1 \input texinfo @c -*-texinfo-*-
2 @c Copyright 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,
3 @c 1999, 2000, 2001, 2002, 2003
4 @c Free Software Foundation, Inc.
7 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
8 @c of @set vars. However, you can override filename with makeinfo -o.
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
25 @c readline appendices use @vindex, @findex and @ftable,
26 @c annotate.texi and gdbmi use @findex.
30 @c !!set GDB manual's edition---not the same as GDB version!
33 @c !!set GDB manual's revision date
36 @c !!set GDB edit command default editor
39 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
41 @c This is a dir.info fragment to support semi-automated addition of
42 @c manuals to an info tree.
43 @dircategory Programming & development tools.
45 * Gdb: (gdb). The @sc{gnu} debugger.
49 This file documents the @sc{gnu} debugger @value{GDBN}.
52 This is the @value{EDITION} Edition, @value{DATE},
53 of @cite{Debugging with @value{GDBN}: the @sc{gnu} Source-Level Debugger}
54 for @value{GDBN} Version @value{GDBVN}.
56 Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,@*
57 1999, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
59 Permission is granted to copy, distribute and/or modify this document
60 under the terms of the GNU Free Documentation License, Version 1.1 or
61 any later version published by the Free Software Foundation; with the
62 Invariant Sections being ``Free Software'' and ``Free Software Needs
63 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
64 and with the Back-Cover Texts as in (a) below.
66 (a) The Free Software Foundation's Back-Cover Text is: ``You have
67 freedom to copy and modify this GNU Manual, like GNU software. Copies
68 published by the Free Software Foundation raise funds for GNU
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 Stallman, Roland Pesch, Stan Shebs, et al.
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 @vskip 0pt plus 1filll
89 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995,
90 1996, 1998, 1999, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
92 Published by the Free Software Foundation @*
93 59 Temple Place - Suite 330, @*
94 Boston, MA 02111-1307 USA @*
97 Permission is granted to copy, distribute and/or modify this document
98 under the terms of the GNU Free Documentation License, Version 1.1 or
99 any later version published by the Free Software Foundation; with the
100 Invariant Sections being ``Free Software'' and ``Free Software Needs
101 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
102 and with the Back-Cover Texts as in (a) below.
104 (a) The Free Software Foundation's Back-Cover Text is: ``You have
105 freedom to copy and modify this GNU Manual, like GNU software. Copies
106 published by the Free Software Foundation raise funds for GNU
112 @node Top, Summary, (dir), (dir)
114 @top Debugging with @value{GDBN}
116 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
118 This is the @value{EDITION} Edition, @value{DATE}, for @value{GDBN} Version
121 Copyright (C) 1988-2003 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
134 * Macros:: Preprocessor Macros
135 * Tracepoints:: Debugging remote targets non-intrusively
136 * Overlays:: Debugging programs that use overlays
138 * Languages:: Using @value{GDBN} with different languages
140 * Symbols:: Examining the symbol table
141 * Altering:: Altering execution
142 * GDB Files:: @value{GDBN} files
143 * Targets:: Specifying a debugging target
144 * Remote Debugging:: Debugging remote programs
145 * Configurations:: Configuration-specific information
146 * Controlling GDB:: Controlling @value{GDBN}
147 * Sequences:: Canned sequences of commands
148 * TUI:: @value{GDBN} Text User Interface
149 * Interpreters:: Command Interpreters
150 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
151 * Annotations:: @value{GDBN}'s annotation interface.
152 * GDB/MI:: @value{GDBN}'s Machine Interface.
154 * GDB Bugs:: Reporting bugs in @value{GDBN}
155 * Formatting Documentation:: How to format and print @value{GDBN} documentation
157 * Command Line Editing:: Command Line Editing
158 * Using History Interactively:: Using History Interactively
159 * Installing GDB:: Installing GDB
160 * Maintenance Commands:: Maintenance Commands
161 * Remote Protocol:: GDB Remote Serial Protocol
162 * Copying:: GNU General Public License says
163 how you can copy and share GDB
164 * GNU Free Documentation License:: The license for this documentation
173 @unnumbered Summary of @value{GDBN}
175 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
176 going on ``inside'' another program while it executes---or what another
177 program was doing at the moment it crashed.
179 @value{GDBN} can do four main kinds of things (plus other things in support of
180 these) to help you catch bugs in the act:
184 Start your program, specifying anything that might affect its behavior.
187 Make your program stop on specified conditions.
190 Examine what has happened, when your program has stopped.
193 Change things in your program, so you can experiment with correcting the
194 effects of one bug and go on to learn about another.
197 You can use @value{GDBN} to debug programs written in C and C++.
198 For more information, see @ref{Support,,Supported languages}.
199 For more information, see @ref{C,,C and C++}.
202 Support for Modula-2 is partial. For information on Modula-2, see
203 @ref{Modula-2,,Modula-2}.
206 Debugging Pascal programs which use sets, subranges, file variables, or
207 nested functions does not currently work. @value{GDBN} does not support
208 entering expressions, printing values, or similar features using Pascal
212 @value{GDBN} can be used to debug programs written in Fortran, although
213 it may be necessary to refer to some variables with a trailing
217 * Free Software:: Freely redistributable software
218 * Contributors:: Contributors to GDB
222 @unnumberedsec Free software
224 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
225 General Public License
226 (GPL). The GPL gives you the freedom to copy or adapt a licensed
227 program---but every person getting a copy also gets with it the
228 freedom to modify that copy (which means that they must get access to
229 the source code), and the freedom to distribute further copies.
230 Typical software companies use copyrights to limit your freedoms; the
231 Free Software Foundation uses the GPL to preserve these freedoms.
233 Fundamentally, the General Public License is a license which says that
234 you have these freedoms and that you cannot take these freedoms away
237 @unnumberedsec Free Software Needs Free Documentation
239 The biggest deficiency in the free software community today is not in
240 the software---it is the lack of good free documentation that we can
241 include with the free software. Many of our most important
242 programs do not come with free reference manuals and free introductory
243 texts. Documentation is an essential part of any software package;
244 when an important free software package does not come with a free
245 manual and a free tutorial, that is a major gap. We have many such
248 Consider Perl, for instance. The tutorial manuals that people
249 normally use are non-free. How did this come about? Because the
250 authors of those manuals published them with restrictive terms---no
251 copying, no modification, source files not available---which exclude
252 them from the free software world.
254 That wasn't the first time this sort of thing happened, and it was far
255 from the last. Many times we have heard a GNU user eagerly describe a
256 manual that he is writing, his intended contribution to the community,
257 only to learn that he had ruined everything by signing a publication
258 contract to make it non-free.
260 Free documentation, like free software, is a matter of freedom, not
261 price. The problem with the non-free manual is not that publishers
262 charge a price for printed copies---that in itself is fine. (The Free
263 Software Foundation sells printed copies of manuals, too.) The
264 problem is the restrictions on the use of the manual. Free manuals
265 are available in source code form, and give you permission to copy and
266 modify. Non-free manuals do not allow this.
268 The criteria of freedom for a free manual are roughly the same as for
269 free software. Redistribution (including the normal kinds of
270 commercial redistribution) must be permitted, so that the manual can
271 accompany every copy of the program, both on-line and on paper.
273 Permission for modification of the technical content is crucial too.
274 When people modify the software, adding or changing features, if they
275 are conscientious they will change the manual too---so they can
276 provide accurate and clear documentation for the modified program. A
277 manual that leaves you no choice but to write a new manual to document
278 a changed version of the program is not really available to our
281 Some kinds of limits on the way modification is handled are
282 acceptable. For example, requirements to preserve the original
283 author's copyright notice, the distribution terms, or the list of
284 authors, are ok. It is also no problem to require modified versions
285 to include notice that they were modified. Even entire sections that
286 may not be deleted or changed are acceptable, as long as they deal
287 with nontechnical topics (like this one). These kinds of restrictions
288 are acceptable because they don't obstruct the community's normal use
291 However, it must be possible to modify all the @emph{technical}
292 content of the manual, and then distribute the result in all the usual
293 media, through all the usual channels. Otherwise, the restrictions
294 obstruct the use of the manual, it is not free, and we need another
295 manual to replace it.
297 Please spread the word about this issue. Our community continues to
298 lose manuals to proprietary publishing. If we spread the word that
299 free software needs free reference manuals and free tutorials, perhaps
300 the next person who wants to contribute by writing documentation will
301 realize, before it is too late, that only free manuals contribute to
302 the free software community.
304 If you are writing documentation, please insist on publishing it under
305 the GNU Free Documentation License or another free documentation
306 license. Remember that this decision requires your approval---you
307 don't have to let the publisher decide. Some commercial publishers
308 will use a free license if you insist, but they will not propose the
309 option; it is up to you to raise the issue and say firmly that this is
310 what you want. If the publisher you are dealing with refuses, please
311 try other publishers. If you're not sure whether a proposed license
312 is free, write to @email{licensing@@gnu.org}.
314 You can encourage commercial publishers to sell more free, copylefted
315 manuals and tutorials by buying them, and particularly by buying
316 copies from the publishers that paid for their writing or for major
317 improvements. Meanwhile, try to avoid buying non-free documentation
318 at all. Check the distribution terms of a manual before you buy it,
319 and insist that whoever seeks your business must respect your freedom.
320 Check the history of the book, and try to reward the publishers that
321 have paid or pay the authors to work on it.
323 The Free Software Foundation maintains a list of free documentation
324 published by other publishers, at
325 @url{http://www.fsf.org/doc/other-free-books.html}.
328 @unnumberedsec Contributors to @value{GDBN}
330 Richard Stallman was the original author of @value{GDBN}, and of many
331 other @sc{gnu} programs. Many others have contributed to its
332 development. This section attempts to credit major contributors. One
333 of the virtues of free software is that everyone is free to contribute
334 to it; with regret, we cannot actually acknowledge everyone here. The
335 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
336 blow-by-blow account.
338 Changes much prior to version 2.0 are lost in the mists of time.
341 @emph{Plea:} Additions to this section are particularly welcome. If you
342 or your friends (or enemies, to be evenhanded) have been unfairly
343 omitted from this list, we would like to add your names!
346 So that they may not regard their many labors as thankless, we
347 particularly thank those who shepherded @value{GDBN} through major
349 Andrew Cagney (releases 5.3, 5.2, 5.1 and 5.0);
350 Jim Blandy (release 4.18);
351 Jason Molenda (release 4.17);
352 Stan Shebs (release 4.14);
353 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
354 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
355 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
356 Jim Kingdon (releases 3.5, 3.4, and 3.3);
357 and Randy Smith (releases 3.2, 3.1, and 3.0).
359 Richard Stallman, assisted at various times by Peter TerMaat, Chris
360 Hanson, and Richard Mlynarik, handled releases through 2.8.
362 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
363 in @value{GDBN}, with significant additional contributions from Per
364 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
365 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
366 much general update work leading to release 3.0).
368 @value{GDBN} uses the BFD subroutine library to examine multiple
369 object-file formats; BFD was a joint project of David V.
370 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
372 David Johnson wrote the original COFF support; Pace Willison did
373 the original support for encapsulated COFF.
375 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
377 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
378 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
380 Jean-Daniel Fekete contributed Sun 386i support.
381 Chris Hanson improved the HP9000 support.
382 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
383 David Johnson contributed Encore Umax support.
384 Jyrki Kuoppala contributed Altos 3068 support.
385 Jeff Law contributed HP PA and SOM support.
386 Keith Packard contributed NS32K support.
387 Doug Rabson contributed Acorn Risc Machine support.
388 Bob Rusk contributed Harris Nighthawk CX-UX support.
389 Chris Smith contributed Convex support (and Fortran debugging).
390 Jonathan Stone contributed Pyramid support.
391 Michael Tiemann contributed SPARC support.
392 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
393 Pace Willison contributed Intel 386 support.
394 Jay Vosburgh contributed Symmetry support.
395 Marko Mlinar contributed OpenRISC 1000 support.
397 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
399 Rich Schaefer and Peter Schauer helped with support of SunOS shared
402 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
403 about several machine instruction sets.
405 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
406 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
407 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
408 and RDI targets, respectively.
410 Brian Fox is the author of the readline libraries providing
411 command-line editing and command history.
413 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
414 Modula-2 support, and contributed the Languages chapter of this manual.
416 Fred Fish wrote most of the support for Unix System Vr4.
417 He also enhanced the command-completion support to cover C@t{++} overloaded
420 Hitachi America, Ltd. sponsored the support for H8/300, H8/500, and
423 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
425 Mitsubishi sponsored the support for D10V, D30V, and M32R/D processors.
427 Toshiba sponsored the support for the TX39 Mips processor.
429 Matsushita sponsored the support for the MN10200 and MN10300 processors.
431 Fujitsu sponsored the support for SPARClite and FR30 processors.
433 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
436 Michael Snyder added support for tracepoints.
438 Stu Grossman wrote gdbserver.
440 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
441 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
443 The following people at the Hewlett-Packard Company contributed
444 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
445 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
446 compiler, and the terminal user interface: Ben Krepp, Richard Title,
447 John Bishop, Susan Macchia, Kathy Mann, Satish Pai, India Paul, Steve
448 Rehrauer, and Elena Zannoni. Kim Haase provided HP-specific
449 information in this manual.
451 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
452 Robert Hoehne made significant contributions to the DJGPP port.
454 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
455 development since 1991. Cygnus engineers who have worked on @value{GDBN}
456 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
457 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
458 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
459 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
460 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
461 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
462 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
463 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
464 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
465 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
466 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
467 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
468 Zuhn have made contributions both large and small.
470 Jim Blandy added support for preprocessor macros, while working for Red
474 @chapter A Sample @value{GDBN} Session
476 You can use this manual at your leisure to read all about @value{GDBN}.
477 However, a handful of commands are enough to get started using the
478 debugger. This chapter illustrates those commands.
481 In this sample session, we emphasize user input like this: @b{input},
482 to make it easier to pick out from the surrounding output.
485 @c FIXME: this example may not be appropriate for some configs, where
486 @c FIXME...primary interest is in remote use.
488 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
489 processor) exhibits the following bug: sometimes, when we change its
490 quote strings from the default, the commands used to capture one macro
491 definition within another stop working. In the following short @code{m4}
492 session, we define a macro @code{foo} which expands to @code{0000}; we
493 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
494 same thing. However, when we change the open quote string to
495 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
496 procedure fails to define a new synonym @code{baz}:
505 @b{define(bar,defn(`foo'))}
509 @b{changequote(<QUOTE>,<UNQUOTE>)}
511 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
514 m4: End of input: 0: fatal error: EOF in string
518 Let us use @value{GDBN} to try to see what is going on.
521 $ @b{@value{GDBP} m4}
522 @c FIXME: this falsifies the exact text played out, to permit smallbook
523 @c FIXME... format to come out better.
524 @value{GDBN} is free software and you are welcome to distribute copies
525 of it under certain conditions; type "show copying" to see
527 There is absolutely no warranty for @value{GDBN}; type "show warranty"
530 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
535 @value{GDBN} reads only enough symbol data to know where to find the
536 rest when needed; as a result, the first prompt comes up very quickly.
537 We now tell @value{GDBN} to use a narrower display width than usual, so
538 that examples fit in this manual.
541 (@value{GDBP}) @b{set width 70}
545 We need to see how the @code{m4} built-in @code{changequote} works.
546 Having looked at the source, we know the relevant subroutine is
547 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
548 @code{break} command.
551 (@value{GDBP}) @b{break m4_changequote}
552 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
556 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
557 control; as long as control does not reach the @code{m4_changequote}
558 subroutine, the program runs as usual:
561 (@value{GDBP}) @b{run}
562 Starting program: /work/Editorial/gdb/gnu/m4/m4
570 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
571 suspends execution of @code{m4}, displaying information about the
572 context where it stops.
575 @b{changequote(<QUOTE>,<UNQUOTE>)}
577 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
579 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
583 Now we use the command @code{n} (@code{next}) to advance execution to
584 the next line of the current function.
588 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
593 @code{set_quotes} looks like a promising subroutine. We can go into it
594 by using the command @code{s} (@code{step}) instead of @code{next}.
595 @code{step} goes to the next line to be executed in @emph{any}
596 subroutine, so it steps into @code{set_quotes}.
600 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
602 530 if (lquote != def_lquote)
606 The display that shows the subroutine where @code{m4} is now
607 suspended (and its arguments) is called a stack frame display. It
608 shows a summary of the stack. We can use the @code{backtrace}
609 command (which can also be spelled @code{bt}), to see where we are
610 in the stack as a whole: the @code{backtrace} command displays a
611 stack frame for each active subroutine.
614 (@value{GDBP}) @b{bt}
615 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
617 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
619 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
620 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
622 #4 0x79dc in expand_input () at macro.c:40
623 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
627 We step through a few more lines to see what happens. The first two
628 times, we can use @samp{s}; the next two times we use @code{n} to avoid
629 falling into the @code{xstrdup} subroutine.
633 0x3b5c 532 if (rquote != def_rquote)
635 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
636 def_lquote : xstrdup(lq);
638 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
641 538 len_lquote = strlen(rquote);
645 The last line displayed looks a little odd; we can examine the variables
646 @code{lquote} and @code{rquote} to see if they are in fact the new left
647 and right quotes we specified. We use the command @code{p}
648 (@code{print}) to see their values.
651 (@value{GDBP}) @b{p lquote}
652 $1 = 0x35d40 "<QUOTE>"
653 (@value{GDBP}) @b{p rquote}
654 $2 = 0x35d50 "<UNQUOTE>"
658 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
659 To look at some context, we can display ten lines of source
660 surrounding the current line with the @code{l} (@code{list}) command.
666 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
668 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
671 538 len_lquote = strlen(rquote);
672 539 len_rquote = strlen(lquote);
679 Let us step past the two lines that set @code{len_lquote} and
680 @code{len_rquote}, and then examine the values of those variables.
684 539 len_rquote = strlen(lquote);
687 (@value{GDBP}) @b{p len_lquote}
689 (@value{GDBP}) @b{p len_rquote}
694 That certainly looks wrong, assuming @code{len_lquote} and
695 @code{len_rquote} are meant to be the lengths of @code{lquote} and
696 @code{rquote} respectively. We can set them to better values using
697 the @code{p} command, since it can print the value of
698 any expression---and that expression can include subroutine calls and
702 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
704 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
709 Is that enough to fix the problem of using the new quotes with the
710 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
711 executing with the @code{c} (@code{continue}) command, and then try the
712 example that caused trouble initially:
718 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
725 Success! The new quotes now work just as well as the default ones. The
726 problem seems to have been just the two typos defining the wrong
727 lengths. We allow @code{m4} exit by giving it an EOF as input:
731 Program exited normally.
735 The message @samp{Program exited normally.} is from @value{GDBN}; it
736 indicates @code{m4} has finished executing. We can end our @value{GDBN}
737 session with the @value{GDBN} @code{quit} command.
740 (@value{GDBP}) @b{quit}
744 @chapter Getting In and Out of @value{GDBN}
746 This chapter discusses how to start @value{GDBN}, and how to get out of it.
750 type @samp{@value{GDBP}} to start @value{GDBN}.
752 type @kbd{quit} or @kbd{C-d} to exit.
756 * Invoking GDB:: How to start @value{GDBN}
757 * Quitting GDB:: How to quit @value{GDBN}
758 * Shell Commands:: How to use shell commands inside @value{GDBN}
762 @section Invoking @value{GDBN}
764 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
765 @value{GDBN} reads commands from the terminal until you tell it to exit.
767 You can also run @code{@value{GDBP}} with a variety of arguments and options,
768 to specify more of your debugging environment at the outset.
770 The command-line options described here are designed
771 to cover a variety of situations; in some environments, some of these
772 options may effectively be unavailable.
774 The most usual way to start @value{GDBN} is with one argument,
775 specifying an executable program:
778 @value{GDBP} @var{program}
782 You can also start with both an executable program and a core file
786 @value{GDBP} @var{program} @var{core}
789 You can, instead, specify a process ID as a second argument, if you want
790 to debug a running process:
793 @value{GDBP} @var{program} 1234
797 would attach @value{GDBN} to process @code{1234} (unless you also have a file
798 named @file{1234}; @value{GDBN} does check for a core file first).
800 Taking advantage of the second command-line argument requires a fairly
801 complete operating system; when you use @value{GDBN} as a remote
802 debugger attached to a bare board, there may not be any notion of
803 ``process'', and there is often no way to get a core dump. @value{GDBN}
804 will warn you if it is unable to attach or to read core dumps.
806 You can optionally have @code{@value{GDBP}} pass any arguments after the
807 executable file to the inferior using @code{--args}. This option stops
810 gdb --args gcc -O2 -c foo.c
812 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
813 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
815 You can run @code{@value{GDBP}} without printing the front material, which describes
816 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
823 You can further control how @value{GDBN} starts up by using command-line
824 options. @value{GDBN} itself can remind you of the options available.
834 to display all available options and briefly describe their use
835 (@samp{@value{GDBP} -h} is a shorter equivalent).
837 All options and command line arguments you give are processed
838 in sequential order. The order makes a difference when the
839 @samp{-x} option is used.
843 * File Options:: Choosing files
844 * Mode Options:: Choosing modes
848 @subsection Choosing files
850 When @value{GDBN} starts, it reads any arguments other than options as
851 specifying an executable file and core file (or process ID). This is
852 the same as if the arguments were specified by the @samp{-se} and
853 @samp{-c} (or @samp{-p} options respectively. (@value{GDBN} reads the
854 first argument that does not have an associated option flag as
855 equivalent to the @samp{-se} option followed by that argument; and the
856 second argument that does not have an associated option flag, if any, as
857 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
858 If the second argument begins with a decimal digit, @value{GDBN} will
859 first attempt to attach to it as a process, and if that fails, attempt
860 to open it as a corefile. If you have a corefile whose name begins with
861 a digit, you can prevent @value{GDBN} from treating it as a pid by
862 prefixing it with @file{./}, eg. @file{./12345}.
864 If @value{GDBN} has not been configured to included core file support,
865 such as for most embedded targets, then it will complain about a second
866 argument and ignore it.
868 Many options have both long and short forms; both are shown in the
869 following list. @value{GDBN} also recognizes the long forms if you truncate
870 them, so long as enough of the option is present to be unambiguous.
871 (If you prefer, you can flag option arguments with @samp{--} rather
872 than @samp{-}, though we illustrate the more usual convention.)
874 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
875 @c way, both those who look for -foo and --foo in the index, will find
879 @item -symbols @var{file}
881 @cindex @code{--symbols}
883 Read symbol table from file @var{file}.
885 @item -exec @var{file}
887 @cindex @code{--exec}
889 Use file @var{file} as the executable file to execute when appropriate,
890 and for examining pure data in conjunction with a core dump.
894 Read symbol table from file @var{file} and use it as the executable
897 @item -core @var{file}
899 @cindex @code{--core}
901 Use file @var{file} as a core dump to examine.
903 @item -c @var{number}
904 @item -pid @var{number}
905 @itemx -p @var{number}
908 Connect to process ID @var{number}, as with the @code{attach} command.
909 If there is no such process, @value{GDBN} will attempt to open a core
910 file named @var{number}.
912 @item -command @var{file}
914 @cindex @code{--command}
916 Execute @value{GDBN} commands from file @var{file}. @xref{Command
917 Files,, Command files}.
919 @item -directory @var{directory}
920 @itemx -d @var{directory}
921 @cindex @code{--directory}
923 Add @var{directory} to the path to search for source files.
927 @cindex @code{--mapped}
929 @emph{Warning: this option depends on operating system facilities that are not
930 supported on all systems.}@*
931 If memory-mapped files are available on your system through the @code{mmap}
932 system call, you can use this option
933 to have @value{GDBN} write the symbols from your
934 program into a reusable file in the current directory. If the program you are debugging is
935 called @file{/tmp/fred}, the mapped symbol file is @file{/tmp/fred.syms}.
936 Future @value{GDBN} debugging sessions notice the presence of this file,
937 and can quickly map in symbol information from it, rather than reading
938 the symbol table from the executable program.
940 The @file{.syms} file is specific to the host machine where @value{GDBN}
941 is run. It holds an exact image of the internal @value{GDBN} symbol
942 table. It cannot be shared across multiple host platforms.
946 @cindex @code{--readnow}
948 Read each symbol file's entire symbol table immediately, rather than
949 the default, which is to read it incrementally as it is needed.
950 This makes startup slower, but makes future operations faster.
954 You typically combine the @code{-mapped} and @code{-readnow} options in
955 order to build a @file{.syms} file that contains complete symbol
956 information. (@xref{Files,,Commands to specify files}, for information
957 on @file{.syms} files.) A simple @value{GDBN} invocation to do nothing
958 but build a @file{.syms} file for future use is:
961 gdb -batch -nx -mapped -readnow programname
965 @subsection Choosing modes
967 You can run @value{GDBN} in various alternative modes---for example, in
968 batch mode or quiet mode.
975 Do not execute commands found in any initialization files. Normally,
976 @value{GDBN} executes the commands in these files after all the command
977 options and arguments have been processed. @xref{Command Files,,Command
983 @cindex @code{--quiet}
984 @cindex @code{--silent}
986 ``Quiet''. Do not print the introductory and copyright messages. These
987 messages are also suppressed in batch mode.
990 @cindex @code{--batch}
991 Run in batch mode. Exit with status @code{0} after processing all the
992 command files specified with @samp{-x} (and all commands from
993 initialization files, if not inhibited with @samp{-n}). Exit with
994 nonzero status if an error occurs in executing the @value{GDBN} commands
995 in the command files.
997 Batch mode may be useful for running @value{GDBN} as a filter, for
998 example to download and run a program on another computer; in order to
999 make this more useful, the message
1002 Program exited normally.
1006 (which is ordinarily issued whenever a program running under
1007 @value{GDBN} control terminates) is not issued when running in batch
1012 @cindex @code{--nowindows}
1014 ``No windows''. If @value{GDBN} comes with a graphical user interface
1015 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1016 interface. If no GUI is available, this option has no effect.
1020 @cindex @code{--windows}
1022 If @value{GDBN} includes a GUI, then this option requires it to be
1025 @item -cd @var{directory}
1027 Run @value{GDBN} using @var{directory} as its working directory,
1028 instead of the current directory.
1032 @cindex @code{--fullname}
1034 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1035 subprocess. It tells @value{GDBN} to output the full file name and line
1036 number in a standard, recognizable fashion each time a stack frame is
1037 displayed (which includes each time your program stops). This
1038 recognizable format looks like two @samp{\032} characters, followed by
1039 the file name, line number and character position separated by colons,
1040 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1041 @samp{\032} characters as a signal to display the source code for the
1045 @cindex @code{--epoch}
1046 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1047 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1048 routines so as to allow Epoch to display values of expressions in a
1051 @item -annotate @var{level}
1052 @cindex @code{--annotate}
1053 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1054 effect is identical to using @samp{set annotate @var{level}}
1055 (@pxref{Annotations}).
1056 Annotation level controls how much information does @value{GDBN} print
1057 together with its prompt, values of expressions, source lines, and other
1058 types of output. Level 0 is the normal, level 1 is for use when
1059 @value{GDBN} is run as a subprocess of @sc{gnu} Emacs, level 2 is the
1060 maximum annotation suitable for programs that control @value{GDBN}.
1063 @cindex @code{--async}
1064 Use the asynchronous event loop for the command-line interface.
1065 @value{GDBN} processes all events, such as user keyboard input, via a
1066 special event loop. This allows @value{GDBN} to accept and process user
1067 commands in parallel with the debugged process being
1068 run@footnote{@value{GDBN} built with @sc{djgpp} tools for
1069 MS-DOS/MS-Windows supports this mode of operation, but the event loop is
1070 suspended when the debuggee runs.}, so you don't need to wait for
1071 control to return to @value{GDBN} before you type the next command.
1072 (@emph{Note:} as of version 5.1, the target side of the asynchronous
1073 operation is not yet in place, so @samp{-async} does not work fully
1075 @c FIXME: when the target side of the event loop is done, the above NOTE
1076 @c should be removed.
1078 When the standard input is connected to a terminal device, @value{GDBN}
1079 uses the asynchronous event loop by default, unless disabled by the
1080 @samp{-noasync} option.
1083 @cindex @code{--noasync}
1084 Disable the asynchronous event loop for the command-line interface.
1087 @cindex @code{--args}
1088 Change interpretation of command line so that arguments following the
1089 executable file are passed as command line arguments to the inferior.
1090 This option stops option processing.
1092 @item -baud @var{bps}
1094 @cindex @code{--baud}
1096 Set the line speed (baud rate or bits per second) of any serial
1097 interface used by @value{GDBN} for remote debugging.
1099 @item -tty @var{device}
1100 @itemx -t @var{device}
1101 @cindex @code{--tty}
1103 Run using @var{device} for your program's standard input and output.
1104 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1106 @c resolve the situation of these eventually
1108 @cindex @code{--tui}
1109 Activate the Terminal User Interface when starting.
1110 The Terminal User Interface manages several text windows on the terminal,
1111 showing source, assembly, registers and @value{GDBN} command outputs
1112 (@pxref{TUI, ,@value{GDBN} Text User Interface}).
1113 Do not use this option if you run @value{GDBN} from Emacs
1114 (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1117 @c @cindex @code{--xdb}
1118 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1119 @c For information, see the file @file{xdb_trans.html}, which is usually
1120 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1123 @item -interpreter @var{interp}
1124 @cindex @code{--interpreter}
1125 Use the interpreter @var{interp} for interface with the controlling
1126 program or device. This option is meant to be set by programs which
1127 communicate with @value{GDBN} using it as a back end.
1128 @xref{Interpreters, , Command Interpreters}.
1130 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1131 @value{GDBN} to use the current @dfn{@sc{gdb/mi} interface}
1132 (@pxref{GDB/MI, , The @sc{gdb/mi} Interface}). The previous @sc{gdb/mi}
1133 interface, included in @value{GDBN} version 5.3, can be selected with
1134 @samp{--interpreter=mi1}. Earlier @sc{gdb/mi} interfaces
1138 @cindex @code{--write}
1139 Open the executable and core files for both reading and writing. This
1140 is equivalent to the @samp{set write on} command inside @value{GDBN}
1144 @cindex @code{--statistics}
1145 This option causes @value{GDBN} to print statistics about time and
1146 memory usage after it completes each command and returns to the prompt.
1149 @cindex @code{--version}
1150 This option causes @value{GDBN} to print its version number and
1151 no-warranty blurb, and exit.
1156 @section Quitting @value{GDBN}
1157 @cindex exiting @value{GDBN}
1158 @cindex leaving @value{GDBN}
1161 @kindex quit @r{[}@var{expression}@r{]}
1162 @kindex q @r{(@code{quit})}
1163 @item quit @r{[}@var{expression}@r{]}
1165 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1166 @code{q}), or type an end-of-file character (usually @kbd{C-d}). If you
1167 do not supply @var{expression}, @value{GDBN} will terminate normally;
1168 otherwise it will terminate using the result of @var{expression} as the
1173 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
1174 terminates the action of any @value{GDBN} command that is in progress and
1175 returns to @value{GDBN} command level. It is safe to type the interrupt
1176 character at any time because @value{GDBN} does not allow it to take effect
1177 until a time when it is safe.
1179 If you have been using @value{GDBN} to control an attached process or
1180 device, you can release it with the @code{detach} command
1181 (@pxref{Attach, ,Debugging an already-running process}).
1183 @node Shell Commands
1184 @section Shell commands
1186 If you need to execute occasional shell commands during your
1187 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1188 just use the @code{shell} command.
1192 @cindex shell escape
1193 @item shell @var{command string}
1194 Invoke a standard shell to execute @var{command string}.
1195 If it exists, the environment variable @code{SHELL} determines which
1196 shell to run. Otherwise @value{GDBN} uses the default shell
1197 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1200 The utility @code{make} is often needed in development environments.
1201 You do not have to use the @code{shell} command for this purpose in
1206 @cindex calling make
1207 @item make @var{make-args}
1208 Execute the @code{make} program with the specified
1209 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1213 @chapter @value{GDBN} Commands
1215 You can abbreviate a @value{GDBN} command to the first few letters of the command
1216 name, if that abbreviation is unambiguous; and you can repeat certain
1217 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1218 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1219 show you the alternatives available, if there is more than one possibility).
1222 * Command Syntax:: How to give commands to @value{GDBN}
1223 * Completion:: Command completion
1224 * Help:: How to ask @value{GDBN} for help
1227 @node Command Syntax
1228 @section Command syntax
1230 A @value{GDBN} command is a single line of input. There is no limit on
1231 how long it can be. It starts with a command name, which is followed by
1232 arguments whose meaning depends on the command name. For example, the
1233 command @code{step} accepts an argument which is the number of times to
1234 step, as in @samp{step 5}. You can also use the @code{step} command
1235 with no arguments. Some commands do not allow any arguments.
1237 @cindex abbreviation
1238 @value{GDBN} command names may always be truncated if that abbreviation is
1239 unambiguous. Other possible command abbreviations are listed in the
1240 documentation for individual commands. In some cases, even ambiguous
1241 abbreviations are allowed; for example, @code{s} is specially defined as
1242 equivalent to @code{step} even though there are other commands whose
1243 names start with @code{s}. You can test abbreviations by using them as
1244 arguments to the @code{help} command.
1246 @cindex repeating commands
1247 @kindex RET @r{(repeat last command)}
1248 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1249 repeat the previous command. Certain commands (for example, @code{run})
1250 will not repeat this way; these are commands whose unintentional
1251 repetition might cause trouble and which you are unlikely to want to
1254 The @code{list} and @code{x} commands, when you repeat them with
1255 @key{RET}, construct new arguments rather than repeating
1256 exactly as typed. This permits easy scanning of source or memory.
1258 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1259 output, in a way similar to the common utility @code{more}
1260 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1261 @key{RET} too many in this situation, @value{GDBN} disables command
1262 repetition after any command that generates this sort of display.
1264 @kindex # @r{(a comment)}
1266 Any text from a @kbd{#} to the end of the line is a comment; it does
1267 nothing. This is useful mainly in command files (@pxref{Command
1268 Files,,Command files}).
1270 @cindex repeating command sequences
1271 @kindex C-o @r{(operate-and-get-next)}
1272 The @kbd{C-o} binding is useful for repeating a complex sequence of
1273 commands. This command accepts the current line, like @kbd{RET}, and
1274 then fetches the next line relative to the current line from the history
1278 @section Command completion
1281 @cindex word completion
1282 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1283 only one possibility; it can also show you what the valid possibilities
1284 are for the next word in a command, at any time. This works for @value{GDBN}
1285 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1287 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1288 of a word. If there is only one possibility, @value{GDBN} fills in the
1289 word, and waits for you to finish the command (or press @key{RET} to
1290 enter it). For example, if you type
1292 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1293 @c complete accuracy in these examples; space introduced for clarity.
1294 @c If texinfo enhancements make it unnecessary, it would be nice to
1295 @c replace " @key" by "@key" in the following...
1297 (@value{GDBP}) info bre @key{TAB}
1301 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1302 the only @code{info} subcommand beginning with @samp{bre}:
1305 (@value{GDBP}) info breakpoints
1309 You can either press @key{RET} at this point, to run the @code{info
1310 breakpoints} command, or backspace and enter something else, if
1311 @samp{breakpoints} does not look like the command you expected. (If you
1312 were sure you wanted @code{info breakpoints} in the first place, you
1313 might as well just type @key{RET} immediately after @samp{info bre},
1314 to exploit command abbreviations rather than command completion).
1316 If there is more than one possibility for the next word when you press
1317 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1318 characters and try again, or just press @key{TAB} a second time;
1319 @value{GDBN} displays all the possible completions for that word. For
1320 example, you might want to set a breakpoint on a subroutine whose name
1321 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1322 just sounds the bell. Typing @key{TAB} again displays all the
1323 function names in your program that begin with those characters, for
1327 (@value{GDBP}) b make_ @key{TAB}
1328 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1329 make_a_section_from_file make_environ
1330 make_abs_section make_function_type
1331 make_blockvector make_pointer_type
1332 make_cleanup make_reference_type
1333 make_command make_symbol_completion_list
1334 (@value{GDBP}) b make_
1338 After displaying the available possibilities, @value{GDBN} copies your
1339 partial input (@samp{b make_} in the example) so you can finish the
1342 If you just want to see the list of alternatives in the first place, you
1343 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1344 means @kbd{@key{META} ?}. You can type this either by holding down a
1345 key designated as the @key{META} shift on your keyboard (if there is
1346 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1348 @cindex quotes in commands
1349 @cindex completion of quoted strings
1350 Sometimes the string you need, while logically a ``word'', may contain
1351 parentheses or other characters that @value{GDBN} normally excludes from
1352 its notion of a word. To permit word completion to work in this
1353 situation, you may enclose words in @code{'} (single quote marks) in
1354 @value{GDBN} commands.
1356 The most likely situation where you might need this is in typing the
1357 name of a C@t{++} function. This is because C@t{++} allows function
1358 overloading (multiple definitions of the same function, distinguished
1359 by argument type). For example, when you want to set a breakpoint you
1360 may need to distinguish whether you mean the version of @code{name}
1361 that takes an @code{int} parameter, @code{name(int)}, or the version
1362 that takes a @code{float} parameter, @code{name(float)}. To use the
1363 word-completion facilities in this situation, type a single quote
1364 @code{'} at the beginning of the function name. This alerts
1365 @value{GDBN} that it may need to consider more information than usual
1366 when you press @key{TAB} or @kbd{M-?} to request word completion:
1369 (@value{GDBP}) b 'bubble( @kbd{M-?}
1370 bubble(double,double) bubble(int,int)
1371 (@value{GDBP}) b 'bubble(
1374 In some cases, @value{GDBN} can tell that completing a name requires using
1375 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1376 completing as much as it can) if you do not type the quote in the first
1380 (@value{GDBP}) b bub @key{TAB}
1381 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1382 (@value{GDBP}) b 'bubble(
1386 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1387 you have not yet started typing the argument list when you ask for
1388 completion on an overloaded symbol.
1390 For more information about overloaded functions, see @ref{C plus plus
1391 expressions, ,C@t{++} expressions}. You can use the command @code{set
1392 overload-resolution off} to disable overload resolution;
1393 see @ref{Debugging C plus plus, ,@value{GDBN} features for C@t{++}}.
1397 @section Getting help
1398 @cindex online documentation
1401 You can always ask @value{GDBN} itself for information on its commands,
1402 using the command @code{help}.
1405 @kindex h @r{(@code{help})}
1408 You can use @code{help} (abbreviated @code{h}) with no arguments to
1409 display a short list of named classes of commands:
1413 List of classes of commands:
1415 aliases -- Aliases of other commands
1416 breakpoints -- Making program stop at certain points
1417 data -- Examining data
1418 files -- Specifying and examining files
1419 internals -- Maintenance commands
1420 obscure -- Obscure features
1421 running -- Running the program
1422 stack -- Examining the stack
1423 status -- Status inquiries
1424 support -- Support facilities
1425 tracepoints -- Tracing of program execution without@*
1426 stopping the program
1427 user-defined -- User-defined commands
1429 Type "help" followed by a class name for a list of
1430 commands in that class.
1431 Type "help" followed by command name for full
1433 Command name abbreviations are allowed if unambiguous.
1436 @c the above line break eliminates huge line overfull...
1438 @item help @var{class}
1439 Using one of the general help classes as an argument, you can get a
1440 list of the individual commands in that class. For example, here is the
1441 help display for the class @code{status}:
1444 (@value{GDBP}) help status
1449 @c Line break in "show" line falsifies real output, but needed
1450 @c to fit in smallbook page size.
1451 info -- Generic command for showing things
1452 about the program being debugged
1453 show -- Generic command for showing things
1456 Type "help" followed by command name for full
1458 Command name abbreviations are allowed if unambiguous.
1462 @item help @var{command}
1463 With a command name as @code{help} argument, @value{GDBN} displays a
1464 short paragraph on how to use that command.
1467 @item apropos @var{args}
1468 The @code{apropos @var{args}} command searches through all of the @value{GDBN}
1469 commands, and their documentation, for the regular expression specified in
1470 @var{args}. It prints out all matches found. For example:
1481 set symbol-reloading -- Set dynamic symbol table reloading
1482 multiple times in one run
1483 show symbol-reloading -- Show dynamic symbol table reloading
1484 multiple times in one run
1489 @item complete @var{args}
1490 The @code{complete @var{args}} command lists all the possible completions
1491 for the beginning of a command. Use @var{args} to specify the beginning of the
1492 command you want completed. For example:
1498 @noindent results in:
1509 @noindent This is intended for use by @sc{gnu} Emacs.
1512 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1513 and @code{show} to inquire about the state of your program, or the state
1514 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1515 manual introduces each of them in the appropriate context. The listings
1516 under @code{info} and under @code{show} in the Index point to
1517 all the sub-commands. @xref{Index}.
1522 @kindex i @r{(@code{info})}
1524 This command (abbreviated @code{i}) is for describing the state of your
1525 program. For example, you can list the arguments given to your program
1526 with @code{info args}, list the registers currently in use with @code{info
1527 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1528 You can get a complete list of the @code{info} sub-commands with
1529 @w{@code{help info}}.
1533 You can assign the result of an expression to an environment variable with
1534 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1535 @code{set prompt $}.
1539 In contrast to @code{info}, @code{show} is for describing the state of
1540 @value{GDBN} itself.
1541 You can change most of the things you can @code{show}, by using the
1542 related command @code{set}; for example, you can control what number
1543 system is used for displays with @code{set radix}, or simply inquire
1544 which is currently in use with @code{show radix}.
1547 To display all the settable parameters and their current
1548 values, you can use @code{show} with no arguments; you may also use
1549 @code{info set}. Both commands produce the same display.
1550 @c FIXME: "info set" violates the rule that "info" is for state of
1551 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1552 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1556 Here are three miscellaneous @code{show} subcommands, all of which are
1557 exceptional in lacking corresponding @code{set} commands:
1560 @kindex show version
1561 @cindex version number
1563 Show what version of @value{GDBN} is running. You should include this
1564 information in @value{GDBN} bug-reports. If multiple versions of
1565 @value{GDBN} are in use at your site, you may need to determine which
1566 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1567 commands are introduced, and old ones may wither away. Also, many
1568 system vendors ship variant versions of @value{GDBN}, and there are
1569 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1570 The version number is the same as the one announced when you start
1573 @kindex show copying
1575 Display information about permission for copying @value{GDBN}.
1577 @kindex show warranty
1579 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1580 if your version of @value{GDBN} comes with one.
1585 @chapter Running Programs Under @value{GDBN}
1587 When you run a program under @value{GDBN}, you must first generate
1588 debugging information when you compile it.
1590 You may start @value{GDBN} with its arguments, if any, in an environment
1591 of your choice. If you are doing native debugging, you may redirect
1592 your program's input and output, debug an already running process, or
1593 kill a child process.
1596 * Compilation:: Compiling for debugging
1597 * Starting:: Starting your program
1598 * Arguments:: Your program's arguments
1599 * Environment:: Your program's environment
1601 * Working Directory:: Your program's working directory
1602 * Input/Output:: Your program's input and output
1603 * Attach:: Debugging an already-running process
1604 * Kill Process:: Killing the child process
1606 * Threads:: Debugging programs with multiple threads
1607 * Processes:: Debugging programs with multiple processes
1611 @section Compiling for debugging
1613 In order to debug a program effectively, you need to generate
1614 debugging information when you compile it. This debugging information
1615 is stored in the object file; it describes the data type of each
1616 variable or function and the correspondence between source line numbers
1617 and addresses in the executable code.
1619 To request debugging information, specify the @samp{-g} option when you run
1622 Most compilers do not include information about preprocessor macros in
1623 the debugging information if you specify the @option{-g} flag alone,
1624 because this information is rather large. Version 3.1 of @value{NGCC},
1625 the @sc{gnu} C compiler, provides macro information if you specify the
1626 options @option{-gdwarf-2} and @option{-g3}; the former option requests
1627 debugging information in the Dwarf 2 format, and the latter requests
1628 ``extra information''. In the future, we hope to find more compact ways
1629 to represent macro information, so that it can be included with
1632 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1633 options together. Using those compilers, you cannot generate optimized
1634 executables containing debugging information.
1636 @value{NGCC}, the @sc{gnu} C compiler, supports @samp{-g} with or
1637 without @samp{-O}, making it possible to debug optimized code. We
1638 recommend that you @emph{always} use @samp{-g} whenever you compile a
1639 program. You may think your program is correct, but there is no sense
1640 in pushing your luck.
1642 @cindex optimized code, debugging
1643 @cindex debugging optimized code
1644 When you debug a program compiled with @samp{-g -O}, remember that the
1645 optimizer is rearranging your code; the debugger shows you what is
1646 really there. Do not be too surprised when the execution path does not
1647 exactly match your source file! An extreme example: if you define a
1648 variable, but never use it, @value{GDBN} never sees that
1649 variable---because the compiler optimizes it out of existence.
1651 Some things do not work as well with @samp{-g -O} as with just
1652 @samp{-g}, particularly on machines with instruction scheduling. If in
1653 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1654 please report it to us as a bug (including a test case!).
1656 Older versions of the @sc{gnu} C compiler permitted a variant option
1657 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1658 format; if your @sc{gnu} C compiler has this option, do not use it.
1662 @section Starting your program
1668 @kindex r @r{(@code{run})}
1671 Use the @code{run} command to start your program under @value{GDBN}.
1672 You must first specify the program name (except on VxWorks) with an
1673 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1674 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1675 (@pxref{Files, ,Commands to specify files}).
1679 If you are running your program in an execution environment that
1680 supports processes, @code{run} creates an inferior process and makes
1681 that process run your program. (In environments without processes,
1682 @code{run} jumps to the start of your program.)
1684 The execution of a program is affected by certain information it
1685 receives from its superior. @value{GDBN} provides ways to specify this
1686 information, which you must do @emph{before} starting your program. (You
1687 can change it after starting your program, but such changes only affect
1688 your program the next time you start it.) This information may be
1689 divided into four categories:
1692 @item The @emph{arguments.}
1693 Specify the arguments to give your program as the arguments of the
1694 @code{run} command. If a shell is available on your target, the shell
1695 is used to pass the arguments, so that you may use normal conventions
1696 (such as wildcard expansion or variable substitution) in describing
1698 In Unix systems, you can control which shell is used with the
1699 @code{SHELL} environment variable.
1700 @xref{Arguments, ,Your program's arguments}.
1702 @item The @emph{environment.}
1703 Your program normally inherits its environment from @value{GDBN}, but you can
1704 use the @value{GDBN} commands @code{set environment} and @code{unset
1705 environment} to change parts of the environment that affect
1706 your program. @xref{Environment, ,Your program's environment}.
1708 @item The @emph{working directory.}
1709 Your program inherits its working directory from @value{GDBN}. You can set
1710 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1711 @xref{Working Directory, ,Your program's working directory}.
1713 @item The @emph{standard input and output.}
1714 Your program normally uses the same device for standard input and
1715 standard output as @value{GDBN} is using. You can redirect input and output
1716 in the @code{run} command line, or you can use the @code{tty} command to
1717 set a different device for your program.
1718 @xref{Input/Output, ,Your program's input and output}.
1721 @emph{Warning:} While input and output redirection work, you cannot use
1722 pipes to pass the output of the program you are debugging to another
1723 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1727 When you issue the @code{run} command, your program begins to execute
1728 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1729 of how to arrange for your program to stop. Once your program has
1730 stopped, you may call functions in your program, using the @code{print}
1731 or @code{call} commands. @xref{Data, ,Examining Data}.
1733 If the modification time of your symbol file has changed since the last
1734 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1735 table, and reads it again. When it does this, @value{GDBN} tries to retain
1736 your current breakpoints.
1739 @section Your program's arguments
1741 @cindex arguments (to your program)
1742 The arguments to your program can be specified by the arguments of the
1744 They are passed to a shell, which expands wildcard characters and
1745 performs redirection of I/O, and thence to your program. Your
1746 @code{SHELL} environment variable (if it exists) specifies what shell
1747 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
1748 the default shell (@file{/bin/sh} on Unix).
1750 On non-Unix systems, the program is usually invoked directly by
1751 @value{GDBN}, which emulates I/O redirection via the appropriate system
1752 calls, and the wildcard characters are expanded by the startup code of
1753 the program, not by the shell.
1755 @code{run} with no arguments uses the same arguments used by the previous
1756 @code{run}, or those set by the @code{set args} command.
1761 Specify the arguments to be used the next time your program is run. If
1762 @code{set args} has no arguments, @code{run} executes your program
1763 with no arguments. Once you have run your program with arguments,
1764 using @code{set args} before the next @code{run} is the only way to run
1765 it again without arguments.
1769 Show the arguments to give your program when it is started.
1773 @section Your program's environment
1775 @cindex environment (of your program)
1776 The @dfn{environment} consists of a set of environment variables and
1777 their values. Environment variables conventionally record such things as
1778 your user name, your home directory, your terminal type, and your search
1779 path for programs to run. Usually you set up environment variables with
1780 the shell and they are inherited by all the other programs you run. When
1781 debugging, it can be useful to try running your program with a modified
1782 environment without having to start @value{GDBN} over again.
1786 @item path @var{directory}
1787 Add @var{directory} to the front of the @code{PATH} environment variable
1788 (the search path for executables) that will be passed to your program.
1789 The value of @code{PATH} used by @value{GDBN} does not change.
1790 You may specify several directory names, separated by whitespace or by a
1791 system-dependent separator character (@samp{:} on Unix, @samp{;} on
1792 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
1793 is moved to the front, so it is searched sooner.
1795 You can use the string @samp{$cwd} to refer to whatever is the current
1796 working directory at the time @value{GDBN} searches the path. If you
1797 use @samp{.} instead, it refers to the directory where you executed the
1798 @code{path} command. @value{GDBN} replaces @samp{.} in the
1799 @var{directory} argument (with the current path) before adding
1800 @var{directory} to the search path.
1801 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1802 @c document that, since repeating it would be a no-op.
1806 Display the list of search paths for executables (the @code{PATH}
1807 environment variable).
1809 @kindex show environment
1810 @item show environment @r{[}@var{varname}@r{]}
1811 Print the value of environment variable @var{varname} to be given to
1812 your program when it starts. If you do not supply @var{varname},
1813 print the names and values of all environment variables to be given to
1814 your program. You can abbreviate @code{environment} as @code{env}.
1816 @kindex set environment
1817 @item set environment @var{varname} @r{[}=@var{value}@r{]}
1818 Set environment variable @var{varname} to @var{value}. The value
1819 changes for your program only, not for @value{GDBN} itself. @var{value} may
1820 be any string; the values of environment variables are just strings, and
1821 any interpretation is supplied by your program itself. The @var{value}
1822 parameter is optional; if it is eliminated, the variable is set to a
1824 @c "any string" here does not include leading, trailing
1825 @c blanks. Gnu asks: does anyone care?
1827 For example, this command:
1834 tells the debugged program, when subsequently run, that its user is named
1835 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1836 are not actually required.)
1838 @kindex unset environment
1839 @item unset environment @var{varname}
1840 Remove variable @var{varname} from the environment to be passed to your
1841 program. This is different from @samp{set env @var{varname} =};
1842 @code{unset environment} removes the variable from the environment,
1843 rather than assigning it an empty value.
1846 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
1848 by your @code{SHELL} environment variable if it exists (or
1849 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1850 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1851 @file{.bashrc} for BASH---any variables you set in that file affect
1852 your program. You may wish to move setting of environment variables to
1853 files that are only run when you sign on, such as @file{.login} or
1856 @node Working Directory
1857 @section Your program's working directory
1859 @cindex working directory (of your program)
1860 Each time you start your program with @code{run}, it inherits its
1861 working directory from the current working directory of @value{GDBN}.
1862 The @value{GDBN} working directory is initially whatever it inherited
1863 from its parent process (typically the shell), but you can specify a new
1864 working directory in @value{GDBN} with the @code{cd} command.
1866 The @value{GDBN} working directory also serves as a default for the commands
1867 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1872 @item cd @var{directory}
1873 Set the @value{GDBN} working directory to @var{directory}.
1877 Print the @value{GDBN} working directory.
1881 @section Your program's input and output
1886 By default, the program you run under @value{GDBN} does input and output to
1887 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
1888 to its own terminal modes to interact with you, but it records the terminal
1889 modes your program was using and switches back to them when you continue
1890 running your program.
1893 @kindex info terminal
1895 Displays information recorded by @value{GDBN} about the terminal modes your
1899 You can redirect your program's input and/or output using shell
1900 redirection with the @code{run} command. For example,
1907 starts your program, diverting its output to the file @file{outfile}.
1910 @cindex controlling terminal
1911 Another way to specify where your program should do input and output is
1912 with the @code{tty} command. This command accepts a file name as
1913 argument, and causes this file to be the default for future @code{run}
1914 commands. It also resets the controlling terminal for the child
1915 process, for future @code{run} commands. For example,
1922 directs that processes started with subsequent @code{run} commands
1923 default to do input and output on the terminal @file{/dev/ttyb} and have
1924 that as their controlling terminal.
1926 An explicit redirection in @code{run} overrides the @code{tty} command's
1927 effect on the input/output device, but not its effect on the controlling
1930 When you use the @code{tty} command or redirect input in the @code{run}
1931 command, only the input @emph{for your program} is affected. The input
1932 for @value{GDBN} still comes from your terminal.
1935 @section Debugging an already-running process
1940 @item attach @var{process-id}
1941 This command attaches to a running process---one that was started
1942 outside @value{GDBN}. (@code{info files} shows your active
1943 targets.) The command takes as argument a process ID. The usual way to
1944 find out the process-id of a Unix process is with the @code{ps} utility,
1945 or with the @samp{jobs -l} shell command.
1947 @code{attach} does not repeat if you press @key{RET} a second time after
1948 executing the command.
1951 To use @code{attach}, your program must be running in an environment
1952 which supports processes; for example, @code{attach} does not work for
1953 programs on bare-board targets that lack an operating system. You must
1954 also have permission to send the process a signal.
1956 When you use @code{attach}, the debugger finds the program running in
1957 the process first by looking in the current working directory, then (if
1958 the program is not found) by using the source file search path
1959 (@pxref{Source Path, ,Specifying source directories}). You can also use
1960 the @code{file} command to load the program. @xref{Files, ,Commands to
1963 The first thing @value{GDBN} does after arranging to debug the specified
1964 process is to stop it. You can examine and modify an attached process
1965 with all the @value{GDBN} commands that are ordinarily available when
1966 you start processes with @code{run}. You can insert breakpoints; you
1967 can step and continue; you can modify storage. If you would rather the
1968 process continue running, you may use the @code{continue} command after
1969 attaching @value{GDBN} to the process.
1974 When you have finished debugging the attached process, you can use the
1975 @code{detach} command to release it from @value{GDBN} control. Detaching
1976 the process continues its execution. After the @code{detach} command,
1977 that process and @value{GDBN} become completely independent once more, and you
1978 are ready to @code{attach} another process or start one with @code{run}.
1979 @code{detach} does not repeat if you press @key{RET} again after
1980 executing the command.
1983 If you exit @value{GDBN} or use the @code{run} command while you have an
1984 attached process, you kill that process. By default, @value{GDBN} asks
1985 for confirmation if you try to do either of these things; you can
1986 control whether or not you need to confirm by using the @code{set
1987 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
1991 @section Killing the child process
1996 Kill the child process in which your program is running under @value{GDBN}.
1999 This command is useful if you wish to debug a core dump instead of a
2000 running process. @value{GDBN} ignores any core dump file while your program
2003 On some operating systems, a program cannot be executed outside @value{GDBN}
2004 while you have breakpoints set on it inside @value{GDBN}. You can use the
2005 @code{kill} command in this situation to permit running your program
2006 outside the debugger.
2008 The @code{kill} command is also useful if you wish to recompile and
2009 relink your program, since on many systems it is impossible to modify an
2010 executable file while it is running in a process. In this case, when you
2011 next type @code{run}, @value{GDBN} notices that the file has changed, and
2012 reads the symbol table again (while trying to preserve your current
2013 breakpoint settings).
2016 @section Debugging programs with multiple threads
2018 @cindex threads of execution
2019 @cindex multiple threads
2020 @cindex switching threads
2021 In some operating systems, such as HP-UX and Solaris, a single program
2022 may have more than one @dfn{thread} of execution. The precise semantics
2023 of threads differ from one operating system to another, but in general
2024 the threads of a single program are akin to multiple processes---except
2025 that they share one address space (that is, they can all examine and
2026 modify the same variables). On the other hand, each thread has its own
2027 registers and execution stack, and perhaps private memory.
2029 @value{GDBN} provides these facilities for debugging multi-thread
2033 @item automatic notification of new threads
2034 @item @samp{thread @var{threadno}}, a command to switch among threads
2035 @item @samp{info threads}, a command to inquire about existing threads
2036 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2037 a command to apply a command to a list of threads
2038 @item thread-specific breakpoints
2042 @emph{Warning:} These facilities are not yet available on every
2043 @value{GDBN} configuration where the operating system supports threads.
2044 If your @value{GDBN} does not support threads, these commands have no
2045 effect. For example, a system without thread support shows no output
2046 from @samp{info threads}, and always rejects the @code{thread} command,
2050 (@value{GDBP}) info threads
2051 (@value{GDBP}) thread 1
2052 Thread ID 1 not known. Use the "info threads" command to
2053 see the IDs of currently known threads.
2055 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2056 @c doesn't support threads"?
