences embedded. Single non-printable characters are
6648 printed using the @samp{CHR(@var{nnn})} format.
6651 The assignment operator (@code{:=}) returns the value of its right-hand
6655 All built-in procedures both modify @emph{and} return their argument.
6659 @subsubsection Modula-2 type and range checks
6660 @cindex Modula-2 checks
6663 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
6666 @c FIXME remove warning when type/range checks added
6668 @value{GDBN} considers two Modula-2 variables type equivalent if:
6672 They are of types that have been declared equivalent via a @code{TYPE
6673 @var{t1} = @var{t2}} statement
6676 They have been declared on the same line. (Note: This is true of the
6677 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
6680 As long as type checking is enabled, any attempt to combine variables
6681 whose types are not equivalent is an error.
6683 Range checking is done on all mathematical operations, assignment, array
6684 index bounds, and all built-in functions and procedures.
6687 @subsubsection The scope operators @code{::} and @code{.}
6690 @cindex colon, doubled as scope operator
6692 @kindex colon-colon@r{, in Modula-2}
6693 @c Info cannot handle :: but TeX can.
6699 There are a few subtle differences between the Modula-2 scope operator
6700 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
6705 @var{module} . @var{id}
6706 @var{scope} :: @var{id}
6710 where @var{scope} is the name of a module or a procedure,
6711 @var{module} the name of a module, and @var{id} is any declared
6712 identifier within your program, except another module.
6714 Using the @code{::} operator makes @value{GDBN} search the scope
6715 specified by @var{scope} for the identifier @var{id}. If it is not
6716 found in the specified scope, then @value{GDBN} searches all scopes
6717 enclosing the one specified by @var{scope}.
6719 Using the @code{.} operator makes @value{GDBN} search the current scope for
6720 the identifier specified by @var{id} that was imported from the
6721 definition module specified by @var{module}. With this operator, it is
6722 an error if the identifier @var{id} was not imported from definition
6723 module @var{module}, or if @var{id} is not an identifier in
6727 @subsubsection @value{GDBN} and Modula-2
6729 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
6730 Five subcommands of @code{set print} and @code{show print} apply
6731 specifically to C and C++: @samp{vtbl}, @samp{demangle},
6732 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
6733 apply to C++, and the last to the C @code{union} type, which has no direct
6734 analogue in Modula-2.
6736 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
6737 with any language, is not useful with Modula-2. Its
6738 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
6739 created in Modula-2 as they can in C or C++. However, because an
6740 address can be specified by an integral constant, the construct
6741 @samp{@{@var{type}@}@var{adrexp}} is still useful.
6743 @cindex @code{#} in Modula-2
6744 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
6745 interpreted as the beginning of a comment. Use @code{<>} instead.
6750 The extensions made to @value{GDBN} to support Chill only support output
6751 from the @sc{gnu} Chill compiler. Other Chill compilers are not currently
6752 supported, and attempting to debug executables produced by them is most
6753 likely to give an error as @value{GDBN} reads in the executable's symbol
6756 @c This used to say "... following Chill related topics ...", but since
6757 @c menus are not shown in the printed manual, it would look awkward.
6758 This section covers the Chill related topics and the features
6759 of @value{GDBN} which support these topics.
6762 * How modes are displayed:: How modes are displayed
6763 * Locations:: Locations and their accesses
6764 * Values and their Operations:: Values and their Operations
6765 * Chill type and range checks::
6769 @node How modes are displayed
6770 @subsubsection How modes are displayed
6772 The Chill Datatype- (Mode) support of @value{GDBN} is directly related
6773 with the functionality of the @sc{gnu} Chill compiler, and therefore deviates
6774 slightly from the standard specification of the Chill language. The
6777 @c FIXME: this @table's contents effectively disable @code by using @r
6778 @c on every @item. So why does it need @code?
6780 @item @r{@emph{Discrete modes:}}
6783 @emph{Integer Modes} which are predefined by @code{BYTE, UBYTE, INT,
6786 @emph{Boolean Mode} which is predefined by @code{BOOL},
6788 @emph{Character Mode} which is predefined by @code{CHAR},
6790 @emph{Set Mode} which is displayed by the keyword @code{SET}.
6792 (@value{GDBP}) ptype x
6793 type = SET (karli = 10, susi = 20, fritzi = 100)
6795 If the type is an unnumbered set the set element values are omitted.
6797 @emph{Range Mode} which is displayed by @code{type = <basemode>
6798 (<lower bound> : <upper bound>)}, where @code{<lower bound>, <upper
6799 bound>} can be of any discrete literal expression (e.g. set element
6803 @item @r{@emph{Powerset Mode:}}
6804 A Powerset Mode is displayed by the keyword @code{POWERSET} followed by
6805 the member mode of the powerset. The member mode can be any discrete mode.
6807 (@value{GDBP}) ptype x
6808 type = POWERSET SET (egon, hugo, otto)
6811 @item @r{@emph{Reference Modes:}}
6814 @emph{Bound Reference Mode} which is displayed by the keyword @code{REF}
6815 followed by the mode name to which the reference is bound.
6817 @emph{Free Reference Mode} which is displayed by the keyword @code{PTR}.
6820 @item @r{@emph{Procedure mode}}
6821 The procedure mode is displayed by @code{type = PROC(<parameter list>)
6822 <return mode> EXCEPTIONS (<exception list>)}. The @code{<parameter
6823 list>} is a list of the parameter modes. @code{<return mode>} indicates
6824 the mode of the result of the procedure if any. The exceptionlist lists
6825 all possible exceptions which can be raised by the procedure.
6828 @item @r{@emph{Instance mode}}
6829 The instance mode is represented by a structure, which has a static
6830 type, and is therefore not really of interest.
6833 @item @r{@emph{Synchronization Modes:}}
6836 @emph{Event Mode} which is displayed by @code{EVENT (<event length>)},
6837 where @code{(<event length>)} is optional.
6839 @emph{Buffer Mode} which is displayed by @code{BUFFER (<buffer length>)
6840 <buffer element mode>}, where @code{(<buffer length>)} is optional.
6843 @item @r{@emph{Timing Modes:}}
6846 @emph{Duration Mode} which is predefined by @code{DURATION}
6848 @emph{Absolute Time Mode} which is predefined by @code{TIME}
6851 @item @r{@emph{Real Modes:}}
6852 Real Modes are predefined with @code{REAL} and @code{LONG_REAL}.
6854 @item @r{@emph{String Modes:}}
6857 @emph{Character String Mode} which is displayed by @code{CHARS(<string
6858 length>)}, followed by the keyword @code{VARYING} if the String Mode is
6861 @emph{Bit String Mode} which is displayed by @code{BOOLS(<string
6865 @item @r{@emph{Array Mode:}}
6866 The Array Mode is displayed by the keyword @code{ARRAY(<range>)}
6867 followed by the element mode (which may in turn be an array mode).
6869 (@value{GDBP}) ptype x
6872 SET (karli = 10, susi = 20, fritzi = 100)
6875 @item @r{@emph{Structure Mode}}
6876 The Structure mode is displayed by the keyword @code{STRUCT(<field
6877 list>)}. The @code{<field list>} consists of names and modes of fields
6878 of the structure. Variant structures have the keyword @code{CASE <field>
6879 OF <variant fields> ESAC} in their field list. Since the current version
6880 of the GNU Chill compiler doesn't implement tag processing (no runtime
6881 checks of variant fields, and therefore no debugging info), the output
6882 always displays all variant fields.
6884 (@value{GDBP}) ptype str
6899 @subsubsection Locations and their accesses
6901 A location in Chill is an object which can contain values.
6903 A value of a location is generally accessed by the (declared) name of
6904 the location. The output conforms to the specification of values in
6905 Chill programs. How values are specified
6906 is the topic of the next section, @ref{Values and their Operations}.
6908 The pseudo-location @code{RESULT} (or @code{result}) can be used to
6909 display or change the result of a currently-active procedure:
6916 This does the same as the Chill action @code{RESULT EXPR} (which
6917 is not available in @value{GDBN}).
6919 Values of reference mode locations are printed by @code{PTR(<hex
6920 value>)} in case of a free reference mode, and by @code{(REF <reference
6921 mode>) (<hex-value>)} in case of a bound reference. @code{<hex value>}
6922 represents the address where the reference points to. To access the
6923 value of the location referenced by the pointer, use the dereference
6926 Values of procedure mode locations are displayed by @code{@{ PROC
6927 (<argument modes> ) <return mode> @} <address> <name of procedure
6928 location>}. @code{<argument modes>} is a list of modes according to the
6929 parameter specification of the procedure and @code{<address>} shows the
6930 address of the entry point.
6933 Locations of instance modes are displayed just like a structure with two
6934 fields specifying the @emph{process type} and the @emph{copy number} of
6935 the investigated instance location@footnote{This comes from the current
6936 implementation of instances. They are implemented as a structure (no
6937 na). The output should be something like @code{[<name of the process>;
6938 <instance number>]}.}. The field names are @code{__proc_type} and
6941 Locations of synchronization modes are displayed like a structure with
6942 the field name @code{__event_data} in case of a event mode location, and
6943 like a structure with the field @code{__buffer_data} in case of a buffer
6944 mode location (refer to previous paragraph).
6946 Structure Mode locations are printed by @code{[.<field name>: <value>,
6947 ...]}. The @code{<field name>} corresponds to the structure mode
6948 definition and the layout of @code{<value>} varies depending of the mode
6949 of the field. If the investigated structure mode location is of variant
6950 structure mode, the variant parts of the structure are enclosed in curled
6951 braces (@samp{@{@}}). Fields enclosed by @samp{@{,@}} are residing
6952 on the same memory location and represent the current values of the
6953 memory location in their specific modes. Since no tag processing is done
6954 all variants are displayed. A variant field is printed by
6955 @code{(<variant name>) = .<field name>: <value>}. (who implements the
6958 (@value{GDBP}) print str1 $4 = [.as: 0, .bs: karli, .<TAG>: { (karli) =
6959 [.cs: []], (susi) = [.ds: susi]}]
6963 Substructures of string mode-, array mode- or structure mode-values
6964 (e.g. array slices, fields of structure locations) are accessed using
6965 certain operations which are described in the next section, @ref{Values
6966 and their Operations}.
6968 A location value may be interpreted as having a different mode using the
6969 location conversion. This mode conversion is written as @code{<mode
6970 name>(<location>)}. The user has to consider that the sizes of the modes
6971 have to be equal otherwise an error occurs. Furthermore, no range
6972 checking of the location against the destination mode is performed, and
6973 therefore the result can be quite confusing.
6976 (@value{GDBP}) print int (s(3 up 4)) XXX TO be filled in !! XXX
6979 @node Values and their Operations
6980 @subsubsection Values and their Operations
6982 Values are used to alter locations, to investigate complex structures in
6983 more detail or to filter relevant information out of a large amount of
6984 data. There are several (mode dependent) operations defined which enable
6985 such investigations. These operations are not only applicable to
6986 constant values but also to locations, which can become quite useful
6987 when debugging complex structures. During parsing the command line
6988 (e.g. evaluating an expression) @value{GDBN} treats location names as
6989 the values behind these locations.
6991 This section describes how values have to be specified and which
6992 operations are legal to be used with such values.
6995 @item Literal Values
6996 Literal values are specified in the same manner as in @sc{gnu} Chill programs.
6997 For detailed specification refer to the @sc{gnu} Chill implementation Manual
6999 @c FIXME: if the Chill Manual is a Texinfo documents, the above should
7000 @c be converted to a @ref.
7005 @emph{Integer Literals} are specified in the same manner as in Chill
7006 programs (refer to the Chill Standard z200/88 chpt 5.2.4.2)
7008 @emph{Boolean Literals} are defined by @code{TRUE} and @code{FALSE}.
7010 @emph{Character Literals} are defined by @code{'<character>'}. (e.g.
7013 @emph{Set Literals} are defined by a name which was specified in a set
7014 mode. The value delivered by a Set Literal is the set value. This is
7015 comparable to an enumeration in C/C++ language.
7017 @emph{Emptiness Literal} is predefined by @code{NULL}. The value of the
7018 emptiness literal delivers either the empty reference value, the empty
7019 procedure value or the empty instance value.
7022 @emph{Character String Literals} are defined by a sequence of characters
7023 enclosed in single- or double quotes. If a single- or double quote has
7024 to be part of the string literal it has to be stuffed (specified twice).
7026 @emph{Bitstring Literals} are specified in the same manner as in Chill
7027 programs (refer z200/88 chpt 5.2.4.8).
7029 @emph{Floating point literals} are specified in the same manner as in
7030 (gnu-)Chill programs (refer @sc{gnu} Chill implementation Manual chapter 1.5).
7035 A tuple is specified by @code{<mode name>[<tuple>]}, where @code{<mode
7036 name>} can be omitted if the mode of the tuple is unambiguous. This
7037 unambiguity is derived from the context of a evaluated expression.
7038 @code{<tuple>} can be one of the following:
7041 @item @emph{Powerset Tuple}
7042 @item @emph{Array Tuple}
7043 @item @emph{Structure Tuple}
7044 Powerset tuples, array tuples and structure tuples are specified in the
7045 same manner as in Chill programs refer to z200/88 chpt 5.2.5.
7048 @item String Element Value
7049 A string element value is specified by @code{<string value>(<index>)},
7050 where @code{<index>} is a integer expression. It delivers a character
7051 value which is equivalent to the character indexed by @code{<index>} in
7054 @item String Slice Value
7055 A string slice value is specified by @code{<string value>(<slice
7056 spec>)}, where @code{<slice spec>} can be either a range of integer
7057 expressions or specified by @code{<start expr> up <size>}.
7058 @code{<size>} denotes the number of elements which the slice contains.
7059 The delivered value is a string value, which is part of the specified
7062 @item Array Element Values
7063 An array element value is specified by @code{<array value>(<expr>)} and
7064 delivers a array element value of the mode of the specified array.
7066 @item Array Slice Values
7067 An array slice is specified by @code{<array value>(<slice spec>)}, where
7068 @code{<slice spec>} can be either a range specified by expressions or by
7069 @code{<start expr> up <size>}. @code{<size>} denotes the number of
7070 arrayelements the slice contains. The delivered value is an array value
7071 which is part of the specified array.
7073 @item Structure Field Values
7074 A structure field value is derived by @code{<structure value>.<field
7075 name>}, where @code{<field name>} indicates the name of a field specified
7076 in the mode definition of the structure. The mode of the delivered value
7077 corresponds to this mode definition in the structure definition.
7079 @item Procedure Call Value
7080 The procedure call value is derived from the return value of the
7081 procedure@footnote{If a procedure call is used for instance in an
7082 expression, then this procedure is called with all its side
7083 effects. This can lead to confusing results if used carelessly.}.
7085 Values of duration mode locations are represented by @code{ULONG} literals.
7087 Values of time mode locations are represented by @code{TIME(<secs>:<nsecs>)}.
7090 This is not implemented yet:
7091 @item Built-in Value
7093 The following built in functions are provided:
7105 @item @code{UPPER()}
7106 @item @code{LOWER()}
7107 @item @code{LENGTH()}
7111 @item @code{ARCSIN()}
7112 @item @code{ARCCOS()}
7113 @item @code{ARCTAN()}
7120 For a detailed description refer to the GNU Chill implementation manual
7124 @item Zero-adic Operator Value
7125 The zero-adic operator value is derived from the instance value for the
7126 current active process.
7128 @item Expression Values
7129 The value delivered by an expression is the result of the evaluation of
7130 the specified expression. If there are error conditions (mode
7131 incompatibility, etc.) the evaluation of expressions is aborted with a
7132 corresponding error message. Expressions may be parenthesised which
7133 causes the evaluation of this expression before any other expression
7134 which uses the result of the parenthesised expression. The following
7135 operators are supported by @value{GDBN}:
7138 @item @code{OR, ORIF, XOR}
7139 @itemx @code{AND, ANDIF}
7141 Logical operators defined over operands of boolean mode.
7144 Equality and inequality operators defined over all modes.
7148 Relational operators defined over predefined modes.
7151 @itemx @code{*, /, MOD, REM}
7152 Arithmetic operators defined over predefined modes.
7155 Change sign operator.
7158 String concatenation operator.
7161 String repetition operator.
7164 Referenced location operator which can be used either to take the
7165 address of a location (@code{->loc}), or to dereference a reference
7166 location (@code{loc->}).
7168 @item @code{OR, XOR}
7171 Powerset and bitstring operators.
7175 Powerset inclusion operators.
7178 Membership operator.
7182 @node Chill type and range checks
7183 @subsubsection Chill type and range checks
7185 @value{GDBN} considers two Chill variables mode equivalent if the sizes
7186 of the two modes are equal. This rule applies recursively to more
7187 complex datatypes which means that complex modes are treated
7188 equivalent if all element modes (which also can be complex modes like
7189 structures, arrays, etc.) have the same size.
7191 Range checking is done on all mathematical operations, assignment, array
7192 index bounds and all built in procedures.
7194 Strong type checks are forced using the @value{GDBN} command @code{set
7195 check strong}. This enforces strong type and range checks on all
7196 operations where Chill constructs are used (expressions, built in
7197 functions, etc.) in respect to the semantics as defined in the z.200
7198 language specification.
7200 All checks can be disabled by the @value{GDBN} command @code{set check
7204 @c Deviations from the Chill Standard Z200/88
7205 see last paragraph ?
7208 @node Chill defaults
7209 @subsubsection Chill defaults
7211 If type and range checking are set automatically by @value{GDBN}, they
7212 both default to @code{on} whenever the working language changes to
7213 Chill. This happens regardless of whether you or @value{GDBN}
7214 selected the working language.
7216 If you allow @value{GDBN} to set the language automatically, then entering
7217 code compiled from a file whose name ends with @file{.ch} sets the
7218 working language to Chill. @xref{Automatically, ,Having @value{GDBN} set
7219 the language automatically}, for further details.
7222 @chapter Examining the Symbol Table
7224 The commands described in this chapter allow you to inquire about the
7225 symbols (names of variables, functions and types) defined in your
7226 program. This information is inherent in the text of your program and
7227 does not change as your program executes. @value{GDBN} finds it in your
7228 program's symbol table, in the file indicated when you started @value{GDBN}
7229 (@pxref{File Options, ,Choosing files}), or by one of the
7230 file-management commands (@pxref{Files, ,Commands to specify files}).
7232 @cindex symbol names
7233 @cindex names of symbols
7234 @cindex quoting names
7235 Occasionally, you may need to refer to symbols that contain unusual
7236 characters, which @value{GDBN} ordinarily treats as word delimiters. The
7237 most frequent case is in referring to static variables in other
7238 source files (@pxref{Variables,,Program variables}). File names
7239 are recorded in object files as debugging symbols, but @value{GDBN} would
7240 ordinarily parse a typical file name, like @file{foo.c}, as the three words
7241 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
7242 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
7249 looks up the value of @code{x} in the scope of the file @file{foo.c}.
7252 @kindex info address
7253 @item info address @var{symbol}
7254 Describe where the data for @var{symbol} is stored. For a register
7255 variable, this says which register it is kept in. For a non-register
7256 local variable, this prints the stack-frame offset at which the variable
7259 Note the contrast with @samp{print &@var{symbol}}, which does not work
7260 at all for a register variable, and for a stack local variable prints
7261 the exact address of the current instantiation of the variable.
7264 @item whatis @var{expr}
7265 Print the data type of expression @var{expr}. @var{expr} is not
7266 actually evaluated, and any side-effecting operations (such as
7267 assignments or function calls) inside it do not take place.
7268 @xref{Expressions, ,Expressions}.
7271 Print the data type of @code{$}, the last value in the value history.
7274 @item ptype @var{typename}
7275 Print a description of data type @var{typename}. @var{typename} may be
7276 the name of a type, or for C code it may have the form @samp{class
7277 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
7278 @var{union-tag}} or @samp{enum @var{enum-tag}}.
7280 @item ptype @var{expr}
7282 Print a description of the type of expression @var{expr}. @code{ptype}
7283 differs from @code{whatis} by printing a detailed description, instead
7284 of just the name of the type.
7286 For example, for this variable declaration:
7289 struct complex @{double real; double imag;@} v;
7293 the two commands give this output:
7297 (@value{GDBP}) whatis v
7298 type = struct complex
7299 (@value{GDBP}) ptype v
7300 type = struct complex @{
7308 As with @code{whatis}, using @code{ptype} without an argument refers to
7309 the type of @code{$}, the last value in the value history.
7312 @item info types @var{regexp}
7314 Print a brief description of all types whose names match @var{regexp}
7315 (or all types in your program, if you supply no argument). Each
7316 complete typename is matched as though it were a complete line; thus,
7317 @samp{i type value} gives information on all types in your program whose
7318 names include the string @code{value}, but @samp{i type ^value$} gives
7319 information only on types whose complete name is @code{value}.
7321 This command differs from @code{ptype} in two ways: first, like
7322 @code{whatis}, it does not print a detailed description; second, it
7323 lists all source files where a type is defined.
7327 Show the name of the current source file---that is, the source file for
7328 the function containing the current point of execution---and the language
7331 @kindex info sources
7333 Print the names of all source files in your program for which there is
7334 debugging information, organized into two lists: files whose symbols
7335 have already been read, and files whose symbols will be read when needed.
7337 @kindex info functions
7338 @item info functions
7339 Print the names and data types of all defined functions.
7341 @item info functions @var{regexp}
7342 Print the names and data types of all defined functions
7343 whose names contain a match for regular expression @var{regexp}.
7344 Thus, @samp{info fun step} finds all functions whose names
7345 include @code{step}; @samp{info fun ^step} finds those whose names
7346 start with @code{step}.
7348 @kindex info variables
7349 @item info variables
7350 Print the names and data types of all variables that are declared
7351 outside of functions (i.e., excluding local variables).
7353 @item info variables @var{regexp}
7354 Print the names and data types of all variables (except for local
7355 variables) whose names contain a match for regular expression
7359 This was never implemented.
7360 @kindex info methods
7362 @itemx info methods @var{regexp}
7363 The @code{info methods} command permits the user to examine all defined
7364 methods within C++ program, or (with the @var{regexp} argument) a
7365 specific set of methods found in the various C++ classes. Many
7366 C++ classes provide a large number of methods. Thus, the output
7367 from the @code{ptype} command can be overwhelming and hard to use. The
7368 @code{info-methods} command filters the methods, printing only those
7369 which match the regular-expression @var{regexp}.
7372 @cindex reloading symbols
7373 Some systems allow individual object files that make up your program to
7374 be replaced without stopping and restarting your program. For example,
7375 in VxWorks you can simply recompile a defective object file and keep on
7376 running. If you are running on one of these systems, you can allow
7377 @value{GDBN} to reload the symbols for automatically relinked modules:
7380 @kindex set symbol-reloading
7381 @item set symbol-reloading on
7382 Replace symbol definitions for the corresponding source file when an
7383 object file with a particular name is seen again.