2059 @cindex focus of debugging
2060 @cindex current thread
2061 The @value{GDBN} thread debugging facility allows you to observe all
2062 threads while your program runs---but whenever @value{GDBN} takes
2063 control, one thread in particular is always the focus of debugging.
2064 This thread is called the @dfn{current thread}. Debugging commands show
2065 program information from the perspective of the current thread.
2067 @cindex @code{New} @var{systag} message
2068 @cindex thread identifier (system)
2069 @c FIXME-implementors!! It would be more helpful if the [New...] message
2070 @c included GDB's numeric thread handle, so you could just go to that
2071 @c thread without first checking `info threads'.
2072 Whenever @value{GDBN} detects a new thread in your program, it displays
2073 the target system's identification for the thread with a message in the
2074 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2075 whose form varies depending on the particular system. For example, on
2076 LynxOS, you might see
2079 [New process 35 thread 27]
2083 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2084 the @var{systag} is simply something like @samp{process 368}, with no
2087 @c FIXME!! (1) Does the [New...] message appear even for the very first
2088 @c thread of a program, or does it only appear for the
2089 @c second---i.e.@: when it becomes obvious we have a multithread
2091 @c (2) *Is* there necessarily a first thread always? Or do some
2092 @c multithread systems permit starting a program with multiple
2093 @c threads ab initio?
2095 @cindex thread number
2096 @cindex thread identifier (GDB)
2097 For debugging purposes, @value{GDBN} associates its own thread
2098 number---always a single integer---with each thread in your program.
2101 @kindex info threads
2103 Display a summary of all threads currently in your
2104 program. @value{GDBN} displays for each thread (in this order):
2107 @item the thread number assigned by @value{GDBN}
2109 @item the target system's thread identifier (@var{systag})
2111 @item the current stack frame summary for that thread
2115 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2116 indicates the current thread.
2120 @c end table here to get a little more width for example
2123 (@value{GDBP}) info threads
2124 3 process 35 thread 27 0x34e5 in sigpause ()
2125 2 process 35 thread 23 0x34e5 in sigpause ()
2126 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2132 @cindex thread number
2133 @cindex thread identifier (GDB)
2134 For debugging purposes, @value{GDBN} associates its own thread
2135 number---a small integer assigned in thread-creation order---with each
2136 thread in your program.
2138 @cindex @code{New} @var{systag} message, on HP-UX
2139 @cindex thread identifier (system), on HP-UX
2140 @c FIXME-implementors!! It would be more helpful if the [New...] message
2141 @c included GDB's numeric thread handle, so you could just go to that
2142 @c thread without first checking `info threads'.
2143 Whenever @value{GDBN} detects a new thread in your program, it displays
2144 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2145 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2146 whose form varies depending on the particular system. For example, on
2150 [New thread 2 (system thread 26594)]
2154 when @value{GDBN} notices a new thread.
2157 @kindex info threads
2159 Display a summary of all threads currently in your
2160 program. @value{GDBN} displays for each thread (in this order):
2163 @item the thread number assigned by @value{GDBN}
2165 @item the target system's thread identifier (@var{systag})
2167 @item the current stack frame summary for that thread
2171 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2172 indicates the current thread.
2176 @c end table here to get a little more width for example
2179 (@value{GDBP}) info threads
2180 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2182 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2183 from /usr/lib/libc.2
2184 1 system thread 27905 0x7b003498 in _brk () \@*
2185 from /usr/lib/libc.2
2189 @kindex thread @var{threadno}
2190 @item thread @var{threadno}
2191 Make thread number @var{threadno} the current thread. The command
2192 argument @var{threadno} is the internal @value{GDBN} thread number, as
2193 shown in the first field of the @samp{info threads} display.
2194 @value{GDBN} responds by displaying the system identifier of the thread
2195 you selected, and its current stack frame summary:
2198 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2199 (@value{GDBP}) thread 2
2200 [Switching to process 35 thread 23]
2201 0x34e5 in sigpause ()
2205 As with the @samp{[New @dots{}]} message, the form of the text after
2206 @samp{Switching to} depends on your system's conventions for identifying
2209 @kindex thread apply
2210 @item thread apply [@var{threadno}] [@var{all}] @var{args}
2211 The @code{thread apply} command allows you to apply a command to one or
2212 more threads. Specify the numbers of the threads that you want affected
2213 with the command argument @var{threadno}. @var{threadno} is the internal
2214 @value{GDBN} thread number, as shown in the first field of the @samp{info
2215 threads} display. To apply a command to all threads, use
2216 @code{thread apply all} @var{args}.
2219 @cindex automatic thread selection
2220 @cindex switching threads automatically
2221 @cindex threads, automatic switching
2222 Whenever @value{GDBN} stops your program, due to a breakpoint or a
2223 signal, it automatically selects the thread where that breakpoint or
2224 signal happened. @value{GDBN} alerts you to the context switch with a
2225 message of the form @samp{[Switching to @var{systag}]} to identify the
2228 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
2229 more information about how @value{GDBN} behaves when you stop and start
2230 programs with multiple threads.
2232 @xref{Set Watchpoints,,Setting watchpoints}, for information about
2233 watchpoints in programs with multiple threads.
2236 @section Debugging programs with multiple processes
2238 @cindex fork, debugging programs which call
2239 @cindex multiple processes
2240 @cindex processes, multiple
2241 On most systems, @value{GDBN} has no special support for debugging
2242 programs which create additional processes using the @code{fork}
2243 function. When a program forks, @value{GDBN} will continue to debug the
2244 parent process and the child process will run unimpeded. If you have
2245 set a breakpoint in any code which the child then executes, the child
2246 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2247 will cause it to terminate.
2249 However, if you want to debug the child process there is a workaround
2250 which isn't too painful. Put a call to @code{sleep} in the code which
2251 the child process executes after the fork. It may be useful to sleep
2252 only if a certain environment variable is set, or a certain file exists,
2253 so that the delay need not occur when you don't want to run @value{GDBN}
2254 on the child. While the child is sleeping, use the @code{ps} program to
2255 get its process ID. Then tell @value{GDBN} (a new invocation of
2256 @value{GDBN} if you are also debugging the parent process) to attach to
2257 the child process (@pxref{Attach}). From that point on you can debug
2258 the child process just like any other process which you attached to.
2260 On HP-UX (11.x and later only?), @value{GDBN} provides support for
2261 debugging programs that create additional processes using the
2262 @code{fork} or @code{vfork} function.
2264 By default, when a program forks, @value{GDBN} will continue to debug
2265 the parent process and the child process will run unimpeded.
2267 If you want to follow the child process instead of the parent process,
2268 use the command @w{@code{set follow-fork-mode}}.
2271 @kindex set follow-fork-mode
2272 @item set follow-fork-mode @var{mode}
2273 Set the debugger response to a program call of @code{fork} or
2274 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2275 process. The @var{mode} can be:
2279 The original process is debugged after a fork. The child process runs
2280 unimpeded. This is the default.
2283 The new process is debugged after a fork. The parent process runs
2287 The debugger will ask for one of the above choices.
2290 @item show follow-fork-mode
2291 Display the current debugger response to a @code{fork} or @code{vfork} call.
2294 If you ask to debug a child process and a @code{vfork} is followed by an
2295 @code{exec}, @value{GDBN} executes the new target up to the first
2296 breakpoint in the new target. If you have a breakpoint set on
2297 @code{main} in your original program, the breakpoint will also be set on
2298 the child process's @code{main}.
2300 When a child process is spawned by @code{vfork}, you cannot debug the
2301 child or parent until an @code{exec} call completes.
2303 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2304 call executes, the new target restarts. To restart the parent process,
2305 use the @code{file} command with the parent executable name as its
2308 You can use the @code{catch} command to make @value{GDBN} stop whenever
2309 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2310 Catchpoints, ,Setting catchpoints}.
2313 @chapter Stopping and Continuing
2315 The principal purposes of using a debugger are so that you can stop your
2316 program before it terminates; or so that, if your program runs into
2317 trouble, you can investigate and find out why.
2319 Inside @value{GDBN}, your program may stop for any of several reasons,
2320 such as a signal, a breakpoint, or reaching a new line after a
2321 @value{GDBN} command such as @code{step}. You may then examine and
2322 change variables, set new breakpoints or remove old ones, and then
2323 continue execution. Usually, the messages shown by @value{GDBN} provide
2324 ample explanation of the status of your program---but you can also
2325 explicitly request this information at any time.
2328 @kindex info program
2330 Display information about the status of your program: whether it is
2331 running or not, what process it is, and why it stopped.
2335 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2336 * Continuing and Stepping:: Resuming execution
2338 * Thread Stops:: Stopping and starting multi-thread programs
2342 @section Breakpoints, watchpoints, and catchpoints
2345 A @dfn{breakpoint} makes your program stop whenever a certain point in
2346 the program is reached. For each breakpoint, you can add conditions to
2347 control in finer detail whether your program stops. You can set
2348 breakpoints with the @code{break} command and its variants (@pxref{Set
2349 Breaks, ,Setting breakpoints}), to specify the place where your program
2350 should stop by line number, function name or exact address in the
2353 In HP-UX, SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can set
2354 breakpoints in shared libraries before the executable is run. There is
2355 a minor limitation on HP-UX systems: you must wait until the executable
2356 is run in order to set breakpoints in shared library routines that are
2357 not called directly by the program (for example, routines that are
2358 arguments in a @code{pthread_create} call).
2361 @cindex memory tracing
2362 @cindex breakpoint on memory address
2363 @cindex breakpoint on variable modification
2364 A @dfn{watchpoint} is a special breakpoint that stops your program
2365 when the value of an expression changes. You must use a different
2366 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2367 watchpoints}), but aside from that, you can manage a watchpoint like
2368 any other breakpoint: you enable, disable, and delete both breakpoints
2369 and watchpoints using the same commands.
2371 You can arrange to have values from your program displayed automatically
2372 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2376 @cindex breakpoint on events
2377 A @dfn{catchpoint} is another special breakpoint that stops your program
2378 when a certain kind of event occurs, such as the throwing of a C@t{++}
2379 exception or the loading of a library. As with watchpoints, you use a
2380 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
2381 catchpoints}), but aside from that, you can manage a catchpoint like any
2382 other breakpoint. (To stop when your program receives a signal, use the
2383 @code{handle} command; see @ref{Signals, ,Signals}.)
2385 @cindex breakpoint numbers
2386 @cindex numbers for breakpoints
2387 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
2388 catchpoint when you create it; these numbers are successive integers
2389 starting with one. In many of the commands for controlling various
2390 features of breakpoints you use the breakpoint number to say which
2391 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
2392 @dfn{disabled}; if disabled, it has no effect on your program until you
2395 @cindex breakpoint ranges
2396 @cindex ranges of breakpoints
2397 Some @value{GDBN} commands accept a range of breakpoints on which to
2398 operate. A breakpoint range is either a single breakpoint number, like
2399 @samp{5}, or two such numbers, in increasing order, separated by a
2400 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
2401 all breakpoint in that range are operated on.
2404 * Set Breaks:: Setting breakpoints
2405 * Set Watchpoints:: Setting watchpoints
2406 * Set Catchpoints:: Setting catchpoints
2407 * Delete Breaks:: Deleting breakpoints
2408 * Disabling:: Disabling breakpoints
2409 * Conditions:: Break conditions
2410 * Break Commands:: Breakpoint command lists
2411 * Breakpoint Menus:: Breakpoint menus
2412 * Error in Breakpoints:: ``Cannot insert breakpoints''
2416 @subsection Setting breakpoints
2418 @c FIXME LMB what does GDB do if no code on line of breakpt?
2419 @c consider in particular declaration with/without initialization.
2421 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2424 @kindex b @r{(@code{break})}
2425 @vindex $bpnum@r{, convenience variable}
2426 @cindex latest breakpoint
2427 Breakpoints are set with the @code{break} command (abbreviated
2428 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2429 number of the breakpoint you've set most recently; see @ref{Convenience
2430 Vars,, Convenience variables}, for a discussion of what you can do with
2431 convenience variables.
2433 You have several ways to say where the breakpoint should go.
2436 @item break @var{function}
2437 Set a breakpoint at entry to function @var{function}.
2438 When using source languages that permit overloading of symbols, such as
2439 C@t{++}, @var{function} may refer to more than one possible place to break.
2440 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2442 @item break +@var{offset}
2443 @itemx break -@var{offset}
2444 Set a breakpoint some number of lines forward or back from the position
2445 at which execution stopped in the currently selected @dfn{stack frame}.
2446 (@xref{Frames, ,Frames}, for a description of stack frames.)
2448 @item break @var{linenum}
2449 Set a breakpoint at line @var{linenum} in the current source file.
2450 The current source file is the last file whose source text was printed.
2451 The breakpoint will stop your program just before it executes any of the
2454 @item break @var{filename}:@var{linenum}
2455 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2457 @item break @var{filename}:@var{function}
2458 Set a breakpoint at entry to function @var{function} found in file
2459 @var{filename}. Specifying a file name as well as a function name is
2460 superfluous except when multiple files contain similarly named
2463 @item break *@var{address}
2464 Set a breakpoint at address @var{address}. You can use this to set
2465 breakpoints in parts of your program which do not have debugging
2466 information or source files.
2469 When called without any arguments, @code{break} sets a breakpoint at
2470 the next instruction to be executed in the selected stack frame
2471 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2472 innermost, this makes your program stop as soon as control
2473 returns to that frame. This is similar to the effect of a
2474 @code{finish} command in the frame inside the selected frame---except
2475 that @code{finish} does not leave an active breakpoint. If you use
2476 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2477 the next time it reaches the current location; this may be useful
2480 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2481 least one instruction has been executed. If it did not do this, you
2482 would be unable to proceed past a breakpoint without first disabling the
2483 breakpoint. This rule applies whether or not the breakpoint already
2484 existed when your program stopped.
2486 @item break @dots{} if @var{cond}
2487 Set a breakpoint with condition @var{cond}; evaluate the expression
2488 @var{cond} each time the breakpoint is reached, and stop only if the
2489 value is nonzero---that is, if @var{cond} evaluates as true.
2490 @samp{@dots{}} stands for one of the possible arguments described
2491 above (or no argument) specifying where to break. @xref{Conditions,
2492 ,Break conditions}, for more information on breakpoint conditions.
2495 @item tbreak @var{args}
2496 Set a breakpoint enabled only for one stop. @var{args} are the
2497 same as for the @code{break} command, and the breakpoint is set in the same
2498 way, but the breakpoint is automatically deleted after the first time your
2499 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2502 @item hbreak @var{args}
2503 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2504 @code{break} command and the breakpoint is set in the same way, but the
2505 breakpoint requires hardware support and some target hardware may not
2506 have this support. The main purpose of this is EPROM/ROM code
2507 debugging, so you can set a breakpoint at an instruction without
2508 changing the instruction. This can be used with the new trap-generation
2509 provided by SPARClite DSU and some x86-based targets. These targets
2510 will generate traps when a program accesses some data or instruction
2511 address that is assigned to the debug registers. However the hardware
2512 breakpoint registers can take a limited number of breakpoints. For
2513 example, on the DSU, only two data breakpoints can be set at a time, and
2514 @value{GDBN} will reject this command if more than two are used. Delete
2515 or disable unused hardware breakpoints before setting new ones
2516 (@pxref{Disabling, ,Disabling}). @xref{Conditions, ,Break conditions}.
2519 @item thbreak @var{args}
2520 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2521 are the same as for the @code{hbreak} command and the breakpoint is set in
2522 the same way. However, like the @code{tbreak} command,
2523 the breakpoint is automatically deleted after the
2524 first time your program stops there. Also, like the @code{hbreak}
2525 command, the breakpoint requires hardware support and some target hardware
2526 may not have this support. @xref{Disabling, ,Disabling breakpoints}.
2527 See also @ref{Conditions, ,Break conditions}.
2530 @cindex regular expression
2531 @item rbreak @var{regex}
2532 Set breakpoints on all functions matching the regular expression
2533 @var{regex}. This command sets an unconditional breakpoint on all
2534 matches, printing a list of all breakpoints it set. Once these
2535 breakpoints are set, they are treated just like the breakpoints set with
2536 the @code{break} command. You can delete them, disable them, or make
2537 them conditional the same way as any other breakpoint.
2539 The syntax of the regular expression is the standard one used with tools
2540 like @file{grep}. Note that this is different from the syntax used by
2541 shells, so for instance @code{foo*} matches all functions that include
2542 an @code{fo} followed by zero or more @code{o}s. There is an implicit
2543 @code{.*} leading and trailing the regular expression you supply, so to
2544 match only functions that begin with @code{foo}, use @code{^foo}.
2546 When debugging C@t{++} programs, @code{rbreak} is useful for setting
2547 breakpoints on overloaded functions that are not members of any special
2550 @kindex info breakpoints
2551 @cindex @code{$_} and @code{info breakpoints}
2552 @item info breakpoints @r{[}@var{n}@r{]}
2553 @itemx info break @r{[}@var{n}@r{]}
2554 @itemx info watchpoints @r{[}@var{n}@r{]}
2555 Print a table of all breakpoints, watchpoints, and catchpoints set and
2556 not deleted, with the following columns for each breakpoint:
2559 @item Breakpoint Numbers
2561 Breakpoint, watchpoint, or catchpoint.
2563 Whether the breakpoint is marked to be disabled or deleted when hit.
2564 @item Enabled or Disabled
2565 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2566 that are not enabled.
2568 Where the breakpoint is in your program, as a memory address.
2570 Where the breakpoint is in the source for your program, as a file and
2575 If a breakpoint is conditional, @code{info break} shows the condition on
2576 the line following the affected breakpoint; breakpoint commands, if any,
2577 are listed after that.
2580 @code{info break} with a breakpoint
2581 number @var{n} as argument lists only that breakpoint. The
2582 convenience variable @code{$_} and the default examining-address for
2583 the @code{x} command are set to the address of the last breakpoint
2584 listed (@pxref{Memory, ,Examining memory}).
2587 @code{info break} displays a count of the number of times the breakpoint
2588 has been hit. This is especially useful in conjunction with the
2589 @code{ignore} command. You can ignore a large number of breakpoint
2590 hits, look at the breakpoint info to see how many times the breakpoint
2591 was hit, and then run again, ignoring one less than that number. This
2592 will get you quickly to the last hit of that breakpoint.
2595 @value{GDBN} allows you to set any number of breakpoints at the same place in
2596 your program. There is nothing silly or meaningless about this. When
2597 the breakpoints are conditional, this is even useful
2598 (@pxref{Conditions, ,Break conditions}).
2600 @cindex negative breakpoint numbers
2601 @cindex internal @value{GDBN} breakpoints
2602 @value{GDBN} itself sometimes sets breakpoints in your program for
2603 special purposes, such as proper handling of @code{longjmp} (in C
2604 programs). These internal breakpoints are assigned negative numbers,
2605 starting with @code{-1}; @samp{info breakpoints} does not display them.
2606 You can see these breakpoints with the @value{GDBN} maintenance command
2607 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
2610 @node Set Watchpoints
2611 @subsection Setting watchpoints
2613 @cindex setting watchpoints
2614 @cindex software watchpoints
2615 @cindex hardware watchpoints
2616 You can use a watchpoint to stop execution whenever the value of an
2617 expression changes, without having to predict a particular place where
2620 Depending on your system, watchpoints may be implemented in software or
2621 hardware. @value{GDBN} does software watchpointing by single-stepping your
2622 program and testing the variable's value each time, which is hundreds of
2623 times slower than normal execution. (But this may still be worth it, to
2624 catch errors where you have no clue what part of your program is the
2627 On some systems, such as HP-UX, @sc{gnu}/Linux and some other x86-based targets,
2628 @value{GDBN} includes support for
2629 hardware watchpoints, which do not slow down the running of your
2634 @item watch @var{expr}
2635 Set a watchpoint for an expression. @value{GDBN} will break when @var{expr}
2636 is written into by the program and its value changes.
2639 @item rwatch @var{expr}
2640 Set a watchpoint that will break when watch @var{expr} is read by the program.
2643 @item awatch @var{expr}
2644 Set a watchpoint that will break when @var{expr} is either read or written into
2647 @kindex info watchpoints
2648 @item info watchpoints
2649 This command prints a list of watchpoints, breakpoints, and catchpoints;
2650 it is the same as @code{info break}.
2653 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
2654 watchpoints execute very quickly, and the debugger reports a change in
2655 value at the exact instruction where the change occurs. If @value{GDBN}
2656 cannot set a hardware watchpoint, it sets a software watchpoint, which
2657 executes more slowly and reports the change in value at the next
2658 statement, not the instruction, after the change occurs.
2660 When you issue the @code{watch} command, @value{GDBN} reports
2663 Hardware watchpoint @var{num}: @var{expr}
2667 if it was able to set a hardware watchpoint.
2669 Currently, the @code{awatch} and @code{rwatch} commands can only set
2670 hardware watchpoints, because accesses to data that don't change the
2671 value of the watched expression cannot be detected without examining
2672 every instruction as it is being executed, and @value{GDBN} does not do
2673 that currently. If @value{GDBN} finds that it is unable to set a
2674 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
2675 will print a message like this:
2678 Expression cannot be implemented with read/access watchpoint.
2681 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
2682 data type of the watched expression is wider than what a hardware
2683 watchpoint on the target machine can handle. For example, some systems
2684 can only watch regions that are up to 4 bytes wide; on such systems you
2685 cannot set hardware watchpoints for an expression that yields a
2686 double-precision floating-point number (which is typically 8 bytes
2687 wide). As a work-around, it might be possible to break the large region
2688 into a series of smaller ones and watch them with separate watchpoints.
2690 If you set too many hardware watchpoints, @value{GDBN} might be unable
2691 to insert all of them when you resume the execution of your program.
2692 Since the precise number of active watchpoints is unknown until such
2693 time as the program is about to be resumed, @value{GDBN} might not be
2694 able to warn you about this when you set the watchpoints, and the
2695 warning will be printed only when the program is resumed:
2698 Hardware watchpoint @var{num}: Could not insert watchpoint
2702 If this happens, delete or disable some of the watchpoints.
2704 The SPARClite DSU will generate traps when a program accesses some data
2705 or instruction address that is assigned to the debug registers. For the
2706 data addresses, DSU facilitates the @code{watch} command. However the
2707 hardware breakpoint registers can only take two data watchpoints, and
2708 both watchpoints must be the same kind. For example, you can set two
2709 watchpoints with @code{watch} commands, two with @code{rwatch} commands,
2710 @strong{or} two with @code{awatch} commands, but you cannot set one
2711 watchpoint with one command and the other with a different command.
2712 @value{GDBN} will reject the command if you try to mix watchpoints.
2713 Delete or disable unused watchpoint commands before setting new ones.
2715 If you call a function interactively using @code{print} or @code{call},
2716 any watchpoints you have set will be inactive until @value{GDBN} reaches another
2717 kind of breakpoint or the call completes.
2719 @value{GDBN} automatically deletes watchpoints that watch local
2720 (automatic) variables, or expressions that involve such variables, when
2721 they go out of scope, that is, when the execution leaves the block in
2722 which these variables were defined. In particular, when the program
2723 being debugged terminates, @emph{all} local variables go out of scope,
2724 and so only watchpoints that watch global variables remain set. If you
2725 rerun the program, you will need to set all such watchpoints again. One
2726 way of doing that would be to set a code breakpoint at the entry to the
2727 @code{main} function and when it breaks, set all the watchpoints.
2730 @cindex watchpoints and threads
2731 @cindex threads and watchpoints
2732 @emph{Warning:} In multi-thread programs, watchpoints have only limited
2733 usefulness. With the current watchpoint implementation, @value{GDBN}
2734 can only watch the value of an expression @emph{in a single thread}. If
2735 you are confident that the expression can only change due to the current
2736 thread's activity (and if you are also confident that no other thread
2737 can become current), then you can use watchpoints as usual. However,
2738 @value{GDBN} may not notice when a non-current thread's activity changes
2741 @c FIXME: this is almost identical to the previous paragraph.
2742 @emph{HP-UX Warning:} In multi-thread programs, software watchpoints
2743 have only limited usefulness. If @value{GDBN} creates a software
2744 watchpoint, it can only watch the value of an expression @emph{in a
2745 single thread}. If you are confident that the expression can only
2746 change due to the current thread's activity (and if you are also
2747 confident that no other thread can become current), then you can use
2748 software watchpoints as usual. However, @value{GDBN} may not notice
2749 when a non-current thread's activity changes the expression. (Hardware
2750 watchpoints, in contrast, watch an expression in all threads.)
2753 @node Set Catchpoints
2754 @subsection Setting catchpoints
2755 @cindex catchpoints, setting
2756 @cindex exception handlers
2757 @cindex event handling
2759 You can use @dfn{catchpoints} to cause the debugger to stop for certain
2760 kinds of program events, such as C@t{++} exceptions or the loading of a
2761 shared library. Use the @code{catch} command to set a catchpoint.
2765 @item catch @var{event}
2766 Stop when @var{event} occurs. @var{event} can be any of the following:
2770 The throwing of a C@t{++} exception.
2774 The catching of a C@t{++} exception.
2778 A call to @code{exec}. This is currently only available for HP-UX.
2782 A call to @code{fork}. This is currently only available for HP-UX.
2786 A call to @code{vfork}. This is currently only available for HP-UX.
2789 @itemx load @var{libname}
2791 The dynamic loading of any shared library, or the loading of the library
2792 @var{libname}. This is currently only available for HP-UX.
2795 @itemx unload @var{libname}
2796 @kindex catch unload
2797 The unloading of any dynamically loaded shared library, or the unloading
2798 of the library @var{libname}. This is currently only available for HP-UX.
2801 @item tcatch @var{event}
2802 Set a catchpoint that is enabled only for one stop. The catchpoint is
2803 automatically deleted after the first time the event is caught.
2807 Use the @code{info break} command to list the current catchpoints.
2809 There are currently some limitations to C@t{++} exception handling
2810 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
2814 If you call a function interactively, @value{GDBN} normally returns
2815 control to you when the function has finished executing. If the call
2816 raises an exception, however, the call may bypass the mechanism that
2817 returns control to you and cause your program either to abort or to
2818 simply continue running until it hits a breakpoint, catches a signal
2819 that @value{GDBN} is listening for, or exits. This is the case even if
2820 you set a catchpoint for the exception; catchpoints on exceptions are
2821 disabled within interactive calls.
2824 You cannot raise an exception interactively.
2827 You cannot install an exception handler interactively.
2830 @cindex raise exceptions
2831 Sometimes @code{catch} is not the best way to debug exception handling:
2832 if you need to know exactly where an exception is raised, it is better to
2833 stop @emph{before} the exception handler is called, since that way you
2834 can see the stack before any unwinding takes place. If you set a
2835 breakpoint in an exception handler instead, it may not be easy to find
2836 out where the exception was raised.
2838 To stop just before an exception handler is called, you need some
2839 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
2840 raised by calling a library function named @code{__raise_exception}
2841 which has the following ANSI C interface:
2844 /* @var{addr} is where the exception identifier is stored.
2845 @var{id} is the exception identifier. */
2846 void __raise_exception (void **addr, void *id);
2850 To make the debugger catch all exceptions before any stack
2851 unwinding takes place, set a breakpoint on @code{__raise_exception}
2852 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2854 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2855 that depends on the value of @var{id}, you can stop your program when
2856 a specific exception is raised. You can use multiple conditional
2857 breakpoints to stop your program when any of a number of exceptions are
2862 @subsection Deleting breakpoints
2864 @cindex clearing breakpoints, watchpoints, catchpoints
2865 @cindex deleting breakpoints, watchpoints, catchpoints
2866 It is often necessary to eliminate a breakpoint, watchpoint, or
2867 catchpoint once it has done its job and you no longer want your program
2868 to stop there. This is called @dfn{deleting} the breakpoint. A
2869 breakpoint that has been deleted no longer exists; it is forgotten.
2871 With the @code{clear} command you can delete breakpoints according to
2872 where they are in your program. With the @code{delete} command you can
2873 delete individual breakpoints, watchpoints, or catchpoints by specifying
2874 their breakpoint numbers.
2876 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2877 automatically ignores breakpoints on the first instruction to be executed
2878 when you continue execution without changing the execution address.
2883 Delete any breakpoints at the next instruction to be executed in the
2884 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2885 the innermost frame is selected, this is a good way to delete a
2886 breakpoint where your program just stopped.
2888 @item clear @var{function}
2889 @itemx clear @var{filename}:@var{function}
2890 Delete any breakpoints set at entry to the function @var{function}.
2892 @item clear @var{linenum}
2893 @itemx clear @var{filename}:@var{linenum}
2894 Delete any breakpoints set at or within the code of the specified line.
2896 @cindex delete breakpoints
2898 @kindex d @r{(@code{delete})}
2899 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2900 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
2901 ranges specified as arguments. If no argument is specified, delete all
2902 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
2903 confirm off}). You can abbreviate this command as @code{d}.
2907 @subsection Disabling breakpoints
2909 @kindex disable breakpoints
2910 @kindex enable breakpoints
2911 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
2912 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
2913 it had been deleted, but remembers the information on the breakpoint so
2914 that you can @dfn{enable} it again later.
2916 You disable and enable breakpoints, watchpoints, and catchpoints with
2917 the @code{enable} and @code{disable} commands, optionally specifying one
2918 or more breakpoint numbers as arguments. Use @code{info break} or
2919 @code{info watch} to print a list of breakpoints, watchpoints, and
2920 catchpoints if you do not know which numbers to use.
2922 A breakpoint, watchpoint, or catchpoint can have any of four different
2923 states of enablement:
2927 Enabled. The breakpoint stops your program. A breakpoint set
2928 with the @code{break} command starts out in this state.
2930 Disabled. The breakpoint has no effect on your program.
2932 Enabled once. The breakpoint stops your program, but then becomes
2935 Enabled for deletion. The breakpoint stops your program, but
2936 immediately after it does so it is deleted permanently. A breakpoint
2937 set with the @code{tbreak} command starts out in this state.
2940 You can use the following commands to enable or disable breakpoints,
2941 watchpoints, and catchpoints:
2944 @kindex disable breakpoints
2946 @kindex dis @r{(@code{disable})}
2947 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2948 Disable the specified breakpoints---or all breakpoints, if none are
2949 listed. A disabled breakpoint has no effect but is not forgotten. All
2950 options such as ignore-counts, conditions and commands are remembered in
2951 case the breakpoint is enabled again later. You may abbreviate
2952 @code{disable} as @code{dis}.
2954 @kindex enable breakpoints
2956 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
2957 Enable the specified breakpoints (or all defined breakpoints). They
2958 become effective once again in stopping your program.
2960 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
2961 Enable the specified breakpoints temporarily. @value{GDBN} disables any
2962 of these breakpoints immediately after stopping your program.
2964 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
2965 Enable the specified breakpoints to work once, then die. @value{GDBN}
2966 deletes any of these breakpoints as soon as your program stops there.
2969 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
2970 @c confusing: tbreak is also initially enabled.
2971 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2972 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2973 subsequently, they become disabled or enabled only when you use one of
2974 the commands above. (The command @code{until} can set and delete a
2975 breakpoint of its own, but it does not change the state of your other
2976 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2980 @subsection Break conditions
2981 @cindex conditional breakpoints
2982 @cindex breakpoint conditions
2984 @c FIXME what is scope of break condition expr? Context where wanted?
2985 @c in particular for a watchpoint?
2986 The simplest sort of breakpoint breaks every time your program reaches a
2987 specified place. You can also specify a @dfn{condition} for a
2988 breakpoint. A condition is just a Boolean expression in your
2989 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2990 a condition evaluates the expression each time your program reaches it,
2991 and your program stops only if the condition is @emph{true}.
2993 This is the converse of using assertions for program validation; in that
2994 situation, you want to stop when the assertion is violated---that is,
2995 when the condition is false. In C, if you want to test an assertion expressed
2996 by the condition @var{assert}, you should set the condition
2997 @samp{! @var{assert}} on the appropriate breakpoint.
2999 Conditions are also accepted for watchpoints; you may not need them,
3000 since a watchpoint is inspecting the value of an expression anyhow---but
3001 it might be simpler, say, to just set a watchpoint on a variable name,
3002 and specify a condition that tests whether the new value is an interesting
3005 Break conditions can have side effects, and may even call functions in
3006 your program. This can be useful, for example, to activate functions
3007 that log program progress, or to use your own print functions to
3008 format special data structures. The effects are completely predictable
3009 unless there is another enabled breakpoint at the same address. (In
3010 that case, @value{GDBN} might see the other breakpoint first and stop your
3011 program without checking the condition of this one.) Note that
3012 breakpoint commands are usually more convenient and flexible than break
3014 purpose of performing side effects when a breakpoint is reached
3015 (@pxref{Break Commands, ,Breakpoint command lists}).
3017 Break conditions can be specified when a breakpoint is set, by using
3018 @samp{if} in the arguments to the @code{break} command. @xref{Set
3019 Breaks, ,Setting breakpoints}. They can also be changed at any time
3020 with the @code{condition} command.
3022 You can also use the @code{if} keyword with the @code{watch} command.
3023 The @code{catch} command does not recognize the @code{if} keyword;
3024 @code{condition} is the only way to impose a further condition on a
3029 @item condition @var{bnum} @var{expression}
3030 Specify @var{expression} as the break condition for breakpoint,
3031 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
3032 breakpoint @var{bnum} stops your program only if the value of
3033 @var{expression} is true (nonzero, in C). When you use
3034 @code{condition}, @value{GDBN} checks @var{expression} immediately for
3035 syntactic correctness, and to determine whether symbols in it have
3036 referents in the context of your breakpoint. If @var{expression} uses
3037 symbols not referenced in the context of the breakpoint, @value{GDBN}
3038 prints an error message:
3041 No symbol "foo" in current context.
3046 not actually evaluate @var{expression} at the time the @code{condition}
3047 command (or a command that sets a breakpoint with a condition, like
3048 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
3050 @item condition @var{bnum}
3051 Remove the condition from breakpoint number @var{bnum}. It becomes
3052 an ordinary unconditional breakpoint.
3055 @cindex ignore count (of breakpoint)
3056 A special case of a breakpoint condition is to stop only when the
3057 breakpoint has been reached a certain number of times. This is so
3058 useful that there is a special way to do it, using the @dfn{ignore
3059 count} of the breakpoint. Every breakpoint has an ignore count, which
3060 is an integer. Most of the time, the ignore count is zero, and
3061 therefore has no effect. But if your program reaches a breakpoint whose
3062 ignore count is positive, then instead of stopping, it just decrements
3063 the ignore count by one and continues. As a result, if the ignore count
3064 value is @var{n}, the breakpoint does not stop the next @var{n} times
3065 your program reaches it.
3069 @item ignore @var{bnum} @var{count}
3070 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3071 The next @var{count} times the breakpoint is reached, your program's
3072 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3075 To make the breakpoint stop the next time it is reached, specify
3078 When you use @code{continue} to resume execution of your program from a
3079 breakpoint, you can specify an ignore count directly as an argument to
3080 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3081 Stepping,,Continuing and stepping}.
3083 If a breakpoint has a positive ignore count and a condition, the
3084 condition is not checked. Once the ignore count reaches zero,
3085 @value{GDBN} resumes checking the condition.
3087 You could achieve the effect of the ignore count with a condition such
3088 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3089 is decremented each time. @xref{Convenience Vars, ,Convenience
3093 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3096 @node Break Commands
3097 @subsection Breakpoint command lists
3099 @cindex breakpoint commands
3100 You can give any breakpoint (or watchpoint or catchpoint) a series of
3101 commands to execute when your program stops due to that breakpoint. For
3102 example, you might want to print the values of certain expressions, or
3103 enable other breakpoints.
3108 @item commands @r{[}@var{bnum}@r{]}
3109 @itemx @dots{} @var{command-list} @dots{}
3111 Specify a list of commands for breakpoint number @var{bnum}. The commands
3112 themselves appear on the following lines. Type a line containing just
3113 @code{end} to terminate the commands.
3115 To remove all commands from a breakpoint, type @code{commands} and
3116 follow it immediately with @code{end}; that is, give no commands.
3118 With no @var{bnum} argument, @code{commands} refers to the last
3119 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
3120 recently encountered).
3123 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
3124 disabled within a @var{command-list}.
3126 You can use breakpoint commands to start your program up again. Simply
3127 use the @code{continue} command, or @code{step}, or any other command
3128 that resumes execution.
3130 Any other commands in the command list, after a command that resumes
3131 execution, are ignored. This is because any time you resume execution
3132 (even with a simple @code{next} or @code{step}), you may encounter
3133 another breakpoint---which could have its own command list, leading to
3134 ambiguities about which list to execute.
3137 If the first command you specify in a command list is @code{silent}, the
3138 usual message about stopping at a breakpoint is not printed. This may
3139 be desirable for breakpoints that are to print a specific message and
3140 then continue. If none of the remaining commands print anything, you
3141 see no sign that the breakpoint was reached. @code{silent} is
3142 meaningful only at the beginning of a breakpoint command list.
3144 The commands @code{echo}, @code{output}, and @code{printf} allow you to
3145 print precisely controlled output, and are often useful in silent
3146 breakpoints. @xref{Output, ,Commands for controlled output}.
3148 For example, here is how you could use breakpoint commands to print the
3149 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
3155 printf "x is %d\n",x
3160 One application for breakpoint commands is to compensate for one bug so
3161 you can test for another. Put a breakpoint just after the erroneous line
3162 of code, give it a condition to detect the case in which something
3163 erroneous has been done, and give it commands to assign correct values
3164 to any variables that need them. End with the @code{continue} command
3165 so that your program does not stop, and start with the @code{silent}
3166 command so that no output is produced. Here is an example:
3177 @node Breakpoint Menus
3178 @subsection Breakpoint menus
3180 @cindex symbol overloading
3182 Some programming languages (notably C@t{++}) permit a single function name
3183 to be defined several times, for application in different contexts.
3184 This is called @dfn{overloading}. When a function name is overloaded,
3185 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
3186 a breakpoint. If you realize this is a problem, you can use
3187 something like @samp{break @var{function}(@var{types})} to specify which
3188 particular version of the function you want. Otherwise, @value{GDBN} offers
3189 you a menu of numbered choices for different possible breakpoints, and
3190 waits for your selection with the prompt @samp{>}. The first two
3191 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
3192 sets a breakpoint at each definition of @var{function}, and typing
3193 @kbd{0} aborts the @code{break} command without setting any new
3196 For example, the following session excerpt shows an attempt to set a
3197 breakpoint at the overloaded symbol @code{String::after}.
3198 We choose three particular definitions of that function name:
3200 @c FIXME! This is likely to change to show arg type lists, at least
3203 (@value{GDBP}) b String::after
3206 [2] file:String.cc; line number:867
3207 [3] file:String.cc; line number:860
3208 [4] file:String.cc; line number:875
3209 [5] file:String.cc; line number:853
3210 [6] file:String.cc; line number:846
3211 [7] file:String.cc; line number:735
3213 Breakpoint 1 at 0xb26c: file String.cc, line 867.
3214 Breakpoint 2 at 0xb344: file String.cc, line 875.
3215 Breakpoint 3 at 0xafcc: file String.cc, line 846.
3216 Multiple breakpoints were set.
3217 Use the "delete" command to delete unwanted
3223 @c @ifclear BARETARGET
3224 @node Error in Breakpoints
3225 @subsection ``Cannot insert breakpoints''
3227 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
3229 Under some operating systems, breakpoints cannot be used in a program if
3230 any other process is running that program. In this situation,
3231 attempting to run or continue a program with a breakpoint causes
3232 @value{GDBN} to print an error message:
3235 Cannot insert breakpoints.
3236 The same program may be running in another process.
3239 When this happens, you have three ways to proceed:
3243 Remove or disable the breakpoints, then continue.
3246 Suspend @value{GDBN}, and copy the file containing your program to a new
3247 name. Resume @value{GDBN} and use the @code{exec-file} command to specify
3248 that @value{GDBN} should run your program under that name.
3249 Then start your program again.
3252 Relink your program so that the text segment is nonsharable, using the
3253 linker option @samp{-N}. The operating system limitation may not apply
3254 to nonsharable executables.
3258 A similar message can be printed if you request too many active
3259 hardware-assisted breakpoints and watchpoints:
3261 @c FIXME: the precise wording of this message may change; the relevant
3262 @c source change is not committed yet (Sep 3, 1999).
3264 Stopped; cannot insert breakpoints.
3265 You may have requested too many hardware breakpoints and watchpoints.
3269 This message is printed when you attempt to resume the program, since
3270 only then @value{GDBN} knows exactly how many hardware breakpoints and
3271 watchpoints it needs to insert.
3273 When this message is printed, you need to disable or remove some of the
3274 hardware-assisted breakpoints and watchpoints, and then continue.
3277 @node Continuing and Stepping
3278 @section Continuing and stepping
3282 @cindex resuming execution
3283 @dfn{Continuing} means resuming program execution until your program
3284 completes normally. In contrast, @dfn{stepping} means executing just
3285 one more ``step'' of your program, where ``step'' may mean either one
3286 line of source code, or one machine instruction (depending on what
3287 particular command you use). Either when continuing or when stepping,
3288 your program may stop even sooner, due to a breakpoint or a signal. (If
3289 it stops due to a signal, you may want to use @code{handle}, or use
3290 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
3294 @kindex c @r{(@code{continue})}
3295 @kindex fg @r{(resume foreground execution)}
3296 @item continue @r{[}@var{ignore-count}@r{]}
3297 @itemx c @r{[}@var{ignore-count}@r{]}
3298 @itemx fg @r{[}@var{ignore-count}@r{]}
3299 Resume program execution, at the address where your program last stopped;
3300 any breakpoints set at that address are bypassed. The optional argument
3301 @var{ignore-count} allows you to specify a further number of times to
3302 ignore a breakpoint at this location; its effect is like that of
3303 @code{ignore} (@pxref{Conditions, ,Break conditions}).
3305 The argument @var{ignore-count} is meaningful only when your program
3306 stopped due to a breakpoint. At other times, the argument to
3307 @code{continue} is ignored.
3309 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
3310 debugged program is deemed to be the foreground program) are provided
3311 purely for convenience, and have exactly the same behavior as
3315 To resume execution at a different place, you can use @code{return}
3316 (@pxref{Returning, ,Returning from a function}) to go back to the
3317 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
3318 different address}) to go to an arbitrary location in your program.
3320 A typical technique for using stepping is to set a breakpoint
3321 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and catchpoints}) at the
3322 beginning of the function or the section of your program where a problem
3323 is believed to lie, run your program until it stops at that breakpoint,
3324 and then step through the suspect area, examining the variables that are
3325 interesting, until you see the problem happen.
3329 @kindex s @r{(@code{step})}
3331 Continue running your program until control reaches a different source
3332 line, then stop it and return control to @value{GDBN}. This command is
3333 abbreviated @code{s}.
3336 @c "without debugging information" is imprecise; actually "without line
3337 @c numbers in the debugging information". (gcc -g1 has debugging info but
3338 @c not line numbers). But it seems complex to try to make that
3339 @c distinction here.
3340 @emph{Warning:} If you use the @code{step} command while control is
3341 within a function that was compiled without debugging information,
3342 execution proceeds until control reaches a function that does have
3343 debugging information. Likewise, it will not step into a function which
3344 is compiled without debugging information. To step through functions
3345 without debugging information, use the @code{stepi} command, described
3349 The @code{step} command only stops at the first instruction of a source
3350 line. This prevents the multiple stops that could otherwise occur in
3351 @code{switch} statements, @code{for} loops, etc. @code{step} continues
3352 to stop if a function that has debugging information is called within
3353 the line. In other words, @code{step} @emph{steps inside} any functions
3354 called within the line.
3356 Also, the @code{step} command only enters a function if there is line
3357 number information for the function. Otherwise it acts like the
3358 @code{next} command. This avoids problems when using @code{cc -gl}
3359 on MIPS machines. Previously, @code{step} entered subroutines if there
3360 was any debugging information about the routine.
3362 @item step @var{count}
3363 Continue running as in @code{step}, but do so @var{count} times. If a
3364 breakpoint is reached, or a signal not related to stepping occurs before
3365 @var{count} steps, stepping stops right away.
3368 @kindex n @r{(@code{next})}
3369 @item next @r{[}@var{count}@r{]}
3370 Continue to the next source line in the current (innermost) stack frame.
3371 This is similar to @code{step}, but function calls that appear within
3372 the line of code are executed without stopping. Execution stops when
3373 control reaches a different line of code at the original stack level
3374 that was executing when you gave the @code{next} command. This command
3375 is abbreviated @code{n}.
3377 An argument @var{count} is a repeat count, as for @code{step}.
3380 @c FIX ME!! Do we delete this, or is there a way it fits in with
3381 @c the following paragraph? --- Vctoria
3383 @c @code{next} within a function that lacks debugging information acts like
3384 @c @code{step}, but any function calls appearing within the code of the
3385 @c function are executed without stopping.
3387 The @code{next} command only stops at the first instruction of a
3388 source line. This prevents multiple stops that could otherwise occur in
3389 @code{switch} statements, @code{for} loops, etc.
3391 @kindex set step-mode
3393 @cindex functions without line info, and stepping
3394 @cindex stepping into functions with no line info
3395 @itemx set step-mode on
3396 The @code{set step-mode on} command causes the @code{step} command to
3397 stop at the first instruction of a function which contains no debug line
3398 information rather than stepping over it.
3400 This is useful in cases where you may be interested in inspecting the
3401 machine instructions of a function which has no symbolic info and do not
3402 want @value{GDBN} to automatically skip over this function.
3404 @item set step-mode off
3405 Causes the @code{step} command to step over any functions which contains no
3406 debug information. This is the default.
3410 Continue running until just after function in the selected stack frame
3411 returns. Print the returned value (if any).
3413 Contrast this with the @code{return} command (@pxref{Returning,
3414 ,Returning from a function}).
3417 @kindex u @r{(@code{until})}
3420 Continue running until a source line past the current line, in the
3421 current stack frame, is reached. This command is used to avoid single
3422 stepping through a loop more than once. It is like the @code{next}
3423 command, except that when @code{until} encounters a jump, it
3424 automatically continues execution until the program counter is greater
3425 than the address of the jump.
3427 This means that when you reach the end of a loop after single stepping
3428 though it, @code{until} makes your program continue execution until it
3429 exits the loop. In contrast, a @code{next} command at the end of a loop
3430 simply steps back to the beginning of the loop, which forces you to step
3431 through the next iteration.
3433 @code{until} always stops your program if it attempts to exit the current
3436 @code{until} may produce somewhat counterintuitive results if the order
3437 of machine code does not match the order of the source lines. For
3438 example, in the following excerpt from a debugging session, the @code{f}
3439 (@code{frame}) command shows that execution is stopped at line
3440 @code{206}; yet when we use @code{until}, we get to line @code{195}:
3444 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
3446 (@value{GDBP}) until
3447 195 for ( ; argc > 0; NEXTARG) @{
3450 This happened because, for execution efficiency, the compiler had
3451 generated code for the loop closure test at the end, rather than the
3452 start, of the loop---even though the test in a C @code{for}-loop is
3453 written before the body of the loop. The @code{until} command appeared
3454 to step back to the beginning of the loop when it advanced to this
3455 expression; however, it has not really gone to an earlier
3456 statement---not in terms of the actual machine code.
3458 @code{until} with no argument works by means of single
3459 instruction stepping, and hence is slower than @code{until} with an
3462 @item until @var{location}
3463 @itemx u @var{location}
3464 Continue running your program until either the specified location is
3465 reached, or the current stack frame returns. @var{location} is any of
3466 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
3467 ,Setting breakpoints}). This form of the command uses breakpoints, and
3468 hence is quicker than @code{until} without an argument. The specified
3469 location is actually reached only if it is in the current frame. This
3470 implies that @code{until} can be used to skip over recursive function
3471 invocations. For instance in the code below, if the current location is
3472 line @code{96}, issuing @code{until 99} will execute the program up to
3473 line @code{99} in the same invocation of factorial, i.e. after the inner
3474 invocations have returned.
3477 94 int factorial (int value)
3479 96 if (value > 1) @{
3480 97 value *= factorial (value - 1);
3487 @kindex advance @var{location}
3488 @itemx advance @var{location}
3489 Continue running the program up to the given location. An argument is
3490 required, anything of the same form as arguments for the @code{break}
3491 command. Execution will also stop upon exit from the current stack
3492 frame. This command is similar to @code{until}, but @code{advance} will
3493 not skip over recursive function calls, and the target location doesn't
3494 have to be in the same frame as the current one.
3498 @kindex si @r{(@code{stepi})}
3500 @itemx stepi @var{arg}
3502 Execute one machine instruction, then stop and return to the debugger.
3504 It is often useful to do @samp{display/i $pc} when stepping by machine
3505 instructions. This makes @value{GDBN} automatically display the next
3506 instruction to be executed, each time your program stops. @xref{Auto
3507 Display,, Automatic display}.
3509 An argument is a repeat count, as in @code{step}.
3513 @kindex ni @r{(@code{nexti})}
3515 @itemx nexti @var{arg}
3517 Execute one machine instruction, but if it is a function call,
3518 proceed until the function returns.
3520 An argument is a repeat count, as in @code{next}.
3527 A signal is an asynchronous event that can happen in a program. The
3528 operating system defines the possible kinds of signals, and gives each
3529 kind a name and a number. For example, in Unix @code{SIGINT} is the
3530 signal a program gets when you type an interrupt character (often @kbd{C-c});
3531 @code{SIGSEGV} is the signal a program gets from referencing a place in
3532 memory far away from all the areas in use; @code{SIGALRM} occurs when
3533 the alarm clock timer goes off (which happens only if your program has
3534 requested an alarm).
3536 @cindex fatal signals
3537 Some signals, including @code{SIGALRM}, are a normal part of the
3538 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3539 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
3540 program has not specified in advance some other way to handle the signal.
3541 @code{SIGINT} does not indicate an error in your program, but it is normally
3542 fatal so it can carry out the purpose of the interrupt: to kill the program.
3544 @value{GDBN} has the ability to detect any occurrence of a signal in your
3545 program. You can tell @value{GDBN} in advance what to do for each kind of
3548 @cindex handling signals
3549 Normally, @value{GDBN} is set up to let the non-erroneous signals like
3550 @code{SIGALRM} be silently passed to your program
3551 (so as not to interfere with their role in the program's functioning)
3552 but to stop your program immediately whenever an error signal happens.
3553 You can change these settings with the @code{handle} command.
3556 @kindex info signals
3559 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3560 handle each one. You can use this to see the signal numbers of all
3561 the defined types of signals.
3563 @code{info handle} is an alias for @code{info signals}.
3566 @item handle @var{signal} @var{keywords}@dots{}
3567 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
3568 can be the number of a signal or its name (with or without the
3569 @samp{SIG} at the beginning); a list of signal numbers of the form
3570 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
3571 known signals. The @var{keywords} say what change to make.
3575 The keywords allowed by the @code{handle} command can be abbreviated.
3576 Their full names are:
3580 @value{GDBN} should not stop your program when this signal happens. It may
3581 still print a message telling you that the signal has come in.
3584 @value{GDBN} should stop your program when this signal happens. This implies
3585 the @code{print} keyword as well.
3588 @value{GDBN} should print a message when this signal happens.
3591 @value{GDBN} should not mention the occurrence of the signal at all. This
3592 implies the @code{nostop} keyword as well.
3596 @value{GDBN} should allow your program to see this signal; your program
3597 can handle the signal, or else it may terminate if the signal is fatal
3598 and not handled. @code{pass} and @code{noignore} are synonyms.
3602 @value{GDBN} should not allow your program to see this signal.
3603 @code{nopass} and @code{ignore} are synonyms.
3607 When a signal stops your program, the signal is not visible to the
3609 continue. Your program sees the signal then, if @code{pass} is in
3610 effect for the signal in question @emph{at that time}. In other words,
3611 after @value{GDBN} reports a signal, you can use the @code{handle}
3612 command with @code{pass} or @code{nopass} to control whether your
3613 program sees that signal when you continue.
3615 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
3616 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
3617 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
3620 You can also use the @code{signal} command to prevent your program from
3621 seeing a signal, or cause it to see a signal it normally would not see,
3622 or to give it any signal at any time. For example, if your program stopped
3623 due to some sort of memory reference error, you might store correct
3624 values into the erroneous variables and continue, hoping to see more
3625 execution; but your program would probably terminate immediately as
3626 a result of the fatal signal once it saw the signal. To prevent this,
3627 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3631 @section Stopping and starting multi-thread programs
3633 When your program has multiple threads (@pxref{Threads,, Debugging
3634 programs with multiple threads}), you can choose whether to set
3635 breakpoints on all threads, or on a particular thread.
3638 @cindex breakpoints and threads
3639 @cindex thread breakpoints
3640 @kindex break @dots{} thread @var{threadno}
3641 @item break @var{linespec} thread @var{threadno}
3642 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3643 @var{linespec} specifies source lines; there are several ways of
3644 writing them, but the effect is always to specify some source line.
3646 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3647 to specify that you only want @value{GDBN} to stop the program when a
3648 particular thread reaches this breakpoint. @var{threadno} is one of the
3649 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3650 column of the @samp{info threads} display.
3652 If you do not specify @samp{thread @var{threadno}} when you set a
3653 breakpoint, the breakpoint applies to @emph{all} threads of your
3656 You can use the @code{thread} qualifier on conditional breakpoints as
3657 well; in this case, place @samp{thread @var{threadno}} before the
3658 breakpoint condition, like this:
3661 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
3666 @cindex stopped threads
3667 @cindex threads, stopped
3668 Whenever your program stops under @value{GDBN} for any reason,
3669 @emph{all} threads of execution stop, not just the current thread. This
3670 allows you to examine the overall state of the program, including
3671 switching between threads, without worrying that things may change
3674 @cindex continuing threads
3675 @cindex threads, continuing
3676 Conversely, whenever you restart the program, @emph{all} threads start
3677 executing. @emph{This is true even when single-stepping} with commands
3678 like @code{step} or @code{next}.
3680 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3681 Since thread scheduling is up to your debugging target's operating
3682 system (not controlled by @value{GDBN}), other threads may
3683 execute more than one statement while the current thread completes a
3684 single step. Moreover, in general other threads stop in the middle of a
3685 statement, rather than at a clean statement boundary, when the program
3688 You might even find your program stopped in another thread after
3689 continuing or even single-stepping. This happens whenever some other
3690 thread runs into a breakpoint, a signal, or an exception before the
3691 first thread completes whatever you requested.
3693 On some OSes, you can lock the OS scheduler and thus allow only a single
3697 @item set scheduler-locking @var{mode}
3698 Set the scheduler locking mode. If it is @code{off}, then there is no
3699 locking and any thread may run at any time. If @code{on}, then only the
3700 current thread may run when the inferior is resumed. The @code{step}
3701 mode optimizes for single-stepping. It stops other threads from
3702 ``seizing the prompt'' by preempting the current thread while you are
3703 stepping. Other threads will only rarely (or never) get a chance to run
3704 when you step. They are more likely to run when you @samp{next} over a
3705 function call, and they are completely free to run when you use commands
3706 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
3707 thread hits a breakpoint during its timeslice, they will never steal the
3708 @value{GDBN} prompt away from the thread that you are debugging.
3710 @item show scheduler-locking
3711 Display the current scheduler locking mode.
3716 @chapter Examining the Stack
3718 When your program has stopped, the first thing you need to know is where it
3719 stopped and how it got there.
3722 Each time your program performs a function call, information about the call
3724 That information includes the location of the call in your program,
3725 the arguments of the call,
3726 and the local variables of the function being called.
3727 The information is saved in a block of data called a @dfn{stack frame}.
3728 The stack frames are allocated in a region of memory called the @dfn{call
3731 When your program stops, the @value{GDBN} commands for examining the
3732 stack allow you to see all of this information.
3734 @cindex selected frame
3735 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3736 @value{GDBN} commands refer implicitly to the selected frame. In
3737 particular, whenever you ask @value{GDBN} for the value of a variable in
3738 your program, the value is found in the selected frame. There are
3739 special @value{GDBN} commands to select whichever frame you are
3740 interested in. @xref{Selection, ,Selecting a frame}.
3742 When your program stops, @value{GDBN} automatically selects the
3743 currently executing frame and describes it briefly, similar to the
3744 @code{frame} command (@pxref{Frame Info, ,Information about a frame}).
3747 * Frames:: Stack frames
3748 * Backtrace:: Backtraces
3749 * Selection:: Selecting a frame
3750 * Frame Info:: Information on a frame
3755 @section Stack frames
3757 @cindex frame, definition
3759 The call stack is divided up into contiguous pieces called @dfn{stack
3760 frames}, or @dfn{frames} for short; each frame is the data associated
3761 with one call to one function. The frame contains the arguments given
3762 to the function, the function's local variables, and the address at
3763 which the function is executing.
3765 @cindex initial frame
3766 @cindex outermost frame
3767 @cindex innermost frame
3768 When your program is started, the stack has only one frame, that of the
3769 function @code{main}. This is called the @dfn{initial} frame or the
3770 @dfn{outermost} frame. Each time a function is called, a new frame is
3771 made. Each time a function returns, the frame for that function invocation
3772 is eliminated. If a function is recursive, there can be many frames for
3773 the same function. The frame for the function in which execution is
3774 actually occurring is called the @dfn{innermost} frame. This is the most
3775 recently created of all the stack frames that still exist.