7385 @item set symbol-reloading off
7386 Do not replace symbol definitions when re-encountering object files of
7387 the same name. This is the default state; if you are not running on a
7388 system that permits automatically relinking modules, you should leave
7389 @code{symbol-reloading} off, since otherwise @value{GDBN} may discard symbols
7390 when linking large programs, that may contain several modules (from
7391 different directories or libraries) with the same name.
7393 @kindex show symbol-reloading
7394 @item show symbol-reloading
7395 Show the current @code{on} or @code{off} setting.
7398 @kindex set opaque-type-resolution
7399 @item set opaque-type-resolution on
7400 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
7401 declared as a pointer to a @code{struct}, @code{class}, or
7402 @code{union}---for example, @code{struct MyType *}---that is used in one
7403 source file although the full declaration of @code{struct MyType} is in
7404 another source file. The default is on.
7406 A change in the setting of this subcommand will not take effect until
7407 the next time symbols for a file are loaded.
7409 @item set opaque-type-resolution off
7410 Tell @value{GDBN} not to resolve opaque types. In this case, the type
7411 is printed as follows:
7413 @{<no data fields>@}
7416 @kindex show opaque-type-resolution
7417 @item show opaque-type-resolution
7418 Show whether opaque types are resolved or not.
7420 @kindex maint print symbols
7422 @kindex maint print psymbols
7423 @cindex partial symbol dump
7424 @item maint print symbols @var{filename}
7425 @itemx maint print psymbols @var{filename}
7426 @itemx maint print msymbols @var{filename}
7427 Write a dump of debugging symbol data into the file @var{filename}.
7428 These commands are used to debug the @value{GDBN} symbol-reading code. Only
7429 symbols with debugging data are included. If you use @samp{maint print
7430 symbols}, @value{GDBN} includes all the symbols for which it has already
7431 collected full details: that is, @var{filename} reflects symbols for
7432 only those files whose symbols @value{GDBN} has read. You can use the
7433 command @code{info sources} to find out which files these are. If you
7434 use @samp{maint print psymbols} instead, the dump shows information about
7435 symbols that @value{GDBN} only knows partially---that is, symbols defined in
7436 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
7437 @samp{maint print msymbols} dumps just the minimal symbol information
7438 required for each object file from which @value{GDBN} has read some symbols.
7439 @xref{Files, ,Commands to specify files}, for a discussion of how
7440 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
7444 @chapter Altering Execution
7446 Once you think you have found an error in your program, you might want to
7447 find out for certain whether correcting the apparent error would lead to
7448 correct results in the rest of the run. You can find the answer by
7449 experiment, using the @value{GDBN} features for altering execution of the
7452 For example, you can store new values into variables or memory
7453 locations, give your program a signal, restart it at a different
7454 address, or even return prematurely from a function.
7457 * Assignment:: Assignment to variables
7458 * Jumping:: Continuing at a different address
7459 * Signaling:: Giving your program a signal
7460 * Returning:: Returning from a function
7461 * Calling:: Calling your program's functions
7462 * Patching:: Patching your program
7466 @section Assignment to variables
7469 @cindex setting variables
7470 To alter the value of a variable, evaluate an assignment expression.
7471 @xref{Expressions, ,Expressions}. For example,
7478 stores the value 4 into the variable @code{x}, and then prints the
7479 value of the assignment expression (which is 4).
7480 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
7481 information on operators in supported languages.
7483 @kindex set variable
7484 @cindex variables, setting
7485 If you are not interested in seeing the value of the assignment, use the
7486 @code{set} command instead of the @code{print} command. @code{set} is
7487 really the same as @code{print} except that the expression's value is
7488 not printed and is not put in the value history (@pxref{Value History,
7489 ,Value history}). The expression is evaluated only for its effects.
7491 If the beginning of the argument string of the @code{set} command
7492 appears identical to a @code{set} subcommand, use the @code{set
7493 variable} command instead of just @code{set}. This command is identical
7494 to @code{set} except for its lack of subcommands. For example, if your
7495 program has a variable @code{width}, you get an error if you try to set
7496 a new value with just @samp{set width=13}, because @value{GDBN} has the
7497 command @code{set width}:
7500 (@value{GDBP}) whatis width
7502 (@value{GDBP}) p width
7504 (@value{GDBP}) set width=47
7505 Invalid syntax in expression.
7509 The invalid expression, of course, is @samp{=47}. In
7510 order to actually set the program's variable @code{width}, use
7513 (@value{GDBP}) set var width=47
7516 Because the @code{set} command has many subcommands that can conflict
7517 with the names of program variables, it is a good idea to use the
7518 @code{set variable} command instead of just @code{set}. For example, if
7519 your program has a variable @code{g}, you run into problems if you try
7520 to set a new value with just @samp{set g=4}, because @value{GDBN} has
7521 the command @code{set gnutarget}, abbreviated @code{set g}:
7525 (@value{GDBP}) whatis g
7529 (@value{GDBP}) set g=4
7533 The program being debugged has been started already.
7534 Start it from the beginning? (y or n) y
7535 Starting program: /home/smith/cc_progs/a.out
7536 "/home/smith/cc_progs/a.out": can't open to read symbols: Invalid bfd target.
7537 (@value{GDBP}) show g
7538 The current BFD target is "=4".
7543 The program variable @code{g} did not change, and you silently set the
7544 @code{gnutarget} to an invalid value. In order to set the variable
7548 (@value{GDBP}) set var g=4
7551 @value{GDBN} allows more implicit conversions in assignments than C; you can
7552 freely store an integer value into a pointer variable or vice versa,
7553 and you can convert any structure to any other structure that is the
7554 same length or shorter.
7555 @comment FIXME: how do structs align/pad in these conversions?
7556 @comment /doc@cygnus.com 18dec1990
7558 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
7559 construct to generate a value of specified type at a specified address
7560 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
7561 to memory location @code{0x83040} as an integer (which implies a certain size
7562 and representation in memory), and
7565 set @{int@}0x83040 = 4
7569 stores the value 4 into that memory location.
7572 @section Continuing at a different address
7574 Ordinarily, when you continue your program, you do so at the place where
7575 it stopped, with the @code{continue} command. You can instead continue at
7576 an address of your own choosing, with the following commands:
7580 @item jump @var{linespec}
7581 Resume execution at line @var{linespec}. Execution stops again
7582 immediately if there is a breakpoint there. @xref{List, ,Printing
7583 source lines}, for a description of the different forms of
7584 @var{linespec}. It is common practice to use the @code{tbreak} command
7585 in conjunction with @code{jump}. @xref{Set Breaks, ,Setting
7588 The @code{jump} command does not change the current stack frame, or
7589 the stack pointer, or the contents of any memory location or any
7590 register other than the program counter. If line @var{linespec} is in
7591 a different function from the one currently executing, the results may
7592 be bizarre if the two functions expect different patterns of arguments or
7593 of local variables. For this reason, the @code{jump} command requests
7594 confirmation if the specified line is not in the function currently
7595 executing. However, even bizarre results are predictable if you are
7596 well acquainted with the machine-language code of your program.
7598 @item jump *@var{address}
7599 Resume execution at the instruction at address @var{address}.
7602 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
7603 On many systems, you can get much the same effect as the @code{jump}
7604 command by storing a new value into the register @code{$pc}. The
7605 difference is that this does not start your program running; it only
7606 changes the address of where it @emph{will} run when you continue. For
7614 makes the next @code{continue} command or stepping command execute at
7615 address @code{0x485}, rather than at the address where your program stopped.
7616 @xref{Continuing and Stepping, ,Continuing and stepping}.
7618 The most common occasion to use the @code{jump} command is to back
7619 up---perhaps with more breakpoints set---over a portion of a program
7620 that has already executed, in order to examine its execution in more
7625 @section Giving your program a signal
7629 @item signal @var{signal}
7630 Resume execution where your program stopped, but immediately give it the
7631 signal @var{signal}. @var{signal} can be the name or the number of a
7632 signal. For example, on many systems @code{signal 2} and @code{signal
7633 SIGINT} are both ways of sending an interrupt signal.
7635 Alternatively, if @var{signal} is zero, continue execution without
7636 giving a signal. This is useful when your program stopped on account of
7637 a signal and would ordinary see the signal when resumed with the
7638 @code{continue} command; @samp{signal 0} causes it to resume without a
7641 @code{signal} does not repeat when you press @key{RET} a second time
7642 after executing the command.
7646 Invoking the @code{signal} command is not the same as invoking the
7647 @code{kill} utility from the shell. Sending a signal with @code{kill}
7648 causes @value{GDBN} to decide what to do with the signal depending on
7649 the signal handling tables (@pxref{Signals}). The @code{signal} command
7650 passes the signal directly to your program.
7654 @section Returning from a function
7657 @cindex returning from a function
7660 @itemx return @var{expression}
7661 You can cancel execution of a function call with the @code{return}
7662 command. If you give an
7663 @var{expression} argument, its value is used as the function's return
7667 When you use @code{return}, @value{GDBN} discards the selected stack frame
7668 (and all frames within it). You can think of this as making the
7669 discarded frame return prematurely. If you wish to specify a value to
7670 be returned, give that value as the argument to @code{return}.
7672 This pops the selected stack frame (@pxref{Selection, ,Selecting a
7673 frame}), and any other frames inside of it, leaving its caller as the
7674 innermost remaining frame. That frame becomes selected. The
7675 specified value is stored in the registers used for returning values
7678 The @code{return} command does not resume execution; it leaves the
7679 program stopped in the state that would exist if the function had just
7680 returned. In contrast, the @code{finish} command (@pxref{Continuing
7681 and Stepping, ,Continuing and stepping}) resumes execution until the
7682 selected stack frame returns naturally.
7685 @section Calling program functions
7687 @cindex calling functions
7690 @item call @var{expr}
7691 Evaluate the expression @var{expr} without displaying @code{void}
7695 You can use this variant of the @code{print} command if you want to
7696 execute a function from your program, but without cluttering the output
7697 with @code{void} returned values. If the result is not void, it
7698 is printed and saved in the value history.
7700 For the A29K, a user-controlled variable @code{call_scratch_address},
7701 specifies the location of a scratch area to be used when @value{GDBN}
7702 calls a function in the target. This is necessary because the usual
7703 method of putting the scratch area on the stack does not work in systems
7704 that have separate instruction and data spaces.
7707 @section Patching programs
7709 @cindex patching binaries
7710 @cindex writing into executables
7711 @cindex writing into corefiles
7713 By default, @value{GDBN} opens the file containing your program's
7714 executable code (or the corefile) read-only. This prevents accidental
7715 alterations to machine code; but it also prevents you from intentionally
7716 patching your program's binary.
7718 If you'd like to be able to patch the binary, you can specify that
7719 explicitly with the @code{set write} command. For example, you might
7720 want to turn on internal debugging flags, or even to make emergency
7726 @itemx set write off
7727 If you specify @samp{set write on}, @value{GDBN} opens executable and
7728 core files for both reading and writing; if you specify @samp{set write
7729 off} (the default), @value{GDBN} opens them read-only.
7731 If you have already loaded a file, you must load it again (using the
7732 @code{exec-file} or @code{core-file} command) after changing @code{set
7733 write}, for your new setting to take effect.
7737 Display whether executable files and core files are opened for writing
7742 @chapter @value{GDBN} Files
7744 @value{GDBN} needs to know the file name of the program to be debugged,
7745 both in order to read its symbol table and in order to start your
7746 program. To debug a core dump of a previous run, you must also tell
7747 @value{GDBN} the name of the core dump file.
7750 * Files:: Commands to specify files
7751 * Symbol Errors:: Errors reading symbol files
7755 @section Commands to specify files
7757 @cindex symbol table
7758 @cindex core dump file
7760 You may want to specify executable and core dump file names. The usual
7761 way to do this is at start-up time, using the arguments to
7762 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
7763 Out of @value{GDBN}}).
7765 Occasionally it is necessary to change to a different file during a
7766 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
7767 a file you want to use. In these situations the @value{GDBN} commands
7768 to specify new files are useful.
7771 @cindex executable file
7773 @item file @var{filename}
7774 Use @var{filename} as the program to be debugged. It is read for its
7775 symbols and for the contents of pure memory. It is also the program
7776 executed when you use the @code{run} command. If you do not specify a
7777 directory and the file is not found in the @value{GDBN} working directory,
7778 @value{GDBN} uses the environment variable @code{PATH} as a list of
7779 directories to search, just as the shell does when looking for a program
7780 to run. You can change the value of this variable, for both @value{GDBN}
7781 and your program, using the @code{path} command.
7783 On systems with memory-mapped files, an auxiliary file
7784 @file{@var{filename}.syms} may hold symbol table information for
7785 @var{filename}. If so, @value{GDBN} maps in the symbol table from
7786 @file{@var{filename}.syms}, starting up more quickly. See the
7787 descriptions of the file options @samp{-mapped} and @samp{-readnow}
7788 (available on the command line, and with the commands @code{file},
7789 @code{symbol-file}, or @code{add-symbol-file}, described below),
7790 for more information.
7793 @code{file} with no argument makes @value{GDBN} discard any information it
7794 has on both executable file and the symbol table.
7797 @item exec-file @r{[} @var{filename} @r{]}
7798 Specify that the program to be run (but not the symbol table) is found
7799 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
7800 if necessary to locate your program. Omitting @var{filename} means to
7801 discard information on the executable file.
7804 @item symbol-file @r{[} @var{filename} @r{]}
7805 Read symbol table information from file @var{filename}. @code{PATH} is
7806 searched when necessary. Use the @code{file} command to get both symbol
7807 table and program to run from the same file.
7809 @code{symbol-file} with no argument clears out @value{GDBN} information on your
7810 program's symbol table.
7812 The @code{symbol-file} command causes @value{GDBN} to forget the contents
7813 of its convenience variables, the value history, and all breakpoints and
7814 auto-display expressions. This is because they may contain pointers to
7815 the internal data recording symbols and data types, which are part of
7816 the old symbol table data being discarded inside @value{GDBN}.
7818 @code{symbol-file} does not repeat if you press @key{RET} again after
7821 When @value{GDBN} is configured for a particular environment, it
7822 understands debugging information in whatever format is the standard
7823 generated for that environment; you may use either a @sc{gnu} compiler, or
7824 other compilers that adhere to the local conventions.
7825 Best results are usually obtained from @sc{gnu} compilers; for example,
7826 using @code{@value{GCC}} you can generate debugging information for
7829 For most kinds of object files, with the exception of old SVR3 systems
7830 using COFF, the @code{symbol-file} command does not normally read the
7831 symbol table in full right away. Instead, it scans the symbol table
7832 quickly to find which source files and which symbols are present. The
7833 details are read later, one source file at a time, as they are needed.
7835 The purpose of this two-stage reading strategy is to make @value{GDBN}
7836 start up faster. For the most part, it is invisible except for
7837 occasional pauses while the symbol table details for a particular source
7838 file are being read. (The @code{set verbose} command can turn these
7839 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
7840 warnings and messages}.)
7842 We have not implemented the two-stage strategy for COFF yet. When the
7843 symbol table is stored in COFF format, @code{symbol-file} reads the
7844 symbol table data in full right away. Note that ``stabs-in-COFF''
7845 still does the two-stage strategy, since the debug info is actually
7849 @cindex reading symbols immediately
7850 @cindex symbols, reading immediately
7852 @cindex memory-mapped symbol file
7853 @cindex saving symbol table
7854 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7855 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7856 You can override the @value{GDBN} two-stage strategy for reading symbol
7857 tables by using the @samp{-readnow} option with any of the commands that
7858 load symbol table information, if you want to be sure @value{GDBN} has the
7859 entire symbol table available.
7861 If memory-mapped files are available on your system through the
7862 @code{mmap} system call, you can use another option, @samp{-mapped}, to
7863 cause @value{GDBN} to write the symbols for your program into a reusable
7864 file. Future @value{GDBN} debugging sessions map in symbol information
7865 from this auxiliary symbol file (if the program has not changed), rather
7866 than spending time reading the symbol table from the executable
7867 program. Using the @samp{-mapped} option has the same effect as
7868 starting @value{GDBN} with the @samp{-mapped} command-line option.
7870 You can use both options together, to make sure the auxiliary symbol
7871 file has all the symbol information for your program.
7873 The auxiliary symbol file for a program called @var{myprog} is called
7874 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
7875 than the corresponding executable), @value{GDBN} always attempts to use
7876 it when you debug @var{myprog}; no special options or commands are
7879 The @file{.syms} file is specific to the host machine where you run
7880 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
7881 symbol table. It cannot be shared across multiple host platforms.
7883 @c FIXME: for now no mention of directories, since this seems to be in
7884 @c flux. 13mar1992 status is that in theory GDB would look either in
7885 @c current dir or in same dir as myprog; but issues like competing
7886 @c GDB's, or clutter in system dirs, mean that in practice right now
7887 @c only current dir is used. FFish says maybe a special GDB hierarchy
7888 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
7893 @item core-file @r{[} @var{filename} @r{]}
7894 Specify the whereabouts of a core dump file to be used as the ``contents
7895 of memory''. Traditionally, core files contain only some parts of the
7896 address space of the process that generated them; @value{GDBN} can access the
7897 executable file itself for other parts.
7899 @code{core-file} with no argument specifies that no core file is
7902 Note that the core file is ignored when your program is actually running
7903 under @value{GDBN}. So, if you have been running your program and you
7904 wish to debug a core file instead, you must kill the subprocess in which
7905 the program is running. To do this, use the @code{kill} command
7906 (@pxref{Kill Process, ,Killing the child process}).
7908 @kindex add-symbol-file
7909 @cindex dynamic linking
7910 @item add-symbol-file @var{filename} @var{address}
7911 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7912 @itemx add-symbol-file @var{filename} @var{address} @var{data_address} @var{bss_address}
7913 @itemx add-symbol-file @var{filename} @r{-T}@var{section} @var{address}
7914 The @code{add-symbol-file} command reads additional symbol table
7915 information from the file @var{filename}. You would use this command
7916 when @var{filename} has been dynamically loaded (by some other means)
7917 into the program that is running. @var{address} should be the memory
7918 address at which the file has been loaded; @value{GDBN} cannot figure
7919 this out for itself. You can specify up to three addresses, in which
7920 case they are taken to be the addresses of the text, data, and bss
7921 segments respectively. For complicated cases, you can specify an
7922 arbitrary number of @samp{@r{-T}@var{section} @var{address}} pairs, to
7923 give an explicit section name and base address for that section. You
7924 can specify any @var{address} as an expression.
7926 The symbol table of the file @var{filename} is added to the symbol table
7927 originally read with the @code{symbol-file} command. You can use the
7928 @code{add-symbol-file} command any number of times; the new symbol data
7929 thus read keeps adding to the old. To discard all old symbol data
7930 instead, use the @code{symbol-file} command without any arguments.
7932 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
7934 You can use the @samp{-mapped} and @samp{-readnow} options just as with
7935 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
7936 table information for @var{filename}.
7938 @kindex add-shared-symbol-file
7939 @item add-shared-symbol-file
7940 The @code{add-shared-symbol-file} command can be used only under Harris' CXUX
7941 operating system for the Motorola 88k. @value{GDBN} automatically looks for
7942 shared libraries, however if @value{GDBN} does not find yours, you can run
7943 @code{add-shared-symbol-file}. It takes no arguments.
7947 The @code{section} command changes the base address of section SECTION of
7948 the exec file to ADDR. This can be used if the exec file does not contain
7949 section addresses, (such as in the a.out format), or when the addresses
7950 specified in the file itself are wrong. Each section must be changed
7951 separately. The @code{info files} command, described below, lists all
7952 the sections and their addresses.
7958 @code{info files} and @code{info target} are synonymous; both print the
7959 current target (@pxref{Targets, ,Specifying a Debugging Target}),
7960 including the names of the executable and core dump files currently in
7961 use by @value{GDBN}, and the files from which symbols were loaded. The
7962 command @code{help target} lists all possible targets rather than
7967 All file-specifying commands allow both absolute and relative file names
7968 as arguments. @value{GDBN} always converts the file name to an absolute file
7969 name and remembers it that way.
7971 @cindex shared libraries
7972 @value{GDBN} supports HP-UX, SunOS, SVr4, Irix 5, and IBM RS/6000 shared
7975 @value{GDBN} automatically loads symbol definitions from shared libraries
7976 when you use the @code{run} command, or when you examine a core file.
7977 (Before you issue the @code{run} command, @value{GDBN} does not understand
7978 references to a function in a shared library, however---unless you are
7979 debugging a core file).
7981 On HP-UX, if the program loads a library explicitly, @value{GDBN}
7982 automatically loads the symbols at the time of the @code{shl_load} call.
7984 @c FIXME: some @value{GDBN} release may permit some refs to undef
7985 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
7986 @c FIXME...lib; check this from time to time when updating manual
7989 @kindex info sharedlibrary
7992 @itemx info sharedlibrary
7993 Print the names of the shared libraries which are currently loaded.
7995 @kindex sharedlibrary
7997 @item sharedlibrary @var{regex}
7998 @itemx share @var{regex}
7999 Load shared object library symbols for files matching a
8000 Unix regular expression.
8001 As with files loaded automatically, it only loads shared libraries
8002 required by your program for a core file or after typing @code{run}. If
8003 @var{regex} is omitted all shared libraries required by your program are
8007 On HP-UX systems, @value{GDBN} detects the loading of a shared library
8008 and automatically reads in symbols from the newly loaded library, up to
8009 a threshold that is initially set but that you can modify if you wish.
8011 Beyond that threshold, symbols from shared libraries must be explicitly
8012 loaded. To load these symbols, use the command @code{sharedlibrary
8013 @var{filename}}. The base address of the shared library is determined
8014 automatically by @value{GDBN} and need not be specified.
8016 To display or set the threshold, use the commands:
8019 @kindex set auto-solib-add
8020 @item set auto-solib-add @var{threshold}
8021 Set the autoloading size threshold, in megabytes. If @var{threshold} is
8022 nonzero, symbols from all shared object libraries will be loaded
8023 automatically when the inferior begins execution or when the dynamic
8024 linker informs @value{GDBN} that a new library has been loaded, until
8025 the symbol table of the program and libraries exceeds this threshold.
8026 Otherwise, symbols must be loaded manually, using the
8027 @code{sharedlibrary} command. The default threshold is 100 megabytes.
8029 @kindex show auto-solib-add
8030 @item show auto-solib-add
8031 Display the current autoloading size threshold, in megabytes.
8035 @section Errors reading symbol files
8037 While reading a symbol file, @value{GDBN} occasionally encounters problems,
8038 such as symbol types it does not recognize, or known bugs in compiler
8039 output. By default, @value{GDBN} does not notify you of such problems, since
8040 they are relatively common and primarily of interest to people
8041 debugging compilers. If you are interested in seeing information
8042 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
8043 only one message about each such type of problem, no matter how many
8044 times the problem occurs; or you can ask @value{GDBN} to print more messages,
8045 to see how many times the problems occur, with the @code{set
8046 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
8049 The messages currently printed, and their meanings, include:
8052 @item inner block not inside outer block in @var{symbol}
8054 The symbol information shows where symbol scopes begin and end
8055 (such as at the start of a function or a block of statements). This
8056 error indicates that an inner scope block is not fully contained
8057 in its outer scope blocks.