3777 @cindex frame pointer
3778 Inside your program, stack frames are identified by their addresses. A
3779 stack frame consists of many bytes, each of which has its own address; each
3780 kind of computer has a convention for choosing one byte whose
3781 address serves as the address of the frame. Usually this address is kept
3782 in a register called the @dfn{frame pointer register} while execution is
3783 going on in that frame.
3785 @cindex frame number
3786 @value{GDBN} assigns numbers to all existing stack frames, starting with
3787 zero for the innermost frame, one for the frame that called it,
3788 and so on upward. These numbers do not really exist in your program;
3789 they are assigned by @value{GDBN} to give you a way of designating stack
3790 frames in @value{GDBN} commands.
3792 @c The -fomit-frame-pointer below perennially causes hbox overflow
3793 @c underflow problems.
3794 @cindex frameless execution
3795 Some compilers provide a way to compile functions so that they operate
3796 without stack frames. (For example, the @value{GCC} option
3798 @samp{-fomit-frame-pointer}
3800 generates functions without a frame.)
3801 This is occasionally done with heavily used library functions to save
3802 the frame setup time. @value{GDBN} has limited facilities for dealing
3803 with these function invocations. If the innermost function invocation
3804 has no stack frame, @value{GDBN} nevertheless regards it as though
3805 it had a separate frame, which is numbered zero as usual, allowing
3806 correct tracing of the function call chain. However, @value{GDBN} has
3807 no provision for frameless functions elsewhere in the stack.
3810 @kindex frame@r{, command}
3811 @cindex current stack frame
3812 @item frame @var{args}
3813 The @code{frame} command allows you to move from one stack frame to another,
3814 and to print the stack frame you select. @var{args} may be either the
3815 address of the frame or the stack frame number. Without an argument,
3816 @code{frame} prints the current stack frame.
3818 @kindex select-frame
3819 @cindex selecting frame silently
3821 The @code{select-frame} command allows you to move from one stack frame
3822 to another without printing the frame. This is the silent version of
3831 @cindex stack traces
3832 A backtrace is a summary of how your program got where it is. It shows one
3833 line per frame, for many frames, starting with the currently executing
3834 frame (frame zero), followed by its caller (frame one), and on up the
3839 @kindex bt @r{(@code{backtrace})}
3842 Print a backtrace of the entire stack: one line per frame for all
3843 frames in the stack.
3845 You can stop the backtrace at any time by typing the system interrupt
3846 character, normally @kbd{C-c}.
3848 @item backtrace @var{n}
3850 Similar, but print only the innermost @var{n} frames.
3852 @item backtrace -@var{n}
3854 Similar, but print only the outermost @var{n} frames.
3859 @kindex info s @r{(@code{info stack})}
3860 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3861 are additional aliases for @code{backtrace}.
3863 Each line in the backtrace shows the frame number and the function name.
3864 The program counter value is also shown---unless you use @code{set
3865 print address off}. The backtrace also shows the source file name and
3866 line number, as well as the arguments to the function. The program
3867 counter value is omitted if it is at the beginning of the code for that
3870 Here is an example of a backtrace. It was made with the command
3871 @samp{bt 3}, so it shows the innermost three frames.
3875 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3877 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3878 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3880 (More stack frames follow...)
3885 The display for frame zero does not begin with a program counter
3886 value, indicating that your program has stopped at the beginning of the
3887 code for line @code{993} of @code{builtin.c}.
3889 @kindex set backtrace-below-main
3890 @kindex show backtrace-below-main
3892 Most programs have a standard entry point---a place where system libraries
3893 and startup code transition into user code. For C this is @code{main}.
3894 When @value{GDBN} finds the entry function in a backtrace it will terminate
3895 the backtrace, to avoid tracing into highly system-specific (and generally
3896 uninteresting) code. If you need to examine the startup code, then you can
3897 change this behavior.
3900 @item set backtrace-below-main off
3901 Backtraces will stop when they encounter the user entry point. This is the
3904 @item set backtrace-below-main
3905 @itemx set backtrace-below-main on
3906 Backtraces will continue past the user entry point to the top of the stack.
3908 @item show backtrace-below-main
3909 Display the current backtrace policy.
3913 @section Selecting a frame
3915 Most commands for examining the stack and other data in your program work on
3916 whichever stack frame is selected at the moment. Here are the commands for
3917 selecting a stack frame; all of them finish by printing a brief description
3918 of the stack frame just selected.
3921 @kindex frame@r{, selecting}
3922 @kindex f @r{(@code{frame})}
3925 Select frame number @var{n}. Recall that frame zero is the innermost
3926 (currently executing) frame, frame one is the frame that called the
3927 innermost one, and so on. The highest-numbered frame is the one for
3930 @item frame @var{addr}
3932 Select the frame at address @var{addr}. This is useful mainly if the
3933 chaining of stack frames has been damaged by a bug, making it
3934 impossible for @value{GDBN} to assign numbers properly to all frames. In
3935 addition, this can be useful when your program has multiple stacks and
3936 switches between them.
3938 On the SPARC architecture, @code{frame} needs two addresses to
3939 select an arbitrary frame: a frame pointer and a stack pointer.
3941 On the MIPS and Alpha architecture, it needs two addresses: a stack
3942 pointer and a program counter.
3944 On the 29k architecture, it needs three addresses: a register stack
3945 pointer, a program counter, and a memory stack pointer.
3946 @c note to future updaters: this is conditioned on a flag
3947 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3948 @c as of 27 Jan 1994.
3952 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3953 advances toward the outermost frame, to higher frame numbers, to frames
3954 that have existed longer. @var{n} defaults to one.
3957 @kindex do @r{(@code{down})}
3959 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3960 advances toward the innermost frame, to lower frame numbers, to frames
3961 that were created more recently. @var{n} defaults to one. You may
3962 abbreviate @code{down} as @code{do}.
3965 All of these commands end by printing two lines of output describing the
3966 frame. The first line shows the frame number, the function name, the
3967 arguments, and the source file and line number of execution in that
3968 frame. The second line shows the text of that source line.
3976 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3978 10 read_input_file (argv[i]);
3982 After such a printout, the @code{list} command with no arguments
3983 prints ten lines centered on the point of execution in the frame.
3984 You can also edit the program at the point of execution with your favorite
3985 editing program by typing @code{edit}.
3986 @xref{List, ,Printing source lines},
3990 @kindex down-silently
3992 @item up-silently @var{n}
3993 @itemx down-silently @var{n}
3994 These two commands are variants of @code{up} and @code{down},
3995 respectively; they differ in that they do their work silently, without
3996 causing display of the new frame. They are intended primarily for use
3997 in @value{GDBN} command scripts, where the output might be unnecessary and
4002 @section Information about a frame
4004 There are several other commands to print information about the selected
4010 When used without any argument, this command does not change which
4011 frame is selected, but prints a brief description of the currently
4012 selected stack frame. It can be abbreviated @code{f}. With an
4013 argument, this command is used to select a stack frame.
4014 @xref{Selection, ,Selecting a frame}.
4017 @kindex info f @r{(@code{info frame})}
4020 This command prints a verbose description of the selected stack frame,
4025 the address of the frame
4027 the address of the next frame down (called by this frame)
4029 the address of the next frame up (caller of this frame)
4031 the language in which the source code corresponding to this frame is written
4033 the address of the frame's arguments
4035 the address of the frame's local variables
4037 the program counter saved in it (the address of execution in the caller frame)
4039 which registers were saved in the frame
4042 @noindent The verbose description is useful when
4043 something has gone wrong that has made the stack format fail to fit
4044 the usual conventions.
4046 @item info frame @var{addr}
4047 @itemx info f @var{addr}
4048 Print a verbose description of the frame at address @var{addr}, without
4049 selecting that frame. The selected frame remains unchanged by this
4050 command. This requires the same kind of address (more than one for some
4051 architectures) that you specify in the @code{frame} command.
4052 @xref{Selection, ,Selecting a frame}.
4056 Print the arguments of the selected frame, each on a separate line.
4060 Print the local variables of the selected frame, each on a separate
4061 line. These are all variables (declared either static or automatic)
4062 accessible at the point of execution of the selected frame.
4065 @cindex catch exceptions, list active handlers
4066 @cindex exception handlers, how to list
4068 Print a list of all the exception handlers that are active in the
4069 current stack frame at the current point of execution. To see other
4070 exception handlers, visit the associated frame (using the @code{up},
4071 @code{down}, or @code{frame} commands); then type @code{info catch}.
4072 @xref{Set Catchpoints, , Setting catchpoints}.
4078 @chapter Examining Source Files
4080 @value{GDBN} can print parts of your program's source, since the debugging
4081 information recorded in the program tells @value{GDBN} what source files were
4082 used to build it. When your program stops, @value{GDBN} spontaneously prints
4083 the line where it stopped. Likewise, when you select a stack frame
4084 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
4085 execution in that frame has stopped. You can print other portions of
4086 source files by explicit command.
4088 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
4089 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
4090 @value{GDBN} under @sc{gnu} Emacs}.
4093 * List:: Printing source lines
4094 * Edit:: Editing source files
4095 * Search:: Searching source files
4096 * Source Path:: Specifying source directories
4097 * Machine Code:: Source and machine code
4101 @section Printing source lines
4104 @kindex l @r{(@code{list})}
4105 To print lines from a source file, use the @code{list} command
4106 (abbreviated @code{l}). By default, ten lines are printed.
4107 There are several ways to specify what part of the file you want to print.
4109 Here are the forms of the @code{list} command most commonly used:
4112 @item list @var{linenum}
4113 Print lines centered around line number @var{linenum} in the
4114 current source file.
4116 @item list @var{function}
4117 Print lines centered around the beginning of function
4121 Print more lines. If the last lines printed were printed with a
4122 @code{list} command, this prints lines following the last lines
4123 printed; however, if the last line printed was a solitary line printed
4124 as part of displaying a stack frame (@pxref{Stack, ,Examining the
4125 Stack}), this prints lines centered around that line.
4128 Print lines just before the lines last printed.
4131 By default, @value{GDBN} prints ten source lines with any of these forms of
4132 the @code{list} command. You can change this using @code{set listsize}:
4135 @kindex set listsize
4136 @item set listsize @var{count}
4137 Make the @code{list} command display @var{count} source lines (unless
4138 the @code{list} argument explicitly specifies some other number).
4140 @kindex show listsize
4142 Display the number of lines that @code{list} prints.
4145 Repeating a @code{list} command with @key{RET} discards the argument,
4146 so it is equivalent to typing just @code{list}. This is more useful
4147 than listing the same lines again. An exception is made for an
4148 argument of @samp{-}; that argument is preserved in repetition so that
4149 each repetition moves up in the source file.
4152 In general, the @code{list} command expects you to supply zero, one or two
4153 @dfn{linespecs}. Linespecs specify source lines; there are several ways
4154 of writing them, but the effect is always to specify some source line.
4155 Here is a complete description of the possible arguments for @code{list}:
4158 @item list @var{linespec}
4159 Print lines centered around the line specified by @var{linespec}.
4161 @item list @var{first},@var{last}
4162 Print lines from @var{first} to @var{last}. Both arguments are
4165 @item list ,@var{last}
4166 Print lines ending with @var{last}.
4168 @item list @var{first},
4169 Print lines starting with @var{first}.
4172 Print lines just after the lines last printed.
4175 Print lines just before the lines last printed.
4178 As described in the preceding table.
4181 Here are the ways of specifying a single source line---all the
4186 Specifies line @var{number} of the current source file.
4187 When a @code{list} command has two linespecs, this refers to
4188 the same source file as the first linespec.
4191 Specifies the line @var{offset} lines after the last line printed.
4192 When used as the second linespec in a @code{list} command that has
4193 two, this specifies the line @var{offset} lines down from the
4197 Specifies the line @var{offset} lines before the last line printed.
4199 @item @var{filename}:@var{number}
4200 Specifies line @var{number} in the source file @var{filename}.
4202 @item @var{function}
4203 Specifies the line that begins the body of the function @var{function}.
4204 For example: in C, this is the line with the open brace.
4206 @item @var{filename}:@var{function}
4207 Specifies the line of the open-brace that begins the body of the
4208 function @var{function} in the file @var{filename}. You only need the
4209 file name with a function name to avoid ambiguity when there are
4210 identically named functions in different source files.
4212 @item *@var{address}
4213 Specifies the line containing the program address @var{address}.
4214 @var{address} may be any expression.
4218 @section Editing source files
4219 @cindex editing source files
4222 @kindex e @r{(@code{edit})}
4223 To edit the lines in a source file, use the @code{edit} command.
4224 The editing program of your choice
4225 is invoked with the current line set to
4226 the active line in the program.
4227 Alternatively, there are several ways to specify what part of the file you
4228 want to print if you want to see other parts of the program.
4230 Here are the forms of the @code{edit} command most commonly used:
4234 Edit the current source file at the active line number in the program.
4236 @item edit @var{number}
4237 Edit the current source file with @var{number} as the active line number.
4239 @item edit @var{function}
4240 Edit the file containing @var{function} at the beginning of its definition.
4242 @item edit @var{filename}:@var{number}
4243 Specifies line @var{number} in the source file @var{filename}.
4245 @item edit @var{filename}:@var{function}
4246 Specifies the line that begins the body of the
4247 function @var{function} in the file @var{filename}. You only need the
4248 file name with a function name to avoid ambiguity when there are
4249 identically named functions in different source files.
4251 @item edit *@var{address}
4252 Specifies the line containing the program address @var{address}.
4253 @var{address} may be any expression.
4256 @subsection Choosing your editor
4257 You can customize @value{GDBN} to use any editor you want
4259 The only restriction is that your editor (say @code{ex}), recognizes the
4260 following command-line syntax:
4262 ex +@var{number} file
4264 The optional numeric value +@var{number} designates the active line in
4265 the file.}. By default, it is @value{EDITOR}, but you can change this
4266 by setting the environment variable @code{EDITOR} before using
4267 @value{GDBN}. For example, to configure @value{GDBN} to use the
4268 @code{vi} editor, you could use these commands with the @code{sh} shell:
4274 or in the @code{csh} shell,
4276 setenv EDITOR /usr/bin/vi
4281 @section Searching source files
4283 @kindex reverse-search
4285 There are two commands for searching through the current source file for a
4290 @kindex forward-search
4291 @item forward-search @var{regexp}
4292 @itemx search @var{regexp}
4293 The command @samp{forward-search @var{regexp}} checks each line,
4294 starting with the one following the last line listed, for a match for
4295 @var{regexp}. It lists the line that is found. You can use the
4296 synonym @samp{search @var{regexp}} or abbreviate the command name as
4299 @item reverse-search @var{regexp}
4300 The command @samp{reverse-search @var{regexp}} checks each line, starting
4301 with the one before the last line listed and going backward, for a match
4302 for @var{regexp}. It lists the line that is found. You can abbreviate
4303 this command as @code{rev}.
4307 @section Specifying source directories
4310 @cindex directories for source files
4311 Executable programs sometimes do not record the directories of the source
4312 files from which they were compiled, just the names. Even when they do,
4313 the directories could be moved between the compilation and your debugging
4314 session. @value{GDBN} has a list of directories to search for source files;
4315 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
4316 it tries all the directories in the list, in the order they are present
4317 in the list, until it finds a file with the desired name. Note that
4318 the executable search path is @emph{not} used for this purpose. Neither is
4319 the current working directory, unless it happens to be in the source
4322 If @value{GDBN} cannot find a source file in the source path, and the
4323 object program records a directory, @value{GDBN} tries that directory
4324 too. If the source path is empty, and there is no record of the
4325 compilation directory, @value{GDBN} looks in the current directory as a
4328 Whenever you reset or rearrange the source path, @value{GDBN} clears out
4329 any information it has cached about where source files are found and where
4330 each line is in the file.
4334 When you start @value{GDBN}, its source path includes only @samp{cdir}
4335 and @samp{cwd}, in that order.
4336 To add other directories, use the @code{directory} command.
4339 @item directory @var{dirname} @dots{}
4340 @item dir @var{dirname} @dots{}
4341 Add directory @var{dirname} to the front of the source path. Several
4342 directory names may be given to this command, separated by @samp{:}
4343 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
4344 part of absolute file names) or
4345 whitespace. You may specify a directory that is already in the source
4346 path; this moves it forward, so @value{GDBN} searches it sooner.
4350 @vindex $cdir@r{, convenience variable}
4351 @vindex $cwdr@r{, convenience variable}
4352 @cindex compilation directory
4353 @cindex current directory
4354 @cindex working directory
4355 @cindex directory, current
4356 @cindex directory, compilation
4357 You can use the string @samp{$cdir} to refer to the compilation
4358 directory (if one is recorded), and @samp{$cwd} to refer to the current
4359 working directory. @samp{$cwd} is not the same as @samp{.}---the former
4360 tracks the current working directory as it changes during your @value{GDBN}
4361 session, while the latter is immediately expanded to the current
4362 directory at the time you add an entry to the source path.
4365 Reset the source path to empty again. This requires confirmation.
4367 @c RET-repeat for @code{directory} is explicitly disabled, but since
4368 @c repeating it would be a no-op we do not say that. (thanks to RMS)
4370 @item show directories
4371 @kindex show directories
4372 Print the source path: show which directories it contains.
4375 If your source path is cluttered with directories that are no longer of
4376 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
4377 versions of source. You can correct the situation as follows:
4381 Use @code{directory} with no argument to reset the source path to empty.
4384 Use @code{directory} with suitable arguments to reinstall the
4385 directories you want in the source path. You can add all the
4386 directories in one command.
4390 @section Source and machine code
4392 You can use the command @code{info line} to map source lines to program
4393 addresses (and vice versa), and the command @code{disassemble} to display
4394 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
4395 mode, the @code{info line} command causes the arrow to point to the
4396 line specified. Also, @code{info line} prints addresses in symbolic form as
4401 @item info line @var{linespec}
4402 Print the starting and ending addresses of the compiled code for
4403 source line @var{linespec}. You can specify source lines in any of
4404 the ways understood by the @code{list} command (@pxref{List, ,Printing
4408 For example, we can use @code{info line} to discover the location of
4409 the object code for the first line of function
4410 @code{m4_changequote}:
4412 @c FIXME: I think this example should also show the addresses in
4413 @c symbolic form, as they usually would be displayed.
4415 (@value{GDBP}) info line m4_changequote
4416 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
4420 We can also inquire (using @code{*@var{addr}} as the form for
4421 @var{linespec}) what source line covers a particular address:
4423 (@value{GDBP}) info line *0x63ff
4424 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
4427 @cindex @code{$_} and @code{info line}
4428 @kindex x@r{(examine), and} info line
4429 After @code{info line}, the default address for the @code{x} command
4430 is changed to the starting address of the line, so that @samp{x/i} is
4431 sufficient to begin examining the machine code (@pxref{Memory,
4432 ,Examining memory}). Also, this address is saved as the value of the
4433 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
4438 @cindex assembly instructions
4439 @cindex instructions, assembly
4440 @cindex machine instructions
4441 @cindex listing machine instructions
4443 This specialized command dumps a range of memory as machine
4444 instructions. The default memory range is the function surrounding the
4445 program counter of the selected frame. A single argument to this
4446 command is a program counter value; @value{GDBN} dumps the function
4447 surrounding this value. Two arguments specify a range of addresses
4448 (first inclusive, second exclusive) to dump.
4451 The following example shows the disassembly of a range of addresses of
4452 HP PA-RISC 2.0 code:
4455 (@value{GDBP}) disas 0x32c4 0x32e4
4456 Dump of assembler code from 0x32c4 to 0x32e4:
4457 0x32c4 <main+204>: addil 0,dp
4458 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
4459 0x32cc <main+212>: ldil 0x3000,r31
4460 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
4461 0x32d4 <main+220>: ldo 0(r31),rp
4462 0x32d8 <main+224>: addil -0x800,dp
4463 0x32dc <main+228>: ldo 0x588(r1),r26
4464 0x32e0 <main+232>: ldil 0x3000,r31
4465 End of assembler dump.
4468 Some architectures have more than one commonly-used set of instruction
4469 mnemonics or other syntax.
4472 @kindex set disassembly-flavor
4473 @cindex assembly instructions
4474 @cindex instructions, assembly
4475 @cindex machine instructions
4476 @cindex listing machine instructions
4477 @cindex Intel disassembly flavor
4478 @cindex AT&T disassembly flavor
4479 @item set disassembly-flavor @var{instruction-set}
4480 Select the instruction set to use when disassembling the
4481 program via the @code{disassemble} or @code{x/i} commands.
4483 Currently this command is only defined for the Intel x86 family. You
4484 can set @var{instruction-set} to either @code{intel} or @code{att}.
4485 The default is @code{att}, the AT&T flavor used by default by Unix
4486 assemblers for x86-based targets.
4491 @chapter Examining Data
4493 @cindex printing data
4494 @cindex examining data
4497 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
4498 @c document because it is nonstandard... Under Epoch it displays in a
4499 @c different window or something like that.
4500 The usual way to examine data in your program is with the @code{print}
4501 command (abbreviated @code{p}), or its synonym @code{inspect}. It
4502 evaluates and prints the value of an expression of the language your
4503 program is written in (@pxref{Languages, ,Using @value{GDBN} with
4504 Different Languages}).
4507 @item print @var{expr}
4508 @itemx print /@var{f} @var{expr}
4509 @var{expr} is an expression (in the source language). By default the
4510 value of @var{expr} is printed in a format appropriate to its data type;
4511 you can choose a different format by specifying @samp{/@var{f}}, where
4512 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
4516 @itemx print /@var{f}
4517 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
4518 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
4519 conveniently inspect the same value in an alternative format.
4522 A more low-level way of examining data is with the @code{x} command.
4523 It examines data in memory at a specified address and prints it in a
4524 specified format. @xref{Memory, ,Examining memory}.
4526 If you are interested in information about types, or about how the
4527 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
4528 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
4532 * Expressions:: Expressions
4533 * Variables:: Program variables
4534 * Arrays:: Artificial arrays
4535 * Output Formats:: Output formats
4536 * Memory:: Examining memory
4537 * Auto Display:: Automatic display
4538 * Print Settings:: Print settings
4539 * Value History:: Value history
4540 * Convenience Vars:: Convenience variables
4541 * Registers:: Registers
4542 * Floating Point Hardware:: Floating point hardware
4543 * Vector Unit:: Vector Unit
4544 * Memory Region Attributes:: Memory region attributes
4545 * Dump/Restore Files:: Copy between memory and a file
4546 * Character Sets:: Debugging programs that use a different
4547 character set than GDB does
4551 @section Expressions
4554 @code{print} and many other @value{GDBN} commands accept an expression and
4555 compute its value. Any kind of constant, variable or operator defined
4556 by the programming language you are using is valid in an expression in
4557 @value{GDBN}. This includes conditional expressions, function calls,
4558 casts, and string constants. It also includes preprocessor macros, if
4559 you compiled your program to include this information; see
4562 @value{GDBN} supports array constants in expressions input by
4563 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
4564 you can use the command @code{print @{1, 2, 3@}} to build up an array in
4565 memory that is @code{malloc}ed in the target program.
4567 Because C is so widespread, most of the expressions shown in examples in
4568 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
4569 Languages}, for information on how to use expressions in other
4572 In this section, we discuss operators that you can use in @value{GDBN}
4573 expressions regardless of your programming language.
4575 Casts are supported in all languages, not just in C, because it is so
4576 useful to cast a number into a pointer in order to examine a structure
4577 at that address in memory.
4578 @c FIXME: casts supported---Mod2 true?
4580 @value{GDBN} supports these operators, in addition to those common
4581 to programming languages:
4585 @samp{@@} is a binary operator for treating parts of memory as arrays.
4586 @xref{Arrays, ,Artificial arrays}, for more information.
4589 @samp{::} allows you to specify a variable in terms of the file or
4590 function where it is defined. @xref{Variables, ,Program variables}.
4592 @cindex @{@var{type}@}
4593 @cindex type casting memory
4594 @cindex memory, viewing as typed object
4595 @cindex casts, to view memory
4596 @item @{@var{type}@} @var{addr}
4597 Refers to an object of type @var{type} stored at address @var{addr} in
4598 memory. @var{addr} may be any expression whose value is an integer or
4599 pointer (but parentheses are required around binary operators, just as in
4600 a cast). This construct is allowed regardless of what kind of data is
4601 normally supposed to reside at @var{addr}.
4605 @section Program variables
4607 The most common kind of expression to use is the name of a variable
4610 Variables in expressions are understood in the selected stack frame
4611 (@pxref{Selection, ,Selecting a frame}); they must be either:
4615 global (or file-static)
4622 visible according to the scope rules of the
4623 programming language from the point of execution in that frame
4626 @noindent This means that in the function
4641 you can examine and use the variable @code{a} whenever your program is
4642 executing within the function @code{foo}, but you can only use or
4643 examine the variable @code{b} while your program is executing inside
4644 the block where @code{b} is declared.
4646 @cindex variable name conflict
4647 There is an exception: you can refer to a variable or function whose
4648 scope is a single source file even if the current execution point is not
4649 in this file. But it is possible to have more than one such variable or
4650 function with the same name (in different source files). If that
4651 happens, referring to that name has unpredictable effects. If you wish,
4652 you can specify a static variable in a particular function or file,
4653 using the colon-colon notation:
4655 @cindex colon-colon, context for variables/functions
4657 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4658 @cindex @code{::}, context for variables/functions
4661 @var{file}::@var{variable}
4662 @var{function}::@var{variable}
4666 Here @var{file} or @var{function} is the name of the context for the
4667 static @var{variable}. In the case of file names, you can use quotes to
4668 make sure @value{GDBN} parses the file name as a single word---for example,
4669 to print a global value of @code{x} defined in @file{f2.c}:
4672 (@value{GDBP}) p 'f2.c'::x
4675 @cindex C@t{++} scope resolution
4676 This use of @samp{::} is very rarely in conflict with the very similar
4677 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
4678 scope resolution operator in @value{GDBN} expressions.
4679 @c FIXME: Um, so what happens in one of those rare cases where it's in
4682 @cindex wrong values
4683 @cindex variable values, wrong
4685 @emph{Warning:} Occasionally, a local variable may appear to have the
4686 wrong value at certain points in a function---just after entry to a new
4687 scope, and just before exit.
4689 You may see this problem when you are stepping by machine instructions.
4690 This is because, on most machines, it takes more than one instruction to
4691 set up a stack frame (including local variable definitions); if you are
4692 stepping by machine instructions, variables may appear to have the wrong
4693 values until the stack frame is completely built. On exit, it usually
4694 also takes more than one machine instruction to destroy a stack frame;
4695 after you begin stepping through that group of instructions, local
4696 variable definitions may be gone.
4698 This may also happen when the compiler does significant optimizations.
4699 To be sure of always seeing accurate values, turn off all optimization
4702 @cindex ``No symbol "foo" in current context''
4703 Another possible effect of compiler optimizations is to optimize
4704 unused variables out of existence, or assign variables to registers (as
4705 opposed to memory addresses). Depending on the support for such cases
4706 offered by the debug info format used by the compiler, @value{GDBN}
4707 might not be able to display values for such local variables. If that
4708 happens, @value{GDBN} will print a message like this:
4711 No symbol "foo" in current context.
4714 To solve such problems, either recompile without optimizations, or use a
4715 different debug info format, if the compiler supports several such
4716 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler
4717 usually supports the @option{-gstabs+} option. @option{-gstabs+}
4718 produces debug info in a format that is superior to formats such as
4719 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
4720 an effective form for debug info. @xref{Debugging Options,,Options
4721 for Debugging Your Program or @sc{gnu} CC, gcc.info, Using @sc{gnu} CC}.
4725 @section Artificial arrays
4727 @cindex artificial array
4728 @kindex @@@r{, referencing memory as an array}
4729 It is often useful to print out several successive objects of the
4730 same type in memory; a section of an array, or an array of
4731 dynamically determined size for which only a pointer exists in the
4734 You can do this by referring to a contiguous span of memory as an
4735 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4736 operand of @samp{@@} should be the first element of the desired array
4737 and be an individual object. The right operand should be the desired length
4738 of the array. The result is an array value whose elements are all of
4739 the type of the left argument. The first element is actually the left
4740 argument; the second element comes from bytes of memory immediately
4741 following those that hold the first element, and so on. Here is an
4742 example. If a program says
4745 int *array = (int *) malloc (len * sizeof (int));
4749 you can print the contents of @code{array} with
4755 The left operand of @samp{@@} must reside in memory. Array values made
4756 with @samp{@@} in this way behave just like other arrays in terms of
4757 subscripting, and are coerced to pointers when used in expressions.
4758 Artificial arrays most often appear in expressions via the value history
4759 (@pxref{Value History, ,Value history}), after printing one out.
4761 Another way to create an artificial array is to use a cast.
4762 This re-interprets a value as if it were an array.
4763 The value need not be in memory:
4765 (@value{GDBP}) p/x (short[2])0x12345678
4766 $1 = @{0x1234, 0x5678@}
4769 As a convenience, if you leave the array length out (as in
4770 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
4771 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4773 (@value{GDBP}) p/x (short[])0x12345678
4774 $2 = @{0x1234, 0x5678@}
4777 Sometimes the artificial array mechanism is not quite enough; in
4778 moderately complex data structures, the elements of interest may not
4779 actually be adjacent---for example, if you are interested in the values
4780 of pointers in an array. One useful work-around in this situation is
4781 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4782 variables}) as a counter in an expression that prints the first
4783 interesting value, and then repeat that expression via @key{RET}. For
4784 instance, suppose you have an array @code{dtab} of pointers to
4785 structures, and you are interested in the values of a field @code{fv}
4786 in each structure. Here is an example of what you might type:
4796 @node Output Formats
4797 @section Output formats
4799 @cindex formatted output
4800 @cindex output formats
4801 By default, @value{GDBN} prints a value according to its data type. Sometimes
4802 this is not what you want. For example, you might want to print a number
4803 in hex, or a pointer in decimal. Or you might want to view data in memory
4804 at a certain address as a character string or as an instruction. To do
4805 these things, specify an @dfn{output format} when you print a value.
4807 The simplest use of output formats is to say how to print a value
4808 already computed. This is done by starting the arguments of the
4809 @code{print} command with a slash and a format letter. The format
4810 letters supported are:
4814 Regard the bits of the value as an integer, and print the integer in
4818 Print as integer in signed decimal.
4821 Print as integer in unsigned decimal.
4824 Print as integer in octal.
4827 Print as integer in binary. The letter @samp{t} stands for ``two''.
4828 @footnote{@samp{b} cannot be used because these format letters are also
4829 used with the @code{x} command, where @samp{b} stands for ``byte'';
4830 see @ref{Memory,,Examining memory}.}
4833 @cindex unknown address, locating
4834 @cindex locate address
4835 Print as an address, both absolute in hexadecimal and as an offset from
4836 the nearest preceding symbol. You can use this format used to discover
4837 where (in what function) an unknown address is located:
4840 (@value{GDBP}) p/a 0x54320
4841 $3 = 0x54320 <_initialize_vx+396>
4845 The command @code{info symbol 0x54320} yields similar results.
4846 @xref{Symbols, info symbol}.
4849 Regard as an integer and print it as a character constant.
4852 Regard the bits of the value as a floating point number and print
4853 using typical floating point syntax.
4856 For example, to print the program counter in hex (@pxref{Registers}), type
4863 Note that no space is required before the slash; this is because command
4864 names in @value{GDBN} cannot contain a slash.
4866 To reprint the last value in the value history with a different format,
4867 you can use the @code{print} command with just a format and no
4868 expression. For example, @samp{p/x} reprints the last value in hex.
4871 @section Examining memory
4873 You can use the command @code{x} (for ``examine'') to examine memory in
4874 any of several formats, independently of your program's data types.
4876 @cindex examining memory
4878 @kindex x @r{(examine memory)}
4879 @item x/@var{nfu} @var{addr}
4882 Use the @code{x} command to examine memory.
4885 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4886 much memory to display and how to format it; @var{addr} is an
4887 expression giving the address where you want to start displaying memory.
4888 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4889 Several commands set convenient defaults for @var{addr}.
4892 @item @var{n}, the repeat count
4893 The repeat count is a decimal integer; the default is 1. It specifies
4894 how much memory (counting by units @var{u}) to display.
4895 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4898 @item @var{f}, the display format
4899 The display format is one of the formats used by @code{print},
4900 @samp{s} (null-terminated string), or @samp{i} (machine instruction).
4901 The default is @samp{x} (hexadecimal) initially.
4902 The default changes each time you use either @code{x} or @code{print}.
4904 @item @var{u}, the unit size
4905 The unit size is any of
4911 Halfwords (two bytes).
4913 Words (four bytes). This is the initial default.
4915 Giant words (eight bytes).
4918 Each time you specify a unit size with @code{x}, that size becomes the
4919 default unit the next time you use @code{x}. (For the @samp{s} and
4920 @samp{i} formats, the unit size is ignored and is normally not written.)
4922 @item @var{addr}, starting display address
4923 @var{addr} is the address where you want @value{GDBN} to begin displaying
4924 memory. The expression need not have a pointer value (though it may);
4925 it is always interpreted as an integer address of a byte of memory.
4926 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4927 @var{addr} is usually just after the last address examined---but several
4928 other commands also set the default address: @code{info breakpoints} (to
4929 the address of the last breakpoint listed), @code{info line} (to the
4930 starting address of a line), and @code{print} (if you use it to display
4931 a value from memory).
4934 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4935 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4936 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4937 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4938 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
4940 Since the letters indicating unit sizes are all distinct from the
4941 letters specifying output formats, you do not have to remember whether
4942 unit size or format comes first; either order works. The output
4943 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4944 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4946 Even though the unit size @var{u} is ignored for the formats @samp{s}
4947 and @samp{i}, you might still want to use a count @var{n}; for example,
4948 @samp{3i} specifies that you want to see three machine instructions,
4949 including any operands. The command @code{disassemble} gives an
4950 alternative way of inspecting machine instructions; see @ref{Machine
4951 Code,,Source and machine code}.
4953 All the defaults for the arguments to @code{x} are designed to make it
4954 easy to continue scanning memory with minimal specifications each time
4955 you use @code{x}. For example, after you have inspected three machine
4956 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4957 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4958 the repeat count @var{n} is used again; the other arguments default as
4959 for successive uses of @code{x}.
4961 @cindex @code{$_}, @code{$__}, and value history
4962 The addresses and contents printed by the @code{x} command are not saved
4963 in the value history because there is often too much of them and they
4964 would get in the way. Instead, @value{GDBN} makes these values available for
4965 subsequent use in expressions as values of the convenience variables
4966 @code{$_} and @code{$__}. After an @code{x} command, the last address
4967 examined is available for use in expressions in the convenience variable
4968 @code{$_}. The contents of that address, as examined, are available in
4969 the convenience variable @code{$__}.
4971 If the @code{x} command has a repeat count, the address and contents saved
4972 are from the last memory unit printed; this is not the same as the last
4973 address printed if several units were printed on the last line of output.
4976 @section Automatic display
4977 @cindex automatic display
4978 @cindex display of expressions
4980 If you find that you want to print the value of an expression frequently
4981 (to see how it changes), you might want to add it to the @dfn{automatic
4982 display list} so that @value{GDBN} prints its value each time your program stops.
4983 Each expression added to the list is given a number to identify it;
4984 to remove an expression from the list, you specify that number.
4985 The automatic display looks like this:
4989 3: bar[5] = (struct hack *) 0x3804
4993 This display shows item numbers, expressions and their current values. As with
4994 displays you request manually using @code{x} or @code{print}, you can
4995 specify the output format you prefer; in fact, @code{display} decides
4996 whether to use @code{print} or @code{x} depending on how elaborate your
4997 format specification is---it uses @code{x} if you specify a unit size,
4998 or one of the two formats (@samp{i} and @samp{s}) that are only
4999 supported by @code{x}; otherwise it uses @code{print}.
5003 @item display @var{expr}
5004 Add the expression @var{expr} to the list of expressions to display
5005 each time your program stops. @xref{Expressions, ,Expressions}.
5007 @code{display} does not repeat if you press @key{RET} again after using it.
5009 @item display/@var{fmt} @var{expr}
5010 For @var{fmt} specifying only a display format and not a size or
5011 count, add the expression @var{expr} to the auto-display list but
5012 arrange to display it each time in the specified format @var{fmt}.
5013 @xref{Output Formats,,Output formats}.
5015 @item display/@var{fmt} @var{addr}
5016 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
5017 number of units, add the expression @var{addr} as a memory address to
5018 be examined each time your program stops. Examining means in effect
5019 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
5022 For example, @samp{display/i $pc} can be helpful, to see the machine
5023 instruction about to be executed each time execution stops (@samp{$pc}
5024 is a common name for the program counter; @pxref{Registers, ,Registers}).
5027 @kindex delete display
5029 @item undisplay @var{dnums}@dots{}
5030 @itemx delete display @var{dnums}@dots{}
5031 Remove item numbers @var{dnums} from the list of expressions to display.
5033 @code{undisplay} does not repeat if you press @key{RET} after using it.
5034 (Otherwise you would just get the error @samp{No display number @dots{}}.)
5036 @kindex disable display
5037 @item disable display @var{dnums}@dots{}
5038 Disable the display of item numbers @var{dnums}. A disabled display
5039 item is not printed automatically, but is not forgotten. It may be
5040 enabled again later.
5042 @kindex enable display
5043 @item enable display @var{dnums}@dots{}
5044 Enable display of item numbers @var{dnums}. It becomes effective once
5045 again in auto display of its expression, until you specify otherwise.
5048 Display the current values of the expressions on the list, just as is
5049 done when your program stops.
5051 @kindex info display
5053 Print the list of expressions previously set up to display
5054 automatically, each one with its item number, but without showing the
5055 values. This includes disabled expressions, which are marked as such.
5056 It also includes expressions which would not be displayed right now
5057 because they refer to automatic variables not currently available.
5060 If a display expression refers to local variables, then it does not make
5061 sense outside the lexical context for which it was set up. Such an
5062 expression is disabled when execution enters a context where one of its
5063 variables is not defined. For example, if you give the command
5064 @code{display last_char} while inside a function with an argument
5065 @code{last_char}, @value{GDBN} displays this argument while your program
5066 continues to stop inside that function. When it stops elsewhere---where
5067 there is no variable @code{last_char}---the display is disabled
5068 automatically. The next time your program stops where @code{last_char}
5069 is meaningful, you can enable the display expression once again.
5071 @node Print Settings
5072 @section Print settings
5074 @cindex format options
5075 @cindex print settings
5076 @value{GDBN} provides the following ways to control how arrays, structures,
5077 and symbols are printed.
5080 These settings are useful for debugging programs in any language:
5083 @kindex set print address
5084 @item set print address
5085 @itemx set print address on
5086 @value{GDBN} prints memory addresses showing the location of stack
5087 traces, structure values, pointer values, breakpoints, and so forth,
5088 even when it also displays the contents of those addresses. The default
5089 is @code{on}. For example, this is what a stack frame display looks like with
5090 @code{set print address on}:
5095 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
5097 530 if (lquote != def_lquote)
5101 @item set print address off
5102 Do not print addresses when displaying their contents. For example,
5103 this is the same stack frame displayed with @code{set print address off}:
5107 (@value{GDBP}) set print addr off
5109 #0 set_quotes (lq="<<", rq=">>") at input.c:530
5110 530 if (lquote != def_lquote)
5114 You can use @samp{set print address off} to eliminate all machine
5115 dependent displays from the @value{GDBN} interface. For example, with
5116 @code{print address off}, you should get the same text for backtraces on
5117 all machines---whether or not they involve pointer arguments.
5119 @kindex show print address
5120 @item show print address
5121 Show whether or not addresses are to be printed.
5124 When @value{GDBN} prints a symbolic address, it normally prints the
5125 closest earlier symbol plus an offset. If that symbol does not uniquely
5126 identify the address (for example, it is a name whose scope is a single
5127 source file), you may need to clarify. One way to do this is with
5128 @code{info line}, for example @samp{info line *0x4537}. Alternately,
5129 you can set @value{GDBN} to print the source file and line number when
5130 it prints a symbolic address:
5133 @kindex set print symbol-filename
5134 @item set print symbol-filename on
5135 Tell @value{GDBN} to print the source file name and line number of a
5136 symbol in the symbolic form of an address.
5138 @item set print symbol-filename off
5139 Do not print source file name and line number of a symbol. This is the
5142 @kindex show print symbol-filename
5143 @item show print symbol-filename
5144 Show whether or not @value{GDBN} will print the source file name and
5145 line number of a symbol in the symbolic form of an address.
5148 Another situation where it is helpful to show symbol filenames and line
5149 numbers is when disassembling code; @value{GDBN} shows you the line
5150 number and source file that corresponds to each instruction.
5152 Also, you may wish to see the symbolic form only if the address being
5153 printed is reasonably close to the closest earlier symbol:
5156 @kindex set print max-symbolic-offset
5157 @item set print max-symbolic-offset @var{max-offset}
5158 Tell @value{GDBN} to only display the symbolic form of an address if the
5159 offset between the closest earlier symbol and the address is less than
5160 @var{max-offset}. The default is 0, which tells @value{GDBN}
5161 to always print the symbolic form of an address if any symbol precedes it.
5163 @kindex show print max-symbolic-offset
5164 @item show print max-symbolic-offset
5165 Ask how large the maximum offset is that @value{GDBN} prints in a
5169 @cindex wild pointer, interpreting
5170 @cindex pointer, finding referent
5171 If you have a pointer and you are not sure where it points, try
5172 @samp{set print symbol-filename on}. Then you can determine the name
5173 and source file location of the variable where it points, using
5174 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
5175 For example, here @value{GDBN} shows that a variable @code{ptt} points
5176 at another variable @code{t}, defined in @file{hi2.c}:
5179 (@value{GDBP}) set print symbol-filename on
5180 (@value{GDBP}) p/a ptt
5181 $4 = 0xe008 <t in hi2.c>
5185 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
5186 does not show the symbol name and filename of the referent, even with
5187 the appropriate @code{set print} options turned on.
5190 Other settings control how different kinds of objects are printed:
5193 @kindex set print array
5194 @item set print array
5195 @itemx set print array on
5196 Pretty print arrays. This format is more convenient to read,
5197 but uses more space. The default is off.
5199 @item set print array off
5200 Return to compressed format for arrays.
5202 @kindex show print array
5203 @item show print array
5204 Show whether compressed or pretty format is selected for displaying
5207 @kindex set print elements
5208 @item set print elements @var{number-of-elements}
5209 Set a limit on how many elements of an array @value{GDBN} will print.
5210 If @value{GDBN} is printing a large array, it stops printing after it has
5211 printed the number of elements set by the @code{set print elements} command.
5212 This limit also applies to the display of strings.
5213 When @value{GDBN} starts, this limit is set to 200.
5214 Setting @var{number-of-elements} to zero means that the printing is unlimited.
5216 @kindex show print elements
5217 @item show print elements
5218 Display the number of elements of a large array that @value{GDBN} will print.
5219 If the number is 0, then the printing is unlimited.
5221 @kindex set print null-stop
5222 @item set print null-stop
5223 Cause @value{GDBN} to stop printing the characters of an array when the first
5224 @sc{null} is encountered. This is useful when large arrays actually
5225 contain only short strings.
5228 @kindex set print pretty
5229 @item set print pretty on
5230 Cause @value{GDBN} to print structures in an indented format with one member
5231 per line, like this:
5246 @item set print pretty off
5247 Cause @value{GDBN} to print structures in a compact format, like this:
5251 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
5252 meat = 0x54 "Pork"@}
5257 This is the default format.
5259 @kindex show print pretty
5260 @item show print pretty
5261 Show which format @value{GDBN} is using to print structures.
5263 @kindex set print sevenbit-strings
5264 @item set print sevenbit-strings on
5265 Print using only seven-bit characters; if this option is set,
5266 @value{GDBN} displays any eight-bit characters (in strings or
5267 character values) using the notation @code{\}@var{nnn}. This setting is
5268 best if you are working in English (@sc{ascii}) and you use the
5269 high-order bit of characters as a marker or ``meta'' bit.
5271 @item set print sevenbit-strings off
5272 Print full eight-bit characters. This allows the use of more
5273 international character sets, and is the default.
5275 @kindex show print sevenbit-strings
5276 @item show print sevenbit-strings
5277 Show whether or not @value{GDBN} is printing only seven-bit characters.
5279 @kindex set print union
5280 @item set print union on
5281 Tell @value{GDBN} to print unions which are contained in structures. This
5282 is the default setting.
5284 @item set print union off
5285 Tell @value{GDBN} not to print unions which are contained in structures.
5287 @kindex show print union
5288 @item show print union
5289 Ask @value{GDBN} whether or not it will print unions which are contained in
5292 For example, given the declarations
5295 typedef enum @{Tree, Bug@} Species;
5296 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
5297 typedef enum @{Caterpillar, Cocoon, Butterfly@}
5308 struct thing foo = @{Tree, @{Acorn@}@};
5312 with @code{set print union on} in effect @samp{p foo} would print
5315 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
5319 and with @code{set print union off} in effect it would print
5322 $1 = @{it = Tree, form = @{...@}@}
5328 These settings are of interest when debugging C@t{++} programs:
5332 @kindex set print demangle
5333 @item set print demangle
5334 @itemx set print demangle on
5335 Print C@t{++} names in their source form rather than in the encoded
5336 (``mangled'') form passed to the assembler and linker for type-safe
5337 linkage. The default is on.
5339 @kindex show print demangle
5340 @item show print demangle
5341 Show whether C@t{++} names are printed in mangled or demangled form.
5343 @kindex set print asm-demangle
5344 @item set print asm-demangle
5345 @itemx set print asm-demangle on
5346 Print C@t{++} names in their source form rather than their mangled form, even
5347 in assembler code printouts such as instruction disassemblies.
5350 @kindex show print asm-demangle
5351 @item show print asm-demangle
5352 Show whether C@t{++} names in assembly listings are printed in mangled
5355 @kindex set demangle-style
5356 @cindex C@t{++} symbol decoding style
5357 @cindex symbol decoding style, C@t{++}
5358 @item set demangle-style @var{style}
5359 Choose among several encoding schemes used by different compilers to
5360 represent C@t{++} names. The choices for @var{style} are currently:
5364 Allow @value{GDBN} to choose a decoding style by inspecting your program.
5367 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
5368 This is the default.
5371 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
5374 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
5377 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
5378 @strong{Warning:} this setting alone is not sufficient to allow
5379 debugging @code{cfront}-generated executables. @value{GDBN} would
5380 require further enhancement to permit that.
5383 If you omit @var{style}, you will see a list of possible formats.
5385 @kindex show demangle-style
5386 @item show demangle-style
5387 Display the encoding style currently in use for decoding C@t{++} symbols.
5389 @kindex set print object
5390 @item set print object
5391 @itemx set print object on
5392 When displaying a pointer to an object, identify the @emph{actual}
5393 (derived) type of the object rather than the @emph{declared} type, using
5394 the virtual function table.
5396 @item set print object off
5397 Display only the declared type of objects, without reference to the
5398 virtual function table. This is the default setting.
5400 @kindex show print object
5401 @item show print object
5402 Show whether actual, or declared, object types are displayed.
5404 @kindex set print static-members
5405 @item set print static-members
5406 @itemx set print static-members on
5407 Print static members when displaying a C@t{++} object. The default is on.
5409 @item set print static-members off
5410 Do not print static members when displaying a C@t{++} object.
5412 @kindex show print static-members
5413 @item show print static-members
5414 Show whether C@t{++} static members are printed, or not.
5416 @c These don't work with HP ANSI C++ yet.
5417 @kindex set print vtbl
5418 @item set print vtbl
5419 @itemx set print vtbl on
5420 Pretty print C@t{++} virtual function tables. The default is off.
5421 (The @code{vtbl} commands do not work on programs compiled with the HP
5422 ANSI C@t{++} compiler (@code{aCC}).)
5424 @item set print vtbl off
5425 Do not pretty print C@t{++} virtual function tables.
5427 @kindex show print vtbl
5428 @item show print vtbl
5429 Show whether C@t{++} virtual function tables are pretty printed, or not.
5433 @section Value history
5435 @cindex value history
5436 Values printed by the @code{print} command are saved in the @value{GDBN}
5437 @dfn{value history}. This allows you to refer to them in other expressions.
5438 Values are kept until the symbol table is re-read or discarded
5439 (for example with the @code{file} or @code{symbol-file} commands).
5440 When the symbol table changes, the value history is discarded,
5441 since the values may contain pointers back to the types defined in the
5446 @cindex history number
5447 The values printed are given @dfn{history numbers} by which you can
5448 refer to them. These are successive integers starting with one.
5449 @code{print} shows you the history number assigned to a value by
5450 printing @samp{$@var{num} = } before the value; here @var{num} is the
5453 To refer to any previous value, use @samp{$} followed by the value's
5454 history number. The way @code{print} labels its output is designed to
5455 remind you of this. Just @code{$} refers to the most recent value in
5456 the history, and @code{$$} refers to the value before that.
5457 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
5458 is the value just prior to @code{$$}, @code{$$1} is equivalent to
5459 @code{$$}, and @code{$$0} is equivalent to @code{$}.
5461 For example, suppose you have just printed a pointer to a structure and
5462 want to see the contents of the structure. It suffices to type
5468 If you have a chain of structures where the component @code{next} points
5469 to the next one, you can print the contents of the next one with this:
5476 You can print successive links in the chain by repeating this
5477 command---which you can do by just typing @key{RET}.
5479 Note that the history records values, not expressions. If the value of
5480 @code{x} is 4 and you type these commands:
5488 then the value recorded in the value history by the @code{print} command
5489 remains 4 even though the value of @code{x} has changed.
5494 Print the last ten values in the value history, with their item numbers.
5495 This is like @samp{p@ $$9} repeated ten times, except that @code{show
5496 values} does not change the history.
5498 @item show values @var{n}
5499 Print ten history values centered on history item number @var{n}.
5502 Print ten history values just after the values last printed. If no more
5503 values are available, @code{show values +} produces no display.
5506 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
5507 same effect as @samp{show values +}.
5509 @node Convenience Vars
5510 @section Convenience variables
5512 @cindex convenience variables
5513 @value{GDBN} provides @dfn{convenience variables} that you can use within
5514 @value{GDBN} to hold on to a value and refer to it later. These variables
5515 exist entirely within @value{GDBN}; they are not part of your program, and
5516 setting a convenience variable has no direct effect on further execution
5517 of your program. That is why you can use them freely.
5519 Convenience variables are prefixed with @samp{$}. Any name preceded by
5520 @samp{$} can be used for a convenience variable, unless it is one of
5521 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
5522 (Value history references, in contrast, are @emph{numbers} preceded
5523 by @samp{$}. @xref{Value History, ,Value history}.)
5525 You can save a value in a convenience variable with an assignment
5526 expression, just as you would set a variable in your program.
5530 set $foo = *object_ptr
5534 would save in @code{$foo} the value contained in the object pointed to by
5537 Using a convenience variable for the first time creates it, but its
5538 value is @code{void} until you assign a new value. You can alter the
5539 value with another assignment at any time.
5541 Convenience variables have no fixed types. You can assign a convenience
5542 variable any type of value, including structures and arrays, even if
5543 that variable already has a value of a different type. The convenience
5544 variable, when used as an expression, has the type of its current value.
5547 @kindex show convenience
5548 @item show convenience
5549 Print a list of convenience variables used so far, and their values.
5550 Abbreviated @code{show conv}.
5553 One of the ways to use a convenience variable is as a counter to be
5554 incremented or a pointer to be advanced. For example, to print
5555 a field from successive elements of an array of structures:
5559 print bar[$i++]->contents
5563 Repeat that command by typing @key{RET}.
5565 Some convenience variables are created automatically by @value{GDBN} and given
5566 values likely to be useful.
5569 @vindex $_@r{, convenience variable}
5571 The variable @code{$_} is automatically set by the @code{x} command to
5572 the last address examined (@pxref{Memory, ,Examining memory}). Other
5573 commands which provide a default address for @code{x} to examine also
5574 set @code{$_} to that address; these commands include @code{info line}
5575 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
5576 except when set by the @code{x} command, in which case it is a pointer
5577 to the type of @code{$__}.
5579 @vindex $__@r{, convenience variable}
5581 The variable @code{$__} is automatically set by the @code{x} command
5582 to the value found in the last address examined. Its type is chosen
5583 to match the format in which the data was printed.
5586 @vindex $_exitcode@r{, convenience variable}
5587 The variable @code{$_exitcode} is automatically set to the exit code when
5588 the program being debugged terminates.
5591 On HP-UX systems, if you refer to a function or variable name that
5592 begins with a dollar sign, @value{GDBN} searches for a user or system
5593 name first, before it searches for a convenience variable.
5599 You can refer to machine register contents, in expressions, as variables
5600 with names starting with @samp{$}. The names of registers are different
5601 for each machine; use @code{info registers} to see the names used on
5605 @kindex info registers
5606 @item info registers
5607 Print the names and values of all registers except floating-point
5608 and vector registers (in the selected stack frame).
5610 @kindex info all-registers
5611 @cindex floating point registers
5612 @item info all-registers
5613 Print the names and values of all registers, including floating-point
5614 and vector registers (in the selected stack frame).
5616 @item info registers @var{regname} @dots{}
5617 Print the @dfn{relativized} value of each specified register @var{regname}.
5618 As discussed in detail below, register values are normally relative to
5619 the selected stack frame. @var{regname} may be any register name valid on
5620 the machine you are using, with or without the initial @samp{$}.
5623 @value{GDBN} has four ``standard'' register names that are available (in
5624 expressions) on most machines---whenever they do not conflict with an
5625 architecture's canonical mnemonics for registers. The register names
5626 @code{$pc} and @code{$sp} are used for the program counter register and
5627 the stack pointer. @code{$fp} is used for a register that contains a
5628 pointer to the current stack frame, and @code{$ps} is used for a
5629 register that contains the processor status. For example,
5630 you could print the program counter in hex with
5637 or print the instruction to be executed next with
5644 or add four to the stack pointer@footnote{This is a way of removing
5645 one word from the stack, on machines where stacks grow downward in
5646 memory (most machines, nowadays). This assumes that the innermost
5647 stack frame is selected; setting @code{$sp} is not allowed when other
5648 stack frames are selected. To pop entire frames off the stack,
5649 regardless of machine architecture, use @code{return};
5650 see @ref{Returning, ,Returning from a function}.} with
5656 Whenever possible, these four standard register names are available on
5657 your machine even though the machine has different canonical mnemonics,
5658 so long as there is no conflict. The @code{info registers} command
5659 shows the canonical names. For example, on the SPARC, @code{info
5660 registers} displays the processor status register as @code{$psr} but you
5661 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
5662 is an alias for the @sc{eflags} register.
5664 @value{GDBN} always considers the contents of an ordinary register as an
5665 integer when the register is examined in this way. Some machines have
5666 special registers which can hold nothing but floating point; these
5667 registers are considered to have floating point values. There is no way
5668 to refer to the contents of an ordinary register as floating point value
5669 (although you can @emph{print} it as a floating point value with
5670 @samp{print/f $@var{regname}}).
5672 Some registers have distinct ``raw'' and ``virtual'' data formats. This
5673 means that the data format in which the register contents are saved by
5674 the operating system is not the same one that your program normally
5675 sees. For example, the registers of the 68881 floating point
5676 coprocessor are always saved in ``extended'' (raw) format, but all C
5677 programs expect to work with ``double'' (virtual) format. In such
5678 cases, @value{GDBN} normally works with the virtual format only (the format
5679 that makes sense for your program), but the @code{info registers} command
5680 prints the data in both formats.
5682 Normally, register values are relative to the selected stack frame
5683 (@pxref{Selection, ,Selecting a frame}). This means that you get the
5684 value that the register would contain if all stack frames farther in
5685 were exited and their saved registers restored. In order to see the
5686 true contents of hardware registers, you must select the innermost
5687 frame (with @samp{frame 0}).
5689 However, @value{GDBN} must deduce where registers are saved, from the machine
5690 code generated by your compiler. If some registers are not saved, or if
5691 @value{GDBN} is unable to locate the saved registers, the selected stack
5692 frame makes no difference.
5694 @node Floating Point Hardware
5695 @section Floating point hardware
5696 @cindex floating point
5698 Depending on the configuration, @value{GDBN} may be able to give
5699 you more information about the status of the floating point hardware.
5704 Display hardware-dependent information about the floating
5705 point unit. The exact contents and layout vary depending on the
5706 floating point chip. Currently, @samp{info float} is supported on
5707 the ARM and x86 machines.
5711 @section Vector Unit
5714 Depending on the configuration, @value{GDBN} may be able to give you
5715 more information about the status of the vector unit.
5720 Display information about the vector unit. The exact contents and
5721 layout vary depending on the hardware.
5724 @node Memory Region Attributes
5725 @section Memory region attributes
5726 @cindex memory region attributes
5728 @dfn{Memory region attributes} allow you to describe special handling
5729 required by regions of your target's memory. @value{GDBN} uses attributes
5730 to determine whether to allow certain types of memory accesses; whether to
5731 use specific width accesses; and whether to cache target memory.
5733 Defined memory regions can be individually enabled and disabled. When a
5734 memory region is disabled, @value{GDBN} uses the default attributes when
5735 accessing memory in that region. Similarly, if no memory regions have
5736 been defined, @value{GDBN} uses the default attributes when accessing
5739 When a memory region is defined, it is given a number to identify it;
5740 to enable, disable, or remove a memory region, you specify that number.
5744 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
5745 Define memory region bounded by @var{lower} and @var{upper} with
5746 attributes @var{attributes}@dots{}. Note that @var{upper} == 0 is a
5747 special case: it is treated as the the target's maximum memory address.
5748 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
5751 @item delete mem @var{nums}@dots{}
5752 Remove memory regions @var{nums}@dots{}.
5755 @item disable mem @var{nums}@dots{}
5756 Disable memory regions @var{nums}@dots{}.
5757 A disabled memory region is not forgotten.
5758 It may be enabled again later.
5761 @item enable mem @var{nums}@dots{}
5762 Enable memory regions @var{nums}@dots{}.
5766 Print a table of all defined memory regions, with the following columns
5770 @item Memory Region Number
5771 @item Enabled or Disabled.
5772 Enabled memory regions are marked with @samp{y}.
5773 Disabled memory regions are marked with @samp{n}.
5776 The address defining the inclusive lower bound of the memory region.
5779 The address defining the exclusive upper bound of the memory region.
5782 The list of attributes set for this memory region.
5787 @subsection Attributes
5789 @subsubsection Memory Access Mode
5790 The access mode attributes set whether @value{GDBN} may make read or
5791 write accesses to a memory region.
5793 While these attributes prevent @value{GDBN} from performing invalid
5794 memory accesses, they do nothing to prevent the target system, I/O DMA,
5795 etc. from accessing memory.
5799 Memory is read only.
5801 Memory is write only.
5803 Memory is read/write. This is the default.
5806 @subsubsection Memory Access Size
5807 The acccess size attributes tells @value{GDBN} to use specific sized
5808 accesses in the memory region. Often memory mapped device registers
5809 require specific sized accesses. If no access size attribute is
5810 specified, @value{GDBN} may use accesses of any size.
5814 Use 8 bit memory accesses.
5816 Use 16 bit memory accesses.
5818 Use 32 bit memory accesses.
5820 Use 64 bit memory accesses.
5823 @c @subsubsection Hardware/Software Breakpoints
5824 @c The hardware/software breakpoint attributes set whether @value{GDBN}
5825 @c will use hardware or software breakpoints for the internal breakpoints
5826 @c used by the step, next, finish, until, etc. commands.
5830 @c Always use hardware breakpoints
5831 @c @item swbreak (default)
5834 @subsubsection Data Cache
5835 The data cache attributes set whether @value{GDBN} will cache target
5836 memory. While this generally improves performance by reducing debug
5837 protocol overhead, it can lead to incorrect results because @value{GDBN}
5838 does not know about volatile variables or memory mapped device
5843 Enable @value{GDBN} to cache target memory.
5845 Disable @value{GDBN} from caching target memory. This is the default.
5848 @c @subsubsection Memory Write Verification
5849 @c The memory write verification attributes set whether @value{GDBN}
5850 @c will re-reads data after each write to verify the write was successful.
5854 @c @item noverify (default)
5857 @node Dump/Restore Files
5858 @section Copy between memory and a file
5859 @cindex dump/restore files
5860 @cindex append data to a file
5861 @cindex dump data to a file
5862 @cindex restore data from a file
5867 The commands @code{dump}, @code{append}, and @code{restore} are used
5868 for copying data between target memory and a file. Data is written
5869 into a file using @code{dump} or @code{append}, and restored from a
5870 file into memory by using @code{restore}. Files may be binary, srec,
5871 intel hex, or tekhex (but only binary files can be appended).
5875 @kindex append binary
5876 @item dump binary memory @var{filename} @var{start_addr} @var{end_addr}
5877 Dump contents of memory from @var{start_addr} to @var{end_addr} into
5878 raw binary format file @var{filename}.
5880 @item append binary memory @var{filename} @var{start_addr} @var{end_addr}
5881 Append contents of memory from @var{start_addr} to @var{end_addr} to
5882 raw binary format file @var{filename}.
5884 @item dump binary value @var{filename} @var{expression}
5885 Dump value of @var{expression} into raw binary format file @var{filename}.
5887 @item append binary memory @var{filename} @var{expression}
5888 Append value of @var{expression} to raw binary format file @var{filename}.
5891 @item dump ihex memory @var{filename} @var{start_addr} @var{end_addr}
5892 Dump contents of memory from @var{start_addr} to @var{end_addr} into
5893 intel hex format file @var{filename}.
5895 @item dump ihex value @var{filename} @var{expression}
5896 Dump value of @var{expression} into intel hex format file @var{filename}.
5899 @item dump srec memory @var{filename} @var{start_addr} @var{end_addr}
5900 Dump contents of memory from @var{start_addr} to @var{end_addr} into
5901 srec format file @var{filename}.
5903 @item dump srec value @var{filename} @var{expression}
5904 Dump value of @var{expression} into srec format file @var{filename}.
5907 @item dump tekhex memory @var{filename} @var{start_addr} @var{end_addr}
5908 Dump contents of memory from @var{start_addr} to @var{end_addr} into
5909 tekhex format file @var{filename}.
5911 @item dump tekhex value @var{filename} @var{expression}
5912 Dump value of @var{expression} into tekhex format file @var{filename}.
5914 @item restore @var{filename} [@var{binary}] @var{bias} @var{start} @var{end}
5915 Restore the contents of file @var{filename} into memory. The @code{restore}
5916 command can automatically recognize any known bfd file format, except for
5917 raw binary. To restore a raw binary file you must use the optional argument
5918 @var{binary} after the filename.
5920 If @var{bias} is non-zero, its value will be added to the addresses
5921 contained in the file. Binary files always start at address zero, so
5922 they will be restored at address @var{bias}. Other bfd files have
5923 a built-in location; they will be restored at offset @var{bias}
5926 If @var{start} and/or @var{end} are non-zero, then only data between
5927 file offset @var{start} and file offset @var{end} will be restored.
5928 These offsets are relative to the addresses in the file, before
5929 the @var{bias} argument is applied.
5933 @node Character Sets
5934 @section Character Sets
5935 @cindex character sets
5937 @cindex translating between character sets
5938 @cindex host character set
5939 @cindex target character set
5941 If the program you are debugging uses a different character set to
5942 represent characters and strings than the one @value{GDBN} uses itself,
5943 @value{GDBN} can automatically translate between the character sets for
5944 you. The character set @value{GDBN} uses we call the @dfn{host
5945 character set}; the one the inferior program uses we call the
5946 @dfn{target character set}.
5948 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
5949 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
5950 remote protocol (@pxref{Remote,Remote Debugging}) to debug a program
5951 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
5952 then the host character set is Latin-1, and the target character set is
5953 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
5954 target-charset ebcdic-us}, then @value{GDBN} translates between
5955 @sc{ebcdic} and Latin 1 as you print character or string values, or use
5956 character and string literals in expressions.
5958 @value{GDBN} has no way to automatically recognize which character set
5959 the inferior program uses; you must tell it, using the @code{set
5960 target-charset} command, described below.
5962 Here are the commands for controlling @value{GDBN}'s character set
5966 @item set target-charset @var{charset}
5967 @kindex set target-charset
5968 Set the current target character set to @var{charset}. We list the
5969 character set names @value{GDBN} recognizes below, but if you invoke the
5970 @code{set target-charset} command with no argument, @value{GDBN} lists
5971 the character sets it supports.
5975 @item set host-charset @var{charset}
5976 @kindex set host-charset
5977 Set the current host character set to @var{charset}.
5979 By default, @value{GDBN} uses a host character set appropriate to the
5980 system it is running on; you can override that default using the
5981 @code{set host-charset} command.
5983 @value{GDBN} can only use certain character sets as its host character
5984 set. We list the character set names @value{GDBN} recognizes below, and
5985 indicate which can be host character sets, but if you invoke the
5986 @code{set host-charset} command with no argument, @value{GDBN} lists the
5987 character sets it supports, placing an asterisk (@samp{*}) after those
5988 it can use as a host character set.
5990 @item set charset @var{charset}
5992 Set the current host and target character sets to @var{charset}. If you
5993 invoke the @code{set charset} command with no argument, it lists the
5994 character sets it supports. @value{GDBN} can only use certain character
5995 sets as its host character set; it marks those in the list with an
5996 asterisk (@samp{*}).
5999 @itemx show host-charset
6000 @itemx show target-charset
6001 @kindex show charset
6002 @kindex show host-charset
6003 @kindex show target-charset
6004 Show the current host and target charsets. The @code{show host-charset}
6005 and @code{show target-charset} commands are synonyms for @code{show
6010 @value{GDBN} currently includes support for the following character
6016 @cindex ASCII character set
6017 Seven-bit U.S. @sc{ascii}. @value{GDBN} can use this as its host
6021 @cindex ISO 8859-1 character set
6022 @cindex ISO Latin 1 character set
6023 The ISO Latin 1 character set. This extends ASCII with accented
6024 characters needed for French, German, and Spanish. @value{GDBN} can use
6025 this as its host character set.
6029 @cindex EBCDIC character set
6030 @cindex IBM1047 character set
6031 Variants of the @sc{ebcdic} character set, used on some of IBM's
6032 mainframe operating systems. (@sc{gnu}/Linux on the S/390 uses U.S. @sc{ascii}.)
6033 @value{GDBN} cannot use these as its host character set.
6037 Note that these are all single-byte character sets. More work inside
6038 GDB is needed to support multi-byte or variable-width character
6039 encodings, like the UTF-8 and UCS-2 encodings of Unicode.
6041 Here is an example of @value{GDBN}'s character set support in action.
6042 Assume that the following source code has been placed in the file
6043 @file{charset-test.c}:
6049 = @{72, 101, 108, 108, 111, 44, 32, 119,
6050 111, 114, 108, 100, 33, 10, 0@};
6051 char ibm1047_hello[]
6052 = @{200, 133, 147, 147, 150, 107, 64, 166,
6053 150, 153, 147, 132, 90, 37, 0@};
6057 printf ("Hello, world!\n");
6061 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
6062 containing the string @samp{Hello, world!} followed by a newline,
6063 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
6065 We compile the program, and invoke the debugger on it:
6068 $ gcc -g charset-test.c -o charset-test
6069 $ gdb -nw charset-test
6070 GNU gdb 2001-12-19-cvs
6071 Copyright 2001 Free Software Foundation, Inc.
6076 We can use the @code{show charset} command to see what character sets
6077 @value{GDBN} is currently using to interpret and display characters and
6082 The current host and target character set is `iso-8859-1'.
6086 For the sake of printing this manual, let's use @sc{ascii} as our
6087 initial character set:
6089 (gdb) set charset ascii
6091 The current host and target character set is `ascii'.
6095 Let's assume that @sc{ascii} is indeed the correct character set for our
6096 host system --- in other words, let's assume that if @value{GDBN} prints
6097 characters using the @sc{ascii} character set, our terminal will display
6098 them properly. Since our current target character set is also
6099 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
6102 (gdb) print ascii_hello
6103 $1 = 0x401698 "Hello, world!\n"
6104 (gdb) print ascii_hello[0]
6109 @value{GDBN} uses the target character set for character and string
6110 literals you use in expressions:
6118 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
6121 @value{GDBN} relies on the user to tell it which character set the
6122 target program uses. If we print @code{ibm1047_hello} while our target
6123 character set is still @sc{ascii}, we get jibberish:
6126 (gdb) print ibm1047_hello
6127 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
6128 (gdb) print ibm1047_hello[0]
6133 If we invoke the @code{set target-charset} command without an argument,
6134 @value{GDBN} tells us the character sets it supports:
6137 (gdb) set target-charset
6138 Valid character sets are:
6143 * - can be used as a host character set
6146 We can select @sc{ibm1047} as our target character set, and examine the
6147 program's strings again. Now the @sc{ascii} string is wrong, but
6148 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
6149 target character set, @sc{ibm1047}, to the host character set,
6150 @sc{ascii}, and they display correctly:
6153 (gdb) set target-charset ibm1047
6155 The current host character set is `ascii'.
6156 The current target character set is `ibm1047'.
6157 (gdb) print ascii_hello
6158 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
6159 (gdb) print ascii_hello[0]
6161 (gdb) print ibm1047_hello
6162 $8 = 0x4016a8 "Hello, world!\n"
6163 (gdb) print ibm1047_hello[0]
6168 As above, @value{GDBN} uses the target character set for character and
6169 string literals you use in expressions:
6177 The IBM1047 character set uses the number 78 to encode the @samp{+}
6182 @chapter C Preprocessor Macros
6184 Some languages, such as C and C++, provide a way to define and invoke
6185 ``preprocessor macros'' which expand into strings of tokens.
6186 @value{GDBN} can evaluate expressions containing macro invocations, show
6187 the result of macro expansion, and show a macro's definition, including
6188 where it was defined.
6190 You may need to compile your program specially to provide @value{GDBN}
6191 with information about preprocessor macros. Most compilers do not
6192 include macros in their debugging information, even when you compile
6193 with the @option{-g} flag. @xref{Compilation}.
6195 A program may define a macro at one point, remove that definition later,
6196 and then provide a different definition after that. Thus, at different
6197 points in the program, a macro may have different definitions, or have
6198 no definition at all. If there is a current stack frame, @value{GDBN}
6199 uses the macros in scope at that frame's source code line. Otherwise,
6200 @value{GDBN} uses the macros in scope at the current listing location;
6203 At the moment, @value{GDBN} does not support the @code{##}
6204 token-splicing operator, the @code{#} stringification operator, or
6205 variable-arity macros.
6207 Whenever @value{GDBN} evaluates an expression, it always expands any
6208 macro invocations present in the expression. @value{GDBN} also provides
6209 the following commands for working with macros explicitly.
6213 @kindex macro expand
6214 @cindex macro expansion, showing the results of preprocessor
6215 @cindex preprocessor macro expansion, showing the results of
6216 @cindex expanding preprocessor macros
6217 @item macro expand @var{expression}
6218 @itemx macro exp @var{expression}
6219 Show the results of expanding all preprocessor macro invocations in
6220 @var{expression}. Since @value{GDBN} simply expands macros, but does
6221 not parse the result, @var{expression} need not be a valid expression;
6222 it can be any string of tokens.
6224 @kindex macro expand-once
6225 @item macro expand-once @var{expression}
6226 @itemx macro exp1 @var{expression}
6227 @i{(This command is not yet implemented.)} Show the results of
6228 expanding those preprocessor macro invocations that appear explicitly in
6229 @var{expression}. Macro invocations appearing in that expansion are
6230 left unchanged. This command allows you to see the effect of a
6231 particular macro more clearly, without being confused by further
6232 expansions. Since @value{GDBN} simply expands macros, but does not
6233 parse the result, @var{expression} need not be a valid expression; it
6234 can be any string of tokens.
6237 @cindex macro definition, showing
6238 @cindex definition, showing a macro's
6239 @item info macro @var{macro}
6240 Show the definition of the macro named @var{macro}, and describe the
6241 source location where that definition was established.
6243 @kindex macro define
6244 @cindex user-defined macros
6245 @cindex defining macros interactively
6246 @cindex macros, user-defined
6247 @item macro define @var{macro} @var{replacement-list}
6248 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
6249 @i{(This command is not yet implemented.)} Introduce a definition for a
6250 preprocessor macro named @var{macro}, invocations of which are replaced
6251 by the tokens given in @var{replacement-list}. The first form of this
6252 command defines an ``object-like'' macro, which takes no arguments; the
6253 second form defines a ``function-like'' macro, which takes the arguments
6254 given in @var{arglist}.
6256 A definition introduced by this command is in scope in every expression
6257 evaluated in @value{GDBN}, until it is removed with the @command{macro
6258 undef} command, described below. The definition overrides all
6259 definitions for @var{macro} present in the program being debugged, as
6260 well as any previous user-supplied definition.
6263 @item macro undef @var{macro}
6264 @i{(This command is not yet implemented.)} Remove any user-supplied
6265 definition for the macro named @var{macro}. This command only affects
6266 definitions provided with the @command{macro define} command, described
6267 above; it cannot remove definitions present in the program being
6272 @cindex macros, example of debugging with
6273 Here is a transcript showing the above commands in action. First, we
6274 show our source files:
6282 #define ADD(x) (M + x)
6287 printf ("Hello, world!\n");
6289 printf ("We're so creative.\n");
6291 printf ("Goodbye, world!\n");
6298 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
6299 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
6300 compiler includes information about preprocessor macros in the debugging
6304 $ gcc -gdwarf-2 -g3 sample.c -o sample
6308 Now, we start @value{GDBN} on our sample program:
6312 GNU gdb 2002-05-06-cvs
6313 Copyright 2002 Free Software Foundation, Inc.
6314 GDB is free software, @dots{}
6318 We can expand macros and examine their definitions, even when the
6319 program is not running. @value{GDBN} uses the current listing position
6320 to decide which macro definitions are in scope:
6326 5 #define ADD(x) (M + x)
6331 10 printf ("Hello, world!\n");
6333 12 printf ("We're so creative.\n");
6334 (gdb) info macro ADD
6335 Defined at /home/jimb/gdb/macros/play/sample.c:5
6336 #define ADD(x) (M + x)
6338 Defined at /home/jimb/gdb/macros/play/sample.h:1
6339 included at /home/jimb/gdb/macros/play/sample.c:2
6341 (gdb) macro expand ADD(1)
6342 expands to: (42 + 1)
6343 (gdb) macro expand-once ADD(1)
6344 expands to: once (M + 1)
6348 In the example above, note that @command{macro expand-once} expands only
6349 the macro invocation explicit in the original text --- the invocation of
6350 @code{ADD} --- but does not expand the invocation of the macro @code{M},
6351 which was introduced by @code{ADD}.
6353 Once the program is running, GDB uses the macro definitions in force at
6354 the source line of the current stack frame:
6358 Breakpoint 1 at 0x8048370: file sample.c, line 10.
6360 Starting program: /home/jimb/gdb/macros/play/sample
6362 Breakpoint 1, main () at sample.c:10
6363 10 printf ("Hello, world!\n");
6367 At line 10, the definition of the macro @code{N} at line 9 is in force:
6371 Defined at /home/jimb/gdb/macros/play/sample.c:9
6373 (gdb) macro expand N Q M
6380 As we step over directives that remove @code{N}'s definition, and then
6381 give it a new definition, @value{GDBN} finds the definition (or lack
6382 thereof) in force at each point:
6387 12 printf ("We're so creative.\n");
6389 The symbol `N' has no definition as a C/C++ preprocessor macro
6390 at /home/jimb/gdb/macros/play/sample.c:12
6393 14 printf ("Goodbye, world!\n");
6395 Defined at /home/jimb/gdb/macros/play/sample.c:13
6397 (gdb) macro expand N Q M
6398 expands to: 1729 < 42
6406 @chapter Tracepoints
6407 @c This chapter is based on the documentation written by Michael
6408 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
6411 In some applications, it is not feasible for the debugger to interrupt
6412 the program's execution long enough for the developer to learn
6413 anything helpful about its behavior. If the program's correctness
6414 depends on its real-time behavior, delays introduced by a debugger
6415 might cause the program to change its behavior drastically, or perhaps
6416 fail, even when the code itself is correct. It is useful to be able
6417 to observe the program's behavior without interrupting it.
6419 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
6420 specify locations in the program, called @dfn{tracepoints}, and
6421 arbitrary expressions to evaluate when those tracepoints are reached.
6422 Later, using the @code{tfind} command, you can examine the values
6423 those expressions had when the program hit the tracepoints. The
6424 expressions may also denote objects in memory---structures or arrays,
6425 for example---whose values @value{GDBN} should record; while visiting
6426 a particular tracepoint, you may inspect those objects as if they were
6427 in memory at that moment. However, because @value{GDBN} records these
6428 values without interacting with you, it can do so quickly and
6429 unobtrusively, hopefully not disturbing the program's behavior.
6431 The tracepoint facility is currently available only for remote
6432 targets. @xref{Targets}. In addition, your remote target must know how
6433 to collect trace data. This functionality is implemented in the remote
6434 stub; however, none of the stubs distributed with @value{GDBN} support
6435 tracepoints as of this writing.
6437 This chapter describes the tracepoint commands and features.
6441 * Analyze Collected Data::
6442 * Tracepoint Variables::
6445 @node Set Tracepoints
6446 @section Commands to Set Tracepoints
6448 Before running such a @dfn{trace experiment}, an arbitrary number of
6449 tracepoints can be set. Like a breakpoint (@pxref{Set Breaks}), a
6450 tracepoint has a number assigned to it by @value{GDBN}. Like with
6451 breakpoints, tracepoint numbers are successive integers starting from
6452 one. Many of the commands associated with tracepoints take the
6453 tracepoint number as their argument, to identify which tracepoint to
6456 For each tracepoint, you can specify, in advance, some arbitrary set
6457 of data that you want the target to collect in the trace buffer when
6458 it hits that tracepoint. The collected data can include registers,
6459 local variables, or global data. Later, you can use @value{GDBN}
6460 commands to examine the values these data had at the time the
6463 This section describes commands to set tracepoints and associated
6464 conditions and actions.
6467 * Create and Delete Tracepoints::
6468 * Enable and Disable Tracepoints::
6469 * Tracepoint Passcounts::
6470 * Tracepoint Actions::
6471 * Listing Tracepoints::
6472 * Starting and Stopping Trace Experiment::
6475 @node Create and Delete Tracepoints
6476 @subsection Create and Delete Tracepoints
6479 @cindex set tracepoint
6482 The @code{trace} command is very similar to the @code{break} command.
6483 Its argument can be a source line, a function name, or an address in
6484 the target program. @xref{Set Breaks}. The @code{trace} command
6485 defines a tracepoint, which is a point in the target program where the
6486 debugger will briefly stop, collect some data, and then allow the
6487 program to continue. Setting a tracepoint or changing its commands
6488 doesn't take effect until the next @code{tstart} command; thus, you
6489 cannot change the tracepoint attributes once a trace experiment is
6492 Here are some examples of using the @code{trace} command:
6495 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
6497 (@value{GDBP}) @b{trace +2} // 2 lines forward
6499 (@value{GDBP}) @b{trace my_function} // first source line of function
6501 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
6503 (@value{GDBP}) @b{trace *0x2117c4} // an address
6507 You can abbreviate @code{trace} as @code{tr}.
6510 @cindex last tracepoint number
6511 @cindex recent tracepoint number
6512 @cindex tracepoint number
6513 The convenience variable @code{$tpnum} records the tracepoint number
6514 of the most recently set tracepoint.
6516 @kindex delete tracepoint
6517 @cindex tracepoint deletion
6518 @item delete tracepoint @r{[}@var{num}@r{]}
6519 Permanently delete one or more tracepoints. With no argument, the
6520 default is to delete all tracepoints.
6525 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
6527 (@value{GDBP}) @b{delete trace} // remove all tracepoints
6531 You can abbreviate this command as @code{del tr}.
6534 @node Enable and Disable Tracepoints
6535 @subsection Enable and Disable Tracepoints
6538 @kindex disable tracepoint
6539 @item disable tracepoint @r{[}@var{num}@r{]}
6540 Disable tracepoint @var{num}, or all tracepoints if no argument
6541 @var{num} is given. A disabled tracepoint will have no effect during
6542 the next trace experiment, but it is not forgotten. You can re-enable
6543 a disabled tracepoint using the @code{enable tracepoint} command.
6545 @kindex enable tracepoint
6546 @item enable tracepoint @r{[}@var{num}@r{]}
6547 Enable tracepoint @var{num}, or all tracepoints. The enabled
6548 tracepoints will become effective the next time a trace experiment is
6552 @node Tracepoint Passcounts
6553 @subsection Tracepoint Passcounts
6557 @cindex tracepoint pass count
6558 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
6559 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
6560 automatically stop a trace experiment. If a tracepoint's passcount is
6561 @var{n}, then the trace experiment will be automatically stopped on
6562 the @var{n}'th time that tracepoint is hit. If the tracepoint number
6563 @var{num} is not specified, the @code{passcount} command sets the
6564 passcount of the most recently defined tracepoint. If no passcount is
6565 given, the trace experiment will run until stopped explicitly by the
6571 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
6572 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
6574 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
6575 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
6576 (@value{GDBP}) @b{trace foo}
6577 (@value{GDBP}) @b{pass 3}
6578 (@value{GDBP}) @b{trace bar}
6579 (@value{GDBP}) @b{pass 2}
6580 (@value{GDBP}) @b{trace baz}
6581 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
6582 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
6583 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
6584 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
6588 @node Tracepoint Actions
6589 @subsection Tracepoint Action Lists
6593 @cindex tracepoint actions
6594 @item actions @r{[}@var{num}@r{]}
6595 This command will prompt for a list of actions to be taken when the
6596 tracepoint is hit. If the tracepoint number @var{num} is not
6597 specified, this command sets the actions for the one that was most
6598 recently defined (so that you can define a tracepoint and then say
6599 @code{actions} without bothering about its number). You specify the
6600 actions themselves on the following lines, one action at a time, and
6601 terminate the actions list with a line containing just @code{end}. So
6602 far, the only defined actions are @code{collect} and
6603 @code{while-stepping}.
6605 @cindex remove actions from a tracepoint
6606 To remove all actions from a tracepoint, type @samp{actions @var{num}}
6607 and follow it immediately with @samp{end}.
6610 (@value{GDBP}) @b{collect @var{data}} // collect some data
6612 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
6614 (@value{GDBP}) @b{end} // signals the end of actions.
6617 In the following example, the action list begins with @code{collect}
6618 commands indicating the things to be collected when the tracepoint is
6619 hit. Then, in order to single-step and collect additional data
6620 following the tracepoint, a @code{while-stepping} command is used,
6621 followed by the list of things to be collected while stepping. The
6622 @code{while-stepping} command is terminated by its own separate
6623 @code{end} command. Lastly, the action list is terminated by an
6627 (@value{GDBP}) @b{trace foo}
6628 (@value{GDBP}) @b{actions}
6629 Enter actions for tracepoint 1, one per line:
6638 @kindex collect @r{(tracepoints)}
6639 @item collect @var{expr1}, @var{expr2}, @dots{}
6640 Collect values of the given expressions when the tracepoint is hit.
6641 This command accepts a comma-separated list of any valid expressions.
6642 In addition to global, static, or local variables, the following
6643 special arguments are supported:
6647 collect all registers
6650 collect all function arguments
6653 collect all local variables.
6656 You can give several consecutive @code{collect} commands, each one
6657 with a single argument, or one @code{collect} command with several
6658 arguments separated by commas: the effect is the same.
6660 The command @code{info scope} (@pxref{Symbols, info scope}) is
6661 particularly useful for figuring out what data to collect.
6663 @kindex while-stepping @r{(tracepoints)}
6664 @item while-stepping @var{n}
6665 Perform @var{n} single-step traces after the tracepoint, collecting
6666 new data at each step. The @code{while-stepping} command is
6667 followed by the list of what to collect while stepping (followed by
6668 its own @code{end} command):
6672 > collect $regs, myglobal
6678 You may abbreviate @code{while-stepping} as @code{ws} or
6682 @node Listing Tracepoints
6683 @subsection Listing Tracepoints
6686 @kindex info tracepoints
6687 @cindex information about tracepoints
6688 @item info tracepoints @r{[}@var{num}@r{]}
6689 Display information about the tracepoint @var{num}. If you don't specify
6690 a tracepoint number, displays information about all the tracepoints
6691 defined so far. For each tracepoint, the following information is
6698 whether it is enabled or disabled
6702 its passcount as given by the @code{passcount @var{n}} command
6704 its step count as given by the @code{while-stepping @var{n}} command
6706 where in the source files is the tracepoint set
6708 its action list as given by the @code{actions} command
6712 (@value{GDBP}) @b{info trace}
6713 Num Enb Address PassC StepC What
6714 1 y 0x002117c4 0 0 <gdb_asm>
6715 2 y 0x0020dc64 0 0 in g_test at g_test.c:1375
6716 3 y 0x0020b1f4 0 0 in get_data at ../foo.c:41
6721 This command can be abbreviated @code{info tp}.
6724 @node Starting and Stopping Trace Experiment
6725 @subsection Starting and Stopping Trace Experiment
6729 @cindex start a new trace experiment
6730 @cindex collected data discarded
6732 This command takes no arguments. It starts the trace experiment, and
6733 begins collecting data. This has the side effect of discarding all
6734 the data collected in the trace buffer during the previous trace
6738 @cindex stop a running trace experiment
6740 This command takes no arguments. It ends the trace experiment, and
6741 stops collecting data.
6743 @strong{Note:} a trace experiment and data collection may stop
6744 automatically if any tracepoint's passcount is reached
6745 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
6748 @cindex status of trace data collection
6749 @cindex trace experiment, status of
6751 This command displays the status of the current trace data
6755 Here is an example of the commands we described so far:
6758 (@value{GDBP}) @b{trace gdb_c_test}
6759 (@value{GDBP}) @b{actions}
6760 Enter actions for tracepoint #1, one per line.
6761 > collect $regs,$locals,$args
6766 (@value{GDBP}) @b{tstart}
6767 [time passes @dots{}]
6768 (@value{GDBP}) @b{tstop}
6772 @node Analyze Collected Data
6773 @section Using the collected data
6775 After the tracepoint experiment ends, you use @value{GDBN} commands
6776 for examining the trace data. The basic idea is that each tracepoint
6777 collects a trace @dfn{snapshot} every time it is hit and another
6778 snapshot every time it single-steps. All these snapshots are
6779 consecutively numbered from zero and go into a buffer, and you can
6780 examine them later. The way you examine them is to @dfn{focus} on a
6781 specific trace snapshot. When the remote stub is focused on a trace
6782 snapshot, it will respond to all @value{GDBN} requests for memory and
6783 registers by reading from the buffer which belongs to that snapshot,
6784 rather than from @emph{real} memory or registers of the program being
6785 debugged. This means that @strong{all} @value{GDBN} commands
6786 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
6787 behave as if we were currently debugging the program state as it was
6788 when the tracepoint occurred. Any requests for data that are not in
6789 the buffer will fail.
6792 * tfind:: How to select a trace snapshot
6793 * tdump:: How to display all data for a snapshot
6794 * save-tracepoints:: How to save tracepoints for a future run
6798 @subsection @code{tfind @var{n}}
6801 @cindex select trace snapshot
6802 @cindex find trace snapshot
6803 The basic command for selecting a trace snapshot from the buffer is
6804 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
6805 counting from zero. If no argument @var{n} is given, the next
6806 snapshot is selected.
6808 Here are the various forms of using the @code{tfind} command.
6812 Find the first snapshot in the buffer. This is a synonym for
6813 @code{tfind 0} (since 0 is the number of the first snapshot).
6816 Stop debugging trace snapshots, resume @emph{live} debugging.
6819 Same as @samp{tfind none}.
6822 No argument means find the next trace snapshot.
6825 Find the previous trace snapshot before the current one. This permits
6826 retracing earlier steps.
6828 @item tfind tracepoint @var{num}
6829 Find the next snapshot associated with tracepoint @var{num}. Search
6830 proceeds forward from the last examined trace snapshot. If no
6831 argument @var{num} is given, it means find the next snapshot collected
6832 for the same tracepoint as the current snapshot.
6834 @item tfind pc @var{addr}
6835 Find the next snapshot associated with the value @var{addr} of the
6836 program counter. Search proceeds forward from the last examined trace
6837 snapshot. If no argument @var{addr} is given, it means find the next
6838 snapshot with the same value of PC as the current snapshot.
6840 @item tfind outside @var{addr1}, @var{addr2}
6841 Find the next snapshot whose PC is outside the given range of
6844 @item tfind range @var{addr1}, @var{addr2}
6845 Find the next snapshot whose PC is between @var{addr1} and
6846 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
6848 @item tfind line @r{[}@var{file}:@r{]}@var{n}
6849 Find the next snapshot associated with the source line @var{n}. If
6850 the optional argument @var{file} is given, refer to line @var{n} in
6851 that source file. Search proceeds forward from the last examined
6852 trace snapshot. If no argument @var{n} is given, it means find the
6853 next line other than the one currently being examined; thus saying
6854 @code{tfind line} repeatedly can appear to have the same effect as
6855 stepping from line to line in a @emph{live} debugging session.
6858 The default arguments for the @code{tfind} commands are specifically
6859 designed to make it easy to scan through the trace buffer. For
6860 instance, @code{tfind} with no argument selects the next trace
6861 snapshot, and @code{tfind -} with no argument selects the previous
6862 trace snapshot. So, by giving one @code{tfind} command, and then
6863 simply hitting @key{RET} repeatedly you can examine all the trace
6864 snapshots in order. Or, by saying @code{tfind -} and then hitting
6865 @key{RET} repeatedly you can examine the snapshots in reverse order.
6866 The @code{tfind line} command with no argument selects the snapshot
6867 for the next source line executed. The @code{tfind pc} command with
6868 no argument selects the next snapshot with the same program counter
6869 (PC) as the current frame. The @code{tfind tracepoint} command with
6870 no argument selects the next trace snapshot collected by the same
6871 tracepoint as the current one.
6873 In addition to letting you scan through the trace buffer manually,
6874 these commands make it easy to construct @value{GDBN} scripts that
6875 scan through the trace buffer and print out whatever collected data
6876 you are interested in. Thus, if we want to examine the PC, FP, and SP
6877 registers from each trace frame in the buffer, we can say this:
6880 (@value{GDBP}) @b{tfind start}
6881 (@value{GDBP}) @b{while ($trace_frame != -1)}
6882 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
6883 $trace_frame, $pc, $sp, $fp
6887 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
6888 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
6889 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
6890 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
6891 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
6892 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
6893 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
6894 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
6895 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
6896 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
6897 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
6900 Or, if we want to examine the variable @code{X} at each source line in
6904 (@value{GDBP}) @b{tfind start}
6905 (@value{GDBP}) @b{while ($trace_frame != -1)}
6906 > printf "Frame %d, X == %d\n", $trace_frame, X
6916 @subsection @code{tdump}
6918 @cindex dump all data collected at tracepoint
6919 @cindex tracepoint data, display
6921 This command takes no arguments. It prints all the data collected at
6922 the current trace snapshot.
6925 (@value{GDBP}) @b{trace 444}
6926 (@value{GDBP}) @b{actions}
6927 Enter actions for tracepoint #2, one per line:
6928 > collect $regs, $locals, $args, gdb_long_test
6931 (@value{GDBP}) @b{tstart}
6933 (@value{GDBP}) @b{tfind line 444}
6934 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
6936 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
6938 (@value{GDBP}) @b{tdump}
6939 Data collected at tracepoint 2, trace frame 1:
6940 d0 0xc4aa0085 -995491707
6944 d4 0x71aea3d 119204413
6949 a1 0x3000668 50333288
6952 a4 0x3000698 50333336
6954 fp 0x30bf3c 0x30bf3c
6955 sp 0x30bf34 0x30bf34
6957 pc 0x20b2c8 0x20b2c8
6961 p = 0x20e5b4 "gdb-test"
6968 gdb_long_test = 17 '\021'
6973 @node save-tracepoints
6974 @subsection @code{save-tracepoints @var{filename}}
6975 @kindex save-tracepoints
6976 @cindex save tracepoints for future sessions
6978 This command saves all current tracepoint definitions together with
6979 their actions and passcounts, into a file @file{@var{filename}}
6980 suitable for use in a later debugging session. To read the saved
6981 tracepoint definitions, use the @code{source} command (@pxref{Command
6984 @node Tracepoint Variables
6985 @section Convenience Variables for Tracepoints
6986 @cindex tracepoint variables
6987 @cindex convenience variables for tracepoints
6990 @vindex $trace_frame
6991 @item (int) $trace_frame
6992 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
6993 snapshot is selected.
6996 @item (int) $tracepoint
6997 The tracepoint for the current trace snapshot.
7000 @item (int) $trace_line
7001 The line number for the current trace snapshot.
7004 @item (char []) $trace_file
7005 The source file for the current trace snapshot.
7008 @item (char []) $trace_func
7009 The name of the function containing @code{$tracepoint}.
7012 Note: @code{$trace_file} is not suitable for use in @code{printf},
7013 use @code{output} instead.
7015 Here's a simple example of using these convenience variables for
7016 stepping through all the trace snapshots and printing some of their
7020 (@value{GDBP}) @b{tfind start}
7022 (@value{GDBP}) @b{while $trace_frame != -1}
7023 > output $trace_file
7024 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
7030 @chapter Debugging Programs That Use Overlays
7033 If your program is too large to fit completely in your target system's
7034 memory, you can sometimes use @dfn{overlays} to work around this
7035 problem. @value{GDBN} provides some support for debugging programs that
7039 * How Overlays Work:: A general explanation of overlays.
7040 * Overlay Commands:: Managing overlays in @value{GDBN}.
7041 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
7042 mapped by asking the inferior.
7043 * Overlay Sample Program:: A sample program using overlays.
7046 @node How Overlays Work
7047 @section How Overlays Work
7048 @cindex mapped overlays
7049 @cindex unmapped overlays
7050 @cindex load address, overlay's
7051 @cindex mapped address
7052 @cindex overlay area
7054 Suppose you have a computer whose instruction address space is only 64
7055 kilobytes long, but which has much more memory which can be accessed by
7056 other means: special instructions, segment registers, or memory
7057 management hardware, for example. Suppose further that you want to
7058 adapt a program which is larger than 64 kilobytes to run on this system.
7060 One solution is to identify modules of your program which are relatively
7061 independent, and need not call each other directly; call these modules
7062 @dfn{overlays}. Separate the overlays from the main program, and place
7063 their machine code in the larger memory. Place your main program in
7064 instruction memory, but leave at least enough space there to hold the
7065 largest overlay as well.
7067 Now, to call a function located in an overlay, you must first copy that
7068 overlay's machine code from the large memory into the space set aside
7069 for it in the instruction memory, and then jump to its entry point
7072 @c NB: In the below the mapped area's size is greater or equal to the
7073 @c size of all overlays. This is intentional to remind the developer
7074 @c that overlays don't necessarily need to be the same size.
7078 Data Instruction Larger
7079 Address Space Address Space Address Space
7080 +-----------+ +-----------+ +-----------+
7082 +-----------+ +-----------+ +-----------+<-- overlay 1
7083 | program | | main | .----| overlay 1 | load address
7084 | variables | | program | | +-----------+
7085 | and heap | | | | | |
7086 +-----------+ | | | +-----------+<-- overlay 2
7087 | | +-----------+ | | | load address
7088 +-----------+ | | | .-| overlay 2 |
7090 mapped --->+-----------+ | | +-----------+
7092 | overlay | <-' | | |
7093 | area | <---' +-----------+<-- overlay 3
7094 | | <---. | | load address
7095 +-----------+ `--| overlay 3 |
7102 @anchor{A code overlay}A code overlay
7106 The diagram (@pxref{A code overlay}) shows a system with separate data
7107 and instruction address spaces. To map an overlay, the program copies
7108 its code from the larger address space to the instruction address space.
7109 Since the overlays shown here all use the same mapped address, only one
7110 may be mapped at a time. For a system with a single address space for
7111 data and instructions, the diagram would be similar, except that the
7112 program variables and heap would share an address space with the main
7113 program and the overlay area.
7115 An overlay loaded into instruction memory and ready for use is called a
7116 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
7117 instruction memory. An overlay not present (or only partially present)
7118 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
7119 is its address in the larger memory. The mapped address is also called
7120 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
7121 called the @dfn{load memory address}, or @dfn{LMA}.
7123 Unfortunately, overlays are not a completely transparent way to adapt a
7124 program to limited instruction memory. They introduce a new set of
7125 global constraints you must keep in mind as you design your program:
7130 Before calling or returning to a function in an overlay, your program
7131 must make sure that overlay is actually mapped. Otherwise, the call or
7132 return will transfer control to the right address, but in the wrong
7133 overlay, and your program will probably crash.
7136 If the process of mapping an overlay is expensive on your system, you
7137 will need to choose your overlays carefully to minimize their effect on
7138 your program's performance.
7141 The executable file you load onto your system must contain each
7142 overlay's instructions, appearing at the overlay's load address, not its
7143 mapped address. However, each overlay's instructions must be relocated
7144 and its symbols defined as if the overlay were at its mapped address.
7145 You can use GNU linker scripts to specify different load and relocation
7146 addresses for pieces of your program; see @ref{Overlay Description,,,
7147 ld.info, Using ld: the GNU linker}.
7150 The procedure for loading executable files onto your system must be able
7151 to load their contents into the larger address space as well as the
7152 instruction and data spaces.
7156 The overlay system described above is rather simple, and could be
7157 improved in many ways:
7162 If your system has suitable bank switch registers or memory management
7163 hardware, you could use those facilities to make an overlay's load area
7164 contents simply appear at their mapped address in instruction space.
7165 This would probably be faster than copying the overlay to its mapped
7166 area in the usual way.
7169 If your overlays are small enough, you could set aside more than one
7170 overlay area, and have more than one overlay mapped at a time.
7173 You can use overlays to manage data, as well as instructions. In
7174 general, data overlays are even less transparent to your design than
7175 code overlays: whereas code overlays only require care when you call or
7176 return to functions, data overlays require care every time you access
7177 the data. Also, if you change the contents of a data overlay, you
7178 must copy its contents back out to its load address before you can copy a
7179 different data overlay into the same mapped area.
7184 @node Overlay Commands
7185 @section Overlay Commands
7187 To use @value{GDBN}'s overlay support, each overlay in your program must
7188 correspond to a separate section of the executable file. The section's
7189 virtual memory address and load memory address must be the overlay's
7190 mapped and load addresses. Identifying overlays with sections allows
7191 @value{GDBN} to determine the appropriate address of a function or
7192 variable, depending on whether the overlay is mapped or not.
7194 @value{GDBN}'s overlay commands all start with the word @code{overlay};
7195 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
7200 Disable @value{GDBN}'s overlay support. When overlay support is
7201 disabled, @value{GDBN} assumes that all functions and variables are
7202 always present at their mapped addresses. By default, @value{GDBN}'s
7203 overlay support is disabled.
7205 @item overlay manual
7206 @kindex overlay manual
7207 @cindex manual overlay debugging
7208 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
7209 relies on you to tell it which overlays are mapped, and which are not,
7210 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
7211 commands described below.
7213 @item overlay map-overlay @var{overlay}
7214 @itemx overlay map @var{overlay}
7215 @kindex overlay map-overlay
7216 @cindex map an overlay
7217 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
7218 be the name of the object file section containing the overlay. When an
7219 overlay is mapped, @value{GDBN} assumes it can find the overlay's
7220 functions and variables at their mapped addresses. @value{GDBN} assumes
7221 that any other overlays whose mapped ranges overlap that of
7222 @var{overlay} are now unmapped.
7224 @item overlay unmap-overlay @var{overlay}
7225 @itemx overlay unmap @var{overlay}
7226 @kindex overlay unmap-overlay
7227 @cindex unmap an overlay
7228 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
7229 must be the name of the object file section containing the overlay.
7230 When an overlay is unmapped, @value{GDBN} assumes it can find the
7231 overlay's functions and variables at their load addresses.
7234 @kindex overlay auto
7235 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
7236 consults a data structure the overlay manager maintains in the inferior
7237 to see which overlays are mapped. For details, see @ref{Automatic
7240 @item overlay load-target
7242 @kindex overlay load-target
7243 @cindex reloading the overlay table
7244 Re-read the overlay table from the inferior. Normally, @value{GDBN}
7245 re-reads the table @value{GDBN} automatically each time the inferior
7246 stops, so this command should only be necessary if you have changed the
7247 overlay mapping yourself using @value{GDBN}. This command is only
7248 useful when using automatic overlay debugging.
7250 @item overlay list-overlays
7252 @cindex listing mapped overlays
7253 Display a list of the overlays currently mapped, along with their mapped
7254 addresses, load addresses, and sizes.
7258 Normally, when @value{GDBN} prints a code address, it includes the name
7259 of the function the address falls in:
7263 $3 = @{int ()@} 0x11a0 <main>
7266 When overlay debugging is enabled, @value{GDBN} recognizes code in
7267 unmapped overlays, and prints the names of unmapped functions with
7268 asterisks around them. For example, if @code{foo} is a function in an
7269 unmapped overlay, @value{GDBN} prints it this way:
7273 No sections are mapped.
7275 $5 = @{int (int)@} 0x100000 <*foo*>
7278 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
7283 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
7284 mapped at 0x1016 - 0x104a
7286 $6 = @{int (int)@} 0x1016 <foo>
7289 When overlay debugging is enabled, @value{GDBN} can find the correct
7290 address for functions and variables in an overlay, whether or not the
7291 overlay is mapped. This allows most @value{GDBN} commands, like
7292 @code{break} and @code{disassemble}, to work normally, even on unmapped
7293 code. However, @value{GDBN}'s breakpoint support has some limitations:
7297 @cindex breakpoints in overlays
7298 @cindex overlays, setting breakpoints in
7299 You can set breakpoints in functions in unmapped overlays, as long as
7300 @value{GDBN} can write to the overlay at its load address.
7302 @value{GDBN} can not set hardware or simulator-based breakpoints in
7303 unmapped overlays. However, if you set a breakpoint at the end of your
7304 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
7305 you are using manual overlay management), @value{GDBN} will re-set its
7306 breakpoints properly.
7310 @node Automatic Overlay Debugging
7311 @section Automatic Overlay Debugging
7312 @cindex automatic overlay debugging
7314 @value{GDBN} can automatically track which overlays are mapped and which
7315 are not, given some simple co-operation from the overlay manager in the
7316 inferior. If you enable automatic overlay debugging with the
7317 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
7318 looks in the inferior's memory for certain variables describing the
7319 current state of the overlays.
7321 Here are the variables your overlay manager must define to support
7322 @value{GDBN}'s automatic overlay debugging:
7326 @item @code{_ovly_table}:
7327 This variable must be an array of the following structures:
7332 /* The overlay's mapped address. */
7335 /* The size of the overlay, in bytes. */
7338 /* The overlay's load address. */
7341 /* Non-zero if the overlay is currently mapped;
7343 unsigned long mapped;
7347 @item @code{_novlys}:
7348 This variable must be a four-byte signed integer, holding the total
7349 number of elements in @code{_ovly_table}.
7353 To decide whether a particular overlay is mapped or not, @value{GDBN}
7354 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
7355 @code{lma} members equal the VMA and LMA of the overlay's section in the
7356 executable file. When @value{GDBN} finds a matching entry, it consults
7357 the entry's @code{mapped} member to determine whether the overlay is
7360 In addition, your overlay manager may define a function called
7361 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
7362 will silently set a breakpoint there. If the overlay manager then
7363 calls this function whenever it has changed the overlay table, this
7364 will enable @value{GDBN} to accurately keep track of which overlays
7365 are in program memory, and update any breakpoints that may be set
7366 in overlays. This will allow breakpoints to work even if the
7367 overlays are kept in ROM or other non-writable memory while they
7368 are not being executed.
7370 @node Overlay Sample Program
7371 @section Overlay Sample Program
7372 @cindex overlay example program
7374 When linking a program which uses overlays, you must place the overlays
7375 at their load addresses, while relocating them to run at their mapped
7376 addresses. To do this, you must write a linker script (@pxref{Overlay
7377 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
7378 since linker scripts are specific to a particular host system, target
7379 architecture, and target memory layout, this manual cannot provide
7380 portable sample code demonstrating @value{GDBN}'s overlay support.
7382 However, the @value{GDBN} source distribution does contain an overlaid
7383 program, with linker scripts for a few systems, as part of its test
7384 suite. The program consists of the following files from
7385 @file{gdb/testsuite/gdb.base}:
7389 The main program file.
7391 A simple overlay manager, used by @file{overlays.c}.
7396 Overlay modules, loaded and used by @file{overlays.c}.
7399 Linker scripts for linking the test program on the @code{d10v-elf}
7400 and @code{m32r-elf} targets.
7403 You can build the test program using the @code{d10v-elf} GCC
7404 cross-compiler like this:
7407 $ d10v-elf-gcc -g -c overlays.c
7408 $ d10v-elf-gcc -g -c ovlymgr.c
7409 $ d10v-elf-gcc -g -c foo.c
7410 $ d10v-elf-gcc -g -c bar.c
7411 $ d10v-elf-gcc -g -c baz.c
7412 $ d10v-elf-gcc -g -c grbx.c
7413 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
7414 baz.o grbx.o -Wl,-Td10v.ld -o overlays
7417 The build process is identical for any other architecture, except that
7418 you must substitute the appropriate compiler and linker script for the
7419 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
7423 @chapter Using @value{GDBN} with Different Languages
7426 Although programming languages generally have common aspects, they are
7427 rarely expressed in the same manner. For instance, in ANSI C,
7428 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
7429 Modula-2, it is accomplished by @code{p^}. Values can also be
7430 represented (and displayed) differently. Hex numbers in C appear as
7431 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
7433 @cindex working language
7434 Language-specific information is built into @value{GDBN} for some languages,
7435 allowing you to express operations like the above in your program's
7436 native language, and allowing @value{GDBN} to output values in a manner
7437 consistent with the syntax of your program's native language. The
7438 language you use to build expressions is called the @dfn{working
7442 * Setting:: Switching between source languages
7443 * Show:: Displaying the language
7444 * Checks:: Type and range checks
7445 * Support:: Supported languages
7449 @section Switching between source languages
7451 There are two ways to control the working language---either have @value{GDBN}
7452 set it automatically, or select it manually yourself. You can use the
7453 @code{set language} command for either purpose. On startup, @value{GDBN}
7454 defaults to setting the language automatically. The working language is
7455 used to determine how expressions you type are interpreted, how values
7458 In addition to the working language, every source file that
7459 @value{GDBN} knows about has its own working language. For some object
7460 file formats, the compiler might indicate which language a particular
7461 source file is in. However, most of the time @value{GDBN} infers the
7462 language from the name of the file. The language of a source file
7463 controls whether C@t{++} names are demangled---this way @code{backtrace} can
7464 show each frame appropriately for its own language. There is no way to
7465 set the language of a source file from within @value{GDBN}, but you can
7466 set the language associated with a filename extension. @xref{Show, ,
7467 Displaying the language}.
7469 This is most commonly a problem when you use a program, such
7470 as @code{cfront} or @code{f2c}, that generates C but is written in
7471 another language. In that case, make the
7472 program use @code{#line} directives in its C output; that way
7473 @value{GDBN} will know the correct language of the source code of the original
7474 program, and will display that source code, not the generated C code.
7477 * Filenames:: Filename extensions and languages.
7478 * Manually:: Setting the working language manually
7479 * Automatically:: Having @value{GDBN} infer the source language
7483 @subsection List of filename extensions and languages
7485 If a source file name ends in one of the following extensions, then
7486 @value{GDBN} infers that its language is the one indicated.
7506 Modula-2 source file
7510 Assembler source file. This actually behaves almost like C, but
7511 @value{GDBN} does not skip over function prologues when stepping.
7514 In addition, you may set the language associated with a filename
7515 extension. @xref{Show, , Displaying the language}.
7518 @subsection Setting the working language
7520 If you allow @value{GDBN} to set the language automatically,
7521 expressions are interpreted the same way in your debugging session and
7524 @kindex set language
7525 If you wish, you may set the language manually. To do this, issue the
7526 command @samp{set language @var{lang}}, where @var{lang} is the name of
7528 @code{c} or @code{modula-2}.
7529 For a list of the supported languages, type @samp{set language}.
7531 Setting the language manually prevents @value{GDBN} from updating the working
7532 language automatically. This can lead to confusion if you try
7533 to debug a program when the working language is not the same as the
7534 source language, when an expression is acceptable to both
7535 languages---but means different things. For instance, if the current
7536 source file were written in C, and @value{GDBN} was parsing Modula-2, a
7544 might not have the effect you intended. In C, this means to add
7545 @code{b} and @code{c} and place the result in @code{a}. The result
7546 printed would be the value of @code{a}. In Modula-2, this means to compare
7547 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
7550 @subsection Having @value{GDBN} infer the source language
7552 To have @value{GDBN} set the working language automatically, use
7553 @samp{set language local} or @samp{set language auto}. @value{GDBN}
7554 then infers the working language. That is, when your program stops in a
7555 frame (usually by encountering a breakpoint), @value{GDBN} sets the
7556 working language to the language recorded for the function in that
7557 frame. If the language for a frame is unknown (that is, if the function
7558 or block corresponding to the frame was defined in a source file that
7559 does not have a recognized extension), the current working language is
7560 not changed, and @value{GDBN} issues a warning.
7562 This may not seem necessary for most programs, which are written
7563 entirely in one source language. However, program modules and libraries
7564 written in one source language can be used by a main program written in
7565 a different source language. Using @samp{set language auto} in this
7566 case frees you from having to set the working language manually.
7569 @section Displaying the language
7571 The following commands help you find out which language is the
7572 working language, and also what language source files were written in.
7574 @kindex show language
7575 @kindex info frame@r{, show the source language}
7576 @kindex info source@r{, show the source language}
7579 Display the current working language. This is the
7580 language you can use with commands such as @code{print} to
7581 build and compute expressions that may involve variables in your program.
7584 Display the source language for this frame. This language becomes the
7585 working language if you use an identifier from this frame.
7586 @xref{Frame Info, ,Information about a frame}, to identify the other
7587 information listed here.
7590 Display the source language of this source file.
7591 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
7592 information listed here.
7595 In unusual circumstances, you may have source files with extensions
7596 not in the standard list. You can then set the extension associated
7597 with a language explicitly:
7599 @kindex set extension-language
7600 @kindex info extensions
7602 @item set extension-language @var{.ext} @var{language}
7603 Set source files with extension @var{.ext} to be assumed to be in
7604 the source language @var{language}.
7606 @item info extensions
7607 List all the filename extensions and the associated languages.
7611 @section Type and range checking
7614 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
7615 checking are included, but they do not yet have any effect. This
7616 section documents the intended facilities.
7618 @c FIXME remove warning when type/range code added
7620 Some languages are designed to guard you against making seemingly common
7621 errors through a series of compile- and run-time checks. These include
7622 checking the type of arguments to functions and operators, and making
7623 sure mathematical overflows are caught at run time. Checks such as
7624 these help to ensure a program's correctness once it has been compiled
7625 by eliminating type mismatches, and providing active checks for range
7626 errors when your program is running.
7628 @value{GDBN} can check for conditions like the above if you wish.
7629 Although @value{GDBN} does not check the statements in your program, it
7630 can check expressions entered directly into @value{GDBN} for evaluation via
7631 the @code{print} command, for example. As with the working language,
7632 @value{GDBN} can also decide whether or not to check automatically based on
7633 your program's source language. @xref{Support, ,Supported languages},
7634 for the default settings of supported languages.
7637 * Type Checking:: An overview of type checking
7638 * Range Checking:: An overview of range checking
7641 @cindex type checking
7642 @cindex checks, type
7644 @subsection An overview of type checking
7646 Some languages, such as Modula-2, are strongly typed, meaning that the
7647 arguments to operators and functions have to be of the correct type,
7648 otherwise an error occurs. These checks prevent type mismatch
7649 errors from ever causing any run-time problems. For example,
7657 The second example fails because the @code{CARDINAL} 1 is not
7658 type-compatible with the @code{REAL} 2.3.
7660 For the expressions you use in @value{GDBN} commands, you can tell the
7661 @value{GDBN} type checker to skip checking;
7662 to treat any mismatches as errors and abandon the expression;
7663 or to only issue warnings when type mismatches occur,
7664 but evaluate the expression anyway. When you choose the last of
7665 these, @value{GDBN} evaluates expressions like the second example above, but
7666 also issues a warning.
7668 Even if you turn type checking off, there may be other reasons
7669 related to type that prevent @value{GDBN} from evaluating an expression.
7670 For instance, @value{GDBN} does not know how to add an @code{int} and
7671 a @code{struct foo}. These particular type errors have nothing to do
7672 with the language in use, and usually arise from expressions, such as
7673 the one described above, which make little sense to evaluate anyway.
7675 Each language defines to what degree it is strict about type. For
7676 instance, both Modula-2 and C require the arguments to arithmetical
7677 operators to be numbers. In C, enumerated types and pointers can be
7678 represented as numbers, so that they are valid arguments to mathematical
7679 operators. @xref{Support, ,Supported languages}, for further
7680 details on specific languages.
7682 @value{GDBN} provides some additional commands for controlling the type checker:
7684 @kindex set check@r{, type}
7685 @kindex set check type
7686 @kindex show check type
7688 @item set check type auto
7689 Set type checking on or off based on the current working language.
7690 @xref{Support, ,Supported languages}, for the default settings for
7693 @item set check type on
7694 @itemx set check type off
7695 Set type checking on or off, overriding the default setting for the
7696 current working language. Issue a warning if the setting does not
7697 match the language default. If any type mismatches occur in
7698 evaluating an expression while type checking is on, @value{GDBN} prints a
7699 message and aborts evaluation of the expression.
7701 @item set check type warn
7702 Cause the type checker to issue warnings, but to always attempt to
7703 evaluate the expression. Evaluating the expression may still
7704 be impossible for other reasons. For example, @value{GDBN} cannot add
7705 numbers and structures.
7708 Show the current setting of the type checker, and whether or not @value{GDBN}
7709 is setting it automatically.
7712 @cindex range checking
7713 @cindex checks, range
7714 @node Range Checking
7715 @subsection An overview of range checking
7717 In some languages (such as Modula-2), it is an error to exceed the
7718 bounds of a type; this is enforced with run-time checks. Such range
7719 checking is meant to ensure program correctness by making sure
7720 computations do not overflow, or indices on an array element access do
7721 not exceed the bounds of the array.
7723 For expressions you use in @value{GDBN} commands, you can tell
7724 @value{GDBN} to treat range errors in one of three ways: ignore them,
7725 always treat them as errors and abandon the expression, or issue
7726 warnings but evaluate the expression anyway.