8059 @value{GDBN} circumvents the problem by treating the inner block as if it had
8060 the same scope as the outer block. In the error message, @var{symbol}
8061 may be shown as ``@code{(don't know)}'' if the outer block is not a
8064 @item block at @var{address} out of order
8066 The symbol information for symbol scope blocks should occur in
8067 order of increasing addresses. This error indicates that it does not
8070 @value{GDBN} does not circumvent this problem, and has trouble
8071 locating symbols in the source file whose symbols it is reading. (You
8072 can often determine what source file is affected by specifying
8073 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
8076 @item bad block start address patched
8078 The symbol information for a symbol scope block has a start address
8079 smaller than the address of the preceding source line. This is known
8080 to occur in the SunOS 4.1.1 (and earlier) C compiler.
8082 @value{GDBN} circumvents the problem by treating the symbol scope block as
8083 starting on the previous source line.
8085 @item bad string table offset in symbol @var{n}
8088 Symbol number @var{n} contains a pointer into the string table which is
8089 larger than the size of the string table.
8091 @value{GDBN} circumvents the problem by considering the symbol to have the
8092 name @code{foo}, which may cause other problems if many symbols end up
8095 @item unknown symbol type @code{0x@var{nn}}
8097 The symbol information contains new data types that @value{GDBN} does
8098 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
8099 uncomprehended information, in hexadecimal.
8101 @value{GDBN} circumvents the error by ignoring this symbol information.
8102 This usually allows you to debug your program, though certain symbols
8103 are not accessible. If you encounter such a problem and feel like
8104 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
8105 on @code{complain}, then go up to the function @code{read_dbx_symtab}
8106 and examine @code{*bufp} to see the symbol.
8108 @item stub type has NULL name
8110 @value{GDBN} could not find the full definition for a struct or class.
8112 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
8113 The symbol information for a C++ member function is missing some
8114 information that recent versions of the compiler should have output for
8117 @item info mismatch between compiler and debugger
8119 @value{GDBN} could not parse a type specification output by the compiler.
8124 @chapter Specifying a Debugging Target
8126 @cindex debugging target
8129 A @dfn{target} is the execution environment occupied by your program.
8131 Often, @value{GDBN} runs in the same host environment as your program;
8132 in that case, the debugging target is specified as a side effect when
8133 you use the @code{file} or @code{core} commands. When you need more
8134 flexibility---for example, running @value{GDBN} on a physically separate
8135 host, or controlling a standalone system over a serial port or a
8136 realtime system over a TCP/IP connection---you can use the @code{target}
8137 command to specify one of the target types configured for @value{GDBN}
8138 (@pxref{Target Commands, ,Commands for managing targets}).
8141 * Active Targets:: Active targets
8142 * Target Commands:: Commands for managing targets
8143 * Byte Order:: Choosing target byte order
8144 * Remote:: Remote debugging
8145 * KOD:: Kernel Object Display
8149 @node Active Targets
8150 @section Active targets
8152 @cindex stacking targets
8153 @cindex active targets
8154 @cindex multiple targets
8156 There are three classes of targets: processes, core files, and
8157 executable files. @value{GDBN} can work concurrently on up to three
8158 active targets, one in each class. This allows you to (for example)
8159 start a process and inspect its activity without abandoning your work on
8162 For example, if you execute @samp{gdb a.out}, then the executable file
8163 @code{a.out} is the only active target. If you designate a core file as
8164 well---presumably from a prior run that crashed and coredumped---then
8165 @value{GDBN} has two active targets and uses them in tandem, looking
8166 first in the corefile target, then in the executable file, to satisfy
8167 requests for memory addresses. (Typically, these two classes of target
8168 are complementary, since core files contain only a program's
8169 read-write memory---variables and so on---plus machine status, while
8170 executable files contain only the program text and initialized data.)
8172 When you type @code{run}, your executable file becomes an active process
8173 target as well. When a process target is active, all @value{GDBN}
8174 commands requesting memory addresses refer to that target; addresses in
8175 an active core file or executable file target are obscured while the
8176 process target is active.
8178 Use the @code{core-file} and @code{exec-file} commands to select a new
8179 core file or executable target (@pxref{Files, ,Commands to specify
8180 files}). To specify as a target a process that is already running, use
8181 the @code{attach} command (@pxref{Attach, ,Debugging an already-running
8184 @node Target Commands
8185 @section Commands for managing targets
8188 @item target @var{type} @var{parameters}
8189 Connects the @value{GDBN} host environment to a target machine or
8190 process. A target is typically a protocol for talking to debugging
8191 facilities. You use the argument @var{type} to specify the type or
8192 protocol of the target machine.
8194 Further @var{parameters} are interpreted by the target protocol, but
8195 typically include things like device names or host names to connect
8196 with, process numbers, and baud rates.
8198 The @code{target} command does not repeat if you press @key{RET} again
8199 after executing the command.
8203 Displays the names of all targets available. To display targets
8204 currently selected, use either @code{info target} or @code{info files}
8205 (@pxref{Files, ,Commands to specify files}).
8207 @item help target @var{name}
8208 Describe a particular target, including any parameters necessary to
8211 @kindex set gnutarget
8212 @item set gnutarget @var{args}
8213 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
8214 knows whether it is reading an @dfn{executable},
8215 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
8216 with the @code{set gnutarget} command. Unlike most @code{target} commands,
8217 with @code{gnutarget} the @code{target} refers to a program, not a machine.
8220 @emph{Warning:} To specify a file format with @code{set gnutarget},
8221 you must know the actual BFD name.
8225 @xref{Files, , Commands to specify files}.
8227 @kindex show gnutarget
8228 @item show gnutarget
8229 Use the @code{show gnutarget} command to display what file format
8230 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
8231 @value{GDBN} will determine the file format for each file automatically,
8232 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
8235 Here are some common targets (available, or not, depending on the GDB
8240 @item target exec @var{program}
8241 An executable file. @samp{target exec @var{program}} is the same as
8242 @samp{exec-file @var{program}}.
8245 @item target core @var{filename}
8246 A core dump file. @samp{target core @var{filename}} is the same as
8247 @samp{core-file @var{filename}}.
8249 @kindex target remote
8250 @item target remote @var{dev}
8251 Remote serial target in GDB-specific protocol. The argument @var{dev}
8252 specifies what serial device to use for the connection (e.g.
8253 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}. @code{target remote}
8254 supports the @code{load} command. This is only useful if you have
8255 some other way of getting the stub to the target system, and you can put
8256 it somewhere in memory where it won't get clobbered by the download.
8260 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
8268 works; however, you cannot assume that a specific memory map, device
8269 drivers, or even basic I/O is available, although some simulators do
8270 provide these. For info about any processor-specific simulator details,
8271 see the appropriate section in @ref{Embedded Processors, ,Embedded
8276 Some configurations may include these targets as well:
8281 @item target nrom @var{dev}
8282 NetROM ROM emulator. This target only supports downloading.
8286 Different targets are available on different configurations of @value{GDBN};
8287 your configuration may have more or fewer targets.
8289 Many remote targets require you to download the executable's code
8290 once you've successfully established a connection.
8294 @kindex load @var{filename}
8295 @item load @var{filename}
8296 Depending on what remote debugging facilities are configured into
8297 @value{GDBN}, the @code{load} command may be available. Where it exists, it
8298 is meant to make @var{filename} (an executable) available for debugging
8299 on the remote system---by downloading, or dynamic linking, for example.
8300 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
8301 the @code{add-symbol-file} command.
8303 If your @value{GDBN} does not have a @code{load} command, attempting to
8304 execute it gets the error message ``@code{You can't do that when your
8305 target is @dots{}}''
8307 The file is loaded at whatever address is specified in the executable.
8308 For some object file formats, you can specify the load address when you
8309 link the program; for other formats, like a.out, the object file format
8310 specifies a fixed address.
8311 @c FIXME! This would be a good place for an xref to the GNU linker doc.
8313 @code{load} does not repeat if you press @key{RET} again after using it.
8317 @section Choosing target byte order
8319 @cindex choosing target byte order
8320 @cindex target byte order
8321 @kindex set endian big
8322 @kindex set endian little
8323 @kindex set endian auto
8326 Some types of processors, such as the MIPS, PowerPC, and Hitachi SH,
8327 offer the ability to run either big-endian or little-endian byte
8328 orders. Usually the executable or symbol will include a bit to
8329 designate the endian-ness, and you will not need to worry about
8330 which to use. However, you may still find it useful to adjust
8331 @value{GDBN}'s idea of processor endian-ness manually.
8334 @kindex set endian big
8335 @item set endian big
8336 Instruct @value{GDBN} to assume the target is big-endian.
8338 @kindex set endian little
8339 @item set endian little
8340 Instruct @value{GDBN} to assume the target is little-endian.
8342 @kindex set endian auto
8343 @item set endian auto
8344 Instruct @value{GDBN} to use the byte order associated with the
8348 Display @value{GDBN}'s current idea of the target byte order.
8352 Note that these commands merely adjust interpretation of symbolic
8353 data on the host, and that they have absolutely no effect on the
8357 @section Remote debugging
8358 @cindex remote debugging
8360 If you are trying to debug a program running on a machine that cannot run
8361 @value{GDBN} in the usual way, it is often useful to use remote debugging.
8362 For example, you might use remote debugging on an operating system kernel,
8363 or on a small system which does not have a general purpose operating system
8364 powerful enough to run a full-featured debugger.
8366 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
8367 to make this work with particular debugging targets. In addition,
8368 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
8369 but not specific to any particular target system) which you can use if you
8370 write the remote stubs---the code that runs on the remote system to
8371 communicate with @value{GDBN}.
8373 Other remote targets may be available in your
8374 configuration of @value{GDBN}; use @code{help target} to list them.
8377 * Remote Serial:: @value{GDBN} remote serial protocol
8381 @subsection The @value{GDBN} remote serial protocol
8383 @cindex remote serial debugging, overview
8384 To debug a program running on another machine (the debugging
8385 @dfn{target} machine), you must first arrange for all the usual
8386 prerequisites for the program to run by itself. For example, for a C
8391 A startup routine to set up the C runtime environment; these usually
8392 have a name like @file{crt0}. The startup routine may be supplied by
8393 your hardware supplier, or you may have to write your own.
8396 A C subroutine library to support your program's
8397 subroutine calls, notably managing input and output.
8400 A way of getting your program to the other machine---for example, a
8401 download program. These are often supplied by the hardware
8402 manufacturer, but you may have to write your own from hardware
8406 The next step is to arrange for your program to use a serial port to
8407 communicate with the machine where @value{GDBN} is running (the @dfn{host}
8408 machine). In general terms, the scheme looks like this:
8412 @value{GDBN} already understands how to use this protocol; when everything
8413 else is set up, you can simply use the @samp{target remote} command
8414 (@pxref{Targets,,Specifying a Debugging Target}).
8416 @item On the target,
8417 you must link with your program a few special-purpose subroutines that
8418 implement the @value{GDBN} remote serial protocol. The file containing these
8419 subroutines is called a @dfn{debugging stub}.
8421 On certain remote targets, you can use an auxiliary program
8422 @code{gdbserver} instead of linking a stub into your program.
8423 @xref{Server,,Using the @code{gdbserver} program}, for details.
8426 The debugging stub is specific to the architecture of the remote
8427 machine; for example, use @file{sparc-stub.c} to debug programs on
8430 @cindex remote serial stub list
8431 These working remote stubs are distributed with @value{GDBN}:
8439 For Intel 386 and compatible architectures.
8443 @cindex Motorola 680x0
8445 For Motorola 680x0 architectures.
8451 For Hitachi SH architectures.
8454 @kindex sparc-stub.c
8456 For @sc{sparc} architectures.
8459 @kindex sparcl-stub.c
8462 For Fujitsu @sc{sparclite} architectures.
8466 The @file{README} file in the @value{GDBN} distribution may list other
8467 recently added stubs.
8470 * Stub Contents:: What the stub can do for you
8471 * Bootstrapping:: What you must do for the stub
8472 * Debug Session:: Putting it all together
8473 * Protocol:: Definition of the communication protocol
8474 * Server:: Using the `gdbserver' program
8475 * NetWare:: Using the `gdbserve.nlm' program
8479 @subsubsection What the stub can do for you
8481 @cindex remote serial stub
8482 The debugging stub for your architecture supplies these three
8486 @item set_debug_traps
8487 @kindex set_debug_traps
8488 @cindex remote serial stub, initialization
8489 This routine arranges for @code{handle_exception} to run when your
8490 program stops. You must call this subroutine explicitly near the
8491 beginning of your program.
8493 @item handle_exception
8494 @kindex handle_exception
8495 @cindex remote serial stub, main routine
8496 This is the central workhorse, but your program never calls it
8497 explicitly---the setup code arranges for @code{handle_exception} to
8498 run when a trap is triggered.
8500 @code{handle_exception} takes control when your program stops during
8501 execution (for example, on a breakpoint), and mediates communications
8502 with @value{GDBN} on the host machine. This is where the communications
8503 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
8504 representative on the target machine. It begins by sending summary
8505 information on the state of your program, then continues to execute,
8506 retrieving and transmitting any information @value{GDBN} needs, until you
8507 execute a @value{GDBN} command that makes your program resume; at that point,
8508 @code{handle_exception} returns control to your own code on the target
8512 @cindex @code{breakpoint} subroutine, remote
8513 Use this auxiliary subroutine to make your program contain a
8514 breakpoint. Depending on the particular situation, this may be the only
8515 way for @value{GDBN} to get control. For instance, if your target
8516 machine has some sort of interrupt button, you won't need to call this;
8517 pressing the interrupt button transfers control to
8518 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
8519 simply receiving characters on the serial port may also trigger a trap;
8520 again, in that situation, you don't need to call @code{breakpoint} from
8521 your own program---simply running @samp{target remote} from the host
8522 @value{GDBN} session gets control.
8524 Call @code{breakpoint} if none of these is true, or if you simply want
8525 to make certain your program stops at a predetermined point for the
8526 start of your debugging session.
8530 @subsubsection What you must do for the stub
8532 @cindex remote stub, support routines
8533 The debugging stubs that come with @value{GDBN} are set up for a particular
8534 chip architecture, but they have no information about the rest of your
8535 debugging target machine.
8537 First of all you need to tell the stub how to communicate with the
8541 @item int getDebugChar()
8542 @kindex getDebugChar
8543 Write this subroutine to read a single character from the serial port.
8544 It may be identical to @code{getchar} for your target system; a
8545 different name is used to allow you to distinguish the two if you wish.
8547 @item void putDebugChar(int)
8548 @kindex putDebugChar
8549 Write this subroutine to write a single character to the serial port.
8550 It may be identical to @code{putchar} for your target system; a
8551 different name is used to allow you to distinguish the two if you wish.
8554 @cindex control C, and remote debugging
8555 @cindex interrupting remote targets
8556 If you want @value{GDBN} to be able to stop your program while it is
8557 running, you need to use an interrupt-driven serial driver, and arrange
8558 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
8559 character). That is the character which @value{GDBN} uses to tell the
8560 remote system to stop.
8562 Getting the debugging target to return the proper status to @value{GDBN}
8563 probably requires changes to the standard stub; one quick and dirty way
8564 is to just execute a breakpoint instruction (the ``dirty'' part is that
8565 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
8567 Other routines you need to supply are:
8570 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
8571 @kindex exceptionHandler
8572 Write this function to install @var{exception_address} in the exception
8573 handling tables. You need to do this because the stub does not have any
8574 way of knowing what the exception handling tables on your target system
8575 are like (for example, the processor's table might be in @sc{rom},
8576 containing entries which point to a table in @sc{ram}).
8577 @var{exception_number} is the exception number which should be changed;
8578 its meaning is architecture-dependent (for example, different numbers
8579 might represent divide by zero, misaligned access, etc). When this
8580 exception occurs, control should be transferred directly to
8581 @var{exception_address}, and the processor state (stack, registers,
8582 and so on) should be just as it is when a processor exception occurs. So if
8583 you want to use a jump instruction to reach @var{exception_address}, it
8584 should be a simple jump, not a jump to subroutine.
8586 For the 386, @var{exception_address} should be installed as an interrupt
8587 gate so that interrupts are masked while the handler runs. The gate
8588 should be at privilege level 0 (the most privileged level). The
8589 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
8590 help from @code{exceptionHandler}.
8592 @item void flush_i_cache()
8593 @kindex flush_i_cache
8594 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
8595 instruction cache, if any, on your target machine. If there is no
8596 instruction cache, this subroutine may be a no-op.
8598 On target machines that have instruction caches, @value{GDBN} requires this
8599 function to make certain that the state of your program is stable.
8603 You must also make sure this library routine is available:
8606 @item void *memset(void *, int, int)
8608 This is the standard library function @code{memset} that sets an area of
8609 memory to a known value. If you have one of the free versions of
8610 @code{libc.a}, @code{memset} can be found there; otherwise, you must
8611 either obtain it from your hardware manufacturer, or write your own.
8614 If you do not use the GNU C compiler, you may need other standard
8615 library subroutines as well; this varies from one stub to another,
8616 but in general the stubs are likely to use any of the common library
8617 subroutines which @code{@value{GCC}} generates as inline code.
8621 @subsubsection Putting it all together
8623 @cindex remote serial debugging summary
8624 In summary, when your program is ready to debug, you must follow these
8629 Make sure you have the supporting low-level routines
8630 (@pxref{Bootstrapping,,What you must do for the stub}):
8632 @code{getDebugChar}, @code{putDebugChar},
8633 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
8637 Insert these lines near the top of your program:
8645 For the 680x0 stub only, you need to provide a variable called
8646 @code{exceptionHook}. Normally you just use:
8649 void (*exceptionHook)() = 0;
8653 but if before calling @code{set_debug_traps}, you set it to point to a
8654 function in your program; that function is called when
8655 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
8656 error). The function indicated by @code{exceptionHook} is called with
8657 one parameter: an @code{int} which is the exception number.
8660 Compile and link together: your program, the @value{GDBN} debugging stub for
8661 your target architecture, and the supporting subroutines.
8664 Make sure you have a serial connection between your target machine and
8665 the @value{GDBN} host, and identify the serial port on the host.
8668 @c The "remote" target now provides a `load' command, so we should
8669 @c document that. FIXME.
8670 Download your program to your target machine (or get it there by
8671 whatever means the manufacturer provides), and start it.
8674 To start remote debugging, run @value{GDBN} on the host machine, and specify
8675 as an executable file the program that is running in the remote machine.
8676 This tells @value{GDBN} how to find your program's symbols and the contents
8680 @cindex serial line, @code{target remote}
8681 Establish communication using the @code{target remote} command.
8682 Its argument specifies how to communicate with the target
8683 machine---either via a devicename attached to a direct serial line, or a
8684 TCP port (usually to a terminal server which in turn has a serial line
8685 to the target). For example, to use a serial line connected to the
8686 device named @file{/dev/ttyb}:
8689 target remote /dev/ttyb
8692 @cindex TCP port, @code{target remote}
8693 To use a TCP connection, use an argument of the form
8694 @code{@var{host}:port}. For example, to connect to port 2828 on a
8695 terminal server named @code{manyfarms}:
8698 target remote manyfarms:2828
8702 Now you can use all the usual commands to examine and change data and to
8703 step and continue the remote program.
8705 To resume the remote program and stop debugging it, use the @code{detach}
8708 @cindex interrupting remote programs
8709 @cindex remote programs, interrupting
8710 Whenever @value{GDBN} is waiting for the remote program, if you type the
8711 interrupt character (often @key{C-C}), @value{GDBN} attempts to stop the
8712 program. This may or may not succeed, depending in part on the hardware
8713 and the serial drivers the remote system uses. If you type the
8714 interrupt character once again, @value{GDBN} displays this prompt:
8717 Interrupted while waiting for the program.
8718 Give up (and stop debugging it)? (y or n)
8721 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
8722 (If you decide you want to try again later, you can use @samp{target
8723 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
8724 goes back to waiting.
8727 @subsubsection Communication protocol
8729 @cindex debugging stub, example
8730 @cindex remote stub, example
8731 @cindex stub example, remote debugging
8732 The stub files provided with @value{GDBN} implement the target side of the
8733 communication protocol, and the @value{GDBN} side is implemented in the
8734 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
8735 these subroutines to communicate, and ignore the details. (If you're
8736 implementing your own stub file, you can still ignore the details: start
8737 with one of the existing stub files. @file{sparc-stub.c} is the best
8738 organized, and therefore the easiest to read.)
8740 However, there may be occasions when you need to know something about
8741 the protocol---for example, if there is only one serial port to your
8742 target machine, you might want your program to do something special if
8743 it recognizes a packet meant for @value{GDBN}.
8745 In the examples below, @samp{<-} and @samp{->} are used to indicate
8746 transmitted and received data respectfully.
8748 @cindex protocol, @value{GDBN} remote serial
8749 @cindex serial protocol, @value{GDBN} remote
8750 @cindex remote serial protocol
8751 All @value{GDBN} commands and responses (other than acknowledgments)
8752 are sent as a @var{packet}. A @var{packet} is introduced with the
8753 character @samp{$}, this is followed by an optional two-digit
8754 @var{sequence-id} and the character @samp{:}, the actual
8755 @var{packet-data}, and the terminating character @samp{#} followed by a
8756 two-digit @var{checksum}:
8759 @code{$}@var{packet-data}@code{#}@var{checksum}
8762 or, with the optional @var{sequence-id}:
8764 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
8767 @cindex checksum, for @value{GDBN} remote
8769 The two-digit @var{checksum} is computed as the modulo 256 sum of all
8770 characters between the leading @samp{$} and the trailing @samp{#} (that
8771 consisting of both the optional @var{sequence-id}@code{:} and the actual
8772 @var{packet-data}) (an eight bit unsigned checksum).
8774 @cindex sequence-id, for @value{GDBN} remote
8776 The two-digit @var{sequence-id}, when present, is returned with the
8777 acknowledgment. Beyond that its meaning is poorly defined.
8778 @value{GDBN} is not known to output @var{sequence-id}s.
8780 When either the host or the target machine receives a packet, the first
8781 response expected is an acknowledgment: either @samp{+} (to indicate
8782 the package was received correctly) or @samp{-} (to request
8786 <- @code{$}@var{packet-data}@code{#}@var{checksum}
8790 If the received packet included a @var{sequence-id} than that is
8791 appended to a positive acknowledgment:
8794 <- @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
8795 -> @code{+}@var{sequence-id}
8798 The host (@value{GDBN}) sends @var{command}s, and the target (the
8799 debugging stub incorporated in your program) sends a @var{response}. In
8800 the case of step and continue @var{command}s, the response is only sent
8801 when the operation has completed (the target has again stopped).
8803 @var{packet-data} consists of a sequence of characters with the
8804 exception of @samp{#} and @samp{$} (see @samp{X} packet for an
8805 exception). @samp{:} can not appear as the third character in a packet.
8806 Fields within the packet should be separated using @samp{,} and @samp{;}
8807 (unfortunately some packets chose to use @samp{:}). Except where
8808 otherwise noted all numbers are represented in HEX with leading zeros
8811 Response @var{data} can be run-length encoded to save space. A @samp{*}
8812 means that the next character is an @sc{ascii} encoding giving a repeat count
8813 which stands for that many repetitions of the character preceding the
8814 @samp{*}. The encoding is @code{n+29}, yielding a printable character
8815 where @code{n >=3} (which is where rle starts to win). The printable
8816 characters @samp{$}, @samp{#}, @samp{+} and @samp{-} or with a numeric
8817 value greater than 126 should not be used.
8819 Some remote systems have used a different run-length encoding mechanism
8820 loosely refered to as the cisco encoding. Following the @samp{*}
8821 character are two hex digits that indicate the size of the packet.
8828 means the same as "0000".
8830 The error response, returned for some packets includes a two character
8831 error number. That number is not well defined.
8833 For any @var{command} not supported by the stub, an empty response
8834 (@samp{$#00}) should be returned. That way it is possible to extend the
8835 protocol. A newer @value{GDBN} can tell if a packet is supported based
8838 Below is a complete list of all currently defined @var{command}s and
8839 their corresponding response @var{data}:
8841 @multitable @columnfractions .30 .30 .40
8846 @item extended ops @emph{(optional)}
8849 Use the extended remote protocol. Sticky---only needs to be set once.
8850 The extended remote protocol support the @samp{R} packet.
8854 Stubs that support the extended remote protocol return @samp{} which,
8855 unfortunately, is identical to the response returned by stubs that do not
8856 support protocol extensions.
8861 Indicate the reason the target halted. The reply is the same as for step
8870 @tab Reserved for future use
8872 @item set program arguments @strong{(reserved)} @emph{(optional)}
8873 @tab @code{A}@var{arglen}@code{,}@var{argnum}@code{,}@var{arg}@code{,...}
8875 Initialized @samp{argv[]} array passed into program. @var{arglen}
8876 specifies the number of bytes in the hex encoded byte stream @var{arg}.
8877 See @file{gdbserver} for more details.
8879 @tab reply @code{OK}
8881 @tab reply @code{E}@var{NN}
8883 @item set baud @strong{(deprecated)}
8884 @tab @code{b}@var{baud}
8886 Change the serial line speed to @var{baud}. JTC: @emph{When does the
8887 transport layer state change? When it's received, or after the ACK is
8888 transmitted. In either case, there are problems if the command or the
8889 acknowledgment packet is dropped.} Stan: @emph{If people really wanted
8890 to add something like this, and get it working for the first time, they
8891 ought to modify ser-unix.c to send some kind of out-of-band message to a
8892 specially-setup stub and have the switch happen "in between" packets, so
8893 that from remote protocol's point of view, nothing actually
8896 @item set breakpoint @strong{(deprecated)}
8897 @tab @code{B}@var{addr},@var{mode}
8899 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
8900 breakpoint at @var{addr}. @emph{This has been replaced by the @samp{Z} and
8904 @tab @code{c}@var{addr}
8906 @var{addr} is address to resume. If @var{addr} is omitted, resume at
8912 @item continue with signal @emph{(optional)}
8913 @tab @code{C}@var{sig}@code{;}@var{addr}
8915 Continue with signal @var{sig} (hex signal number). If
8916 @code{;}@var{addr} is omitted, resume at same address.
8921 @item toggle debug @emph{(deprecated)}
8926 @item detach @emph{(optional)}
8929 Detach @value{GDBN} from the remote system. Sent to the remote target before
8930 @value{GDBN} disconnects.
8932 @tab reply @emph{no response}
8934 @value{GDBN} does not check for any response after sending this packet
8938 @tab Reserved for future use
8942 @tab Reserved for future use
8946 @tab Reserved for future use
8950 @tab Reserved for future use
8952 @item read registers
8954 @tab Read general registers.
8956 @tab reply @var{XX...}
8958 Each byte of register data is described by two hex digits. The bytes
8959 with the register are transmitted in target byte order. The size of
8960 each register and their position within the @samp{g} @var{packet} are
8961 determined by the @value{GDBN} internal macros @var{REGISTER_RAW_SIZE} and
8962 @var{REGISTER_NAME} macros. The specification of several standard
8963 @code{g} packets is specified below.
8965 @tab @code{E}@var{NN}
8969 @tab @code{G}@var{XX...}
8971 See @samp{g} for a description of the @var{XX...} data.
8973 @tab reply @code{OK}
8976 @tab reply @code{E}@var{NN}
8981 @tab Reserved for future use
8983 @item set thread @emph{(optional)}
8984 @tab @code{H}@var{c}@var{t...}
8986 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
8987 @samp{G}, et.al.). @var{c} = @samp{c} for thread used in step and
8988 continue; @var{t...} can be -1 for all threads. @var{c} = @samp{g} for
8989 thread used in other operations. If zero, pick a thread, any thread.
8991 @tab reply @code{OK}
8994 @tab reply @code{E}@var{NN}
8998 @c 'H': How restrictive (or permissive) is the thread model. If a
8999 @c thread is selected and stopped, are other threads allowed
9000 @c to continue to execute? As I mentioned above, I think the
9001 @c semantics of each command when a thread is selected must be
9002 @c described. For example:
9004 @c 'g': If the stub supports threads and a specific thread is
9005 @c selected, returns the register block from that thread;
9006 @c otherwise returns current registers.
9008 @c 'G' If the stub supports threads and a specific thread is
9009 @c selected, sets the registers of the register block of
9010 @c that thread; otherwise sets current registers.
9012 @item cycle step @strong{(draft)} @emph{(optional)}
9013 @tab @code{i}@var{addr}@code{,}@var{nnn}
9015 Step the remote target by a single clock cycle. If @code{,}@var{nnn} is
9016 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
9017 step starting at that address.
9019 @item signal then cycle step @strong{(reserved)} @emph{(optional)}
9022 See @samp{i} and @samp{S} for likely syntax and semantics.
9026 @tab Reserved for future use
9030 @tab Reserved for future use
9032 @item kill request @emph{(optional)}
9035 FIXME: @emph{There is no description of how operate when a specific
9036 thread context has been selected (ie. does 'k' kill only that thread?)}.
9040 @tab Reserved for future use
9044 @tab Reserved for future use
9047 @tab @code{m}@var{addr}@code{,}@var{length}
9049 Read @var{length} bytes of memory starting at address @var{addr}.
9050 Neither @value{GDBN} nor the stub assume that sized memory transfers are assumed
9051 using word alligned accesses. FIXME: @emph{A word aligned memory
9052 transfer mechanism is needed.}
9054 @tab reply @var{XX...}
9056 @var{XX...} is mem contents. Can be fewer bytes than requested if able
9057 to read only part of the data. Neither @value{GDBN} nor the stub assume that
9058 sized memory transfers are assumed using word alligned accesses. FIXME:
9059 @emph{A word aligned memory transfer mechanism is needed.}
9061 @tab reply @code{E}@var{NN}
9062 @tab @var{NN} is errno
9065 @tab @code{M}@var{addr},@var{length}@code{:}@var{XX...}
9067 Write @var{length} bytes of memory starting at address @var{addr}.
9068 @var{XX...} is the data.
9070 @tab reply @code{OK}
9073 @tab reply @code{E}@var{NN}
9075 for an error (this includes the case where only part of the data was
9080 @tab Reserved for future use
9084 @tab Reserved for future use
9088 @tab Reserved for future use
9092 @tab Reserved for future use
9094 @item read reg @strong{(reserved)}
9095 @tab @code{p}@var{n...}
9099 @tab return @var{r....}
9100 @tab The hex encoded value of the register in target byte order.
9102 @item write reg @emph{(optional)}
9103 @tab @code{P}@var{n...}@code{=}@var{r...}
9105 Write register @var{n...} with value @var{r...}, which contains two hex
9106 digits for each byte in the register (target byte order).
9108 @tab reply @code{OK}
9111 @tab reply @code{E}@var{NN}
9114 @item general query @emph{(optional)}
9115 @tab @code{q}@var{query}
9117 Request info about @var{query}. In general @value{GDBN} @var{query}'s
9118 have a leading upper case letter. Custom vendor queries should use a
9119 company prefix (in lower case) ex: @samp{qfsf.var}. @var{query} may
9120 optionally be followed by a @samp{,} or @samp{;} separated list. Stubs
9121 must ensure that they match the full @var{query} name.
9123 @tab reply @code{XX...}
9124 @tab Hex encoded data from query. The reply can not be empty.
9126 @tab reply @code{E}@var{NN}
9130 @tab Indicating an unrecognized @var{query}.
9132 @item general set @emph{(optional)}
9133 @tab @code{Q}@var{var}@code{=}@var{val}
9135 Set value of @var{var} to @var{val}. See @samp{q} for a discussing of
9138 @item reset @emph{(deprecated)}
9141 Reset the entire system.
9143 @item remote restart @emph{(optional)}
9144 @tab @code{R}@var{XX}
9146 Restart the remote server. @var{XX} while needed has no clear
9147 definition. FIXME: @emph{An example interaction explaining how this
9148 packet is used in extended-remote mode is needed}.
9150 @item step @emph{(optional)}
9151 @tab @code{s}@var{addr}
9153 @var{addr} is address to resume. If @var{addr} is omitted, resume at
9159 @item step with signal @emph{(optional)}
9160 @tab @code{S}@var{sig}@code{;}@var{addr}
9162 Like @samp{C} but step not continue.
9167 @item search @emph{(optional)}
9168 @tab @code{t}@var{addr}@code{:}@var{PP}@code{,}@var{MM}
9170 Search backwards starting at address @var{addr} for a match with pattern
9171 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4
9172 bytes. @var{addr} must be at least 3 digits.
9174 @item thread alive @emph{(optional)}
9175 @tab @code{T}@var{XX}
9176 @tab Find out if the thread XX is alive.
9178 @tab reply @code{OK}
9179 @tab thread is still alive
9181 @tab reply @code{E}@var{NN}
9186 @tab Reserved for future use
9190 @tab Reserved for future use
9194 @tab Reserved for future use
9198 @tab Reserved for future use
9202 @tab Reserved for future use
9206 @tab Reserved for future use
9210 @tab Reserved for future use
9212 @item write mem (binary) @emph{(optional)}
9213 @tab @code{X}@var{addr}@code{,}@var{length}@var{:}@var{XX...}
9215 @var{addr} is address, @var{length} is number of bytes, @var{XX...} is
9216 binary data. The characters @code{$}, @code{#}, and @code{0x7d} are
9217 escaped using @code{0x7d}.
9219 @tab reply @code{OK}
9222 @tab reply @code{E}@var{NN}
9227 @tab Reserved for future use
9231 @tab Reserved for future use
9233 @item remove break or watchpoint @strong{(draft)} @emph{(optional)}
9234 @tab @code{z}@var{t}@code{,}@var{addr}@code{,}@var{length}
9238 @item insert break or watchpoint @strong{(draft)} @emph{(optional)}
9239 @tab @code{Z}@var{t}@code{,}@var{addr}@code{,}@var{length}
9241 @var{t} is type: @samp{0} - software breakpoint, @samp{1} - hardware
9242 breakpoint, @samp{2} - write watchpoint, @samp{3} - read watchpoint,
9243 @samp{4} - access watchpoint; @var{addr} is address; @var{length} is in
9244 bytes. For a software breakpoint, @var{length} specifies the size of
9245 the instruction to be patched. For hardware breakpoints and watchpoints
9246 @var{length} specifies the memory region to be monitored. To avoid
9247 potential problems with duplicate packets, the operations should be
9248 implemented in an ident-potentent way.
9250 @tab reply @code{E}@var{NN}
9253 @tab reply @code{OK}
9257 @tab If not supported.
9261 @tab Reserved for future use
9265 The @samp{C}, @samp{c}, @samp{S}, @samp{s} and @samp{?} packets can
9266 receive any of the below as a reply. In the case of the @samp{C},
9267 @samp{c}, @samp{S} and @samp{s} packets, that reply is only returned
9268 when the target halts. In the below the exact meaning of @samp{signal
9269 number} is poorly defined. In general one of the UNIX signal numbering
9270 conventions is used.
9272 @multitable @columnfractions .4 .6
9274 @item @code{S}@var{AA}
9275 @tab @var{AA} is the signal number
9277 @item @code{T}@var{AA}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}
9279 @var{AA} = two hex digit signal number; @var{n...} = register number
9280 (hex), @var{r...} = target byte ordered register contents, size defined
9281 by @code{REGISTER_RAW_SIZE}; @var{n...} = @samp{thread}, @var{r...} =
9282 thread process ID, this is a hex integer; @var{n...} = other string not
9283 starting with valid hex digit. @value{GDBN} should ignore this
9284 @var{n...}, @var{r...} pair and go on to the next. This way we can
9285 extend the protocol.
9287 @item @code{W}@var{AA}
9289 The process exited, and @var{AA} is the exit status. This is only
9290 applicable for certains sorts of targets.
9292 @item @code{X}@var{AA}
9294 The process terminated with signal @var{AA}.
9296 @item @code{N}@var{AA}@code{;}@var{tttttttt}@code{;}@var{dddddddd}@code{;}@var{bbbbbbbb} @strong{(obsolete)}
9298 @var{AA} = signal number; @var{tttttttt} = address of symbol "_start";
9299 @var{dddddddd} = base of data section; @var{bbbbbbbb} = base of bss
9300 section. @emph{Note: only used by Cisco Systems targets. The difference
9301 between this reply and the "qOffsets" query is that the 'N' packet may
9302 arrive spontaneously whereas the 'qOffsets' is a query initiated by the
9305 @item @code{O}@var{XX...}
9307 @var{XX...} is hex encoding of @sc{ascii} data. This can happen at any time
9308 while the program is running and the debugger should continue to wait
9313 The following set and query packets have already been defined.
9315 @multitable @columnfractions .2 .2 .6
9317 @item current thread
9318 @tab @code{q}@code{C}
9319 @tab Return the current thread id.
9321 @tab reply @code{QC}@var{pid}
9323 Where @var{pid} is a HEX encoded 16 bit process id.
9326 @tab Any other reply implies the old pid.
9328 @item compute CRC of memory block
9329 @tab @code{q}@code{CRC:}@var{addr}@code{,}@var{length}
9332 @tab reply @code{E}@var{NN}
9333 @tab An error (such as memory fault)
9335 @tab reply @code{C}@var{CRC32}
9336 @tab A 32 bit cyclic redundancy check of the specified memory region.
9338 @item query @var{LIST} or @var{threadLIST} @strong{(deprecated)}
9339 @tab @code{q}@code{L}@var{startflag}@var{threadcount}@var{nextthread}
9341 Obtain thread information from RTOS. Where: @var{startflag} (one hex
9342 digit) is one to indicate the first query and zero to indicate a
9343 subsequent query; @var{threadcount} (two hex digits) is the maximum
9344 number of threads the response packet can contain; and @var{nextthread}
9345 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
9346 returned in the response as @var{argthread}.
9348 @tab reply @code{q}@code{M}@var{count}@var{done}@var{argthread}@var{thread...}
9350 Where: @var{count} (two hex digits) is the number of threads being
9351 returned; @var{done} (one hex digit) is zero to indicate more threads
9352 and one indicates no further threads; @var{argthreadid} (eight hex
9353 digits) is @var{nextthread} from the request packet; @var{thread...} is
9354 a sequence of thread IDs from the target. @var{threadid} (eight hex
9355 digits). See @code{remote.c:parse_threadlist_response()}.
9357 @item query sect offs
9358 @tab @code{q}@code{Offsets}
9360 Get section offsets that the target used when re-locating the downloaded
9361 image. @emph{Note: while a @code{Bss} offset is included in the
9362 response, @value{GDBN} ignores this and instead applies the @code{Data}
9363 offset to the @code{Bss} section.}
9365 @tab reply @code{Text=}@var{xxx}@code{;Data=}@var{yyy}@code{;Bss=}@var{zzz}
9367 @item thread info request
9368 @tab @code{q}@code{P}@var{mode}@var{threadid}
9370 Returns information on @var{threadid}. Where: @var{mode} is a hex
9371 encoded 32 bit mode; @var{threadid} is a hex encoded 64 bit thread ID.
9375 See @code{remote.c:remote_unpack_thread_info_response()}.
9377 @item remote command
9378 @tab @code{q}@code{Rcmd,}@var{COMMAND}
9380 @var{COMMAND} (hex encoded) is passed to the local interpreter for
9381 execution. Invalid commands should be reported using the output string.
9382 Before the final result packet, the target may also respond with a
9383 number of intermediate @code{O}@var{OUTPUT} console output
9384 packets. @emph{Implementors should note that providing access to a
9385 stubs's interpreter may have security implications}.
9387 @tab reply @code{OK}
9389 A command response with no output.
9391 @tab reply @var{OUTPUT}
9393 A command response with the hex encoded output string @var{OUTPUT}.
9395 @tab reply @code{E}@var{NN}
9397 Indicate a badly formed request.
9402 When @samp{q}@samp{Rcmd} is not recognized.
9406 The following @samp{g}/@samp{G} packets have previously been defined.
9407 In the below, some thirty-two bit registers are transferred as sixty-four
9408 bits. Those registers should be zero/sign extended (which?) to fill the
9409 space allocated. Register bytes are transfered in target byte order.
9410 The two nibbles within a register byte are transfered most-significant -
9413 @multitable @columnfractions .5 .5
9417 All registers are transfered as thirty-two bit quantities in the order:
9418 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
9419 registers; fsr; fir; fp.
9423 All registers are transfered as sixty-four bit quantities (including
9424 thirty-two bit registers such as @code{sr}). The ordering is the same
9429 Example sequence of a target being re-started. Notice how the restart
9430 does not get any direct output:
9435 @emph{target restarts}
9438 -> @code{T001:1234123412341234}
9442 Example sequence of a target being stepped by a single instruction:
9450 -> @code{T001:1234123412341234}
9458 @kindex set remotedebug@r{, serial protocol}
9459 @kindex show remotedebug@r{, serial protocol}
9460 @cindex packets, reporting on stdout
9461 @cindex serial connections, debugging
9462 If you have trouble with the serial connection, you can use the command
9463 @code{set remotedebug}. This makes @value{GDBN} report on all packets sent
9464 back and forth across the serial line to the remote machine. The
9465 packet-debugging information is printed on the @value{GDBN} standard output
9466 stream. @code{set remotedebug off} turns it off, and @code{show
9467 remotedebug} shows you its current state.
9470 @subsubsection Using the @code{gdbserver} program
9473 @cindex remote connection without stubs
9474 @code{gdbserver} is a control program for Unix-like systems, which
9475 allows you to connect your program with a remote @value{GDBN} via
9476 @code{target remote}---but without linking in the usual debugging stub.
9478 @code{gdbserver} is not a complete replacement for the debugging stubs,
9479 because it requires essentially the same operating-system facilities
9480 that @value{GDBN} itself does. In fact, a system that can run
9481 @code{gdbserver} to connect to a remote @value{GDBN} could also run
9482 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
9483 because it is a much smaller program than @value{GDBN} itself. It is
9484 also easier to port than all of @value{GDBN}, so you may be able to get
9485 started more quickly on a new system by using @code{gdbserver}.
9486 Finally, if you develop code for real-time systems, you may find that
9487 the tradeoffs involved in real-time operation make it more convenient to
9488 do as much development work as possible on another system, for example
9489 by cross-compiling. You can use @code{gdbserver} to make a similar
9490 choice for debugging.
9492 @value{GDBN} and @code{gdbserver} communicate via either a serial line
9493 or a TCP connection, using the standard @value{GDBN} remote serial
9497 @item On the target machine,
9498 you need to have a copy of the program you want to debug.
9499 @code{gdbserver} does not need your program's symbol table, so you can
9500 strip the program if necessary to save space. @value{GDBN} on the host
9501 system does all the symbol handling.
9503 To use the server, you must tell it how to communicate with @value{GDBN};
9504 the name of your program; and the arguments for your program. The
9508 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
9511 @var{comm} is either a device name (to use a serial line) or a TCP
9512 hostname and portnumber. For example, to debug Emacs with the argument
9513 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
9517 target> gdbserver /dev/com1 emacs foo.txt
9520 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
9523 To use a TCP connection instead of a serial line:
9526 target> gdbserver host:2345 emacs foo.txt
9529 The only difference from the previous example is the first argument,
9530 specifying that you are communicating with the host @value{GDBN} via
9531 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
9532 expect a TCP connection from machine @samp{host} to local TCP port 2345.
9533 (Currently, the @samp{host} part is ignored.) You can choose any number
9534 you want for the port number as long as it does not conflict with any
9535 TCP ports already in use on the target system (for example, @code{23} is
9536 reserved for @code{telnet}).@footnote{If you choose a port number that
9537 conflicts with another service, @code{gdbserver} prints an error message
9538 and exits.} You must use the same port number with the host @value{GDBN}
9539 @code{target remote} command.
9541 @item On the @value{GDBN} host machine,
9542 you need an unstripped copy of your program, since @value{GDBN} needs
9543 symbols and debugging information. Start up @value{GDBN} as usual,
9544 using the name of the local copy of your program as the first argument.
9545 (You may also need the @w{@samp{--baud}} option if the serial line is
9546 running at anything other than 9600@dmn{bps}.) After that, use @code{target
9547 remote} to establish communications with @code{gdbserver}. Its argument
9548 is either a device name (usually a serial device, like
9549 @file{/dev/ttyb}), or a TCP port descriptor in the form
9550 @code{@var{host}:@var{PORT}}. For example:
9553 (@value{GDBP}) target remote /dev/ttyb
9557 communicates with the server via serial line @file{/dev/ttyb}, and
9560 (@value{GDBP}) target remote the-target:2345
9564 communicates via a TCP connection to port 2345 on host @w{@file{the-target}}.
9565 For TCP connections, you must start up @code{gdbserver} prior to using
9566 the @code{target remote} command. Otherwise you may get an error whose
9567 text depends on the host system, but which usually looks something like
9568 @samp{Connection refused}.
9572 @subsubsection Using the @code{gdbserve.nlm} program
9574 @kindex gdbserve.nlm
9575 @code{gdbserve.nlm} is a control program for NetWare systems, which
9576 allows you to connect your program with a remote @value{GDBN} via
9577 @code{target remote}.
9579 @value{GDBN} and @code{gdbserve.nlm} communicate via a serial line,
9580 using the standard @value{GDBN} remote serial protocol.