7728 A range error can result from numerical overflow, from exceeding an
7729 array index bound, or when you type a constant that is not a member
7730 of any type. Some languages, however, do not treat overflows as an
7731 error. In many implementations of C, mathematical overflow causes the
7732 result to ``wrap around'' to lower values---for example, if @var{m} is
7733 the largest integer value, and @var{s} is the smallest, then
7736 @var{m} + 1 @result{} @var{s}
7739 This, too, is specific to individual languages, and in some cases
7740 specific to individual compilers or machines. @xref{Support, ,
7741 Supported languages}, for further details on specific languages.
7743 @value{GDBN} provides some additional commands for controlling the range checker:
7745 @kindex set check@r{, range}
7746 @kindex set check range
7747 @kindex show check range
7749 @item set check range auto
7750 Set range checking on or off based on the current working language.
7751 @xref{Support, ,Supported languages}, for the default settings for
7754 @item set check range on
7755 @itemx set check range off
7756 Set range checking on or off, overriding the default setting for the
7757 current working language. A warning is issued if the setting does not
7758 match the language default. If a range error occurs and range checking is on,
7759 then a message is printed and evaluation of the expression is aborted.
7761 @item set check range warn
7762 Output messages when the @value{GDBN} range checker detects a range error,
7763 but attempt to evaluate the expression anyway. Evaluating the
7764 expression may still be impossible for other reasons, such as accessing
7765 memory that the process does not own (a typical example from many Unix
7769 Show the current setting of the range checker, and whether or not it is
7770 being set automatically by @value{GDBN}.
7774 @section Supported languages
7776 @value{GDBN} supports C, C@t{++}, Fortran, Java, assembly, and Modula-2.
7777 @c This is false ...
7778 Some @value{GDBN} features may be used in expressions regardless of the
7779 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
7780 and the @samp{@{type@}addr} construct (@pxref{Expressions,
7781 ,Expressions}) can be used with the constructs of any supported
7784 The following sections detail to what degree each source language is
7785 supported by @value{GDBN}. These sections are not meant to be language
7786 tutorials or references, but serve only as a reference guide to what the
7787 @value{GDBN} expression parser accepts, and what input and output
7788 formats should look like for different languages. There are many good
7789 books written on each of these languages; please look to these for a
7790 language reference or tutorial.
7794 * Modula-2:: Modula-2
7798 @subsection C and C@t{++}
7800 @cindex C and C@t{++}
7801 @cindex expressions in C or C@t{++}
7803 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
7804 to both languages. Whenever this is the case, we discuss those languages
7808 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
7809 @cindex @sc{gnu} C@t{++}
7810 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
7811 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
7812 effectively, you must compile your C@t{++} programs with a supported
7813 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
7814 compiler (@code{aCC}).
7816 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
7817 format; if it doesn't work on your system, try the stabs+ debugging
7818 format. You can select those formats explicitly with the @code{g++}
7819 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
7820 @xref{Debugging Options,,Options for Debugging Your Program or @sc{gnu}
7821 CC, gcc.info, Using @sc{gnu} CC}.
7824 * C Operators:: C and C@t{++} operators
7825 * C Constants:: C and C@t{++} constants
7826 * C plus plus expressions:: C@t{++} expressions
7827 * C Defaults:: Default settings for C and C@t{++}
7828 * C Checks:: C and C@t{++} type and range checks
7829 * Debugging C:: @value{GDBN} and C
7830 * Debugging C plus plus:: @value{GDBN} features for C@t{++}
7834 @subsubsection C and C@t{++} operators
7836 @cindex C and C@t{++} operators
7838 Operators must be defined on values of specific types. For instance,
7839 @code{+} is defined on numbers, but not on structures. Operators are
7840 often defined on groups of types.
7842 For the purposes of C and C@t{++}, the following definitions hold:
7847 @emph{Integral types} include @code{int} with any of its storage-class
7848 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
7851 @emph{Floating-point types} include @code{float}, @code{double}, and
7852 @code{long double} (if supported by the target platform).
7855 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
7858 @emph{Scalar types} include all of the above.
7863 The following operators are supported. They are listed here
7864 in order of increasing precedence:
7868 The comma or sequencing operator. Expressions in a comma-separated list
7869 are evaluated from left to right, with the result of the entire
7870 expression being the last expression evaluated.
7873 Assignment. The value of an assignment expression is the value
7874 assigned. Defined on scalar types.
7877 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
7878 and translated to @w{@code{@var{a} = @var{a op b}}}.
7879 @w{@code{@var{op}=}} and @code{=} have the same precedence.
7880 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
7881 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
7884 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
7885 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
7889 Logical @sc{or}. Defined on integral types.
7892 Logical @sc{and}. Defined on integral types.
7895 Bitwise @sc{or}. Defined on integral types.
7898 Bitwise exclusive-@sc{or}. Defined on integral types.
7901 Bitwise @sc{and}. Defined on integral types.
7904 Equality and inequality. Defined on scalar types. The value of these
7905 expressions is 0 for false and non-zero for true.
7907 @item <@r{, }>@r{, }<=@r{, }>=
7908 Less than, greater than, less than or equal, greater than or equal.
7909 Defined on scalar types. The value of these expressions is 0 for false
7910 and non-zero for true.
7913 left shift, and right shift. Defined on integral types.
7916 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
7919 Addition and subtraction. Defined on integral types, floating-point types and
7922 @item *@r{, }/@r{, }%
7923 Multiplication, division, and modulus. Multiplication and division are
7924 defined on integral and floating-point types. Modulus is defined on
7928 Increment and decrement. When appearing before a variable, the
7929 operation is performed before the variable is used in an expression;
7930 when appearing after it, the variable's value is used before the
7931 operation takes place.
7934 Pointer dereferencing. Defined on pointer types. Same precedence as
7938 Address operator. Defined on variables. Same precedence as @code{++}.
7940 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
7941 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
7942 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
7943 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
7947 Negative. Defined on integral and floating-point types. Same
7948 precedence as @code{++}.
7951 Logical negation. Defined on integral types. Same precedence as
7955 Bitwise complement operator. Defined on integral types. Same precedence as
7960 Structure member, and pointer-to-structure member. For convenience,
7961 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
7962 pointer based on the stored type information.
7963 Defined on @code{struct} and @code{union} data.
7966 Dereferences of pointers to members.
7969 Array indexing. @code{@var{a}[@var{i}]} is defined as
7970 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
7973 Function parameter list. Same precedence as @code{->}.
7976 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
7977 and @code{class} types.
7980 Doubled colons also represent the @value{GDBN} scope operator
7981 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
7985 If an operator is redefined in the user code, @value{GDBN} usually
7986 attempts to invoke the redefined version instead of using the operator's
7994 @subsubsection C and C@t{++} constants
7996 @cindex C and C@t{++} constants
7998 @value{GDBN} allows you to express the constants of C and C@t{++} in the
8003 Integer constants are a sequence of digits. Octal constants are
8004 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
8005 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
8006 @samp{l}, specifying that the constant should be treated as a
8010 Floating point constants are a sequence of digits, followed by a decimal
8011 point, followed by a sequence of digits, and optionally followed by an
8012 exponent. An exponent is of the form:
8013 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
8014 sequence of digits. The @samp{+} is optional for positive exponents.
8015 A floating-point constant may also end with a letter @samp{f} or
8016 @samp{F}, specifying that the constant should be treated as being of
8017 the @code{float} (as opposed to the default @code{double}) type; or with
8018 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
8022 Enumerated constants consist of enumerated identifiers, or their
8023 integral equivalents.
8026 Character constants are a single character surrounded by single quotes
8027 (@code{'}), or a number---the ordinal value of the corresponding character
8028 (usually its @sc{ascii} value). Within quotes, the single character may
8029 be represented by a letter or by @dfn{escape sequences}, which are of
8030 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
8031 of the character's ordinal value; or of the form @samp{\@var{x}}, where
8032 @samp{@var{x}} is a predefined special character---for example,
8033 @samp{\n} for newline.
8036 String constants are a sequence of character constants surrounded by
8037 double quotes (@code{"}). Any valid character constant (as described
8038 above) may appear. Double quotes within the string must be preceded by
8039 a backslash, so for instance @samp{"a\"b'c"} is a string of five
8043 Pointer constants are an integral value. You can also write pointers
8044 to constants using the C operator @samp{&}.
8047 Array constants are comma-separated lists surrounded by braces @samp{@{}
8048 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
8049 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
8050 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
8054 * C plus plus expressions::
8061 @node C plus plus expressions
8062 @subsubsection C@t{++} expressions
8064 @cindex expressions in C@t{++}
8065 @value{GDBN} expression handling can interpret most C@t{++} expressions.
8067 @cindex debugging C@t{++} programs
8068 @cindex C@t{++} compilers
8069 @cindex debug formats and C@t{++}
8070 @cindex @value{NGCC} and C@t{++}
8072 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
8073 proper compiler and the proper debug format. Currently, @value{GDBN}
8074 works best when debugging C@t{++} code that is compiled with
8075 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
8076 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
8077 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
8078 stabs+ as their default debug format, so you usually don't need to
8079 specify a debug format explicitly. Other compilers and/or debug formats
8080 are likely to work badly or not at all when using @value{GDBN} to debug
8086 @cindex member functions
8088 Member function calls are allowed; you can use expressions like
8091 count = aml->GetOriginal(x, y)
8094 @vindex this@r{, inside C@t{++} member functions}
8095 @cindex namespace in C@t{++}
8097 While a member function is active (in the selected stack frame), your
8098 expressions have the same namespace available as the member function;
8099 that is, @value{GDBN} allows implicit references to the class instance
8100 pointer @code{this} following the same rules as C@t{++}.
8102 @cindex call overloaded functions
8103 @cindex overloaded functions, calling
8104 @cindex type conversions in C@t{++}
8106 You can call overloaded functions; @value{GDBN} resolves the function
8107 call to the right definition, with some restrictions. @value{GDBN} does not
8108 perform overload resolution involving user-defined type conversions,
8109 calls to constructors, or instantiations of templates that do not exist
8110 in the program. It also cannot handle ellipsis argument lists or
8113 It does perform integral conversions and promotions, floating-point
8114 promotions, arithmetic conversions, pointer conversions, conversions of
8115 class objects to base classes, and standard conversions such as those of
8116 functions or arrays to pointers; it requires an exact match on the
8117 number of function arguments.
8119 Overload resolution is always performed, unless you have specified
8120 @code{set overload-resolution off}. @xref{Debugging C plus plus,
8121 ,@value{GDBN} features for C@t{++}}.
8123 You must specify @code{set overload-resolution off} in order to use an
8124 explicit function signature to call an overloaded function, as in
8126 p 'foo(char,int)'('x', 13)
8129 The @value{GDBN} command-completion facility can simplify this;
8130 see @ref{Completion, ,Command completion}.
8132 @cindex reference declarations
8134 @value{GDBN} understands variables declared as C@t{++} references; you can use
8135 them in expressions just as you do in C@t{++} source---they are automatically
8138 In the parameter list shown when @value{GDBN} displays a frame, the values of
8139 reference variables are not displayed (unlike other variables); this
8140 avoids clutter, since references are often used for large structures.
8141 The @emph{address} of a reference variable is always shown, unless
8142 you have specified @samp{set print address off}.
8145 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
8146 expressions can use it just as expressions in your program do. Since
8147 one scope may be defined in another, you can use @code{::} repeatedly if
8148 necessary, for example in an expression like
8149 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
8150 resolving name scope by reference to source files, in both C and C@t{++}
8151 debugging (@pxref{Variables, ,Program variables}).
8154 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
8155 calling virtual functions correctly, printing out virtual bases of
8156 objects, calling functions in a base subobject, casting objects, and
8157 invoking user-defined operators.
8160 @subsubsection C and C@t{++} defaults
8162 @cindex C and C@t{++} defaults
8164 If you allow @value{GDBN} to set type and range checking automatically, they
8165 both default to @code{off} whenever the working language changes to
8166 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
8167 selects the working language.
8169 If you allow @value{GDBN} to set the language automatically, it
8170 recognizes source files whose names end with @file{.c}, @file{.C}, or
8171 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
8172 these files, it sets the working language to C or C@t{++}.
8173 @xref{Automatically, ,Having @value{GDBN} infer the source language},
8174 for further details.
8176 @c Type checking is (a) primarily motivated by Modula-2, and (b)
8177 @c unimplemented. If (b) changes, it might make sense to let this node
8178 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
8181 @subsubsection C and C@t{++} type and range checks
8183 @cindex C and C@t{++} checks
8185 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
8186 is not used. However, if you turn type checking on, @value{GDBN}
8187 considers two variables type equivalent if:
8191 The two variables are structured and have the same structure, union, or
8195 The two variables have the same type name, or types that have been
8196 declared equivalent through @code{typedef}.
8199 @c leaving this out because neither J Gilmore nor R Pesch understand it.
8202 The two @code{struct}, @code{union}, or @code{enum} variables are
8203 declared in the same declaration. (Note: this may not be true for all C
8208 Range checking, if turned on, is done on mathematical operations. Array
8209 indices are not checked, since they are often used to index a pointer
8210 that is not itself an array.
8213 @subsubsection @value{GDBN} and C
8215 The @code{set print union} and @code{show print union} commands apply to
8216 the @code{union} type. When set to @samp{on}, any @code{union} that is
8217 inside a @code{struct} or @code{class} is also printed. Otherwise, it
8218 appears as @samp{@{...@}}.
8220 The @code{@@} operator aids in the debugging of dynamic arrays, formed
8221 with pointers and a memory allocation function. @xref{Expressions,
8225 * Debugging C plus plus::
8228 @node Debugging C plus plus
8229 @subsubsection @value{GDBN} features for C@t{++}
8231 @cindex commands for C@t{++}
8233 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
8234 designed specifically for use with C@t{++}. Here is a summary:
8237 @cindex break in overloaded functions
8238 @item @r{breakpoint menus}
8239 When you want a breakpoint in a function whose name is overloaded,
8240 @value{GDBN} breakpoint menus help you specify which function definition
8241 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
8243 @cindex overloading in C@t{++}
8244 @item rbreak @var{regex}
8245 Setting breakpoints using regular expressions is helpful for setting
8246 breakpoints on overloaded functions that are not members of any special
8248 @xref{Set Breaks, ,Setting breakpoints}.
8250 @cindex C@t{++} exception handling
8253 Debug C@t{++} exception handling using these commands. @xref{Set
8254 Catchpoints, , Setting catchpoints}.
8257 @item ptype @var{typename}
8258 Print inheritance relationships as well as other information for type
8260 @xref{Symbols, ,Examining the Symbol Table}.
8262 @cindex C@t{++} symbol display
8263 @item set print demangle
8264 @itemx show print demangle
8265 @itemx set print asm-demangle
8266 @itemx show print asm-demangle
8267 Control whether C@t{++} symbols display in their source form, both when
8268 displaying code as C@t{++} source and when displaying disassemblies.
8269 @xref{Print Settings, ,Print settings}.
8271 @item set print object
8272 @itemx show print object
8273 Choose whether to print derived (actual) or declared types of objects.
8274 @xref{Print Settings, ,Print settings}.
8276 @item set print vtbl
8277 @itemx show print vtbl
8278 Control the format for printing virtual function tables.
8279 @xref{Print Settings, ,Print settings}.
8280 (The @code{vtbl} commands do not work on programs compiled with the HP
8281 ANSI C@t{++} compiler (@code{aCC}).)
8283 @kindex set overload-resolution
8284 @cindex overloaded functions, overload resolution
8285 @item set overload-resolution on
8286 Enable overload resolution for C@t{++} expression evaluation. The default
8287 is on. For overloaded functions, @value{GDBN} evaluates the arguments
8288 and searches for a function whose signature matches the argument types,
8289 using the standard C@t{++} conversion rules (see @ref{C plus plus expressions, ,C@t{++}
8290 expressions}, for details). If it cannot find a match, it emits a
8293 @item set overload-resolution off
8294 Disable overload resolution for C@t{++} expression evaluation. For
8295 overloaded functions that are not class member functions, @value{GDBN}
8296 chooses the first function of the specified name that it finds in the
8297 symbol table, whether or not its arguments are of the correct type. For
8298 overloaded functions that are class member functions, @value{GDBN}
8299 searches for a function whose signature @emph{exactly} matches the
8302 @item @r{Overloaded symbol names}
8303 You can specify a particular definition of an overloaded symbol, using
8304 the same notation that is used to declare such symbols in C@t{++}: type
8305 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
8306 also use the @value{GDBN} command-line word completion facilities to list the
8307 available choices, or to finish the type list for you.
8308 @xref{Completion,, Command completion}, for details on how to do this.
8312 @subsection Modula-2
8314 @cindex Modula-2, @value{GDBN} support
8316 The extensions made to @value{GDBN} to support Modula-2 only support
8317 output from the @sc{gnu} Modula-2 compiler (which is currently being
8318 developed). Other Modula-2 compilers are not currently supported, and
8319 attempting to debug executables produced by them is most likely
8320 to give an error as @value{GDBN} reads in the executable's symbol
8323 @cindex expressions in Modula-2
8325 * M2 Operators:: Built-in operators
8326 * Built-In Func/Proc:: Built-in functions and procedures
8327 * M2 Constants:: Modula-2 constants
8328 * M2 Defaults:: Default settings for Modula-2
8329 * Deviations:: Deviations from standard Modula-2
8330 * M2 Checks:: Modula-2 type and range checks
8331 * M2 Scope:: The scope operators @code{::} and @code{.}
8332 * GDB/M2:: @value{GDBN} and Modula-2
8336 @subsubsection Operators
8337 @cindex Modula-2 operators
8339 Operators must be defined on values of specific types. For instance,
8340 @code{+} is defined on numbers, but not on structures. Operators are
8341 often defined on groups of types. For the purposes of Modula-2, the
8342 following definitions hold:
8347 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
8351 @emph{Character types} consist of @code{CHAR} and its subranges.
8354 @emph{Floating-point types} consist of @code{REAL}.
8357 @emph{Pointer types} consist of anything declared as @code{POINTER TO
8361 @emph{Scalar types} consist of all of the above.
8364 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
8367 @emph{Boolean types} consist of @code{BOOLEAN}.
8371 The following operators are supported, and appear in order of
8372 increasing precedence:
8376 Function argument or array index separator.
8379 Assignment. The value of @var{var} @code{:=} @var{value} is
8383 Less than, greater than on integral, floating-point, or enumerated
8387 Less than or equal to, greater than or equal to
8388 on integral, floating-point and enumerated types, or set inclusion on
8389 set types. Same precedence as @code{<}.
8391 @item =@r{, }<>@r{, }#
8392 Equality and two ways of expressing inequality, valid on scalar types.
8393 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
8394 available for inequality, since @code{#} conflicts with the script
8398 Set membership. Defined on set types and the types of their members.
8399 Same precedence as @code{<}.
8402 Boolean disjunction. Defined on boolean types.
8405 Boolean conjunction. Defined on boolean types.
8408 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
8411 Addition and subtraction on integral and floating-point types, or union
8412 and difference on set types.
8415 Multiplication on integral and floating-point types, or set intersection
8419 Division on floating-point types, or symmetric set difference on set
8420 types. Same precedence as @code{*}.
8423 Integer division and remainder. Defined on integral types. Same
8424 precedence as @code{*}.
8427 Negative. Defined on @code{INTEGER} and @code{REAL} data.
8430 Pointer dereferencing. Defined on pointer types.
8433 Boolean negation. Defined on boolean types. Same precedence as
8437 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
8438 precedence as @code{^}.
8441 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
8444 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
8448 @value{GDBN} and Modula-2 scope operators.
8452 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
8453 treats the use of the operator @code{IN}, or the use of operators
8454 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
8455 @code{<=}, and @code{>=} on sets as an error.
8459 @node Built-In Func/Proc
8460 @subsubsection Built-in functions and procedures
8461 @cindex Modula-2 built-ins
8463 Modula-2 also makes available several built-in procedures and functions.
8464 In describing these, the following metavariables are used:
8469 represents an @code{ARRAY} variable.
8472 represents a @code{CHAR} constant or variable.
8475 represents a variable or constant of integral type.
8478 represents an identifier that belongs to a set. Generally used in the
8479 same function with the metavariable @var{s}. The type of @var{s} should
8480 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
8483 represents a variable or constant of integral or floating-point type.
8486 represents a variable or constant of floating-point type.
8492 represents a variable.
8495 represents a variable or constant of one of many types. See the
8496 explanation of the function for details.
8499 All Modula-2 built-in procedures also return a result, described below.
8503 Returns the absolute value of @var{n}.
8506 If @var{c} is a lower case letter, it returns its upper case
8507 equivalent, otherwise it returns its argument.
8510 Returns the character whose ordinal value is @var{i}.
8513 Decrements the value in the variable @var{v} by one. Returns the new value.
8515 @item DEC(@var{v},@var{i})
8516 Decrements the value in the variable @var{v} by @var{i}. Returns the
8519 @item EXCL(@var{m},@var{s})
8520 Removes the element @var{m} from the set @var{s}. Returns the new
8523 @item FLOAT(@var{i})
8524 Returns the floating point equivalent of the integer @var{i}.
8527 Returns the index of the last member of @var{a}.
8530 Increments the value in the variable @var{v} by one. Returns the new value.
8532 @item INC(@var{v},@var{i})
8533 Increments the value in the variable @var{v} by @var{i}. Returns the
8536 @item INCL(@var{m},@var{s})
8537 Adds the element @var{m} to the set @var{s} if it is not already
8538 there. Returns the new set.
8541 Returns the maximum value of the type @var{t}.
8544 Returns the minimum value of the type @var{t}.
8547 Returns boolean TRUE if @var{i} is an odd number.
8550 Returns the ordinal value of its argument. For example, the ordinal
8551 value of a character is its @sc{ascii} value (on machines supporting the
8552 @sc{ascii} character set). @var{x} must be of an ordered type, which include
8553 integral, character and enumerated types.
8556 Returns the size of its argument. @var{x} can be a variable or a type.
8558 @item TRUNC(@var{r})
8559 Returns the integral part of @var{r}.
8561 @item VAL(@var{t},@var{i})
8562 Returns the member of the type @var{t} whose ordinal value is @var{i}.
8566 @emph{Warning:} Sets and their operations are not yet supported, so
8567 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
8571 @cindex Modula-2 constants
8573 @subsubsection Constants
8575 @value{GDBN} allows you to express the constants of Modula-2 in the following
8581 Integer constants are simply a sequence of digits. When used in an
8582 expression, a constant is interpreted to be type-compatible with the
8583 rest of the expression. Hexadecimal integers are specified by a
8584 trailing @samp{H}, and octal integers by a trailing @samp{B}.
8587 Floating point constants appear as a sequence of digits, followed by a
8588 decimal point and another sequence of digits. An optional exponent can
8589 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
8590 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
8591 digits of the floating point constant must be valid decimal (base 10)
8595 Character constants consist of a single character enclosed by a pair of
8596 like quotes, either single (@code{'}) or double (@code{"}). They may
8597 also be expressed by their ordinal value (their @sc{ascii} value, usually)
8598 followed by a @samp{C}.
8601 String constants consist of a sequence of characters enclosed by a
8602 pair of like quotes, either single (@code{'}) or double (@code{"}).
8603 Escape sequences in the style of C are also allowed. @xref{C
8604 Constants, ,C and C@t{++} constants}, for a brief explanation of escape
8608 Enumerated constants consist of an enumerated identifier.
8611 Boolean constants consist of the identifiers @code{TRUE} and
8615 Pointer constants consist of integral values only.
8618 Set constants are not yet supported.
8622 @subsubsection Modula-2 defaults
8623 @cindex Modula-2 defaults
8625 If type and range checking are set automatically by @value{GDBN}, they
8626 both default to @code{on} whenever the working language changes to
8627 Modula-2. This happens regardless of whether you or @value{GDBN}
8628 selected the working language.
8630 If you allow @value{GDBN} to set the language automatically, then entering
8631 code compiled from a file whose name ends with @file{.mod} sets the
8632 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
8633 the language automatically}, for further details.
8636 @subsubsection Deviations from standard Modula-2
8637 @cindex Modula-2, deviations from
8639 A few changes have been made to make Modula-2 programs easier to debug.
8640 This is done primarily via loosening its type strictness:
8644 Unlike in standard Modula-2, pointer constants can be formed by
8645 integers. This allows you to modify pointer variables during
8646 debugging. (In standard Modula-2, the actual address contained in a
8647 pointer variable is hidden from you; it can only be modified
8648 through direct assignment to another pointer variable or expression that
8649 returned a pointer.)
8652 C escape sequences can be used in strings and characters to represent
8653 non-printable characters. @value{GDBN} prints out strings with these
8654 escape sequences embedded. Single non-printable characters are
8655 printed using the @samp{CHR(@var{nnn})} format.
8658 The assignment operator (@code{:=}) returns the value of its right-hand
8662 All built-in procedures both modify @emph{and} return their argument.
8666 @subsubsection Modula-2 type and range checks
8667 @cindex Modula-2 checks
8670 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
8673 @c FIXME remove warning when type/range checks added
8675 @value{GDBN} considers two Modula-2 variables type equivalent if:
8679 They are of types that have been declared equivalent via a @code{TYPE
8680 @var{t1} = @var{t2}} statement
8683 They have been declared on the same line. (Note: This is true of the
8684 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
8687 As long as type checking is enabled, any attempt to combine variables
8688 whose types are not equivalent is an error.
8690 Range checking is done on all mathematical operations, assignment, array
8691 index bounds, and all built-in functions and procedures.
8694 @subsubsection The scope operators @code{::} and @code{.}
8696 @cindex @code{.}, Modula-2 scope operator
8697 @cindex colon, doubled as scope operator
8699 @vindex colon-colon@r{, in Modula-2}
8700 @c Info cannot handle :: but TeX can.
8703 @vindex ::@r{, in Modula-2}
8706 There are a few subtle differences between the Modula-2 scope operator
8707 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
8712 @var{module} . @var{id}
8713 @var{scope} :: @var{id}
8717 where @var{scope} is the name of a module or a procedure,
8718 @var{module} the name of a module, and @var{id} is any declared
8719 identifier within your program, except another module.
8721 Using the @code{::} operator makes @value{GDBN} search the scope
8722 specified by @var{scope} for the identifier @var{id}. If it is not
8723 found in the specified scope, then @value{GDBN} searches all scopes
8724 enclosing the one specified by @var{scope}.
8726 Using the @code{.} operator makes @value{GDBN} search the current scope for
8727 the identifier specified by @var{id} that was imported from the
8728 definition module specified by @var{module}. With this operator, it is
8729 an error if the identifier @var{id} was not imported from definition
8730 module @var{module}, or if @var{id} is not an identifier in
8734 @subsubsection @value{GDBN} and Modula-2
8736 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
8737 Five subcommands of @code{set print} and @code{show print} apply
8738 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
8739 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
8740 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
8741 analogue in Modula-2.
8743 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
8744 with any language, is not useful with Modula-2. Its
8745 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
8746 created in Modula-2 as they can in C or C@t{++}. However, because an
8747 address can be specified by an integral constant, the construct
8748 @samp{@{@var{type}@}@var{adrexp}} is still useful.
8750 @cindex @code{#} in Modula-2
8751 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
8752 interpreted as the beginning of a comment. Use @code{<>} instead.
8755 @chapter Examining the Symbol Table
8757 The commands described in this chapter allow you to inquire about the
8758 symbols (names of variables, functions and types) defined in your
8759 program. This information is inherent in the text of your program and
8760 does not change as your program executes. @value{GDBN} finds it in your
8761 program's symbol table, in the file indicated when you started @value{GDBN}
8762 (@pxref{File Options, ,Choosing files}), or by one of the
8763 file-management commands (@pxref{Files, ,Commands to specify files}).
8765 @cindex symbol names
8766 @cindex names of symbols
8767 @cindex quoting names
8768 Occasionally, you may need to refer to symbols that contain unusual
8769 characters, which @value{GDBN} ordinarily treats as word delimiters. The
8770 most frequent case is in referring to static variables in other
8771 source files (@pxref{Variables,,Program variables}). File names
8772 are recorded in object files as debugging symbols, but @value{GDBN} would
8773 ordinarily parse a typical file name, like @file{foo.c}, as the three words
8774 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
8775 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
8782 looks up the value of @code{x} in the scope of the file @file{foo.c}.
8785 @kindex info address
8786 @cindex address of a symbol
8787 @item info address @var{symbol}
8788 Describe where the data for @var{symbol} is stored. For a register
8789 variable, this says which register it is kept in. For a non-register
8790 local variable, this prints the stack-frame offset at which the variable
8793 Note the contrast with @samp{print &@var{symbol}}, which does not work
8794 at all for a register variable, and for a stack local variable prints
8795 the exact address of the current instantiation of the variable.
8798 @cindex symbol from address
8799 @item info symbol @var{addr}
8800 Print the name of a symbol which is stored at the address @var{addr}.
8801 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
8802 nearest symbol and an offset from it:
8805 (@value{GDBP}) info symbol 0x54320
8806 _initialize_vx + 396 in section .text
8810 This is the opposite of the @code{info address} command. You can use
8811 it to find out the name of a variable or a function given its address.
8814 @item whatis @var{expr}
8815 Print the data type of expression @var{expr}. @var{expr} is not
8816 actually evaluated, and any side-effecting operations (such as
8817 assignments or function calls) inside it do not take place.
8818 @xref{Expressions, ,Expressions}.
8821 Print the data type of @code{$}, the last value in the value history.
8824 @item ptype @var{typename}
8825 Print a description of data type @var{typename}. @var{typename} may be
8826 the name of a type, or for C code it may have the form @samp{class
8827 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
8828 @var{union-tag}} or @samp{enum @var{enum-tag}}.
8830 @item ptype @var{expr}
8832 Print a description of the type of expression @var{expr}. @code{ptype}
8833 differs from @code{whatis} by printing a detailed description, instead
8834 of just the name of the type.
8836 For example, for this variable declaration:
8839 struct complex @{double real; double imag;@} v;
8843 the two commands give this output:
8847 (@value{GDBP}) whatis v
8848 type = struct complex
8849 (@value{GDBP}) ptype v
8850 type = struct complex @{
8858 As with @code{whatis}, using @code{ptype} without an argument refers to
8859 the type of @code{$}, the last value in the value history.
8862 @item info types @var{regexp}
8864 Print a brief description of all types whose names match @var{regexp}
8865 (or all types in your program, if you supply no argument). Each
8866 complete typename is matched as though it were a complete line; thus,
8867 @samp{i type value} gives information on all types in your program whose
8868 names include the string @code{value}, but @samp{i type ^value$} gives
8869 information only on types whose complete name is @code{value}.
8871 This command differs from @code{ptype} in two ways: first, like
8872 @code{whatis}, it does not print a detailed description; second, it
8873 lists all source files where a type is defined.
8876 @cindex local variables
8877 @item info scope @var{addr}
8878 List all the variables local to a particular scope. This command
8879 accepts a location---a function name, a source line, or an address
8880 preceded by a @samp{*}, and prints all the variables local to the
8881 scope defined by that location. For example:
8884 (@value{GDBP}) @b{info scope command_line_handler}
8885 Scope for command_line_handler:
8886 Symbol rl is an argument at stack/frame offset 8, length 4.
8887 Symbol linebuffer is in static storage at address 0x150a18, length 4.
8888 Symbol linelength is in static storage at address 0x150a1c, length 4.
8889 Symbol p is a local variable in register $esi, length 4.
8890 Symbol p1 is a local variable in register $ebx, length 4.
8891 Symbol nline is a local variable in register $edx, length 4.
8892 Symbol repeat is a local variable at frame offset -8, length 4.
8896 This command is especially useful for determining what data to collect
8897 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
8902 Show information about the current source file---that is, the source file for
8903 the function containing the current point of execution:
8906 the name of the source file, and the directory containing it,
8908 the directory it was compiled in,
8910 its length, in lines,
8912 which programming language it is written in,
8914 whether the executable includes debugging information for that file, and
8915 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
8917 whether the debugging information includes information about
8918 preprocessor macros.
8922 @kindex info sources
8924 Print the names of all source files in your program for which there is
8925 debugging information, organized into two lists: files whose symbols
8926 have already been read, and files whose symbols will be read when needed.
8928 @kindex info functions
8929 @item info functions
8930 Print the names and data types of all defined functions.
8932 @item info functions @var{regexp}
8933 Print the names and data types of all defined functions
8934 whose names contain a match for regular expression @var{regexp}.
8935 Thus, @samp{info fun step} finds all functions whose names
8936 include @code{step}; @samp{info fun ^step} finds those whose names
8937 start with @code{step}. If a function name contains characters
8938 that conflict with the regular expression language (eg.
8939 @samp{operator*()}), they may be quoted with a backslash.
8941 @kindex info variables
8942 @item info variables
8943 Print the names and data types of all variables that are declared
8944 outside of functions (i.e.@: excluding local variables).
8946 @item info variables @var{regexp}
8947 Print the names and data types of all variables (except for local
8948 variables) whose names contain a match for regular expression
8952 This was never implemented.
8953 @kindex info methods
8955 @itemx info methods @var{regexp}
8956 The @code{info methods} command permits the user to examine all defined
8957 methods within C@t{++} program, or (with the @var{regexp} argument) a
8958 specific set of methods found in the various C@t{++} classes. Many
8959 C@t{++} classes provide a large number of methods. Thus, the output
8960 from the @code{ptype} command can be overwhelming and hard to use. The
8961 @code{info-methods} command filters the methods, printing only those
8962 which match the regular-expression @var{regexp}.
8965 @cindex reloading symbols
8966 Some systems allow individual object files that make up your program to
8967 be replaced without stopping and restarting your program. For example,
8968 in VxWorks you can simply recompile a defective object file and keep on
8969 running. If you are running on one of these systems, you can allow
8970 @value{GDBN} to reload the symbols for automatically relinked modules:
8973 @kindex set symbol-reloading
8974 @item set symbol-reloading on
8975 Replace symbol definitions for the corresponding source file when an
8976 object file with a particular name is seen again.
8978 @item set symbol-reloading off
8979 Do not replace symbol definitions when encountering object files of the
8980 same name more than once. This is the default state; if you are not
8981 running on a system that permits automatic relinking of modules, you
8982 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
8983 may discard symbols when linking large programs, that may contain
8984 several modules (from different directories or libraries) with the same
8987 @kindex show symbol-reloading
8988 @item show symbol-reloading
8989 Show the current @code{on} or @code{off} setting.
8992 @kindex set opaque-type-resolution
8993 @item set opaque-type-resolution on
8994 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
8995 declared as a pointer to a @code{struct}, @code{class}, or
8996 @code{union}---for example, @code{struct MyType *}---that is used in one
8997 source file although the full declaration of @code{struct MyType} is in
8998 another source file. The default is on.
9000 A change in the setting of this subcommand will not take effect until
9001 the next time symbols for a file are loaded.
9003 @item set opaque-type-resolution off
9004 Tell @value{GDBN} not to resolve opaque types. In this case, the type
9005 is printed as follows:
9007 @{<no data fields>@}
9010 @kindex show opaque-type-resolution
9011 @item show opaque-type-resolution
9012 Show whether opaque types are resolved or not.
9014 @kindex maint print symbols
9016 @kindex maint print psymbols
9017 @cindex partial symbol dump
9018 @item maint print symbols @var{filename}
9019 @itemx maint print psymbols @var{filename}
9020 @itemx maint print msymbols @var{filename}
9021 Write a dump of debugging symbol data into the file @var{filename}.
9022 These commands are used to debug the @value{GDBN} symbol-reading code. Only
9023 symbols with debugging data are included. If you use @samp{maint print
9024 symbols}, @value{GDBN} includes all the symbols for which it has already
9025 collected full details: that is, @var{filename} reflects symbols for
9026 only those files whose symbols @value{GDBN} has read. You can use the
9027 command @code{info sources} to find out which files these are. If you
9028 use @samp{maint print psymbols} instead, the dump shows information about
9029 symbols that @value{GDBN} only knows partially---that is, symbols defined in
9030 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
9031 @samp{maint print msymbols} dumps just the minimal symbol information
9032 required for each object file from which @value{GDBN} has read some symbols.
9033 @xref{Files, ,Commands to specify files}, for a discussion of how
9034 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
9038 @chapter Altering Execution
9040 Once you think you have found an error in your program, you might want to
9041 find out for certain whether correcting the apparent error would lead to
9042 correct results in the rest of the run. You can find the answer by
9043 experiment, using the @value{GDBN} features for altering execution of the
9046 For example, you can store new values into variables or memory
9047 locations, give your program a signal, restart it at a different
9048 address, or even return prematurely from a function.
9051 * Assignment:: Assignment to variables
9052 * Jumping:: Continuing at a different address
9053 * Signaling:: Giving your program a signal
9054 * Returning:: Returning from a function
9055 * Calling:: Calling your program's functions
9056 * Patching:: Patching your program
9060 @section Assignment to variables
9063 @cindex setting variables
9064 To alter the value of a variable, evaluate an assignment expression.
9065 @xref{Expressions, ,Expressions}. For example,
9072 stores the value 4 into the variable @code{x}, and then prints the
9073 value of the assignment expression (which is 4).
9074 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
9075 information on operators in supported languages.
9077 @kindex set variable
9078 @cindex variables, setting
9079 If you are not interested in seeing the value of the assignment, use the
9080 @code{set} command instead of the @code{print} command. @code{set} is
9081 really the same as @code{print} except that the expression's value is
9082 not printed and is not put in the value history (@pxref{Value History,
9083 ,Value history}). The expression is evaluated only for its effects.
9085 If the beginning of the argument string of the @code{set} command
9086 appears identical to a @code{set} subcommand, use the @code{set
9087 variable} command instead of just @code{set}. This command is identical
9088 to @code{set} except for its lack of subcommands. For example, if your
9089 program has a variable @code{width}, you get an error if you try to set
9090 a new value with just @samp{set width=13}, because @value{GDBN} has the
9091 command @code{set width}:
9094 (@value{GDBP}) whatis width
9096 (@value{GDBP}) p width
9098 (@value{GDBP}) set width=47
9099 Invalid syntax in expression.
9103 The invalid expression, of course, is @samp{=47}. In
9104 order to actually set the program's variable @code{width}, use
9107 (@value{GDBP}) set var width=47
9110 Because the @code{set} command has many subcommands that can conflict
9111 with the names of program variables, it is a good idea to use the
9112 @code{set variable} command instead of just @code{set}. For example, if
9113 your program has a variable @code{g}, you run into problems if you try
9114 to set a new value with just @samp{set g=4}, because @value{GDBN} has
9115 the command @code{set gnutarget}, abbreviated @code{set g}:
9119 (@value{GDBP}) whatis g
9123 (@value{GDBP}) set g=4
9127 The program being debugged has been started already.
9128 Start it from the beginning? (y or n) y
9129 Starting program: /home/smith/cc_progs/a.out
9130 "/home/smith/cc_progs/a.out": can't open to read symbols:
9132 (@value{GDBP}) show g
9133 The current BFD target is "=4".
9138 The program variable @code{g} did not change, and you silently set the
9139 @code{gnutarget} to an invalid value. In order to set the variable
9143 (@value{GDBP}) set var g=4
9146 @value{GDBN} allows more implicit conversions in assignments than C; you can
9147 freely store an integer value into a pointer variable or vice versa,
9148 and you can convert any structure to any other structure that is the
9149 same length or shorter.
9150 @comment FIXME: how do structs align/pad in these conversions?
9151 @comment /doc@cygnus.com 18dec1990
9153 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
9154 construct to generate a value of specified type at a specified address
9155 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
9156 to memory location @code{0x83040} as an integer (which implies a certain size
9157 and representation in memory), and
9160 set @{int@}0x83040 = 4
9164 stores the value 4 into that memory location.
9167 @section Continuing at a different address
9169 Ordinarily, when you continue your program, you do so at the place where
9170 it stopped, with the @code{continue} command. You can instead continue at
9171 an address of your own choosing, with the following commands:
9175 @item jump @var{linespec}
9176 Resume execution at line @var{linespec}. Execution stops again
9177 immediately if there is a breakpoint there. @xref{List, ,Printing
9178 source lines}, for a description of the different forms of
9179 @var{linespec}. It is common practice to use the @code{tbreak} command
9180 in conjunction with @code{jump}. @xref{Set Breaks, ,Setting
9183 The @code{jump} command does not change the current stack frame, or
9184 the stack pointer, or the contents of any memory location or any
9185 register other than the program counter. If line @var{linespec} is in
9186 a different function from the one currently executing, the results may
9187 be bizarre if the two functions expect different patterns of arguments or
9188 of local variables. For this reason, the @code{jump} command requests
9189 confirmation if the specified line is not in the function currently
9190 executing. However, even bizarre results are predictable if you are
9191 well acquainted with the machine-language code of your program.
9193 @item jump *@var{address}
9194 Resume execution at the instruction at address @var{address}.
9197 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
9198 On many systems, you can get much the same effect as the @code{jump}
9199 command by storing a new value into the register @code{$pc}. The
9200 difference is that this does not start your program running; it only
9201 changes the address of where it @emph{will} run when you continue. For
9209 makes the next @code{continue} command or stepping command execute at
9210 address @code{0x485}, rather than at the address where your program stopped.
9211 @xref{Continuing and Stepping, ,Continuing and stepping}.
9213 The most common occasion to use the @code{jump} command is to back
9214 up---perhaps with more breakpoints set---over a portion of a program
9215 that has already executed, in order to examine its execution in more
9220 @section Giving your program a signal
9224 @item signal @var{signal}
9225 Resume execution where your program stopped, but immediately give it the
9226 signal @var{signal}. @var{signal} can be the name or the number of a
9227 signal. For example, on many systems @code{signal 2} and @code{signal
9228 SIGINT} are both ways of sending an interrupt signal.
9230 Alternatively, if @var{signal} is zero, continue execution without
9231 giving a signal. This is useful when your program stopped on account of
9232 a signal and would ordinary see the signal when resumed with the
9233 @code{continue} command; @samp{signal 0} causes it to resume without a
9236 @code{signal} does not repeat when you press @key{RET} a second time
9237 after executing the command.
9241 Invoking the @code{signal} command is not the same as invoking the
9242 @code{kill} utility from the shell. Sending a signal with @code{kill}
9243 causes @value{GDBN} to decide what to do with the signal depending on
9244 the signal handling tables (@pxref{Signals}). The @code{signal} command
9245 passes the signal directly to your program.
9249 @section Returning from a function
9252 @cindex returning from a function
9255 @itemx return @var{expression}
9256 You can cancel execution of a function call with the @code{return}
9257 command. If you give an
9258 @var{expression} argument, its value is used as the function's return
9262 When you use @code{return}, @value{GDBN} discards the selected stack frame
9263 (and all frames within it). You can think of this as making the
9264 discarded frame return prematurely. If you wish to specify a value to
9265 be returned, give that value as the argument to @code{return}.
9267 This pops the selected stack frame (@pxref{Selection, ,Selecting a
9268 frame}), and any other frames inside of it, leaving its caller as the
9269 innermost remaining frame. That frame becomes selected. The
9270 specified value is stored in the registers used for returning values
9273 The @code{return} command does not resume execution; it leaves the
9274 program stopped in the state that would exist if the function had just
9275 returned. In contrast, the @code{finish} command (@pxref{Continuing
9276 and Stepping, ,Continuing and stepping}) resumes execution until the
9277 selected stack frame returns naturally.
9280 @section Calling program functions
9282 @cindex calling functions
9285 @item call @var{expr}
9286 Evaluate the expression @var{expr} without displaying @code{void}
9290 You can use this variant of the @code{print} command if you want to
9291 execute a function from your program, but without cluttering the output
9292 with @code{void} returned values. If the result is not void, it
9293 is printed and saved in the value history.
9296 @section Patching programs
9298 @cindex patching binaries
9299 @cindex writing into executables
9300 @cindex writing into corefiles
9302 By default, @value{GDBN} opens the file containing your program's
9303 executable code (or the corefile) read-only. This prevents accidental
9304 alterations to machine code; but it also prevents you from intentionally
9305 patching your program's binary.
9307 If you'd like to be able to patch the binary, you can specify that
9308 explicitly with the @code{set write} command. For example, you might
9309 want to turn on internal debugging flags, or even to make emergency
9315 @itemx set write off
9316 If you specify @samp{set write on}, @value{GDBN} opens executable and
9317 core files for both reading and writing; if you specify @samp{set write
9318 off} (the default), @value{GDBN} opens them read-only.
9320 If you have already loaded a file, you must load it again (using the
9321 @code{exec-file} or @code{core-file} command) after changing @code{set
9322 write}, for your new setting to take effect.
9326 Display whether executable files and core files are opened for writing
9331 @chapter @value{GDBN} Files
9333 @value{GDBN} needs to know the file name of the program to be debugged,
9334 both in order to read its symbol table and in order to start your
9335 program. To debug a core dump of a previous run, you must also tell
9336 @value{GDBN} the name of the core dump file.
9339 * Files:: Commands to specify files
9340 * Separate Debug Files:: Debugging information in separate files
9341 * Symbol Errors:: Errors reading symbol files
9345 @section Commands to specify files
9347 @cindex symbol table
9348 @cindex core dump file
9350 You may want to specify executable and core dump file names. The usual
9351 way to do this is at start-up time, using the arguments to
9352 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
9353 Out of @value{GDBN}}).
9355 Occasionally it is necessary to change to a different file during a
9356 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
9357 a file you want to use. In these situations the @value{GDBN} commands
9358 to specify new files are useful.
9361 @cindex executable file
9363 @item file @var{filename}
9364 Use @var{filename} as the program to be debugged. It is read for its
9365 symbols and for the contents of pure memory. It is also the program
9366 executed when you use the @code{run} command. If you do not specify a
9367 directory and the file is not found in the @value{GDBN} working directory,
9368 @value{GDBN} uses the environment variable @code{PATH} as a list of
9369 directories to search, just as the shell does when looking for a program
9370 to run. You can change the value of this variable, for both @value{GDBN}
9371 and your program, using the @code{path} command.
9373 On systems with memory-mapped files, an auxiliary file named
9374 @file{@var{filename}.syms} may hold symbol table information for
9375 @var{filename}. If so, @value{GDBN} maps in the symbol table from
9376 @file{@var{filename}.syms}, starting up more quickly. See the
9377 descriptions of the file options @samp{-mapped} and @samp{-readnow}
9378 (available on the command line, and with the commands @code{file},
9379 @code{symbol-file}, or @code{add-symbol-file}, described below),
9380 for more information.
9383 @code{file} with no argument makes @value{GDBN} discard any information it
9384 has on both executable file and the symbol table.
9387 @item exec-file @r{[} @var{filename} @r{]}
9388 Specify that the program to be run (but not the symbol table) is found
9389 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
9390 if necessary to locate your program. Omitting @var{filename} means to
9391 discard information on the executable file.
9394 @item symbol-file @r{[} @var{filename} @r{]}
9395 Read symbol table information from file @var{filename}. @code{PATH} is
9396 searched when necessary. Use the @code{file} command to get both symbol
9397 table and program to run from the same file.
9399 @code{symbol-file} with no argument clears out @value{GDBN} information on your
9400 program's symbol table.
9402 The @code{symbol-file} command causes @value{GDBN} to forget the contents
9403 of its convenience variables, the value history, and all breakpoints and
9404 auto-display expressions. This is because they may contain pointers to
9405 the internal data recording symbols and data types, which are part of
9406 the old symbol table data being discarded inside @value{GDBN}.
9408 @code{symbol-file} does not repeat if you press @key{RET} again after
9411 When @value{GDBN} is configured for a particular environment, it
9412 understands debugging information in whatever format is the standard
9413 generated for that environment; you may use either a @sc{gnu} compiler, or
9414 other compilers that adhere to the local conventions.
9415 Best results are usually obtained from @sc{gnu} compilers; for example,
9416 using @code{@value{GCC}} you can generate debugging information for
9419 For most kinds of object files, with the exception of old SVR3 systems
9420 using COFF, the @code{symbol-file} command does not normally read the
9421 symbol table in full right away. Instead, it scans the symbol table
9422 quickly to find which source files and which symbols are present. The
9423 details are read later, one source file at a time, as they are needed.
9425 The purpose of this two-stage reading strategy is to make @value{GDBN}
9426 start up faster. For the most part, it is invisible except for
9427 occasional pauses while the symbol table details for a particular source
9428 file are being read. (The @code{set verbose} command can turn these
9429 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
9430 warnings and messages}.)
9432 We have not implemented the two-stage strategy for COFF yet. When the
9433 symbol table is stored in COFF format, @code{symbol-file} reads the
9434 symbol table data in full right away. Note that ``stabs-in-COFF''
9435 still does the two-stage strategy, since the debug info is actually
9439 @cindex reading symbols immediately
9440 @cindex symbols, reading immediately
9442 @cindex memory-mapped symbol file
9443 @cindex saving symbol table
9444 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
9445 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
9446 You can override the @value{GDBN} two-stage strategy for reading symbol
9447 tables by using the @samp{-readnow} option with any of the commands that
9448 load symbol table information, if you want to be sure @value{GDBN} has the
9449 entire symbol table available.
9451 If memory-mapped files are available on your system through the
9452 @code{mmap} system call, you can use another option, @samp{-mapped}, to
9453 cause @value{GDBN} to write the symbols for your program into a reusable
9454 file. Future @value{GDBN} debugging sessions map in symbol information
9455 from this auxiliary symbol file (if the program has not changed), rather
9456 than spending time reading the symbol table from the executable
9457 program. Using the @samp{-mapped} option has the same effect as
9458 starting @value{GDBN} with the @samp{-mapped} command-line option.
9460 You can use both options together, to make sure the auxiliary symbol
9461 file has all the symbol information for your program.
9463 The auxiliary symbol file for a program called @var{myprog} is called
9464 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
9465 than the corresponding executable), @value{GDBN} always attempts to use
9466 it when you debug @var{myprog}; no special options or commands are
9469 The @file{.syms} file is specific to the host machine where you run
9470 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
9471 symbol table. It cannot be shared across multiple host platforms.
9473 @c FIXME: for now no mention of directories, since this seems to be in
9474 @c flux. 13mar1992 status is that in theory GDB would look either in
9475 @c current dir or in same dir as myprog; but issues like competing
9476 @c GDB's, or clutter in system dirs, mean that in practice right now
9477 @c only current dir is used. FFish says maybe a special GDB hierarchy
9478 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
9483 @item core-file @r{[} @var{filename} @r{]}
9484 Specify the whereabouts of a core dump file to be used as the ``contents
9485 of memory''. Traditionally, core files contain only some parts of the
9486 address space of the process that generated them; @value{GDBN} can access the
9487 executable file itself for other parts.
9489 @code{core-file} with no argument specifies that no core file is
9492 Note that the core file is ignored when your program is actually running
9493 under @value{GDBN}. So, if you have been running your program and you
9494 wish to debug a core file instead, you must kill the subprocess in which
9495 the program is running. To do this, use the @code{kill} command
9496 (@pxref{Kill Process, ,Killing the child process}).
9498 @kindex add-symbol-file
9499 @cindex dynamic linking
9500 @item add-symbol-file @var{filename} @var{address}
9501 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
9502 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
9503 The @code{add-symbol-file} command reads additional symbol table
9504 information from the file @var{filename}. You would use this command
9505 when @var{filename} has been dynamically loaded (by some other means)
9506 into the program that is running. @var{address} should be the memory
9507 address at which the file has been loaded; @value{GDBN} cannot figure
9508 this out for itself. You can additionally specify an arbitrary number
9509 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
9510 section name and base address for that section. You can specify any
9511 @var{address} as an expression.
9513 The symbol table of the file @var{filename} is added to the symbol table
9514 originally read with the @code{symbol-file} command. You can use the
9515 @code{add-symbol-file} command any number of times; the new symbol data
9516 thus read keeps adding to the old. To discard all old symbol data
9517 instead, use the @code{symbol-file} command without any arguments.
9519 @cindex relocatable object files, reading symbols from
9520 @cindex object files, relocatable, reading symbols from
9521 @cindex reading symbols from relocatable object files
9522 @cindex symbols, reading from relocatable object files
9523 @cindex @file{.o} files, reading symbols from
9524 Although @var{filename} is typically a shared library file, an
9525 executable file, or some other object file which has been fully
9526 relocated for loading into a process, you can also load symbolic
9527 information from relocatable @file{.o} files, as long as:
9531 the file's symbolic information refers only to linker symbols defined in
9532 that file, not to symbols defined by other object files,
9534 every section the file's symbolic information refers to has actually
9535 been loaded into the inferior, as it appears in the file, and
9537 you can determine the address at which every section was loaded, and
9538 provide these to the @code{add-symbol-file} command.
9542 Some embedded operating systems, like Sun Chorus and VxWorks, can load
9543 relocatable files into an already running program; such systems
9544 typically make the requirements above easy to meet. However, it's
9545 important to recognize that many native systems use complex link
9546 procedures (@code{.linkonce} section factoring and C++ constructor table
9547 assembly, for example) that make the requirements difficult to meet. In
9548 general, one cannot assume that using @code{add-symbol-file} to read a
9549 relocatable object file's symbolic information will have the same effect
9550 as linking the relocatable object file into the program in the normal
9553 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
9555 You can use the @samp{-mapped} and @samp{-readnow} options just as with
9556 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
9557 table information for @var{filename}.
9559 @kindex add-shared-symbol-file
9560 @item add-shared-symbol-file
9561 The @code{add-shared-symbol-file} command can be used only under Harris' CXUX
9562 operating system for the Motorola 88k. @value{GDBN} automatically looks for
9563 shared libraries, however if @value{GDBN} does not find yours, you can run
9564 @code{add-shared-symbol-file}. It takes no arguments.
9568 The @code{section} command changes the base address of section SECTION of
9569 the exec file to ADDR. This can be used if the exec file does not contain
9570 section addresses, (such as in the a.out format), or when the addresses
9571 specified in the file itself are wrong. Each section must be changed
9572 separately. The @code{info files} command, described below, lists all
9573 the sections and their addresses.
9579 @code{info files} and @code{info target} are synonymous; both print the
9580 current target (@pxref{Targets, ,Specifying a Debugging Target}),
9581 including the names of the executable and core dump files currently in
9582 use by @value{GDBN}, and the files from which symbols were loaded. The
9583 command @code{help target} lists all possible targets rather than
9586 @kindex maint info sections
9587 @item maint info sections
9588 Another command that can give you extra information about program sections
9589 is @code{maint info sections}. In addition to the section information
9590 displayed by @code{info files}, this command displays the flags and file
9591 offset of each section in the executable and core dump files. In addition,
9592 @code{maint info sections} provides the following command options (which
9593 may be arbitrarily combined):
9597 Display sections for all loaded object files, including shared libraries.
9598 @item @var{sections}
9599 Display info only for named @var{sections}.
9600 @item @var{section-flags}
9601 Display info only for sections for which @var{section-flags} are true.
9602 The section flags that @value{GDBN} currently knows about are:
9605 Section will have space allocated in the process when loaded.
9606 Set for all sections except those containing debug information.
9608 Section will be loaded from the file into the child process memory.
9609 Set for pre-initialized code and data, clear for @code{.bss} sections.
9611 Section needs to be relocated before loading.
9613 Section cannot be modified by the child process.
9615 Section contains executable code only.
9617 Section contains data only (no executable code).
9619 Section will reside in ROM.
9621 Section contains data for constructor/destructor lists.
9623 Section is not empty.
9625 An instruction to the linker to not output the section.
9626 @item COFF_SHARED_LIBRARY
9627 A notification to the linker that the section contains
9628 COFF shared library information.
9630 Section contains common symbols.
9633 @kindex set trust-readonly-sections
9634 @item set trust-readonly-sections on
9635 Tell @value{GDBN} that readonly sections in your object file
9636 really are read-only (i.e.@: that their contents will not change).
9637 In that case, @value{GDBN} can fetch values from these sections
9638 out of the object file, rather than from the target program.
9639 For some targets (notably embedded ones), this can be a significant
9640 enhancement to debugging performance.
9644 @item set trust-readonly-sections off
9645 Tell @value{GDBN} not to trust readonly sections. This means that
9646 the contents of the section might change while the program is running,
9647 and must therefore be fetched from the target when needed.
9650 All file-specifying commands allow both absolute and relative file names
9651 as arguments. @value{GDBN} always converts the file name to an absolute file
9652 name and remembers it that way.
9654 @cindex shared libraries
9655 @value{GDBN} supports HP-UX, SunOS, SVr4, Irix 5, and IBM RS/6000 shared
9658 @value{GDBN} automatically loads symbol definitions from shared libraries
9659 when you use the @code{run} command, or when you examine a core file.
9660 (Before you issue the @code{run} command, @value{GDBN} does not understand
9661 references to a function in a shared library, however---unless you are
9662 debugging a core file).
9664 On HP-UX, if the program loads a library explicitly, @value{GDBN}
9665 automatically loads the symbols at the time of the @code{shl_load} call.
9667 @c FIXME: some @value{GDBN} release may permit some refs to undef
9668 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
9669 @c FIXME...lib; check this from time to time when updating manual
9671 There are times, however, when you may wish to not automatically load
9672 symbol definitions from shared libraries, such as when they are
9673 particularly large or there are many of them.
9675 To control the automatic loading of shared library symbols, use the
9679 @kindex set auto-solib-add
9680 @item set auto-solib-add @var{mode}
9681 If @var{mode} is @code{on}, symbols from all shared object libraries
9682 will be loaded automatically when the inferior begins execution, you
9683 attach to an independently started inferior, or when the dynamic linker
9684 informs @value{GDBN} that a new library has been loaded. If @var{mode}
9685 is @code{off}, symbols must be loaded manually, using the
9686 @code{sharedlibrary} command. The default value is @code{on}.