9583 @item On the target machine,
9584 you need to have a copy of the program you want to debug.
9585 @code{gdbserve.nlm} does not need your program's symbol table, so you
9586 can strip the program if necessary to save space. @value{GDBN} on the
9587 host system does all the symbol handling.
9589 To use the server, you must tell it how to communicate with
9590 @value{GDBN}; the name of your program; and the arguments for your
9591 program. The syntax is:
9594 load gdbserve [ BOARD=@var{board} ] [ PORT=@var{port} ]
9595 [ BAUD=@var{baud} ] @var{program} [ @var{args} @dots{} ]
9598 @var{board} and @var{port} specify the serial line; @var{baud} specifies
9599 the baud rate used by the connection. @var{port} and @var{node} default
9600 to 0, @var{baud} defaults to 9600@dmn{bps}.
9602 For example, to debug Emacs with the argument @samp{foo.txt}and
9603 communicate with @value{GDBN} over serial port number 2 or board 1
9604 using a 19200@dmn{bps} connection:
9607 load gdbserve BOARD=1 PORT=2 BAUD=19200 emacs foo.txt
9610 @item On the @value{GDBN} host machine,
9611 you need an unstripped copy of your program, since @value{GDBN} needs
9612 symbols and debugging information. Start up @value{GDBN} as usual,
9613 using the name of the local copy of your program as the first argument.
9614 (You may also need the @w{@samp{--baud}} option if the serial line is
9615 running at anything other than 9600@dmn{bps}. After that, use @code{target
9616 remote} to establish communications with @code{gdbserve.nlm}. Its
9617 argument is a device name (usually a serial device, like
9618 @file{/dev/ttyb}). For example:
9621 (@value{GDBP}) target remote /dev/ttyb
9625 communications with the server via serial line @file{/dev/ttyb}.
9629 @section Kernel Object Display
9631 @cindex kernel object display
9632 @cindex kernel object
9635 Some targets support kernel object display. Using this facility,
9636 @value{GDBN} communicates specially with the underlying operating system
9637 and can display information about operating system-level objects such as
9638 mutexes and other synchronization objects. Exactly which objects can be
9639 displayed is determined on a per-OS basis.
9641 Use the @code{set os} command to set the operating system. This tells
9642 @value{GDBN} which kernel object display module to initialize:
9645 (@value{GDBP}) set os cisco
9648 If @code{set os} succeeds, @value{GDBN} will display some information
9649 about the operating system, and will create a new @code{info} command
9650 which can be used to query the target. The @code{info} command is named
9651 after the operating system:
9654 (@value{GDBP}) info cisco
9655 List of Cisco Kernel Objects
9657 any Any and all objects
9660 Further subcommands can be used to query about particular objects known
9663 There is currently no way to determine whether a given operating system
9664 is supported other than to try it.
9667 @node Configurations
9668 @chapter Configuration-Specific Information
9670 While nearly all @value{GDBN} commands are available for all native and
9671 cross versions of the debugger, there are some exceptions. This chapter
9672 describes things that are only available in certain configurations.
9674 There are three major categories of configurations: native
9675 configurations, where the host and target are the same, embedded
9676 operating system configurations, which are usually the same for several
9677 different processor architectures, and bare embedded processors, which
9678 are quite different from each other.
9683 * Embedded Processors::
9690 This section describes details specific to particular native
9695 * SVR4 Process Information:: SVR4 process information
9701 On HP-UX systems, if you refer to a function or variable name that
9702 begins with a dollar sign, @value{GDBN} searches for a user or system
9703 name first, before it searches for a convenience variable.
9705 @node SVR4 Process Information
9706 @subsection SVR4 process information
9709 @cindex process image
9711 Many versions of SVR4 provide a facility called @samp{/proc} that can be
9712 used to examine the image of a running process using file-system
9713 subroutines. If @value{GDBN} is configured for an operating system with
9714 this facility, the command @code{info proc} is available to report on
9715 several kinds of information about the process running your program.
9716 @code{info proc} works only on SVR4 systems that include the
9717 @code{procfs} code. This includes OSF/1 (Digital Unix), Solaris, Irix,
9718 and Unixware, but not HP-UX or Linux, for example.
9723 Summarize available information about the process.
9725 @kindex info proc mappings
9726 @item info proc mappings
9727 Report on the address ranges accessible in the program, with information
9728 on whether your program may read, write, or execute each range.
9730 @kindex info proc times
9731 @item info proc times
9732 Starting time, user CPU time, and system CPU time for your program and
9735 @kindex info proc id
9737 Report on the process IDs related to your program: its own process ID,
9738 the ID of its parent, the process group ID, and the session ID.
9740 @kindex info proc status
9741 @item info proc status
9742 General information on the state of the process. If the process is
9743 stopped, this report includes the reason for stopping, and any signal
9747 Show all the above information about the process.
9751 @section Embedded Operating Systems
9753 This section describes configurations involving the debugging of
9754 embedded operating systems that are available for several different
9758 * VxWorks:: Using @value{GDBN} with VxWorks
9761 @value{GDBN} includes the ability to debug programs running on
9762 various real-time operating systems.
9765 @subsection Using @value{GDBN} with VxWorks
9771 @kindex target vxworks
9772 @item target vxworks @var{machinename}
9773 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
9774 is the target system's machine name or IP address.
9778 On VxWorks, @code{load} links @var{filename} dynamically on the
9779 current target system as well as adding its symbols in @value{GDBN}.
9781 @value{GDBN} enables developers to spawn and debug tasks running on networked
9782 VxWorks targets from a Unix host. Already-running tasks spawned from
9783 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
9784 both the Unix host and on the VxWorks target. The program
9785 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
9786 installed with the name @code{vxgdb}, to distinguish it from a
9787 @value{GDBN} for debugging programs on the host itself.)
9790 @item VxWorks-timeout @var{args}
9791 @kindex vxworks-timeout
9792 All VxWorks-based targets now support the option @code{vxworks-timeout}.
9793 This option is set by the user, and @var{args} represents the number of
9794 seconds @value{GDBN} waits for responses to rpc's. You might use this if
9795 your VxWorks target is a slow software simulator or is on the far side
9796 of a thin network line.
9799 The following information on connecting to VxWorks was current when
9800 this manual was produced; newer releases of VxWorks may use revised
9804 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
9805 to include the remote debugging interface routines in the VxWorks
9806 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
9807 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
9808 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
9809 source debugging task @code{tRdbTask} when VxWorks is booted. For more
9810 information on configuring and remaking VxWorks, see the manufacturer's
9812 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
9814 Once you have included @file{rdb.a} in your VxWorks system image and set
9815 your Unix execution search path to find @value{GDBN}, you are ready to
9816 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
9817 @code{vxgdb}, depending on your installation).
9819 @value{GDBN} comes up showing the prompt:
9826 * VxWorks Connection:: Connecting to VxWorks
9827 * VxWorks Download:: VxWorks download
9828 * VxWorks Attach:: Running tasks
9831 @node VxWorks Connection
9832 @subsubsection Connecting to VxWorks
9834 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
9835 network. To connect to a target whose host name is ``@code{tt}'', type:
9838 (vxgdb) target vxworks tt
9842 @value{GDBN} displays messages like these:
9845 Attaching remote machine across net...
9850 @value{GDBN} then attempts to read the symbol tables of any object modules
9851 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
9852 these files by searching the directories listed in the command search
9853 path (@pxref{Environment, ,Your program's environment}); if it fails
9854 to find an object file, it displays a message such as:
9857 prog.o: No such file or directory.
9860 When this happens, add the appropriate directory to the search path with
9861 the @value{GDBN} command @code{path}, and execute the @code{target}
9864 @node VxWorks Download
9865 @subsubsection VxWorks download
9867 @cindex download to VxWorks
9868 If you have connected to the VxWorks target and you want to debug an
9869 object that has not yet been loaded, you can use the @value{GDBN}
9870 @code{load} command to download a file from Unix to VxWorks
9871 incrementally. The object file given as an argument to the @code{load}
9872 command is actually opened twice: first by the VxWorks target in order
9873 to download the code, then by @value{GDBN} in order to read the symbol
9874 table. This can lead to problems if the current working directories on
9875 the two systems differ. If both systems have NFS mounted the same
9876 filesystems, you can avoid these problems by using absolute paths.
9877 Otherwise, it is simplest to set the working directory on both systems
9878 to the directory in which the object file resides, and then to reference
9879 the file by its name, without any path. For instance, a program
9880 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
9881 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
9882 program, type this on VxWorks:
9885 -> cd "@var{vxpath}/vw/demo/rdb"
9889 Then, in @value{GDBN}, type:
9892 (vxgdb) cd @var{hostpath}/vw/demo/rdb
9896 @value{GDBN} displays a response similar to this:
9899 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
9902 You can also use the @code{load} command to reload an object module
9903 after editing and recompiling the corresponding source file. Note that
9904 this makes @value{GDBN} delete all currently-defined breakpoints,
9905 auto-displays, and convenience variables, and to clear the value
9906 history. (This is necessary in order to preserve the integrity of
9907 debugger's data structures that reference the target system's symbol
9910 @node VxWorks Attach
9911 @subsubsection Running tasks
9913 @cindex running VxWorks tasks
9914 You can also attach to an existing task using the @code{attach} command as
9918 (vxgdb) attach @var{task}
9922 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
9923 or suspended when you attach to it. Running tasks are suspended at
9924 the time of attachment.
9926 @node Embedded Processors
9927 @section Embedded Processors
9929 This section goes into details specific to particular embedded
9933 * A29K Embedded:: AMD A29K Embedded
9935 * H8/300:: Hitachi H8/300
9936 * H8/500:: Hitachi H8/500
9938 * M32R/D:: Mitsubishi M32R/D
9939 * M68K:: Motorola M68K
9940 * M88K:: Motorola M88K
9941 * MIPS Embedded:: MIPS Embedded
9942 * PA:: HP PA Embedded
9945 * Sparclet:: Tsqware Sparclet
9946 * Sparclite:: Fujitsu Sparclite
9947 * ST2000:: Tandem ST2000
9948 * Z8000:: Zilog Z8000
9952 @subsection AMD A29K Embedded
9957 * Comms (EB29K):: Communications setup
9958 * gdb-EB29K:: EB29K cross-debugging
9959 * Remote Log:: Remote log
9964 @kindex target adapt
9965 @item target adapt @var{dev}
9966 Adapt monitor for A29K.
9968 @kindex target amd-eb
9969 @item target amd-eb @var{dev} @var{speed} @var{PROG}
9971 Remote PC-resident AMD EB29K board, attached over serial lines.
9972 @var{dev} is the serial device, as for @code{target remote};
9973 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
9974 name of the program to be debugged, as it appears to DOS on the PC.
9975 @xref{A29K EB29K, ,EBMON protocol for AMD29K}.
9980 @subsubsection A29K UDI
9983 @cindex AMD29K via UDI
9985 @value{GDBN} supports AMD's UDI (``Universal Debugger Interface'')
9986 protocol for debugging the a29k processor family. To use this
9987 configuration with AMD targets running the MiniMON monitor, you need the
9988 program @code{MONTIP}, available from AMD at no charge. You can also
9989 use @value{GDBN} with the UDI-conformant a29k simulator program
9990 @code{ISSTIP}, also available from AMD.
9993 @item target udi @var{keyword}
9995 Select the UDI interface to a remote a29k board or simulator, where
9996 @var{keyword} is an entry in the AMD configuration file @file{udi_soc}.
9997 This file contains keyword entries which specify parameters used to
9998 connect to a29k targets. If the @file{udi_soc} file is not in your
9999 working directory, you must set the environment variable @samp{UDICONF}
10004 @subsubsection EBMON protocol for AMD29K
10006 @cindex EB29K board
10007 @cindex running 29K programs
10009 AMD distributes a 29K development board meant to fit in a PC, together
10010 with a DOS-hosted monitor program called @code{EBMON}. As a shorthand
10011 term, this development system is called the ``EB29K''. To use
10012 @value{GDBN} from a Unix system to run programs on the EB29K board, you
10013 must first connect a serial cable between the PC (which hosts the EB29K
10014 board) and a serial port on the Unix system. In the following, we
10015 assume you've hooked the cable between the PC's @file{COM1} port and
10016 @file{/dev/ttya} on the Unix system.
10018 @node Comms (EB29K)
10019 @subsubsection Communications setup
10021 The next step is to set up the PC's port, by doing something like this
10025 C:\> MODE com1:9600,n,8,1,none
10029 This example---run on an MS DOS 4.0 system---sets the PC port to 9600
10030 bps, no parity, eight data bits, one stop bit, and no ``retry'' action;
10031 you must match the communications parameters when establishing the Unix
10032 end of the connection as well.
10033 @c FIXME: Who knows what this "no retry action" crud from the DOS manual may
10034 @c mean? It's optional; leave it out? ---doc@cygnus.com, 25feb91
10036 @c It's optional, but it's unwise to omit it: who knows what is the
10037 @c default value set when the DOS machines boots? "No retry" means that
10038 @c the DOS serial device driver won't retry the operation if it fails;
10039 @c I understand that this is needed because the GDB serial protocol
10040 @c handles any errors and retransmissions itself. ---Eli Zaretskii, 3sep99
10042 To give control of the PC to the Unix side of the serial line, type
10043 the following at the DOS console:
10050 (Later, if you wish to return control to the DOS console, you can use
10051 the command @code{CTTY con}---but you must send it over the device that
10052 had control, in our example over the @file{COM1} serial line.)
10054 From the Unix host, use a communications program such as @code{tip} or
10055 @code{cu} to communicate with the PC; for example,
10058 cu -s 9600 -l /dev/ttya
10062 The @code{cu} options shown specify, respectively, the linespeed and the
10063 serial port to use. If you use @code{tip} instead, your command line
10064 may look something like the following:
10067 tip -9600 /dev/ttya
10071 Your system may require a different name where we show
10072 @file{/dev/ttya} as the argument to @code{tip}. The communications
10073 parameters, including which port to use, are associated with the
10074 @code{tip} argument in the ``remote'' descriptions file---normally the
10075 system table @file{/etc/remote}.
10076 @c FIXME: What if anything needs doing to match the "n,8,1,none" part of
10077 @c the DOS side's comms setup? cu can support -o (odd
10078 @c parity), -e (even parity)---apparently no settings for no parity or
10079 @c for character size. Taken from stty maybe...? John points out tip
10080 @c can set these as internal variables, eg ~s parity=none; man stty
10081 @c suggests that it *might* work to stty these options with stdin or
10082 @c stdout redirected... ---doc@cygnus.com, 25feb91
10084 @c There's nothing to be done for the "none" part of the DOS MODE
10085 @c command. The rest of the parameters should be matched by the
10086 @c baudrate, bits, and parity used by the Unix side. ---Eli Zaretskii, 3Sep99
10089 Using the @code{tip} or @code{cu} connection, change the DOS working
10090 directory to the directory containing a copy of your 29K program, then
10091 start the PC program @code{EBMON} (an EB29K control program supplied
10092 with your board by AMD). You should see an initial display from
10093 @code{EBMON} similar to the one that follows, ending with the
10094 @code{EBMON} prompt @samp{#}---
10099 G:\> CD \usr\joe\work29k
10101 G:\USR\JOE\WORK29K> EBMON
10102 Am29000 PC Coprocessor Board Monitor, version 3.0-18
10103 Copyright 1990 Advanced Micro Devices, Inc.
10104 Written by Gibbons and Associates, Inc.
10106 Enter '?' or 'H' for help
10108 PC Coprocessor Type = EB29K
10110 Memory Base = 0xd0000
10112 Data Memory Size = 2048KB
10113 Available I-RAM Range = 0x8000 to 0x1fffff
10114 Available D-RAM Range = 0x80002000 to 0x801fffff
10117 Register Stack Size = 0x800
10118 Memory Stack Size = 0x1800
10121 Am29027 Available = No
10122 Byte Write Available = Yes
10127 Then exit the @code{cu} or @code{tip} program (done in the example by
10128 typing @code{~.} at the @code{EBMON} prompt). @code{EBMON} keeps
10129 running, ready for @value{GDBN} to take over.
10131 For this example, we've assumed what is probably the most convenient
10132 way to make sure the same 29K program is on both the PC and the Unix
10133 system: a PC/NFS connection that establishes ``drive @file{G:}'' on the
10134 PC as a file system on the Unix host. If you do not have PC/NFS or
10135 something similar connecting the two systems, you must arrange some
10136 other way---perhaps floppy-disk transfer---of getting the 29K program
10137 from the Unix system to the PC; @value{GDBN} does @emph{not} download it over the
10141 @subsubsection EB29K cross-debugging
10143 Finally, @code{cd} to the directory containing an image of your 29K
10144 program on the Unix system, and start @value{GDBN}---specifying as argument the
10145 name of your 29K program:
10148 cd /usr/joe/work29k
10153 Now you can use the @code{target} command:
10156 target amd-eb /dev/ttya 9600 MYFOO
10157 @c FIXME: test above 'target amd-eb' as spelled, with caps! caps are meant to
10158 @c emphasize that this is the name as seen by DOS (since I think DOS is
10159 @c single-minded about case of letters). ---doc@cygnus.com, 25feb91
10163 In this example, we've assumed your program is in a file called
10164 @file{myfoo}. Note that the filename given as the last argument to
10165 @code{target amd-eb} should be the name of the program as it appears to DOS.
10166 In our example this is simply @code{MYFOO}, but in general it can include
10167 a DOS path, and depending on your transfer mechanism may not resemble
10168 the name on the Unix side.
10170 At this point, you can set any breakpoints you wish; when you are ready
10171 to see your program run on the 29K board, use the @value{GDBN} command
10174 To stop debugging the remote program, use the @value{GDBN} @code{detach}
10177 To return control of the PC to its console, use @code{tip} or @code{cu}
10178 once again, after your @value{GDBN} session has concluded, to attach to
10179 @code{EBMON}. You can then type the command @code{q} to shut down
10180 @code{EBMON}, returning control to the DOS command-line interpreter.
10181 Type @kbd{CTTY con} to return command input to the main DOS console,
10182 and type @kbd{~.} to leave @code{tip} or @code{cu}.
10185 @subsubsection Remote log
10187 @cindex log file for EB29K
10189 The @code{target amd-eb} command creates a file @file{eb.log} in the
10190 current working directory, to help debug problems with the connection.
10191 @file{eb.log} records all the output from @code{EBMON}, including echoes
10192 of the commands sent to it. Running @samp{tail -f} on this file in
10193 another window often helps to understand trouble with @code{EBMON}, or
10194 unexpected events on the PC side of the connection.
10202 @item target rdi @var{dev}
10203 ARM Angel monitor, via RDI library interface to ADP protocol. You may
10204 use this target to communicate with both boards running the Angel
10205 monitor, or with the EmbeddedICE JTAG debug device.
10208 @item target rdp @var{dev}
10214 @subsection Hitachi H8/300
10218 @kindex target hms@r{, with H8/300}
10219 @item target hms @var{dev}
10220 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
10221 Use special commands @code{device} and @code{speed} to control the serial
10222 line and the communications speed used.
10224 @kindex target e7000@r{, with H8/300}
10225 @item target e7000 @var{dev}
10226 E7000 emulator for Hitachi H8 and SH.
10228 @kindex target sh3@r{, with H8/300}
10229 @kindex target sh3e@r{, with H8/300}
10230 @item target sh3 @var{dev}
10231 @itemx target sh3e @var{dev}
10232 Hitachi SH-3 and SH-3E target systems.
10236 @cindex download to H8/300 or H8/500
10237 @cindex H8/300 or H8/500 download
10238 @cindex download to Hitachi SH
10239 @cindex Hitachi SH download
10240 When you select remote debugging to a Hitachi SH, H8/300, or H8/500
10241 board, the @code{load} command downloads your program to the Hitachi
10242 board and also opens it as the current executable target for
10243 @value{GDBN} on your host (like the @code{file} command).
10245 @value{GDBN} needs to know these things to talk to your
10246 Hitachi SH, H8/300, or H8/500:
10250 that you want to use @samp{target hms}, the remote debugging interface
10251 for Hitachi microprocessors, or @samp{target e7000}, the in-circuit
10252 emulator for the Hitachi SH and the Hitachi 300H. (@samp{target hms} is
10253 the default when @value{GDBN} is configured specifically for the Hitachi SH,
10254 H8/300, or H8/500.)
10257 what serial device connects your host to your Hitachi board (the first
10258 serial device available on your host is the default).
10261 what speed to use over the serial device.
10265 * Hitachi Boards:: Connecting to Hitachi boards.
10266 * Hitachi ICE:: Using the E7000 In-Circuit Emulator.
10267 * Hitachi Special:: Special @value{GDBN} commands for Hitachi micros.
10270 @node Hitachi Boards
10271 @subsubsection Connecting to Hitachi boards
10273 @c only for Unix hosts
10275 @cindex serial device, Hitachi micros
10276 Use the special @code{@value{GDBN}} command @samp{device @var{port}} if you
10277 need to explicitly set the serial device. The default @var{port} is the
10278 first available port on your host. This is only necessary on Unix
10279 hosts, where it is typically something like @file{/dev/ttya}.
10282 @cindex serial line speed, Hitachi micros
10283 @code{@value{GDBN}} has another special command to set the communications
10284 speed: @samp{speed @var{bps}}. This command also is only used from Unix
10285 hosts; on DOS hosts, set the line speed as usual from outside @value{GDBN} with
10286 the DOS @code{mode} command (for instance,
10287 @w{@kbd{mode com2:9600,n,8,1,p}} for a 9600@dmn{bps} connection).
10289 The @samp{device} and @samp{speed} commands are available only when you
10290 use a Unix host to debug your Hitachi microprocessor programs. If you
10292 @value{GDBN} depends on an auxiliary terminate-and-stay-resident program
10293 called @code{asynctsr} to communicate with the development board
10294 through a PC serial port. You must also use the DOS @code{mode} command
10295 to set up the serial port on the DOS side.
10297 The following sample session illustrates the steps needed to start a
10298 program under @value{GDBN} control on an H8/300. The example uses a
10299 sample H8/300 program called @file{t.x}. The procedure is the same for
10300 the Hitachi SH and the H8/500.
10302 First hook up your development board. In this example, we use a
10303 board attached to serial port @code{COM2}; if you use a different serial
10304 port, substitute its name in the argument of the @code{mode} command.
10305 When you call @code{asynctsr}, the auxiliary comms program used by the
10306 debugger, you give it just the numeric part of the serial port's name;
10307 for example, @samp{asyncstr 2} below runs @code{asyncstr} on
10311 C:\H8300\TEST> asynctsr 2
10312 C:\H8300\TEST> mode com2:9600,n,8,1,p
10314 Resident portion of MODE loaded
10316 COM2: 9600, n, 8, 1, p
10321 @emph{Warning:} We have noticed a bug in PC-NFS that conflicts with
10322 @code{asynctsr}. If you also run PC-NFS on your DOS host, you may need to
10323 disable it, or even boot without it, to use @code{asynctsr} to control
10324 your development board.