9688 @kindex show auto-solib-add
9689 @item show auto-solib-add
9690 Display the current autoloading mode.
9693 To explicitly load shared library symbols, use the @code{sharedlibrary}
9697 @kindex info sharedlibrary
9700 @itemx info sharedlibrary
9701 Print the names of the shared libraries which are currently loaded.
9703 @kindex sharedlibrary
9705 @item sharedlibrary @var{regex}
9706 @itemx share @var{regex}
9707 Load shared object library symbols for files matching a
9708 Unix regular expression.
9709 As with files loaded automatically, it only loads shared libraries
9710 required by your program for a core file or after typing @code{run}. If
9711 @var{regex} is omitted all shared libraries required by your program are
9715 On some systems, such as HP-UX systems, @value{GDBN} supports
9716 autoloading shared library symbols until a limiting threshold size is
9717 reached. This provides the benefit of allowing autoloading to remain on
9718 by default, but avoids autoloading excessively large shared libraries,
9719 up to a threshold that is initially set, but which you can modify if you
9722 Beyond that threshold, symbols from shared libraries must be explicitly
9723 loaded. To load these symbols, use the command @code{sharedlibrary
9724 @var{filename}}. The base address of the shared library is determined
9725 automatically by @value{GDBN} and need not be specified.
9727 To display or set the threshold, use the commands:
9730 @kindex set auto-solib-limit
9731 @item set auto-solib-limit @var{threshold}
9732 Set the autoloading size threshold, in an integral number of megabytes.
9733 If @var{threshold} is nonzero and shared library autoloading is enabled,
9734 symbols from all shared object libraries will be loaded until the total
9735 size of the loaded shared library symbols exceeds this threshold.
9736 Otherwise, symbols must be loaded manually, using the
9737 @code{sharedlibrary} command. The default threshold is 100 (i.e.@: 100
9740 @kindex show auto-solib-limit
9741 @item show auto-solib-limit
9742 Display the current autoloading size threshold, in megabytes.
9745 Shared libraries are also supported in many cross or remote debugging
9746 configurations. A copy of the target's libraries need to be present on the
9747 host system; they need to be the same as the target libraries, although the
9748 copies on the target can be stripped as long as the copies on the host are
9751 You need to tell @value{GDBN} where the target libraries are, so that it can
9752 load the correct copies---otherwise, it may try to load the host's libraries.
9753 @value{GDBN} has two variables to specify the search directories for target
9757 @kindex set solib-absolute-prefix
9758 @item set solib-absolute-prefix @var{path}
9759 If this variable is set, @var{path} will be used as a prefix for any
9760 absolute shared library paths; many runtime loaders store the absolute
9761 paths to the shared library in the target program's memory. If you use
9762 @samp{solib-absolute-prefix} to find shared libraries, they need to be laid
9763 out in the same way that they are on the target, with e.g.@: a
9764 @file{/usr/lib} hierarchy under @var{path}.
9766 You can set the default value of @samp{solib-absolute-prefix} by using the
9767 configure-time @samp{--with-sysroot} option.
9769 @kindex show solib-absolute-prefix
9770 @item show solib-absolute-prefix
9771 Display the current shared library prefix.
9773 @kindex set solib-search-path
9774 @item set solib-search-path @var{path}
9775 If this variable is set, @var{path} is a colon-separated list of directories
9776 to search for shared libraries. @samp{solib-search-path} is used after
9777 @samp{solib-absolute-prefix} fails to locate the library, or if the path to
9778 the library is relative instead of absolute. If you want to use
9779 @samp{solib-search-path} instead of @samp{solib-absolute-prefix}, be sure to
9780 set @samp{solib-absolute-prefix} to a nonexistant directory to prevent
9781 @value{GDBN} from finding your host's libraries.
9783 @kindex show solib-search-path
9784 @item show solib-search-path
9785 Display the current shared library search path.
9789 @node Separate Debug Files
9790 @section Debugging Information in Separate Files
9791 @cindex separate debugging information files
9792 @cindex debugging information in separate files
9793 @cindex @file{.debug} subdirectories
9794 @cindex debugging information directory, global
9795 @cindex global debugging information directory
9797 @value{GDBN} allows you to put a program's debugging information in a
9798 file separate from the executable itself, in a way that allows
9799 @value{GDBN} to find and load the debugging information automatically.
9800 Since debugging information can be very large --- sometimes larger
9801 than the executable code itself --- some systems distribute debugging
9802 information for their executables in separate files, which users can
9803 install only when they need to debug a problem.
9805 If an executable's debugging information has been extracted to a
9806 separate file, the executable should contain a @dfn{debug link} giving
9807 the name of the debugging information file (with no directory
9808 components), and a checksum of its contents. (The exact form of a
9809 debug link is described below.) If the full name of the directory
9810 containing the executable is @var{execdir}, and the executable has a
9811 debug link that specifies the name @var{debugfile}, then @value{GDBN}
9812 will automatically search for the debugging information file in three
9817 the directory containing the executable file (that is, it will look
9818 for a file named @file{@var{execdir}/@var{debugfile}},
9820 a subdirectory of that directory named @file{.debug} (that is, the
9821 file @file{@var{execdir}/.debug/@var{debugfile}}, and
9823 a subdirectory of the global debug file directory that includes the
9824 executable's full path, and the name from the link (that is, the file
9825 @file{@var{globaldebugdir}/@var{execdir}/@var{debugfile}}, where
9826 @var{globaldebugdir} is the global debug file directory, and
9827 @var{execdir} has been turned into a relative path).
9830 @value{GDBN} checks under each of these names for a debugging
9831 information file whose checksum matches that given in the link, and
9832 reads the debugging information from the first one it finds.
9834 So, for example, if you ask @value{GDBN} to debug @file{/usr/bin/ls},
9835 which has a link containing the name @file{ls.debug}, and the global
9836 debug directory is @file{/usr/lib/debug}, then @value{GDBN} will look
9837 for debug information in @file{/usr/bin/ls.debug},
9838 @file{/usr/bin/.debug/ls.debug}, and
9839 @file{/usr/lib/debug/usr/bin/ls.debug}.
9841 You can set the global debugging info directory's name, and view the
9842 name @value{GDBN} is currently using.
9846 @kindex set debug-file-directory
9847 @item set debug-file-directory @var{directory}
9848 Set the directory which @value{GDBN} searches for separate debugging
9849 information files to @var{directory}.
9851 @kindex show debug-file-directory
9852 @item show debug-file-directory
9853 Show the directory @value{GDBN} searches for separate debugging
9858 @cindex @code{.gnu_debuglink} sections
9860 A debug link is a special section of the executable file named
9861 @code{.gnu_debuglink}. The section must contain:
9865 A filename, with any leading directory components removed, followed by
9868 zero to three bytes of padding, as needed to reach the next four-byte
9869 boundary within the section, and
9871 a four-byte CRC checksum, stored in the same endianness used for the
9872 executable file itself. The checksum is computed on the debugging
9873 information file's full contents by the function given below, passing
9874 zero as the @var{crc} argument.
9877 Any executable file format can carry a debug link, as long as it can
9878 contain a section named @code{.gnu_debuglink} with the contents
9881 The debugging information file itself should be an ordinary
9882 executable, containing a full set of linker symbols, sections, and
9883 debugging information. The sections of the debugging information file
9884 should have the same names, addresses and sizes as the original file,
9885 but they need not contain any data --- much like a @code{.bss} section
9886 in an ordinary executable.
9888 As of December 2002, there is no standard GNU utility to produce
9889 separated executable / debugging information file pairs. Ulrich
9890 Drepper's @file{elfutils} package, starting with version 0.53,
9891 contains a version of the @code{strip} command such that the command
9892 @kbd{strip foo -f foo.debug} removes the debugging information from
9893 the executable file @file{foo}, places it in the file
9894 @file{foo.debug}, and leaves behind a debug link in @file{foo}.
9896 Since there are many different ways to compute CRC's (different
9897 polynomials, reversals, byte ordering, etc.), the simplest way to
9898 describe the CRC used in @code{.gnu_debuglink} sections is to give the
9899 complete code for a function that computes it:
9901 @kindex @code{gnu_debuglink_crc32}
9904 gnu_debuglink_crc32 (unsigned long crc,
9905 unsigned char *buf, size_t len)
9907 static const unsigned long crc32_table[256] =
9909 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
9910 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
9911 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
9912 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
9913 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
9914 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
9915 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
9916 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
9917 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
9918 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
9919 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
9920 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
9921 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
9922 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
9923 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
9924 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
9925 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
9926 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
9927 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
9928 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
9929 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
9930 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
9931 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
9932 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
9933 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
9934 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
9935 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
9936 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
9937 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
9938 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
9939 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
9940 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
9941 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
9942 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
9943 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
9944 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
9945 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
9946 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
9947 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
9948 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
9949 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
9950 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
9951 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
9952 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
9953 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
9954 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
9955 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
9956 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
9957 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
9958 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
9959 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
9964 crc = ~crc & 0xffffffff;
9965 for (end = buf + len; buf < end; ++buf)
9966 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
9967 return ~crc & 0xffffffff;;
9973 @section Errors reading symbol files
9975 While reading a symbol file, @value{GDBN} occasionally encounters problems,
9976 such as symbol types it does not recognize, or known bugs in compiler
9977 output. By default, @value{GDBN} does not notify you of such problems, since
9978 they are relatively common and primarily of interest to people
9979 debugging compilers. If you are interested in seeing information
9980 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
9981 only one message about each such type of problem, no matter how many
9982 times the problem occurs; or you can ask @value{GDBN} to print more messages,
9983 to see how many times the problems occur, with the @code{set
9984 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
9987 The messages currently printed, and their meanings, include:
9990 @item inner block not inside outer block in @var{symbol}
9992 The symbol information shows where symbol scopes begin and end
9993 (such as at the start of a function or a block of statements). This
9994 error indicates that an inner scope block is not fully contained
9995 in its outer scope blocks.
9997 @value{GDBN} circumvents the problem by treating the inner block as if it had
9998 the same scope as the outer block. In the error message, @var{symbol}
9999 may be shown as ``@code{(don't know)}'' if the outer block is not a
10002 @item block at @var{address} out of order
10004 The symbol information for symbol scope blocks should occur in
10005 order of increasing addresses. This error indicates that it does not
10008 @value{GDBN} does not circumvent this problem, and has trouble
10009 locating symbols in the source file whose symbols it is reading. (You
10010 can often determine what source file is affected by specifying
10011 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
10014 @item bad block start address patched
10016 The symbol information for a symbol scope block has a start address
10017 smaller than the address of the preceding source line. This is known
10018 to occur in the SunOS 4.1.1 (and earlier) C compiler.
10020 @value{GDBN} circumvents the problem by treating the symbol scope block as
10021 starting on the previous source line.
10023 @item bad string table offset in symbol @var{n}
10026 Symbol number @var{n} contains a pointer into the string table which is
10027 larger than the size of the string table.
10029 @value{GDBN} circumvents the problem by considering the symbol to have the
10030 name @code{foo}, which may cause other problems if many symbols end up
10033 @item unknown symbol type @code{0x@var{nn}}
10035 The symbol information contains new data types that @value{GDBN} does
10036 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
10037 uncomprehended information, in hexadecimal.
10039 @value{GDBN} circumvents the error by ignoring this symbol information.
10040 This usually allows you to debug your program, though certain symbols
10041 are not accessible. If you encounter such a problem and feel like
10042 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
10043 on @code{complain}, then go up to the function @code{read_dbx_symtab}
10044 and examine @code{*bufp} to see the symbol.
10046 @item stub type has NULL name
10048 @value{GDBN} could not find the full definition for a struct or class.
10050 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
10051 The symbol information for a C@t{++} member function is missing some
10052 information that recent versions of the compiler should have output for
10055 @item info mismatch between compiler and debugger
10057 @value{GDBN} could not parse a type specification output by the compiler.
10062 @chapter Specifying a Debugging Target
10064 @cindex debugging target
10067 A @dfn{target} is the execution environment occupied by your program.
10069 Often, @value{GDBN} runs in the same host environment as your program;
10070 in that case, the debugging target is specified as a side effect when
10071 you use the @code{file} or @code{core} commands. When you need more
10072 flexibility---for example, running @value{GDBN} on a physically separate
10073 host, or controlling a standalone system over a serial port or a
10074 realtime system over a TCP/IP connection---you can use the @code{target}
10075 command to specify one of the target types configured for @value{GDBN}
10076 (@pxref{Target Commands, ,Commands for managing targets}).
10079 * Active Targets:: Active targets
10080 * Target Commands:: Commands for managing targets
10081 * Byte Order:: Choosing target byte order
10082 * Remote:: Remote debugging
10083 * KOD:: Kernel Object Display
10087 @node Active Targets
10088 @section Active targets
10090 @cindex stacking targets
10091 @cindex active targets
10092 @cindex multiple targets
10094 There are three classes of targets: processes, core files, and
10095 executable files. @value{GDBN} can work concurrently on up to three
10096 active targets, one in each class. This allows you to (for example)
10097 start a process and inspect its activity without abandoning your work on
10100 For example, if you execute @samp{gdb a.out}, then the executable file
10101 @code{a.out} is the only active target. If you designate a core file as
10102 well---presumably from a prior run that crashed and coredumped---then
10103 @value{GDBN} has two active targets and uses them in tandem, looking
10104 first in the corefile target, then in the executable file, to satisfy
10105 requests for memory addresses. (Typically, these two classes of target
10106 are complementary, since core files contain only a program's
10107 read-write memory---variables and so on---plus machine status, while
10108 executable files contain only the program text and initialized data.)
10110 When you type @code{run}, your executable file becomes an active process
10111 target as well. When a process target is active, all @value{GDBN}
10112 commands requesting memory addresses refer to that target; addresses in
10113 an active core file or executable file target are obscured while the
10114 process target is active.
10116 Use the @code{core-file} and @code{exec-file} commands to select a new
10117 core file or executable target (@pxref{Files, ,Commands to specify
10118 files}). To specify as a target a process that is already running, use
10119 the @code{attach} command (@pxref{Attach, ,Debugging an already-running
10122 @node Target Commands
10123 @section Commands for managing targets
10126 @item target @var{type} @var{parameters}
10127 Connects the @value{GDBN} host environment to a target machine or
10128 process. A target is typically a protocol for talking to debugging
10129 facilities. You use the argument @var{type} to specify the type or
10130 protocol of the target machine.
10132 Further @var{parameters} are interpreted by the target protocol, but
10133 typically include things like device names or host names to connect
10134 with, process numbers, and baud rates.
10136 The @code{target} command does not repeat if you press @key{RET} again
10137 after executing the command.
10139 @kindex help target
10141 Displays the names of all targets available. To display targets
10142 currently selected, use either @code{info target} or @code{info files}
10143 (@pxref{Files, ,Commands to specify files}).
10145 @item help target @var{name}
10146 Describe a particular target, including any parameters necessary to
10149 @kindex set gnutarget
10150 @item set gnutarget @var{args}
10151 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
10152 knows whether it is reading an @dfn{executable},
10153 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
10154 with the @code{set gnutarget} command. Unlike most @code{target} commands,
10155 with @code{gnutarget} the @code{target} refers to a program, not a machine.
10158 @emph{Warning:} To specify a file format with @code{set gnutarget},
10159 you must know the actual BFD name.
10163 @xref{Files, , Commands to specify files}.
10165 @kindex show gnutarget
10166 @item show gnutarget
10167 Use the @code{show gnutarget} command to display what file format
10168 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
10169 @value{GDBN} will determine the file format for each file automatically,
10170 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
10173 Here are some common targets (available, or not, depending on the GDB
10177 @kindex target exec
10178 @item target exec @var{program}
10179 An executable file. @samp{target exec @var{program}} is the same as
10180 @samp{exec-file @var{program}}.
10182 @kindex target core
10183 @item target core @var{filename}
10184 A core dump file. @samp{target core @var{filename}} is the same as
10185 @samp{core-file @var{filename}}.
10187 @kindex target remote
10188 @item target remote @var{dev}
10189 Remote serial target in GDB-specific protocol. The argument @var{dev}
10190 specifies what serial device to use for the connection (e.g.
10191 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}. @code{target remote}
10192 supports the @code{load} command. This is only useful if you have
10193 some other way of getting the stub to the target system, and you can put
10194 it somewhere in memory where it won't get clobbered by the download.
10198 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
10206 works; however, you cannot assume that a specific memory map, device
10207 drivers, or even basic I/O is available, although some simulators do
10208 provide these. For info about any processor-specific simulator details,
10209 see the appropriate section in @ref{Embedded Processors, ,Embedded
10214 Some configurations may include these targets as well:
10218 @kindex target nrom
10219 @item target nrom @var{dev}
10220 NetROM ROM emulator. This target only supports downloading.
10224 Different targets are available on different configurations of @value{GDBN};
10225 your configuration may have more or fewer targets.
10227 Many remote targets require you to download the executable's code
10228 once you've successfully established a connection.
10232 @kindex load @var{filename}
10233 @item load @var{filename}
10234 Depending on what remote debugging facilities are configured into
10235 @value{GDBN}, the @code{load} command may be available. Where it exists, it
10236 is meant to make @var{filename} (an executable) available for debugging
10237 on the remote system---by downloading, or dynamic linking, for example.
10238 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
10239 the @code{add-symbol-file} command.
10241 If your @value{GDBN} does not have a @code{load} command, attempting to
10242 execute it gets the error message ``@code{You can't do that when your
10243 target is @dots{}}''
10245 The file is loaded at whatever address is specified in the executable.
10246 For some object file formats, you can specify the load address when you
10247 link the program; for other formats, like a.out, the object file format
10248 specifies a fixed address.
10249 @c FIXME! This would be a good place for an xref to the GNU linker doc.
10251 @code{load} does not repeat if you press @key{RET} again after using it.
10255 @section Choosing target byte order
10257 @cindex choosing target byte order
10258 @cindex target byte order
10260 Some types of processors, such as the MIPS, PowerPC, and Hitachi SH,
10261 offer the ability to run either big-endian or little-endian byte
10262 orders. Usually the executable or symbol will include a bit to
10263 designate the endian-ness, and you will not need to worry about
10264 which to use. However, you may still find it useful to adjust
10265 @value{GDBN}'s idea of processor endian-ness manually.
10268 @kindex set endian big
10269 @item set endian big
10270 Instruct @value{GDBN} to assume the target is big-endian.
10272 @kindex set endian little
10273 @item set endian little
10274 Instruct @value{GDBN} to assume the target is little-endian.
10276 @kindex set endian auto
10277 @item set endian auto
10278 Instruct @value{GDBN} to use the byte order associated with the
10282 Display @value{GDBN}'s current idea of the target byte order.
10286 Note that these commands merely adjust interpretation of symbolic
10287 data on the host, and that they have absolutely no effect on the
10291 @section Remote debugging
10292 @cindex remote debugging
10294 If you are trying to debug a program running on a machine that cannot run
10295 @value{GDBN} in the usual way, it is often useful to use remote debugging.
10296 For example, you might use remote debugging on an operating system kernel,
10297 or on a small system which does not have a general purpose operating system
10298 powerful enough to run a full-featured debugger.
10300 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
10301 to make this work with particular debugging targets. In addition,
10302 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
10303 but not specific to any particular target system) which you can use if you
10304 write the remote stubs---the code that runs on the remote system to
10305 communicate with @value{GDBN}.
10307 Other remote targets may be available in your
10308 configuration of @value{GDBN}; use @code{help target} to list them.
10311 @section Kernel Object Display
10313 @cindex kernel object display
10314 @cindex kernel object
10317 Some targets support kernel object display. Using this facility,
10318 @value{GDBN} communicates specially with the underlying operating system
10319 and can display information about operating system-level objects such as
10320 mutexes and other synchronization objects. Exactly which objects can be
10321 displayed is determined on a per-OS basis.
10323 Use the @code{set os} command to set the operating system. This tells
10324 @value{GDBN} which kernel object display module to initialize:
10327 (@value{GDBP}) set os cisco
10330 If @code{set os} succeeds, @value{GDBN} will display some information
10331 about the operating system, and will create a new @code{info} command
10332 which can be used to query the target. The @code{info} command is named
10333 after the operating system:
10336 (@value{GDBP}) info cisco
10337 List of Cisco Kernel Objects
10339 any Any and all objects
10342 Further subcommands can be used to query about particular objects known
10345 There is currently no way to determine whether a given operating system
10346 is supported other than to try it.
10349 @node Remote Debugging
10350 @chapter Debugging remote programs
10353 * Server:: Using the gdbserver program
10354 * NetWare:: Using the gdbserve.nlm program
10355 * remote stub:: Implementing a remote stub
10359 @section Using the @code{gdbserver} program
10362 @cindex remote connection without stubs
10363 @code{gdbserver} is a control program for Unix-like systems, which
10364 allows you to connect your program with a remote @value{GDBN} via
10365 @code{target remote}---but without linking in the usual debugging stub.
10367 @code{gdbserver} is not a complete replacement for the debugging stubs,
10368 because it requires essentially the same operating-system facilities
10369 that @value{GDBN} itself does. In fact, a system that can run
10370 @code{gdbserver} to connect to a remote @value{GDBN} could also run
10371 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
10372 because it is a much smaller program than @value{GDBN} itself. It is
10373 also easier to port than all of @value{GDBN}, so you may be able to get
10374 started more quickly on a new system by using @code{gdbserver}.
10375 Finally, if you develop code for real-time systems, you may find that
10376 the tradeoffs involved in real-time operation make it more convenient to
10377 do as much development work as possible on another system, for example
10378 by cross-compiling. You can use @code{gdbserver} to make a similar
10379 choice for debugging.
10381 @value{GDBN} and @code{gdbserver} communicate via either a serial line
10382 or a TCP connection, using the standard @value{GDBN} remote serial
10386 @item On the target machine,
10387 you need to have a copy of the program you want to debug.
10388 @code{gdbserver} does not need your program's symbol table, so you can
10389 strip the program if necessary to save space. @value{GDBN} on the host
10390 system does all the symbol handling.
10392 To use the server, you must tell it how to communicate with @value{GDBN};
10393 the name of your program; and the arguments for your program. The usual
10397 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
10400 @var{comm} is either a device name (to use a serial line) or a TCP
10401 hostname and portnumber. For example, to debug Emacs with the argument
10402 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
10406 target> gdbserver /dev/com1 emacs foo.txt
10409 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
10412 To use a TCP connection instead of a serial line:
10415 target> gdbserver host:2345 emacs foo.txt
10418 The only difference from the previous example is the first argument,
10419 specifying that you are communicating with the host @value{GDBN} via
10420 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
10421 expect a TCP connection from machine @samp{host} to local TCP port 2345.
10422 (Currently, the @samp{host} part is ignored.) You can choose any number
10423 you want for the port number as long as it does not conflict with any
10424 TCP ports already in use on the target system (for example, @code{23} is
10425 reserved for @code{telnet}).@footnote{If you choose a port number that
10426 conflicts with another service, @code{gdbserver} prints an error message
10427 and exits.} You must use the same port number with the host @value{GDBN}
10428 @code{target remote} command.
10430 On some targets, @code{gdbserver} can also attach to running programs.
10431 This is accomplished via the @code{--attach} argument. The syntax is:
10434 target> gdbserver @var{comm} --attach @var{pid}
10437 @var{pid} is the process ID of a currently running process. It isn't necessary
10438 to point @code{gdbserver} at a binary for the running process.
10440 @item On the @value{GDBN} host machine,
10441 you need an unstripped copy of your program, since @value{GDBN} needs
10442 symbols and debugging information. Start up @value{GDBN} as usual,
10443 using the name of the local copy of your program as the first argument.
10444 (You may also need the @w{@samp{--baud}} option if the serial line is
10445 running at anything other than 9600@dmn{bps}.) After that, use @code{target
10446 remote} to establish communications with @code{gdbserver}. Its argument
10447 is either a device name (usually a serial device, like
10448 @file{/dev/ttyb}), or a TCP port descriptor in the form
10449 @code{@var{host}:@var{PORT}}. For example:
10452 (@value{GDBP}) target remote /dev/ttyb
10456 communicates with the server via serial line @file{/dev/ttyb}, and
10459 (@value{GDBP}) target remote the-target:2345
10463 communicates via a TCP connection to port 2345 on host @w{@file{the-target}}.
10464 For TCP connections, you must start up @code{gdbserver} prior to using
10465 the @code{target remote} command. Otherwise you may get an error whose
10466 text depends on the host system, but which usually looks something like
10467 @samp{Connection refused}.
10471 @section Using the @code{gdbserve.nlm} program
10473 @kindex gdbserve.nlm
10474 @code{gdbserve.nlm} is a control program for NetWare systems, which
10475 allows you to connect your program with a remote @value{GDBN} via
10476 @code{target remote}.
10478 @value{GDBN} and @code{gdbserve.nlm} communicate via a serial line,
10479 using the standard @value{GDBN} remote serial protocol.
10482 @item On the target machine,
10483 you need to have a copy of the program you want to debug.
10484 @code{gdbserve.nlm} does not need your program's symbol table, so you
10485 can strip the program if necessary to save space. @value{GDBN} on the
10486 host system does all the symbol handling.
10488 To use the server, you must tell it how to communicate with
10489 @value{GDBN}; the name of your program; and the arguments for your
10490 program. The syntax is:
10493 load gdbserve [ BOARD=@var{board} ] [ PORT=@var{port} ]
10494 [ BAUD=@var{baud} ] @var{program} [ @var{args} @dots{} ]
10497 @var{board} and @var{port} specify the serial line; @var{baud} specifies
10498 the baud rate used by the connection. @var{port} and @var{node} default
10499 to 0, @var{baud} defaults to 9600@dmn{bps}.
10501 For example, to debug Emacs with the argument @samp{foo.txt}and
10502 communicate with @value{GDBN} over serial port number 2 or board 1
10503 using a 19200@dmn{bps} connection:
10506 load gdbserve BOARD=1 PORT=2 BAUD=19200 emacs foo.txt
10509 @item On the @value{GDBN} host machine,
10510 you need an unstripped copy of your program, since @value{GDBN} needs
10511 symbols and debugging information. Start up @value{GDBN} as usual,
10512 using the name of the local copy of your program as the first argument.
10513 (You may also need the @w{@samp{--baud}} option if the serial line is
10514 running at anything other than 9600@dmn{bps}. After that, use @code{target
10515 remote} to establish communications with @code{gdbserve.nlm}. Its
10516 argument is a device name (usually a serial device, like
10517 @file{/dev/ttyb}). For example:
10520 (@value{GDBP}) target remote /dev/ttyb
10524 communications with the server via serial line @file{/dev/ttyb}.
10528 @section Implementing a remote stub
10530 @cindex debugging stub, example
10531 @cindex remote stub, example
10532 @cindex stub example, remote debugging
10533 The stub files provided with @value{GDBN} implement the target side of the
10534 communication protocol, and the @value{GDBN} side is implemented in the
10535 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
10536 these subroutines to communicate, and ignore the details. (If you're
10537 implementing your own stub file, you can still ignore the details: start
10538 with one of the existing stub files. @file{sparc-stub.c} is the best
10539 organized, and therefore the easiest to read.)
10541 @cindex remote serial debugging, overview
10542 To debug a program running on another machine (the debugging
10543 @dfn{target} machine), you must first arrange for all the usual
10544 prerequisites for the program to run by itself. For example, for a C
10549 A startup routine to set up the C runtime environment; these usually
10550 have a name like @file{crt0}. The startup routine may be supplied by
10551 your hardware supplier, or you may have to write your own.
10554 A C subroutine library to support your program's
10555 subroutine calls, notably managing input and output.
10558 A way of getting your program to the other machine---for example, a
10559 download program. These are often supplied by the hardware
10560 manufacturer, but you may have to write your own from hardware
10564 The next step is to arrange for your program to use a serial port to
10565 communicate with the machine where @value{GDBN} is running (the @dfn{host}
10566 machine). In general terms, the scheme looks like this:
10570 @value{GDBN} already understands how to use this protocol; when everything
10571 else is set up, you can simply use the @samp{target remote} command
10572 (@pxref{Targets,,Specifying a Debugging Target}).
10574 @item On the target,
10575 you must link with your program a few special-purpose subroutines that
10576 implement the @value{GDBN} remote serial protocol. The file containing these
10577 subroutines is called a @dfn{debugging stub}.
10579 On certain remote targets, you can use an auxiliary program
10580 @code{gdbserver} instead of linking a stub into your program.
10581 @xref{Server,,Using the @code{gdbserver} program}, for details.
10584 The debugging stub is specific to the architecture of the remote
10585 machine; for example, use @file{sparc-stub.c} to debug programs on
10588 @cindex remote serial stub list
10589 These working remote stubs are distributed with @value{GDBN}:
10594 @cindex @file{i386-stub.c}
10597 For Intel 386 and compatible architectures.
10600 @cindex @file{m68k-stub.c}
10601 @cindex Motorola 680x0
10603 For Motorola 680x0 architectures.
10606 @cindex @file{sh-stub.c}
10609 For Hitachi SH architectures.
10612 @cindex @file{sparc-stub.c}
10614 For @sc{sparc} architectures.
10616 @item sparcl-stub.c
10617 @cindex @file{sparcl-stub.c}
10620 For Fujitsu @sc{sparclite} architectures.
10624 The @file{README} file in the @value{GDBN} distribution may list other
10625 recently added stubs.
10628 * Stub Contents:: What the stub can do for you
10629 * Bootstrapping:: What you must do for the stub
10630 * Debug Session:: Putting it all together
10633 @node Stub Contents
10634 @subsection What the stub can do for you
10636 @cindex remote serial stub
10637 The debugging stub for your architecture supplies these three
10641 @item set_debug_traps
10642 @kindex set_debug_traps
10643 @cindex remote serial stub, initialization
10644 This routine arranges for @code{handle_exception} to run when your
10645 program stops. You must call this subroutine explicitly near the
10646 beginning of your program.
10648 @item handle_exception
10649 @kindex handle_exception
10650 @cindex remote serial stub, main routine
10651 This is the central workhorse, but your program never calls it
10652 explicitly---the setup code arranges for @code{handle_exception} to
10653 run when a trap is triggered.
10655 @code{handle_exception} takes control when your program stops during
10656 execution (for example, on a breakpoint), and mediates communications
10657 with @value{GDBN} on the host machine. This is where the communications
10658 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
10659 representative on the target machine. It begins by sending summary
10660 information on the state of your program, then continues to execute,
10661 retrieving and transmitting any information @value{GDBN} needs, until you
10662 execute a @value{GDBN} command that makes your program resume; at that point,
10663 @code{handle_exception} returns control to your own code on the target
10667 @cindex @code{breakpoint} subroutine, remote
10668 Use this auxiliary subroutine to make your program contain a
10669 breakpoint. Depending on the particular situation, this may be the only
10670 way for @value{GDBN} to get control. For instance, if your target
10671 machine has some sort of interrupt button, you won't need to call this;
10672 pressing the interrupt button transfers control to
10673 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
10674 simply receiving characters on the serial port may also trigger a trap;
10675 again, in that situation, you don't need to call @code{breakpoint} from
10676 your own program---simply running @samp{target remote} from the host
10677 @value{GDBN} session gets control.
10679 Call @code{breakpoint} if none of these is true, or if you simply want
10680 to make certain your program stops at a predetermined point for the
10681 start of your debugging session.
10684 @node Bootstrapping
10685 @subsection What you must do for the stub
10687 @cindex remote stub, support routines
10688 The debugging stubs that come with @value{GDBN} are set up for a particular
10689 chip architecture, but they have no information about the rest of your
10690 debugging target machine.
10692 First of all you need to tell the stub how to communicate with the
10696 @item int getDebugChar()
10697 @kindex getDebugChar
10698 Write this subroutine to read a single character from the serial port.
10699 It may be identical to @code{getchar} for your target system; a
10700 different name is used to allow you to distinguish the two if you wish.
10702 @item void putDebugChar(int)
10703 @kindex putDebugChar
10704 Write this subroutine to write a single character to the serial port.
10705 It may be identical to @code{putchar} for your target system; a
10706 different name is used to allow you to distinguish the two if you wish.
10709 @cindex control C, and remote debugging
10710 @cindex interrupting remote targets
10711 If you want @value{GDBN} to be able to stop your program while it is
10712 running, you need to use an interrupt-driven serial driver, and arrange
10713 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
10714 character). That is the character which @value{GDBN} uses to tell the
10715 remote system to stop.
10717 Getting the debugging target to return the proper status to @value{GDBN}
10718 probably requires changes to the standard stub; one quick and dirty way
10719 is to just execute a breakpoint instruction (the ``dirty'' part is that
10720 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
10722 Other routines you need to supply are:
10725 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
10726 @kindex exceptionHandler
10727 Write this function to install @var{exception_address} in the exception
10728 handling tables. You need to do this because the stub does not have any
10729 way of knowing what the exception handling tables on your target system
10730 are like (for example, the processor's table might be in @sc{rom},
10731 containing entries which point to a table in @sc{ram}).
10732 @var{exception_number} is the exception number which should be changed;
10733 its meaning is architecture-dependent (for example, different numbers
10734 might represent divide by zero, misaligned access, etc). When this
10735 exception occurs, control should be transferred directly to
10736 @var{exception_address}, and the processor state (stack, registers,
10737 and so on) should be just as it is when a processor exception occurs. So if
10738 you want to use a jump instruction to reach @var{exception_address}, it
10739 should be a simple jump, not a jump to subroutine.
10741 For the 386, @var{exception_address} should be installed as an interrupt
10742 gate so that interrupts are masked while the handler runs. The gate
10743 should be at privilege level 0 (the most privileged level). The
10744 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
10745 help from @code{exceptionHandler}.
10747 @item void flush_i_cache()
10748 @kindex flush_i_cache
10749 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
10750 instruction cache, if any, on your target machine. If there is no
10751 instruction cache, this subroutine may be a no-op.
10753 On target machines that have instruction caches, @value{GDBN} requires this
10754 function to make certain that the state of your program is stable.
10758 You must also make sure this library routine is available:
10761 @item void *memset(void *, int, int)
10763 This is the standard library function @code{memset} that sets an area of
10764 memory to a known value. If you have one of the free versions of
10765 @code{libc.a}, @code{memset} can be found there; otherwise, you must
10766 either obtain it from your hardware manufacturer, or write your own.
10769 If you do not use the GNU C compiler, you may need other standard
10770 library subroutines as well; this varies from one stub to another,
10771 but in general the stubs are likely to use any of the common library
10772 subroutines which @code{@value{GCC}} generates as inline code.
10775 @node Debug Session
10776 @subsection Putting it all together
10778 @cindex remote serial debugging summary
10779 In summary, when your program is ready to debug, you must follow these
10784 Make sure you have defined the supporting low-level routines
10785 (@pxref{Bootstrapping,,What you must do for the stub}):
10787 @code{getDebugChar}, @code{putDebugChar},
10788 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
10792 Insert these lines near the top of your program:
10800 For the 680x0 stub only, you need to provide a variable called
10801 @code{exceptionHook}. Normally you just use:
10804 void (*exceptionHook)() = 0;
10808 but if before calling @code{set_debug_traps}, you set it to point to a
10809 function in your program, that function is called when
10810 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
10811 error). The function indicated by @code{exceptionHook} is called with
10812 one parameter: an @code{int} which is the exception number.
10815 Compile and link together: your program, the @value{GDBN} debugging stub for
10816 your target architecture, and the supporting subroutines.
10819 Make sure you have a serial connection between your target machine and
10820 the @value{GDBN} host, and identify the serial port on the host.
10823 @c The "remote" target now provides a `load' command, so we should
10824 @c document that. FIXME.
10825 Download your program to your target machine (or get it there by
10826 whatever means the manufacturer provides), and start it.
10829 To start remote debugging, run @value{GDBN} on the host machine, and specify
10830 as an executable file the program that is running in the remote machine.
10831 This tells @value{GDBN} how to find your program's symbols and the contents
10835 @cindex serial line, @code{target remote}
10836 Establish communication using the @code{target remote} command.
10837 Its argument specifies how to communicate with the target
10838 machine---either via a devicename attached to a direct serial line, or a
10839 TCP or UDP port (usually to a terminal server which in turn has a serial line
10840 to the target). For example, to use a serial line connected to the
10841 device named @file{/dev/ttyb}:
10844 target remote /dev/ttyb
10847 @cindex TCP port, @code{target remote}
10848 To use a TCP connection, use an argument of the form
10849 @code{@var{host}:@var{port}} or @code{tcp:@var{host}:@var{port}}.
10850 For example, to connect to port 2828 on a
10851 terminal server named @code{manyfarms}:
10854 target remote manyfarms:2828
10857 If your remote target is actually running on the same machine as
10858 your debugger session (e.g.@: a simulator of your target running on
10859 the same host), you can omit the hostname. For example, to connect
10860 to port 1234 on your local machine:
10863 target remote :1234
10867 Note that the colon is still required here.
10869 @cindex UDP port, @code{target remote}
10870 To use a UDP connection, use an argument of the form
10871 @code{udp:@var{host}:@var{port}}. For example, to connect to UDP port 2828
10872 on a terminal server named @code{manyfarms}:
10875 target remote udp:manyfarms:2828
10878 When using a UDP connection for remote debugging, you should keep in mind
10879 that the `U' stands for ``Unreliable''. UDP can silently drop packets on
10880 busy or unreliable networks, which will cause havoc with your debugging
10885 Now you can use all the usual commands to examine and change data and to
10886 step and continue the remote program.
10888 To resume the remote program and stop debugging it, use the @code{detach}
10891 @cindex interrupting remote programs
10892 @cindex remote programs, interrupting
10893 Whenever @value{GDBN} is waiting for the remote program, if you type the
10894 interrupt character (often @key{C-C}), @value{GDBN} attempts to stop the
10895 program. This may or may not succeed, depending in part on the hardware
10896 and the serial drivers the remote system uses. If you type the
10897 interrupt character once again, @value{GDBN} displays this prompt:
10900 Interrupted while waiting for the program.
10901 Give up (and stop debugging it)? (y or n)
10904 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
10905 (If you decide you want to try again later, you can use @samp{target
10906 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
10907 goes back to waiting.
10910 @node Configurations
10911 @chapter Configuration-Specific Information
10913 While nearly all @value{GDBN} commands are available for all native and
10914 cross versions of the debugger, there are some exceptions. This chapter
10915 describes things that are only available in certain configurations.
10917 There are three major categories of configurations: native
10918 configurations, where the host and target are the same, embedded
10919 operating system configurations, which are usually the same for several
10920 different processor architectures, and bare embedded processors, which
10921 are quite different from each other.
10926 * Embedded Processors::
10933 This section describes details specific to particular native
10938 * SVR4 Process Information:: SVR4 process information
10939 * DJGPP Native:: Features specific to the DJGPP port
10940 * Cygwin Native:: Features specific to the Cygwin port
10946 On HP-UX systems, if you refer to a function or variable name that
10947 begins with a dollar sign, @value{GDBN} searches for a user or system
10948 name first, before it searches for a convenience variable.
10950 @node SVR4 Process Information
10951 @subsection SVR4 process information
10954 @cindex process image
10956 Many versions of SVR4 provide a facility called @samp{/proc} that can be
10957 used to examine the image of a running process using file-system
10958 subroutines. If @value{GDBN} is configured for an operating system with
10959 this facility, the command @code{info proc} is available to report on
10960 several kinds of information about the process running your program.
10961 @code{info proc} works only on SVR4 systems that include the
10962 @code{procfs} code. This includes OSF/1 (Digital Unix), Solaris, Irix,
10963 and Unixware, but not HP-UX or @sc{gnu}/Linux, for example.
10968 Summarize available information about the process.
10970 @kindex info proc mappings
10971 @item info proc mappings
10972 Report on the address ranges accessible in the program, with information
10973 on whether your program may read, write, or execute each range.
10975 @comment These sub-options of 'info proc' were not included when
10976 @comment procfs.c was re-written. Keep their descriptions around
10977 @comment against the day when someone finds the time to put them back in.
10978 @kindex info proc times
10979 @item info proc times
10980 Starting time, user CPU time, and system CPU time for your program and
10983 @kindex info proc id
10985 Report on the process IDs related to your program: its own process ID,
10986 the ID of its parent, the process group ID, and the session ID.
10988 @kindex info proc status
10989 @item info proc status
10990 General information on the state of the process. If the process is
10991 stopped, this report includes the reason for stopping, and any signal
10994 @item info proc all
10995 Show all the above information about the process.
11000 @subsection Features for Debugging @sc{djgpp} Programs
11001 @cindex @sc{djgpp} debugging
11002 @cindex native @sc{djgpp} debugging
11003 @cindex MS-DOS-specific commands
11005 @sc{djgpp} is the port of @sc{gnu} development tools to MS-DOS and
11006 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
11007 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
11008 top of real-mode DOS systems and their emulations.
11010 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
11011 defines a few commands specific to the @sc{djgpp} port. This
11012 subsection describes those commands.
11017 This is a prefix of @sc{djgpp}-specific commands which print
11018 information about the target system and important OS structures.
11021 @cindex MS-DOS system info
11022 @cindex free memory information (MS-DOS)
11023 @item info dos sysinfo
11024 This command displays assorted information about the underlying
11025 platform: the CPU type and features, the OS version and flavor, the
11026 DPMI version, and the available conventional and DPMI memory.
11031 @cindex segment descriptor tables
11032 @cindex descriptor tables display
11034 @itemx info dos ldt
11035 @itemx info dos idt
11036 These 3 commands display entries from, respectively, Global, Local,
11037 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
11038 tables are data structures which store a descriptor for each segment
11039 that is currently in use. The segment's selector is an index into a
11040 descriptor table; the table entry for that index holds the
11041 descriptor's base address and limit, and its attributes and access
11044 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
11045 segment (used for both data and the stack), and a DOS segment (which
11046 allows access to DOS/BIOS data structures and absolute addresses in
11047 conventional memory). However, the DPMI host will usually define
11048 additional segments in order to support the DPMI environment.
11050 @cindex garbled pointers
11051 These commands allow to display entries from the descriptor tables.
11052 Without an argument, all entries from the specified table are
11053 displayed. An argument, which should be an integer expression, means
11054 display a single entry whose index is given by the argument. For
11055 example, here's a convenient way to display information about the
11056 debugged program's data segment:
11059 @exdent @code{(@value{GDBP}) info dos ldt $ds}
11060 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
11064 This comes in handy when you want to see whether a pointer is outside
11065 the data segment's limit (i.e.@: @dfn{garbled}).
11067 @cindex page tables display (MS-DOS)
11069 @itemx info dos pte
11070 These two commands display entries from, respectively, the Page
11071 Directory and the Page Tables. Page Directories and Page Tables are
11072 data structures which control how virtual memory addresses are mapped
11073 into physical addresses. A Page Table includes an entry for every
11074 page of memory that is mapped into the program's address space; there
11075 may be several Page Tables, each one holding up to 4096 entries. A
11076 Page Directory has up to 4096 entries, one each for every Page Table
11077 that is currently in use.
11079 Without an argument, @kbd{info dos pde} displays the entire Page
11080 Directory, and @kbd{info dos pte} displays all the entries in all of
11081 the Page Tables. An argument, an integer expression, given to the
11082 @kbd{info dos pde} command means display only that entry from the Page
11083 Directory table. An argument given to the @kbd{info dos pte} command
11084 means display entries from a single Page Table, the one pointed to by
11085 the specified entry in the Page Directory.
11087 @cindex direct memory access (DMA) on MS-DOS
11088 These commands are useful when your program uses @dfn{DMA} (Direct
11089 Memory Access), which needs physical addresses to program the DMA
11092 These commands are supported only with some DPMI servers.
11094 @cindex physical address from linear address
11095 @item info dos address-pte @var{addr}
11096 This command displays the Page Table entry for a specified linear
11097 address. The argument linear address @var{addr} should already have the
11098 appropriate segment's base address added to it, because this command
11099 accepts addresses which may belong to @emph{any} segment. For
11100 example, here's how to display the Page Table entry for the page where
11101 the variable @code{i} is stored:
11104 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
11105 @exdent @code{Page Table entry for address 0x11a00d30:}
11106 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
11110 This says that @code{i} is stored at offset @code{0xd30} from the page
11111 whose physical base address is @code{0x02698000}, and prints all the
11112 attributes of that page.
11114 Note that you must cast the addresses of variables to a @code{char *},
11115 since otherwise the value of @code{__djgpp_base_address}, the base
11116 address of all variables and functions in a @sc{djgpp} program, will
11117 be added using the rules of C pointer arithmetics: if @code{i} is
11118 declared an @code{int}, @value{GDBN} will add 4 times the value of
11119 @code{__djgpp_base_address} to the address of @code{i}.
11121 Here's another example, it displays the Page Table entry for the
11125 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
11126 @exdent @code{Page Table entry for address 0x29110:}
11127 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
11131 (The @code{+ 3} offset is because the transfer buffer's address is the
11132 3rd member of the @code{_go32_info_block} structure.) The output of
11133 this command clearly shows that addresses in conventional memory are
11134 mapped 1:1, i.e.@: the physical and linear addresses are identical.
11136 This command is supported only with some DPMI servers.
11139 @node Cygwin Native
11140 @subsection Features for Debugging MS Windows PE executables
11141 @cindex MS Windows debugging
11142 @cindex native Cygwin debugging
11143 @cindex Cygwin-specific commands
11145 @value{GDBN} supports native debugging of MS Windows programs, and
11146 defines a few commands specific to the Cygwin port. This
11147 subsection describes those commands.
11152 This is a prefix of MS Windows specific commands which print
11153 information about the target system and important OS structures.
11155 @item info w32 selector
11156 This command displays information returned by
11157 the Win32 API @code{GetThreadSelectorEntry} function.
11158 It takes an optional argument that is evaluated to
11159 a long value to give the information about this given selector.
11160 Without argument, this command displays information
11161 about the the six segment registers.
11165 This is a Cygwin specific alias of info shared.
11167 @kindex dll-symbols
11169 This command loads symbols from a dll similarly to
11170 add-sym command but without the need to specify a base address.
11172 @kindex set new-console
11173 @item set new-console @var{mode}
11174 If @var{mode} is @code{on} the debuggee will
11175 be started in a new console on next start.
11176 If @var{mode} is @code{off}i, the debuggee will
11177 be started in the same console as the debugger.
11179 @kindex show new-console
11180 @item show new-console
11181 Displays whether a new console is used
11182 when the debuggee is started.
11184 @kindex set new-group
11185 @item set new-group @var{mode}
11186 This boolean value controls whether the debuggee should
11187 start a new group or stay in the same group as the debugger.
11188 This affects the way the Windows OS handles
11191 @kindex show new-group
11192 @item show new-group
11193 Displays current value of new-group boolean.
11195 @kindex set debugevents
11196 @item set debugevents
11197 This boolean value adds debug output concerning events seen by the debugger.
11199 @kindex set debugexec
11200 @item set debugexec
11201 This boolean value adds debug output concerning execute events
11202 seen by the debugger.
11204 @kindex set debugexceptions
11205 @item set debugexceptions
11206 This boolean value adds debug ouptut concerning exception events
11207 seen by the debugger.
11209 @kindex set debugmemory
11210 @item set debugmemory
11211 This boolean value adds debug ouptut concerning memory events
11212 seen by the debugger.
11216 This boolean values specifies whether the debuggee is called
11217 via a shell or directly (default value is on).
11221 Displays if the debuggee will be started with a shell.
11226 @section Embedded Operating Systems
11228 This section describes configurations involving the debugging of
11229 embedded operating systems that are available for several different
11233 * VxWorks:: Using @value{GDBN} with VxWorks
11236 @value{GDBN} includes the ability to debug programs running on
11237 various real-time operating systems.
11240 @subsection Using @value{GDBN} with VxWorks
11246 @kindex target vxworks
11247 @item target vxworks @var{machinename}
11248 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
11249 is the target system's machine name or IP address.
11253 On VxWorks, @code{load} links @var{filename} dynamically on the
11254 current target system as well as adding its symbols in @value{GDBN}.
11256 @value{GDBN} enables developers to spawn and debug tasks running on networked
11257 VxWorks targets from a Unix host. Already-running tasks spawned from
11258 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
11259 both the Unix host and on the VxWorks target. The program
11260 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
11261 installed with the name @code{vxgdb}, to distinguish it from a
11262 @value{GDBN} for debugging programs on the host itself.)
11265 @item VxWorks-timeout @var{args}
11266 @kindex vxworks-timeout
11267 All VxWorks-based targets now support the option @code{vxworks-timeout}.
11268 This option is set by the user, and @var{args} represents the number of
11269 seconds @value{GDBN} waits for responses to rpc's. You might use this if
11270 your VxWorks target is a slow software simulator or is on the far side
11271 of a thin network line.
11274 The following information on connecting to VxWorks was current when
11275 this manual was produced; newer releases of VxWorks may use revised
11278 @kindex INCLUDE_RDB
11279 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
11280 to include the remote debugging interface routines in the VxWorks
11281 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
11282 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
11283 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
11284 source debugging task @code{tRdbTask} when VxWorks is booted. For more
11285 information on configuring and remaking VxWorks, see the manufacturer's
11287 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
11289 Once you have included @file{rdb.a} in your VxWorks system image and set
11290 your Unix execution search path to find @value{GDBN}, you are ready to
11291 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
11292 @code{vxgdb}, depending on your installation).
11294 @value{GDBN} comes up showing the prompt:
11301 * VxWorks Connection:: Connecting to VxWorks
11302 * VxWorks Download:: VxWorks download
11303 * VxWorks Attach:: Running tasks
11306 @node VxWorks Connection
11307 @subsubsection Connecting to VxWorks
11309 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
11310 network. To connect to a target whose host name is ``@code{tt}'', type:
11313 (vxgdb) target vxworks tt
11317 @value{GDBN} displays messages like these:
11320 Attaching remote machine across net...
11325 @value{GDBN} then attempts to read the symbol tables of any object modules
11326 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
11327 these files by searching the directories listed in the command search
11328 path (@pxref{Environment, ,Your program's environment}); if it fails
11329 to find an object file, it displays a message such as:
11332 prog.o: No such file or directory.
11335 When this happens, add the appropriate directory to the search path with
11336 the @value{GDBN} command @code{path}, and execute the @code{target}
11339 @node VxWorks Download
11340 @subsubsection VxWorks download
11342 @cindex download to VxWorks
11343 If you have connected to the VxWorks target and you want to debug an
11344 object that has not yet been loaded, you can use the @value{GDBN}
11345 @code{load} command to download a file from Unix to VxWorks
11346 incrementally. The object file given as an argument to the @code{load}
11347 command is actually opened twice: first by the VxWorks target in order
11348 to download the code, then by @value{GDBN} in order to read the symbol
11349 table. This can lead to problems if the current working directories on
11350 the two systems differ. If both systems have NFS mounted the same
11351 filesystems, you can avoid these problems by using absolute paths.
11352 Otherwise, it is simplest to set the working directory on both systems
11353 to the directory in which the object file resides, and then to reference
11354 the file by its name, without any path. For instance, a program
11355 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
11356 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
11357 program, type this on VxWorks:
11360 -> cd "@var{vxpath}/vw/demo/rdb"
11364 Then, in @value{GDBN}, type:
11367 (vxgdb) cd @var{hostpath}/vw/demo/rdb
11368 (vxgdb) load prog.o
11371 @value{GDBN} displays a response similar to this:
11374 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
11377 You can also use the @code{load} command to reload an object module
11378 after editing and recompiling the corresponding source file. Note that
11379 this makes @value{GDBN} delete all currently-defined breakpoints,
11380 auto-displays, and convenience variables, and to clear the value
11381 history. (This is necessary in order to preserve the integrity of
11382 debugger's data structures that reference the target system's symbol
11385 @node VxWorks Attach
11386 @subsubsection Running tasks
11388 @cindex running VxWorks tasks
11389 You can also attach to an existing task using the @code{attach} command as
11393 (vxgdb) attach @var{task}
11397 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
11398 or suspended when you attach to it. Running tasks are suspended at
11399 the time of attachment.
11401 @node Embedded Processors
11402 @section Embedded Processors
11404 This section goes into details specific to particular embedded
11410 * H8/300:: Hitachi H8/300
11411 * H8/500:: Hitachi H8/500
11412 * M32R/D:: Mitsubishi M32R/D
11413 * M68K:: Motorola M68K
11414 * MIPS Embedded:: MIPS Embedded
11415 * OpenRISC 1000:: OpenRisc 1000
11416 * PA:: HP PA Embedded
11419 * Sparclet:: Tsqware Sparclet
11420 * Sparclite:: Fujitsu Sparclite
11421 * ST2000:: Tandem ST2000
11422 * Z8000:: Zilog Z8000
11431 @item target rdi @var{dev}
11432 ARM Angel monitor, via RDI library interface to ADP protocol. You may
11433 use this target to communicate with both boards running the Angel
11434 monitor, or with the EmbeddedICE JTAG debug device.
11437 @item target rdp @var{dev}
11443 @subsection Hitachi H8/300
11447 @kindex target hms@r{, with H8/300}
11448 @item target hms @var{dev}
11449 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
11450 Use special commands @code{device} and @code{speed} to control the serial
11451 line and the communications speed used.
11453 @kindex target e7000@r{, with H8/300}
11454 @item target e7000 @var{dev}
11455 E7000 emulator for Hitachi H8 and SH.
11457 @kindex target sh3@r{, with H8/300}
11458 @kindex target sh3e@r{, with H8/300}
11459 @item target sh3 @var{dev}
11460 @itemx target sh3e @var{dev}
11461 Hitachi SH-3 and SH-3E target systems.
11465 @cindex download to H8/300 or H8/500
11466 @cindex H8/300 or H8/500 download
11467 @cindex download to Hitachi SH
11468 @cindex Hitachi SH download
11469 When you select remote debugging to a Hitachi SH, H8/300, or H8/500
11470 board, the @code{load} command downloads your program to the Hitachi
11471 board and also opens it as the current executable target for
11472 @value{GDBN} on your host (like the @code{file} command).
11474 @value{GDBN} needs to know these things to talk to your
11475 Hitachi SH, H8/300, or H8/500:
11479 that you want to use @samp{target hms}, the remote debugging interface
11480 for Hitachi microprocessors, or @samp{target e7000}, the in-circuit
11481 emulator for the Hitachi SH and the Hitachi 300H. (@samp{target hms} is
11482 the default when @value{GDBN} is configured specifically for the Hitachi SH,
11483 H8/300, or H8/500.)
11486 what serial device connects your host to your Hitachi board (the first
11487 serial device available on your host is the default).
11490 what speed to use over the serial device.
11494 * Hitachi Boards:: Connecting to Hitachi boards.
11495 * Hitachi ICE:: Using the E7000 In-Circuit Emulator.
11496 * Hitachi Special:: Special @value{GDBN} commands for Hitachi micros.
11499 @node Hitachi Boards
11500 @subsubsection Connecting to Hitachi boards
11502 @c only for Unix hosts
11504 @cindex serial device, Hitachi micros
11505 Use the special @code{@value{GDBN}} command @samp{device @var{port}} if you
11506 need to explicitly set the serial device. The default @var{port} is the
11507 first available port on your host. This is only necessary on Unix
11508 hosts, where it is typically something like @file{/dev/ttya}.
11511 @cindex serial line speed, Hitachi micros
11512 @code{@value{GDBN}} has another special command to set the communications
11513 speed: @samp{speed @var{bps}}. This command also is only used from Unix
11514 hosts; on DOS hosts, set the line speed as usual from outside @value{GDBN} with
11515 the DOS @code{mode} command (for instance,
11516 @w{@kbd{mode com2:9600,n,8,1,p}} for a 9600@dmn{bps} connection).
11518 The @samp{device} and @samp{speed} commands are available only when you
11519 use a Unix host to debug your Hitachi microprocessor programs. If you
11521 @value{GDBN} depends on an auxiliary terminate-and-stay-resident program
11522 called @code{asynctsr} to communicate with the development board
11523 through a PC serial port. You must also use the DOS @code{mode} command
11524 to set up the serial port on the DOS side.
11526 The following sample session illustrates the steps needed to start a
11527 program under @value{GDBN} control on an H8/300. The example uses a
11528 sample H8/300 program called @file{t.x}. The procedure is the same for
11529 the Hitachi SH and the H8/500.
11531 First hook up your development board. In this example, we use a
11532 board attached to serial port @code{COM2}; if you use a different serial
11533 port, substitute its name in the argument of the @code{mode} command.
11534 When you call @code{asynctsr}, the auxiliary comms program used by the
11535 debugger, you give it just the numeric part of the serial port's name;
11536 for example, @samp{asyncstr 2} below runs @code{asyncstr} on
11540 C:\H8300\TEST> asynctsr 2
11541 C:\H8300\TEST> mode com2:9600,n,8,1,p
11543 Resident portion of MODE loaded
11545 COM2: 9600, n, 8, 1, p
11550 @emph{Warning:} We have noticed a bug in PC-NFS that conflicts with
11551 @code{asynctsr}. If you also run PC-NFS on your DOS host, you may need to
11552 disable it, or even boot without it, to use @code{asynctsr} to control
11553 your development board.
11556 @kindex target hms@r{, and serial protocol}
11557 Now that serial communications are set up, and the development board is
11558 connected, you can start up @value{GDBN}. Call @code{@value{GDBP}} with
11559 the name of your program as the argument. @code{@value{GDBN}} prompts
11560 you, as usual, with the prompt @samp{(@value{GDBP})}. Use two special
11561 commands to begin your debugging session: @samp{target hms} to specify
11562 cross-debugging to the Hitachi board, and the @code{load} command to
11563 download your program to the board. @code{load} displays the names of
11564 the program's sections, and a @samp{*} for each 2K of data downloaded.
11565 (If you want to refresh @value{GDBN} data on symbols or on the
11566 executable file without downloading, use the @value{GDBN} commands
11567 @code{file} or @code{symbol-file}. These commands, and @code{load}
11568 itself, are described in @ref{Files,,Commands to specify files}.)
11571 (eg-C:\H8300\TEST) @value{GDBP} t.x
11572 @value{GDBN} is free software and you are welcome to distribute copies
11573 of it under certain conditions; type "show copying" to see
11575 There is absolutely no warranty for @value{GDBN}; type "show warranty"
11577 @value{GDBN} @value{GDBVN}, Copyright 1992 Free Software Foundation, Inc...
11578 (@value{GDBP}) target hms
11579 Connected to remote H8/300 HMS system.
11580 (@value{GDBP}) load t.x
11581 .text : 0x8000 .. 0xabde ***********
11582 .data : 0xabde .. 0xad30 *
11583 .stack : 0xf000 .. 0xf014 *
11586 At this point, you're ready to run or debug your program. From here on,
11587 you can use all the usual @value{GDBN} commands. The @code{break} command
11588 sets breakpoints; the @code{run} command starts your program;
11589 @code{print} or @code{x} display data; the @code{continue} command
11590 resumes execution after stopping at a breakpoint. You can use the
11591 @code{help} command at any time to find out more about @value{GDBN} commands.
11593 Remember, however, that @emph{operating system} facilities aren't
11594 available on your development board; for example, if your program hangs,
11595 you can't send an interrupt---but you can press the @sc{reset} switch!
11597 Use the @sc{reset} button on the development board
11600 to interrupt your program (don't use @kbd{ctl-C} on the DOS host---it has
11601 no way to pass an interrupt signal to the development board); and
11604 to return to the @value{GDBN} command prompt after your program finishes
11605 normally. The communications protocol provides no other way for @value{GDBN}
11606 to detect program completion.
11609 In either case, @value{GDBN} sees the effect of a @sc{reset} on the
11610 development board as a ``normal exit'' of your program.
11613 @subsubsection Using the E7000 in-circuit emulator
11615 @kindex target e7000@r{, with Hitachi ICE}
11616 You can use the E7000 in-circuit emulator to develop code for either the
11617 Hitachi SH or the H8/300H. Use one of these forms of the @samp{target
11618 e7000} command to connect @value{GDBN} to your E7000:
11621 @item target e7000 @var{port} @var{speed}
11622 Use this form if your E7000 is connected to a serial port. The
11623 @var{port} argument identifies what serial port to use (for example,
11624 @samp{com2}). The third argument is the line speed in bits per second
11625 (for example, @samp{9600}).
11627 @item target e7000 @var{hostname}
11628 If your E7000 is installed as a host on a TCP/IP network, you can just
11629 specify its hostname; @value{GDBN} uses @code{telnet} to connect.
11632 @node Hitachi Special
11633 @subsubsection Special @value{GDBN} commands for Hitachi micros
11635 Some @value{GDBN} commands are available only for the H8/300:
11639 @kindex set machine
11640 @kindex show machine
11641 @item set machine h8300
11642 @itemx set machine h8300h
11643 Condition @value{GDBN} for one of the two variants of the H8/300
11644 architecture with @samp{set machine}. You can use @samp{show machine}
11645 to check which variant is currently in effect.
11654 @kindex set memory @var{mod}
11655 @cindex memory models, H8/500
11656 @item set memory @var{mod}
11658 Specify which H8/500 memory model (@var{mod}) you are using with
11659 @samp{set memory}; check which memory model is in effect with @samp{show
11660 memory}. The accepted values for @var{mod} are @code{small},
11661 @code{big}, @code{medium}, and @code{compact}.
11666 @subsection Mitsubishi M32R/D
11670 @kindex target m32r
11671 @item target m32r @var{dev}
11672 Mitsubishi M32R/D ROM monitor.
11679 The Motorola m68k configuration includes ColdFire support, and
11680 target command for the following ROM monitors.
11684 @kindex target abug
11685 @item target abug @var{dev}
11686 ABug ROM monitor for M68K.
11688 @kindex target cpu32bug
11689 @item target cpu32bug @var{dev}
11690 CPU32BUG monitor, running on a CPU32 (M68K) board.
11692 @kindex target dbug
11693 @item target dbug @var{dev}
11694 dBUG ROM monitor for Motorola ColdFire.
11697 @item target est @var{dev}
11698 EST-300 ICE monitor, running on a CPU32 (M68K) board.
11700 @kindex target rom68k
11701 @item target rom68k @var{dev}
11702 ROM 68K monitor, running on an M68K IDP board.
11708 @kindex target rombug
11709 @item target rombug @var{dev}
11710 ROMBUG ROM monitor for OS/9000.
11714 @node MIPS Embedded
11715 @subsection MIPS Embedded
11717 @cindex MIPS boards
11718 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
11719 MIPS board attached to a serial line. This is available when
11720 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
11723 Use these @value{GDBN} commands to specify the connection to your target board:
11726 @item target mips @var{port}
11727 @kindex target mips @var{port}
11728 To run a program on the board, start up @code{@value{GDBP}} with the
11729 name of your program as the argument. To connect to the board, use the
11730 command @samp{target mips @var{port}}, where @var{port} is the name of
11731 the serial port connected to the board. If the program has not already
11732 been downloaded to the board, you may use the @code{load} command to
11733 download it. You can then use all the usual @value{GDBN} commands.
11735 For example, this sequence connects to the target board through a serial
11736 port, and loads and runs a program called @var{prog} through the
11740 host$ @value{GDBP} @var{prog}
11741 @value{GDBN} is free software and @dots{}
11742 (@value{GDBP}) target mips /dev/ttyb
11743 (@value{GDBP}) load @var{prog}
11747 @item target mips @var{hostname}:@var{portnumber}
11748 On some @value{GDBN} host configurations, you can specify a TCP
11749 connection (for instance, to a serial line managed by a terminal
11750 concentrator) instead of a serial port, using the syntax
11751 @samp{@var{hostname}:@var{portnumber}}.
11753 @item target pmon @var{port}
11754 @kindex target pmon @var{port}
11757 @item target ddb @var{port}
11758 @kindex target ddb @var{port}
11759 NEC's DDB variant of PMON for Vr4300.
11761 @item target lsi @var{port}
11762 @kindex target lsi @var{port}
11763 LSI variant of PMON.
11765 @kindex target r3900
11766 @item target r3900 @var{dev}
11767 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
11769 @kindex target array
11770 @item target array @var{dev}
11771 Array Tech LSI33K RAID controller board.
11777 @value{GDBN} also supports these special commands for MIPS targets:
11780 @item set processor @var{args}
11781 @itemx show processor
11782 @kindex set processor @var{args}
11783 @kindex show processor
11784 Use the @code{set processor} command to set the type of MIPS
11785 processor when you want to access processor-type-specific registers.
11786 For example, @code{set processor @var{r3041}} tells @value{GDBN}
11787 to use the CPU registers appropriate for the 3041 chip.
11788 Use the @code{show processor} command to see what MIPS processor @value{GDBN}
11789 is using. Use the @code{info reg} command to see what registers
11790 @value{GDBN} is using.
11792 @item set mipsfpu double
11793 @itemx set mipsfpu single
11794 @itemx set mipsfpu none
11795 @itemx show mipsfpu
11796 @kindex set mipsfpu
11797 @kindex show mipsfpu
11798 @cindex MIPS remote floating point
11799 @cindex floating point, MIPS remote
11800 If your target board does not support the MIPS floating point
11801 coprocessor, you should use the command @samp{set mipsfpu none} (if you
11802 need this, you may wish to put the command in your @value{GDBN} init
11803 file). This tells @value{GDBN} how to find the return value of
11804 functions which return floating point values. It also allows
11805 @value{GDBN} to avoid saving the floating point registers when calling
11806 functions on the board. If you are using a floating point coprocessor
11807 with only single precision floating point support, as on the @sc{r4650}
11808 processor, use the command @samp{set mipsfpu single}. The default
11809 double precision floating point coprocessor may be selected using
11810 @samp{set mipsfpu double}.
11812 In previous versions the only choices were double precision or no
11813 floating point, so @samp{set mipsfpu on} will select double precision
11814 and @samp{set mipsfpu off} will select no floating point.
11816 As usual, you can inquire about the @code{mipsfpu} variable with
11817 @samp{show mipsfpu}.
11819 @item set remotedebug @var{n}
11820 @itemx show remotedebug
11821 @kindex set remotedebug@r{, MIPS protocol}
11822 @kindex show remotedebug@r{, MIPS protocol}
11823 @cindex @code{remotedebug}, MIPS protocol
11824 @cindex MIPS @code{remotedebug} protocol
11825 @c FIXME! For this to be useful, you must know something about the MIPS
11826 @c FIXME...protocol. Where is it described?
11827 You can see some debugging information about communications with the board
11828 by setting the @code{remotedebug} variable. If you set it to @code{1} using
11829 @samp{set remotedebug 1}, every packet is displayed. If you set it
11830 to @code{2}, every character is displayed. You can check the current value
11831 at any time with the command @samp{show remotedebug}.
11833 @item set timeout @var{seconds}
11834 @itemx set retransmit-timeout @var{seconds}
11835 @itemx show timeout
11836 @itemx show retransmit-timeout
11837 @cindex @code{timeout}, MIPS protocol
11838 @cindex @code{retransmit-timeout}, MIPS protocol
11839 @kindex set timeout
11840 @kindex show timeout
11841 @kindex set retransmit-timeout
11842 @kindex show retransmit-timeout
11843 You can control the timeout used while waiting for a packet, in the MIPS
11844 remote protocol, with the @code{set timeout @var{seconds}} command. The
11845 default is 5 seconds. Similarly, you can control the timeout used while
11846 waiting for an acknowledgement of a packet with the @code{set
11847 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
11848 You can inspect both values with @code{show timeout} and @code{show
11849 retransmit-timeout}. (These commands are @emph{only} available when
11850 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
11852 The timeout set by @code{set timeout} does not apply when @value{GDBN}
11853 is waiting for your program to stop. In that case, @value{GDBN} waits
11854 forever because it has no way of knowing how long the program is going
11855 to run before stopping.
11858 @node OpenRISC 1000
11859 @subsection OpenRISC 1000
11860 @cindex OpenRISC 1000
11862 @cindex or1k boards
11863 See OR1k Architecture document (@uref{www.opencores.org}) for more information
11864 about platform and commands.
11868 @kindex target jtag
11869 @item target jtag jtag://@var{host}:@var{port}
11871 Connects to remote JTAG server.
11872 JTAG remote server can be either an or1ksim or JTAG server,
11873 connected via parallel port to the board.
11875 Example: @code{target jtag jtag://localhost:9999}
11878 @item or1ksim @var{command}
11879 If connected to @code{or1ksim} OpenRISC 1000 Architectural
11880 Simulator, proprietary commands can be executed.
11882 @kindex info or1k spr
11883 @item info or1k spr
11884 Displays spr groups.
11886 @item info or1k spr @var{group}
11887 @itemx info or1k spr @var{groupno}
11888 Displays register names in selected group.
11890 @item info or1k spr @var{group} @var{register}
11891 @itemx info or1k spr @var{register}
11892 @itemx info or1k spr @var{groupno} @var{registerno}
11893 @itemx info or1k spr @var{registerno}
11894 Shows information about specified spr register.
11897 @item spr @var{group} @var{register} @var{value}
11898 @itemx spr @var{register @var{value}}
11899 @itemx spr @var{groupno} @var{registerno @var{value}}
11900 @itemx spr @var{registerno @var{value}}
11901 Writes @var{value} to specified spr register.
11904 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
11905 It is very similar to @value{GDBN} trace, except it does not interfere with normal
11906 program execution and is thus much faster. Hardware breakpoints/watchpoint
11907 triggers can be set using:
11910 Load effective address/data
11912 Store effective address/data
11914 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
11919 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
11920 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
11922 @code{htrace} commands:
11923 @cindex OpenRISC 1000 htrace
11926 @item hwatch @var{conditional}
11927 Set hardware watchpoint on combination of Load/Store Effecive Address(es)
11928 or Data. For example:
11930 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
11932 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
11934 @kindex htrace info
11936 Display information about current HW trace configuration.
11938 @kindex htrace trigger
11939 @item htrace trigger @var{conditional}
11940 Set starting criteria for HW trace.
11942 @kindex htrace qualifier
11943 @item htrace qualifier @var{conditional}
11944 Set acquisition qualifier for HW trace.
11946 @kindex htrace stop
11947 @item htrace stop @var{conditional}
11948 Set HW trace stopping criteria.
11950 @kindex htrace record
11951 @item htrace record [@var{data}]*
11952 Selects the data to be recorded, when qualifier is met and HW trace was
11955 @kindex htrace enable
11956 @item htrace enable
11957 @kindex htrace disable
11958 @itemx htrace disable
11959 Enables/disables the HW trace.
11961 @kindex htrace rewind
11962 @item htrace rewind [@var{filename}]
11963 Clears currently recorded trace data.
11965 If filename is specified, new trace file is made and any newly collected data
11966 will be written there.
11968 @kindex htrace print
11969 @item htrace print [@var{start} [@var{len}]]
11970 Prints trace buffer, using current record configuration.
11972 @kindex htrace mode continuous
11973 @item htrace mode continuous
11974 Set continuous trace mode.
11976 @kindex htrace mode suspend
11977 @item htrace mode suspend
11978 Set suspend trace mode.
11983 @subsection PowerPC
11987 @kindex target dink32
11988 @item target dink32 @var{dev}
11989 DINK32 ROM monitor.
11991 @kindex target ppcbug
11992 @item target ppcbug @var{dev}
11993 @kindex target ppcbug1
11994 @item target ppcbug1 @var{dev}
11995 PPCBUG ROM monitor for PowerPC.
11998 @item target sds @var{dev}
11999 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
12004 @subsection HP PA Embedded
12008 @kindex target op50n
12009 @item target op50n @var{dev}
12010 OP50N monitor, running on an OKI HPPA board.
12012 @kindex target w89k
12013 @item target w89k @var{dev}
12014 W89K monitor, running on a Winbond HPPA board.
12019 @subsection Hitachi SH
12023 @kindex target hms@r{, with Hitachi SH}
12024 @item target hms @var{dev}
12025 A Hitachi SH board attached via serial line to your host. Use special
12026 commands @code{device} and @code{speed} to control the serial line and
12027 the communications speed used.
12029 @kindex target e7000@r{, with Hitachi SH}
12030 @item target e7000 @var{dev}
12031 E7000 emulator for Hitachi SH.
12033 @kindex target sh3@r{, with SH}
12034 @kindex target sh3e@r{, with SH}
12035 @item target sh3 @var{dev}
12036 @item target sh3e @var{dev}
12037 Hitachi SH-3 and SH-3E target systems.
12042 @subsection Tsqware Sparclet
12046 @value{GDBN} enables developers to debug tasks running on
12047 Sparclet targets from a Unix host.
12048 @value{GDBN} uses code that runs on
12049 both the Unix host and on the Sparclet target. The program
12050 @code{@value{GDBP}} is installed and executed on the Unix host.
12053 @item remotetimeout @var{args}
12054 @kindex remotetimeout
12055 @value{GDBN} supports the option @code{remotetimeout}.
12056 This option is set by the user, and @var{args} represents the number of
12057 seconds @value{GDBN} waits for responses.
12060 @cindex compiling, on Sparclet
12061 When compiling for debugging, include the options @samp{-g} to get debug
12062 information and @samp{-Ttext} to relocate the program to where you wish to
12063 load it on the target. You may also want to add the options @samp{-n} or
12064 @samp{-N} in order to reduce the size of the sections. Example:
12067 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
12070 You can use @code{objdump} to verify that the addresses are what you intended:
12073 sparclet-aout-objdump --headers --syms prog
12076 @cindex running, on Sparclet
12078 your Unix execution search path to find @value{GDBN}, you are ready to
12079 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
12080 (or @code{sparclet-aout-gdb}, depending on your installation).
12082 @value{GDBN} comes up showing the prompt:
12089 * Sparclet File:: Setting the file to debug
12090 * Sparclet Connection:: Connecting to Sparclet
12091 * Sparclet Download:: Sparclet download
12092 * Sparclet Execution:: Running and debugging
12095 @node Sparclet File
12096 @subsubsection Setting file to debug
12098 The @value{GDBN} command @code{file} lets you choose with program to debug.
12101 (gdbslet) file prog
12105 @value{GDBN} then attempts to read the symbol table of @file{prog}.
12106 @value{GDBN} locates
12107 the file by searching the directories listed in the command search
12109 If the file was compiled with debug information (option "-g"), source
12110 files will be searched as well.
12111 @value{GDBN} locates
12112 the source files by searching the directories listed in the directory search
12113 path (@pxref{Environment, ,Your program's environment}).
12115 to find a file, it displays a message such as:
12118 prog: No such file or directory.
12121 When this happens, add the appropriate directories to the search paths with
12122 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
12123 @code{target} command again.
12125 @node Sparclet Connection
12126 @subsubsection Connecting to Sparclet
12128 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
12129 To connect to a target on serial port ``@code{ttya}'', type:
12132 (gdbslet) target sparclet /dev/ttya
12133 Remote target sparclet connected to /dev/ttya
12134 main () at ../prog.c:3
12138 @value{GDBN} displays messages like these:
12144 @node Sparclet Download
12145 @subsubsection Sparclet download
12147 @cindex download to Sparclet
12148 Once connected to the Sparclet target,
12149 you can use the @value{GDBN}
12150 @code{load} command to download the file from the host to the target.
12151 The file name and load offset should be given as arguments to the @code{load}
12153 Since the file format is aout, the program must be loaded to the starting
12154 address. You can use @code{objdump} to find out what this value is. The load
12155 offset is an offset which is added to the VMA (virtual memory address)
12156 of each of the file's sections.
12157 For instance, if the program
12158 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
12159 and bss at 0x12010170, in @value{GDBN}, type:
12162 (gdbslet) load prog 0x12010000
12163 Loading section .text, size 0xdb0 vma 0x12010000
12166 If the code is loaded at a different address then what the program was linked
12167 to, you may need to use the @code{section} and @code{add-symbol-file} commands
12168 to tell @value{GDBN} where to map the symbol table.
12170 @node Sparclet Execution
12171 @subsubsection Running and debugging
12173 @cindex running and debugging Sparclet programs
12174 You can now begin debugging the task using @value{GDBN}'s execution control
12175 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
12176 manual for the list of commands.
12180 Breakpoint 1 at 0x12010000: file prog.c, line 3.
12182 Starting program: prog
12183 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
12184 3 char *symarg = 0;
12186 4 char *execarg = "hello!";
12191 @subsection Fujitsu Sparclite
12195 @kindex target sparclite
12196 @item target sparclite @var{dev}
12197 Fujitsu sparclite boards, used only for the purpose of loading.
12198 You must use an additional command to debug the program.
12199 For example: target remote @var{dev} using @value{GDBN} standard
12205 @subsection Tandem ST2000
12207 @value{GDBN} may be used with a Tandem ST2000 phone switch, running Tandem's
12210 To connect your ST2000 to the host system, see the manufacturer's
12211 manual. Once the ST2000 is physically attached, you can run:
12214 target st2000 @var{dev} @var{speed}
12218 to establish it as your debugging environment. @var{dev} is normally
12219 the name of a serial device, such as @file{/dev/ttya}, connected to the
12220 ST2000 via a serial line. You can instead specify @var{dev} as a TCP
12221 connection (for example, to a serial line attached via a terminal
12222 concentrator) using the syntax @code{@var{hostname}:@var{portnumber}}.
12224 The @code{load} and @code{attach} commands are @emph{not} defined for
12225 this target; you must load your program into the ST2000 as you normally
12226 would for standalone operation. @value{GDBN} reads debugging information
12227 (such as symbols) from a separate, debugging version of the program
12228 available on your host computer.
12229 @c FIXME!! This is terribly vague; what little content is here is
12230 @c basically hearsay.
12232 @cindex ST2000 auxiliary commands
12233 These auxiliary @value{GDBN} commands are available to help you with the ST2000
12237 @item st2000 @var{command}
12238 @kindex st2000 @var{cmd}
12239 @cindex STDBUG commands (ST2000)
12240 @cindex commands to STDBUG (ST2000)
12241 Send a @var{command} to the STDBUG monitor. See the manufacturer's
12242 manual for available commands.
12245 @cindex connect (to STDBUG)
12246 Connect the controlling terminal to the STDBUG command monitor. When
12247 you are done interacting with STDBUG, typing either of two character
12248 sequences gets you back to the @value{GDBN} command prompt:
12249 @kbd{@key{RET}~.} (Return, followed by tilde and period) or
12250 @kbd{@key{RET}~@key{C-d}} (Return, followed by tilde and control-D).
12254 @subsection Zilog Z8000
12257 @cindex simulator, Z8000
12258 @cindex Zilog Z8000 simulator
12260 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
12263 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
12264 unsegmented variant of the Z8000 architecture) or the Z8001 (the
12265 segmented variant). The simulator recognizes which architecture is
12266 appropriate by inspecting the object code.
12269 @item target sim @var{args}
12271 @kindex target sim@r{, with Z8000}
12272 Debug programs on a simulated CPU. If the simulator supports setup
12273 options, specify them via @var{args}.
12277 After specifying this target, you can debug programs for the simulated
12278 CPU in the same style as programs for your host computer; use the
12279 @code{file} command to load a new program image, the @code{run} command
12280 to run your program, and so on.
12282 As well as making available all the usual machine registers
12283 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
12284 additional items of information as specially named registers:
12289 Counts clock-ticks in the simulator.
12292 Counts instructions run in the simulator.
12295 Execution time in 60ths of a second.
12299 You can refer to these values in @value{GDBN} expressions with the usual
12300 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
12301 conditional breakpoint that suspends only after at least 5000
12302 simulated clock ticks.
12304 @node Architectures
12305 @section Architectures
12307 This section describes characteristics of architectures that affect
12308 all uses of @value{GDBN} with the architecture, both native and cross.
12321 @kindex set rstack_high_address
12322 @cindex AMD 29K register stack
12323 @cindex register stack, AMD29K
12324 @item set rstack_high_address @var{address}
12325 On AMD 29000 family processors, registers are saved in a separate
12326 @dfn{register stack}. There is no way for @value{GDBN} to determine the
12327 extent of this stack. Normally, @value{GDBN} just assumes that the
12328 stack is ``large enough''. This may result in @value{GDBN} referencing
12329 memory locations that do not exist. If necessary, you can get around
12330 this problem by specifying the ending address of the register stack with
12331 the @code{set rstack_high_address} command. The argument should be an
12332 address, which you probably want to precede with @samp{0x} to specify in
12335 @kindex show rstack_high_address
12336 @item show rstack_high_address
12337 Display the current limit of the register stack, on AMD 29000 family
12345 See the following section.
12350 @cindex stack on Alpha
12351 @cindex stack on MIPS
12352 @cindex Alpha stack
12354 Alpha- and MIPS-based computers use an unusual stack frame, which
12355 sometimes requires @value{GDBN} to search backward in the object code to
12356 find the beginning of a function.
12358 @cindex response time, MIPS debugging
12359 To improve response time (especially for embedded applications, where
12360 @value{GDBN} may be restricted to a slow serial line for this search)
12361 you may want to limit the size of this search, using one of these
12365 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
12366 @item set heuristic-fence-post @var{limit}
12367 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
12368 search for the beginning of a function. A value of @var{0} (the
12369 default) means there is no limit. However, except for @var{0}, the
12370 larger the limit the more bytes @code{heuristic-fence-post} must search
12371 and therefore the longer it takes to run.
12373 @item show heuristic-fence-post
12374 Display the current limit.
12378 These commands are available @emph{only} when @value{GDBN} is configured
12379 for debugging programs on Alpha or MIPS processors.
12382 @node Controlling GDB
12383 @chapter Controlling @value{GDBN}
12385 You can alter the way @value{GDBN} interacts with you by using the
12386 @code{set} command. For commands controlling how @value{GDBN} displays
12387 data, see @ref{Print Settings, ,Print settings}. Other settings are
12392 * Editing:: Command editing
12393 * History:: Command history
12394 * Screen Size:: Screen size
12395 * Numbers:: Numbers
12396 * ABI:: Configuring the current ABI
12397 * Messages/Warnings:: Optional warnings and messages
12398 * Debugging Output:: Optional messages about internal happenings
12406 @value{GDBN} indicates its readiness to read a command by printing a string
12407 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
12408 can change the prompt string with the @code{set prompt} command. For
12409 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
12410 the prompt in one of the @value{GDBN} sessions so that you can always tell
12411 which one you are talking to.
12413 @emph{Note:} @code{set prompt} does not add a space for you after the
12414 prompt you set. This allows you to set a prompt which ends in a space
12415 or a prompt that does not.
12419 @item set prompt @var{newprompt}
12420 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
12422 @kindex show prompt
12424 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
12428 @section Command editing
12430 @cindex command line editing
12432 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
12433 @sc{gnu} library provides consistent behavior for programs which provide a
12434 command line interface to the user. Advantages are @sc{gnu} Emacs-style
12435 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
12436 substitution, and a storage and recall of command history across
12437 debugging sessions.
12439 You may control the behavior of command line editing in @value{GDBN} with the
12440 command @code{set}.
12443 @kindex set editing
12446 @itemx set editing on
12447 Enable command line editing (enabled by default).
12449 @item set editing off
12450 Disable command line editing.
12452 @kindex show editing
12454 Show whether command line editing is enabled.
12458 @section Command history
12460 @value{GDBN} can keep track of the commands you type during your
12461 debugging sessions, so that you can be certain of precisely what
12462 happened. Use these commands to manage the @value{GDBN} command
12466 @cindex history substitution
12467 @cindex history file
12468 @kindex set history filename
12469 @kindex GDBHISTFILE
12470 @item set history filename @var{fname}
12471 Set the name of the @value{GDBN} command history file to @var{fname}.
12472 This is the file where @value{GDBN} reads an initial command history
12473 list, and where it writes the command history from this session when it
12474 exits. You can access this list through history expansion or through
12475 the history command editing characters listed below. This file defaults
12476 to the value of the environment variable @code{GDBHISTFILE}, or to
12477 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
12480 @cindex history save
12481 @kindex set history save
12482 @item set history save
12483 @itemx set history save on
12484 Record command history in a file, whose name may be specified with the
12485 @code{set history filename} command. By default, this option is disabled.
12487 @item set history save off
12488 Stop recording command history in a file.
12490 @cindex history size
12491 @kindex set history size
12492 @item set history size @var{size}
12493 Set the number of commands which @value{GDBN} keeps in its history list.
12494 This defaults to the value of the environment variable
12495 @code{HISTSIZE}, or to 256 if this variable is not set.
12498 @cindex history expansion
12499 History expansion assigns special meaning to the character @kbd{!}.
12500 @ifset have-readline-appendices
12501 @xref{Event Designators}.
12504 Since @kbd{!} is also the logical not operator in C, history expansion
12505 is off by default. If you decide to enable history expansion with the
12506 @code{set history expansion on} command, you may sometimes need to
12507 follow @kbd{!} (when it is used as logical not, in an expression) with
12508 a space or a tab to prevent it from being expanded. The readline
12509 history facilities do not attempt substitution on the strings
12510 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
12512 The commands to control history expansion are:
12515 @kindex set history expansion
12516 @item set history expansion on
12517 @itemx set history expansion
12518 Enable history expansion. History expansion is off by default.
12520 @item set history expansion off
12521 Disable history expansion.
12523 The readline code comes with more complete documentation of
12524 editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs
12525 or @code{vi} may wish to read it.
12526 @ifset have-readline-appendices
12527 @xref{Command Line Editing}.
12531 @kindex show history
12533 @itemx show history filename
12534 @itemx show history save
12535 @itemx show history size
12536 @itemx show history expansion
12537 These commands display the state of the @value{GDBN} history parameters.
12538 @code{show history} by itself displays all four states.
12544 @item show commands
12545 Display the last ten commands in the command history.
12547 @item show commands @var{n}
12548 Print ten commands centered on command number @var{n}.
12550 @item show commands +
12551 Print ten commands just after the commands last printed.
12555 @section Screen size
12556 @cindex size of screen
12557 @cindex pauses in output
12559 Certain commands to @value{GDBN} may produce large amounts of
12560 information output to the screen. To help you read all of it,
12561 @value{GDBN} pauses and asks you for input at the end of each page of
12562 output. Type @key{RET} when you want to continue the output, or @kbd{q}
12563 to discard the remaining output. Also, the screen width setting
12564 determines when to wrap lines of output. Depending on what is being
12565 printed, @value{GDBN} tries to break the line at a readable place,
12566 rather than simply letting it overflow onto the following line.
12568 Normally @value{GDBN} knows the size of the screen from the terminal
12569 driver software. For example, on Unix @value{GDBN} uses the termcap data base
12570 together with the value of the @code{TERM} environment variable and the
12571 @code{stty rows} and @code{stty cols} settings. If this is not correct,
12572 you can override it with the @code{set height} and @code{set
12579 @kindex show height
12580 @item set height @var{lpp}
12582 @itemx set width @var{cpl}
12584 These @code{set} commands specify a screen height of @var{lpp} lines and
12585 a screen width of @var{cpl} characters. The associated @code{show}
12586 commands display the current settings.
12588 If you specify a height of zero lines, @value{GDBN} does not pause during
12589 output no matter how long the output is. This is useful if output is to a
12590 file or to an editor buffer.
12592 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
12593 from wrapping its output.
12598 @cindex number representation
12599 @cindex entering numbers
12601 You can always enter numbers in octal, decimal, or hexadecimal in
12602 @value{GDBN} by the usual conventions: octal numbers begin with
12603 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
12604 begin with @samp{0x}. Numbers that begin with none of these are, by
12605 default, entered in base 10; likewise, the default display for
12606 numbers---when no particular format is specified---is base 10. You can
12607 change the default base for both input and output with the @code{set
12611 @kindex set input-radix
12612 @item set input-radix @var{base}
12613 Set the default base for numeric input. Supported choices
12614 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
12615 specified either unambiguously or using the current default radix; for
12625 sets the base to decimal. On the other hand, @samp{set radix 10}
12626 leaves the radix unchanged no matter what it was.
12628 @kindex set output-radix
12629 @item set output-radix @var{base}
12630 Set the default base for numeric display. Supported choices
12631 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
12632 specified either unambiguously or using the current default radix.
12634 @kindex show input-radix
12635 @item show input-radix
12636 Display the current default base for numeric input.
12638 @kindex show output-radix
12639 @item show output-radix
12640 Display the current default base for numeric display.
12644 @section Configuring the current ABI
12646 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
12647 application automatically. However, sometimes you need to override its
12648 conclusions. Use these commands to manage @value{GDBN}'s view of the
12655 One @value{GDBN} configuration can debug binaries for multiple operating
12656 system targets, either via remote debugging or native emulation.
12657 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
12658 but you can override its conclusion using the @code{set osabi} command.
12659 One example where this is useful is in debugging of binaries which use
12660 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
12661 not have the same identifying marks that the standard C library for your
12666 Show the OS ABI currently in use.
12669 With no argument, show the list of registered available OS ABI's.
12671 @item set osabi @var{abi}
12672 Set the current OS ABI to @var{abi}.
12675 @cindex float promotion
12676 @kindex set coerce-float-to-double
12678 Generally, the way that an argument of type @code{float} is passed to a
12679 function depends on whether the function is prototyped. For a prototyped
12680 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
12681 according to the architecture's convention for @code{float}. For unprototyped
12682 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
12683 @code{double} and then passed.
12685 Unfortunately, some forms of debug information do not reliably indicate whether
12686 a function is prototyped. If @value{GDBN} calls a function that is not marked
12687 as prototyped, it consults @kbd{set coerce-float-to-double}.
12690 @item set coerce-float-to-double
12691 @itemx set coerce-float-to-double on
12692 Arguments of type @code{float} will be promoted to @code{double} when passed
12693 to an unprototyped function. This is the default setting.
12695 @item set coerce-float-to-double off
12696 Arguments of type @code{float} will be passed directly to unprototyped
12700 @node Messages/Warnings
12701 @section Optional warnings and messages
12703 By default, @value{GDBN} is silent about its inner workings. If you are
12704 running on a slow machine, you may want to use the @code{set verbose}
12705 command. This makes @value{GDBN} tell you when it does a lengthy
12706 internal operation, so you will not think it has crashed.
12708 Currently, the messages controlled by @code{set verbose} are those
12709 which announce that the symbol table for a source file is being read;
12710 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
12713 @kindex set verbose
12714 @item set verbose on
12715 Enables @value{GDBN} output of certain informational messages.
12717 @item set verbose off
12718 Disables @value{GDBN} output of certain informational messages.
12720 @kindex show verbose
12722 Displays whether @code{set verbose} is on or off.
12725 By default, if @value{GDBN} encounters bugs in the symbol table of an
12726 object file, it is silent; but if you are debugging a compiler, you may
12727 find this information useful (@pxref{Symbol Errors, ,Errors reading
12732 @kindex set complaints
12733 @item set complaints @var{limit}
12734 Permits @value{GDBN} to output @var{limit} complaints about each type of
12735 unusual symbols before becoming silent about the problem. Set
12736 @var{limit} to zero to suppress all complaints; set it to a large number
12737 to prevent complaints from being suppressed.
12739 @kindex show complaints
12740 @item show complaints
12741 Displays how many symbol complaints @value{GDBN} is permitted to produce.
12745 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
12746 lot of stupid questions to confirm certain commands. For example, if
12747 you try to run a program which is already running:
12751 The program being debugged has been started already.
12752 Start it from the beginning? (y or n)
12755 If you are willing to unflinchingly face the consequences of your own
12756 commands, you can disable this ``feature'':
12760 @kindex set confirm
12762 @cindex confirmation
12763 @cindex stupid questions
12764 @item set confirm off
12765 Disables confirmation requests.
12767 @item set confirm on
12768 Enables confirmation requests (the default).
12770 @kindex show confirm
12772 Displays state of confirmation requests.
12776 @node Debugging Output
12777 @section Optional messages about internal happenings
12779 @kindex set debug arch
12780 @item set debug arch
12781 Turns on or off display of gdbarch debugging info. The default is off
12782 @kindex show debug arch
12783 @item show debug arch
12784 Displays the current state of displaying gdbarch debugging info.
12785 @kindex set debug event
12786 @item set debug event
12787 Turns on or off display of @value{GDBN} event debugging info. The
12789 @kindex show debug event
12790 @item show debug event
12791 Displays the current state of displaying @value{GDBN} event debugging
12793 @kindex set debug expression
12794 @item set debug expression
12795 Turns on or off display of @value{GDBN} expression debugging info. The
12797 @kindex show debug expression
12798 @item show debug expression
12799 Displays the current state of displaying @value{GDBN} expression
12801 @kindex set debug overload
12802 @item set debug overload
12803 Turns on or off display of @value{GDBN} C@t{++} overload debugging
12804 info. This includes info such as ranking of functions, etc. The default
12806 @kindex show debug overload
12807 @item show debug overload
12808 Displays the current state of displaying @value{GDBN} C@t{++} overload
12810 @kindex set debug remote
12811 @cindex packets, reporting on stdout
12812 @cindex serial connections, debugging
12813 @item set debug remote
12814 Turns on or off display of reports on all packets sent back and forth across
12815 the serial line to the remote machine. The info is printed on the
12816 @value{GDBN} standard output stream. The default is off.
12817 @kindex show debug remote
12818 @item show debug remote
12819 Displays the state of display of remote packets.
12820 @kindex set debug serial
12821 @item set debug serial
12822 Turns on or off display of @value{GDBN} serial debugging info. The
12824 @kindex show debug serial
12825 @item show debug serial
12826 Displays the current state of displaying @value{GDBN} serial debugging
12828 @kindex set debug target
12829 @item set debug target
12830 Turns on or off display of @value{GDBN} target debugging info. This info
12831 includes what is going on at the target level of GDB, as it happens. The
12833 @kindex show debug target
12834 @item show debug target
12835 Displays the current state of displaying @value{GDBN} target debugging
12837 @kindex set debug varobj
12838 @item set debug varobj
12839 Turns on or off display of @value{GDBN} variable object debugging
12840 info. The default is off.
12841 @kindex show debug varobj
12842 @item show debug varobj
12843 Displays the current state of displaying @value{GDBN} variable object
12848 @chapter Canned Sequences of Commands
12850 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
12851 command lists}), @value{GDBN} provides two ways to store sequences of
12852 commands for execution as a unit: user-defined commands and command
12856 * Define:: User-defined commands
12857 * Hooks:: User-defined command hooks
12858 * Command Files:: Command files
12859 * Output:: Commands for controlled output
12863 @section User-defined commands
12865 @cindex user-defined command
12866 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
12867 which you assign a new name as a command. This is done with the
12868 @code{define} command. User commands may accept up to 10 arguments
12869 separated by whitespace. Arguments are accessed within the user command
12870 via @var{$arg0@dots{}$arg9}. A trivial example:
12874 print $arg0 + $arg1 + $arg2
12878 To execute the command use:
12885 This defines the command @code{adder}, which prints the sum of
12886 its three arguments. Note the arguments are text substitutions, so they may
12887 reference variables, use complex expressions, or even perform inferior
12893 @item define @var{commandname}
12894 Define a command named @var{commandname}. If there is already a command
12895 by that name, you are asked to confirm that you want to redefine it.
12897 The definition of the command is made up of other @value{GDBN} command lines,
12898 which are given following the @code{define} command. The end of these
12899 commands is marked by a line containing @code{end}.
12904 Takes a single argument, which is an expression to evaluate.
12905 It is followed by a series of commands that are executed
12906 only if the expression is true (nonzero).
12907 There can then optionally be a line @code{else}, followed
12908 by a series of commands that are only executed if the expression
12909 was false. The end of the list is marked by a line containing @code{end}.
12913 The syntax is similar to @code{if}: the command takes a single argument,
12914 which is an expression to evaluate, and must be followed by the commands to
12915 execute, one per line, terminated by an @code{end}.
12916 The commands are executed repeatedly as long as the expression
12920 @item document @var{commandname}
12921 Document the user-defined command @var{commandname}, so that it can be
12922 accessed by @code{help}. The command @var{commandname} must already be
12923 defined. This command reads lines of documentation just as @code{define}
12924 reads the lines of the command definition, ending with @code{end}.
12925 After the @code{document} command is finished, @code{help} on command
12926 @var{commandname} displays the documentation you have written.
12928 You may use the @code{document} command again to change the
12929 documentation of a command. Redefining the command with @code{define}
12930 does not change the documentation.
12932 @kindex help user-defined
12933 @item help user-defined
12934 List all user-defined commands, with the first line of the documentation
12939 @itemx show user @var{commandname}
12940 Display the @value{GDBN} commands used to define @var{commandname} (but
12941 not its documentation). If no @var{commandname} is given, display the
12942 definitions for all user-defined commands.
12944 @kindex show max-user-call-depth
12945 @kindex set max-user-call-depth
12946 @item show max-user-call-depth
12947 @itemx set max-user-call-depth
12948 The value of @code{max-user-call-depth} controls how many recursion
12949 levels are allowed in user-defined commands before GDB suspects an
12950 infinite recursion and aborts the command.
12954 When user-defined commands are executed, the
12955 commands of the definition are not printed. An error in any command
12956 stops execution of the user-defined command.
12958 If used interactively, commands that would ask for confirmation proceed
12959 without asking when used inside a user-defined command. Many @value{GDBN}
12960 commands that normally print messages to say what they are doing omit the
12961 messages when used in a user-defined command.
12964 @section User-defined command hooks
12965 @cindex command hooks
12966 @cindex hooks, for commands
12967 @cindex hooks, pre-command
12971 You may define @dfn{hooks}, which are a special kind of user-defined
12972 command. Whenever you run the command @samp{foo}, if the user-defined
12973 command @samp{hook-foo} exists, it is executed (with no arguments)
12974 before that command.
12976 @cindex hooks, post-command
12979 A hook may also be defined which is run after the command you executed.
12980 Whenever you run the command @samp{foo}, if the user-defined command
12981 @samp{hookpost-foo} exists, it is executed (with no arguments) after
12982 that command. Post-execution hooks may exist simultaneously with
12983 pre-execution hooks, for the same command.
12985 It is valid for a hook to call the command which it hooks. If this
12986 occurs, the hook is not re-executed, thereby avoiding infinte recursion.
12988 @c It would be nice if hookpost could be passed a parameter indicating
12989 @c if the command it hooks executed properly or not. FIXME!
12991 @kindex stop@r{, a pseudo-command}
12992 In addition, a pseudo-command, @samp{stop} exists. Defining
12993 (@samp{hook-stop}) makes the associated commands execute every time
12994 execution stops in your program: before breakpoint commands are run,
12995 displays are printed, or the stack frame is printed.
12997 For example, to ignore @code{SIGALRM} signals while
12998 single-stepping, but treat them normally during normal execution,
13003 handle SIGALRM nopass
13007 handle SIGALRM pass
13010 define hook-continue
13011 handle SIGLARM pass
13015 As a further example, to hook at the begining and end of the @code{echo}
13016 command, and to add extra text to the beginning and end of the message,
13024 define hookpost-echo
13028 (@value{GDBP}) echo Hello World
13029 <<<---Hello World--->>>
13034 You can define a hook for any single-word command in @value{GDBN}, but
13035 not for command aliases; you should define a hook for the basic command
13036 name, e.g. @code{backtrace} rather than @code{bt}.
13037 @c FIXME! So how does Joe User discover whether a command is an alias
13039 If an error occurs during the execution of your hook, execution of
13040 @value{GDBN} commands stops and @value{GDBN} issues a prompt
13041 (before the command that you actually typed had a chance to run).
13043 If you try to define a hook which does not match any known command, you
13044 get a warning from the @code{define} command.
13046 @node Command Files
13047 @section Command files
13049 @cindex command files
13050 A command file for @value{GDBN} is a file of lines that are @value{GDBN}
13051 commands. Comments (lines starting with @kbd{#}) may also be included.
13052 An empty line in a command file does nothing; it does not mean to repeat
13053 the last command, as it would from the terminal.
13056 @cindex @file{.gdbinit}
13057 @cindex @file{gdb.ini}
13058 When you start @value{GDBN}, it automatically executes commands from its
13059 @dfn{init files}, normally called @file{.gdbinit}@footnote{The DJGPP
13060 port of @value{GDBN} uses the name @file{gdb.ini} instead, due to the
13061 limitations of file names imposed by DOS filesystems.}.
13062 During startup, @value{GDBN} does the following:
13066 Reads the init file (if any) in your home directory@footnote{On
13067 DOS/Windows systems, the home directory is the one pointed to by the
13068 @code{HOME} environment variable.}.
13071 Processes command line options and operands.
13074 Reads the init file (if any) in the current working directory.
13077 Reads command files specified by the @samp{-x} option.
13080 The init file in your home directory can set options (such as @samp{set
13081 complaints}) that affect subsequent processing of command line options
13082 and operands. Init files are not executed if you use the @samp{-nx}
13083 option (@pxref{Mode Options, ,Choosing modes}).
13085 @cindex init file name
13086 On some configurations of @value{GDBN}, the init file is known by a
13087 different name (these are typically environments where a specialized
13088 form of @value{GDBN} may need to coexist with other forms, hence a
13089 different name for the specialized version's init file). These are the
13090 environments with special init file names:
13092 @cindex @file{.vxgdbinit}
13095 VxWorks (Wind River Systems real-time OS): @file{.vxgdbinit}
13097 @cindex @file{.os68gdbinit}
13099 OS68K (Enea Data Systems real-time OS): @file{.os68gdbinit}
13101 @cindex @file{.esgdbinit}
13103 ES-1800 (Ericsson Telecom AB M68000 emulator): @file{.esgdbinit}
13106 You can also request the execution of a command file with the
13107 @code{source} command:
13111 @item source @var{filename}
13112 Execute the command file @var{filename}.
13115 The lines in a command file are executed sequentially. They are not
13116 printed as they are executed. An error in any command terminates
13117 execution of the command file and control is returned to the console.
13119 Commands that would ask for confirmation if used interactively proceed
13120 without asking when used in a command file. Many @value{GDBN} commands that
13121 normally print messages to say what they are doing omit the messages
13122 when called from command files.
13124 @value{GDBN} also accepts command input from standard input. In this
13125 mode, normal output goes to standard output and error output goes to
13126 standard error. Errors in a command file supplied on standard input do
13127 not terminate execution of the command file --- execution continues with
13131 gdb < cmds > log 2>&1
13134 (The syntax above will vary depending on the shell used.) This example
13135 will execute commands from the file @file{cmds}. All output and errors
13136 would be directed to @file{log}.
13139 @section Commands for controlled output
13141 During the execution of a command file or a user-defined command, normal
13142 @value{GDBN} output is suppressed; the only output that appears is what is
13143 explicitly printed by the commands in the definition. This section
13144 describes three commands useful for generating exactly the output you
13149 @item echo @var{text}
13150 @c I do not consider backslash-space a standard C escape sequence
13151 @c because it is not in ANSI.
13152 Print @var{text}. Nonprinting characters can be included in
13153 @var{text} using C escape sequences, such as @samp{\n} to print a
13154 newline. @strong{No newline is printed unless you specify one.}
13155 In addition to the standard C escape sequences, a backslash followed
13156 by a space stands for a space. This is useful for displaying a
13157 string with spaces at the beginning or the end, since leading and
13158 trailing spaces are otherwise trimmed from all arguments.
13159 To print @samp{@w{ }and foo =@w{ }}, use the command
13160 @samp{echo \@w{ }and foo = \@w{ }}.
13162 A backslash at the end of @var{text} can be used, as in C, to continue
13163 the command onto subsequent lines. For example,
13166 echo This is some text\n\
13167 which is continued\n\
13168 onto several lines.\n
13171 produces the same output as
13174 echo This is some text\n
13175 echo which is continued\n
13176 echo onto several lines.\n
13180 @item output @var{expression}
13181 Print the value of @var{expression} and nothing but that value: no
13182 newlines, no @samp{$@var{nn} = }. The value is not entered in the
13183 value history either. @xref{Expressions, ,Expressions}, for more information
13186 @item output/@var{fmt} @var{expression}
13187 Print the value of @var{expression} in format @var{fmt}. You can use
13188 the same formats as for @code{print}. @xref{Output Formats,,Output
13189 formats}, for more information.
13192 @item printf @var{string}, @var{expressions}@dots{}
13193 Print the values of the @var{expressions} under the control of
13194 @var{string}. The @var{expressions} are separated by commas and may be
13195 either numbers or pointers. Their values are printed as specified by
13196 @var{string}, exactly as if your program were to execute the C
13198 @c FIXME: the above implies that at least all ANSI C formats are
13199 @c supported, but it isn't true: %E and %G don't work (or so it seems).
13200 @c Either this is a bug, or the manual should document what formats are
13204 printf (@var{string}, @var{expressions}@dots{});
13207 For example, you can print two values in hex like this:
13210 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
13213 The only backslash-escape sequences that you can use in the format
13214 string are the simple ones that consist of backslash followed by a
13219 @chapter Command Interpreters
13220 @cindex command interpreters
13222 @value{GDBN} supports multiple command interpreters, and some command
13223 infrastructure to allow users or user interface writers to switch
13224 between interpreters or run commands in other interpreters.
13226 @value{GDBN} currently supports two command interpreters, the console
13227 interpreter (sometimes called the command-line interpreter or @sc{cli})
13228 and the machine interface interpreter (or @sc{gdb/mi}). This manual
13229 describes both of these interfaces in great detail.
13231 By default, @value{GDBN} will start with the console interpreter.
13232 However, the user may choose to start @value{GDBN} with another
13233 interpreter by specifying the @option{-i} or @option{--interpreter}
13234 startup options. Defined interpreters include:
13238 @cindex console interpreter
13239 The traditional console or command-line interpreter. This is the most often
13240 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
13241 @value{GDBN} will use this interpreter.
13244 @cindex mi interpreter
13245 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
13246 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
13247 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
13251 @cindex mi2 interpreter
13252 The current @sc{gdb/mi} interface.
13255 @cindex mi1 interpreter
13256 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
13260 @cindex invoke another interpreter
13261 The interpreter being used by @value{GDBN} may not be dynamically
13262 switched at runtime. Although possible, this could lead to a very
13263 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
13264 enters the command "interpreter-set console" in a console view,
13265 @value{GDBN} would switch to using the console interpreter, rendering
13266 the IDE inoperable!
13268 @kindex interpreter-exec
13269 Although you may only choose a single interpreter at startup, you may execute
13270 commands in any interpreter from the current interpreter using the appropriate
13271 command. If you are running the console interpreter, simply use the
13272 @code{interpreter-exec} command:
13275 interpreter-exec mi "-data-list-register-names"
13278 @sc{gdb/mi} has a similar command, although it is only available in versions of
13279 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
13282 @chapter @value{GDBN} Text User Interface
13286 * TUI Overview:: TUI overview
13287 * TUI Keys:: TUI key bindings
13288 * TUI Single Key Mode:: TUI single key mode
13289 * TUI Commands:: TUI specific commands
13290 * TUI Configuration:: TUI configuration variables
13293 The @value{GDBN} Text User Interface, TUI in short,
13294 is a terminal interface which uses the @code{curses} library
13295 to show the source file, the assembly output, the program registers
13296 and @value{GDBN} commands in separate text windows.
13297 The TUI is available only when @value{GDBN} is configured
13298 with the @code{--enable-tui} configure option (@pxref{Configure Options}).
13301 @section TUI overview
13303 The TUI has two display modes that can be switched while
13308 A curses (or TUI) mode in which it displays several text
13309 windows on the terminal.
13312 A standard mode which corresponds to the @value{GDBN} configured without
13316 In the TUI mode, @value{GDBN} can display several text window
13321 This window is the @value{GDBN} command window with the @value{GDBN}
13322 prompt and the @value{GDBN} outputs. The @value{GDBN} input is still
13323 managed using readline but through the TUI. The @emph{command}
13324 window is always visible.
13327 The source window shows the source file of the program. The current
13328 line as well as active breakpoints are displayed in this window.
13331 The assembly window shows the disassembly output of the program.
13334 This window shows the processor registers. It detects when
13335 a register is changed and when this is the case, registers that have
13336 changed are highlighted.
13340 The source and assembly windows show the current program position
13341 by highlighting the current line and marking them with the @samp{>} marker.
13342 Breakpoints are also indicated with two markers. A first one
13343 indicates the breakpoint type:
13347 Breakpoint which was hit at least once.
13350 Breakpoint which was never hit.
13353 Hardware breakpoint which was hit at least once.
13356 Hardware breakpoint which was never hit.
13360 The second marker indicates whether the breakpoint is enabled or not:
13364 Breakpoint is enabled.
13367 Breakpoint is disabled.
13371 The source, assembly and register windows are attached to the thread
13372 and the frame position. They are updated when the current thread
13373 changes, when the frame changes or when the program counter changes.
13374 These three windows are arranged by the TUI according to several
13375 layouts. The layout defines which of these three windows are visible.
13376 The following layouts are available:
13386 source and assembly
13389 source and registers
13392 assembly and registers
13396 On top of the command window a status line gives various information
13397 concerning the current process begin debugged. The status line is
13398 updated when the information it shows changes. The following fields
13403 Indicates the current gdb target
13404 (@pxref{Targets, ,Specifying a Debugging Target}).
13407 Gives information about the current process or thread number.
13408 When no process is being debugged, this field is set to @code{No process}.
13411 Gives the current function name for the selected frame.
13412 The name is demangled if demangling is turned on (@pxref{Print Settings}).
13413 When there is no symbol corresponding to the current program counter
13414 the string @code{??} is displayed.
13417 Indicates the current line number for the selected frame.
13418 When the current line number is not known the string @code{??} is displayed.
13421 Indicates the current program counter address.
13426 @section TUI Key Bindings
13427 @cindex TUI key bindings
13429 The TUI installs several key bindings in the readline keymaps
13430 (@pxref{Command Line Editing}).
13431 They allow to leave or enter in the TUI mode or they operate
13432 directly on the TUI layout and windows. The TUI also provides
13433 a @emph{SingleKey} keymap which binds several keys directly to
13434 @value{GDBN} commands. The following key bindings
13435 are installed for both TUI mode and the @value{GDBN} standard mode.
13444 Enter or leave the TUI mode. When the TUI mode is left,
13445 the curses window management is left and @value{GDBN} operates using
13446 its standard mode writing on the terminal directly. When the TUI
13447 mode is entered, the control is given back to the curses windows.
13448 The screen is then refreshed.
13452 Use a TUI layout with only one window. The layout will
13453 either be @samp{source} or @samp{assembly}. When the TUI mode
13454 is not active, it will switch to the TUI mode.
13456 Think of this key binding as the Emacs @kbd{C-x 1} binding.
13460 Use a TUI layout with at least two windows. When the current
13461 layout shows already two windows, a next layout with two windows is used.
13462 When a new layout is chosen, one window will always be common to the
13463 previous layout and the new one.
13465 Think of it as the Emacs @kbd{C-x 2} binding.
13469 Use the TUI @emph{SingleKey} keymap that binds single key to gdb commands
13470 (@pxref{TUI Single Key Mode}).
13474 The following key bindings are handled only by the TUI mode:
13479 Scroll the active window one page up.
13483 Scroll the active window one page down.
13487 Scroll the active window one line up.
13491 Scroll the active window one line down.
13495 Scroll the active window one column left.
13499 Scroll the active window one column right.
13503 Refresh the screen.
13507 In the TUI mode, the arrow keys are used by the active window
13508 for scrolling. This means they are not available for readline. It is
13509 necessary to use other readline key bindings such as @key{C-p}, @key{C-n},
13510 @key{C-b} and @key{C-f}.
13512 @node TUI Single Key Mode
13513 @section TUI Single Key Mode
13514 @cindex TUI single key mode
13516 The TUI provides a @emph{SingleKey} mode in which it installs a particular
13517 key binding in the readline keymaps to connect single keys to
13521 @kindex c @r{(SingleKey TUI key)}
13525 @kindex d @r{(SingleKey TUI key)}
13529 @kindex f @r{(SingleKey TUI key)}
13533 @kindex n @r{(SingleKey TUI key)}
13537 @kindex q @r{(SingleKey TUI key)}
13539 exit the @emph{SingleKey} mode.
13541 @kindex r @r{(SingleKey TUI key)}
13545 @kindex s @r{(SingleKey TUI key)}
13549 @kindex u @r{(SingleKey TUI key)}
13553 @kindex v @r{(SingleKey TUI key)}
13557 @kindex w @r{(SingleKey TUI key)}
13563 Other keys temporarily switch to the @value{GDBN} command prompt.
13564 The key that was pressed is inserted in the editing buffer so that
13565 it is possible to type most @value{GDBN} commands without interaction
13566 with the TUI @emph{SingleKey} mode. Once the command is entered the TUI
13567 @emph{SingleKey} mode is restored. The only way to permanently leave
13568 this mode is by hitting @key{q} or @samp{@key{C-x} @key{s}}.
13572 @section TUI specific commands
13573 @cindex TUI commands
13575 The TUI has specific commands to control the text windows.
13576 These commands are always available, that is they do not depend on
13577 the current terminal mode in which @value{GDBN} runs. When @value{GDBN}
13578 is in the standard mode, using these commands will automatically switch
13584 List and give the size of all displayed windows.
13587 @kindex layout next
13588 Display the next layout.
13591 @kindex layout prev
13592 Display the previous layout.
13596 Display the source window only.
13600 Display the assembly window only.
13603 @kindex layout split
13604 Display the source and assembly window.
13607 @kindex layout regs
13608 Display the register window together with the source or assembly window.
13610 @item focus next | prev | src | asm | regs | split
13612 Set the focus to the named window.
13613 This command allows to change the active window so that scrolling keys
13614 can be affected to another window.
13618 Refresh the screen. This is similar to using @key{C-L} key.
13622 Update the source window and the current execution point.
13624 @item winheight @var{name} +@var{count}
13625 @itemx winheight @var{name} -@var{count}
13627 Change the height of the window @var{name} by @var{count}
13628 lines. Positive counts increase the height, while negative counts
13633 @node TUI Configuration
13634 @section TUI configuration variables
13635 @cindex TUI configuration variables
13637 The TUI has several configuration variables that control the
13638 appearance of windows on the terminal.
13641 @item set tui border-kind @var{kind}
13642 @kindex set tui border-kind
13643 Select the border appearance for the source, assembly and register windows.
13644 The possible values are the following:
13647 Use a space character to draw the border.
13650 Use ascii characters + - and | to draw the border.
13653 Use the Alternate Character Set to draw the border. The border is
13654 drawn using character line graphics if the terminal supports them.
13658 @item set tui active-border-mode @var{mode}
13659 @kindex set tui active-border-mode
13660 Select the attributes to display the border of the active window.
13661 The possible values are @code{normal}, @code{standout}, @code{reverse},
13662 @code{half}, @code{half-standout}, @code{bold} and @code{bold-standout}.