10327 @kindex target hms@r{, and serial protocol}
10328 Now that serial communications are set up, and the development board is
10329 connected, you can start up @value{GDBN}. Call @code{@value{GDBP}} with
10330 the name of your program as the argument. @code{@value{GDBN}} prompts
10331 you, as usual, with the prompt @samp{(@value{GDBP})}. Use two special
10332 commands to begin your debugging session: @samp{target hms} to specify
10333 cross-debugging to the Hitachi board, and the @code{load} command to
10334 download your program to the board. @code{load} displays the names of
10335 the program's sections, and a @samp{*} for each 2K of data downloaded.
10336 (If you want to refresh @value{GDBN} data on symbols or on the
10337 executable file without downloading, use the @value{GDBN} commands
10338 @code{file} or @code{symbol-file}. These commands, and @code{load}
10339 itself, are described in @ref{Files,,Commands to specify files}.)
10342 (eg-C:\H8300\TEST) @value{GDBP} t.x
10343 @value{GDBN} is free software and you are welcome to distribute copies
10344 of it under certain conditions; type "show copying" to see
10346 There is absolutely no warranty for @value{GDBN}; type "show warranty"
10348 @value{GDBN} @value{GDBVN}, Copyright 1992 Free Software Foundation, Inc...
10349 (@value{GDBP}) target hms
10350 Connected to remote H8/300 HMS system.
10351 (@value{GDBP}) load t.x
10352 .text : 0x8000 .. 0xabde ***********
10353 .data : 0xabde .. 0xad30 *
10354 .stack : 0xf000 .. 0xf014 *
10357 At this point, you're ready to run or debug your program. From here on,
10358 you can use all the usual @value{GDBN} commands. The @code{break} command
10359 sets breakpoints; the @code{run} command starts your program;
10360 @code{print} or @code{x} display data; the @code{continue} command
10361 resumes execution after stopping at a breakpoint. You can use the
10362 @code{help} command at any time to find out more about @value{GDBN} commands.
10364 Remember, however, that @emph{operating system} facilities aren't
10365 available on your development board; for example, if your program hangs,
10366 you can't send an interrupt---but you can press the @sc{reset} switch!
10368 Use the @sc{reset} button on the development board
10371 to interrupt your program (don't use @kbd{ctl-C} on the DOS host---it has
10372 no way to pass an interrupt signal to the development board); and
10375 to return to the @value{GDBN} command prompt after your program finishes
10376 normally. The communications protocol provides no other way for @value{GDBN}
10377 to detect program completion.
10380 In either case, @value{GDBN} sees the effect of a @sc{reset} on the
10381 development board as a ``normal exit'' of your program.
10384 @subsubsection Using the E7000 in-circuit emulator
10386 @kindex target e7000@r{, with Hitachi ICE}
10387 You can use the E7000 in-circuit emulator to develop code for either the
10388 Hitachi SH or the H8/300H. Use one of these forms of the @samp{target
10389 e7000} command to connect @value{GDBN} to your E7000:
10392 @item target e7000 @var{port} @var{speed}
10393 Use this form if your E7000 is connected to a serial port. The
10394 @var{port} argument identifies what serial port to use (for example,
10395 @samp{com2}). The third argument is the line speed in bits per second
10396 (for example, @samp{9600}).
10398 @item target e7000 @var{hostname}
10399 If your E7000 is installed as a host on a TCP/IP network, you can just
10400 specify its hostname; @value{GDBN} uses @code{telnet} to connect.
10403 @node Hitachi Special
10404 @subsubsection Special @value{GDBN} commands for Hitachi micros
10406 Some @value{GDBN} commands are available only for the H8/300:
10410 @kindex set machine
10411 @kindex show machine
10412 @item set machine h8300
10413 @itemx set machine h8300h
10414 Condition @value{GDBN} for one of the two variants of the H8/300
10415 architecture with @samp{set machine}. You can use @samp{show machine}
10416 to check which variant is currently in effect.
10425 @kindex set memory @var{mod}
10426 @cindex memory models, H8/500
10427 @item set memory @var{mod}
10429 Specify which H8/500 memory model (@var{mod}) you are using with
10430 @samp{set memory}; check which memory model is in effect with @samp{show
10431 memory}. The accepted values for @var{mod} are @code{small},
10432 @code{big}, @code{medium}, and @code{compact}.
10437 @subsection Intel i960
10441 @kindex target mon960
10442 @item target mon960 @var{dev}
10443 MON960 monitor for Intel i960.
10445 @item target nindy @var{devicename}
10446 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
10447 the name of the serial device to use for the connection, e.g.
10454 @dfn{Nindy} is a ROM Monitor program for Intel 960 target systems. When
10455 @value{GDBN} is configured to control a remote Intel 960 using Nindy, you can
10456 tell @value{GDBN} how to connect to the 960 in several ways:
10460 Through command line options specifying serial port, version of the
10461 Nindy protocol, and communications speed;
10464 By responding to a prompt on startup;
10467 By using the @code{target} command at any point during your @value{GDBN}
10468 session. @xref{Target Commands, ,Commands for managing targets}.
10470 @kindex target nindy
10471 @item target nindy @var{devicename}
10472 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
10473 the name of the serial device to use for the connection, e.g.
10478 @cindex download to Nindy-960
10479 With the Nindy interface to an Intel 960 board, @code{load}
10480 downloads @var{filename} to the 960 as well as adding its symbols in
10484 * Nindy Startup:: Startup with Nindy
10485 * Nindy Options:: Options for Nindy
10486 * Nindy Reset:: Nindy reset command
10489 @node Nindy Startup
10490 @subsubsection Startup with Nindy
10492 If you simply start @code{@value{GDBP}} without using any command-line
10493 options, you are prompted for what serial port to use, @emph{before} you
10494 reach the ordinary @value{GDBN} prompt:
10497 Attach /dev/ttyNN -- specify NN, or "quit" to quit:
10501 Respond to the prompt with whatever suffix (after @samp{/dev/tty})
10502 identifies the serial port you want to use. You can, if you choose,
10503 simply start up with no Nindy connection by responding to the prompt
10504 with an empty line. If you do this and later wish to attach to Nindy,
10505 use @code{target} (@pxref{Target Commands, ,Commands for managing targets}).
10507 @node Nindy Options
10508 @subsubsection Options for Nindy
10510 These are the startup options for beginning your @value{GDBN} session with a
10511 Nindy-960 board attached:
10514 @item -r @var{port}
10515 Specify the serial port name of a serial interface to be used to connect
10516 to the target system. This option is only available when @value{GDBN} is
10517 configured for the Intel 960 target architecture. You may specify
10518 @var{port} as any of: a full pathname (e.g. @samp{-r /dev/ttya}), a
10519 device name in @file{/dev} (e.g. @samp{-r ttya}), or simply the unique
10520 suffix for a specific @code{tty} (e.g. @samp{-r a}).
10523 (An uppercase letter ``O'', not a zero.) Specify that @value{GDBN} should use
10524 the ``old'' Nindy monitor protocol to connect to the target system.
10525 This option is only available when @value{GDBN} is configured for the Intel 960
10526 target architecture.
10529 @emph{Warning:} if you specify @samp{-O}, but are actually trying to
10530 connect to a target system that expects the newer protocol, the connection
10531 fails, appearing to be a speed mismatch. @value{GDBN} repeatedly
10532 attempts to reconnect at several different line speeds. You can abort
10533 this process with an interrupt.
10537 Specify that @value{GDBN} should first send a @code{BREAK} signal to the target
10538 system, in an attempt to reset it, before connecting to a Nindy target.
10541 @emph{Warning:} Many target systems do not have the hardware that this
10542 requires; it only works with a few boards.
10546 The standard @samp{-b} option controls the line speed used on the serial
10551 @subsubsection Nindy reset command
10556 For a Nindy target, this command sends a ``break'' to the remote target
10557 system; this is only useful if the target has been equipped with a
10558 circuit to perform a hard reset (or some other interesting action) when
10559 a break is detected.
10564 @subsection Mitsubishi M32R/D
10568 @kindex target m32r
10569 @item target m32r @var{dev}
10570 Mitsubishi M32R/D ROM monitor.
10577 The Motorola m68k configuration includes ColdFire support, and
10578 target command for the following ROM monitors.
10582 @kindex target abug
10583 @item target abug @var{dev}
10584 ABug ROM monitor for M68K.
10586 @kindex target cpu32bug
10587 @item target cpu32bug @var{dev}
10588 CPU32BUG monitor, running on a CPU32 (M68K) board.
10590 @kindex target dbug
10591 @item target dbug @var{dev}
10592 dBUG ROM monitor for Motorola ColdFire.
10595 @item target est @var{dev}
10596 EST-300 ICE monitor, running on a CPU32 (M68K) board.
10598 @kindex target rom68k
10599 @item target rom68k @var{dev}
10600 ROM 68K monitor, running on an M68K IDP board.
10604 If @value{GDBN} is configured with @code{m68*-ericsson-*}, it will
10605 instead have only a single special target command:
10609 @kindex target es1800
10610 @item target es1800 @var{dev}
10611 ES-1800 emulator for M68K.
10619 @kindex target rombug
10620 @item target rombug @var{dev}
10621 ROMBUG ROM monitor for OS/9000.
10631 @item target bug @var{dev}
10632 BUG monitor, running on a MVME187 (m88k) board.
10636 @node MIPS Embedded
10637 @subsection MIPS Embedded
10639 @cindex MIPS boards
10640 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
10641 MIPS board attached to a serial line. This is available when
10642 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
10645 Use these @value{GDBN} commands to specify the connection to your target board:
10648 @item target mips @var{port}
10649 @kindex target mips @var{port}
10650 To run a program on the board, start up @code{@value{GDBP}} with the
10651 name of your program as the argument. To connect to the board, use the
10652 command @samp{target mips @var{port}}, where @var{port} is the name of
10653 the serial port connected to the board. If the program has not already
10654 been downloaded to the board, you may use the @code{load} command to
10655 download it. You can then use all the usual @value{GDBN} commands.
10657 For example, this sequence connects to the target board through a serial
10658 port, and loads and runs a program called @var{prog} through the
10662 host$ @value{GDBP} @var{prog}
10663 @value{GDBN} is free software and @dots{}
10664 (@value{GDBP}) target mips /dev/ttyb
10665 (@value{GDBP}) load @var{prog}
10669 @item target mips @var{hostname}:@var{portnumber}
10670 On some @value{GDBN} host configurations, you can specify a TCP
10671 connection (for instance, to a serial line managed by a terminal
10672 concentrator) instead of a serial port, using the syntax
10673 @samp{@var{hostname}:@var{portnumber}}.
10675 @item target pmon @var{port}
10676 @kindex target pmon @var{port}
10679 @item target ddb @var{port}
10680 @kindex target ddb @var{port}
10681 NEC's DDB variant of PMON for Vr4300.
10683 @item target lsi @var{port}
10684 @kindex target lsi @var{port}
10685 LSI variant of PMON.
10687 @kindex target r3900
10688 @item target r3900 @var{dev}
10689 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
10691 @kindex target array
10692 @item target array @var{dev}
10693 Array Tech LSI33K RAID controller board.
10699 @value{GDBN} also supports these special commands for MIPS targets:
10702 @item set processor @var{args}
10703 @itemx show processor
10704 @kindex set processor @var{args}
10705 @kindex show processor
10706 Use the @code{set processor} command to set the type of MIPS
10707 processor when you want to access processor-type-specific registers.
10708 For example, @code{set processor @var{r3041}} tells @value{GDBN}
10709 to use the CPO registers appropriate for the 3041 chip.
10710 Use the @code{show processor} command to see what MIPS processor @value{GDBN}
10711 is using. Use the @code{info reg} command to see what registers
10712 @value{GDBN} is using.
10714 @item set mipsfpu double
10715 @itemx set mipsfpu single
10716 @itemx set mipsfpu none
10717 @itemx show mipsfpu
10718 @kindex set mipsfpu
10719 @kindex show mipsfpu
10720 @cindex MIPS remote floating point
10721 @cindex floating point, MIPS remote
10722 If your target board does not support the MIPS floating point
10723 coprocessor, you should use the command @samp{set mipsfpu none} (if you
10724 need this, you may wish to put the command in your @value{GDBN} init
10725 file). This tells @value{GDBN} how to find the return value of
10726 functions which return floating point values. It also allows
10727 @value{GDBN} to avoid saving the floating point registers when calling
10728 functions on the board. If you are using a floating point coprocessor
10729 with only single precision floating point support, as on the @sc{r4650}
10730 processor, use the command @samp{set mipsfpu single}. The default
10731 double precision floating point coprocessor may be selected using
10732 @samp{set mipsfpu double}.
10734 In previous versions the only choices were double precision or no
10735 floating point, so @samp{set mipsfpu on} will select double precision
10736 and @samp{set mipsfpu off} will select no floating point.
10738 As usual, you can inquire about the @code{mipsfpu} variable with
10739 @samp{show mipsfpu}.
10741 @item set remotedebug @var{n}
10742 @itemx show remotedebug
10743 @kindex set remotedebug@r{, MIPS protocol}
10744 @kindex show remotedebug@r{, MIPS protocol}
10745 @cindex @code{remotedebug}, MIPS protocol
10746 @cindex MIPS @code{remotedebug} protocol
10747 @c FIXME! For this to be useful, you must know something about the MIPS
10748 @c FIXME...protocol. Where is it described?
10749 You can see some debugging information about communications with the board
10750 by setting the @code{remotedebug} variable. If you set it to @code{1} using
10751 @samp{set remotedebug 1}, every packet is displayed. If you set it
10752 to @code{2}, every character is displayed. You can check the current value
10753 at any time with the command @samp{show remotedebug}.
10755 @item set timeout @var{seconds}
10756 @itemx set retransmit-timeout @var{seconds}
10757 @itemx show timeout
10758 @itemx show retransmit-timeout
10759 @cindex @code{timeout}, MIPS protocol
10760 @cindex @code{retransmit-timeout}, MIPS protocol
10761 @kindex set timeout
10762 @kindex show timeout
10763 @kindex set retransmit-timeout
10764 @kindex show retransmit-timeout
10765 You can control the timeout used while waiting for a packet, in the MIPS
10766 remote protocol, with the @code{set timeout @var{seconds}} command. The
10767 default is 5 seconds. Similarly, you can control the timeout used while
10768 waiting for an acknowledgement of a packet with the @code{set
10769 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
10770 You can inspect both values with @code{show timeout} and @code{show
10771 retransmit-timeout}. (These commands are @emph{only} available when
10772 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
10774 The timeout set by @code{set timeout} does not apply when @value{GDBN}
10775 is waiting for your program to stop. In that case, @value{GDBN} waits
10776 forever because it has no way of knowing how long the program is going
10777 to run before stopping.
10781 @subsection PowerPC
10785 @kindex target dink32
10786 @item target dink32 @var{dev}
10787 DINK32 ROM monitor.
10789 @kindex target ppcbug
10790 @item target ppcbug @var{dev}
10791 @kindex target ppcbug1
10792 @item target ppcbug1 @var{dev}
10793 PPCBUG ROM monitor for PowerPC.
10796 @item target sds @var{dev}
10797 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
10802 @subsection HP PA Embedded
10806 @kindex target op50n
10807 @item target op50n @var{dev}
10808 OP50N monitor, running on an OKI HPPA board.
10810 @kindex target w89k
10811 @item target w89k @var{dev}
10812 W89K monitor, running on a Winbond HPPA board.
10817 @subsection Hitachi SH
10821 @kindex target hms@r{, with Hitachi SH}
10822 @item target hms @var{dev}
10823 A Hitachi SH board attached via serial line to your host. Use special
10824 commands @code{device} and @code{speed} to control the serial line and
10825 the communications speed used.
10827 @kindex target e7000@r{, with Hitachi SH}
10828 @item target e7000 @var{dev}
10829 E7000 emulator for Hitachi SH.
10831 @kindex target sh3@r{, with SH}
10832 @kindex target sh3e@r{, with SH}
10833 @item target sh3 @var{dev}
10834 @item target sh3e @var{dev}
10835 Hitachi SH-3 and SH-3E target systems.
10840 @subsection Tsqware Sparclet
10844 @value{GDBN} enables developers to debug tasks running on
10845 Sparclet targets from a Unix host.
10846 @value{GDBN} uses code that runs on
10847 both the Unix host and on the Sparclet target. The program
10848 @code{@value{GDBP}} is installed and executed on the Unix host.
10851 @item timeout @var{args}
10852 @kindex remotetimeout
10853 @value{GDBN} supports the option @code{remotetimeout}.
10854 This option is set by the user, and @var{args} represents the number of
10855 seconds @value{GDBN} waits for responses.
10859 When compiling for debugging, include the options @samp{-g} to get debug
10860 information and @samp{-Ttext} to relocate the program to where you wish to
10861 load it on the target. You may also want to add the options @samp{-n} or
10862 @samp{-N} in order to reduce the size of the sections. Example:
10865 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
10868 You can use @code{objdump} to verify that the addresses are what you intended:
10871 sparclet-aout-objdump --headers --syms prog
10876 your Unix execution search path to find @value{GDBN}, you are ready to
10877 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
10878 (or @code{sparclet-aout-gdb}, depending on your installation).
10880 @value{GDBN} comes up showing the prompt:
10887 * Sparclet File:: Setting the file to debug
10888 * Sparclet Connection:: Connecting to Sparclet
10889 * Sparclet Download:: Sparclet download
10890 * Sparclet Execution:: Running and debugging
10893 @node Sparclet File
10894 @subsubsection Setting file to debug
10896 The @value{GDBN} command @code{file} lets you choose with program to debug.
10899 (gdbslet) file prog
10903 @value{GDBN} then attempts to read the symbol table of @file{prog}.
10904 @value{GDBN} locates
10905 the file by searching the directories listed in the command search
10907 If the file was compiled with debug information (option "-g"), source
10908 files will be searched as well.
10909 @value{GDBN} locates
10910 the source files by searching the directories listed in the directory search
10911 path (@pxref{Environment, ,Your program's environment}).
10913 to find a file, it displays a message such as:
10916 prog: No such file or directory.
10919 When this happens, add the appropriate directories to the search paths with
10920 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
10921 @code{target} command again.
10923 @node Sparclet Connection
10924 @subsubsection Connecting to Sparclet
10926 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
10927 To connect to a target on serial port ``@code{ttya}'', type:
10930 (gdbslet) target sparclet /dev/ttya
10931 Remote target sparclet connected to /dev/ttya
10932 main () at ../prog.c:3
10936 @value{GDBN} displays messages like these:
10942 @node Sparclet Download
10943 @subsubsection Sparclet download
10945 @cindex download to Sparclet
10946 Once connected to the Sparclet target,
10947 you can use the @value{GDBN}
10948 @code{load} command to download the file from the host to the target.
10949 The file name and load offset should be given as arguments to the @code{load}
10951 Since the file format is aout, the program must be loaded to the starting
10952 address. You can use @code{objdump} to find out what this value is. The load
10953 offset is an offset which is added to the VMA (virtual memory address)
10954 of each of the file's sections.
10955 For instance, if the program
10956 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
10957 and bss at 0x12010170, in @value{GDBN}, type:
10960 (gdbslet) load prog 0x12010000
10961 Loading section .text, size 0xdb0 vma 0x12010000
10964 If the code is loaded at a different address then what the program was linked
10965 to, you may need to use the @code{section} and @code{add-symbol-file} commands
10966 to tell @value{GDBN} where to map the symbol table.
10968 @node Sparclet Execution
10969 @subsubsection Running and debugging
10971 @cindex running and debugging Sparclet programs
10972 You can now begin debugging the task using @value{GDBN}'s execution control
10973 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
10974 manual for the list of commands.
10978 Breakpoint 1 at 0x12010000: file prog.c, line 3.
10980 Starting program: prog
10981 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
10982 3 char *symarg = 0;
10984 4 char *execarg = "hello!";
10989 @subsection Fujitsu Sparclite
10993 @kindex target sparclite
10994 @item target sparclite @var{dev}
10995 Fujitsu sparclite boards, used only for the purpose of loading.
10996 You must use an additional command to debug the program.
10997 For example: target remote @var{dev} using @value{GDBN} standard
11003 @subsection Tandem ST2000
11005 @value{GDBN} may be used with a Tandem ST2000 phone switch, running Tandem's
11008 To connect your ST2000 to the host system, see the manufacturer's
11009 manual. Once the ST2000 is physically attached, you can run:
11012 target st2000 @var{dev} @var{speed}
11016 to establish it as your debugging environment. @var{dev} is normally
11017 the name of a serial device, such as @file{/dev/ttya}, connected to the
11018 ST2000 via a serial line. You can instead specify @var{dev} as a TCP
11019 connection (for example, to a serial line attached via a terminal
11020 concentrator) using the syntax @code{@var{hostname}:@var{portnumber}}.
11022 The @code{load} and @code{attach} commands are @emph{not} defined for
11023 this target; you must load your program into the ST2000 as you normally
11024 would for standalone operation. @value{GDBN} reads debugging information
11025 (such as symbols) from a separate, debugging version of the program
11026 available on your host computer.
11027 @c FIXME!! This is terribly vague; what little content is here is
11028 @c basically hearsay.
11030 @cindex ST2000 auxiliary commands
11031 These auxiliary @value{GDBN} commands are available to help you with the ST2000
11035 @item st2000 @var{command}
11036 @kindex st2000 @var{cmd}
11037 @cindex STDBUG commands (ST2000)
11038 @cindex commands to STDBUG (ST2000)
11039 Send a @var{command} to the STDBUG monitor. See the manufacturer's
11040 manual for available commands.
11043 @cindex connect (to STDBUG)
11044 Connect the controlling terminal to the STDBUG command monitor. When
11045 you are done interacting with STDBUG, typing either of two character
11046 sequences gets you back to the @value{GDBN} command prompt:
11047 @kbd{@key{RET}~.} (Return, followed by tilde and period) or
11048 @kbd{@key{RET}~@key{C-d}} (Return, followed by tilde and control-D).
11052 @subsection Zilog Z8000
11055 @cindex simulator, Z8000
11056 @cindex Zilog Z8000 simulator
11058 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
11061 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
11062 unsegmented variant of the Z8000 architecture) or the Z8001 (the
11063 segmented variant). The simulator recognizes which architecture is
11064 appropriate by inspecting the object code.
11067 @item target sim @var{args}
11069 @kindex target sim@r{, with Z8000}
11070 Debug programs on a simulated CPU. If the simulator supports setup
11071 options, specify them via @var{args}.
11075 After specifying this target, you can debug programs for the simulated
11076 CPU in the same style as programs for your host computer; use the
11077 @code{file} command to load a new program image, the @code{run} command
11078 to run your program, and so on.