13664 @item set tui border-mode @var{mode}
13665 @kindex set tui border-mode
13666 Select the attributes to display the border of other windows.
13667 The @var{mode} can be one of the following:
13670 Use normal attributes to display the border.
13676 Use reverse video mode.
13679 Use half bright mode.
13681 @item half-standout
13682 Use half bright and standout mode.
13685 Use extra bright or bold mode.
13687 @item bold-standout
13688 Use extra bright or bold and standout mode.
13695 @chapter Using @value{GDBN} under @sc{gnu} Emacs
13698 @cindex @sc{gnu} Emacs
13699 A special interface allows you to use @sc{gnu} Emacs to view (and
13700 edit) the source files for the program you are debugging with
13703 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
13704 executable file you want to debug as an argument. This command starts
13705 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
13706 created Emacs buffer.
13707 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
13709 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
13714 All ``terminal'' input and output goes through the Emacs buffer.
13717 This applies both to @value{GDBN} commands and their output, and to the input
13718 and output done by the program you are debugging.
13720 This is useful because it means that you can copy the text of previous
13721 commands and input them again; you can even use parts of the output
13724 All the facilities of Emacs' Shell mode are available for interacting
13725 with your program. In particular, you can send signals the usual
13726 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
13731 @value{GDBN} displays source code through Emacs.
13734 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
13735 source file for that frame and puts an arrow (@samp{=>}) at the
13736 left margin of the current line. Emacs uses a separate buffer for
13737 source display, and splits the screen to show both your @value{GDBN} session
13740 Explicit @value{GDBN} @code{list} or search commands still produce output as
13741 usual, but you probably have no reason to use them from Emacs.
13744 @emph{Warning:} If the directory where your program resides is not your
13745 current directory, it can be easy to confuse Emacs about the location of
13746 the source files, in which case the auxiliary display buffer does not
13747 appear to show your source. @value{GDBN} can find programs by searching your
13748 environment's @code{PATH} variable, so the @value{GDBN} input and output
13749 session proceeds normally; but Emacs does not get enough information
13750 back from @value{GDBN} to locate the source files in this situation. To
13751 avoid this problem, either start @value{GDBN} mode from the directory where
13752 your program resides, or specify an absolute file name when prompted for the
13753 @kbd{M-x gdb} argument.
13755 A similar confusion can result if you use the @value{GDBN} @code{file} command to
13756 switch to debugging a program in some other location, from an existing
13757 @value{GDBN} buffer in Emacs.
13760 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
13761 you need to call @value{GDBN} by a different name (for example, if you keep
13762 several configurations around, with different names) you can set the
13763 Emacs variable @code{gdb-command-name}; for example,
13766 (setq gdb-command-name "mygdb")
13770 (preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or
13771 in your @file{.emacs} file) makes Emacs call the program named
13772 ``@code{mygdb}'' instead.
13774 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
13775 addition to the standard Shell mode commands:
13779 Describe the features of Emacs' @value{GDBN} Mode.
13782 Execute to another source line, like the @value{GDBN} @code{step} command; also
13783 update the display window to show the current file and location.
13786 Execute to next source line in this function, skipping all function
13787 calls, like the @value{GDBN} @code{next} command. Then update the display window
13788 to show the current file and location.
13791 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
13792 display window accordingly.
13794 @item M-x gdb-nexti
13795 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
13796 display window accordingly.
13799 Execute until exit from the selected stack frame, like the @value{GDBN}
13800 @code{finish} command.
13803 Continue execution of your program, like the @value{GDBN} @code{continue}
13806 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
13809 Go up the number of frames indicated by the numeric argument
13810 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
13811 like the @value{GDBN} @code{up} command.
13813 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
13816 Go down the number of frames indicated by the numeric argument, like the
13817 @value{GDBN} @code{down} command.
13819 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
13822 Read the number where the cursor is positioned, and insert it at the end
13823 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
13824 around an address that was displayed earlier, type @kbd{disassemble};
13825 then move the cursor to the address display, and pick up the
13826 argument for @code{disassemble} by typing @kbd{C-x &}.
13828 You can customize this further by defining elements of the list
13829 @code{gdb-print-command}; once it is defined, you can format or
13830 otherwise process numbers picked up by @kbd{C-x &} before they are
13831 inserted. A numeric argument to @kbd{C-x &} indicates that you
13832 wish special formatting, and also acts as an index to pick an element of the
13833 list. If the list element is a string, the number to be inserted is
13834 formatted using the Emacs function @code{format}; otherwise the number
13835 is passed as an argument to the corresponding list element.
13838 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
13839 tells @value{GDBN} to set a breakpoint on the source line point is on.
13841 If you accidentally delete the source-display buffer, an easy way to get
13842 it back is to type the command @code{f} in the @value{GDBN} buffer, to
13843 request a frame display; when you run under Emacs, this recreates
13844 the source buffer if necessary to show you the context of the current
13847 The source files displayed in Emacs are in ordinary Emacs buffers
13848 which are visiting the source files in the usual way. You can edit
13849 the files with these buffers if you wish; but keep in mind that @value{GDBN}
13850 communicates with Emacs in terms of line numbers. If you add or
13851 delete lines from the text, the line numbers that @value{GDBN} knows cease
13852 to correspond properly with the code.
13854 @c The following dropped because Epoch is nonstandard. Reactivate
13855 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
13857 @kindex Emacs Epoch environment
13861 Version 18 of @sc{gnu} Emacs has a built-in window system
13862 called the @code{epoch}
13863 environment. Users of this environment can use a new command,
13864 @code{inspect} which performs identically to @code{print} except that
13865 each value is printed in its own window.
13868 @include annotate.texi
13869 @include gdbmi.texinfo
13872 @chapter Reporting Bugs in @value{GDBN}
13873 @cindex bugs in @value{GDBN}
13874 @cindex reporting bugs in @value{GDBN}
13876 Your bug reports play an essential role in making @value{GDBN} reliable.
13878 Reporting a bug may help you by bringing a solution to your problem, or it
13879 may not. But in any case the principal function of a bug report is to help
13880 the entire community by making the next version of @value{GDBN} work better. Bug
13881 reports are your contribution to the maintenance of @value{GDBN}.
13883 In order for a bug report to serve its purpose, you must include the
13884 information that enables us to fix the bug.
13887 * Bug Criteria:: Have you found a bug?
13888 * Bug Reporting:: How to report bugs
13892 @section Have you found a bug?
13893 @cindex bug criteria
13895 If you are not sure whether you have found a bug, here are some guidelines:
13898 @cindex fatal signal
13899 @cindex debugger crash
13900 @cindex crash of debugger
13902 If the debugger gets a fatal signal, for any input whatever, that is a
13903 @value{GDBN} bug. Reliable debuggers never crash.
13905 @cindex error on valid input
13907 If @value{GDBN} produces an error message for valid input, that is a
13908 bug. (Note that if you're cross debugging, the problem may also be
13909 somewhere in the connection to the target.)
13911 @cindex invalid input
13913 If @value{GDBN} does not produce an error message for invalid input,
13914 that is a bug. However, you should note that your idea of
13915 ``invalid input'' might be our idea of ``an extension'' or ``support
13916 for traditional practice''.
13919 If you are an experienced user of debugging tools, your suggestions
13920 for improvement of @value{GDBN} are welcome in any case.
13923 @node Bug Reporting
13924 @section How to report bugs
13925 @cindex bug reports
13926 @cindex @value{GDBN} bugs, reporting
13928 A number of companies and individuals offer support for @sc{gnu} products.
13929 If you obtained @value{GDBN} from a support organization, we recommend you
13930 contact that organization first.
13932 You can find contact information for many support companies and
13933 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
13935 @c should add a web page ref...
13937 In any event, we also recommend that you submit bug reports for
13938 @value{GDBN}. The prefered method is to submit them directly using
13939 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
13940 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
13943 @strong{Do not send bug reports to @samp{info-gdb}, or to
13944 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
13945 not want to receive bug reports. Those that do have arranged to receive
13948 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
13949 serves as a repeater. The mailing list and the newsgroup carry exactly
13950 the same messages. Often people think of posting bug reports to the
13951 newsgroup instead of mailing them. This appears to work, but it has one
13952 problem which can be crucial: a newsgroup posting often lacks a mail
13953 path back to the sender. Thus, if we need to ask for more information,
13954 we may be unable to reach you. For this reason, it is better to send
13955 bug reports to the mailing list.
13957 The fundamental principle of reporting bugs usefully is this:
13958 @strong{report all the facts}. If you are not sure whether to state a
13959 fact or leave it out, state it!
13961 Often people omit facts because they think they know what causes the
13962 problem and assume that some details do not matter. Thus, you might
13963 assume that the name of the variable you use in an example does not matter.
13964 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
13965 stray memory reference which happens to fetch from the location where that
13966 name is stored in memory; perhaps, if the name were different, the contents
13967 of that location would fool the debugger into doing the right thing despite
13968 the bug. Play it safe and give a specific, complete example. That is the
13969 easiest thing for you to do, and the most helpful.
13971 Keep in mind that the purpose of a bug report is to enable us to fix the
13972 bug. It may be that the bug has been reported previously, but neither
13973 you nor we can know that unless your bug report is complete and
13976 Sometimes people give a few sketchy facts and ask, ``Does this ring a
13977 bell?'' Those bug reports are useless, and we urge everyone to
13978 @emph{refuse to respond to them} except to chide the sender to report
13981 To enable us to fix the bug, you should include all these things:
13985 The version of @value{GDBN}. @value{GDBN} announces it if you start
13986 with no arguments; you can also print it at any time using @code{show
13989 Without this, we will not know whether there is any point in looking for
13990 the bug in the current version of @value{GDBN}.
13993 The type of machine you are using, and the operating system name and
13997 What compiler (and its version) was used to compile @value{GDBN}---e.g.
13998 ``@value{GCC}--2.8.1''.
14001 What compiler (and its version) was used to compile the program you are
14002 debugging---e.g. ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
14003 C Compiler''. For GCC, you can say @code{gcc --version} to get this
14004 information; for other compilers, see the documentation for those
14008 The command arguments you gave the compiler to compile your example and
14009 observe the bug. For example, did you use @samp{-O}? To guarantee
14010 you will not omit something important, list them all. A copy of the
14011 Makefile (or the output from make) is sufficient.
14013 If we were to try to guess the arguments, we would probably guess wrong
14014 and then we might not encounter the bug.
14017 A complete input script, and all necessary source files, that will
14021 A description of what behavior you observe that you believe is
14022 incorrect. For example, ``It gets a fatal signal.''
14024 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
14025 will certainly notice it. But if the bug is incorrect output, we might
14026 not notice unless it is glaringly wrong. You might as well not give us
14027 a chance to make a mistake.
14029 Even if the problem you experience is a fatal signal, you should still
14030 say so explicitly. Suppose something strange is going on, such as, your
14031 copy of @value{GDBN} is out of synch, or you have encountered a bug in
14032 the C library on your system. (This has happened!) Your copy might
14033 crash and ours would not. If you told us to expect a crash, then when
14034 ours fails to crash, we would know that the bug was not happening for
14035 us. If you had not told us to expect a crash, then we would not be able
14036 to draw any conclusion from our observations.
14039 If you wish to suggest changes to the @value{GDBN} source, send us context
14040 diffs. If you even discuss something in the @value{GDBN} source, refer to
14041 it by context, not by line number.
14043 The line numbers in our development sources will not match those in your
14044 sources. Your line numbers would convey no useful information to us.
14048 Here are some things that are not necessary:
14052 A description of the envelope of the bug.
14054 Often people who encounter a bug spend a lot of time investigating
14055 which changes to the input file will make the bug go away and which
14056 changes will not affect it.
14058 This is often time consuming and not very useful, because the way we
14059 will find the bug is by running a single example under the debugger
14060 with breakpoints, not by pure deduction from a series of examples.
14061 We recommend that you save your time for something else.
14063 Of course, if you can find a simpler example to report @emph{instead}
14064 of the original one, that is a convenience for us. Errors in the
14065 output will be easier to spot, running under the debugger will take
14066 less time, and so on.
14068 However, simplification is not vital; if you do not want to do this,
14069 report the bug anyway and send us the entire test case you used.
14072 A patch for the bug.
14074 A patch for the bug does help us if it is a good one. But do not omit
14075 the necessary information, such as the test case, on the assumption that
14076 a patch is all we need. We might see problems with your patch and decide
14077 to fix the problem another way, or we might not understand it at all.
14079 Sometimes with a program as complicated as @value{GDBN} it is very hard to
14080 construct an example that will make the program follow a certain path
14081 through the code. If you do not send us the example, we will not be able
14082 to construct one, so we will not be able to verify that the bug is fixed.
14084 And if we cannot understand what bug you are trying to fix, or why your
14085 patch should be an improvement, we will not install it. A test case will
14086 help us to understand.
14089 A guess about what the bug is or what it depends on.
14091 Such guesses are usually wrong. Even we cannot guess right about such
14092 things without first using the debugger to find the facts.
14095 @c The readline documentation is distributed with the readline code
14096 @c and consists of the two following files:
14098 @c inc-hist.texinfo
14099 @c Use -I with makeinfo to point to the appropriate directory,
14100 @c environment var TEXINPUTS with TeX.
14101 @include rluser.texinfo
14102 @include inc-hist.texinfo
14105 @node Formatting Documentation
14106 @appendix Formatting Documentation
14108 @cindex @value{GDBN} reference card
14109 @cindex reference card
14110 The @value{GDBN} 4 release includes an already-formatted reference card, ready
14111 for printing with PostScript or Ghostscript, in the @file{gdb}
14112 subdirectory of the main source directory@footnote{In
14113 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
14114 release.}. If you can use PostScript or Ghostscript with your printer,
14115 you can print the reference card immediately with @file{refcard.ps}.
14117 The release also includes the source for the reference card. You
14118 can format it, using @TeX{}, by typing:
14124 The @value{GDBN} reference card is designed to print in @dfn{landscape}
14125 mode on US ``letter'' size paper;
14126 that is, on a sheet 11 inches wide by 8.5 inches
14127 high. You will need to specify this form of printing as an option to
14128 your @sc{dvi} output program.
14130 @cindex documentation
14132 All the documentation for @value{GDBN} comes as part of the machine-readable
14133 distribution. The documentation is written in Texinfo format, which is
14134 a documentation system that uses a single source file to produce both
14135 on-line information and a printed manual. You can use one of the Info
14136 formatting commands to create the on-line version of the documentation
14137 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
14139 @value{GDBN} includes an already formatted copy of the on-line Info
14140 version of this manual in the @file{gdb} subdirectory. The main Info
14141 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
14142 subordinate files matching @samp{gdb.info*} in the same directory. If
14143 necessary, you can print out these files, or read them with any editor;
14144 but they are easier to read using the @code{info} subsystem in @sc{gnu}
14145 Emacs or the standalone @code{info} program, available as part of the
14146 @sc{gnu} Texinfo distribution.
14148 If you want to format these Info files yourself, you need one of the
14149 Info formatting programs, such as @code{texinfo-format-buffer} or
14152 If you have @code{makeinfo} installed, and are in the top level
14153 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
14154 version @value{GDBVN}), you can make the Info file by typing:
14161 If you want to typeset and print copies of this manual, you need @TeX{},
14162 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
14163 Texinfo definitions file.
14165 @TeX{} is a typesetting program; it does not print files directly, but
14166 produces output files called @sc{dvi} files. To print a typeset
14167 document, you need a program to print @sc{dvi} files. If your system
14168 has @TeX{} installed, chances are it has such a program. The precise
14169 command to use depends on your system; @kbd{lpr -d} is common; another
14170 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
14171 require a file name without any extension or a @samp{.dvi} extension.
14173 @TeX{} also requires a macro definitions file called
14174 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
14175 written in Texinfo format. On its own, @TeX{} cannot either read or
14176 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
14177 and is located in the @file{gdb-@var{version-number}/texinfo}
14180 If you have @TeX{} and a @sc{dvi} printer program installed, you can
14181 typeset and print this manual. First switch to the the @file{gdb}
14182 subdirectory of the main source directory (for example, to
14183 @file{gdb-@value{GDBVN}/gdb}) and type:
14189 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
14191 @node Installing GDB
14192 @appendix Installing @value{GDBN}
14193 @cindex configuring @value{GDBN}
14194 @cindex installation
14195 @cindex configuring @value{GDBN}, and source tree subdirectories
14197 @value{GDBN} comes with a @code{configure} script that automates the process
14198 of preparing @value{GDBN} for installation; you can then use @code{make} to
14199 build the @code{gdb} program.
14201 @c irrelevant in info file; it's as current as the code it lives with.
14202 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
14203 look at the @file{README} file in the sources; we may have improved the
14204 installation procedures since publishing this manual.}
14207 The @value{GDBN} distribution includes all the source code you need for
14208 @value{GDBN} in a single directory, whose name is usually composed by
14209 appending the version number to @samp{gdb}.
14211 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
14212 @file{gdb-@value{GDBVN}} directory. That directory contains:
14215 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
14216 script for configuring @value{GDBN} and all its supporting libraries
14218 @item gdb-@value{GDBVN}/gdb
14219 the source specific to @value{GDBN} itself
14221 @item gdb-@value{GDBVN}/bfd
14222 source for the Binary File Descriptor library
14224 @item gdb-@value{GDBVN}/include
14225 @sc{gnu} include files
14227 @item gdb-@value{GDBVN}/libiberty
14228 source for the @samp{-liberty} free software library
14230 @item gdb-@value{GDBVN}/opcodes
14231 source for the library of opcode tables and disassemblers
14233 @item gdb-@value{GDBVN}/readline
14234 source for the @sc{gnu} command-line interface
14236 @item gdb-@value{GDBVN}/glob
14237 source for the @sc{gnu} filename pattern-matching subroutine
14239 @item gdb-@value{GDBVN}/mmalloc
14240 source for the @sc{gnu} memory-mapped malloc package
14243 The simplest way to configure and build @value{GDBN} is to run @code{configure}
14244 from the @file{gdb-@var{version-number}} source directory, which in
14245 this example is the @file{gdb-@value{GDBVN}} directory.
14247 First switch to the @file{gdb-@var{version-number}} source directory
14248 if you are not already in it; then run @code{configure}. Pass the
14249 identifier for the platform on which @value{GDBN} will run as an
14255 cd gdb-@value{GDBVN}
14256 ./configure @var{host}
14261 where @var{host} is an identifier such as @samp{sun4} or
14262 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
14263 (You can often leave off @var{host}; @code{configure} tries to guess the
14264 correct value by examining your system.)
14266 Running @samp{configure @var{host}} and then running @code{make} builds the
14267 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
14268 libraries, then @code{gdb} itself. The configured source files, and the
14269 binaries, are left in the corresponding source directories.
14272 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
14273 system does not recognize this automatically when you run a different
14274 shell, you may need to run @code{sh} on it explicitly:
14277 sh configure @var{host}
14280 If you run @code{configure} from a directory that contains source
14281 directories for multiple libraries or programs, such as the
14282 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
14283 creates configuration files for every directory level underneath (unless
14284 you tell it not to, with the @samp{--norecursion} option).
14286 You should run the @code{configure} script from the top directory in the
14287 source tree, the @file{gdb-@var{version-number}} directory. If you run
14288 @code{configure} from one of the subdirectories, you will configure only
14289 that subdirectory. That is usually not what you want. In particular,
14290 if you run the first @code{configure} from the @file{gdb} subdirectory
14291 of the @file{gdb-@var{version-number}} directory, you will omit the
14292 configuration of @file{bfd}, @file{readline}, and other sibling
14293 directories of the @file{gdb} subdirectory. This leads to build errors
14294 about missing include files such as @file{bfd/bfd.h}.
14296 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
14297 However, you should make sure that the shell on your path (named by
14298 the @samp{SHELL} environment variable) is publicly readable. Remember
14299 that @value{GDBN} uses the shell to start your program---some systems refuse to
14300 let @value{GDBN} debug child processes whose programs are not readable.
14303 * Separate Objdir:: Compiling @value{GDBN} in another directory
14304 * Config Names:: Specifying names for hosts and targets
14305 * Configure Options:: Summary of options for configure
14308 @node Separate Objdir
14309 @section Compiling @value{GDBN} in another directory
14311 If you want to run @value{GDBN} versions for several host or target machines,
14312 you need a different @code{gdb} compiled for each combination of
14313 host and target. @code{configure} is designed to make this easy by
14314 allowing you to generate each configuration in a separate subdirectory,
14315 rather than in the source directory. If your @code{make} program
14316 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
14317 @code{make} in each of these directories builds the @code{gdb}
14318 program specified there.
14320 To build @code{gdb} in a separate directory, run @code{configure}
14321 with the @samp{--srcdir} option to specify where to find the source.
14322 (You also need to specify a path to find @code{configure}
14323 itself from your working directory. If the path to @code{configure}
14324 would be the same as the argument to @samp{--srcdir}, you can leave out
14325 the @samp{--srcdir} option; it is assumed.)
14327 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
14328 separate directory for a Sun 4 like this:
14332 cd gdb-@value{GDBVN}
14335 ../gdb-@value{GDBVN}/configure sun4
14340 When @code{configure} builds a configuration using a remote source
14341 directory, it creates a tree for the binaries with the same structure
14342 (and using the same names) as the tree under the source directory. In
14343 the example, you'd find the Sun 4 library @file{libiberty.a} in the
14344 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
14345 @file{gdb-sun4/gdb}.
14347 Make sure that your path to the @file{configure} script has just one
14348 instance of @file{gdb} in it. If your path to @file{configure} looks
14349 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
14350 one subdirectory of @value{GDBN}, not the whole package. This leads to
14351 build errors about missing include files such as @file{bfd/bfd.h}.
14353 One popular reason to build several @value{GDBN} configurations in separate
14354 directories is to configure @value{GDBN} for cross-compiling (where
14355 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
14356 programs that run on another machine---the @dfn{target}).
14357 You specify a cross-debugging target by
14358 giving the @samp{--target=@var{target}} option to @code{configure}.
14360 When you run @code{make} to build a program or library, you must run
14361 it in a configured directory---whatever directory you were in when you
14362 called @code{configure} (or one of its subdirectories).
14364 The @code{Makefile} that @code{configure} generates in each source
14365 directory also runs recursively. If you type @code{make} in a source
14366 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
14367 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
14368 will build all the required libraries, and then build GDB.
14370 When you have multiple hosts or targets configured in separate
14371 directories, you can run @code{make} on them in parallel (for example,
14372 if they are NFS-mounted on each of the hosts); they will not interfere
14376 @section Specifying names for hosts and targets
14378 The specifications used for hosts and targets in the @code{configure}
14379 script are based on a three-part naming scheme, but some short predefined
14380 aliases are also supported. The full naming scheme encodes three pieces
14381 of information in the following pattern:
14384 @var{architecture}-@var{vendor}-@var{os}
14387 For example, you can use the alias @code{sun4} as a @var{host} argument,
14388 or as the value for @var{target} in a @code{--target=@var{target}}
14389 option. The equivalent full name is @samp{sparc-sun-sunos4}.
14391 The @code{configure} script accompanying @value{GDBN} does not provide
14392 any query facility to list all supported host and target names or
14393 aliases. @code{configure} calls the Bourne shell script
14394 @code{config.sub} to map abbreviations to full names; you can read the
14395 script, if you wish, or you can use it to test your guesses on
14396 abbreviations---for example:
14399 % sh config.sub i386-linux
14401 % sh config.sub alpha-linux
14402 alpha-unknown-linux-gnu
14403 % sh config.sub hp9k700
14405 % sh config.sub sun4
14406 sparc-sun-sunos4.1.1
14407 % sh config.sub sun3
14408 m68k-sun-sunos4.1.1
14409 % sh config.sub i986v
14410 Invalid configuration `i986v': machine `i986v' not recognized
14414 @code{config.sub} is also distributed in the @value{GDBN} source
14415 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
14417 @node Configure Options
14418 @section @code{configure} options
14420 Here is a summary of the @code{configure} options and arguments that
14421 are most often useful for building @value{GDBN}. @code{configure} also has
14422 several other options not listed here. @inforef{What Configure
14423 Does,,configure.info}, for a full explanation of @code{configure}.
14426 configure @r{[}--help@r{]}
14427 @r{[}--prefix=@var{dir}@r{]}
14428 @r{[}--exec-prefix=@var{dir}@r{]}
14429 @r{[}--srcdir=@var{dirname}@r{]}
14430 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
14431 @r{[}--target=@var{target}@r{]}
14436 You may introduce options with a single @samp{-} rather than
14437 @samp{--} if you prefer; but you may abbreviate option names if you use
14442 Display a quick summary of how to invoke @code{configure}.
14444 @item --prefix=@var{dir}
14445 Configure the source to install programs and files under directory
14448 @item --exec-prefix=@var{dir}
14449 Configure the source to install programs under directory
14452 @c avoid splitting the warning from the explanation:
14454 @item --srcdir=@var{dirname}
14455 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
14456 @code{make} that implements the @code{VPATH} feature.}@*
14457 Use this option to make configurations in directories separate from the
14458 @value{GDBN} source directories. Among other things, you can use this to
14459 build (or maintain) several configurations simultaneously, in separate
14460 directories. @code{configure} writes configuration specific files in
14461 the current directory, but arranges for them to use the source in the
14462 directory @var{dirname}. @code{configure} creates directories under
14463 the working directory in parallel to the source directories below
14466 @item --norecursion
14467 Configure only the directory level where @code{configure} is executed; do not
14468 propagate configuration to subdirectories.
14470 @item --target=@var{target}
14471 Configure @value{GDBN} for cross-debugging programs running on the specified
14472 @var{target}. Without this option, @value{GDBN} is configured to debug
14473 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
14475 There is no convenient way to generate a list of all available targets.
14477 @item @var{host} @dots{}
14478 Configure @value{GDBN} to run on the specified @var{host}.
14480 There is no convenient way to generate a list of all available hosts.
14483 There are many other options available as well, but they are generally
14484 needed for special purposes only.
14486 @node Maintenance Commands
14487 @appendix Maintenance Commands
14488 @cindex maintenance commands
14489 @cindex internal commands
14491 In addition to commands intended for @value{GDBN} users, @value{GDBN}
14492 includes a number of commands intended for @value{GDBN} developers.
14493 These commands are provided here for reference.
14496 @kindex maint info breakpoints
14497 @item @anchor{maint info breakpoints}maint info breakpoints
14498 Using the same format as @samp{info breakpoints}, display both the
14499 breakpoints you've set explicitly, and those @value{GDBN} is using for
14500 internal purposes. Internal breakpoints are shown with negative
14501 breakpoint numbers. The type column identifies what kind of breakpoint
14506 Normal, explicitly set breakpoint.
14509 Normal, explicitly set watchpoint.
14512 Internal breakpoint, used to handle correctly stepping through
14513 @code{longjmp} calls.
14515 @item longjmp resume
14516 Internal breakpoint at the target of a @code{longjmp}.
14519 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
14522 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
14525 Shared library events.
14529 @kindex maint internal-error
14530 @kindex maint internal-warning
14531 @item maint internal-error
14532 @itemx maint internal-warning
14533 Cause @value{GDBN} to call the internal function @code{internal_error}
14534 or @code{internal_warning} and hence behave as though an internal error
14535 or internal warning has been detected. In addition to reporting the
14536 internal problem, these functions give the user the opportunity to
14537 either quit @value{GDBN} or create a core file of the current
14538 @value{GDBN} session.
14541 (gdb) @kbd{maint internal-error testing, 1, 2}
14542 @dots{}/maint.c:121: internal-error: testing, 1, 2
14543 A problem internal to GDB has been detected. Further
14544 debugging may prove unreliable.
14545 Quit this debugging session? (y or n) @kbd{n}
14546 Create a core file? (y or n) @kbd{n}
14550 Takes an optional parameter that is used as the text of the error or
14553 @kindex maint print registers
14554 @kindex maint print raw-registers
14555 @kindex maint print cooked-registers
14556 @kindex maint print register-groups
14557 @item maint print registers
14558 @itemx maint print raw-registers
14559 @itemx maint print cooked-registers
14560 @itemx maint print register-groups
14561 Print @value{GDBN}'s internal register data structures.
14563 The command @code{maint print raw-registers} includes the contents of
14564 the raw register cache; the command @code{maint print cooked-registers}
14565 includes the (cooked) value of all registers; and the command
14566 @code{maint print register-groups} includes the groups that each
14567 register is a member of. @xref{Registers,, Registers, gdbint,
14568 @value{GDBN} Internals}.
14570 Takes an optional file parameter.
14572 @kindex maint print reggroups
14573 @item maint print reggroups
14574 Print @value{GDBN}'s internal register group data structures.
14576 Takes an optional file parameter.
14579 (gdb) @kbd{maint print reggroups}
14590 @kindex maint set profile
14591 @kindex maint show profile
14592 @cindex profiling GDB
14593 @item maint set profile
14594 @itemx maint show profile
14595 Control profiling of @value{GDBN}.
14597 Profiling will be disabled until you use the @samp{maint set profile}
14598 command to enable it. When you enable profiling, the system will begin
14599 collecting timing and execution count data; when you disable profiling or
14600 exit @value{GDBN}, the results will be written to a log file. Remember that
14601 if you use profiling, @value{GDBN} will overwrite the profiling log file
14602 (often called @file{gmon.out}). If you have a record of important profiling
14603 data in a @file{gmon.out} file, be sure to move it to a safe location.
14605 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
14606 compiled with the @samp{-pg} compiler option.
14611 @node Remote Protocol
14612 @appendix @value{GDBN} Remote Serial Protocol
14617 * Stop Reply Packets::
14618 * General Query Packets::
14619 * Register Packet Format::
14626 There may be occasions when you need to know something about the
14627 protocol---for example, if there is only one serial port to your target
14628 machine, you might want your program to do something special if it
14629 recognizes a packet meant for @value{GDBN}.
14631 In the examples below, @samp{->} and @samp{<-} are used to indicate
14632 transmitted and received data respectfully.
14634 @cindex protocol, @value{GDBN} remote serial
14635 @cindex serial protocol, @value{GDBN} remote
14636 @cindex remote serial protocol
14637 All @value{GDBN} commands and responses (other than acknowledgments) are
14638 sent as a @var{packet}. A @var{packet} is introduced with the character
14639 @samp{$}, the actual @var{packet-data}, and the terminating character
14640 @samp{#} followed by a two-digit @var{checksum}:
14643 @code{$}@var{packet-data}@code{#}@var{checksum}
14647 @cindex checksum, for @value{GDBN} remote
14649 The two-digit @var{checksum} is computed as the modulo 256 sum of all
14650 characters between the leading @samp{$} and the trailing @samp{#} (an
14651 eight bit unsigned checksum).
14653 Implementors should note that prior to @value{GDBN} 5.0 the protocol
14654 specification also included an optional two-digit @var{sequence-id}:
14657 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
14660 @cindex sequence-id, for @value{GDBN} remote
14662 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
14663 has never output @var{sequence-id}s. Stubs that handle packets added
14664 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
14666 @cindex acknowledgment, for @value{GDBN} remote
14667 When either the host or the target machine receives a packet, the first
14668 response expected is an acknowledgment: either @samp{+} (to indicate
14669 the package was received correctly) or @samp{-} (to request
14673 -> @code{$}@var{packet-data}@code{#}@var{checksum}
14678 The host (@value{GDBN}) sends @var{command}s, and the target (the
14679 debugging stub incorporated in your program) sends a @var{response}. In
14680 the case of step and continue @var{command}s, the response is only sent
14681 when the operation has completed (the target has again stopped).
14683 @var{packet-data} consists of a sequence of characters with the
14684 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
14687 Fields within the packet should be separated using @samp{,} @samp{;} or
14688 @cindex remote protocol, field separator
14689 @samp{:}. Except where otherwise noted all numbers are represented in
14690 @sc{hex} with leading zeros suppressed.
14692 Implementors should note that prior to @value{GDBN} 5.0, the character
14693 @samp{:} could not appear as the third character in a packet (as it
14694 would potentially conflict with the @var{sequence-id}).
14696 Response @var{data} can be run-length encoded to save space. A @samp{*}
14697 means that the next character is an @sc{ascii} encoding giving a repeat count
14698 which stands for that many repetitions of the character preceding the
14699 @samp{*}. The encoding is @code{n+29}, yielding a printable character
14700 where @code{n >=3} (which is where rle starts to win). The printable
14701 characters @samp{$}, @samp{#}, @samp{+} and @samp{-} or with a numeric
14702 value greater than 126 should not be used.
14704 Some remote systems have used a different run-length encoding mechanism
14705 loosely refered to as the cisco encoding. Following the @samp{*}
14706 character are two hex digits that indicate the size of the packet.
14713 means the same as "0000".
14715 The error response returned for some packets includes a two character
14716 error number. That number is not well defined.
14718 For any @var{command} not supported by the stub, an empty response
14719 (@samp{$#00}) should be returned. That way it is possible to extend the
14720 protocol. A newer @value{GDBN} can tell if a packet is supported based
14723 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
14724 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
14730 The following table provides a complete list of all currently defined
14731 @var{command}s and their corresponding response @var{data}.
14735 @item @code{!} --- extended mode
14736 @cindex @code{!} packet
14738 Enable extended mode. In extended mode, the remote server is made
14739 persistent. The @samp{R} packet is used to restart the program being
14745 The remote target both supports and has enabled extended mode.
14748 @item @code{?} --- last signal
14749 @cindex @code{?} packet
14751 Indicate the reason the target halted. The reply is the same as for
14755 @xref{Stop Reply Packets}, for the reply specifications.
14757 @item @code{a} --- reserved
14759 Reserved for future use.
14761 @item @code{A}@var{arglen}@code{,}@var{argnum}@code{,}@var{arg}@code{,@dots{}} --- set program arguments @strong{(reserved)}
14762 @cindex @code{A} packet
14764 Initialized @samp{argv[]} array passed into program. @var{arglen}
14765 specifies the number of bytes in the hex encoded byte stream @var{arg}.
14766 See @code{gdbserver} for more details.
14774 @item @code{b}@var{baud} --- set baud @strong{(deprecated)}
14775 @cindex @code{b} packet
14777 Change the serial line speed to @var{baud}.
14779 JTC: @emph{When does the transport layer state change? When it's
14780 received, or after the ACK is transmitted. In either case, there are
14781 problems if the command or the acknowledgment packet is dropped.}
14783 Stan: @emph{If people really wanted to add something like this, and get
14784 it working for the first time, they ought to modify ser-unix.c to send
14785 some kind of out-of-band message to a specially-setup stub and have the
14786 switch happen "in between" packets, so that from remote protocol's point
14787 of view, nothing actually happened.}
14789 @item @code{B}@var{addr},@var{mode} --- set breakpoint @strong{(deprecated)}
14790 @cindex @code{B} packet
14792 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
14793 breakpoint at @var{addr}.
14795 This packet has been replaced by the @samp{Z} and @samp{z} packets
14796 (@pxref{insert breakpoint or watchpoint packet}).
14798 @item @code{c}@var{addr} --- continue
14799 @cindex @code{c} packet
14801 @var{addr} is address to resume. If @var{addr} is omitted, resume at
14805 @xref{Stop Reply Packets}, for the reply specifications.
14807 @item @code{C}@var{sig}@code{;}@var{addr} --- continue with signal
14808 @cindex @code{C} packet
14810 Continue with signal @var{sig} (hex signal number). If
14811 @code{;}@var{addr} is omitted, resume at same address.
14814 @xref{Stop Reply Packets}, for the reply specifications.
14816 @item @code{d} --- toggle debug @strong{(deprecated)}
14817 @cindex @code{d} packet
14821 @item @code{D} --- detach
14822 @cindex @code{D} packet
14824 Detach @value{GDBN} from the remote system. Sent to the remote target
14825 before @value{GDBN} disconnects.
14829 @item @emph{no response}
14830 @value{GDBN} does not check for any response after sending this packet.
14833 @item @code{e} --- reserved
14835 Reserved for future use.
14837 @item @code{E} --- reserved
14839 Reserved for future use.
14841 @item @code{f} --- reserved
14843 Reserved for future use.
14845 @item @code{F} --- reserved
14847 Reserved for future use.
14849 @item @code{g} --- read registers
14850 @anchor{read registers packet}
14851 @cindex @code{g} packet
14853 Read general registers.
14857 @item @var{XX@dots{}}
14858 Each byte of register data is described by two hex digits. The bytes
14859 with the register are transmitted in target byte order. The size of
14860 each register and their position within the @samp{g} @var{packet} are
14861 determined by the @value{GDBN} internal macros @var{REGISTER_RAW_SIZE}
14862 and @var{REGISTER_NAME} macros. The specification of several standard
14863 @code{g} packets is specified below.
14868 @item @code{G}@var{XX@dots{}} --- write regs
14869 @cindex @code{G} packet
14871 @xref{read registers packet}, for a description of the @var{XX@dots{}}
14882 @item @code{h} --- reserved
14884 Reserved for future use.
14886 @item @code{H}@var{c}@var{t@dots{}} --- set thread
14887 @cindex @code{H} packet
14889 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
14890 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
14891 should be @samp{c} for step and continue operations, @samp{g} for other
14892 operations. The thread designator @var{t@dots{}} may be -1, meaning all
14893 the threads, a thread number, or zero which means pick any thread.
14904 @c 'H': How restrictive (or permissive) is the thread model. If a
14905 @c thread is selected and stopped, are other threads allowed
14906 @c to continue to execute? As I mentioned above, I think the
14907 @c semantics of each command when a thread is selected must be
14908 @c described. For example:
14910 @c 'g': If the stub supports threads and a specific thread is
14911 @c selected, returns the register block from that thread;
14912 @c otherwise returns current registers.
14914 @c 'G' If the stub supports threads and a specific thread is
14915 @c selected, sets the registers of the register block of
14916 @c that thread; otherwise sets current registers.
14918 @item @code{i}@var{addr}@code{,}@var{nnn} --- cycle step @strong{(draft)}
14919 @anchor{cycle step packet}
14920 @cindex @code{i} packet
14922 Step the remote target by a single clock cycle. If @code{,}@var{nnn} is
14923 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
14924 step starting at that address.
14926 @item @code{I} --- signal then cycle step @strong{(reserved)}
14927 @cindex @code{I} packet
14929 @xref{step with signal packet}. @xref{cycle step packet}.
14931 @item @code{j} --- reserved
14933 Reserved for future use.
14935 @item @code{J} --- reserved
14937 Reserved for future use.
14939 @item @code{k} --- kill request
14940 @cindex @code{k} packet
14942 FIXME: @emph{There is no description of how to operate when a specific
14943 thread context has been selected (i.e.@: does 'k' kill only that
14946 @item @code{K} --- reserved
14948 Reserved for future use.
14950 @item @code{l} --- reserved
14952 Reserved for future use.
14954 @item @code{L} --- reserved
14956 Reserved for future use.
14958 @item @code{m}@var{addr}@code{,}@var{length} --- read memory
14959 @cindex @code{m} packet
14961 Read @var{length} bytes of memory starting at address @var{addr}.
14962 Neither @value{GDBN} nor the stub assume that sized memory transfers are
14963 assumed using word aligned accesses. FIXME: @emph{A word aligned memory
14964 transfer mechanism is needed.}
14968 @item @var{XX@dots{}}
14969 @var{XX@dots{}} is mem contents. Can be fewer bytes than requested if able
14970 to read only part of the data. Neither @value{GDBN} nor the stub assume
14971 that sized memory transfers are assumed using word aligned
14972 accesses. FIXME: @emph{A word aligned memory transfer mechanism is
14978 @item @code{M}@var{addr},@var{length}@code{:}@var{XX@dots{}} --- write mem
14979 @cindex @code{M} packet
14981 Write @var{length} bytes of memory starting at address @var{addr}.
14982 @var{XX@dots{}} is the data.
14989 for an error (this includes the case where only part of the data was
14993 @item @code{n} --- reserved
14995 Reserved for future use.
14997 @item @code{N} --- reserved
14999 Reserved for future use.
15001 @item @code{o} --- reserved
15003 Reserved for future use.
15005 @item @code{O} --- reserved
15007 Reserved for future use.
15009 @item @code{p}@var{n@dots{}} --- read reg @strong{(reserved)}
15010 @cindex @code{p} packet
15012 @xref{write register packet}.
15016 @item @var{r@dots{}.}
15017 The hex encoded value of the register in target byte order.
15020 @item @code{P}@var{n@dots{}}@code{=}@var{r@dots{}} --- write register
15021 @anchor{write register packet}
15022 @cindex @code{P} packet
15024 Write register @var{n@dots{}} with value @var{r@dots{}}, which contains two hex
15025 digits for each byte in the register (target byte order).
15035 @item @code{q}@var{query} --- general query
15036 @anchor{general query packet}
15037 @cindex @code{q} packet
15039 Request info about @var{query}. In general @value{GDBN} queries have a
15040 leading upper case letter. Custom vendor queries should use a company
15041 prefix (in lower case) ex: @samp{qfsf.var}. @var{query} may optionally
15042 be followed by a @samp{,} or @samp{;} separated list. Stubs must ensure
15043 that they match the full @var{query} name.
15047 @item @var{XX@dots{}}
15048 Hex encoded data from query. The reply can not be empty.
15052 Indicating an unrecognized @var{query}.
15055 @item @code{Q}@var{var}@code{=}@var{val} --- general set
15056 @cindex @code{Q} packet
15058 Set value of @var{var} to @var{val}.
15060 @xref{general query packet}, for a discussion of naming conventions.
15062 @item @code{r} --- reset @strong{(deprecated)}
15063 @cindex @code{r} packet
15065 Reset the entire system.
15067 @item @code{R}@var{XX} --- remote restart
15068 @cindex @code{R} packet
15070 Restart the program being debugged. @var{XX}, while needed, is ignored.
15071 This packet is only available in extended mode.
15075 @item @emph{no reply}
15076 The @samp{R} packet has no reply.
15079 @item @code{s}@var{addr} --- step
15080 @cindex @code{s} packet
15082 @var{addr} is address to resume. If @var{addr} is omitted, resume at
15086 @xref{Stop Reply Packets}, for the reply specifications.
15088 @item @code{S}@var{sig}@code{;}@var{addr} --- step with signal
15089 @anchor{step with signal packet}
15090 @cindex @code{S} packet
15092 Like @samp{C} but step not continue.
15095 @xref{Stop Reply Packets}, for the reply specifications.
15097 @item @code{t}@var{addr}@code{:}@var{PP}@code{,}@var{MM} --- search
15098 @cindex @code{t} packet
15100 Search backwards starting at address @var{addr} for a match with pattern
15101 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
15102 @var{addr} must be at least 3 digits.
15104 @item @code{T}@var{XX} --- thread alive
15105 @cindex @code{T} packet
15107 Find out if the thread XX is alive.
15112 thread is still alive
15117 @item @code{u} --- reserved
15119 Reserved for future use.
15121 @item @code{U} --- reserved
15123 Reserved for future use.
15125 @item @code{v} --- reserved
15127 Reserved for future use.
15129 @item @code{V} --- reserved
15131 Reserved for future use.
15133 @item @code{w} --- reserved
15135 Reserved for future use.
15137 @item @code{W} --- reserved
15139 Reserved for future use.
15141 @item @code{x} --- reserved
15143 Reserved for future use.
15145 @item @code{X}@var{addr}@code{,}@var{length}@var{:}@var{XX@dots{}} --- write mem (binary)
15146 @cindex @code{X} packet
15148 @var{addr} is address, @var{length} is number of bytes, @var{XX@dots{}}
15149 is binary data. The characters @code{$}, @code{#}, and @code{0x7d} are
15150 escaped using @code{0x7d}.
15160 @item @code{y} --- reserved
15162 Reserved for future use.
15164 @item @code{Y} reserved
15166 Reserved for future use.
15168 @item @code{z}@var{type}@code{,}@var{addr}@code{,}@var{length} --- remove breakpoint or watchpoint @strong{(draft)}
15169 @itemx @code{Z}@var{type}@code{,}@var{addr}@code{,}@var{length} --- insert breakpoint or watchpoint @strong{(draft)}
15170 @anchor{insert breakpoint or watchpoint packet}
15171 @cindex @code{z} packet
15172 @cindex @code{Z} packets
15174 Insert (@code{Z}) or remove (@code{z}) a @var{type} breakpoint or
15175 watchpoint starting at address @var{address} and covering the next
15176 @var{length} bytes.
15178 Each breakpoint and watchpoint packet @var{type} is documented
15181 @emph{Implementation notes: A remote target shall return an empty string
15182 for an unrecognized breakpoint or watchpoint packet @var{type}. A
15183 remote target shall support either both or neither of a given
15184 @code{Z}@var{type}@dots{} and @code{z}@var{type}@dots{} packet pair. To
15185 avoid potential problems with duplicate packets, the operations should
15186 be implemented in an idempotent way.}
15188 @item @code{z}@code{0}@code{,}@var{addr}@code{,}@var{length} --- remove memory breakpoint @strong{(draft)}
15189 @item @code{Z}@code{0}@code{,}@var{addr}@code{,}@var{length} --- insert memory breakpoint @strong{(draft)}
15190 @cindex @code{z0} packet
15191 @cindex @code{Z0} packet
15193 Insert (@code{Z0}) or remove (@code{z0}) a memory breakpoint at address
15194 @code{addr} of size @code{length}.
15196 A memory breakpoint is implemented by replacing the instruction at
15197 @var{addr} with a software breakpoint or trap instruction. The
15198 @code{length} is used by targets that indicates the size of the
15199 breakpoint (in bytes) that should be inserted (e.g., the @sc{arm} and
15200 @sc{mips} can insert either a 2 or 4 byte breakpoint).
15202 @emph{Implementation note: It is possible for a target to copy or move
15203 code that contains memory breakpoints (e.g., when implementing
15204 overlays). The behavior of this packet, in the presence of such a
15205 target, is not defined.}
15217 @item @code{z}@code{1}@code{,}@var{addr}@code{,}@var{length} --- remove hardware breakpoint @strong{(draft)}
15218 @item @code{Z}@code{1}@code{,}@var{addr}@code{,}@var{length} --- insert hardware breakpoint @strong{(draft)}
15219 @cindex @code{z1} packet
15220 @cindex @code{Z1} packet
15222 Insert (@code{Z1}) or remove (@code{z1}) a hardware breakpoint at
15223 address @code{addr} of size @code{length}.
15225 A hardware breakpoint is implemented using a mechanism that is not
15226 dependant on being able to modify the target's memory.
15228 @emph{Implementation note: A hardware breakpoint is not affected by code
15241 @item @code{z}@code{2}@code{,}@var{addr}@code{,}@var{length} --- remove write watchpoint @strong{(draft)}
15242 @item @code{Z}@code{2}@code{,}@var{addr}@code{,}@var{length} --- insert write watchpoint @strong{(draft)}
15243 @cindex @code{z2} packet
15244 @cindex @code{Z2} packet
15246 Insert (@code{Z2}) or remove (@code{z2}) a write watchpoint.
15258 @item @code{z}@code{3}@code{,}@var{addr}@code{,}@var{length} --- remove read watchpoint @strong{(draft)}
15259 @item @code{Z}@code{3}@code{,}@var{addr}@code{,}@var{length} --- insert read watchpoint @strong{(draft)}
15260 @cindex @code{z3} packet
15261 @cindex @code{Z3} packet
15263 Insert (@code{Z3}) or remove (@code{z3}) a read watchpoint.
15275 @item @code{z}@code{4}@code{,}@var{addr}@code{,}@var{length} --- remove access watchpoint @strong{(draft)}
15276 @item @code{Z}@code{4}@code{,}@var{addr}@code{,}@var{length} --- insert access watchpoint @strong{(draft)}
15277 @cindex @code{z4} packet
15278 @cindex @code{Z4} packet
15280 Insert (@code{Z4}) or remove (@code{z4}) an access watchpoint.
15294 @node Stop Reply Packets
15295 @section Stop Reply Packets
15296 @cindex stop reply packets
15298 The @samp{C}, @samp{c}, @samp{S}, @samp{s} and @samp{?} packets can
15299 receive any of the below as a reply. In the case of the @samp{C},
15300 @samp{c}, @samp{S} and @samp{s} packets, that reply is only returned
15301 when the target halts. In the below the exact meaning of @samp{signal
15302 number} is poorly defined. In general one of the UNIX signal numbering
15303 conventions is used.
15308 @var{AA} is the signal number
15310 @item @code{T}@var{AA}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}@var{n...}@code{:}@var{r...}@code{;}
15311 @cindex @code{T} packet reply
15313 @var{AA} = two hex digit signal number; @var{n...} = register number
15314 (hex), @var{r...} = target byte ordered register contents, size defined
15315 by @code{REGISTER_RAW_SIZE}; @var{n...} = @samp{thread}, @var{r...} =
15316 thread process ID, this is a hex integer; @var{n...} = (@samp{watch} |
15317 @samp{rwatch} | @samp{awatch}, @var{r...} = data address, this is a hex
15318 integer; @var{n...} = other string not starting with valid hex digit.
15319 @value{GDBN} should ignore this @var{n...}, @var{r...} pair and go on
15320 to the next. This way we can extend the protocol.
15324 The process exited, and @var{AA} is the exit status. This is only
15325 applicable to certain targets.
15329 The process terminated with signal @var{AA}.
15331 @item N@var{AA};@var{t@dots{}};@var{d@dots{}};@var{b@dots{}} @strong{(obsolete)}
15333 @var{AA} = signal number; @var{t@dots{}} = address of symbol
15334 @code{_start}; @var{d@dots{}} = base of data section; @var{b@dots{}} =
15335 base of bss section. @emph{Note: only used by Cisco Systems targets.
15336 The difference between this reply and the @samp{qOffsets} query is that
15337 the @samp{N} packet may arrive spontaneously whereas the @samp{qOffsets}
15338 is a query initiated by the host debugger.}
15340 @item O@var{XX@dots{}}
15342 @var{XX@dots{}} is hex encoding of @sc{ascii} data. This can happen at
15343 any time while the program is running and the debugger should continue
15344 to wait for @samp{W}, @samp{T}, etc.
15348 @node General Query Packets
15349 @section General Query Packets
15351 The following set and query packets have already been defined.
15355 @item @code{q}@code{C} --- current thread
15357 Return the current thread id.
15361 @item @code{QC}@var{pid}
15362 Where @var{pid} is a HEX encoded 16 bit process id.
15364 Any other reply implies the old pid.
15367 @item @code{q}@code{fThreadInfo} -- all thread ids
15369 @code{q}@code{sThreadInfo}
15371 Obtain a list of active thread ids from the target (OS). Since there
15372 may be too many active threads to fit into one reply packet, this query
15373 works iteratively: it may require more than one query/reply sequence to
15374 obtain the entire list of threads. The first query of the sequence will
15375 be the @code{qf}@code{ThreadInfo} query; subsequent queries in the
15376 sequence will be the @code{qs}@code{ThreadInfo} query.
15378 NOTE: replaces the @code{qL} query (see below).
15382 @item @code{m}@var{id}
15384 @item @code{m}@var{id},@var{id}@dots{}
15385 a comma-separated list of thread ids
15387 (lower case 'el') denotes end of list.
15390 In response to each query, the target will reply with a list of one or
15391 more thread ids, in big-endian hex, separated by commas. @value{GDBN}
15392 will respond to each reply with a request for more thread ids (using the
15393 @code{qs} form of the query), until the target responds with @code{l}
15394 (lower-case el, for @code{'last'}).
15396 @item @code{q}@code{ThreadExtraInfo}@code{,}@var{id} --- extra thread info
15398 Where @var{id} is a thread-id in big-endian hex. Obtain a printable
15399 string description of a thread's attributes from the target OS. This
15400 string may contain anything that the target OS thinks is interesting for
15401 @value{GDBN} to tell the user about the thread. The string is displayed
15402 in @value{GDBN}'s @samp{info threads} display. Some examples of
15403 possible thread extra info strings are ``Runnable'', or ``Blocked on
15408 @item @var{XX@dots{}}
15409 Where @var{XX@dots{}} is a hex encoding of @sc{ascii} data, comprising
15410 the printable string containing the extra information about the thread's
15414 @item @code{q}@code{L}@var{startflag}@var{threadcount}@var{nextthread} --- query @var{LIST} or @var{threadLIST} @strong{(deprecated)}
15416 Obtain thread information from RTOS. Where: @var{startflag} (one hex
15417 digit) is one to indicate the first query and zero to indicate a
15418 subsequent query; @var{threadcount} (two hex digits) is the maximum
15419 number of threads the response packet can contain; and @var{nextthread}
15420 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
15421 returned in the response as @var{argthread}.
15423 NOTE: this query is replaced by the @code{q}@code{fThreadInfo} query
15428 @item @code{q}@code{M}@var{count}@var{done}@var{argthread}@var{thread@dots{}}
15429 Where: @var{count} (two hex digits) is the number of threads being
15430 returned; @var{done} (one hex digit) is zero to indicate more threads
15431 and one indicates no further threads; @var{argthreadid} (eight hex
15432 digits) is @var{nextthread} from the request packet; @var{thread@dots{}}
15433 is a sequence of thread IDs from the target. @var{threadid} (eight hex
15434 digits). See @code{remote.c:parse_threadlist_response()}.
15437 @item @code{q}@code{CRC:}@var{addr}@code{,}@var{length} --- compute CRC of memory block
15441 @item @code{E}@var{NN}
15442 An error (such as memory fault)
15443 @item @code{C}@var{CRC32}
15444 A 32 bit cyclic redundancy check of the specified memory region.
15447 @item @code{q}@code{Offsets} --- query sect offs
15449 Get section offsets that the target used when re-locating the downloaded
15450 image. @emph{Note: while a @code{Bss} offset is included in the
15451 response, @value{GDBN} ignores this and instead applies the @code{Data}
15452 offset to the @code{Bss} section.}
15456 @item @code{Text=}@var{xxx}@code{;Data=}@var{yyy}@code{;Bss=}@var{zzz}
15459 @item @code{q}@code{P}@var{mode}@var{threadid} --- thread info request
15461 Returns information on @var{threadid}. Where: @var{mode} is a hex
15462 encoded 32 bit mode; @var{threadid} is a hex encoded 64 bit thread ID.
15469 See @code{remote.c:remote_unpack_thread_info_response()}.
15471 @item @code{q}@code{Rcmd,}@var{command} --- remote command
15473 @var{command} (hex encoded) is passed to the local interpreter for
15474 execution. Invalid commands should be reported using the output string.
15475 Before the final result packet, the target may also respond with a
15476 number of intermediate @code{O}@var{output} console output packets.
15477 @emph{Implementors should note that providing access to a stubs's
15478 interpreter may have security implications}.
15483 A command response with no output.
15485 A command response with the hex encoded output string @var{OUTPUT}.
15486 @item @code{E}@var{NN}
15487 Indicate a badly formed request.
15489 When @samp{q}@samp{Rcmd} is not recognized.
15492 @item @code{qSymbol::} --- symbol lookup
15494 Notify the target that @value{GDBN} is prepared to serve symbol lookup
15495 requests. Accept requests from the target for the values of symbols.
15500 The target does not need to look up any (more) symbols.
15501 @item @code{qSymbol:}@var{sym_name}
15502 The target requests the value of symbol @var{sym_name} (hex encoded).
15503 @value{GDBN} may provide the value by using the
15504 @code{qSymbol:}@var{sym_value}:@var{sym_name} message, described below.
15507 @item @code{qSymbol:}@var{sym_value}:@var{sym_name} --- symbol value
15509 Set the value of @var{sym_name} to @var{sym_value}.
15511 @var{sym_name} (hex encoded) is the name of a symbol whose value the
15512 target has previously requested.
15514 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
15515 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
15521 The target does not need to look up any (more) symbols.
15522 @item @code{qSymbol:}@var{sym_name}
15523 The target requests the value of a new symbol @var{sym_name} (hex
15524 encoded). @value{GDBN} will continue to supply the values of symbols
15525 (if available), until the target ceases to request them.
15530 @node Register Packet Format
15531 @section Register Packet Format
15533 The following @samp{g}/@samp{G} packets have previously been defined.
15534 In the below, some thirty-two bit registers are transferred as
15535 sixty-four bits. Those registers should be zero/sign extended (which?)
15536 to fill the space allocated. Register bytes are transfered in target
15537 byte order. The two nibbles within a register byte are transfered
15538 most-significant - least-significant.
15544 All registers are transfered as thirty-two bit quantities in the order:
15545 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
15546 registers; fsr; fir; fp.
15550 All registers are transfered as sixty-four bit quantities (including
15551 thirty-two bit registers such as @code{sr}). The ordering is the same
15559 Example sequence of a target being re-started. Notice how the restart
15560 does not get any direct output:
15565 @emph{target restarts}
15568 <- @code{T001:1234123412341234}
15572 Example sequence of a target being stepped by a single instruction:
15575 -> @code{G1445@dots{}}
15580 <- @code{T001:1234123412341234}
15584 <- @code{1455@dots{}}
15598 % I think something like @colophon should be in texinfo. In the
15600 \long\def\colophon{\hbox to0pt{}\vfill
15601 \centerline{The body of this manual is set in}
15602 \centerline{\fontname\tenrm,}
15603 \centerline{with headings in {\bf\fontname\tenbf}}
15604 \centerline{and examples in {\tt\fontname\tentt}.}
15605 \centerline{{\it\fontname\tenit\/},}
15606 \centerline{{\bf\fontname\tenbf}, and}
15607 \centerline{{\sl\fontname\tensl\/}}
15608 \centerline{are used for emphasis.}\vfill}
15610 % Blame: doc@cygnus.com, 1991.