11080 As well as making available all the usual machine registers
11081 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
11082 additional items of information as specially named registers:
11087 Counts clock-ticks in the simulator.
11090 Counts instructions run in the simulator.
11093 Execution time in 60ths of a second.
11097 You can refer to these values in @value{GDBN} expressions with the usual
11098 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
11099 conditional breakpoint that suspends only after at least 5000
11100 simulated clock ticks.
11102 @node Architectures
11103 @section Architectures
11105 This section describes characteristics of architectures that affect
11106 all uses of @value{GDBN} with the architecture, both native and cross.
11119 @kindex set rstack_high_address
11120 @cindex AMD 29K register stack
11121 @cindex register stack, AMD29K
11122 @item set rstack_high_address @var{address}
11123 On AMD 29000 family processors, registers are saved in a separate
11124 @dfn{register stack}. There is no way for @value{GDBN} to determine the
11125 extent of this stack. Normally, @value{GDBN} just assumes that the
11126 stack is ``large enough''. This may result in @value{GDBN} referencing
11127 memory locations that do not exist. If necessary, you can get around
11128 this problem by specifying the ending address of the register stack with
11129 the @code{set rstack_high_address} command. The argument should be an
11130 address, which you probably want to precede with @samp{0x} to specify in
11133 @kindex show rstack_high_address
11134 @item show rstack_high_address
11135 Display the current limit of the register stack, on AMD 29000 family
11143 See the following section.
11148 @cindex stack on Alpha
11149 @cindex stack on MIPS
11150 @cindex Alpha stack
11152 Alpha- and MIPS-based computers use an unusual stack frame, which
11153 sometimes requires @value{GDBN} to search backward in the object code to
11154 find the beginning of a function.
11156 @cindex response time, MIPS debugging
11157 To improve response time (especially for embedded applications, where
11158 @value{GDBN} may be restricted to a slow serial line for this search)
11159 you may want to limit the size of this search, using one of these
11163 @cindex @code{heuristic-fence-post} (Alpha,MIPS)
11164 @item set heuristic-fence-post @var{limit}
11165 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
11166 search for the beginning of a function. A value of @var{0} (the
11167 default) means there is no limit. However, except for @var{0}, the
11168 larger the limit the more bytes @code{heuristic-fence-post} must search
11169 and therefore the longer it takes to run.
11171 @item show heuristic-fence-post
11172 Display the current limit.
11176 These commands are available @emph{only} when @value{GDBN} is configured
11177 for debugging programs on Alpha or MIPS processors.
11180 @node Controlling GDB
11181 @chapter Controlling @value{GDBN}
11183 You can alter the way @value{GDBN} interacts with you by using the
11184 @code{set} command. For commands controlling how @value{GDBN} displays
11185 data, see @ref{Print Settings, ,Print settings}. Other settings are
11190 * Editing:: Command editing
11191 * History:: Command history
11192 * Screen Size:: Screen size
11193 * Numbers:: Numbers
11194 * Messages/Warnings:: Optional warnings and messages
11202 @value{GDBN} indicates its readiness to read a command by printing a string
11203 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
11204 can change the prompt string with the @code{set prompt} command. For
11205 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
11206 the prompt in one of the @value{GDBN} sessions so that you can always tell
11207 which one you are talking to.
11209 @emph{Note:} @code{set prompt} does not add a space for you after the
11210 prompt you set. This allows you to set a prompt which ends in a space
11211 or a prompt that does not.
11215 @item set prompt @var{newprompt}
11216 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
11218 @kindex show prompt
11220 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
11224 @section Command editing
11226 @cindex command line editing
11228 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
11229 @sc{gnu} library provides consistent behavior for programs which provide a
11230 command line interface to the user. Advantages are @sc{gnu} Emacs-style
11231 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
11232 substitution, and a storage and recall of command history across
11233 debugging sessions.
11235 You may control the behavior of command line editing in @value{GDBN} with the
11236 command @code{set}.
11239 @kindex set editing
11242 @itemx set editing on
11243 Enable command line editing (enabled by default).
11245 @item set editing off
11246 Disable command line editing.
11248 @kindex show editing
11250 Show whether command line editing is enabled.
11254 @section Command history
11256 @value{GDBN} can keep track of the commands you type during your
11257 debugging sessions, so that you can be certain of precisely what
11258 happened. Use these commands to manage the @value{GDBN} command
11262 @cindex history substitution
11263 @cindex history file
11264 @kindex set history filename
11265 @kindex GDBHISTFILE
11266 @item set history filename @var{fname}
11267 Set the name of the @value{GDBN} command history file to @var{fname}.
11268 This is the file where @value{GDBN} reads an initial command history
11269 list, and where it writes the command history from this session when it
11270 exits. You can access this list through history expansion or through
11271 the history command editing characters listed below. This file defaults
11272 to the value of the environment variable @code{GDBHISTFILE}, or to
11273 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
11276 @cindex history save
11277 @kindex set history save
11278 @item set history save
11279 @itemx set history save on
11280 Record command history in a file, whose name may be specified with the
11281 @code{set history filename} command. By default, this option is disabled.
11283 @item set history save off
11284 Stop recording command history in a file.
11286 @cindex history size
11287 @kindex set history size
11288 @item set history size @var{size}
11289 Set the number of commands which @value{GDBN} keeps in its history list.
11290 This defaults to the value of the environment variable
11291 @code{HISTSIZE}, or to 256 if this variable is not set.
11294 @cindex history expansion
11295 History expansion assigns special meaning to the character @kbd{!}.
11296 @ifset have-readline-appendices
11297 @xref{Event Designators}.
11300 Since @kbd{!} is also the logical not operator in C, history expansion
11301 is off by default. If you decide to enable history expansion with the
11302 @code{set history expansion on} command, you may sometimes need to
11303 follow @kbd{!} (when it is used as logical not, in an expression) with
11304 a space or a tab to prevent it from being expanded. The readline
11305 history facilities do not attempt substitution on the strings
11306 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
11308 The commands to control history expansion are:
11311 @kindex set history expansion
11312 @item set history expansion on
11313 @itemx set history expansion
11314 Enable history expansion. History expansion is off by default.
11316 @item set history expansion off
11317 Disable history expansion.
11319 The readline code comes with more complete documentation of
11320 editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs
11321 or @code{vi} may wish to read it.
11322 @ifset have-readline-appendices
11323 @xref{Command Line Editing}.
11327 @kindex show history
11329 @itemx show history filename
11330 @itemx show history save
11331 @itemx show history size
11332 @itemx show history expansion
11333 These commands display the state of the @value{GDBN} history parameters.
11334 @code{show history} by itself displays all four states.
11339 @kindex show commands
11340 @item show commands
11341 Display the last ten commands in the command history.
11343 @item show commands @var{n}
11344 Print ten commands centered on command number @var{n}.
11346 @item show commands +
11347 Print ten commands just after the commands last printed.
11351 @section Screen size
11352 @cindex size of screen
11353 @cindex pauses in output
11355 Certain commands to @value{GDBN} may produce large amounts of
11356 information output to the screen. To help you read all of it,
11357 @value{GDBN} pauses and asks you for input at the end of each page of
11358 output. Type @key{RET} when you want to continue the output, or @kbd{q}
11359 to discard the remaining output. Also, the screen width setting
11360 determines when to wrap lines of output. Depending on what is being
11361 printed, @value{GDBN} tries to break the line at a readable place,
11362 rather than simply letting it overflow onto the following line.
11364 Normally @value{GDBN} knows the size of the screen from the terminal
11365 driver software. For example, on Unix @value{GDBN} uses the termcap data base
11366 together with the value of the @code{TERM} environment variable and the
11367 @code{stty rows} and @code{stty cols} settings. If this is not correct,
11368 you can override it with the @code{set height} and @code{set
11375 @kindex show height
11376 @item set height @var{lpp}
11378 @itemx set width @var{cpl}
11380 These @code{set} commands specify a screen height of @var{lpp} lines and
11381 a screen width of @var{cpl} characters. The associated @code{show}
11382 commands display the current settings.
11384 If you specify a height of zero lines, @value{GDBN} does not pause during
11385 output no matter how long the output is. This is useful if output is to a
11386 file or to an editor buffer.
11388 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
11389 from wrapping its output.
11394 @cindex number representation
11395 @cindex entering numbers
11397 You can always enter numbers in octal, decimal, or hexadecimal in
11398 @value{GDBN} by the usual conventions: octal numbers begin with
11399 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
11400 begin with @samp{0x}. Numbers that begin with none of these are, by
11401 default, entered in base 10; likewise, the default display for
11402 numbers---when no particular format is specified---is base 10. You can
11403 change the default base for both input and output with the @code{set
11407 @kindex set input-radix
11408 @item set input-radix @var{base}
11409 Set the default base for numeric input. Supported choices
11410 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
11411 specified either unambiguously or using the current default radix; for
11421 sets the base to decimal. On the other hand, @samp{set radix 10}
11422 leaves the radix unchanged no matter what it was.
11424 @kindex set output-radix
11425 @item set output-radix @var{base}
11426 Set the default base for numeric display. Supported choices
11427 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
11428 specified either unambiguously or using the current default radix.
11430 @kindex show input-radix
11431 @item show input-radix
11432 Display the current default base for numeric input.
11434 @kindex show output-radix
11435 @item show output-radix
11436 Display the current default base for numeric display.
11439 @node Messages/Warnings
11440 @section Optional warnings and messages
11442 By default, @value{GDBN} is silent about its inner workings. If you are
11443 running on a slow machine, you may want to use the @code{set verbose}
11444 command. This makes @value{GDBN} tell you when it does a lengthy
11445 internal operation, so you will not think it has crashed.
11447 Currently, the messages controlled by @code{set verbose} are those
11448 which announce that the symbol table for a source file is being read;
11449 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
11452 @kindex set verbose
11453 @item set verbose on
11454 Enables @value{GDBN} output of certain informational messages.
11456 @item set verbose off
11457 Disables @value{GDBN} output of certain informational messages.
11459 @kindex show verbose
11461 Displays whether @code{set verbose} is on or off.
11464 By default, if @value{GDBN} encounters bugs in the symbol table of an
11465 object file, it is silent; but if you are debugging a compiler, you may
11466 find this information useful (@pxref{Symbol Errors, ,Errors reading
11471 @kindex set complaints
11472 @item set complaints @var{limit}
11473 Permits @value{GDBN} to output @var{limit} complaints about each type of
11474 unusual symbols before becoming silent about the problem. Set
11475 @var{limit} to zero to suppress all complaints; set it to a large number
11476 to prevent complaints from being suppressed.
11478 @kindex show complaints
11479 @item show complaints
11480 Displays how many symbol complaints @value{GDBN} is permitted to produce.
11484 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
11485 lot of stupid questions to confirm certain commands. For example, if
11486 you try to run a program which is already running:
11490 The program being debugged has been started already.
11491 Start it from the beginning? (y or n)
11494 If you are willing to unflinchingly face the consequences of your own
11495 commands, you can disable this ``feature'':
11499 @kindex set confirm
11501 @cindex confirmation
11502 @cindex stupid questions
11503 @item set confirm off
11504 Disables confirmation requests.
11506 @item set confirm on
11507 Enables confirmation requests (the default).
11509 @kindex show confirm
11511 Displays state of confirmation requests.
11516 @chapter Canned Sequences of Commands
11518 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
11519 command lists}), @value{GDBN} provides two ways to store sequences of
11520 commands for execution as a unit: user-defined commands and command
11524 * Define:: User-defined commands
11525 * Hooks:: User-defined command hooks
11526 * Command Files:: Command files
11527 * Output:: Commands for controlled output
11531 @section User-defined commands
11533 @cindex user-defined command
11534 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
11535 which you assign a new name as a command. This is done with the
11536 @code{define} command. User commands may accept up to 10 arguments
11537 separated by whitespace. Arguments are accessed within the user command
11538 via @var{$arg0@dots{}$arg9}. A trivial example:
11542 print $arg0 + $arg1 + $arg2
11546 To execute the command use:
11553 This defines the command @code{adder}, which prints the sum of
11554 its three arguments. Note the arguments are text substitutions, so they may
11555 reference variables, use complex expressions, or even perform inferior
11561 @item define @var{commandname}
11562 Define a command named @var{commandname}. If there is already a command
11563 by that name, you are asked to confirm that you want to redefine it.
11565 The definition of the command is made up of other @value{GDBN} command lines,
11566 which are given following the @code{define} command. The end of these
11567 commands is marked by a line containing @code{end}.
11572 Takes a single argument, which is an expression to evaluate.
11573 It is followed by a series of commands that are executed
11574 only if the expression is true (nonzero).
11575 There can then optionally be a line @code{else}, followed
11576 by a series of commands that are only executed if the expression
11577 was false. The end of the list is marked by a line containing @code{end}.
11581 The syntax is similar to @code{if}: the command takes a single argument,
11582 which is an expression to evaluate, and must be followed by the commands to
11583 execute, one per line, terminated by an @code{end}.
11584 The commands are executed repeatedly as long as the expression
11588 @item document @var{commandname}
11589 Document the user-defined command @var{commandname}, so that it can be
11590 accessed by @code{help}. The command @var{commandname} must already be
11591 defined. This command reads lines of documentation just as @code{define}
11592 reads the lines of the command definition, ending with @code{end}.
11593 After the @code{document} command is finished, @code{help} on command
11594 @var{commandname} displays the documentation you have written.
11596 You may use the @code{document} command again to change the
11597 documentation of a command. Redefining the command with @code{define}
11598 does not change the documentation.
11600 @kindex help user-defined
11601 @item help user-defined
11602 List all user-defined commands, with the first line of the documentation
11607 @itemx show user @var{commandname}
11608 Display the @value{GDBN} commands used to define @var{commandname} (but
11609 not its documentation). If no @var{commandname} is given, display the
11610 definitions for all user-defined commands.
11614 When user-defined commands are executed, the
11615 commands of the definition are not printed. An error in any command
11616 stops execution of the user-defined command.
11618 If used interactively, commands that would ask for confirmation proceed
11619 without asking when used inside a user-defined command. Many @value{GDBN}
11620 commands that normally print messages to say what they are doing omit the
11621 messages when used in a user-defined command.
11624 @section User-defined command hooks
11625 @cindex command hooks
11626 @cindex hooks, for commands
11628 You may define @emph{hooks}, which are a special kind of user-defined
11629 command. Whenever you run the command @samp{foo}, if the user-defined
11630 command @samp{hook-foo} exists, it is executed (with no arguments)
11631 before that command.
11633 @kindex stop@r{, a pseudo-command}
11634 In addition, a pseudo-command, @samp{stop} exists. Defining
11635 (@samp{hook-stop}) makes the associated commands execute every time
11636 execution stops in your program: before breakpoint commands are run,
11637 displays are printed, or the stack frame is printed.
11639 For example, to ignore @code{SIGALRM} signals while
11640 single-stepping, but treat them normally during normal execution,
11645 handle SIGALRM nopass
11649 handle SIGALRM pass
11652 define hook-continue
11653 handle SIGLARM pass
11657 You can define a hook for any single-word command in @value{GDBN}, but
11658 not for command aliases; you should define a hook for the basic command
11659 name, e.g. @code{backtrace} rather than @code{bt}.
11660 @c FIXME! So how does Joe User discover whether a command is an alias
11662 If an error occurs during the execution of your hook, execution of
11663 @value{GDBN} commands stops and @value{GDBN} issues a prompt
11664 (before the command that you actually typed had a chance to run).
11666 If you try to define a hook which does not match any known command, you
11667 get a warning from the @code{define} command.
11669 @node Command Files
11670 @section Command files
11672 @cindex command files
11673 A command file for @value{GDBN} is a file of lines that are @value{GDBN}
11674 commands. Comments (lines starting with @kbd{#}) may also be included.
11675 An empty line in a command file does nothing; it does not mean to repeat
11676 the last command, as it would from the terminal.
11679 @cindex @file{.gdbinit}
11680 @cindex @file{gdb.ini}
11681 When you start @value{GDBN}, it automatically executes commands from its
11682 @dfn{init files}. These are files named @file{.gdbinit} on Unix, or
11683 @file{gdb.ini} on DOS/Windows. @value{GDBN} reads the init file (if
11684 any) in your home directory@footnote{On DOS/Windows systems, the home
11685 directory is the one pointed to by the @code{HOME} environment
11686 variable.}, then processes command line options and operands, and then
11687 reads the init file (if any) in the current working directory. This is
11688 so the init file in your home directory can set options (such as
11689 @code{set complaints}) which affect the processing of the command line
11690 options and operands. The init files are not executed if you use the
11691 @samp{-nx} option; @pxref{Mode Options, ,Choosing modes}.
11693 @cindex init file name
11694 On some configurations of @value{GDBN}, the init file is known by a
11695 different name (these are typically environments where a specialized
11696 form of @value{GDBN} may need to coexist with other forms, hence a
11697 different name for the specialized version's init file). These are the
11698 environments with special init file names:
11703 VxWorks (Wind River Systems real-time OS): @samp{.vxgdbinit}
11705 @kindex .os68gdbinit
11707 OS68K (Enea Data Systems real-time OS): @samp{.os68gdbinit}
11711 ES-1800 (Ericsson Telecom AB M68000 emulator): @samp{.esgdbinit}
11714 You can also request the execution of a command file with the
11715 @code{source} command:
11719 @item source @var{filename}
11720 Execute the command file @var{filename}.
11723 The lines in a command file are executed sequentially. They are not
11724 printed as they are executed. An error in any command terminates execution
11725 of the command file.
11727 Commands that would ask for confirmation if used interactively proceed
11728 without asking when used in a command file. Many @value{GDBN} commands that
11729 normally print messages to say what they are doing omit the messages
11730 when called from command files.
11733 @section Commands for controlled output
11735 During the execution of a command file or a user-defined command, normal
11736 @value{GDBN} output is suppressed; the only output that appears is what is
11737 explicitly printed by the commands in the definition. This section
11738 describes three commands useful for generating exactly the output you
11743 @item echo @var{text}
11744 @c I do not consider backslash-space a standard C escape sequence
11745 @c because it is not in ANSI.
11746 Print @var{text}. Nonprinting characters can be included in
11747 @var{text} using C escape sequences, such as @samp{\n} to print a
11748 newline. @strong{No newline is printed unless you specify one.}
11749 In addition to the standard C escape sequences, a backslash followed
11750 by a space stands for a space. This is useful for displaying a
11751 string with spaces at the beginning or the end, since leading and
11752 trailing spaces are otherwise trimmed from all arguments.
11753 To print @samp{@w{ }and foo =@w{ }}, use the command
11754 @samp{echo \@w{ }and foo = \@w{ }}.
11756 A backslash at the end of @var{text} can be used, as in C, to continue
11757 the command onto subsequent lines. For example,
11760 echo This is some text\n\
11761 which is continued\n\
11762 onto several lines.\n
11765 produces the same output as
11768 echo This is some text\n
11769 echo which is continued\n
11770 echo onto several lines.\n
11774 @item output @var{expression}
11775 Print the value of @var{expression} and nothing but that value: no
11776 newlines, no @samp{$@var{nn} = }. The value is not entered in the
11777 value history either. @xref{Expressions, ,Expressions}, for more information
11780 @item output/@var{fmt} @var{expression}
11781 Print the value of @var{expression} in format @var{fmt}. You can use
11782 the same formats as for @code{print}. @xref{Output Formats,,Output
11783 formats}, for more information.
11786 @item printf @var{string}, @var{expressions}@dots{}
11787 Print the values of the @var{expressions} under the control of
11788 @var{string}. The @var{expressions} are separated by commas and may be
11789 either numbers or pointers. Their values are printed as specified by
11790 @var{string}, exactly as if your program were to execute the C
11792 @c FIXME: the above implies that at least all ANSI C formats are
11793 @c supported, but it isn't true: %E and %G don't work (or so it seems).
11794 @c Either this is a bug, or the manual should document what formats are
11798 printf (@var{string}, @var{expressions}@dots{});
11801 For example, you can print two values in hex like this:
11804 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
11807 The only backslash-escape sequences that you can use in the format
11808 string are the simple ones that consist of backslash followed by a
11813 @chapter Using @value{GDBN} under @sc{gnu} Emacs
11816 @cindex @sc{gnu} Emacs
11817 A special interface allows you to use @sc{gnu} Emacs to view (and
11818 edit) the source files for the program you are debugging with
11821 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
11822 executable file you want to debug as an argument. This command starts
11823 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
11824 created Emacs buffer.
11825 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
11827 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
11832 All ``terminal'' input and output goes through the Emacs buffer.
11835 This applies both to @value{GDBN} commands and their output, and to the input
11836 and output done by the program you are debugging.
11838 This is useful because it means that you can copy the text of previous
11839 commands and input them again; you can even use parts of the output
11842 All the facilities of Emacs' Shell mode are available for interacting
11843 with your program. In particular, you can send signals the usual
11844 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
11849 @value{GDBN} displays source code through Emacs.
11852 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
11853 source file for that frame and puts an arrow (@samp{=>}) at the
11854 left margin of the current line. Emacs uses a separate buffer for
11855 source display, and splits the screen to show both your @value{GDBN} session
11858 Explicit @value{GDBN} @code{list} or search commands still produce output as
11859 usual, but you probably have no reason to use them from Emacs.
11862 @emph{Warning:} If the directory where your program resides is not your
11863 current directory, it can be easy to confuse Emacs about the location of
11864 the source files, in which case the auxiliary display buffer does not
11865 appear to show your source. @value{GDBN} can find programs by searching your
11866 environment's @code{PATH} variable, so the @value{GDBN} input and output
11867 session proceeds normally; but Emacs does not get enough information
11868 back from @value{GDBN} to locate the source files in this situation. To
11869 avoid this problem, either start @value{GDBN} mode from the directory where
11870 your program resides, or specify an absolute file name when prompted for the
11871 @kbd{M-x gdb} argument.
11873 A similar confusion can result if you use the @value{GDBN} @code{file} command to
11874 switch to debugging a program in some other location, from an existing
11875 @value{GDBN} buffer in Emacs.
11878 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
11879 you need to call @value{GDBN} by a different name (for example, if you keep
11880 several configurations around, with different names) you can set the
11881 Emacs variable @code{gdb-command-name}; for example,
11884 (setq gdb-command-name "mygdb")
11888 (preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or
11889 in your @file{.emacs} file) makes Emacs call the program named
11890 ``@code{mygdb}'' instead.
11892 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
11893 addition to the standard Shell mode commands:
11897 Describe the features of Emacs' @value{GDBN} Mode.
11900 Execute to another source line, like the @value{GDBN} @code{step} command; also
11901 update the display window to show the current file and location.
11904 Execute to next source line in this function, skipping all function
11905 calls, like the @value{GDBN} @code{next} command. Then update the display window
11906 to show the current file and location.
11909 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
11910 display window accordingly.
11912 @item M-x gdb-nexti
11913 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
11914 display window accordingly.
11917 Execute until exit from the selected stack frame, like the @value{GDBN}
11918 @code{finish} command.
11921 Continue execution of your program, like the @value{GDBN} @code{continue}
11924 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
11927 Go up the number of frames indicated by the numeric argument
11928 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
11929 like the @value{GDBN} @code{up} command.
11931 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
11934 Go down the number of frames indicated by the numeric argument, like the
11935 @value{GDBN} @code{down} command.
11937 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
11940 Read the number where the cursor is positioned, and insert it at the end
11941 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
11942 around an address that was displayed earlier, type @kbd{disassemble};
11943 then move the cursor to the address display, and pick up the
11944 argument for @code{disassemble} by typing @kbd{C-x &}.
11946 You can customize this further by defining elements of the list
11947 @code{gdb-print-command}; once it is defined, you can format or
11948 otherwise process numbers picked up by @kbd{C-x &} before they are
11949 inserted. A numeric argument to @kbd{C-x &} indicates that you
11950 wish special formatting, and also acts as an index to pick an element of the
11951 list. If the list element is a string, the number to be inserted is
11952 formatted using the Emacs function @code{format}; otherwise the number
11953 is passed as an argument to the corresponding list element.
11956 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
11957 tells @value{GDBN} to set a breakpoint on the source line point is on.
11959 If you accidentally delete the source-display buffer, an easy way to get
11960 it back is to type the command @code{f} in the @value{GDBN} buffer, to
11961 request a frame display; when you run under Emacs, this recreates
11962 the source buffer if necessary to show you the context of the current
11965 The source files displayed in Emacs are in ordinary Emacs buffers
11966 which are visiting the source files in the usual way. You can edit
11967 the files with these buffers if you wish; but keep in mind that @value{GDBN}
11968 communicates with Emacs in terms of line numbers. If you add or
11969 delete lines from the text, the line numbers that @value{GDBN} knows cease
11970 to correspond properly with the code.
11972 @c The following dropped because Epoch is nonstandard. Reactivate
11973 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
11975 @kindex Emacs Epoch environment
11979 Version 18 of @sc{gnu} Emacs has a built-in window system
11980 called the @code{epoch}
11981 environment. Users of this environment can use a new command,
11982 @code{inspect} which performs identically to @code{print} except that
11983 each value is printed in its own window.
11986 @include annotate.texi
11989 @chapter Reporting Bugs in @value{GDBN}
11990 @cindex bugs in @value{GDBN}
11991 @cindex reporting bugs in @value{GDBN}
11993 Your bug reports play an essential role in making @value{GDBN} reliable.
11995 Reporting a bug may help you by bringing a solution to your problem, or it
11996 may not. But in any case the principal function of a bug report is to help
11997 the entire community by making the next version of @value{GDBN} work better. Bug
11998 reports are your contribution to the maintenance of @value{GDBN}.
12000 In order for a bug report to serve its purpose, you must include the
12001 information that enables us to fix the bug.
12004 * Bug Criteria:: Have you found a bug?
12005 * Bug Reporting:: How to report bugs
12009 @section Have you found a bug?
12010 @cindex bug criteria
12012 If you are not sure whether you have found a bug, here are some guidelines:
12015 @cindex fatal signal
12016 @cindex debugger crash
12017 @cindex crash of debugger
12019 If the debugger gets a fatal signal, for any input whatever, that is a
12020 @value{GDBN} bug. Reliable debuggers never crash.
12022 @cindex error on valid input
12024 If @value{GDBN} produces an error message for valid input, that is a
12025 bug. (Note that if you're cross debugging, the problem may also be
12026 somewhere in the connection to the target.)
12028 @cindex invalid input
12030 If @value{GDBN} does not produce an error message for invalid input,
12031 that is a bug. However, you should note that your idea of
12032 ``invalid input'' might be our idea of ``an extension'' or ``support
12033 for traditional practice''.
12036 If you are an experienced user of debugging tools, your suggestions
12037 for improvement of @value{GDBN} are welcome in any case.
12040 @node Bug Reporting
12041 @section How to report bugs
12042 @cindex bug reports
12043 @cindex @value{GDBN} bugs, reporting
12045 A number of companies and individuals offer support for @sc{gnu} products.
12046 If you obtained @value{GDBN} from a support organization, we recommend you
12047 contact that organization first.
12049 You can find contact information for many support companies and
12050 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
12052 @c should add a web page ref...
12054 In any event, we also recommend that you send bug reports for
12055 @value{GDBN} to this addresses:
12061 @strong{Do not send bug reports to @samp{info-gdb}, or to
12062 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
12063 not want to receive bug reports. Those that do have arranged to receive
12066 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
12067 serves as a repeater. The mailing list and the newsgroup carry exactly
12068 the same messages. Often people think of posting bug reports to the
12069 newsgroup instead of mailing them. This appears to work, but it has one
12070 problem which can be crucial: a newsgroup posting often lacks a mail
12071 path back to the sender. Thus, if we need to ask for more information,
12072 we may be unable to reach you. For this reason, it is better to send
12073 bug reports to the mailing list.
12075 As a last resort, send bug reports on paper to:
12078 @sc{gnu} Debugger Bugs
12079 Free Software Foundation Inc.
12080 59 Temple Place - Suite 330
12081 Boston, MA 02111-1307
12085 The fundamental principle of reporting bugs usefully is this:
12086 @strong{report all the facts}. If you are not sure whether to state a
12087 fact or leave it out, state it!
12089 Often people omit facts because they think they know what causes the
12090 problem and assume that some details do not matter. Thus, you might
12091 assume that the name of the variable you use in an example does not matter.
12092 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
12093 stray memory reference which happens to fetch from the location where that
12094 name is stored in memory; perhaps, if the name were different, the contents
12095 of that location would fool the debugger into doing the right thing despite
12096 the bug. Play it safe and give a specific, complete example. That is the
12097 easiest thing for you to do, and the most helpful.
12099 Keep in mind that the purpose of a bug report is to enable us to fix the
12100 bug. It may be that the bug has been reported previously, but neither
12101 you nor we can know that unless your bug report is complete and
12104 Sometimes people give a few sketchy facts and ask, ``Does this ring a
12105 bell?'' Those bug reports are useless, and we urge everyone to
12106 @emph{refuse to respond to them} except to chide the sender to report
12109 To enable us to fix the bug, you should include all these things:
12113 The version of @value{GDBN}. @value{GDBN} announces it if you start
12114 with no arguments; you can also print it at any time using @code{show
12117 Without this, we will not know whether there is any point in looking for
12118 the bug in the current version of @value{GDBN}.
12121 The type of machine you are using, and the operating system name and
12125 What compiler (and its version) was used to compile @value{GDBN}---e.g.
12126 ``@value{GCC}--2.8.1''.
12129 What compiler (and its version) was used to compile the program you are
12130 debugging---e.g. ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
12131 C Compiler''. For GCC, you can say @code{gcc --version} to get this
12132 information; for other compilers, see the documentation for those
12136 The command arguments you gave the compiler to compile your example and
12137 observe the bug. For example, did you use @samp{-O}? To guarantee
12138 you will not omit something important, list them all. A copy of the
12139 Makefile (or the output from make) is sufficient.
12141 If we were to try to guess the arguments, we would probably guess wrong
12142 and then we might not encounter the bug.
12145 A complete input script, and all necessary source files, that will
12149 A description of what behavior you observe that you believe is
12150 incorrect. For example, ``It gets a fatal signal.''
12152 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
12153 will certainly notice it. But if the bug is incorrect output, we might
12154 not notice unless it is glaringly wrong. You might as well not give us
12155 a chance to make a mistake.
12157 Even if the problem you experience is a fatal signal, you should still
12158 say so explicitly. Suppose something strange is going on, such as, your
12159 copy of @value{GDBN} is out of synch, or you have encountered a bug in
12160 the C library on your system. (This has happened!) Your copy might
12161 crash and ours would not. If you told us to expect a crash, then when
12162 ours fails to crash, we would know that the bug was not happening for
12163 us. If you had not told us to expect a crash, then we would not be able
12164 to draw any conclusion from our observations.
12167 If you wish to suggest changes to the @value{GDBN} source, send us context
12168 diffs. If you even discuss something in the @value{GDBN} source, refer to
12169 it by context, not by line number.
12171 The line numbers in our development sources will not match those in your
12172 sources. Your line numbers would convey no useful information to us.
12176 Here are some things that are not necessary:
12180 A description of the envelope of the bug.
12182 Often people who encounter a bug spend a lot of time investigating
12183 which changes to the input file will make the bug go away and which
12184 changes will not affect it.
12186 This is often time consuming and not very useful, because the way we
12187 will find the bug is by running a single example under the debugger
12188 with breakpoints, not by pure deduction from a series of examples.
12189 We recommend that you save your time for something else.
12191 Of course, if you can find a simpler example to report @emph{instead}
12192 of the original one, that is a convenience for us. Errors in the
12193 output will be easier to spot, running under the debugger will take
12194 less time, and so on.
12196 However, simplification is not vital; if you do not want to do this,
12197 report the bug anyway and send us the entire test case you used.
12200 A patch for the bug.
12202 A patch for the bug does help us if it is a good one. But do not omit
12203 the necessary information, such as the test case, on the assumption that
12204 a patch is all we need. We might see problems with your patch and decide
12205 to fix the problem another way, or we might not understand it at all.
12207 Sometimes with a program as complicated as @value{GDBN} it is very hard to
12208 construct an example that will make the program follow a certain path
12209 through the code. If you do not send us the example, we will not be able
12210 to construct one, so we will not be able to verify that the bug is fixed.
12212 And if we cannot understand what bug you are trying to fix, or why your
12213 patch should be an improvement, we will not install it. A test case will
12214 help us to understand.
12217 A guess about what the bug is or what it depends on.
12219 Such guesses are usually wrong. Even we cannot guess right about such
12220 things without first using the debugger to find the facts.
12223 @c The readline documentation is distributed with the readline code
12224 @c and consists of the two following files:
12226 @c inc-hist.texinfo
12227 @c Use -I with makeinfo to point to the appropriate directory,
12228 @c environment var TEXINPUTS with TeX.
12229 @include rluser.texinfo
12230 @include inc-hist.texinfo
12233 @node Formatting Documentation
12234 @appendix Formatting Documentation
12236 @cindex @value{GDBN} reference card
12237 @cindex reference card
12238 The @value{GDBN} 4 release includes an already-formatted reference card, ready
12239 for printing with PostScript or Ghostscript, in the @file{gdb}
12240 subdirectory of the main source directory@footnote{In
12241 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
12242 release.}. If you can use PostScript or Ghostscript with your printer,
12243 you can print the reference card immediately with @file{refcard.ps}.
12245 The release also includes the source for the reference card. You
12246 can format it, using @TeX{}, by typing:
12252 The @value{GDBN} reference card is designed to print in @dfn{landscape}
12253 mode on US ``letter'' size paper;
12254 that is, on a sheet 11 inches wide by 8.5 inches
12255 high. You will need to specify this form of printing as an option to
12256 your @sc{dvi} output program.
12258 @cindex documentation
12260 All the documentation for @value{GDBN} comes as part of the machine-readable
12261 distribution. The documentation is written in Texinfo format, which is
12262 a documentation system that uses a single source file to produce both
12263 on-line information and a printed manual. You can use one of the Info
12264 formatting commands to create the on-line version of the documentation
12265 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
12267 @value{GDBN} includes an already formatted copy of the on-line Info
12268 version of this manual in the @file{gdb} subdirectory. The main Info
12269 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
12270 subordinate files matching @samp{gdb.info*} in the same directory. If
12271 necessary, you can print out these files, or read them with any editor;
12272 but they are easier to read using the @code{info} subsystem in @sc{gnu}
12273 Emacs or the standalone @code{info} program, available as part of the
12274 @sc{gnu} Texinfo distribution.
12276 If you want to format these Info files yourself, you need one of the
12277 Info formatting programs, such as @code{texinfo-format-buffer} or
12280 If you have @code{makeinfo} installed, and are in the top level
12281 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
12282 version @value{GDBVN}), you can make the Info file by typing:
12289 If you want to typeset and print copies of this manual, you need @TeX{},
12290 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
12291 Texinfo definitions file.
12293 @TeX{} is a typesetting program; it does not print files directly, but
12294 produces output files called @sc{dvi} files. To print a typeset
12295 document, you need a program to print @sc{dvi} files. If your system
12296 has @TeX{} installed, chances are it has such a program. The precise
12297 command to use depends on your system; @kbd{lpr -d} is common; another
12298 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
12299 require a file name without any extension or a @samp{.dvi} extension.
12301 @TeX{} also requires a macro definitions file called
12302 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
12303 written in Texinfo format. On its own, @TeX{} cannot either read or
12304 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
12305 and is located in the @file{gdb-@var{version-number}/texinfo}
12308 If you have @TeX{} and a @sc{dvi} printer program installed, you can
12309 typeset and print this manual. First switch to the the @file{gdb}
12310 subdirectory of the main source directory (for example, to
12311 @file{gdb-@value{GDBVN}/gdb}) and type:
12317 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
12319 @node Installing GDB
12320 @appendix Installing @value{GDBN}
12321 @cindex configuring @value{GDBN}
12322 @cindex installation
12324 @value{GDBN} comes with a @code{configure} script that automates the process
12325 of preparing @value{GDBN} for installation; you can then use @code{make} to
12326 build the @code{gdb} program.
12328 @c irrelevant in info file; it's as current as the code it lives with.
12329 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
12330 look at the @file{README} file in the sources; we may have improved the
12331 installation procedures since publishing this manual.}
12334 The @value{GDBN} distribution includes all the source code you need for
12335 @value{GDBN} in a single directory, whose name is usually composed by
12336 appending the version number to @samp{gdb}.
12338 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
12339 @file{gdb-@value{GDBVN}} directory. That directory contains:
12342 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
12343 script for configuring @value{GDBN} and all its supporting libraries
12345 @item gdb-@value{GDBVN}/gdb
12346 the source specific to @value{GDBN} itself
12348 @item gdb-@value{GDBVN}/bfd
12349 source for the Binary File Descriptor library
12351 @item gdb-@value{GDBVN}/include
12352 @sc{gnu} include files
12354 @item gdb-@value{GDBVN}/libiberty
12355 source for the @samp{-liberty} free software library
12357 @item gdb-@value{GDBVN}/opcodes
12358 source for the library of opcode tables and disassemblers
12360 @item gdb-@value{GDBVN}/readline
12361 source for the @sc{gnu} command-line interface
12363 @item gdb-@value{GDBVN}/glob
12364 source for the @sc{gnu} filename pattern-matching subroutine
12366 @item gdb-@value{GDBVN}/mmalloc
12367 source for the @sc{gnu} memory-mapped malloc package
12370 The simplest way to configure and build @value{GDBN} is to run @code{configure}
12371 from the @file{gdb-@var{version-number}} source directory, which in
12372 this example is the @file{gdb-@value{GDBVN}} directory.
12374 First switch to the @file{gdb-@var{version-number}} source directory
12375 if you are not already in it; then run @code{configure}. Pass the
12376 identifier for the platform on which @value{GDBN} will run as an
12382 cd gdb-@value{GDBVN}
12383 ./configure @var{host}
12388 where @var{host} is an identifier such as @samp{sun4} or
12389 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
12390 (You can often leave off @var{host}; @code{configure} tries to guess the
12391 correct value by examining your system.)
12393 Running @samp{configure @var{host}} and then running @code{make} builds the
12394 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
12395 libraries, then @code{gdb} itself. The configured source files, and the
12396 binaries, are left in the corresponding source directories.
12399 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
12400 system does not recognize this automatically when you run a different
12401 shell, you may need to run @code{sh} on it explicitly:
12404 sh configure @var{host}
12407 If you run @code{configure} from a directory that contains source
12408 directories for multiple libraries or programs, such as the
12409 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
12410 creates configuration files for every directory level underneath (unless
12411 you tell it not to, with the @samp{--norecursion} option).
12413 You can run the @code{configure} script from any of the
12414 subordinate directories in the @value{GDBN} distribution if you only want to
12415 configure that subdirectory, but be sure to specify a path to it.
12417 For example, with version @value{GDBVN}, type the following to configure only
12418 the @code{bfd} subdirectory:
12422 cd gdb-@value{GDBVN}/bfd
12423 ../configure @var{host}
12427 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
12428 However, you should make sure that the shell on your path (named by
12429 the @samp{SHELL} environment variable) is publicly readable. Remember
12430 that @value{GDBN} uses the shell to start your program---some systems refuse to
12431 let @value{GDBN} debug child processes whose programs are not readable.
12434 * Separate Objdir:: Compiling @value{GDBN} in another directory
12435 * Config Names:: Specifying names for hosts and targets
12436 * Configure Options:: Summary of options for configure
12439 @node Separate Objdir
12440 @section Compiling @value{GDBN} in another directory
12442 If you want to run @value{GDBN} versions for several host or target machines,
12443 you need a different @code{gdb} compiled for each combination of
12444 host and target. @code{configure} is designed to make this easy by
12445 allowing you to generate each configuration in a separate subdirectory,
12446 rather than in the source directory. If your @code{make} program
12447 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
12448 @code{make} in each of these directories builds the @code{gdb}
12449 program specified there.
12451 To build @code{gdb} in a separate directory, run @code{configure}
12452 with the @samp{--srcdir} option to specify where to find the source.
12453 (You also need to specify a path to find @code{configure}
12454 itself from your working directory. If the path to @code{configure}
12455 would be the same as the argument to @samp{--srcdir}, you can leave out
12456 the @samp{--srcdir} option; it is assumed.)
12458 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
12459 separate directory for a Sun 4 like this:
12463 cd gdb-@value{GDBVN}
12466 ../gdb-@value{GDBVN}/configure sun4
12471 When @code{configure} builds a configuration using a remote source
12472 directory, it creates a tree for the binaries with the same structure
12473 (and using the same names) as the tree under the source directory. In
12474 the example, you'd find the Sun 4 library @file{libiberty.a} in the
12475 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
12476 @file{gdb-sun4/gdb}.
12478 One popular reason to build several @value{GDBN} configurations in separate
12479 directories is to configure @value{GDBN} for cross-compiling (where
12480 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
12481 programs that run on another machine---the @dfn{target}).
12482 You specify a cross-debugging target by
12483 giving the @samp{--target=@var{target}} option to @code{configure}.
12485 When you run @code{make} to build a program or library, you must run
12486 it in a configured directory---whatever directory you were in when you
12487 called @code{configure} (or one of its subdirectories).
12489 The @code{Makefile} that @code{configure} generates in each source
12490 directory also runs recursively. If you type @code{make} in a source
12491 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
12492 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
12493 will build all the required libraries, and then build GDB.
12495 When you have multiple hosts or targets configured in separate
12496 directories, you can run @code{make} on them in parallel (for example,
12497 if they are NFS-mounted on each of the hosts); they will not interfere
12501 @section Specifying names for hosts and targets
12503 The specifications used for hosts and targets in the @code{configure}
12504 script are based on a three-part naming scheme, but some short predefined
12505 aliases are also supported. The full naming scheme encodes three pieces
12506 of information in the following pattern:
12509 @var{architecture}-@var{vendor}-@var{os}
12512 For example, you can use the alias @code{sun4} as a @var{host} argument,
12513 or as the value for @var{target} in a @code{--target=@var{target}}
12514 option. The equivalent full name is @samp{sparc-sun-sunos4}.
12516 The @code{configure} script accompanying @value{GDBN} does not provide
12517 any query facility to list all supported host and target names or
12518 aliases. @code{configure} calls the Bourne shell script
12519 @code{config.sub} to map abbreviations to full names; you can read the
12520 script, if you wish, or you can use it to test your guesses on
12521 abbreviations---for example:
12524 % sh config.sub i386-linux
12526 % sh config.sub alpha-linux
12527 alpha-unknown-linux-gnu
12528 % sh config.sub hp9k700
12530 % sh config.sub sun4
12531 sparc-sun-sunos4.1.1
12532 % sh config.sub sun3
12533 m68k-sun-sunos4.1.1
12534 % sh config.sub i986v
12535 Invalid configuration `i986v': machine `i986v' not recognized
12539 @code{config.sub} is also distributed in the @value{GDBN} source
12540 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
12542 @node Configure Options
12543 @section @code{configure} options
12545 Here is a summary of the @code{configure} options and arguments that
12546 are most often useful for building @value{GDBN}. @code{configure} also has
12547 several other options not listed here. @inforef{What Configure
12548 Does,,configure.info}, for a full explanation of @code{configure}.
12551 configure @r{[}--help@r{]}
12552 @r{[}--prefix=@var{dir}@r{]}
12553 @r{[}--exec-prefix=@var{dir}@r{]}
12554 @r{[}--srcdir=@var{dirname}@r{]}
12555 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
12556 @r{[}--target=@var{target}@r{]}
12561 You may introduce options with a single @samp{-} rather than
12562 @samp{--} if you prefer; but you may abbreviate option names if you use
12567 Display a quick summary of how to invoke @code{configure}.
12569 @item --prefix=@var{dir}
12570 Configure the source to install programs and files under directory
12573 @item --exec-prefix=@var{dir}
12574 Configure the source to install programs under directory
12577 @c avoid splitting the warning from the explanation:
12579 @item --srcdir=@var{dirname}
12580 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
12581 @code{make} that implements the @code{VPATH} feature.}@*
12582 Use this option to make configurations in directories separate from the
12583 @value{GDBN} source directories. Among other things, you can use this to
12584 build (or maintain) several configurations simultaneously, in separate
12585 directories. @code{configure} writes configuration specific files in
12586 the current directory, but arranges for them to use the source in the
12587 directory @var{dirname}. @code{configure} creates directories under
12588 the working directory in parallel to the source directories below
12591 @item --norecursion
12592 Configure only the directory level where @code{configure} is executed; do not
12593 propagate configuration to subdirectories.
12595 @item --target=@var{target}
12596 Configure @value{GDBN} for cross-debugging programs running on the specified
12597 @var{target}. Without this option, @value{GDBN} is configured to debug
12598 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
12600 There is no convenient way to generate a list of all available targets.
12602 @item @var{host} @dots{}
12603 Configure @value{GDBN} to run on the specified @var{host}.
12605 There is no convenient way to generate a list of all available hosts.
12608 There are many other options available as well, but they are generally
12609 needed for special purposes only.
12617 % I think something like @colophon should be in texinfo. In the
12619 \long\def\colophon{\hbox to0pt{}\vfill
12620 \centerline{The body of this manual is set in}
12621 \centerline{\fontname\tenrm,}
12622 \centerline{with headings in {\bf\fontname\tenbf}}
12623 \centerline{and examples in {\tt\fontname\tentt}.}
12624 \centerline{{\it\fontname\tenit\/},}
12625 \centerline{{\bf\fontname\tenbf}, and}
12626 \centerline{{\sl\fontname\tensl\/}}
12627 \centerline{are used for emphasis.}\vfill}
12629 % Blame: doc@cygnus.com, 1991.